DLD Weblog

This is an old-fashioned weblog, in that it is a list of links I've been reading, with excerpts and notes to myself - sometimes I have lost the original reference. It covers readings around genetics, physics, epidemiology, statistics, programming, philosophy, SF, bushwalking and rockclimbing. It is a cutdown version of an internal weblog, and is updated spasmodically. Additions to earlier topics are often made to the original text.

20210818: Rhys Chatham

The notes for his Sydney Festival contribution:

In the mid 1970s Rhys Chatham was working alongside sonic experimentalists including Philip Glass, LaMonte Young and Steve Reich in New York, when his world was turned upside down by the Ramones' self-titled debut album.

Was involved in No Wave, playing with Glen Branca (recall), and including the Band of Susans members and Nina Canal (punk/noise band Ut) in his ensembles [Wikipedia]. The piece A Crimson Grail is for a 400 guitar orchestra. There are several effects in the latter that remind me of King Crimson (specifically in Starless and The Night Watch).

20210813: Liquid Tension Experiment

Commelin and Scholze decided to call their Lean project the Liquid Tensor Experiment, in an homage to progressive-rock band Liquid Tension Experiment, of which both mathematicians are fans.

Includes Jordan Rudess, Tony Levin, Mike Portnoy, John Petrucci. Portnoy and Petrucci were with Dream Theater, which "along with Queensrÿche and Fates Warning...has been referred to as one of the "big three" of the progressive metal genre." LTE formed 1999. Petrucci is "#17 greatest guitarist in Guitar World Magazine".

20210706: Intension (per Preston Stovall)

A response to Bright on the "end of analytic philosophy" comments:

Carnap was clear that his notion of intension...as functions from state descriptions to extensions... wasn't meant to do the work that older notions of intension (sense, comprehension, connotation, content, etc.) were able to do..."intensional semantics" is now, in many idiolects, coextensional with "possible-worlds semantics"...[but] Kit Fine's work on essences in the 1990s and early 2000s, for instance, is predicated on the observation that possible worlds cannot by themselves distinguish the asymmetric ontological dependence (so the thought runs) of the singleton set {Socrates} on the human being Socrates. Each exist at exactly the same worlds, and so the ontological dependence of a set on its members cannot be modelled in those terms. Consequently, models restricted to possible worlds cannot explain the different truth-conditional meanings of "Socrates is essential to {Socrates}" and "{Socrates} is essential to Socrates".

Rather than try to build a notion of intension inside model theory, then, it seems more promising to treat model theory and proof theory as two formalisms for reconstructing the old extension/intension distinction concerning complementary notions of meaning - one ontological, and concerned with word-world (and, as I'll note in a moment, world-word) relations, and the other deontic, concerned with word-word relations.

20210620: Balancing Rock and Spring Mountain

D and K. Left car at park off Woodline Drive ~1300, back at 1700.

Anticlockwise circuit - Yuddamun track, Magical Mystery Tour, off-track then found new track along ridge to Balancing Rock (1500). Nice track right on top of ridge to Frogmouth trail, then Spring Mt trail and Spring Mt summit (1530). Back via Spring Mt trail, Ring Rd, David Savage Trail. Saw only 2 walkers and 2 cyclists.

20210618: 10-year CHD risk

Obisesan et al [2021]:

CAC [coronary artery calcification] is typically quantified using the Agatston score - a sum of the attenuation (in Hounsfield units) and area of all CAC lesions in the coronary arteries - which is then categorized into very low risk (CAC = 0), mildly increased risk (CAC = 1-99), moderately increased risk (CAC = 100-299), and moderate to severely increased risk (CAC ≥ 300).

Akintoye et al [2021] pooled the ARIC, CHS and MESA studies. Of ~14000 participents with low or borderline risk (mean age 55), 38% had CAC (>0 Agatston units).

One very useful tool is the MESA Risk Score Calculator:

According to the Cardiac Society ANZ, "The evidence for pharmacotherapy is less robust in patients at intermediate levels of CAC 100-400, with modest benefit for aspirin use; though statins may be reasonable if they are above 75th centile."

20210519: Tim Williamson re HTT

There is plenty of good mathematics in homotopy type theory, which can of course be applied to natural science in the usual way. But when the theory is dressed up as a new foundation for mathematics, it is combined with philosophical claims about the mathematics on which those applications don't really make sense. For example, mathematics is claimed to be interpreted in terms of proofs, but mathematical equations about the physical world can't be proved in the relevant sense. Isomorphic systems are claimed to be identical, but physical systems can have the same structure without being identical

20210505: Terry Riley and Don Cherry

From The Birth of Loop:

Riley frequently played solo all night concerts 1968-9 using keyboards, saxophone and tape delay...Pandit Pran Nath's other students were Don Cherry and Jon Hassell. His Music for The Gift (1963) is the first piece using a delay loop.

20210504 Rodrigo y Gabriela

From Wikipedia

Acoustic guitar duo Rodrigo y Gabriela covered "Echoes" on their 2019 album Mettavolution, one of 7 tracks which won the album an award for Best Contemporary Instrumental Album at the 2020 Grammy Awards.

20210503 Mt Philps at Glen Rock

D and K. Left Gabba 0900, Glen Rock 1050, Mt Philps 1300 (phone coverage just at summit), back at car 1510. Knee-high grass everywhere with rain (hiding rocks). Followed pad ~20m S along road past information hut that joined correct fire trail (took lower fork after 5 min) and climbed to ridge top at 1200. To summit steep at top, dislodged a few rocks up to 30cm diameter. Dropped down northern ridge to fenceline and track.

20210416 Maroon

Stroll with MG over to base of Graveyard Ridge and up. Where ascend past top cliffs traversed to Tiger Wall Annex. Started 10:40, Caldera at 1500, top of tourist track 1600, car at 17:40.

20210416 Frege-Geach argument

The Wikipedia article on Expressivism says the Frege-Geach Argument [Geach 1960] is that complex statements involving (embedding) simpler, say, ethical statements eg Modus Ponens "If it is wrong to tell lies, then it is wrong to get your little brother to tell lies" seem to work even though a noncognitivist should find it hard to defend this (Uchii in his example, "if you support the rule of law, then you should not drive without a licence", seems to think this is not a problem). Brandom characterizes the declaration (in naive expressivism) to have imperative force, and testable by substitution into similar complex sentences. He says second-wave expressivism (Blackburne, Gibbard) attempts an answer.

Schroeder [2012] in the Routledge Encyclopedia of Philosophy says:

...Simon Blackburn in Spreading the Word (1984). According to a slight adaptation of Blackburn's central idea, what it is to think that if P, then Q, is just to "boo" thinking that P but not thinking that Q... Though Blackburn's account may be able to explain why some arguments are valid, it has been argued to predict too many arguments to be valid.

Yeah, but consider theological arguments, or astrological inferences - how are they any different?

20210414: PRS AUC

Zheng et al [2020] mention AUC = qnorm(sqrt(0.5*h2).

And see https://github.com/yandorazhang/CancerEffectSize.

20210407: Counterfactual theory of explanation

Woodwards's definition of causation underpins his definition of explanation:

An explanation ought to be such that it enables us to see what sort of difference it would have made for the explanandum if the factors cited in the explanans had been different in various possible ways.

He [Woodward 2018 ] and Reutzinger and colleagues [2020] suggest that the factors involved might not have to be causes, in order for an explanation to successfully follow this form, giving a counterfactual explanatory monism.

Older examples of a unitary theory of explanation are Hempel's "covering law" and Kitcher's "unification". Ross [2020] comments that non-causal explanations (under a dualistic model) often involve topological features of the system (eg Konigsberg bridges), but that in fact topology is causal in many scientific settings, even though we normally model these as timeless. Lange [2014,2016] considers that the counterfactuals in a mathematical explanation (ie proofs) are necessarily trivial (quodlibet) in the standard semantics. Baron et al [ 2020] investigate nonstandard truth-closeness-based approaches to the latter, in their example by creating a structural equation model including a counterfactual variable. Since their example was a Euclidean geometry problem, I wonder if the free variable could have been explicitly indexing the geometry or topology in which the problem was set.

20210325: UKBB Telomere length

Telomere length for all 500,000 participants...

20201217: Recommendations in recent philosophy

A list of best papers 2010-20

20201209: Intel fortran compiler

Steve Lionel writes:

Intel has released its oneAPI toolkit, which includes full support for Fortran 2018. Even better news, it's free if you don't mind "community support". (If you want Priority Support, you can buy it.)

There are two Fortran compilers here. ifort is the next version of what had been in Intel Parallel Studio XE, and is the one with the full Fortran 2018 support. ifx is a new, LLVM-based compiler and is still considered "beta".


As is usually the case, there may be divots in the F2018 support - these should be reported to Intel. I have been trying the beta and it's very good.

20201127: safecopy

safecopy - rescue data from a source that causes IO errors --stage1 Preset to rescue most of the data fast, using no retries and avoiding bad areas.

20201118: Parkinson's Disease PRS

Nalls et al [2019] is a meta-analysis of 17 GWAS including UKBB relatives of cases, which "identified 90 independent genome-wide significant risk signals across 78 genomic regions...explain[ing] 16-36% of the heritable risk".

Jacobs et al [2020] use PRS and clinical data in 1267 UKBB cases to assess prediction. Predictors included "a positive family history of PD, a positive family history of dementia, non-smoking, low alcohol consumption, depression, daytime somnolence, epilepsy and earlier menarche".

20201116: The pragmatic maxim


if one can define accurately all the conceivable experimental phenomena which the affirmation or denial of a concept could imply, one will have therein a complete definition of the concept, and there is absolutely nothing more in it
And (also via David Matthews (U Adel):

...scientific inquiry is characterised by "contrite fallibilism" and...the scientist is distinguished by "his readiness to dump the whole cartload of his beliefs the moment experience is against them"...truth is "that concordance of a statement with the ideal limit towards which endless investigation would tend to bring scientific beliefs".

20201116: Against mental epiphenomenalism

Place [1988] suggests if there is "no causal connection between a mental event and the description that is purportedly given of that mental event in the subject's introspective report...it cannot be a genuine description of the event in question". A riposte might be that the verbal description is emitted from the "causal layer". Another might be that introspection is worthless.

20201109: Formalizing natural language

On n-Category Cafe:

To the Davidsonian analysis of event structure, Peter Hacker proposed the example of
He wisely apologised.

Evidently there is a context for such a statement that contains the identity of the apologiser, the act for which the apologising is required, the people who were offended by the act, some idea of how they might seek redress, and so on. A story of a certain shape needs to be supplied. Ranta's ideas on narratives as DTT contexts seem to me appropriate.

How about: "An apology was in order".

20201106: Square Limit

Henderson [1982, 2002] Functional Geometry (re-)reconstructs the Escher woodcut in a functional picture language. His R version is at https://mhenderson.github.io/funcgeo/.

20201102: Supervaluation

Smith and Brogaard [2000] summarize this material in a section "The Problem of the Many"...

A proposition about a imprecisely defined entity can be true about the many almost identical entities you might have specified using more information. Van Fraasen [1966] and Fine [1975] suggest there are supertrue sentences in many cases, while if they are only mostly true there is a "supervaluational truth-value gap". What is the referent for "Mont Blanc"? Are rabbits in holes of that region included?

20201030: Direct reference (and ostention)

Bensusan and da Silva [2020], presenting an indexical account of propositions, comment:

Among the many groundbreaking developments in the philosophy of language and thought in the last decades motivated by the idea of direct reference, an important one was the clear disconnection of the intentional and the cognitive (and the perceptive). A language user can denote something or successfully point at something, with little or no knowledge of what is denoted. An utterance can refer even while no one knows how to correctly describe its referent. Intentionality is detached from cognition.

Michaelson's SEP summary of reference includes overlapping models: descriptivism ("words refer in virtue of being associated with a specific descriptive content"); causal model; character model (conforming to the rules of reference - "indexical tokens acquire their reference because they are uttered in a particular context, presumably intentionally. But they acquire their reference independently of anything having to do with the speaker's mental state"); intentionalist (a la Grice, so you interact to clarify what we are currently talking about). And Quine, of course.

20201026: Phronesis

Assorted quotes

virtue is "a state of character concerned with choice, lying in a mean, i.e., the mean relative to us, this being determined by a rational principle, and by that principle by which the man of practical wisdom would determine it. That we know this "golden mean" by examining the choices made by the man of practical wisdom and also recognize the man of practical wisdom by the fact that he typically selects the mean is a tautology often noted by students of the Ethics.

Thus, we must return to the principle by which the man of practical wisdom operates. Though difficult to operationalize, it is clear. It requires in every particular situation that a balance of what is desirable and what is reasonable be determined through deliberation. "Choice is deliberate desire, therefore both the reasoning must be true and the desire right, if the choice is to be good and the latter must pursue just what the former asserts" (NE 1139a23-26). Although phronesis has universal applicability, it is not "concerned with universals only - it must also recognize the particulars; for it is practical, and practice is concerned with particulars" (NE 1141b14-16). As Ronald Milo observes, Aristotle's notion of good deliberation "presupposes both correct reasoning and reasoning with a view to a good end."

phronesis and moral virtue, or goodness of character, are closely and reciprocally related, in the sense that neither can exist without the other

Aristotle: "perhaps one's own good cannot exist without household management nor without a form of government."

The Greek concept phronesis is famously thematized by Aristotle in the Nicomachean Ethics, and is usually translated as "practical wisdom", as contrasted to technical skill in arts and crafts (techne), the knowledge of science (episteme), the theoretical wisdom of philosophy (sophia), and intuitive reason (nous).

The rulers of the polis are required to render judgment in complicated particular cases in which general laws provide inadequate guidance concerning the right way to proceed, and in which the aim of the decision is the good action and the good life in itself and not some other thing, such as health.

20201026: Pan UKBB association results


Summary statistics MatrixTable: gs://ukb-diverse-pops-public/sumstats_release/results_full.mt (12.78 T)
    Meta-analysis MatrixTable: gs://ukb-diverse-pops-public/sumstats_release/meta_analysis.mt (12.54 T)

Essential Tremor (code 1525) N.Cases = 133

20201021: 100 best fantasy novels

From Time magazine.

Slants to younger audiences and best selling series...

Susan Cooper The Dark is Rising
Natalie Babbitt (1975) Tuck Everlasting
Brian Jacques (1986) Redwall (Series, younger)
Diana Gabaldon (1986) Outlander (Series, younger)
Cassandra Clare City of Glass (Mortal Instruments v 3)
Ken Liu
Isabel Ibañez (2020) Woven in Moonlight

James Nicoll suggests Melanie Card (2011) Ward Against Death (series).

20201017: Logans's Ridge

With K. Left Yellowpinch 0745. Summit ~12:15. Back at car ~16:10. Two detours from usual route, last being ~100 m from top where traversed W on pad, then technical scramble up below big gully, and had to backtrack slightly to get back onto ridge. Returned Peasant's, so I could collect and drink another 4 litres of water.

20201016: enthymeme

A syllogism with an unspoken premise or conclusion. Uncharitably, hiding an unreasonable assumption; charitably, enjoining the listener to perform and agree with the judgment entailed in the hidden material: "rhetoric finds its end in judgment" - Aristotle.

20201014: Du Fu

Tang era official 712-770 and the "greatest" of poets (the poet sage, shi sheng), caught up by the An Lushan Rebellion. When Commissioner of Education in Huazhou, a poem includes:

I am about to scream madly in the office
Especially when they bring more papers to pile higher on my desk.

Discussed with interesting commentary at Language Log:

Spring View

Country damaged mountains rivers here/continue
City spring grass trees deep
Feel moment flower splash tears
Regret parting bird startle heart
Beacon fires join three months
Family letters worth ten thousand metal
White head scratch become thin
Virtually about to not bear hairpin

David Hinton's translation:

The country in ruins, rivers and mountains
continue. The city grows lush with spring.

Blossoms scatter tears for us, and all these
separations in a bird's cry startle the heart.

Beacon-fires three months ablaze: by now
a mere letter's worth ten thousand in gold,

and worry's thinned my hair to such white
confusion I can't even keep this hairpin in.

Conal Boyce comments the morphosyllabic language means each line is "like a slideshow". The literal gloss does this, but "lands...somewhere between the New Age Chinoiserie of the 1970s and just plain goofy".

Papers and books of Richard John Lynn .

20201008: On semantics

Via DN, the linguist Chris Potts, discussing GPT-3, contrasts the Lewisian view that "Semantics with no treatment of truth conditions is not semantics" with that of Jerrold Katz, who wrote that "'[t]he arbitrariness of the distinction between form and matter reveals itself', ...argu[ing] throughout his career for an expansive semantic theory in which words and phrases are essentially their own meanings" (generative semantics).

Lewis [1970]:

A meaning for a sentence is something that determines the conditions under which the sentence is true or false. It determines the truth-value of the sentence in various possible states of affairs, at various times, at various places, for various speakers, and so on...[T]he truth-value of a sentence [is] the extension of that sentence; the thing named by a name [is] the extension of that name; the set of things to which a common noun applies [is] the extension of that common noun. [...An] input package of relevant factors [is] an index; any function from indices to appropriate extensions for a sentence, name, or common noun [is] an intension...The plan to construe intensions as extension determining functions originated with Carnap (1947).

Deery "briefly sketches" Putnam-Kripke causal-historical externalism as implying (but perhaps not entailing) essentialism, in that reference to a natural kind is to a natural kind ("essential nature becomes an empirical matter").

20200930: Financialization

Baud and Durand [2012]

(1) financialization of objectives refers to the implementation of shareholder value norms, whose concrete consequences are an increase of the financial flows from non-financial corporations to the financial sector; (2) financialization of investment refers to the increasing share of financial assets owned by non-financial firms; and (3) financialization of operations refers to the development of financial activities and relationships, notably offered to customers and/or imposed on workers and suppliers, by non-financial firms.

"From the beginning of the 1990s up until 2007, leading retailers experienced a slowdown of their growth in domestic markets and yet generated an opposite upward trend in return on equity (ROE)...[via] foreign expansion, financialization of assets, deterioration of suppliers' and workers' positions and the use of working capital management to transform market power into financial gains."

20200914: Hong Lee and others' CORE GREML


The CORE GREML approach fits the Cholesky decomposition of kernel matrices in an LMM, to estimate the covariance between a given pair of random effects. Using extensive simulations, we show that CORE GREML outperforms GREML, providing estimates of variance and covariance components that are free from bias due to correlated random effects.

They report results from a GTE model that "decomposes phenotypic effects into the random effects of the genome and the imputed transcriptome and residuals, i.e., y=g+t+e, [where] the imputed PrediXcan transcriptome consists of expression levels of 227,664 genes from 43 non-sex-specific tissues" [Gamazon et al 2015]. The GTE model always fitted better than a GE model, and better than a GGTE model, where GT was derived from the unweighted genotypes of the 1316391 SNPs used to impute the transcriptome values. Unsurprisingly, G and T are correlated.

In passing, Shor et al [2019] compared their Haseman-Elston based program to Wombat, finding the latter 13 times faster for estimating heritability (AE model) for a dataset containing 250000 records, but crashing when they tried 500000 records. However their package could handle epistatic components without matrix inversion.

20200910: Consequentalization

Campbell Brown in Consequentialize This:

To 'consequentialize' is to take a putatively nonconsequentialist moral theory and show that it is actually just another form of consequentialism... My strategy is to decompose consequentialism into three conditions, which I call 'agent neutrality', 'no moral dilemmas', and 'dominance', and then to exhibit some moral theories which violate each of these.

...You say you're concerned about the guy's rights? No worries; we'll just build that into your theory of the good. Then you can be a consequentialist too. And just like that, you've been assimilated.

Dreier is one such, because it is commonsensical. And Jennie Louise [2004] says "[s]ince we are now all under the consequentialist umbrella, the question now becomes not whether we should be consequentialists or not, but whether we should be value-neutralists or value-relativists". Schroeder [2017] says the "Compelling Idea" behind consequentialism is that "it is always permissible to bring about the best outcome...This attractive structure, though, is paired with deeply counter-intuitive verdicts on individual cases". Douglas Portmore states as a maxim that we have more reason to perform one action over another iff, and in virtue of the fact that, we have more reason to desire the outcome associated with the former (disputed by several authors). One problem with consequentializing is that the transformation may be ugly eg absolute injunctions imply infinite value. Colyvan et al [2010] specifically ask whether decision theory can successfully represent the motivation and reasoning of "agents who subscribe to [nonconsequentalist] ethical theories", and mentions Hajek's [2003] hyperreal models of Pascal's Wager (eg avoiding where any mixed strategies still have infinite support).

20200828: Darren Bradley

Yet another interesting interview by Richard Marshall, where Darren Bradley talks about Carnap, formal epistemology, Bayesianism, many-worlds.

Let's start with relevant alternatives theories of knowledge. The idea is that you don't simply know that p; you know that p rather than q. For example, you don't know that grass is green; you know that grass is green rather than white, black, red etc. This theory is useful for answering skepticism about the external world. The skeptic points out that you might be a brain-in-a-vat, and infers that you don't know you have hands. The relevant alternatives theorist can acknowledge a sense in which the skeptic is right - you don't know you have hands rather than being a brain-in-a-vat. However, you know you have hands rather than having hooks.

Yes I think that a debate in recent philosophy of biology was solved by Carnap. The point can be made using the slogan that natural selection operates through the survival of the fittest. Some object that "survival of the fittest" is an a priori tautology, so can't be a law of nature. My position is that it is an a priori tautology and also a law of nature...

For a simpler example, think about carburetors. What it is to be a carburetor is to be a device for mixing air and fuel. Consider "the carburetor mixes the air and fuel". That's an a priori tautology, as it follows from the definition of "carburetor". But it can still be a law, and it can still be explanatory. If I don't know that there is a device that mixes air and fuel, or don't know that air is needed for the engine to work, it could be explanatory to be told "the carburetor mixes the air and fuel".

In Bradley [2018], he argues why philosophers should prefer simpler theories against Huemer [2009], who had put forward four counterarguments, where what might work in the sciences will be inadequate for metaphysics etc: Empiricist: science lacks criteria to show that theories preferred as simple are really better; Likelihood: likelihood-based tests fail for philosophical hypotheses eg dualism; Numerousness: larger number of complex hypotheses; Bounded Asymmetry: alternative theory complexity can only go up [?].

Huemer would have been better off just claiming Knightian uncertainty, it seems to me. In another 2018 publication, he argues that Carnap's dismissal of metaphysics is because Carnap felt metaphysical hypotheses could not be justified, rather than verified. Bradley's asks what conceivable evidence could support S over ~S, commenting that in "many cases, metaphysicians engaged in a debate intend that their competing hypotheses should be empirically equivalent":

1C. If we can have justification for S over T then evidence which would support S over T is conceivable.
2C. It is not the case that evidence which would justify S over T is conceivable.
3C. Therefore we cannot have [epistemic] justification for S over T. [though pragmatic justification is quite possible]

This reminds me of Hume on miracles.

20200825i: Scientific theories

From a review of Hardt's Economics Without Laws, where Hands writes:

The first major theme is that not only have philosophers moved away from the Humean, event-regularity, view of scientific laws that dominated the literature of the nineteenth century, they have also steadily moved away from the syntactic view of scientific theories associated with logical empiricism: theories as axiomatized systems of sentences linking theoretical terms and observational terms, generally through some version of correspondence rules. There are many aspects to the recent philosophical developments Hardt discusses, but three of the most important are (i) an increased concern with the actual practice of science (I would call this the naturalistic turn), (ii) an increased emphasis on scientific models and a decreased emphasis on scientific theories, and (iii) the development of various versions of the semantic (or model-based, or non-statement, or predicate) view of scientific theories in which scientific theories are simply families of models.


Under the believable world conception "a model is an entity containing mechanisms that are believed to be similar to the ones operating in the real world" and they "are similar because a model of a mechanism demonstrates the reality of a mechanism by isolating it" (148). Hardt's believable world conception draws heavily (with proper recognition) on both Cartwright's nomological machines and Robert Sugden's notion of a credible world or model (Sugden, 2000; 2009; 2011), but is not identical to either. Hardt summarizes his account of the relationship between idealized economic models and scientific explanation toward the end of the book:
models explain by depicting structures which enable the workings of mechanisms, or models are just mechanisms' descriptions. And such models of mechanisms produce beliefs about the real world and thus those beliefs are always true in models producing them. Here my answer to the question of how models explain refers closely to Cartwright's idea of models as blueprints for nomological machines that produce "insights" about regularities in the actual world. (150-151)

And points to Marc Lange's What Makes a Scientific Explanation Distinctively Mathematical?

Here is a very simple example (inspired by Braine [1972], p. 144):The fact that twenty-three cannot be divided evenly by three explains why it is that Mother fails every time she tries to distribute exactly twenty-three strawberries evenly among her three children without cutting any (strawberries!)...[A] manipulability account of causal relations (such as Woodward's) says roughly that C is a cause of E exactly when systematic changes in E can be brought about by suitable interventions on C. Clearly, manipulation of the numbers of strawberries or children would bring about corresponding changes in the outcome of Mother's attempt.

...I agree with Mancosu, Lipton, and Kitcher that distinctively mathematical explanations in science are non-causal explanations... the categorical ground for a key's power to open a distant lock resides in the key's structure and the lock's structure, but the lock's structure is not a cause of the key's power (on pain of action at a distance on the cheap).

...not all explanations are causal...Spatiotemporal symmetries and the principle of relativity do not describe causal relations, even at a coarse-grained or abstract level. Rather, they impose constraints on the laws of nature...

20200824: Law of comparative advantage

Krugman's Ricardo's difficult idea describes it as "simple and compelling to those who understand it, but about which intelligent people somehow manage to get confused time and time again".

20200817: Heterodox economics

The Harvard-based "radical" or "heterodox" economist Stephen A. Marglin interviewed here regarding his book Raising Keynes.

The General Theory...never went beyond a simplified model [of interest] with only two financial assets - cash and bonds... central banks provide the missing link in Keynes's theory. Keynes disputed the mainstream view that a market equilibrium meant full employment; he believed instead that equilibrium would usually fall short of full employment. But if you don't have a theory for the level of the rate of interest, you don't have a theory of equilibrium at all. On its own terms Keynes' model was incomplete. We are thrown back on a non-market theory of interest rates!

The final innovation of the book is more directly related to current policy proposals like the Green New Deal. This is the section of the book that deals with "functional" finance... the [GND] requisite tax increases would not be to balance the budget for the sake of balancing the budget (sound finance), but to keep aggregate demand in line with the productive capacity of the economy (functional finance)...in the context of full employment, it will require a reduction in demand somewhere along the line, starting with the reversal of Trump's tax "reform".

Economics claims to be primarily a descriptive subject, with a secondary normative function. But I think the relationship is exactly the opposite - economics is basically a normative discipline, and those normative beliefs shape our descriptive tools and analyses. That is not a criticism; it is rather a call for honesty about our political and moral presuppositions, and how they shape the way we think about the economy.

There's another aspect to economics, which I refer to in an earlier book, The Dismal Science, which is the constructive function. Economists not only describe and make judgments; they also try to shape the economy in the image of their economics. Given the predominance of the mainstream, this means shaping the economy in the image of a self-regulating market and curing any market failures by more and better markets. This would less pernicious if economists were upfront about their ideological presuppositions rather than claiming their economics is ideologically neutral.

Collaborated with Sen at one time.

Hein [2017] suggests there are five Post-Keynesian strands: fundamental Keynesian (via Hyman Minsky, GLS Shackle, and Sydney Weintraub); Kaleckian; Kaldorian; Sraffian/neo-Ricardian; and Institutionalist (via Veblen, Lerner, Galbraith). Aside from these, the Marxian approach is still an important heterodoxy, and see Harrod. Lavoie [2019] points out that:

[t]he main MMT authors - Randall Wray, Matt Forstater, Stephanie Bell-Kelton, Pavlina Tcherneva, Andrew Watts, Eric Tymoigne - were all tied to post-Keynesian economics from the very start. The only exceptions would be Scott Fullwiler, who came from the Institutionalist tradition, and William Mitchell, who was closer to the Marxian tradition

The stock-flow consistent approach is at the heart of post-Keynesian economics since the mid-1990s, and it was a critical contribution of Godley and Cripps [Macroeconomics (1983); Lavoie was a student of Godley].

The latter gives the macroeconomic balance sheet, with K the fixed capital stock,the only asset that belongs to firms and W, the financial wealth stock, which is ultimately attributed to households.
Public Debt +W -W1-WE0
Fixed Capital +K K
Σ +W+K -W1-WEK

Alex Williams comments in a review [he is a Levy Institute graduate, a post-Keynesian centre, the home of Randall Wray]:

...the macroeconomic detective is armed with the knowledge that balance sheets always balance. This simple insight, that every transaction has two sides, means that there are certain aggregate relationships between transactions that must obtain for the world economy. Knowing this, it's possible to chase actors across seemingly unrelated balance sheets to find where the system as a whole was forced to balance. From here, the skillful economist can identify the long-run tendencies that a given balance is likely to create. (Wynne Godley famously predicted the Global Financial Crisis in just this way, following US mortgage debt around the world and back.)

Specifically, on Michal Kalecki [Hein 2017]:

Kalecki's (1939) principle of effective demand, as an earlier alternative to Keynes's (1936) approach, has the profit share (or the wage share) as a determinant of the multiplier effect of short-run exogenous expenditures (investment, government expenditures or exports), together with the propensity to save out of profits...[ie] wage-led growth...a major contribution...has been his theory of distribution based on mark-up price setting of firms in incompletely competitive markets.

Gross profits net of taxes = Gross investment + Capitalists' consumption + Government budget deficit + Export surplus - Workers' saving

Again from Alex Williams:

Pettis and Klein [2020] note in passing that the era of scarcity as the fundamental economic fact has ended, and in its place is the fundamental fact of demand. We have known since Keynes that the limiting factor in developed economies is not a scarcity of resources, but rather a scarcity of demand for finished output.

Hein is a little more circumspect, noting that the empirical support for a wage-led growth and demand are mixed, even within the same country, probably reflecting the analysis, viz short-term (more profit-led) v. longer-term scale (more wage led) effects.

Post-Keynesian-Kalecki models can include investment plans of firms that can allow saving and investment to equalize at equilibrium, so that
g = S/K,
remains true, where K is the amount of homogenous capital, S the total savings in the economy, and g the growth rate of capital. Dutt [ 2016] presents a particular PKK model with two groups "the top and the rest", which allows "financialization and the increasing importance and income of top managers (or CEOs and their allies and close subordinates) [to] explain both lower rates of growth, and rising wealth and income inequality especially with deficient aggregate demand".

Recall that Piketty's main interest is the historical relation between g (growth of the whole economy) and r, the rate of return on capital, see for example, Piketty [2015]:

In a representative-agent framework, what r > g means is simply that in steady-state each family only needs to reinvest a fraction g/r of its capital income in order to ensure that its capital stock will grow at the same rate g as the size of the economy, and the family can then consume a fraction 1-g/r. For example, if r = 5 percent and g = 1 percent, then each family will reinvest 20 percent of its capital income and can consume 80 percent.

NCM = New Consensus/Classical Macroeconomics [Clarida,Gali,Gertler; Goodfriend, King] - the current model underlying central bank interest as major lever.

More from Hein:

In a post-Keynesian macroeconomic policy mix, fiscal policies have a major impact on economic activity and the distribution of disposable income, and should thus actively take care of real stabilisation of the economy in the short and the long run, using government expenditures and taxation as tools without any autonomous government deficit targets. It thus means to follow a functional finance approach in the tradition of Lerner (1943) (Arestis/Sawyer 2003, 2004b, Setterfield 2009b). Potential limits to government debt in this kind of approach are a matter of controversy between those sympathetic to neo-chartalism and functional finance - what is now called "modern money theory" (MMT) (Wray 2012) - and the critics of such an approach (Palley 2015b). The relevance of government debt limits will depend on the precise institutional link between the government and the central bank, the international acceptance of the national currency, whether private and public debt is denominated in the domestic currency and so on (Lavoie 2013). In particular, if central banks act as a lender of last resort for the government and guarantee government debt, and private agents thus do not have to fear the illiquidity or insolvency of the government, the level of government debt or government debt-income ratios should be of minor concern, as has been pointed by the proponents of MMT.

Added 20200828: Perry Anderson's review of Tooze's Crashed discusses Tooze's use of stock-flow consistency in analysing 2008, and Cedric Durand's criticism that Godley saw a key advantage of his approach being that it integrated the financial and real economies, and Tooze did not discuss that nexus.

20200816: Postphenomenology (Don Ihde)

Homo faber Revisited: Postphenomenology and Material Engagement Theory

An interest in technology as prosthesis (a la extended mind). The phenomenology of sensing the world once the intermediation becomes transparent, or is the experience. And the centrality of technology to being an anatomically modern human, "How Things Shape the Mind" [Malafouris 2013], and coevolution of the cognitive niche. Thinks an emphasis on the computational mind ignoring the accidents and constraints of matter is deceptive. Ihde points to Merleau-Ponty [1962]:

The blind man's stick has ceased to be an object for him, and is no longer perceived for itself; its point has become an area of sensitivity, extending the scope and active radius of touch, and providing a parallel to sight. In the exploration of things, the length of the stick does not enter expressly as a middle term: the blind man is rather aware of it through the position of objects than of the position of objects through it. The position of things is immediately given through the extent of the reach that carries him to it, which comprises, besides the arm's reach, the stick's range of action.

This skips to some extent the primary/secondary qualities distinction, ISTM. Sensuous aesthetics is of the primary.

20200810: Tibro with G and V

Zeitgeist; DLD lead all. Started ~10:30. P1 crux harder than all of us remembered - looks like 2-3 holds have come off. Topped out ~15:00. Rapped in 2 60m falls, walking out at 17:00.

20200804: Game semantic models of computation

See also Hintikka's work on game-theoretic semantics.

The game-semantic model of computation (Abramsky, Hylan, McCluskey) gives "an intrinsic (i.e., without recourse to another model of computation), non-inductive, non-axiomatic [model], which is similar to classic Turing machines, yet beyond computation on natural numbers (which let us call classical computation), e.g., higher-order computation" [Yamada 2019]. Such higher-order computation is carried out in typed functional language eg PCF - "computation that may take (as an input) or produce (as an output) another computation...Existing models of higher-order computation are either syntactic (such as programming languages and nested sequential procedures), axiomatic (such as Kleene's schemata S1-S9) or extrinsic (i.e.,reduced to classical computation by numbering whose 'effective computability' is often left imprecise" [ Yamada 2017 ].

Yamada [2017] "point[s] out that computation is an intensional concept in the sense that not only what its result (or extension) is but also how it computes, i.e., its algorithm, matters in an informal sense. For instance, the programs double(succ(5)) and succ(succ(double(5))) clearly have the same value, namely 12, but different algorithms...these are extensionally equal but intensionally different. Let us call informally the equality of algorithms intensional equality of computation."

20200804: Shooting Room Paradox

Reviewed with a solution by Bartha and Hitchcock [1999]

Introduced by John Leslie as a thought experiment relevant to his Doomsday Argument. Sequential trial where batches of 1, 9, 90, 900... individuals drawn from an countably infinite population will terminate with 1/36 probability per batch. There is a infinitesimally small probability of "90%" of the population losing, even though each individual has a 35/36 probability of surviving the episode they are sampled into.

20200727 William James on belief

Mentioned here:

...two goals that we have when we have beliefs - believing truths and avoiding errors. When we have a belief, it gives us a chance of being right, but it also runs the risk of being wrong. In constrast, when we withhold judgment on a proposition, we run no risk of being wrong, but we give ourselves no chance of being right.

20200723: Post-classical probability

Barnum and Wilce [2012] introduce post-classical probability theory as a very abstract approach developed for quantum mechanics. There is a finite test space X of all outcomes along with the covering of these by non-empty sets (the tests), and probability weights summing to unity for each test, with Ω the set of all weights for X. A quantum version has a Hilbert space. A probabilistic model is a structure {X,Ω}. Scandolo [2019] applies such to "the information-theoretic foundations of thermodynamics and statistical mechanics...in arbitrary physical theories...[G]eneral probabilistic theories identif[y] the two main ingredients of any physical theory to be its compositional structure (how to build experiments) and its probabilistic structure (how to assign probabilities to experimental observations)". Acceptable theories need to hew to four information-theoretic axioms, informally stated as follows: Causality - No signal can be sent from the future to the past. Purity Preservation - The composition of two pure transformations is a pure transformation. Pure Sharpness - Every system has at least one pure sharp observable. Purification - Every state can be modelled as the marginal of a pure state, uniquely up to local reversible transformations"

20200723: Information causality principle

Weilenmann and Colbeck [2020]:

Information causality is a candidate principle for singling out quantum theory. Roughly speaking the principle is that sending n bits of classical information from one party to another cannot give the recipient access to more than n bits of previously unknown information regardless of any pre-shared resources the parties may have.

Pawlowski et al [2009], in the paper that introduces the concept:

[I]nformation causality holds true even if the quantum bits are transmitted provided that they are disentangled from the systems of the receiver. This follows from the Holevo bound, which limits information gain after transmission of m such qubits to m classical bits.

20200720: flang

Playing with flang-7. This is the Ubuntu-packaged older flang - recall that LLVM 10.0.0 24 released March 2020). Some clang options do not work as advertised. Here is some interpreted (actually JIT, I think) Fortran:

> cat test.f90
program test
 write(*,*) 'Hello from Fortran'
end program test
> flang -emit-llvm -c -o test.bc test.f90
> lli -version
LLVM (http://llvm.org/):
  LLVM version 10.0.0
  Optimized build.
  Default target: x86_64-pc-linux-gnu
  Host CPU: nehalem

> lli -entry-function=MAIN_ -load=/usr/lib/x86_64-linux-gnu/libflangrti.so -load=/usr/lib/x86_64-linux-gnu/libflang.so test.bc
 Hello from Fortran
> llvm-bcanalyzer test.bc
Summary of test.bc:
         Total size: 19104b/2388.00B/597W
	Stream type: LLVM IR
   # Toplevel Blocks: 4
> llvm-dis < test.bc | less
> llc test.bc -o test.s

Also see https://www.llvm.org/docs/BitCodeFormat.html

20200713: Copernicus and Ibn al-Shatir

Saliba [2002] summarizes the refinements that Islamic astronomers made to the Ptolemaic model, following an avalanche of contradictory observations from the 9th century CE onwards (exaggerated equinoctal precession and obliquity of the ecliptic). In the 13th century, Urdi's lemma gave a solution for the planetary equant - "a deferent that moved uniformly in place around the axis of the centre of the planet", and al-Tusi the Tusi Couple, which in the steam engine is the sun-and-planet mechanism, converting linear to circular motion.

Ibn al-Shatir (born ~1305, muezzin and time-keeper in Damascus) uses both of these in a refined geocentric model. Copernicus uses same Tusi Couple, and his diagram is lettered in the same order as that in al-Shatir's MS. Abbud [1962] showed that the numerical agreement between al-Shatir and Copernicus is also extremely tight.

Swerdlow and Neugebauer [1984] argued that this implied Copernicus had seen a translation or summary of al-Shatir, maybe via the Jewish The Light of the World [Morrison 2017]. The 11th Century Alfonsine Tables, for example, are the work of Jewish astronomers in Spain.

Copernicus probably jumped from here to a geostatic heliocentric system for all the other planets and then a straight heliostatic system. Blasio [2014] argues that the parallels are just that, given that "Copernicus happily cites numerous earlier Islamic sources - and there is virtually no evidence that [the later Islamic astronomical works] were available to him". Nikgahm and Ragep [2019] think the agreement is not explainable this way, suggesting Copernicus had seen diagrams of al-Shatir's "individualistic" model for Mercury, the planet that has the most complex Ptolemaic model, and that had a large number of quite different previously published alternatives.

20200713: Greg Cochran, Henry Harpending, Jason Hardy

The authors [2005] of Natural History of Ashkenazi Intelligence. Cochran has a blog https://westhunt.wordpress.com/ discussing human evolutionary genetics. Cochran and Harpending [2009] argue for increasing recent selection on human genomes in their book The 10,000 Year Explosion.

From the end of Cochran's review of Plomin's Blueprint:

Indeed, social scientists have done such a terrible job that it's hard to see how the field can be repaired. They wanted the false results they got, and they still do. I'm sure their descendants will as well. Isn't heritability grand?

20200710: The Mip*=r.e. preprint

That is, Ji, Natarajan, Vidick, Wright and Yuen, who showed that "the complexity of approximating the quantum value of a non-local game G is equivalent to the complexity of the Halting problem."

A two-player non-local game is played between a verifier and two cooperating players named Alice and Bob who cannot communicate with each other once the game starts. During the game, the verifier samples a pair of questions (x,y) from a joint distribution μ, sends x to Alice and y to Bob, who respond with answers a and b respectively. The verifier accepts if and only if D(x,y,a,b)=1 for some predicate D. The quantum value of a non-local game G, denoted by wq(G), is defined to be the supremum of the verifier's acceptance probability over all possible finite dimensional quantum strategies of Alice and Bob for the game G.

Scott Aaronson: "To say it more simply, entangled provers can convince a polynomial-time verifier that an arbitrary Turing machine halts."

Recursively enumerable (or computably enumerable) is

...the simplest of all complexity classes, a language is in RE if there is some Turing machine M such that x is in L if and only if M on input x accepts. For x not in L, M on x can reject or run forever. The classic halting problem, the set of descriptions of Turing machines that halt on empty input, is RE-complete.

MIP* is the set of things provable to a classically random polynomial-time verifier by two separated provers with an unlimited number of quantumly entangled qubits.

[An earlier] paper [by Natarajan and Wright] showed that quantum entanglement actually gets more, much more, than classical provers...[this] get[s] a much stronger and tight result, and...disproving the Connes' embedding conjecture.

Recall that

interactive proofs [are] where the verifier verifies the correctness of a statement, by interacting with the prover and by using randomness. As was shown later by Lund, Fortnow, Karloff, Nisan, and Shamir, interactive proofs seem to be much more powerful than standard proofs, as every language in PSPACE can be verified efficiently via an interactive proof, whereas only languages in NP can be verified efficiently via a standard proof.

[In] multi-prover interactive proofs (MIPs)...there are several [only need 2] provers that are proving a statement to a single polynomial time verifier, and the assumption is that these provers do not communicate with each other during the proof...These can be converted into probabilistically checkable proofs (PCPs).

20200707: Bennett's logical depth

Delahaye and Vihal [2018] write in the American Philosophical Association Newsletter on Philosophy and Computers about the increase in complexity of the universe ("Big History") via the concept of logical depth:

A first attempt to formulate [Charles H] Bennett's idea is to say that the logical depth of S, LD(S) is the time it takes for the shortest program of S, S*, to produce S [the decompression time]...The concept of thermodynamic depth introduced by Seth Lloyd & Heinz Pagels (1988) is defined as "the amount of entropy produced during a state's actual evolution" [the difference between the system's coarse- and fine-grained entropy]. It is a first attempt to translate Bennett's idea in a more physical context.

20200705: East Ridge of Beerwah

Paint spots now mark a route that traverses a little more westerly and then ascends to below the razorback. Descent really busy. Guys climbing Stainless Climb (the grade 27 free climb), trying to free the third (roof) pitch of Anticlimb.

20200629: UKBB twins

Code Description Count
1777 Part of a multiple birth 11846
2734 Number of live births 1.8 (mean)
O30 ICD10 Multiple Gestation 108
132212Date O30 first reported 641
O31 ICD10 Complications of Multiple Gestation21
132214Date O31 first reported 35
O84 ICD10 Multiple Delivery 22

20200629: Latour's Modes of Existence


...it would do no good to settle for saying that it is simply a matter of different "language games". Were we to do so, our generosity would actually be a cover for extreme stinginess, since it is to language, but still not to being, that we would be entrusting the task of accounting for diversity.

20200627: Bulimba Creek Bikeway

After crossing the Gateway bridge, rode to Cannon Hill Plaza. From there, regained Bulimba Creek Bikeway at Wynnum Rd, and went right through to where SE Freeway crosses Logan Rd just South of Garden City. Back along SE Freeway bikeway, where new section at Gaza Rd is now open. OK Coffee at Beartown Coffee House, Gabba.

20200618: BLUPF90 for GWAS

The BLUPF90 webpage is more up to date than the PDF manual, notably for SSGWAS analysis, where there is no mention of the snp_p_value option described by Aguilar et al [2019]. This paper confirms that EMMAX/GCTA type analyses are equivalent to GBLUP, and that a single-step SNP-BLUP analysis can incorporate ungenotyped but phenotyped individuals. The resulting estimates include individual prediction error variances for all SNP effect estimates in one calculation. To obtain this analysis, one uses renumf90 to prepare the data files, and then runs blupf90 and postGSf90 in sequence.

Simple univariate example using Sib-pair:

read bin dataset.bin.gz
unique_id sequential
set print 01110
set mis "0"
write ssgwas.ped
undrop trait
set print 01000
set ple -2
file delete ssgwas.dat
out ssgwas.dat
print where trait ^= x
set loc typed aff
if (protyp > 0.9) then typed = y
write blupf90 ssgwas.geno typed
write map blupf90 ssgwas.map
# Call renumf90
file delete tt_renum.job
out tt_renum.job
echo ssgwas.dat
echo 2
echo 1
echo WEIGHT(S)
echo 0.5
echo 1  cross alpha
echo animal
echo FILE
echo ssgwas.ped
echo FILE_POS 
echo 1 2 3 0 0
echo ssgwas.geno
echo  0.50
echo OPTION map_file ssgwas.map
echo OPTION snp_p_value
$ renumf90 tt_renum.job
$ preGSf90 renf90.par
$ blupf90 renf90.par
$ postGSf90 renf90.par

The postGSf90 output file chrsnp_pval contains columns {trait, effect, -log10_P, SNP, chromosome, position} [see here].

20200615: Function in genomics

Linquist et al [2020] summarize the nosology of "selected effect" functions and "causal role" functions (latter is any downstream effect). Also discuss Constructive Neutral Evolution [Stoltzfus 1999] where epistasis via effects such as buffering can allow a "neutral evolutionary ratchet", that is, buffering allows occult variation that then makes buffering compulsory ( "irremediable complexity"). Maynard Smith and Szathmary invoke this as the contingent irreversibility of major evolutionary transitions eg multicellularity. [via Larry Moran].

20200609: Stuck DVD tray

Belt, lubricant, locktab out of position are usual problems.


Unplug the unit for an hour
press and hold eject button while plugging in

invert the unit and quickly unplug when mechanical noise starts
try ejecting when returned upright or
while inverted eject and unplug just before starts moving

Latter was effective in this case.

20200608: Kant on beauty refuted!

Brielmann and Pelli [2017] find that "pleasure amplitude increases linearly with the feeling of beauty, To test Kant's claim of a need for thought, we reduce cognitive capacity by adding a "2-back" task. This added task greatly reduces the beauty and pleasure experienced from stimuli that otherwise produce strong pleasure, and spares that of less-pleasant stimuli. We also find that strong pleasure is always beautiful, whether produced reliably by beautiful stimuli, or just occasionally by sensuous stimuli. In sum, we confirm Kant's claim that only the pleasure associated with feeling beauty requires thought and disconfirm his claim that sensuous pleasures cannot be beautiful."

Luoto [2017] counter that "visually pleasing stimuli can cause affective and sexual responses even without conscious awareness [posing a] challenge to their argument (Ponseti and Bosinski, 2010; Gillath and Collins, 2016; cf. Chatterjee et al., 2009). The counterargument that thought is not a prerequisite for an affective response to visual stimuli is also supported by the finding that visual exposure to faces from out-group ethnical groups can elicit interracial affective bias outside conscious awareness (Yuan et al., 2017)...there are integrative processes that can occur outside of conscious awareness (Mudrik et al., 2014)... [A]n aesthetic judgment of beauty is firmly grounded in sensory processes (Jacobsen, 2006), and there are no empirical grounds to cleave visually mediated appreciation of beauty from sensory pleasures. Beauty can be amplified by cognitive processes (Vessel et al., 2012)¿such as integration with novel associations, integration over higher semantic levels, or integration over multiple modalities (Mudrik et al., 2014) - processes which can be particularly important for the experience of art (Nadal, 2013)."

20200604: Lust for Liberty?

Samuel K. Cohn Jr [2006] in Lust for Liberty: The Politics of Social Revolt in Medieval Europe, 1200-1425: Italy, France, and Flanders claims:

The first outright victory of this lower stratum in Italy appears in Bologna in 1289, when "the people without underpants" (servants, apprentices, "and others of a lower sort") booted Bologna's highest officer, the lord podesta, out of office and led him personally to the city gates. No retaliation followed, either from the city government, the forces of the podesta, or any faction of the ruling elites.
I think this looks like a misunderstanding of the local government of that time in Bologna. Carniello [2002]:

Bologna's guild-based government from 1282 to 1292 offers an important case for the study of Italian popular reform movements...Rolandino Passaggeri, renowned Bolognese master of notarial arts, authored the [anti-Magnate] Sacred Ordinances, launching the popular government, and reorganised the notaries' guild as part of the reform initiative... Sometime before 1243, when conflicts turned to violence in the streets of the town, magnate families had already coalesced into two factions, the Guelf (called the pars Ieremiensium after the Geremei) and the Ghibelline (called the pars Lambertaciorum after the Lambertazzi)...The popular movement's response to magnate violence was the establishment of its own political structure, the populus, which was completed in 1255 with the establishment of the captaincy of the Popolo. The Ghibelline magnates [were] expelled in 1274 after the civil war, when the Ferrarans came in support of the Geremei... The Geremei, far from enjoying the fruits of victory in Bologna, followed their enemies, took their own crippling blow in battle against the combined Ghibelline forces in 1275, and had to turn to Charles of Anjou in 1276 for military assistance...The Geremei were not occupied with ruling Bologna, much less governing the town and working to free the streets from violence...[After being returned by papal intervention in 1279, the violence also returned and the] Lambertazzi expulsion had to be repeated...

The new government, the Popolo, comprised ministers from the Guilds, popular arms societies (militias), and magnates, but only in a personal capacity. A list of magnate family members had to pay a good behaviour bond, and pay restitution for the civil war. Serfdom was abolished.

The episode in 1289 may or may not be the Tumult of the Fullers. The Podestà is an outsider who served as chief magistrate hired by the city, and rotated out after each six month period. Corso Donati was a leader of the Florentine Black Guelph faction (related to Dante and mentioned in the Divine Comedy), and was Podestà of Bologna in July-December 1288. For 1289, they were Antonio de Fixiraga and Zachanus de Zachanis.

20200526: Naturalization of normativity

I thought I would put together a few more references here:

First, from Richard Pettigrew's remarks about Bayesian updating

[Re the] Bayesian norm of updating [,] [s]ome pay attention to the pragmatic costs of updating any other way (Brown 1976; Lewis 1999); some pay attention to the epistemic costs, which are spelled out in terms of the accuracy of the credences that result from the updating plans (Greaves & Wallace 2006; Briggs & Pettigrew 2018); others show that updating as the Bayesian requires, and only updating in that way, preserves as much as possible about the prior credences while still respecting the new evidence (Diaconis & Zabell 1982; Dietrich, List, and Bradley 2016). And then there are the symmetry arguments...

I can immediately see Darwinian(-like) arguments along the lines of those for success semantics. Poor reasoners, that is those who fail to follow appropriate norms of reason, will be underrepresented in the population, and those who can recognize that there are such norms (a type of "reflective knowledge") should have a further advantage. https://www.frontiersin.org/articles/10.3389/fpsyg.2018.01291/full

20200526: Doggy reasoning

Rico the collie dog had a vocabulary of 200 words and would return an unfamiliar object from among a collection of familiar objects on command if asked for it by a novel word. I doubt he was up for polysemy.

20200520: Hintikka on the analytic-synthetic distinction

One of the greatest disasters that befell twentieth-century analytic philosophy was Quine's (1953) rejection of the distinction between analytic and synthetic truths as an "untenable dualism"...the terms "analytic" and "synthetic" are most unfortunate from a historical point of view. What is meant is in fact a distinction between conceptual and factual information. Quine is right in effect pointing out that one cannot tell from a person's behavior whether the information he is relying on is factual or conceptual...[but this] does not mean that one cannot define the distinction by some other means. Quine's way of thinking, and that of many other contemporary philosophers, is hence predicated on a distinction between logical and nonlogical constants.

The first observation that can be made here is that nonlogical analytical truths sometimes turn out to be logical ones when their structure is analyzed properly. In his Tractatus, Wittgenstein apparently assumed that this can be done for all conceptual truths. Within his truth-functional logic this amounted to the independence of atomic (elementary) propositions of each other. He ran into difficulties, however, in connection with color concepts, and was ultimately led to change his entire philosophy because of such difficulties. Wittgenstein despaired too soon, however, for at least in his paradigm case, the conceptual incompatibility of color terms can be turned into a logical truth simply by conceptualizing the concept of color as a function mapping points in a visual space into color space. (See Hintikka and Hintikka 1986, pp. 123-132) This is an instructive example of how nonlogical but "analytic" truths can be interpreted as logical ones. Then they are uninformative ("tautological") by the same token and in the same sense as logical truths.

He then introduces his Game-Theoretic Semantics (GTS) and independence-friendly (IF) alias hyperclassical logics.

The KK thesis in epistemic logic is that if an agent knows X, then they know that they know X. GTS can allow for imperfect memory and finite processing power ("liberated from classical linearity and perfect information").

20200516: Redcliffe Ride

Via Brighton, Thomas St Clontarf to Elizabeth Ave and Kipparing. Back via Petrie, Leitch's Crossing, Old Northern Rd (65 k). Notes at bora ring at Nudgee Waterhole mention Kippa Ring is a bora ring for initiation of boys (kippas).

20200420: Updated compiler timings

Compiler Code Time Comment
gfortran 9.2.1rs17293443.in16m31.097s-O2, openmp
sunf95 12.6 rs17293443.in17m23.626s-O3, openmp
Flang 7.0.1 rs17293443.in10m56.857s-O2

Compiler Code Time Comment
gfortran 9.2.1weil.in ae0m4.244s
sunf95 12.3 weil.in ae0m3.916s
Flang 7.0.1 weil.in ae0m12.290s
gfortran 9.2.1weil.in ce0m14.624s
sunf95 12.3 weil.in ce0m13.531s
Flang 7.0.1 weil.in ce0m39.146s

Flang and gfortran, it turned out, differ in treatment of signed zeros by sign().

20200415: The Murder Mystery

Song from The Velvet Underground [1969] - the first post-Cale album. It has simultaneous recitations from Reed and Morrison of different sets of lyrics (one twice as fast) over organ and guitars, interspersed with Maureen Tucker and Doug Yule singing a more lyrical chorus in lagged counterpoint. Reminds me of Jefferson Airplane/West Coast (raga) crossing Steve Reich (especially the last two minutes).

20200409: ebriety

Ebriety \E*bri"e*ty\, n.; pl. {Ebrieties}. [L. ebrietas, from.
ebrius intoxicated: cf. F. ['e]bri['e]te. Cf. {So?er}.]
Drunkenness; intoxication by spirituous liquors; inebriety.
"Ruinous ebriety." --Cowper.

20200407: Roman Jakobson (1896-1982)

Semiotician, phonologist (eg classified childhood aphasias) and linguist (Russia, Czechoskovakia, US) (and Formalist) whose six functions of language are:
aesthetic (materiality of means of communication)
conative (rhetorical, persuasive)
metalingual (reflexive)

These might be compared to the Peircian classification of relationships between sign and signified: Icon, Index, Symbol. Note that any given pairing may be multimodal (as above). Jakobson argued that the aesthetic is an additional type.

The Russian Formalists (eg cognitive estrangement) see poetry as disautomatizing and refreshing expressions, images and themes [Winner 1987], so Jakobson [1960] summarises the aesthetic function as "orientation toward the utterance", using "self-valuable" words, visual or acoustic materials. Prague Linguistic Circle and Prague School aesthetics extends this, the work of art as a structure, and art as a sign. Jakobson saw the aesthetic function as autonomous to the other functions - "introversive semiosis" with minimal or polysemous cultural references.

Influenced Levi-Strauss. Caton [1987] is a review of his contributions:

classification of phonemes (consonants and vowels) using 12 binary features of spectral analysis (compact, grave, strident...), which then give dimensions for analysis of morphemes and further up the hierarchy, phrases.

a teleological view of language, where reference is not the primary goal, so "the sound system...cannot be analysed without taking into account the purpose which that system serves..".

Contextual meanings (Grundbedeutung) "cannot be acknowledged without the existence of [relevant] invariant meanings (Gesamtbedeutung). These correspond to semantics and pragmatics, and the influence of behaviourism split these apart in US linguistics. In the example of "shifters", the pronoun "I" has an invariant meaning (the addressor of the speech act), and a contextual Indicial, "the utterer is existentially related to his utterance [so "I" indexes him]" - deixis essentially starts as a field of study following Jakobson's 1957 paper. Jakobson also worked on iconicity of grammar (see below).

Parallelism is a formal aesthetic process eg in poetry parallel the rhyme of words with parallels in semantic category eg "shame" and "blame" as moral judgments and as nouns. Similarly, play off. This extends to grammatical structure, eg order inversions. He published several very close readings of poems eg Shakespearian sonnets 127, 129. Wallot and Menninghaus experimentally test effects of this on semantic processing:

In general, multiparallelistic sentences and texts are linguistic analogues to multilayered structures of symmetry, repetition, and variation that are well established as core features of the aesthetic appeal of both music and visual objects.

[...] Roman Jakobson stipulated: "Ambiguity is an intrinsic, inalienable feature in any self-focused message".

Umberto Eco [1979] introduces ostention as a semiotic process - the signified of the ostended object is its class. Barry comments in the context of a semiotic analysis of one abstract painting that the significand might be intrinsically hard to comprehend (eg space and time).

Charles Morris is the logical empiricist (associate editor of the United Encyclopedia of Unified Science) who tried to present a semiotics of aesthetics esp literature, with the aesthetic sign as "an icon whose designatum is a value" [Steiner 1979]. He was not popular with the New Critics.

20200331: Louis and Bebe Barron

The Self-Destructing Modules Behind Revolutionary 1956 Soundtrack of Forbidden Planet

Rather than just tape effects/musique concrete, far more complex.

20200305: quadratic form in normal variables

These are needed to assess the significance of the SKAT statistic, but applications are wide.

Chen and Lumley [2019] compare algorithms, and find the saddle-point approximation of Kuonen [1999, 2001] well behaved. There is an implementation in the R survey package.

Gold standard methods are those of Davies, Imhof, an implementation by Farebrother), and Rice [1980].

20200209: SPEN spen family transciptional repressor

SPEN (gene spans ~100 kbp on 1p36) is the Xist-binding protein that is essential for X chromosome inactivation [reviewed by Trotman & Calabrese, 2020]. SPEN is suggested to bind active promotors and enhancers of target genes, complexed with Xist and then HDAC3 and NuRD. gnomAD records 4 (expect 129) pLoF variants, not increased in melanoma.

20200127: Wealth, Health and Child Development

Cesarini et al [2016]:

We use administrative data on Swedish lottery players to estimate the causal impact of substantial wealth shocks on players' own health and their children's health and developmental outcomes...Overall, our findings suggest that in affluent countries with extensive social safety nets, causal effects of wealth are not a major source of the wealth-mortality gradients, nor of the observed relationships between child developmental outcomes and household income.

Cited by Hill et al [2019] "Genome-wide analysis identifies molecular systems and 149 genetic loci associated with income".

20200125: Within and between family PRS comparisons

Selzam et al [2019]

There were three main findings: (1) DZ twin GPS differences predicted DZ differences in height, BMI, intelligence, educational achievement, and ADHD symptoms; (2) target and cross-trait analyses indicated that GPS prediction estimates for cognitive traits (intelligence and educational achievement) were on average 60% greater between families than within families, but this was not the case for non-cognitive traits; and (3) much of this within- and between-family difference for cognitive traits disappeared after controlling for family socio-economic status (SES), suggesting that SES is a major source of between-family prediction through rGE mechanisms

...[P]revious within-family analyses have revealed substantial reductions in individual SNP effect sizes. For example, there was an effect size attenuation of ~40% compared to between-family associations in the most recent GWA study on educational attainment. Most of this reduction has been attributed to prGE; no similar deflation of effect sizes was found for height, indicating that prGE is not likely at play. A novel method relying on closely and distantly related individuals, and that is applied to very large populations, detected a similar reduction of SNP-heritability estimates of educational achievement (~40%).

Regression predicting SES from polygenic risk scores [Table S12]
ADHD -0.168-0.214-0.13 1.32e-28
BMI -0.19 -0.216-0.1343.80e-38
EA 0.433 0.409 0.4792.35e-204
Height0.069 0.02 0.102 6.51e-06
IQ 0.231 0.213 0.291 4.28e-55
Neurot-0.06 -0.129-0.0435.35e-05
SCZ 0.04 -0.0150.066 0.008
Health0.302 0.238 0.318 3.86e-89

Kong et al [2018] similarly quantify the effects of "genetic nurture" in Iceland: the polygenic score computed for the nontransmitted alleles of 21,637 probands with at least one parent genotyped has an estimated effect on the educational attainment of the proband that is 29.9% (P=1.6×10-14) of that of the transmitted polygenic score. Specifically, transmitted alleles: θT=0.223 with R2=5%, and θNT=0.067 with R2=2.5%. Adjusting for parental educational attainment halves the effect of the nontransmitted allele EA PRS on offspring EA. There was no significant difference between paternal and maternal contributions.

Cheesman et al [2019] found in a comparison of 6311 adopted versus nonadopted children in UKBB that "polygenic scores are twice as predictive of years of education in non-adopted individuals compared to adoptees (R2= 0.074 vs 0.037, difference test p= 8.23 × 10¿24).

In passing, Cox et al [2019] in UKBB (N=7201) found that "[t]he association between (age- and sex- corrected) total brain volume and a latent factor of general intelligence is r=0.276, 95% CI 0.252-0.300... largest brain regional correlates of g were volumes of the insula, frontal, anterior/superior and medial temporal, posterior and paracingulate, lateral occipital cortices, thalamic volume, and the white matter microstructure of thalamic and association fibres, and of the forceps minor."

20200125: Human mutation rates

Tian, Browning and Browning [2019] give 1.3×10-8
mutations per base pair per meiosis with a 95% confidence interval of
1.0×10-8, 1.6×10-8.

20200123: Historical racism

My postings in the discussion of John Wilkin's thoughtful essay Is racism Christian? (qv) include one on Sweet [1997], who argues Christian attitudes to black Africans were transmitted from the Moors:

By the ninth century, Muslims were making distinctions between black and white slaves...The white mamnuk commanded a higher price than the black 'abd because he could bring a substantial Christian ransom or be exchanged for a Muslim captive. The differing treatment of white and black slaves reflected their relative worth. The mamnuk was viewed as an investment to protect,.. [w]herever there was back-breaking work to be done in the Arab world, black slaves were made to do it...[f]rom ninth-century Iraqi land reclamation projects to fourteenth-century Saharan salt and copper mines...White slaves were...usually household servants.

...In the eleventh century,Toledo historian Sd'id al-Andalusi wrote [pretty much as per Aristotle on natural slavery]:.

For those peoples...who live near and beyond the equinoctal line to the limit of the inhabited world in the south, the long presence of the sun at the zenith makes the air hot and the atmosphere thin. Because of this their temperaments become hot and their humors fiery, their color black and their hair woolly. Thus, they lack self-control and steadiness of mind and are overcome by fickleness, foolishness, and ignorance. Such are the blacks, who live at the extremity of the land of Ethiopia, the Nubians, the Zanj and the like.

From an initial differentiation on ransom price, Sweet argues Iberian Muslims moved rapidly to an Aristotelian climate justification as well as the religious "Race of Ham" argument about why slavery was acceptable, even for some black fellow Muslims:

Islamic interpretations of Noah's curse varied, but a tenth-century Persian historian, Tabari, presented a typically racial response "Ham begot all blacks and people with crinkly hair...Noah put a curse on Ham, according to which the hair of his descendants would not extend over their ears and they would be enslaved wherever they were encountered."

20200120: BAZ1B and human self-domestication

Zanella et al [2019] report that BAZB1 (alias Williams syndrome transcription factor WSTF) patterns the human face via effects on neural crest precursors, and that regulatory mutations in downstream target genes seem more common in modern humans than in Neanderthals and Denisovans.

Wilkins et al [2014] previously hypothesized that "all the facets of the domestication syndrome can be traced to mild neural crest cell deficits". Zanella et al point out that "[a]mong the genes downstream of BAZ1B... uncovered in this study, FOXP2, ROBO1, and ROBO2 have long been implicated in brain wiring processes critical for vocal learning in several species, including humans, and will warrant further mechanistic dissection in light of the distinctive linguistic profile of Williams-Beuren [7q11-23 deletion or duplication] syndrome individuals". WBS is characterised by facial reduction and retraction, pronounced friendliness, and reduced reactive aggression, and "[s]tructural variants in WBS genes, for example in the case of GTF2I and its paralogs, have been shown to underlie stereotypical hypersociability in domestic dogs and foxes".

20200106: de dicto and de re (Wikipedia)

De dicto: Necessarily, some x is such that it is A
De re: Some x is such that it is necessarily A
De se: Of oneself, but quasi-indexed (and see logophoricity)

20191128: Timings

Is a first generation i7 much different from a Xeon (Sky Lake)?

tinonee i7 920 @ 2.67 GHz (8 GB RAM)
hpcnode063Xeon Gold 6242 @ 2.8GHz (32 GB RAM)

Task tinonee hpcnode063
Sib-pair test suite 5 s oopsie! [1]
Compiling sib-pair 72 s 74 s
simulation with lamp [2] 64.8 s 40.7 s

[1] "sp-00001505.kin" Fortran runtime error: Reading more data than the record size (RECL)

[2] Sib-pair script powersim2.in

20191113: Cancer Genomics Cloud and the PanCanAtlas

An overview is at https://datascience.cancer.gov/data-commons/cloud-resources


The TCGA PanCanAtlas is hosted at https://isb-cgc.appspot.com/, but ISB-CGC offers access to all of TCGA, as well as several other datasets. The link to the PanCanAtlas in Huang et al [2018] still points to an uncompleted page. But see https://github.com/ding-lab/PanCanAtlasGermline/blob/master/README.md.

A glossy overview of PanCanAtlas publications:


20191113: The GDC Data Transfer Tool

gdc-client download -m ~/Downloads/gdc_manifest_20191113_012256.txt -t  ~/Downloads/gdc-user-token.2019-11-13T01_25_27.008Z.txt

20191106: iCite

NIH bibliometric tool: https://icite.od.nih.gov/analysis. The Relative Citation Ratio is cites/year v. NIH funded papers in same field and year. They also give an NIH percentile, relative to NIH funded papers.

20190903: Gianfranco Basti

From formal logic to formal ontology: The new dual paradigm in natural sciences .

The possibilist modal logic KD45 is an optimal candidate for a formal ontology of Quantum Field Theory, that gives a semantic information based on the Boltzmann-Schrodinger notion, where

the macroscopic "ordered state", of which information measurement corresponds to the variation of the density distribution of the so-called "Nambu-Goldstone bosons" (NGB) - "phonons" in the crystal state of matter, "magnons" in the ferromagnetic state, DWQ [Dipole Wave Quanta] in the living state of matter.

...Dipole Wave Quanta -- corresponding, at the mesoscopic level, to the long-range correlation waves observed in brain dynamics -- depends on the triggering action of the external stimulus for the symmetry breakdown of the quantum vacuum of the corresponding brain state.

The International Research Area on Foundations of the Sciences [https://irafs.org/ is based at the Pontifical Lateran University.

20190825: New History of (US) Capitalism

Gavin Wright comments:

In case any economic historian has been asleep or on Mars for the past three years, you may want to know that the economics-of-slavery culture wars have broken out again. Though only a pale shadow of the dust-up we had back in the 1970s, the aggressive assertions of the "new history of capitalism" regarding the centrality of slavery for U.S. economic development, and critiques of this work by economic historians, have generated much commotion in academic circles,

Beckert & Rockman Slavery's Capitalism: A New History of American Economic Development
Sven Beckert Empire of Cotton
Walter Johnson River of Dark Dreams
Edward Baptist The Half Has Never Been Told

20190822: Richard Tieszen

After Gödel. Platonism and Rationalism in Mathematics and Logic [2011] joins up phenomenology to mathematics.

van Atten et al similarly discusses the influence of Husserl and Kant on Brouwer (and Weyl). For intuitionists, the reals are "incomplete", as they are "indefinitely proceedable sequence[s] of 'nested' intervals" rather than atomic "finished" points. Sequences (including the above) can be lawful or lawless, but even the lawless represent a sequence of free choices. This is why intuitionist logic eschews the Principle of the Excluded Middle -

consider a lawless sequence α of which so far the intial segment 1,2,3, has been generated, and the statement P='The number 4 occurs in α'...we cannot say that P OR not P holds...[and] extensional identity of choice sequences (i.e. having the same values at the same places) is not generally decidable...[A] choice sequence α can be taken as an argument of a total function because in that case the function assignment must be contructable from just a suitable initial segment of α.

20190820: Success semantics

Blackburne [2004] reviews the pragmatist-derived idea (Frank Ramsey, Jamie Whyte [1990]) that practical success follows from correct representation:

the truth condition of a belief [is] that condition that guarantees the success of desires based on that belief.

Blackburne suggests "guarantee" might be too strong, and that "aid" is more correct. He argues that this approach is not obligatorily teleosemantic. One of Papineau's comments is that success or satisfaction of a goal is itself a "representational notion", so how does this help the ontology of representation? And what about pragmatic success based on a misunderstanding or a fiction?

20190815: MTAP expression

Kong et al [2018] find that in islet cells,

p15 abundance did not correlate with p16 and only marginally correlated with p14 and ANRIL. MTAP expression was marginally correlated with p14, p16, and ANRIL, but highly correlated with p15 expression. [r=0.68]

MTAP and p15 transcripts levels were not correlated with age, cf p16.

Recall Sangalli et al [2017] found the rs7023954*A allele to be more expressed than the G in fibroblasts, and that that haplotype was less methylated.

20190813: More webnovels/fanfic

Worm [Dave DeLaney "not grimdark, but crapsack"]
See https://tvtropes.org/pmwiki/pmwiki.php/Recap/Worm

Taylor Varga
Mauling Snarks

20190722: liftover


discusses the chain file format. The header for each chain gives the chromosome (fields 3 and 7), start of the current interval (field 6) and offset between maps (field 11 - field 6). The following lines of the chain give the end of the current interval (start + field 1), the new start, (start + field 1 + field 2), and the new offset (old_offset + field 3 - field 2).

For example:

chain 3231099988 chr22 50818468 + 16367188 50806138 chr22 51304566 + 16847850 51244566 23
19744   0       40
36      1       1


chr22 16367188 16386932 480662
chr22 16386932 16386968 480702
chr22 16386969 16387000 480702

I have implemented this in Sib-pair, and all seems to work.

20190716: Word2vec for scientific inference


Here we show that materials science knowledge present in the published literature can be efficiently encoded as information-dense word embeddings (vector representations of words) without human labelling or supervision. Without any explicit insertion of chemical knowledge, these embeddings capture complex materials science concepts such as the underlying structure of the periodic table and structure-property relationships in materials. Furthermore, we demonstrate that an unsupervised method can recommend materials for functional applications several years before their discovery.

...To train the embeddings, we collected and processed approximately 3.3 million scientific abstracts published between 1922 and 2018 in more than 1,000 journals deemed likely to contain materials-related research, resulting in a vocabulary of approximately 500,000 words. We then applied the skip-gram variation of Word2vec, which is trained to predict context words that appear in the proximity of the target word as a means to learn the 200-dimensional embedding of that target word, to our text corpus...

We find that, even though no chemical information or interpretation is added to the algorithm, the obtained word embeddings behave consistently with chemical intuition when they are combined using various vector operations (projection, addition, subtraction). For example, many words in our corpus represent chemical compositions of materials, and the five materials most similar to LiCoO2 (a well-known lithium-ion cathode compound) can be determined through a dot product (projection) of normalized word embeddings...[For example, ]

ferromagnetic - NiFe + IrMn = antiferromagnetic

...For instance, CsAgGa2Se4 has high likelihood of appearing next to "chalcogenide", "band gap", "optoelectronic" and "photovoltaic applications": many good thermoelectrics are chalcogenides, the existence of a bandgap is crucial for the majority of thermoelectrics, and there is a large overlap between optoelectronic, photovoltaic and thermoelectric materials...

20190705: Higher Criticism and the Berk-Jones statistic

Liu and Xie [2018] mention

Various methods have been developed to improve power for detecting sparse alternatives in this situation. The Tippett's minimum p-value test (Tippett, 1931), the higher criticism test (Donoho and Jin, 2004), and the Berk-Jones test (Berk and Jones, 1979) are particularly popular

The Higher Criticism test is based on the distribution of the P-values as order statistics under a normalised empirical process,

HC*n = max(i=1..pcrit*n) n0.5 [ i/n-p(i) ]/ (p(i)(1-p(i))0.5

The test is constructed for the alternative hypothesis of a two-component mixture, and critical threshold approximately

(2 log log n)0.5

due to "Law of the Iterated Logarithm" [Robbins 1970]. Robbins and Siegmund [1970,1973,1974] produced a similar sequential test, the Power-One Test, which in this setting is used for the exponential hypothesis. The R GHC package is one implementation of Generalized Higher Criticism for GWAS data.

Sun et al [2019] discuss set-based hypothesis testing of a number of tests, where the Berk-Jones statistic is Zmax (assuming independence of tests),

...the maximum of a set of likelihood ratio tests performed on S(t) [number of test statistics greater than threshold t] at all observed test statistic magnitudes greater than or equal to the median observed magnitude.

The Sun et al version incorporates a correlation matrix - in their GWAS setting, the marker LD - to give an analytic P, which they claim comparable to GSEA, SKAT etc.

Liu and Xie transform and sum P-values as Cauchy-transformed - Pillai and Meng [2016] found that the sum of perfectly dependent Cauchy variables follows the same distribution as if they were i.i.d., and under more complex dependence patterns, the (heavy) tails are not greatly affected.

T = S wi tan { 1-0.5piπ } with the w's summing to unity.

20190705: When does a physical system compute?

Horsman et al [2014] suggest:

...the use of a physical system to predict the outcome of an abstract evolution.

as opposed to an experiment, where evolution of a physical system tests the adequacy of the abstract model.

Scott Aaronson suggests it is whenever you can get a wrong answer (contra pancomputationalism etc).

Millhouse [2019] considers the collection of functions mapping physical states to abstract machine states and those mapping inputs to the abstract machine to interventions in the physical system. The least (algorithmically) complex interpretation is best.

A physical system, P, implements a machine, M, to the extent that the simplest interpretation of P as M, argminI in IK(I), is simple relative to the complexity of M, K (M). I is simple relative to M to the extent that it minimizes K(I)/K(M).

20190626: Shuggie Otis

A guitarist and singer/songwriter playing blues, jazz etc from a young age with father - a famous composition is Strawberry Letter 23.

20190620: de Finetti's Lottery and Orgulity

It was often argued that "all rational probability judgments are countably additive, but de Finetti denies this" [ Nielsen et al 2019]

Consider a lottery consisting of countably many tickets, one for each integer. According to de Finetti, it should be open to a rational agent to consider such a lottery fair (de Finetti 1974). To do so, each ticket must be assigned equal probability. But if probabilities are countably additive, this is not possible.

I am still thinking about this. It seems to me that a nonstandard analysis is possible a la Edward Nelson, following his derivation of the ordinal and cardinal versions of the Borel-Cantelli theorem (Theorems 7.3 and 7.4 of Nelson [1987]). Ai...Av are the events where one of the first v tickets wins, where v is a nonstandard integer infinitely close to +infinity. The question is whether the probability of a win of any of the first v tickets converges or not. If a rational agent has accepted that this is a fair lottery, this should converge to 1, and the AN are independent. Then

Π(n=i..v) (1-Pr An) ≤ exp(-Σ(n=i..v) Pr An)

as 1-x ≤ exp(-x). From this, the sum Σ(n=i..v) Pr An will be infinitely close to zero, even though the product on the left approaches unity. It is now possible to assign infinitesimal probabilities to each ticket (Robinson's lemma is that if x1..xn are infinitesimals, then there is one nonstandard v where xn are infinitesimals for all nv. An infinitesimal x is smaller than 1/n for all n in standard N), and two infinitesimals are infinitely close to each other.

Pruss [2018] tries to argue that this type of infinitesimal probability gives rise to inconsistencies, eg comparing two infinite lotteries - one where there is one winning number to one where there are two winning numbers (his example is where winning is a coarsening - double and round). Then one would prefer the second, even though it is only "weakly greater". But being offered a finite number of tickets in multiple infinite lotteries still offers only an infinitesimal chance of winning. Some of these problems arise after conditioning on having already drawn a ticket, so that the "face value" is a standard natural number. Stipulating that the values can be negative fixes some problems. Further stipulating that it is irrational to care about infinitesimal differences also helps.

A related argument is that Bayesian rationality cannot be normative in the face of infinities and topological set concepts such as meagreness, as they can be epistemically immodest approaching the limit of a series on conditionalizations (Bayesian Orgulity).

20190522: Two Melbourne bands


The Omnific is a progressive instrumental band from Melbourne, Australia featuring two bass players and a drummer.

Include tapping a la stick or Warr. Usually add synth backing/melismata, often ending up with a sound that reminds me of Metheny's Orchestrion.

Tim Willis and The End are a 5-8 piece jazz-rock group formed in 2010, who play with "post-rock and minimalism".

The End are Jack Beeche on alto sax; Jon Crompton (Wangaratta Jazz Awards Runner Pp 2009) on alto-sax; John Felstead (University Medalist, Lee Barker, Cam McCalister) on tenor sax; Tim Willis (composer) on lead guitar; bassist Gareth Hill (Michelangelo & The Tin Star, Bob Sedegreen); and Nick Martyn (Whitesploitation) on drums.

20190521: John Lilburne (1615-57)

The Leveller, "Free-born John", who was so disputatious that

If the World was emptied of all but John Lilburne, Lilburne would quarrel with John, and John with Lilburne.

He was imprisoned on multiple occasions, escaping execution for returning from exile by "successfully" arguing "that the prosecution was unable to prove he was the same John Lilburne who had been banished". At an earlier treason trial

...before the court of Star Chamber, he refused to take the oath. "It is this trial that has been cited by constitutional jurists and scholars in the United States of America as being the historical foundation of the Fifth Amendment to the United States Constitution. It is also cited within the 1966 majority opinion of Miranda v Arizona by the U.S. Supreme Court."

[He]...refused to take an Oath to answer Interrogatories, saying it was the Oath ex Officio, and that no free-born English man ought to take it, not being bound by the Law to accuse himself, (whence ever after he was called Free-born John)...

20190521: Another web fanfic novel

Like HP and the Methods of Rationality:

https://tvtropes.org/pmwiki/pmwiki.php/Fanfic/Luminosity http://luminous.elcenia.com/

I enjoyed this - only a few dud notes. However, I have not read the originals.

20190519: SSSS

Athena Andreadis recommends Minna Sundberg's Stand Still, Stay Silent webcomic


Really good.

20190502: GeCKO

Lau et al [2019]

...performed genome-scale CRISPR knockout (GeCKO) screen. We mutagenised the HAP1 cells with the GeCKO v2 library, which targets 19,050 human genes with 123,411 unique guide sgRNA sequences [ Sanjana et al 2014], and then selected these knockout pools with a lethal concentration of [box jellyfish] venom for 14 days.

Previously, we used a Genome-scale CRISPR Knock-Out (GeCKO) library to identify loss-of-function mutations conferring vemurafenib resistance in a melanoma model [ Shalem et al 2014].

20190501: Michael Waltzer

Politicial philosopher from communitarian and socialist traditions. Key book is Spheres of justice: A defense of pluralism and equality (1983).

From Joshua Cohen's review:

The central thesis of [his] theory of value is a version of communitarianism:

(C) The subjects [bearers] of values are in the first instance political communities, and not the individual members of those communities.

(C1) The objects that are socially valued are different for different political communities.

(C2) Communities typically have pluralistic values. That is, they value a variety of social goods - for example, money, political power, education, free time, love - which are unordered, in that there is no ranking of their relative value.

The second main element of TCE fits this theory of value into an account of the justification of distributive norms. (N) Each of the heterogeneous goods in a society is associated with a correct distributive norm, and that distributive norm is contained in the socially shared understanding of that good.

[Unordered plurality entails] a set of distinct "spheres of justice," each with its own internal regulative principle. What justice then requires is the "autonomy" of these spheres.

Trappenberg [2014] summarises:

[J]ustice takes a different shape in different societal spheres. In the sphere of education justice has to do with creating equal opportunities (in primary education) and with rewarding according to merit (in secondary education). In the sphere of money and commodities, justice takes the shape of free exchange. In the sphere of welfare, goods are distributed according to `socially recognized needs'. In the sphere of politics justice is about procedures: democratic elections, the will of the majority, gaining the public's favour and so on. Each sphere of justice has its own `internal moral logic'. A political community should be ordered in such a way that its spheres of justice can uphold their internal moral logic.

If a political community manages to keep its spheres of justice apart, such a community accomplishes an ideal called `complex equality'. Simple egalitarians are egalitarians who abhor income inequalities as such. Complex egalitarians on the other hand can put up with quite a lot of inequality in the sphere of money and commodities as long as this inequality is confined within that particular sphere. Rich people should not be able to buy political power, love and friendship, better education for their children, or preferential treatment by doctors, judges and policemen. But if these conditions are fulfilled there is nothing wrong with richness per se.

Walzer's friendly critics, the mitigated pluralists, seem to share a certain dislike for...the principle of spherical autonomy. They try to improve Spheres of Justice by seeking or constructing `across spheres criteria',`overarching principles' `underlying notions that go beyond local autonomy' and `non sphere-specific considerations

20190413: Kenneth Boulding

Economist (1910-1993) with interests in General System Theory (9 levels of organisation), ecological and evolutionary economics, peace studies, ethics of economics and economics as a moral science. A student of Keynes.

"Keynes saw himself writing in the Cambridge tradition of economics as a 'moral science'"

20190411: Camille Paglia in Salon 1998

What I hate about Foucault ;)

20190411: On a post from John Danaher

Here, John Danaher analyses a paper by Neil Levy, on how consciousness is important to moral value. I wrote:

Yes, as an attempt at a reductio wrt phenomenal consciousness and
Siewert's wager.

P1a. One must be sentient to have some kind of moral standing.
P1b. Minimally, one must be able to suffer.
P2. Suffering requires phenomenal consciousness (PC).

C1. PC is a key feature of personhood.

P3. Vivid access consciousness (memory, imagination) is
experientially equivalent to (indistinguishable from) PC for
some humans, by all accounts.
P4. For other humans, similar mental function is present
but as a less vivid access consciousness.
P5. Focal loss of the capacity for such experiences due to neurological
diseases does occur, without effects on other facets of consciousness.
(Affected individuals notice the effect, and find it annoying and
a loss of function).

C2: Access and phenomenal consciousness can be
functionally interchangeable in particular domains between
persons. I think this is stronger than for perceptual faculties
eg cortical blindness (v. blindsight).

C3: By a stepwise replacement model, it seems plausible to me that we can now
posit being a person with no phenomenal experiences. Say, for example,
this person accesses visual information about the world only via working
or short term memory. Hey, isn't that all of us absent-minded people
;)? Maybe PC is incoherent? At the very least, C1 seems contestable.

Consider suffering due to acute tissue injury, psychogenic pain, and
anticipated pain, and types of consciousness in each case.

20190331: Truthmaker Semantics

Fine [ 2014] proposes a new realist semantics crossing contructivist and Kripkean (1965) "condition-oriented" semantics. In Brouwer-Heyting-Kolmogorov semantics, "a construction establishes B .and. C if it is the combination of a construction that establishes B and a construction that establishes C", while for Kripke, "a state verifies [or forces] B .and. C if it verifies B and verifies C". For the constructivist, the semantics is exact, but in the latter case, one can add (extend) irrelevant content to the state without changing the verification: "the state of the ball being red and of its raining in Timbuktu will also verify that the ball is red" - in the case of possible worlds verifying X, then there are relevant and irrelevant facts, though tricky regarding negation.

In intuitionistic logic, the latter is less of a problem. Fine defines a conditional connection between states (of nature) s -> t, that is s "leads to" t, where s is an exact verifier of, say, B. Now logically equivalent statements may not be verified by the same states. For example, the logically equivalent statements p and p .or. (p .and. q). A state might exactly verify just (p .and. q) but not p. Truth of a statement requires exact verification by an actual state ie a fact.

Ciaredelli derived a very similar generalized or inquisitive semantics [2009,2011,2013], under the model that propositions are proposals, which are defined as the set of all maximal states (roughly independent states) that support the proposition. A proposition can be contradictory, thus inviting an informative response to which arm is correct.

20190329: Collaborative Cross

Eight founder strains: A/J (A), C57BL/6J (B), 129S1/SvImJ (C),NOD/ShiLtJ (D), NZO/H1LtJ (E), CAST/EiJ (F), PWK/PhJ (G),and WSB/EiJ (H). The 3 wild-type founders are F-H. There are ~70 RI strains currently available, as interstrain allelic incompatibility has limited the number of viable lines.

The Mouse Universal Genotyping Array has gone through three generations: 7851 markers, MegaMUGA 77808 markers, and now the GigaMUGA (Infinium HD) with 143259 markers.

Broman et al [2019] describe the updated R/qtl2 package, which can handle "multiparent populations derived from more than two founder strains, such as the Collaborative Cross and Diversity Outbred mice". This requires SNP genotypes for individuals and founders, coded as 1 and 3 for the different homozygotes (heterozygotes=2 are ignored) in the CSV genotype file

Power calculations for Haley-Knott regression of CC strain means are presented by Keele et al [2019]. They note the QTL support interval is roughly 10 Mbp, after allowing for LD. For r replicates,

h2QTL(r=1) = h 2QTL / (h2QTL + h2strain + s2/r)

For example, a mapping experiment on strain means with QTL effect size h2QTL=0.3, h2strain=0.4, s2=0.3, and r=10, is equivalent to our simulation of a single-observation with no strain effect but QTL effect size h2QTL=0.41.

QTL h2Power (50 strains)
1 Obs3 rep*5 rep*2 alleles3 alleles8 alleles
0.3 0.1250.0790.1050.1180.116
0.4 0.1820.1180.2980.3350.383
0.5 0.2500.1670.6200.6300.712
* Assuming no strain effect

Note that there are "inconsistencies on CC haplotype probability file in build 38 compared to build 37. The two errors are as follows: On chromosome 5, there are two markers (SAbGeoEUCOMM001 SAbGeoEUCOMM002) that should not be given genome position. On chromosome 13, there is a problem with the last set of approximately 80 markers that result in an inconsistent pattern of founder haplotype reconstruction. (Beginning near UNC23486670 CH13:118447298 to UNC23498758 CH13:119480991)" [http://csbio.unc.edu/CCstatus].

20190329: Hair greying in the mouse

Example of phenotype Fialkowski et al [2019] use Tg(DctSox10) transgenic mouse line to find hair greying loci - rs6160140 (mmu7), rs6411274 (mmu13), rs13483929 (X). The candidates in these intervals were Oca2, Trpm1 (DBA/1J has known missense mutations), Bloc1s5, Dtnbp1, Kif13a, and Atp7a.

20190329: Terrence Deacon

Anthropologist-cum-neuroscientist-cum-biosemiotician currently at UC Berkeley - key books are The Symbolic Species: The Coevolution of Language and the Brain and Incomplete Nature: The Emergence of Mind from Matter, the latter summarised in detail in its Wikipedia article.

In Incomplete Nature, he presents:

...three modes of system dynamics that are distinguished by their hierarchic (i.e., nested) dependencies and their reversals of spontaneous global dynamical tendencies to reach different kinds of stable end-states (or attractors), if they are provided with the required time to do so. These dynamical modes include homeodynamics (e.g., processes at or near thermodynamic equilibrium), morphodynamics (e.g., non-chaotic dissipating processes such as exemplified by self-organizing systems), and teleodynamics (e.g., self-preserving processes such as exemplified by living systems).
[Deacon and Koutroufinis 2014]

On reference, thinks it can be formalized in a physical model,

...reference is made possible by the susceptibility of a given information medium to reflect the effect of work with respect to an extrinsic context, and that the sign of this effect - i.e. whether there is an increase or decrease of medium constraint - will depend on whether this work originates in the interpretive process or in its extrinsic physical context.

20190315: Syncytins

The HERVs are 8% of the human genome, and have contributed two important coding genes. Syncytin-1 (ERVWE1 7q21) induces syncytia, and is highly expressed in the syncytiotrophoblast. It may modulate the immune response in pregnancy. Syncytin-2 (6q24) expression is limited to the villous cytotrophoblasts, and its receptor MFSD2 is in the syncytiotrophoblasts.

20190315: De novo gene evolution

Numerous examples of de novo emergence of genes from previously noncoding sequences are now recognised - excluding horizontal gene transfer and poor annotation. "the RNA-first model, in which the formation of an ORF occurs in a region that is already transcriptionally active, is supported by previous reports on de novo genes and by both pervasive transcription and pervasive translation" [Vakirlis et al 2018]. Carvunis et al [2012] considered a "continuum" model that there is a large pool of protogenes being transcribed at low levels, some of which are then selected.

Vakirlis et al [2018] screened yeast data to find 366 taxonomically restricted genes that represented likely true de novo events - 0.5-1% of loci. A point mutation giving rise to an ORF could be identified in 30 cases. In one example in the cod, the promotor region for an antifreeze glycoprotein became associated following a translocation of the microsatellite (9 bp tripeptide repeat). Understandably, de novo genes tend to be shorter, and "biosynthetic cost is also lower than those of noncoding sequences, in agreement with an intermediate stage from a noncoding to a coding state". They tend to be reverse orientation to 5' neighbours and arising in GC rich areas, given that AT-rich stop codons are less common.

Chen et al [2015] describe 64 hominoid-specific "motherless" or "orphan" protein coding genes that arose from GC-rich lncRNA genes. Of these, 43 encode human-specific proteins, and 21 encoding similar proteins in human and chimpanzee but not in rhesus macaque. Protein expression was confirmed by mass spectrometry. In the macaque, the orthologous loci for the lncRNA precursors of human de novo genes are not subjected to strong selective constraints, ie "the ancestor of de novo genes may not be particularly distinct in terms of functional importance before the proteins arise". The human de novo genes do show NS/S ratios consistent with selection. Several studies have shown that the protein structure tends to be "disordered", but can still be functional.

Ruiz-Orera et al [2015] report "over five thousand new multiexonic transcriptional events in human and/or chimpanzee that are not observed in" macaque or mouse. Most did not evidence purifying selection. In human large-scale RNA-seq, "unannotated genes represented 0.5-2% of the transcriptional cost depending on the tissue...The vast majority of de novo transcripts were expressed in testis (93.8-94.5%), as were transcripts from phylogenetically conserved genes. In contrast, in brain, liver and heart, transcripts from de novo genes were underrepresented when compared to transcripts from conserved genes." Further, the de novo genes were AT rich compared to older genes. Only 6 were found to produce protein.

20190308: Dopamine and musical pleasure

Via http://mindblog.dericbownds.net/ Ferreri et al [2019] show levodopa increases musical pleasure, and risperidone diminishes it. And earlier, Blood and Zatorre showed musical "chills" to be associated "with increased blood flow in the ventral striatum, the amygdala, and other brain regions associated with emotion and reward".

20190306: Chromatin marks

Stitzel et al [2015]:

Enhancer sites H3K4me1 enriched. Active H3K27ac
Stretch enhancers stretches of enhancers > 3 kb cell-type-specific identity
Super enhancers long stretches of H3K27ac active enhancer modifications mark LCRs
Broad domains contiguous H3K4me3 promoter marks >4 kb cell-type-specific identity/function eg G6PC2

Roadmap Epigenomics Consortium [2015]:

The active states (associated with expressed genes) consist of active TSS-proximal promoter states (TssA, TssAFlnk), a transcribed state at the 5' and 3' end of genes showing both promoter and enhancer signatures (TxFlnk), actively-transcribed states (Tx, TxWk), enhancer states (Enh, EnhG), and a state associated with zinc finger protein genes (ZNF/Rpts). The inactive states consist of constitutive heterochromatin (Het), bivalent regulatory states (TssBiv, BivFlnk, EnhBiv), repressed Polycomb states (ReprPC, ReprPCWk), and a quiescent state (Quies) which covers on average 68% of each reference epigenome. Enhancer and promoter states cover approximately 5% of each reference epigenome on average, and show enrichment for evolutionarily-conserved non-coding regions41.


Active TSS 0.7%
Flanking active TSS 0.5%
3' 5' transcription 0.1%
Strong transcription 3.6%
Weak transcription 11.6%
Genic enhancers 0.4%
Enhancers 2.8%
ZNF genes and repeats 0.2%
Heterochromatin 2.6%
Bivalent/poised TSS 0.1%
Flanking Bivalent/poised TSS 0.1%
Bivalent Enhancer 0.1%
Repressed Polycomb 1.2%
Weak Repressed Polycomb 8.3%
Quiescent/Low 67.8%

Polycomb repressive complex-2 (PRC2, subunits EZH2, SUZ12, EED, RBBP4, AEBP2) is a histone methyltransferase required for epigenetic silencing - adds three methyl groups to lysine 27 of histone H3.

Finucane et al [2015] partitioning of heritability of 11 traits by functional class:

Berthelot et al [2015]

...breakpoints statistically associate with features including regions of high gene density, high GC content, and high repeat content... Here, we show that the distribution of rearrangements can be accurately explained as misrepaired breaks between open chromatin regions in non-coding regions that are brought into contact by the three-dimensional conformation of chromosomes in the nucleus, which also provides a direct explanation for their mechanism of occurrence.

For rice, 72 Mb DNAse sites, 389 Mb genome.

Interval,Rice genome

20190304: Grid-LMM

https://github.com/deruncie/GridLMM , as the name implies, tests for marker association (and GxE) against a grid of random effects values. Authors argue performant for case of multiple REs.

20190227: Systemic atopic dermatitis therapies

From Alexander et al [2019]:

Existing systemic therapies for AD: cyclosporine, azathioprine (higher side effects), methotrexate and mycophenolate mofetil ("blocks de novo guanine synthesis via the inhibition of inosine monophosphate dehydrogenase leading to impaired leucocyte proliferation").

Novel systemic therapies (~25 trials to date)

Mepolizumab monoclonal antibody to interleukin-4 (IL-4).
Omalizumab anti IgE antibody.
Dupilumab IL-4 receptor-alpha antagonist (IL-4/IL-13)
Lebrikizumabanti IL-13
tralokinumabanti IL-13
Ustekinumab p40 subunit of IL-12 and IL-23 (Th1 and Th17)
Nemolizumab anti IL-31 receptor A.
Fezakinumab IL-22 antagonist.
Baricitinib (JAK1 and JAK2 inhibitor.

NICE has recommended dupilumab as an option for treating moderate-to-severe AD in adults where the disease has not responded to at least one conventional systemic therapy or where these are contraindicated or not tolerated.

20190226: Shoe repair adhesive

Sikaflex marine & polyurethane based construction adhesive
310 ml for $24.26 at Bunnings.

Gear Aid Freesole (aka Aquaseal) urethane formula: K2 $17.95 for 28 g.

Original Gorilla, Barge Cement

20190221: UCSC v. Ensembl marker map database

Some microsatellites are still "not mapped to the assembly in the current Ensembl database", but are on UCSC B37.

20190218: NEDD4 and keloid formation

Japanese studies (Nakashima et al 2010, Ogawa et al 2014) pointed to rs8032158 on chromosome 15 as being associated with keloid scarring. More recently, the risk genotype C/C been found to increase expression of the transcript variant (TV) 3 of NEDD4 in keloid tissue. A quick look at UKBB finds the C allele associated with "M72 fibroblastic disorders" (P=10-14) as well as decreased sitting height. The ICD10 classification M72 covers fasciitis (eosinophilic, necrotizing, pseudosarcomatous), fibromatosis, and Duyputrens.

Marneros [2019] comments:

Higher transcript levels of NEDD4-TV3 are associated with increased activation of NF-kB and STAT3 in keratinocytes and fibroblasts of keloid lesions when compared with normal skin in individuals without this genetic risk allele. Overexpression of NEDD4-TV3 in primary human keratinocytes increased NF-kB activity, whereas its knockdown had the opposite effect.

20190216: Thelma Plum


Indigenous Australian singer/songwriter from Delungra NSW...won Triple J's National Indigenous Music Awards competition in 2012.

Clair de Lune and Candle good.

20190211: ROC AUC

Bamber [1975] showed that AUC = Pr(Y ≥ X), that is the probability that the value for a randomly sampled case ≥ that of a control. It pretty directly follows that the Mann-Whitney U is a nonparametric estimator of AUC, allowing asymptotic intervals that tend to be less accurate when AUC is high and N small.

20190211: Another variant consequences predictor

Huang et al [Nat Gen 2017] published their LINSIGHT scores for all sites, which they claim improves predictions for noncoding sequence via inference of negative selection (similarly to fitCons).

20190211: MELPREDICT

Probability of being a CDKN2A carrier [Taylor et al 2019]:


That is, sporadic case in 60 y.o., Pr=0.003; affected relative, two primaries, 20 y.o., Pr=0.11.

20190207: Rates of cuckoldry

Larmuseau et al [2013]

[F]irst large-scale, unbiased genetic study of historical EPP rates in a Western European human population based on combining Y-chromosomal data to infer genetic patrilineages with genealogical and surname data, which reflect known historical presumed paternity. Using two independent methods, we estimate that over the last few centuries, EPP rates in Flanders (Belgium) were only around 1-2% per generation.

Larmuseau et al [2015]

Based on the analysis of 68 representative genealogical pairs, separated by a total of 1013 fertilization events, we estimated that the historical EPP rate for the Dutch population over the last 400 years was 0.96% per generation (95% confidence interval 0.46%-1.76%).

Greeff and Erasmus [2015]

Here we...investigate 1273 conceptions over a period of 330 years in 23 families of the Afrikaner population in South Africa. We use haplotype frequency and diversity and coalescent simulations to show that the male population did not undergo a severe bottleneck and that paternity exclusion rates are high for this population. The rate of cuckoldry in this Western population was 0.9% (95% confidence interval 0.4-1.5%)...

20190129: Genetic correlations for solid cancers

Jiang et al [2019] use LDSC


20190114: Synergy Indices

The epiR package contains the useful epi.interaction() function, which calculates relative excess risk due to interaction, proportion attributable to interaction, and the synergy index (RRAB-1)/(RRAb-1)+(RRaB-1)).

20190109: Niklas Luhmann

Niklas Luhmann (1927-1998) was a German social scientist working on the "theory of social systems". This centred on decision communications as autopoiesis in organisations, specifically as uncertainty absorption.

Quotes from Luhmann taken from the English language review by Schoeneborn [2011]

Accordingly, social systems are not comprised of persons and actions but of communications...Social systems use communications as their particular mode of autopoietic reproduction. Their elements are communications which are recursively produced and reproduced by a network of communications, and which cannot exist outside the network...

Organised social systems can be understood as systems made up of decisions [...]. Decision is not understood as a psychological mechanism, but as a matter of communication, not as a psychological event in the form of an internally conscious definition of the self, but as a social event. That makes it impossible to state that decisions already taken still have to be communicated. Decisions are communications; something that clearly does not preclude that one can communicate about decisions.

Was not translated into English for many years. David Spivak in his work "Toward a mathematical foundation for autopoiesis" cites Luhmann. In Fong et al [2018], they apply topos theory/categorical logic to mereology of (complex) systems.

20181224: Exchangeability

A few half-baked thoughts that appeared on Electric Agora.

Exchangeability is a concept that is closely bound up with ideas of individuality and identity. That is to say that one's head starts hurting once one examines it too closely. In probability and statistics, it is a crucial formalized idea that underlies whole areas like Bayesianism and subjective probability. In that mathematical context, exchangeable events or variables can be permuted without altering the properties that you are interested of the that system. A classic example is the number of heads out a series of coin tosses. You are not interested in what order the successes appear, just the number, so any pair of events could have been swapped around in the sequence without altering the relevant outcome.

In the statistical mechanics of quantum physics (QFT), the exchangeability of fundamental particles is more than just a matter of the interests of the observer - they are an essential property of matter. As you might guess, exchangeability is a type of symmetry relation in physics and mathematics. Failure to correctly model exchangeability in (semi-classical) thermodynamics leads to Gibbs Paradox.

The concept of a sortal in philosophy has a similar flavour, in that there is numerosity, with the things being counted identical to each other in respect of the properties used to define the sortal eg "three brown dogs".

So, can two entities be exchangeable without being identical? That is, is it a weaker concept than being perfectly indiscernably alike? I have recently been enjoying Rodin's Axiomatic Method and Category Theory, where in Chapter 5 he runs over some of the complications of defining and using the identity relation. He comments:

We see that Plato, Frege and Geach propose three different views of identity in mathematics. Plato notes that the sense of the "same" as applied to mathematical objects and to the ideas is different: properly speaking, sameness (identity) applies only to ideas while in mathematics sameness means equality or some other equivalence relation. Although Plato certainly recognizes essential links between mathematical objects and Ideas (recall the ideal numbers) he keeps the two domains apart. Unlike Plato Frege supposes that identity is a purely logical and domain- independent notion, which mathematicians must rely upon in order to talk about the sameness or difference of mathematical objects, or any other kind at all. Geach's [Theory of Relative Identity] has the opposite aim: to provide a logical justification for the way of thinking about the (relativized) notions of sameness and difference which he takes to be usual in mathematical contexts and then extend it to contexts outside mathematics..."Any equivalence relation ... can be used to specify a criterion of relative identity."

Rodin then introduces what he variously describes as a constructivist (in that one specifies a procedure) or substantialist interpretation of the identity x = y: that there is an invertible transformation from x to y and y to x. His discussion now segues into categorical logic and category theory (which some folks might be impressed to see is Hegelian), but I will wander off in a simpler non-mathematical way back to exchangeability of x and y.

I am thinking of x and y each sitting within their relationships with everything around them, and we perform symmetrical tranformations of x to y and y to x. Let's say these transformations are simple translations (movements) in space or time. So I might swap one coin for the other coin in two separate sequences of coin throws. In the traditional statistical setup for subjective probability, I might swap coin tosses from different times in a single sequence of throws. And if I were Black (1952) I might have swapped the two identical spheres that are all a particular universe contains.

So at the practical level, no-one involved may notice there has been an exchange, or there were obvious physical changes in whichever object you were paying attention to, or subtle changes in behaviour detectable by prolonged observation eg the statistical properties of the sequence of coin tosses in a change point model. I might consider how much work is required to detect my substitution in an informational sense. In the Leibnitzian idea of identity of indiscernibles, there is an implication that there is an absolute true state of nature (I won't go off into quantum mechanics again) or the observer doing the discernment can spend an infinite amount of time and energy getting to the bottom of things while taking no time at all (like one of those hypercomputers).

Is there a point to this? I see it as a nice way to think about various hoary thought experiments regarding identity:

So is exchangeability just another term for equivalence in these examples? I find it useful because of its constructivist quality - that is it specifies what action is to be taken to demonstrate that two entities are exchangeable. Time travel might be a little hard to realise - in the statistical settings the permutation is epistemic (and counterfactual) rather than ontological. That the testing of equivalence is via checking both sides of the swap is an interesting feature that again speaks to a pragmatic definition rather than just a purely a priori attempt to specify identity or equality, for certain values of pragmatic. It seems to lend itself to interpretations of these various thought experiments that strike me as common-sensical. Where it can't actually answer the question that two entities are actually exchangeable, it can suggest forms of test one might use in the future.

I should ask whether identity has somehow been smuggled in as the identity-preserving transformation. In the comments, we discussed the lump of clay <-> statue. It seems to me that the necessary transformations are quite information heavy in only one direction, that is clay->statue contains all you need to convert any lump of clay to the specified statue, while statue->clay is also suitably general but lots easier to implement. Rodin presents the argument that it is morphisms that are the "real" objects.

20181211: Neolithic genetics

Gonzalez-Fortes et al [2017] report results for six Eneolithic and Mesolithic samples (4 Romanian 5.4-8.8 kya, 2 Spanish)

Gomes et al [2017 describe one sample from the Chalcolithic
SC2_MesoRom 8800 bpT/T
AsturiasSpain 3480 bpC/C

The rs12203592*T variant is especially common in modern Irish, and is associated with dark hair, light irides and skin, and increased freckling. T carriers are less likely to be wild type for MC1R, at least in the Australian population.

20181209: Pirinen et al (2012)

Matti Pirinen's software

Pirinen M, Donnelly P and Spencer CCA (2012):
Including known covariates can reduce power to detect genetic effects in case-control studies.
Nat Genet 44: 848-851.

20181208: O'Connor & Price Latent Causal Variable model

O'Connor and Price [2017, 2018] take advantage of the fact that "if trait 1 is partially genetically causal for trait 2, then most SNPs affecting trait 1 will have proportional effects on trait 2, but not vice versa...".

SNP effect ak = qkp + gk
E(a13a2) = κpq13 q2 + 3 rg

where κp is the kurtosis of p
gcp = log(abs(q2)-log(abs(q1))/ (log(abs(q2)+log(abs(q1)))

where gpc=1 when "trait 1 is fully genetically causal for trait 2: q1 = 1 and q2=rg"

They estimate rg via a LD score bivariate regression, and normalise the estimates of a via a separate LD score regression (jackknifing for standard errors) since:

[o]ur method exploits this excess kurtosis; when κp is zero (such as when p is normally distributed), we are unable to test for partial causality or to estimate gcp (indeed, the model is not identifiable when p is normally distributed...However, it turns out that the Gaussian case is the only non-identifiable case, assuming that (p, g1 , g2) are independent. The following proposition asserts that under an independence assumption, the model is identifiable if and only if does not follow a normal distribution. It does not matter what the marginal distributions of g1 and g2 are. This result echoes similar results in Independent Components Analysis, which separates independent, additive signals exploiting non-Gaussianity.

...[T]he LCV model includes only a single intermediary and can be confounded in the presence of multiple intermediaries, in particular when the intermediaries have differential polygenicity. Indeed, some trait pairs [in UKBB] appear to show evidence for multiple intermediaries. Nonetheless, causality or partially causality provide a more parsimonious explanation for estimated causal effects, especially when the gcp estimate is high.

20181204: Wang Wei

Poet (painter, musician, official), lived Tang Dynasty 699-759.

Wingceltis goldenrain shine empty bend
Fresh and green ripple ripples ripples
Secret enter Shang hill road
Woodcutter not able know

20181130: Meta-reinforcement learning

Stadie et al [2018]:

Meta RL goes by many different names: learning to learn, multi-task learning, lifelong learning, transfer learning. The goal, however, is usually the same - we wish to train the agents to learn transferable knowledge that helps it generalize to new situations...

This problem definition induces an interesting consequence: during meta-learning, we are no longer under the obligation to optimize for maximal reward during training. Instead, we can optimize for a sampling process that maximally informs the meta-learner how it should adapt to new environments. In the context of gradient based algorithms, this means that one principled approach for learning an optimal sampling strategy is to differentiate the meta RL agent's per-task sampling process with respect to the goal of maximizing the reward attained by the agent post-adaption.

Formalism of a reinforcement learner is a Markov Decision Process ({States}, {Actions}, p, γ), with p the dynamics, with p(r,s' | s,a) the probability of reward r, state s' following action a in state s, and γ the discounting of future rewards. The classic architecture of a reinforcement learner is as a LSTM (long short term memory) recurrent layer NN [Hochreiter and Schmidhuber 1997].

AlphaZero is a deep reinforcement learner. "While the results are impressive, they were achieved on one task at the time, each task requiring to train a new agent instance from scratch...Parallel multi-task learning has recently achieved remarkable success in enabling a single system to learn a large number of diverse tasks...We... automatically adapt the contribution of each task to the agent's updates, so that all tasks have a similar impact on the learning dynamics. This resulted in state of the art performance on learning to play all games in a set of 57 diverse Atari games. Excitingly, our method learned a single trained policy - with a single set of weights - that exceeds median human performance. To our knowledge, this was the first time a single agent surpassed human-level performance on this multi-task domain." [Hessel et al 2018]

Wang et al [2018] (Deep Mind) describe a meta-reinforcement-learning algorithm that mimics the pattern of learning of monkeys in the Harlow task, where two stimuli (objects) are presented, one of which hides a reward (in a hole under it). The pair of objects changes in blocks of six, but the rules remain the same. Monkeys learn the abstract rules of the game, so they quickly find the reward regardless of the actual objects. Wang's model is that dopamine represents the local NN reward prediction error (RPE) signal for synaptic learning in eg dorsal striatum thalamus. In the prefrontal cortex a separate network uses the same mechanism, with dopamine (RPE) representing "the expected values of actions, objects and states [by] dynamically [encoding] conversions from reward and choice history to object value, and from object value to object choice". This modulates the former RL process via the corticostriatal pathways - "learning to learn".

Crucially, the emergent algorithm is a full-fledged RL procedure: It copes with the exploration-exploitation tradeoff, maintains a representation of the value function, and progressively adjusts the action policy. In view of this point, and in recognition of some precursor research [Schmidhuber et al, 1996, Simple principles of metalearning], we refer to the overall effect as meta-reinforcement learning.

Added 20171207: New Science paper on more AlphaZero chess games versus Stockfish "A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play".

20181125: References on semiotics etc


And Parthmore [2013] on the relationship between Gardenfor's conceptual space and semiotics.

And Hermann-Pillath [2010] on entropy and naturalising semiotics.

20181125: Rorty and irony (not that kind)

Rorty [1989] Contingency, Irony and Solidarity mentions "humour" exactly three times, discussing Nietzsche and Heidegger (!?), and "jokes" (5 times) only when discussing Derrida (and Nabokov).

He takes his epigraph from Kundera's The Art of the Novel:

The agélastes [Rabelais's word for those who do not laugh], the non- thought of received ideas, and kitsch are one and the same, the three- headed enemy of the art born as the echo of God's laughter, the art that created the fascinating imaginative realm where no one owns the truth and everyone has the right to be understood.


...sketches a figure whom I call the "liberal ironist." I borrow my definition of "liberal" from Judith Shklar, who says that liberals are the people who think that cruelty is the worst thing we do. I use "ironist" to name the sort of person who faces up to the contingency of his or her own most central beliefs and desires - someone sufficiently historicist and nominalist to have abandoned the idea that those central beliefs and desires refer back to something beyond the reach of time and chance. Liberal ironists are people who include among these ungroundable desires their own hope that suffering will be diminished, that the humiliation of human beings by other human beings may cease.

[...] One of my aims in this book is to suggest the possibility of a liberal utopia: one in which ironism, in the relevant sense, is universal... Solidarity is not discovered by reflection but created. It is created by increasing our sensitivity to the particular details of the pain and humiliation of other, unfamiliar sorts of people.

20181107: Tracy-Widom Distribution

From Lototsky's notes:

Longest increasing subsequences: Let Z = (Z1..Zn) be i.i.d. U(0,1) random variables and let Sn be the length of the longest increasing subsequence of Z . Then

(Sn-2n1/2) / (n1/6) -> TW (as n -> inf)

where TW is the Tracy-Widom distribution. [...]

Largest principal component: The largest principal component of the covariance matrix XT X, where X is an n × p data matrix all of whose entries are independent standard normal has TW as the limiting distribution after proper centering and scaling.

Tracy CA and Widom H. (1994). Level spacing distributions and the bessel kernel. Commun Math Phys. 161 :289--309.

Tracy, C. A. and Widom, H., Distribution Functions for Largest Eigenvalues and Their Applications , ICM 2012, Vol. I, 587-596, 2012.

Distribution functions well approximated by Gamma [ Chiani, 2012].

A common usage is to select the number of significant eigenvalues (smartpca uses it):

Patterson N, Price AL and Reich D. (2006). Population structure and eigenanalysis. PLoS Genet. 2 :20.

And see the review of ancestry programs by Liu et al [2013].

For a general discussion of random matrix theory (from the Wikipedia article), see Deift [2005]. Deift was a coauthor on the longest increasing subsequence paper.

Majumbar and Schehr [2014] discuss this as another extreme value distribution for strongly correlated random variables, specifically

the probability distribution of the top eigenvalue lambda-max of an N × N Gaussian random matrix for large N. While the typical small fluctuations of lambda-max of order ~O(N-2/3) around its mean lambda-max=sqrt(2) are described by Tracy-Widom distributions, atypically large fluctuations of O(1) are described by two different large deviation functions respectively on the left and on the right of the mean. These two tails correspond to very different physics in terms of [an] underlying Coulomb gas describing the eigenvalues: the left large deviation corresponds to a pushed Coulomb gas, while the right large deviation corresponds to a pulled Coulomb gas.

The T-W distribution in this setup represents the distribution around a phase transition of the system from strongly coupled ("unstable") to weakly coupled ("stable").

20181106: Raymond Ruyer

Very little in the Anglophone literature, but Deleuze and Weber both quote his work. Very similar flavour to the enactivists, though theistic neofinalist existentialist. From Wiklund [1960]:

Ruyer foursquarely maintains that meaning and direction (sens) are inherent in the inorganic and organic world, and that the activity of conscious man, firmly rooted in that world, is its natural prolongation. ...[A]n action, as it unfolds itself in a spatiotemporal world of cause and effect, cannot be understood without reference to its goal. This finality of the action gives meaning to all that is purely a succession of causes and effects in it.

His axiological cogito is "I am seeking a first truth, a rock-bottom certitude, therefore I am free. Freedom is then the certitude I sought, because the search for knowledge implies freedom, which is the positive condition of such a search." [Ruyer, Neo-finalisme 1952].

To deny freedom, pretending in so doing, to enounce a verity, that is ultimately to affirm freedom. The determinist is evidently not pushed into activity a tergo, as he would maintain, but rather initiates an argument to a purpose, and, in the very act, frees himself of the fetters he purports to prove are there. His activity is manifestly not purely mechanistic, but essentially axiological, i.e., it affirms a value; it strives toward a goal.

Neofinalist concepts are existence, freedom, finality, work, invention, value - that is goal-direction, that is life.

Neo-finalisme has a 2016 English edition and the above seems a good summary of his theses. For example, the form of his axiological cogito is based on the double dilemma type argument of Renouvier eg

  1. Being a pure set of processes, I affirm that my activity is senseless.
  2. Pursuing senseful ends, I affirm the absurd nature of my activity.
  3. Being a pure set of processes, I affirm that my activity has a sense.
  4. Pursuing senseful ends, I affirm that my activity has a sense.

Assertions 1 and 3 eliminate themselves. The fact that assertion 2 is an assertion completely undermines it. So assertion 4 remains.

Translating sens as sense might not be optimal here, given it is also direction and meaning.

20181030: Shklar and the view from nowhere

From The Liberalism of Fear:

To step outside these customs is not, as the relativist claims, particularly insolent and intrusive. Only the challenge from nowhere and the claims of universal humanity and rational argument cast in general terms can be put to the test of general scrutiny and public criticism.

20180701: Spring Mountain

Left Horse Float car park ~1020
White Rock ~1100
Spring Mountain summit ~1300
Left summit ~1345
Back at car ~1700
Total ~16.3 km.

Didn't rain, though cloudy and hazy to boot. Lots of families on the White Rock summit. We continued along main track to Spring Mountain, which we ascended from the N side.

Excellent views of Brisbane, Ipswich, Springfield. Has two solar-powered transmitters/repeaters and old survey marker. Returned via "Daisy's track", along ridge heading NNW, with little cliffs along and one razorback section. Return along Yuddamun track very undulating - all took longer than expected.

In blossom: banksias, some gums. Lots of smaller birds.

20180605: Revolt des anges

La guerre et le romantisme, fléaux effroyables
War and Romanticism, what terrible scourges!

20180603: Sverdrup

a Sverdrup (Sv) is a flow of a million cubic meters per second.

If you doubt that the AMOC has weakened, read this

For 1994-2013, Rossby et al. (2013) - at the Oleander line between 32° and 40° North - found a decrease in the upper 2000m transport of the Gulf Stream by 0.8 Sverdrup...

20180602: The preface paradox

From Ney [2016]:

The preface paradox (Makinson 1965) is that facing the writer of the work of nonfiction who after carefully researching all of the very many claims made in her book recognizes that given human fallibility, it is likely she made at least one mistake somewhere. So to be honest, she acknowledges this fact in her book's preface. In this (common) situation, the author has excellent reason for making each claim in the main text of the book. But she also has excellent reason to believe at least one of these claims is false.

In other words, FDR.

20180527: Commodity fetishism

Hudson and Hudson [2003]:

Commodity fetishism is the tendency, in a capitalist commodity system, for social relations between people to appear as a relationship between things (Marx, 1867/1976, p. 164). For example, capital appears to have a life of its own, capable of commanding a return and hiring and firing labor. However, capital is no more than the product of labor, organized and exploited through a social relation between the capitalist and the worker. All commodities appear to trade in relation with each other, to have an inherent value compared to all other commodities, but this merely masks the social relations hidden in the production of these commodities. Therefore, the thing becomes the bearer of value, not the labor that went into its creation (Geras, 1986, p. 59). This makes it appear as though there exists an invisible hand of the market that operates according to scientific laws, outside the realm of human control, when this market is merely the product of human relationships (Sweezy, 1942, p. 36). The organization of production through the purchase of commodities provides a highly effective mask over the exploitative class rela- tionships within a capitalist economy. It appears as though all actors in the economic system are the owners of a commodity, whether a specific product or an input into production, and as such, they stand as equals, each with something to sell in a voluntary exchange (Sweezy, 1942, p. 39).

20180521: On the Brains website

Carrie Figdor on her new book Pieces of Mind: The Proper Domain of Psychological Predicates

Hamlyn comments:

...the fundamental mereological principle that the parts that enable a decision maker to make decisions cannot themselves be decision makers.

and Figdor ripostes

Furthermore, and I also argue this at length in the book, the so-called mereological fallacy isn't a fallacy - I suspect it arises by taking compositional principles that appear to hold for objects and extending them, mistakenly, to activities. Elliot Sober also pointed this out long ago. Planets and atomic nuclei both rotate (to use his example, though the point is quite general) - a planet doesn't (and, maybe, cannot) have planets as parts, but there's nothing wrong with a planet rotating and its parts rotating.

20180517: GWAS meta-analysis under heterogeneity

Pritchard and Przeworski showed that the noncentrality parameter for the binary trait association Z-score is approximately rλN1/2. Han and Eskin [2011] invert this to estimate the apparent relative risk for the tag SNP (of a similar allele frequency) as:

RRM ~ [((RR-1)p - RR + 1)r + (1-RR)p-1] / [(RR-1)pr+(1-RR)p-1]

They point out that the fixed-effects meta-analysis combining Z values needs to use the SNP allele frequencies to correctly match the results from the full data Mantel-Haenszel test or inverse-variance weighted odds ratio.

Their main proposal is for a new random-effects approach that assumes there is no heterogeneity under the null hypothesis, that is that μ=0 and τ=0. In the presence of heterogeneity, they suggest a genomic control approach where they separately adjust the heterogeneity component and the main effect component of their statistic for multiple testing.

Their Metasoft program is at:


20180517: in silico v. in vivo mutation effects

Miosge et al [2015] assessed 33 de novo mouse mutations in "essential immune system genes" in vivo and all possible TP53 mutations in vitro measured as p21WAF1 transcriptional enhancer activity. For the mouse mutations:

PolyPhen SIFT
phenoBenignPoss.Dam.Prob.Dam. Tolerated Delet
No 11 (42%)8 (31%)7 (27%) 15 (58%) 11 (42%)
Yes 0 (0%) 1 (25%)3 (75%) 0 (0%) 4 (100%)

These are PolyPhen AUC c=0.81, and SIFT c=0.82.

For TP53:

Of the 1102 mutations predicted to be deleterious with [Polyphen2] score of 0.8 or greater, 42% had good TA activity measured in yeast with a reporter carrying the TP53-binding sequence from p21WAF1. When these 1102 predicted deleterious mutations were tested for activity against TP53-binding sequences from other target genes, the fraction that were FP predictions ranged from 34% for MDM2 sequences to 61% for P53R2 sequences. By contrast, false-negative (FN) predictions, where mutants have less than 50% of WT TA activity yet are predicted not to be damaging (PolyPhen2 0.2), accounted for only 93 of the 2026 mutations (4.6%).

20180516: Middle Kobble Creek

Fire breaks all signposted now. Lots of cobbler's pegs growing beside and often in the middle of roads. Round trip ~11 km (1140-1600) - 2.2 km along Lepidozamia Road to turnoff. Waterfall flowing. Newish shelter shed, water tanks and fireplace at Lepidozamia Park.

20180515: Soluble adenylate cyclase (ADCY10)

Pozdniakova and Ladilov [2018] review the role of soluble adenylate cyclase. The first 9 ADCY genes code the familiar transmembrane adenylate cyclases, whose produced cAMP are found only close to the plasma membrane (phosphodiesterases etc mop up). The cAMP present in organelles and the nucleus are produced by soluble adenylate cyclase. This is sensitive to bicarbonate, so among other roles acts as a pH sensor. Rahman et al [2016] showed that sAC is "essential for lysosomal acidification. In the absence of sAC, V-ATPase does not properly localize to lysosomes, [and] lysosomes fail to fully acidify". Specifically, sAC complexes with V-ATPase in many cell types, and is required for correct PKA-dependent translocation.

20180426: TOOLPACK epigraph

"To each is given a bag of tools,
A shapeless mass and a book of rules.
And each must make, ere life is flown,
A stumbling-block or a stepping-stone." R. L. Sharpe,

20180410: Synaesthesia genes

Tilot et al [2018]:

...we applied whole-exome sequencing to three families with sound-color (auditory-visual) synesthesia affecting multiple relatives across three or more generations. We identified rare genetic variants that fully cosegregate with synesthesia in each family, uncovering 37 genes of interest. Consistent with reports indicating genetic heterogeneity, no variants were shared across families. Gene ontology analyses highlighted six genes - COL4A1, ITGA2, MYO10, ROBO3, SLC9A6, and SLIT2 - associated with axonogenesis and expressed during early childhood when synesthetic associations are formed.

20180404: "UMP" Bayesian Tests

Johnson [2013] writes in the PNAS to suggest a critical P of 0.005:

Recently, Johnson [2013] proposed a new method for specifying alternative hypotheses. When used to test simple null hypotheses in common testing scenarios, this method produces default Bayesian procedures that are uniformly most powerful in the sense that they maximize the probability that the Bayes factor in favor of the alternative hypothesis exceeds a specified threshold. A critical feature of these Bayesian tests is that their rejection regions can be matched exactly to the rejection regions of classical hypothesis tests.

Andrew Gelman and many others commented unfavourably at the time.

20180403: Reliability of psychometric tests

Test-retest correlation:

Study Instrument r N Notes
Catron and Thompson [1979] WAIS 0.74-0.90 76 1 wk to 4 mo
Snow et al [1988] WAIS 0.90 101 1 y
Spitz et al [1983] WAIS 0.75 42 Young MR
Spitz et al [1983] WISC 0.84 24 Young MR
Rae and Olson [2018]IAT 0.34-0.48 519 Children
Bar-Anan and Nosek [2014] IAT 0.45 116 1 h

Practice effects for the WAIS are well-known, 5-10 points for FSIQ after 3 weeks to 6 month intervals. Estevis et al [2012] suggest that the individual FSIQ level predictive half-interval is 11.

Rae and Olson [2018] comment "[t]he Implicit Association Test (IAT) is increasingly used in developmental research despite minimal evidence of whether children's IAT scores are reliable across time or predictive of behavior".

The IAT-anxiety test has test-retest of 0.5 over 1 year [Egloff et al 2005].

20180403: Self-domestication in Homo sapiens

Theofanopoulou et al [2017]:

A broad consensus is now emerging that "commensal" and "mutualistic" processes can lead to domestication, whereby both the domesticator and domesticated species seek out and benefit from cohabitation...Many of the species that have ultimately come to inhabit domestic niches are widely considered to have done so largely autonomously; in other words, to have self-domesticated...[phenotypes include] depigmentation; floppy, reduced ears; shorter muzzles; curly tails; smaller teeth; smaller cranial capacities (and concomitant brain size reduction); paedomorphosis; neotenous (juvenile) behavior; reduction of sexual dimorphism (feminization); docility; and more frequent estrous cycles.

A subset of the genes flagged as selected overlapping between humans and other domesticated species:


20180323: Artificial Intelligence

Minsky in Semantic information processing [1968]:

the science of making machines do the things that would require intelligence if done by men.

Digging a hole? Came up when chasing the Emotions-Beliefs-Desires-Intentions model for agent-oriented programs and agent-based (social) simulations.

20180315: And see

a new website

20180228: The graphic novel

Patrick [2012]:

Richard Kyle coined the phrase "graphic novel" in CAPA-alpha, an American comic fanzine, in November 1964, and used it to describe "long-form" comic book stories of broadly serious artistic merit. The term was adopted to promote a cluster of science-fiction/fantasy titles, including Richard Corben's Bloodstar (Corben 1976), George Metzger's Beyond Time and Again (Metzger 1976) and Don McGregor and Paul Gulacy's Sabre (McGregor and Gulacy 1978), which were published as paperback books, and distributed via specialty comic shops. However, it was left to Will Eisner's A Contract with God (Eisner 1978) - a work erroneously cited as the first American graphic novel - to demonstrate the medium's potential...

...popularized 1985 [onwards]...Batman: The Dark Knight Returns... Watchmen...Maus

20180226: FOXQ1 and melanoma

Bagati et al [2018] (includes Neil Box) previously reported on a role for FOXQ1, previously implicated in numerous tumour types, as a tumour suppressor gene for melanoma. In the present paper, they show FOXQ1 acting upstream from MITF to modulate melanocyte differentiation.

They find the MITF promotor contains FOXQ1 binding sites, confirmed on ChIP. Forskolin was already known to upregulate MITF levels, and they show that depleting FOXQ1 abrogates this (either directly - in culture, or in a FOXQ1 KO mouse model). FOXQ1 may be the intermediary for CREB1 regulation of MITF.

...a potential functional cooperation between FOXQ1 and CREB1 similar to that between SOX10 and its target gene PAX3. Similar to FOXQ1, SOX10 also directly activates MITF [36] and is also required for full-scale activation of MITF by PAX3.

Foxq1-null mice retain baseline pigmentation. This observation is consistent with other studies demonstrating that mice deficient of CREB1 or another MITF activator PGC1-alpha also demonstrate "no coat-color" phenotype, underlining the complexity of basal versus induced pigmentation in mouse skin...

UV -> keratinocytes -> MSH -> CREB1 -> MITF

In human and mouse melanocytes, BRAF*600E decreased FOXQ1 (as did depletion of beta-catenin, the most likely downstream target of V600E). Bumping up FOXQ1 levels in these systems seemed to return the cell phenotype to normal.

Germline coding variants in FOXQ1 are not increased in familial melanoma according to Artomov et al [2017]. The best SNPs from the melanoma and nevus metanalyses are rs35434757 (mela P=0.0008) and rs2317902 (mela P=0.000939); no associations in the UKBB for melanoma or pigmentation phenotype.

20180221: Meaning eliminativism

Callum Hackett comments:

The upshot of this view (known as "meaning eliminativism" in a branch of pragmatics called "relevance theory"), is that words in contexts of use are excellent for referring to concepts, but words themselves do not achieve this by denoting concepts. Instead, words have non-conceptual potentials for conceptual reference, and humans use their pragmatic capacities to fix conceptual referents in context (it is too much to discuss here, but it is important to note that a non-conceptual potential is something quite different from polysemy, which is just a many-to-one correspondence between a word and possible concepts - the claim is instead that words denote something that is materially different from conceptual concept, though use of a word is perceived to have conceptual reference).

Four contextualist positions from Recanati [2004] summarised by Carsten [2012]:

quasi-contextualism word meaning may contribute directly, or be modified by context
pragmatic compositionword meaning "overwritten" by context
"wrong format" word meanings must be translated into concepts
meaning eliminativismonly specific utterings of words have meaning

Underlying these,

(a) collections of memory traces or exemplars of previous uses (tokenings) and (b) bundles of contingent encyclopaedic information about the things in the world the word is used to refer to.

So how does analogy-making fit in?

20180219: Imputation using association statistics

Benner et al [2017]:

During the past few years, various novel statistical methods have been developed for fine-mapping with the use of summary statistics from genome-wide association studies (GWASs). Although these approaches require information about the linkage disequilibrium (LD) between variants, there has not been a comprehensive evaluation of how estimation of the LD structure from reference genotype panels performs in comparison with that from the original individual-level GWAS data. Using population genotype data from Finland and the UK Biobank, we show here that a reference panel of 1,000 individuals from the target population is adequate for a GWAS cohort of up to 10,000 individuals, whereas smaller panels, such as those from the 1000 Genomes Project, should be avoided. We also show, both theoretically and empirically, that the size of the reference panel needs to scale with the GWAS sample size; this has important consequences for the application of these methods in ongoing GWAS meta-analyses and large biobank studies. We conclude by providing software tools and by recommending practices for sharing LD information to more efficiently exploit summary statistics in genetics research.


20180216: Representation

Franz Burto summarises:

...mental representations represent (1) linguistically (in whatever way language represents, without the intervention of sensory modalities); (2) pictorially (in whatever way pictures represent, via quasi-spatial features and the involvement of sensory modalities); (3) a mixture of (1) and (2); (4) magically (they just represent, with no further questions asked on how they do it).

Some cognitive psychologists, like Allan Paivio, have defended a "dual coding" theory of mental representations whereby some represent linguistically, some pictorially. People like Kosslyn and others attach great importance to representations of kind (2); others, like Fodor or Pylyshyn, claim that it all reduces to (1)...

I have an argument by cases. Roughly: if the conceivability of P at issue for Humeans involves representation of kind (1), we can conceive the impossible...

People often call "Meinongians" (from the Austrian philosopher Alexius Meinong) those philosophers who claim that some things do not exist... Meinongians are opposed to Quineans, who, on the contrary, claim that everything exists.

His (along with Rossi and Tagliabue) 2010 book The Mathematics of Models of Reference describes an ambitious analysis of a reversible CA system, where the claim is made that recursive self-reference can be implemented:

Recursive self-reference takes place when a MoR [Model of Reference] not only refers to itself, but is aware of such a self-reference. A system implementing a self-referentially recursive MoR can therefore be aware of what it does via that very MoR.

Again, this is not semantic animism. For such expressions as "being aware" can be given a precise mathematical meaning. Much research at iLabs is guided by the persuasion that recursive self-reference is at the basis of what people ordinarily call "consciousness": a key difference between conscious thoughts and any other computational procedure is that our mind, as (self-)conscious, can think about and have a viewpoint on itself (albeit with arguably limited powers to operate on its own source code). If Artificial Intelligence is to be real, this will be achieved by means of recursive self-reference - or this is our bet.

The [Kleene (Strong)] Recursion Theorems, applied to our recursive MsoR, guarantee that we can define partial MsoR which are recursively self-referential, for they include their own code in their recursive definition. These are simply classical fixed-point definitions. Since numeric codes are perceptions taken as inputs by (meta-)models of reference, which can also emulate the thought procedures performed by the encoded MsoR, recursive self-referential MsoR can perceive themselves in a precise mathematical fashion, and represent the computational procedure in which they consist within themselves.

20180212: HERRA (SNP heritability from glmnet)

Gorfine et al [2017] estimate the SNP heritability via:

Since the lasso possesses the oracle property, it can be shown (details in S1 and S2 Text) that, as the sample size increases, [this estimate] converges to the true heritability value.

They argue that current approaches will underestimate the heritability especially when the number of causative variants is small.

Seminal concept for predictor selection is the sure screening property.

"given n samples for each of p variables, we will use the term "high dimensional" to mean p = O(nk) for some k > 0, and the term "ultra-high dimensional" to mean log(p) = O(nk) [Reese et al 2018]." Saldana and Feng [2017]:

Fan and Lv (2008) introduced a new framework for variable screening via independent correlation learning that tackles the...challenges of ultrahigh dimensional linear models. Their proposed sure independence screening (SIS) is a two-stage procedure; first filtering out the features that have weak marginal correlation with the response, effectively reducing the dimensionality p to a moderate scale below the sample size n, and then performing variable selection and parameter estimation simultaneously through a lower dimensional penalized least squares method such as SCAD or LASSO. Under certain regularity conditions, Fan and Lv (2008) showed surprisingly that this fast feature selection method has a "sure screening property"; that is, with probability tending to 1, the independence screening technique retains all of the important features in the model.
Fan and Lv [2010]

As shown in Fan and Fan (2008), even for the independence classification rule described in Section 4.2, classification using all features can be as bad as a random guess due to noise accumulation in estimating the population centroids in high dimensional feature space. Therefore, variable selection is fundamentally important to high dimensional statistical modeling, including regression and classification.

In the correlation screening setup, they show it is possible to reduce the number of included covariates to a submodel with a very high specifiable probability of including all the important predictors, given that a sparse model is correct. Several screening techniques can be shown to have the sure screening property. Once the set is reduced to a moderate scale, other methods can now be used without computational problems. Using a split-half approach, since the sure screening property is present for each analysis, choosing intersecting predictors reduces the false-positive rate.

Fan J, Lv J. Sure independence screening for ultrahigh dimensional feature space. Journal of the Royal Statistical Society: Series B. 2008;70(5):849¿911.

Fan J, Guo S, Hao N. Variance estimation using refitted cross-validation in ultrahigh dimensional regression. Journal of the Royal Statistical Society: Series B. 2012;74(1):37-65.

The R SIS package provides this approach to multiple types of regreesion. The TSGSIS package is an experimental package for pruning SNP pairwise interactions.

Reese et al [2018] discuss these approaches using an extended Cochrane-Armitage test, and compare this to a distance correlation based method. Again, it uses just the screening statistic for each covariate (SNP) in turn.

Pan et al [2017] discuss the Ball correlation, a nonparametric correlation measure (in Banach space). The R Ball package provides a bcorsis() function for screening, using n/log(n) as the default number of variables to retain (or can use jackknifey n-1).

20180211: amodal completion

Rauschenberger & Yantis [2001]:

When one object is partly occluded by another, its occluded parts are perceptually "filled in", that is, the occluded object appears to continue behind its occluder. This process is known as amodal completion... the input to visual search is much more complex than previously assumed ...the entry level for vision (that is, entire objects or individual features) can be quite high in many cases,

...In our experiments, participants searched for a notched disk target among complete disks and squares. With unlimited exposure duration, when the notched target disk abuts a square (adjacent condition), search is inefficient because the notched target is rendered similar to the complete distractor disks by amodal completion...

20180206: The Nonidentity Problem

John Danaher discusses this -

that is, the moral standing of future generations.

  1. An action cannot be wrong unless it does harm.
  2. Some risky policy (X) does no harm.
  3. The risky policy (X) is wrong.

20180201: Duncan Murdoch's tables package

# Dealing with incomplete data
N <- function(x) sum(complete.cases(x))
MEAN <- function(x) mean(x, na.rm=T)
SD <- function(x) sd(x, na.rm=T)

f <- formula(cmm * study ~ (n=1) + Height*(mean + sd))
tabular(f, data=subset(x, complete.cases(Height, study ,sex)))

tabs <- list()
tabs[[1]] <- tabular((Factor(BigHx, "History")+1) ~
              ((n=1) + Format(digits=3)*Percent()),
              data=subset(x, !is.na(BigHx)))
tabs[[2]] <- tabular((Factor(Hair_colour)+1) ~
              ((n=1) + Format(digits=3)*Percent()),
              data=subset(x, !is.na(BigHx)))
tabs[[3]] < tabular((1+Factor(sex)) * (Age+TNC5+reflectance_axilla + 
                                         reflectance_outer + reflectance_inner) ~
          Factor(CM)*(N+Format(digits=2)*(MEAN+SD)), data=x)

f <- "tables.html"
con <- file(f, "wt")
writeLines(myheader, con)
for(i in seq(1,length(tabs))) {

", con) html(tabs[[i]], con) } writeLines("", con) close(con)

20180125: PRMT5 and MTA

PRMT5 is at HSA 14:23.33-23.45. In UKBB, the only significant SNP association is rs11157931 (RBC distribution width).

PRMT5 is a Type II arginine methyltransferase involved in post-translational protein modification. It acts as a transcriptional corepressor via methylation of histones H4R3me2s and H3R8me2s, and prepares proteins for their assembly into mature small nuclear ribonucleoproteins (snRNPs). Unlike other PRMTs, PRMT5 complexes with other proteins, most importantly MEP50, but also RioK1 and PlCln. (review by Blanc and Richard [2017]). The H4R3me2s is a flag for DNMT3a to then methylate the associated DNA.

PRMT5 is inhibited by 5'-O-methylthioadenosine (MTA), the substrate for methylthioadenosine phosphorylase (MTAP), a competitive inhibitor of S-adenosylmethionine.

In mammals, arginine methylation is a modification as common as phosphorylation and ubiquitination. The PRMTs produce the activating H4R3me2a, H3R2me2s, H3R17me2a, H3R26me2a and repressive H3R2me2a, H3R8me2a, H3R8me2s, H4R3me2s) histone marks.

Tamiya et al [2018]:

SHARPIN, an adaptor for the linear ubiquitin chain assembly complex (LUBAC), plays important roles in NF-kappaB signaling and inflammation. Here, we have demonstrated a LUBAC-independent role for SHARPIN in regulating melanoma growth. We observed that SHARPIN interacted with PRMT5...and increased its multiprotein complex and methyltransferase activity. Activated PRMT5 controlled the expression of the transcription factors SOX10 and MITF by SHARPIN-dependent arginine dimethylation and inhibition of the transcriptional corepressor SKI.

Linear ubiquination is involved in NF-kappaB signalling. SHARPIN binds the PRMT5/MEP50 complex and increases its methyltransferase activity. SHARPIN is at HSA8:141Mbp. Like PRMT5, the SNP associations are with blood cell properties (RBC and platelet), but also height.

20180123: Vini Reilly and The Durutti Column

From Wikipedia:

The Durutti Column are an English post-punk band formed in 1978 in Manchester, England. The band is a project of guitarist and occasional pianist Vini Reilly who is often accompanied by Bruce Mitchell on drums and Keir Stewart on bass, keyboards and harmonica.

Very melodic. Shredding, I guess.

20180119: Androgen Insensitivity

Major cause, especially of complete insensitivity, is AR mutation. FLNA (Filamin 1, Xq28) is an androgen cofactor where mutation leads to diminished translocation of the AR. FLNA mutation has many other pathologies eg female early progressive obstructive lung disease, FLNA-associated neuronal periventricular nodular heterotopia, seizures, otopalatodigital syndromes (OPD1, OPD2), frontometaphyseal dysplasia (FMD), Melnick-Needles syndrome (MNS), and X-linked congenital idiopathic intestinal pseudoobstruction (CIIPX).

Mutations in SRD5A2 (steroid 5 alpha-reductase 2, 2p23.1) affect dihydrotestosterone conversion, and are more commonly associated with disorders of sexual development such as hypospadias.

NR5A1 (steroidogenic factor-1, 9q33.3) mutations are associated with testicular dysgenesis. XY gonadal dysgenesis is also called Swyer syndrome. Other Swyer genes include DHH, NR0B1, SOX9 and MAP3K1.

SRY mutations similarly lead to female sex reversal and dysgerminoma in 10-15% of cases. SRY is intronless, and mutations of 80 of its 204 amino acids causes disease.

Savic et al [2017] report that in complete androgen insensitivity syndrome (CAIS):

Compared to men, CAIS women displayed a "female" pattern by having thicker parietal and occipital cortices, lower fractional anisotropy (FA) in the right corticospinal, superior and inferior longitudinal tracts, and corpus callosum. Their functional connectivity from the amygdala to the medial prefrontal cortex, was stronger and amygdala-connections to the motor cortex weaker than in control men. CAIS and control women also showed stronger posterior cingulate and precuneus connections in the default mode network. Thickness of the motor cortex, the caudate volume, and the FA in the callosal body followed, however, a "male" pattern.

Altogether, these data suggest that testosterone modulates the microstructure of somatosensory and visual cortices and their axonal connections to the frontal cortex. Testosterone also influenced functional connections from the amygdala, whereas the motor cortex could, in agreement with our previous reports, be moderated by processes linked to X-chromosome gene dosage.

20180118: The Golden Age of SF

Ahasuerus writes on r.a.sf.w:

As a general observation, Ernst Haeckel's "ontogeny recapitulates phylogeny" is not a bad first-order approximation for magazine SF in the 20th century. Roughly:

11: 1910s (Ralph 124C 41+)
12: 1920s (Ray Cummings)
13: 1930s (Doc Smith)
14: 1940s (the Golden Age)
15: 1950s (Galaxy)
16: 1960s (the New Wave)

20180111: The Alternative Natural Philosophy Association (ANPA)

Nick Rossiter and Michael Heather have a number of papers on the application of category theory to metaphysics. It bears some resemblance to Castonguay set-theoretic approach [1972, editor Mario Bunge] with mappings from the Formal System to the Natural System (intension being relations between concepts in the FS - or "set of constructs C", and extension the mapping from FS to NS "set of objects O"), as per the 17th century Port Royal School [Arnauld and Nicole, La logique ou l'art de penser] (more famous for their Jansenism). Rossiter and Heather [2015] extend Rosen's category theoretical models for anticipatory systems (Rosen considered this a cardinal feature of life). If mapping is inperfect, adjointness is the appropriate concept.

Lawvere and Schanuel [2009]. Conceptual Mathematics: A First Introduction to Categories.
Lawvere and Rosebrugh [2003] Sets for Mathematics

20180104: Personality over time and culture

Bilstedt et al [2014] report on 153 women completing Eysenck Personality Inventory in 1968 and 2005 (37 y): N r=0.50, E r=0.57; and in 274 women 1968 to 1992 (24 y): N r=0.58, E r=0.63.

Briley and Tucker-Drob [2014] performed a meta-analysis of the stability of personality in 21057 sib pairs. They note

...an influential meta-analysis of test-retest data from 152 longitudinal studies of personality found increases in 7-year stability coefficients from .3 in early childhood to .6 by early adulthood, and to .7 by later adulthood [ Roberts & DelVecchio, 2000]. One of the remarkable findings of Roberts and DelVecchio (2000) is that there are practically no differences between personality traits in terms of differential stability.

Briley and Tucker-Drob found the phenotypic stability of personality was ~0.5 at age 15, stabilizing at ~0.7 by age 30. The genetic stability was ~0.8 at age 15, and unity after age 30. The environmental stability was ~0.4 at age 15, but corrected for measurement error was ~0.5. From Widger et al The Oxford Handbook of the Five Factor Model

McCrae et al [1998] studied Chinese undergraduates in Hong Kong and Vancouver. Canadian-born Chinese were closer to other Canadians than were recent emigrants. Other migration studies are Gungor et al 2013, Soldner et al 2012, Zimmermann and Neyer 2013. Twenge [2000, 2001, 2002, 2007, 2008] find that American students over 50 years are ~1 SD higher on both N and E (NEO).

Yu et al [2009] reextracted factors from a large panel of descriptors in 610 Chinese students, finding Intelligent, Emotional, Conscientious, Unsocial and Agreeable. Zhu et al [2009] ended up with 7 factors, Extraversion, Conscientiousness, Unselfishness, Emotional volatility, Negative Valence, Positive Valence/Intellect, Fragility.

20171227: Pro and con Evolutionary Psychology

Various articles on EP from Frontiers in Psychology

20171220: African skin pigmentation

Crawford et al [2017] reported some novel loci in African populations, and shortly after, Martin et al [2017] report on a skin reflectance GWAS in 465 Khoe-San (Khomani San, Nama). The Khoe-San are lighter skinned than surrounding populations, and have brown, hazel and green eyes. A big effect of SLC24A5 was confirmed - rs1426654 again (but also SLC45A2 rs16891982). A sliding window weighted Fst comparing KhoeSan to Bantu and Europeans flagged SLC24A5 and OCA2. The OCA2 rs1800417 SNP was associated with eye colour but not skin colour.

Both studies use DermaSpectrometer DSMII Martin log10[1/%red reflectance] Irish=25 Ghana=100 Crawford 100*log10[1/prop_red] 45-145 Bantu=

From Table S1 Crawford et al and Martin et al

Study Population Lat N Mean melanin indexSD
Crawford Botswana Bantu 292 81.3
Crawford Botswana San 358 69.2
Crawford Botswana San/Bantu 106 74.7
Crawford "Coloured"* 13 65.3
Crawford Cushitic 292 84.3
Crawford Ethiopia Nilosaharan 112 119.9
Crawford Ethiopia Omotic 155 98.2
Crawford Ethiopia Semitic 193 79.6
Crawford Tanzania Hadza 379 101.5
Crawford Tanzania Nilosaharan 103 99.9
Crawford Tanzania Sandawe 98 82.4
Candille Ireland 53.35 42 27.10 1.95
Candille Poland 52.23 45 30.54 1.86
Parra European 48.14 135 28.32 2.72
Candille Italy 41.90 45 31.35 2.06
Candille Portugal 41.16 43 30.73 2.46
Parra East Asian 35.86 116 31.07 2.98
Coussens Xhosa 34.03 50 67.10 7.50
Coussens Cape Mixed 33.91 42 53.10 8.50
Basu Mallick Ror 29.69 44 42.16 4.11
Martin Nama 28.82 223 52.12 8.93
90+% KhoeSan Nama 28.82 15 51.70 5.74
Martin Khomani 28.40 278 57.57 10.12
90+% KhoeSan 28.40 41 56.10 6.93
Sandoval/Moreno Easter Island 27.13 81 44.27 5.57
Parra South Asian 21.13 117 37.48 5.48
Basu Mallick Nihali 20.94 33 58.01 6.66
Basu Mallick Korku 20.94 35 52.46 7.92
Basu Mallick Kapu 17.39 272 43.41 4.87
Basu Mallick Naidu 17.39 112 43.57 4.87
Basu Mallick Reddy 17.05 844 43.72 4.87
Sandoval/Moreno Peruvians 15.85 117 51.88 4.58
Beleza Cape Verde 15.11 684 55.39 13.00
Q1 African ancestry 15.11 171 46.23 7.73
Q2 African ancestry 15.11 171 51.98 9.48
Q3 African ancestry 15.11 171 54.31 9.83
Q4 African ancestry 15.11 171 69.09 12.10
Basu Mallick Toda 11.41 11 42.96 4.12
Basu Mallick Kurumba 11.41 24 56.41 5.29
Basu Mallick Badaga 11.41 31 46.18 6.15
Basu Mallick Kota 11.41 27 45.13 6.30
Basu Mallick Saurashtrian 9.92 25 42.36 4.90
Basu Mallick Brahmin 9.92 20 41.07 5.51
Basu Mallick Yadava 9.92 27 58.21 8.03
Durazo-Arvizu Ghana 6.67 237 96.04 10.94
Norton Bougainville 6.05 153 89.84 9.52
Norton Papua New Guinae 5.90 21 67.94 9.42
Norton New Britain 5.56 491 67.92 6.80
Norton New Ireland 3.32 242 74.22 8.06
Norton Lavongai 2.52 102 77.12 7.51
Norton Mussau 1.41 35 65.48 7.95

Lasisi and Shriver [2018] comment:

For their global comparison of skin pigmentation and latitude, Martin et al. extracted skin reflectance measures from the literature that were made with two reflectometers, which employ different approaches to measuring skin color. Although they are manufactured by one company (Cortex Technology, Hadsund, Denmark) and both refer to their readings as "M index", one uses a red light-emitting diode (LED) to produce a narrow peak centered on 655 nm (DermaSpec) and the other uses white LED illumination extracting the R component of the tristimulus red, green, blue (RGB) color system as a measure of reflected red light (DSMII). Though these measures are similar, there are clear differences that should be accounted for in analyses that combine data from more than one device.

From https://khomani.co.za/

The Bushmen or San as they are known to Science, are part of the so-called Khoi-San races. The word KhoiSan denotes the two related groups of people, namely the larger KhoiKhoi people, also called Hotentotte who were a physically larger and more yellow people than the San or Bushmen, and who were cattle-herders and pastoralists. The term "San" was historically applied by their ethnic relatives and historic rivals, the Khoikhoi. This term means "outsider" in the Nama language.

...The two largest San groups in South Africa are immigrants from Angola via Namibia. These are the !Xû (3500) and the Khwe (1100). The next largest group (~1500 but more diffuse) are...those from the Southern Kalahari around Witdraai, called the "Khomani San" by scientists, although the remaining speakers of the language, of which 25 have now been identified, say that this refers to a specific clan. Academics from the turn of the century had used the name Khomani san for them, but the community says it is the name of a related clan, that does not really apply to them. They refer to themselves in the aggregate as "Saasi", and their language as "!Kabee". Most speak Khoekhoegowap or Afrikaans.

There is a group of about 70 adult !Kung San living across the border from South Africa at Masetleng and Ngwaatle Pans in Botswana. These people originally lived next to the Khomani in what became the Kalahari Gemsbok National Park (KGNP). They were displaced by the KGNP and driven into Botswana. In 2007, a group of San won their court case in Botswana regarding expulsion from the Central Kalahari Game Reserve.

There are also thousands of people in the Northern Cape who are to some degree aware that they are direct descendants of the largest South African San population of the 18th and 19th centuries, the !Xam.

The Khoisan languages of Africa are unique in incorporating a large variety of click consonants. The Hadza and Sandawe of Tanzania are the only Khoisan-speaking populations to reside outside of southern Africa. However, inclusion of the Hadza and Sandawe in the Khoisan language family is highly contentious among linguists. Despite their small population sizes (the Hadza number only ~1000 individuals), the Khoisan-speakers exhibit a large amount of Y-chromosomal and mitochondrial genetic diversity. The southern KhoeSan or "Bushmen" populations (see Materials and Methods for population nomenclature) today occupy the Kalahari Desert region of Botswana, Namibia, and northern South Africa.

Henn et al [2010] note that "KhoeSan individuals retain many of the deep divergence mtDNA haplogroups...[but] ~35% of the paternal hunter-gatherer lineages were of either European or Bantu origin. Across all three click-speaking populations the contribution of maternal haplogroups from putative agriculturalists varied from 0 to 28%, whereas the corresponding paternal contribution of agriculturalists varied from 30% to 42%".

Schlebush et al [2012] report 2.3M SNPs from 220 individuals from 11 Southern African populations. There was distinct genetic separation between Northern San (Ju, !Xun aka !Kung, Ju/'hoani) and Southern Khoe-San (Tuu, Khoe speakers - Kerretjie, Khomani and Nama)

Here are maximum likelihood estimates of haplotype frequencies for a 41 kbp haplotype spanning SLC24A5, rs1834640-rs1426654-rs2470102, combining the SNP genotyping from Schlebush et al [2012] with 1000 Genomes data. The A is European light skin allele, G is that present in chimp, orangutan and macaque. The haplotypes seem to be:

SNP Build38 Build37
rs1834640 48099968 48392165
rs1426654 48134287 48426484
rs2470102 48141297 48433494

G-G-G, the ancestral dark-skin haplotype;
A-A-A, the European light-skin haplotype;
G-G-A, an Asian haplotype presumably associated with darker skin;
G-A-A, a South Asian haplotype presumably associated with light skin;
A-G-G, an East Asian haplotype presumably associated with darker skin.

Denisovan sequence: G-G-G19 reads for rs1426654
Vindija Vi33.26 Neandertal:G-G-?Single read

Note the expected Asian haplotypes in the Southern African "Coloured" samples.

Presumed phylogeny G-G-G

Sample NA-A-A G-A-A A-G-A G-G-A A-A-G G-A-G A-G-G G-G-G
Khomani+Nama 53 0.35850.00000.00000.00000.00000.00000.00940.6321
Other KhoeSan 80 0.09380.00000.00000.00000.00000.00000.00000.9062
"Coloureds" 38 0.38160.03950.00000.03950.00000.00000.00000.5395
Other African 706 0.07370.00070.00000.00140.00070.00000.00010.9235
Yoruba (NIG) 108 0.0139 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.9861
Esan (NIG) 99 0.0253 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.9747
Luhya (KEN) 101 0.0693 0.0050 0.0000 0.0000 0.0000 0.0000 0.0000 0.9257
Gambia 113 0.0752 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.9248
Mende (SL) 84 0.0893 0.0000 0.0000 0.0060 0.0000 0.0000 0.0000 0.9048
Bantu 32 0.0938 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.9062
AfroCaribbean 96 0.1042 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.8958
AfroAmerican 65 0.1845 0.0000 0.0000 0.0078 0.0078 0.0000 0.0000 0.7999
Europe 505 0.98710.00500.00100.00100.00500.00020.00100.0002
Finnish 99 0.9747 0.0051 0.0051 0.0050 0.0101 0.0001 0.0001 0.0001
CEU 99 0.9747 0.0101 0.0001 0.0001 0.0152 0.0001 0.0001 0.0001
Tuscany 108 0.9907 0.0046 0.0000 0.0001 0.0000 0.0001 0.0046 0.0001
Spain 107 0.9953 0.0047 0.0002 0.0002 0.0002 0.0002 0.0002 0.0002
British 92 1.0000 0.0002 0.0002 0.0002 0.0002 0.0002 0.0002 0.0002
South Asia 494 0.67500.01230.00000.04250.00000.00000.01220.2580
Tamil 103 0.4756 0.0098 0.0000 0.0485 0.0000 0.0000 0.0098 0.4562
Bengali 86 0.5289 0.0060 0.0020 0.0445 0.0000 0.0000 0.0331 0.3855
India Telugu 103 0.6355 0.0198 0.0000 0.0874 0.0000 0.0000 0.0198 0.2375
Punjabi 96 0.7656 0.0156 0.0000 0.0260 0.0000 0.0000 0.0000 0.1927
Gujarati 106 0.9434 0.0094 0.0000 0.0047 0.0000 0.0000 0.0000 0.0425
East Asia 515 0.01060.00100.00260.23230.00000.00000.06540.6880
South Han 108 0.0185 0.0000 0.0082 0.2557 0.0000 0.0000 0.0844 0.6332
Han Chinese 103 0.0238 0.0053 0.0000 0.2039 0.0000 0.0000 0.0733 0.6937
China Dai 99 0.0001 0.0001 0.0000 0.2222 0.0001 0.0001 0.0606 0.7172
Japan 104 0.0048 0.0000 0.0052 0.2352 0.0001 0.0001 0.0284 0.7264
Vietnam Kinh 101 0.0050 0.0001 0.0024 0.2402 0.0000 0.0001 0.0768 0.6757
Americas 350 0.58540.00310.00230.07060.00150.00000.09650.2405
Peru 86 0.2849 0.0000 0.0000 0.1395 0.0000 0.0000 0.1628 0.4128
Mexican 65 0.5154 0.0000 0.0125 0.0952 0.0000 0.0000 0.1260 0.2509
Columbian 94 0.7012 0.0114 0.0000 0.0321 0.0055 0.0000 0.0752 0.1746
Puerto Rico 105 0.7714 0.0000 0.0000 0.0333 0.0000 0.0000 0.0429 0.1524
Overall 2741 0.4142 0.0045 0.0012 0.0613 0.0013 0.0000 0.0270 0.4907

From Haak et al [2015], compare this to rs2250072-rs12441154, which can be used to impute rs1426654
Sample N rs2250072-rs12441154 Haplotype Imputed rs1426654 genotypes
A-C G-C A-T G-T A/A A/G G/G rs1426654*A
Nonhuman 3 0.0001 1.0000 0.0001 0.0001 0 (.000) 0 (.000) 3 (1.00) 0.0000
Neandertal Vi33.261 1.0000 0.0001 0.0001 0.0001 1 (1.00) 0 (.000) 0 (.000) 1.0000
Denisovan 1 0.0001 1.0000 0.0001 0.0001 0 (.000) 0 (.000) 1 (1.00) 0.0000
Pleistocene HG 3 0.0001 0.3333 0.0001 0.6667 1 (.333) 0 (.000) 2 (.667) 0.3333
Holocene HG 9 0.5556 0.0556 0.0000 0.3889 5 (.556) 0 (.000) 4 (.444) 0.5556
Early Neolithic 11 0.6364 0.2727 0.0909 0.0000 11 (1.00) 0 (.000) 0 (.000) 1.0000
Middle Neolithic 3 0.3333 0.6667 0.0001 0.0001 4 (1.00) 0 (.000) 0 (.000) 1.0000
Late Copper 5 1.0000 0.0001 0.0001 0.0001 5 (1.00) 0 (.000) 0 (.000) 1.0000
Late Neolithic 4 0.7500 0.0001 0.0001 0.2500 5 (.833) 0 (.000) 1 (.167) 0.8333
Early Bronze 5 1.0000 0.0001 0.0001 0.0001 5 (1.00) 0 (.000) 0 (.000) 1.0000
Late Bronze 1 1.0000 0.0001 0.0001 0.0001 1 (1.00) 0 (.000) 0 (.000) 1.0000
Iron Age 1 - - - - 1 (1.00) 0 (.000) 0 (.000) 1.0000
Modern 1988 0.5772 0.1097 0.0093 0.30381049 (.542)259 (.134)628 (.324) 0.6087

For South American pigmentation data, see Ruiz-Linares et al ]2014]

20171130: some nice looking R packages and a cancer genome browser

http://www.sthda.com/english/: Statistical tools for high-throughput data analysis.


The cBioPortal for Cancer Genomics provides visualization, analysis and download of large-scale cancer genomics data sets. Please cite Gao et al. Sci. Signal. 2013 &amp; Cerami et al. Cancer Discov. 2012 when publishing results based on cBioPortal.

1000 Genomes Selection Breowser

20171122: MC1R

Ding et al [2014] suggest that V92M might be Neanderthal in origin. They base this on "59 Neanderthal AIMs ...identified within the MC1R gene and its left and right 100-kb flanking regions" - actually concentrating on 16:89941438-90008296. They found that "all putative introgressive haplotypes carry the derived allele of the Val92Met variant (rs2228479*A)", but that this is absent in the Altai Neanderthal woman. This is explained by the hypothesis that the founder mutation postdates that individual. They also see the derived allele in one non-introgression haplotype, and have to invoke recurrent mutation.

20171114: Egocentric thought and self-representation

Conference proceedings on egocentric thought and self-representation:

The aim of the conference is to shed light on the complex relationship between the salient epistemic properties of egology - or what we may call, after Lewis 1979, "de se thoughts" - and the putative existence and nature of an under­lying self-representational structure. Egocentric thoughts, like those expressed by a subject when she declares "I wish a woman could win the election» or "I see a canary", can be conveniently singled out in terms of their specific epistemic proper­ties. These thoughts are reputed to be invulnerable to certain forms of error: they display what Shoemaker has famously called "immunity to error through misidentification", as well as, in core cases, immunity to error through misascription of properties. Other remarkable epistemic features of I-thoughts include their apparent groundlessness, or exemption from the need of an epistemic justification, the privilege of first-personal access to such thoughts, and the subsequent authority attached to first-personal reports on them.

...[T]wo antagonist paths to account for these peculiarities...the first option stems from a positive answer to this question [self-representation as self-reference and token-reflexivity] According to a number of authors, who appeal to a plausible principle of isomorphism between I-thoughts and first-personal utterances, the well-documented asymmetry between first- and third-personal mental states is due to the peculiar way in which a subject is represented in her own thoughts in the first case...

The second option, whose origin could be traced back to Wittgenstein [1953] and Anscombe [1975], consists in an endeavor to explain the epistemic peculiarities of I-thoughts by the absence of self-representation. As Perry [1993/2000] states it, one's basic knowledge of one­self can be considered as "intrinsically selfless" ...representing oneself would be, in many cases, redundant and needlessly costly in terms of cognitive investment.

20171107: Roslin UK BioBank GeneATLAS



PheWAS for SYNE1:
Sitting height rs2635443 152520494 7.88719398273
Weight rs2635463 152517608 5.13783282866
fat-free mass rs2635463 152517608 7.99046412338
BMR rs2635463 152517608 7.38470240571
Standing height rs910415 152473181 9.3356871601
uterine problem rs11751190 152558197 10.3568237807
Endometriosis rs11751190 152558197 6.99576463364

20171106: McCain & Turner Causal Calculus

A non-monotonic logic for reasoning about action and change. Originally described in McCain & Turner [1997], Lifschitz [1997], Turner [1999]. In Bochman & Lifschitz [2015], Pearl's causal models are re-expressed in this formalism, which is supposed to allow prediction, abduction and transduction - interventions and counterfactuals - by giving rules to extract relevant submodels. This also corresponds to a modal logic, so that A causes B is equivalent to A implies B is necessarily so. Points out that this makes some of the assumptions interpreting SEMs clear:

The fact that Pearl's causal models are interpreted in this study as nonmonotonic causal theories allows us to clarify a large part of confusions surrounding a causal reading of structural equations. As we discussed in Section 3, the causal calculus is a two-layered construction. On the first level it has the underlying causal logic, a fully logical, though nonclassical, formalism of causal inference relations that has its own (possible worlds) logical semantics... Above this layer, however, the causal calculus includes a nonmonotonic "overhead" that is determined by the corresponding nonmonotonic semantics.

Then points out it is still grossly restricted compared to the usual uses of Pearl's models for causal inference.

White [2012] uses this type of approach to address the frame problem as well as the "grue" paradox of induction, arguing one reduces to the other.

20171106: Citing Shalizi and Moore [2003]

Papo [2015]:

[P]erceptuo-motor functions neural activity is shaped by the input's statistical properties, and processing is carried out at high speed in hardwired spatially segregated modules...[I]n reasoning, neural activity is driven by internal dynamics and processing times, stages, and functional brain geometry are largely unconstrained.

Event-related neural activity associated with the solution of riddles with insight was found to be related to properties of preceding resting activity (Kounios et al., 2006, 2008).

Wiesner et al [2011]:

We derive the minimum energy cost of simulating a complex data set with a stochastic computation. We show that it is proportional to the difference between the statistical complexity [Crutchfield & Young 1989] and the predictive information of the data ["the mutual information between the past and the future of a time series", Bialek et al 2001]. We derive this difference as the amount of information erased during the computation.

Walker [2017]:

Reductionism is what makes emergence possible: if we could not reduce reality to the study of a few component building blocks, we would not be able to describe how those building blocks come together to create more complex "higher-levels"

Recently, Marshall et al. (Marshall et. al., 2017) performed causal analysis on the fission yeast cell cycle regulatory network Boolean model using IIT. The analysis uncovered that the cell cycle in its primary attractor state is an integrated whole...The most irreducible structure (the global maxima) in the cause-effect structure includes all nodes of the cell cycle, exclusive of SK [Start Kinase, complex of Cdc2/Cig1/Cig2/Puc1 cyclins plus Cdc25], which is the start node and has no other feedback on the network. The analysis also revealed a local maxima of cause-effect structure, which corresponds to the set of nodes that when individually intervened on (by setting their state to their value in the primary attractor state at each step) expand the size of the primary basin of attraction and therefore increase its functional robustness. In other words, these nodes increase the goal-directed behavior of the network. The network was also shown to be integrated as a whole (all 8 nodes excepting the `start' node SK) through each state of the biological attractor. This network therefore provides an example of a system that maintains causal borders as an integrated entity in executing a goal-directed function (converging to an attractor).

20171103: Prediction of gene variant effects

Liu et al [2017]:

Toward this goal, computational approaches have been proposed to predict functionally damaging effects of genetic variants in the whole genome level [5, 6, 7]. For example, CADD [8] and GWAVA [9] integrated a number of genomic and epigenomic annotations to predict functional implications of all possible genetic variants in the human genome under the binary classification framework. Some methods mainly focus on variants occurring in such specific type of regulatory regions as transcription factor (TF) binding sites because it has been revealed that genetic variants occur in transcription factors binding sites can affect cellular phenotype and gene expression [10]. To mention a few, ChroMos, an integrated web-tool for SNPs classification and prioritization with the combination of genetic and epigenetic data [11]. HaploReg, another tool based on quantifying the difference between reference and alternate alleles in genetic context of canonical TF binding motifs [12]. DeepBind [13] identify sequence specificities of DNA- and RNA-binding proteins from experimental data by using the deep learning technology.

5. Kang HM, Sul JH, Service SK, Zaitlen NA, Kong S-y, Freimer NB, Sabatti C, Eskin E. Variance component model to account for sample structure in genome-wide association studies. Nat Genet. 2010;42:348¿54.

6. McCarthy MI, Abecasis GR, Cardon LR, Goldstein DB, Little J, Ioannidis JP, Hirschhorn JN. Genome-wide association studies for complex traits: consensus, uncertainty and challenges. Nat Rev Genet. 2008;9:356¿69.

7. Hirschhorn JN, Daly MJ. Genome-wide association studies for common diseases and complex traits. Nat Rev Genet. 2005;6:95-108.

8. Kircher M, Witten DM, Jain P, O'Roak BJ, Cooper GM, Shendure J. A general framework for estimating the relative pathogenicity of human genetic variants. Nat Genet. 2014;46:310-5.

9. Ritchie GR, Dunham I, Zeggini E, Flicek P. Functional annotation of noncoding sequence variants. Nat Methods. 2014;11:294-6.

10. Ward LD, Kellis M. HaploReg: a resource for exploring chromatin states, conservation, and regulatory motif alterations within sets of genetically linked variants. Nucleic Acids Res. 2012;40:D930-4.

11. Barenboim M, Manke T. ChroMoS: an integrated web tool for SNP classification, prioritization and functional interpretation. Bioinformatics. 2013;29:2197-8.

12. Ward LD, Kellis M. HaploReg v4: systematic mining of putative causal variants, cell types, regulators and target genes for human complex traits and disease. Nucleic Acids Res. 2016;44:D877-81.

13. Alipanahi B, Delong A, Weirauch MT, Frey BJ. Predicting the sequence specificities of DNA- and RNA-binding proteins by deep learning. Nat Biotechnol. 2015;33:831-8.

20171103: yardangs

Wind-sculpted (bedrock) pillars (etc)

20171101: Is-Ought gap

Woods and MacGuire [2017]:

Somewhat unfortunately, the simplest logical characterization of Hume's dictum - no descriptive proposition entails an ethical proposition - is demonstrably false. A.N. Prior (1960) showed this with several simple counterexamples. Let N be an ethical sentence, say, that it is impermissible to starve the Irish, and let M be a descriptive claim, say, that most people cannot digest grass. Presumably, ¬M is also descriptive. If N v M is descriptive, then the descriptive ¬M together with N v M entails the ethical N. If N v M is ethical, then it is entailed by the descriptive M. Assuming N v M is either ethical or descriptive, we have a decisive argument that the descriptive can entail the ethical. These entailments are impeccable; in the ordinary sense of entailment, we really do have a case of the ethical being entailed by the descriptive.

...One class of attempts treats the is-ought gap as an instance of a more general property of logical conservativeness - anytime we have a Prior-type entailment like the above, the conclusion is ethically irrelevant in a particular sense they define (Pigden 1989; Schurz 1997).

[Accepting these types of defences requires that one] deny there are conceptual or metaphysically necessary truths about the ethical.

Searle did not think there is an is-ought gap:

Suppose, following Searle (1964), that we can impose ethical demands on ourselves by uttering "I promise to give you 5 bucks" and this follows from facts analytic of the notion of promising. If Searle is right, we can analytically infer from the existence of a piece of descriptive behavior - an utterance of "I promise to give you five bucks" (in the appropriate circumstances) - to the existence of an ethical demand on ourselves.

...[P]romissory behavior, given the analytic connection between it and our obligations, is not norm invariant for Searle. The premises here - facts about speech behavior - thus count as ethical facts on [Daniel] Singer's taxonomy. Intuitively, this is incorrect; these are intuitively descriptive facts.

20171101: Rights again

John Danaher writes:

...the standard Hohfeldian account of rights - one that has been used for over 100 years. According to this account, rights claims - e.g. the claim that you have a right to privacy - can be broken down into a set of four possible incidents: (i) a privilege; (ii) a claim [about the duties of others]; (iii) a power [to waive]; and (iv) an immunity [to others waiving on your behalf].

20171028: Semantic maps on the cortex

Huth et al [2016] Natural speech reveals the semantic maps that tile human cerebral cortex

Interactive results

Here we systematically map semantic selectivity across the cortex using voxel-wise modelling of functional MRI (fMRI) data collected while subjects listened to hours of narrative stories

Deen et al [2017] "report that the visual cortex of 4-6 month-old infants contains regions that respond preferentially to abstract categories (faces and scenes), with a spatial organization similar to adults. However, precise response profiles and patterns of activity across multiple visual categories differ between infants and adults. These results demonstrate that the large-scale organization of category preferences in visual cortex is adult-like within a few months after birth."

Wang et al [2017] give 42 "neurally plausible semantic features" that they map fMRI BOLD at 600 voxels in the 5 seconds after 4 presentations of 240 sentences:
Visual perception
Positive affective valence
Negative affective valence
High affective arousal
Change of location
Arts and literature
Social norms
Transfer of possession
Social interaction
Social support
Physical action
Change of physical state
Physical impact
Mental action
Perceptual action

NPSF features performed reliably better than any of ... features derived from word co-occurrence norms in text corpora: GloVe vectors for word representation [Pennington et al., 2014], a documents-based and a dependencies- based vector space model (doc VSM and dep VSM) [Fyshe et al., 2013], and Latent Semantic Analysis [Landauer and Dumais, 1997].

...the model can predict the activation pattern of a sentence that is new to the model with a mean rank accuracy of 0.86 among 240 alternatives, where chance is 0.50...

...the sentence prediction accuracies are primarily based on the mapping between brain activation patterns and semantic and thematic features, rather than on the mapping to observed activation patterns of particular word concepts. In addition, this analysis establishes the model's ability to predict activation for the previously unseen words.

20171028: On words and pictures

Lovely post at philpercs from Axel Barcelo:

These are some of the many ways philosophers have proposed for distinguishing (at least some sorts of) pictures from words just in the last few decades:

  1. By their content: Words have conceptual content, Pictures have non-conceptual content
  2. By their persuasive force: Words are Apollinean, Pictures are Dionysian
  3. By their structure: Words have a recursive syntatcic/semantic structure, Pictures are dense
  4. By what makes them representations: Words belong to languages, Pictures are autonomous depictions
  5. By how they are related to what they represent: Words are artificially related to what they represent, Pictures are naturally related to what they represent
  6. By how we grasp their content: The content of words is interpreted according to linguistic conventions plus contextual information, the content of pictures is seen in them
  7. By their modality: Pictures are visual, Words need not be
  8. By their phenomenology: Seeing a picture of X feels similar to seeing or being in the presence of X itself; reading a word meaning X, less so.

20171026: EPQ genetics

Okbay et al [2016] meta-analysed neuroticism (N=170910) from UK BioBank (12 item EPQ N), GERA and Genetics of Personality Consortium. The SNP heritability of N was 9.1%

Docherty et al [2016] give the SNP heritability of N in 10064 Han Chinese as 10% (CONVERGE Study).

20171024: ASPM

ASPM and MCPH1 were both fingered as being involved in "ongoing adaptive evolution" by Evans et al [2005].

ASPM mutations cause primary AR microcephaly. Passemard et al [2016] report the volume of the cortices is reduced by 50%, although the hippocampus and medial temporal regions are relatively spared (with OK memory). In the mouse [Williams et al 2015], Aspm is expressed in cerebellar granular neuron progenitors (under Shh control). In humans, ASPM regulates WNT.

Wong et al [2012] mention that there is no association between rs41310927 (Ser2562Gly) in ASPM or rs930557 (Asp314His) in MCPH1 with brain size or performance eg Mekel-Bobrov et al [2009] (Martin, Boomsma).

Montgomery et al [2011] report evidence of selection for ASPM and brain size (especially neonatal) across 21 primate species (also regressed root-to-tip dN/dS and phenotypes).

20171024: The cognitive niche and the social brain

Pinker [2010] on the cognitive niche [Tooby and DeVore 1987], and why analogical thinking is so useful.

Dunbar and Shultz [2017] discuss the large brains of primates (commenting that total brain size seems the best phenotype), and list 9 hypotheses:


Big brains allow one to exploit the environment more efficiently.


In primates [only], social group size [esp number of females] correlates with a wide range of brain indices, including absolute and relative cranial volume, brain volume, neocortex volume, non-striate neocortex volume and frontal lobe volume...

[In] humans and macaques...individual differences in social network size (variously indexed as sociability, core social network size, number of Facebook friends and living group size) correlate with the absolute volume of core regions in the frontal and temporal lobes of the brain...

Suggests predation risk is the main driver for group size ("group size is adjusted to the level of experienced predation risk" [Bettridge et al 2010]), and that bigger brains may merely be required because they allow large groups to form. The large brains require more calories, and so intelligent foraging is then selected for, but as exaptation:

What evidence is there to suggest that ecological or technical decision-making has actual fitness consequences? Only one study has assessed this directly: Altmann [1991] showed that female baboons who were better able to match an optimal diet (in terms of energy and protein intake) as yearlings survived longer, had longer reproductive careers and produced more offspring (whether indexed as total number of births or the number that survived to 12 months). Although the sample size is small (n = 6 females), the results are remarkably linear and convincing. Large brains certainly provide the capacity to engage in efficient trial-and-error problem-solving or insightful one-trial learning.

Johnson [2002] find among female baboons that "the benefit of rank to reproductive success shown...is 0.83 additional offspring." Clutton-Brock and Huchard [2013] review female social competition more generally, especially in matrilineal social animals (infanticide to reduce opposing matriline sizes etc). Female-female competition more likely to be resource based, as opposed to mate-availability based.

Lewis and Barton [2004] report an (unsurprising) correlation between brain size and frequency of play across primate species. Allomaternal care, where juvenile females help care for others' offspring is correlated with later reproductive success.

Potts et al [2015] compare two chimpanzee communities in Kibale National Park...mean foraging efficiency was higher in the larger, denser group, and variance lower.

Hogberg & Gardenfors [2015] Knowing, learning and teaching - How homo became docens

20171024: Examples of gene-culture coevolution

Hünemeier et al [2012]

[In Native Americans], two more recently reported examples might be associated with positive selection: (1) Tovo-Rodrigues et al. investigated the distribution of D4 dopamine receptor (DRD4) alleles in several South Amerindian populations and found a significant difference in the allelic distributions between hunter-gatherers and agriculturalists, with an increase of the 7R allele among the former; and (2) Acuña-Alonzo et al. showed that the 230Cys allele (Arg230Cys, rs9282541) of the ATP-binding cassette transporter A1 (ABCA1) gene, which was previously associated with low HDL-cholesterol levels and obesity-related comorbidities, was exclusively present in Native American and mestizo individuals. These authors verified that cells expressing the ABCA1*230Cys allele showed a 27% cholesterol efflux reduction, confirming that this Native American autochthonous variant has a functional effect in vitro. Other investigations have shown that the presence of ABCA1*230Cys explains almost 4% of the variation in plasma HDL-C concentrations in Mexican admixed populations [23]. This variation in HDL-C concentration was the highest one associated with a single nucleotide polymorphism (SNP) among different continental populations in these genome-wide association studies, corroborating its functionality...230Cys resides in a haplotype that was the target of an ongoing directional selective sweep...

A significant correlation was observed between the ABCA1*230Cys allele frequencies and the distribution of the Zea pollen relics in Mesoamerica...

Peng et al. presented evidence for a similar case of gene-culture coevolution, suggesting that positive selection for the ADH1B*47His allele was caused by the emergence and expansion of rice domestication in East Asia.

Mutation age ~6000 BCE as estimated by LD methods, so coincides with cultivation of maize. However ABCA1 alleles may be selected on other phenotypes, such as infection resistance (dengue, malaria). Note also that many other variants affect CVD risk in other populations.

20171023: Kwong and Pemberton [2014]

Kwong and Pemberton [2014] describe "a dataset of 5,795 human and 84 chimpanzee individuals with genotypes at 246 human-derived autosomal microsatellites". Here I plot the first 3 principal components from their data:

And here are mean Fst's as a measure of interpopulation distances. Like any other correlation-based measure, the scaling depend upon the variation across the individuals sampled, which is why being able to include comparable markers for the different species is so useful.

AfricaAsia 0.03
AfricaChimp 0.16
EuropeAsia 0.01
EuropeChimp 0.18
Asia Chimp 0.18

Bonobo Central 0.17
Bonobo Eastern 0.16
Bonobo Hybrid 0.22
Bonobo Western 0.16
Bonobo Unreported0.21
CentralEastern 0.04
CentralHybrid 0.11
CentralWestern 0.01
EasternHybrid 0.10
EasternWestern 0.03
Hybrid Western 0.06
Hybrid Unreported0.00

There are lots of peculiarities of the mutation mechanism for microsatellites that make interpretation of this less than straightforward. However the broad picture fits in with other types of sequence data. My main interest was how the distances between modern human populations compare to those between humans and other primates, as well as those between other primate species. Usually, these kinds of raw data are converted into divergence times or phylogenies in publications.

20171019: Phoresy

That form of commensalism in which one organism uses another for the
purpose of transport.

20171016: Abduction

From Bateson, G., and Bateson, M. C. (2004). Angels fear: Towards an epistemology of the sacred. New York: Hampton Press.

1. Men are mortal,
2. Grass is mortal,
3. Men are grass.

20171004: Pigs

Need certain amount of water (300 mm rain), not good for nomads, compete for same foods as humans so less attractive when high intensity agriculture, breed fast, less efficient than cattle if add milk protein, leather, but Egyptians did not use milk, better suited to local consumption versus tithing by authorities. Philistines were pork eaters but temporal variability, Greeks regarded pigs as unsuitable for sacrifice, Egyptian consumption fell off into New Kingdom era, Jews avoided especially after Babylonian exile (not eaten in Judah). Redding [1991] thinks the chicken outcompeted the pig for that niche in the subsistence economy in Egypt and the Levant. Pig, donkey and hippo are associated with Seth in Greco-Roman Egypt, when became more frequently consumed again.

Sapir-Hen et al [2013]:

The biblical authors insisted that all Hebrews who dwelt in the territories of both, Israel and Judah, must acknowledge the rule of the Davidic Dynasty and worship in a sole Temple in Jerusalem. The pig taboo could have been another Judahite cultural trait that was opposed to the situation in the north, and which the authors wished to impose on the entire Israelite population.

20170929: DNA repair


Even the FANC's go from FANCA to FANCP, as well as FANCA-associated FAAP20 and FAAP24.

20170927: Maximin, Leximin

Ravallion [2015]:

...[O]ne can defend Rawls's difference principle without accepting his rationale in terms of a social contract formed behind the veil of ignorance. Hammond (1976) showed that a generalized lexicographic version of maximin, dubbed leximin in the literature, can be derived from a set of axioms including a requirement that reducing the disparities in welfare between the rich and the poor is socially preferred, other things being equal. Similarly, Fleurbaey and Maniquet (2011, Chapter 3) showed that leximin is implied by what they termed the "priority among equals" axiom. Again this requires that more equitable allocations are socially preferred but that (echoing Rawls) this never trumps efficiency in the sense that a situation in which everyone is better off is always preferred. Roemer (2014) argued for leximin but from a somewhat different starting point, namely the desire to equalize opportunities.

Fleurbaey M, Maniquet F [2011]. A Theory of Fairness and Social Welfare. Cambridge: Cambridge University Press.

Hammond P [1976]. Equity, Arrow's Conditions and Rawls' Difference Principle Econometrica 44: 793-804.

Roemer J [2014]. Economic Development as Opportunity Equalization. World Bank Economic Review 28(2): 189-209.

Kameda T, et al. [2016]. Maximin rule as a common cognitive anchor in distributive justice and risky decisions: Rawls in our minds. Proc Natl Acad Sci USA 113:11817-11822.

Takesue et al [2017] . Human pursuance of equality hinges on mental processes of projecting oneself into the perspectives of others and into future situations. Sci Rep. 2017; 7: 5878.

Also, compare Parfitt's prioritarianism where "weight assigned to each potential benefit also reflects how much worse off its beneficiary is".

20170927: Water evaporation as a renewable energy resource

Cavusoglu et al [2017]:

[W]ater-responsive materials can be incorporated into evaporation-driven engines that harness energy when placed above a body of evaporating water ...through a 4 stage cycle: (I) With the upper shutters closed, the water-responsive material swells, absorbing water vapor at the high chemical potential us. (II) At maximum absorption, the upper shutters open as the bottom shutters close. (III) With the upper shutter open, the water-responsive material shrinks, releasing water that evaporates away into the atmosphere at a lower chemical potential (IV) At maximum desorption, the upper shutters close as the bottom shutters open, restarting the engine cycle.

When restricted to existing lakes and reservoirs larger than 0.1 km2 in the contiguous United States (excluding the Great Lakes), we estimate the total power available to be up to 325 GW, which is over 69% of the US electrical energy generation rate in 2015.

Since thermal storage by water, can match output to peak electricity usage times. Also, controls evaporative losses. Note that 400 GL/yr of fresh water is evaporated by Australian coal-fired power plant cooling towers.

Staedter [2017] article:

Ozgur Sahin, a biophysicist at Columbia who heads the team and was senior author on the paper. In 2015, Sahin and his colleagues unveiled an evaporation engine they created by depositing bacteria spores onto thin plastic films and layering the films into a palm-sized stack. [ Chen et al, 2015].

The films behaved like artificial muscles that flex in response to fluctuations in the amount of moisture in the air. When the air was humid, the spores expanded, causing the films to straighten; when the air was dry, the spores contracted, causing them to bend. In experiments, the movement generated enough electricity to power an LED light bulb.

20170916: autosfruttamento

Is self-exploitation.

What's Love Got to Do with it? Precarious Academic Labour Forces and the Role of Passion in Italian Universities is an interesting article on post-GFC Italian academia.

20170825: Nonsurgical treatment of appendicitis

Talan et al [2017]

16 (53.3%) were randomized to antibiotics-first and 14 (46.7%) to appendectomy. Median age was 33 years (range 9 to 73 years), median WBC count was 15,000/uL (range 6,200 to 23,100/uL), and median computed tomography appendiceal diameter was 10 mm (range 7 to 18 mm). Of 15 antibiotic-treated adults, 14 (93.3%) were discharged from the ED and all had symptom resolution. At 1 month, major complications occurred in 2 appendectomy participants (14.3%; 95% confidence interval [CI] 1.8% to 42.8%) and 1 antibiotics-first participant (6.3%; 95% CI 0.2% to 30.2%). Antibiotics-first participants had less total hospital time than appendectomy participants, 16.2 versus 42.1 hours, respectively. Antibiotics-first-treated participants had less pain and disability. During median 12-month follow-up, 2 of 15 antibiotics-first-treated participants (13.3%; 95% CI 3.7% to 37.9%) developed appendicitis and 1 was treated successfully with antibiotics; 1 had appendectomy. No more major complications occurred in either group.
Georgiou et al [2017]:

Ten articles reporting 413 children receiving non-operative treatment [NOT] were included. Six, including 1 randomized controlled trial, compared NOT with appendectomy. The remaining 4 reported outcomes of children receiving NOT without a comparison group. NOT was effective as the initial treatment in 97% of children (95% confidence interval [CI] 96% to 99%). Initial length of hospital stay was shorter in children treated with appendectomy compared with NOT (mean difference 0.5 days [95% CI 0.2 to 0.8]; P = .002). At final reported follow-up (range 8 weeks to 4 years), NOT remained effective (no appendectomy performed) in 82% of children (95% CI 77% to 87%). Recurrent appendicitis occurred in 14% (95% CI 7% to 21%). Complications and total length of hospital stay during follow-up were similar for NOT and appendectomy. No serious adverse events related to NOT were reported.

20170825: Tasters

TAS2R38 contains three coding SNPs accounting for 85% of the bimodal taste response (PROP, PTC).

rs713598 A49P 0.41
rs1726866 A262V0.55

The P-A-V haplotype (US frequency 0.42) is the taster allele (dominant over A-V-I, frequency 0.53).

20170820: Minyon Falls ride

Circuit Federal-Eureka-Minyon-Federal (44 km, ~1000 m climbing).

20170816: CI Lewis An analysis of knowledge and valuation [1946]

Bunge cites this for his comment that:

Evaluation may involve emotion - but so does cognition. Evaluation is a kind of cognition (Lewis 1946). A rat that presses a lever for more food pellets of a certain kind knows that the food is good for him - but he will revise this evaluation if the last time he sampled food laced with poison. A person who has learned that keeping promises is necessary to stay in good social standing knows something about certain values, and she has come to know it on the strength of experience and reason. And a person who realizes that she has made mistakes in value attributions will correct the latter much in the same way as she corrects perceptual or calculation errors.

Value judgments can be justified or criticized, rather than accepted or rejected dogmatically, when they are rooted to basic needs or legitimate wants. In this case they can be shown to be true or false.

Hempel reviews the the first two volumes of this 3 volume work (Volume 3 is "Valuation").

Every term has 4 modes of meaning: Denotation/extension (class of all actual things to which the term applies); Comprehension (all non-actual but consistently thinkable things to which the term would apply; each of the latter is said to be named by the term); Signification (that characteristic of a thing that allows the term to be correctly applied to it); Connotation/Intension (a term B is connoted if it automatically would apply if term A is usable about a thing, eg man, mammal).

An analytic proposition is one applicable to every possible world, so universal comprehension and zero connotation [?]. All a priori truths are analytic, and the relations between meanings for these are universal, and not convention-dependent.

Sense meaning is central to the discussion of empirical knowledge: the sense meaning of a verifiable statement of objective fact is exhibitable in an inexhaustible set of judgments (couched in an expressive language) that doing A always results in B. This is a terminating judgment that predicts not only if-then but the outcomes of counterfactuals, but is not logical per se but closer to natural law. The more usual non-terminating judgments are probabilistic.

Added 20171114:

David Kaplan takes up Frege's denotation and sense, splitting the latter into content (proposition or intension) and character, which is the property of the term that determines how it interacts with the context of the utterance eg "character of 'I' would then be represented by the function (or rule, if you prefer) which assigns to each context that content which is represented by the constant function from possible worlds to the agent of the context" [Kaplan 1990].

Added 20180111:

Bertolet [2008] discussing meaning in the context of science:

...[Hilary] Putnam offered a multi-faceted picture of what the meaning of a natural kind term involves, suggesting that a type of "normal form for the description of meaning" for the term "water" would include at least syntactic markers (mass noun, concrete), semantic markers (natural kind, liquid), stereotype (colorless, transparent, tasteless, thirst­quenching ...), and extension (H2O give or take impurities). Moreover, he conjectured that these components, except for the extension, are part of the competence of the individual speaker. So competent speakers need to know that "water" picks out a natural kind in liquid form that is transparent, colorless, odorless, and so forth, though they do not need to know that what it picks out is H2O (as we did not in 1750).

In this section I consider two problems that are prominent in the literature, and then a complete rejection of the Kripke-Putnam position...reference of "tiger" might be fixed by appeal to something being of the same biological kind as a passing tiger. But a passing tiger is a passing cat and a passing mammal and a passing animal: which level counts? This is the qua problem: if a term is bestowed, qua what - tiger, mammal - is that thing construed?

The Kripkean "rigid designator" also fails for this example, given that "tiger" can have different extension in different possible worlds. In "two dimensionalism" [Frank Jackson], there are primary and secondary intensions, where the first is our personal acquaintance, and secondary intension is the scientific or "actual" reference, which makes it a rigid designator, and a posteriori analytic and contingent.

20170815: names of monoclonal antibodies

Fudin [2015]

Due to International Nonproprietary Names Working Group (WHO).

prefix unconstrained
target -ami- amyloid
-b(a)- bacterial
-c(i)- cardiovascular
-f(u)- fungal
-k(i)- interleukin
-l(i)- immunomodulating
-n(e)- neural
-(o)s- bone
-tox(a)- toxin
-t(u)- or -ta- tumour
-vet- veterinary use
-v(i)- viral
origin a rat
axo rat/mouse
e hamster
i primate
o mouse
u human
xi chimeric
xizu chimeric/humanized (<85% human sequence)
zu humanized (≥85% human sequence)
suffix mab [pab=polyclonal]


mepo - li - zu - mab is a humanized anti-IL5.

20170725: WD Ross

From Cowan's discussion of Rossian Conceptual Intuitionism [ 2017]. Other flavours are Empirical and Rational Intuitionism, and Perceptualism, an account of how self-evident the principles actually are.

A plurality of fundamental principles, each of which are overridable but are ineradicable,:

fidelity, reparation, gratitude, beneficence or benevolence, justice and self-improvement, and noninjury. (Audi adds liberty, respectfulness).

[A] self-evident proposition, p, is a truth such that adequate understanding of p entails powerful (but defeasible) noninferential justification for believing it (even if it isn't believed); if p is believed wholly on the basis of adequate understanding, then it is noninferentially known.

If it known, then it can be applied, used for inference, and differentiated from other concepts. Cowan also spends time on nonsubstantive (roughly, analytic) propositions eg "murder is wrongful killing" comparing these to substantive (roughly, synthetic) propositions eg "less populated happy worlds are worse than overpopulated less-happy worlds". The former are often thought of more accessible to intuition.

20170724: Simon Blackburn on Kant

From a review of a biography:

Surprisingly, in spite of such unpromising doctrines, Kant is undoubtedly the most influential moral and political philosopher of modern times. At present he probably has more, and more crusading, defenders among professional moral and political philosophers than ever before. Partly, he is a foil to "utilitarianism", which is equated in many minds with a fearsome social engineering that puts the individual firmly at the service of the collective. Better still, he directs attention away from any very demanding educational needs. So, according to the Greek tradition virtue is rare and requires the most careful cultivation and practise. Furthermore, for the Greeks, democracy requires virtuous citizens. For Kant on the other hand, people have the possibility of autonomy, or freedom, and above all they deserve respect, just like that. They do not have to work to earn respect. And however dim or dumb they may be, democratic republicanism is the right form of government.

20170724: Cellular gateway

Cellular Bluetooth Gateway
Connect/Link to cell phone system


PhoneLabs Dock-N-Talk
Xtreme Tech XLINK cellular bluetooth gateway (2007)
Siemens Gigaset One Bluetooth Gateway (rebadged XLINK)
VTech Connect-to-Cell Cordless Phone System (2011)
Panasonic DECT 6.0 PLUS Link-to-Cell Bluetooth Cordless Phone

20170724: MC1R

Chen et al [2017]:

MC1Re/e mice, which carry an inactivating mutation of MC1R and mimic the human redhead phenotype, have compromised nigrostriatal dopaminergic neuronal integrity, and they are more susceptible to dopaminergic neuron toxins 6-hydroxydopamine and 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP). Furthermore, a selective MC1R agonist protects against MPTP-induced dopaminergic neurotoxicity.

Nick Hayward [ Johannson et al 2017]:

We observed statistically significant associations between mutation burden in melanoma and body site, age at onset and MC1R genotype, for both ultraviolet radiation (UVR) signature changes (C>T and CC>TT) and non-UVR base pair substitutions, as well as with overall variant load.

D'Orazio [ Horrell et al 2017]:

Loss-of-function melanocortin 1 receptor (MC1R) polymorphisms are common in UV-sensitive fair-skinned individuals and are associated with blunted cAMP second messenger signalling and higher lifetime risk of melanoma because of diminished ability of melanocytes to cope with UV damage. cAMP signalling positions melanocytes to resist UV injury by upregulating synthesis of UV-blocking eumelanin pigment and by enhancing the repair of UV-induced DNA damage. cAMP enhances melanocyte nucleotide excision repair (NER), the genome maintenance pathway responsible for the removal of mutagenic UV photolesions, through cAMP-activated protein kinase (protein kinase A)-mediated phosphorylation of the ataxia telangiectasia-mutated and Rad3-related (ATR) protein on the S435 residue. We investigated the interdependence of cAMP-mediated melanin upregulation and cAMP-enhanced DNA repair in primary human melanocytes and a melanoma cell line. We observed that the ATR-dependent molecular pathway linking cAMP signalling to the NER pathway is independent of MITF activation. Similarly, cAMP-mediated upregulation of pigment synthesis is independent of ATR, suggesting that the key molecular events driving MC1R-mediated enhancement of genome maintenance (eg PKA-mediated phosphorylation of ATR) and MC1R-induced pigment induction (eg MITF activation) are distinct.

20170719: Genomics MOOCs from HarvardX

From Stuart McGregor: http://rafalab.github.io/pages/harvardx.html

And from there https://kasperdanielhansen.github.io/genbioconductor

20170710: DNA polymerase mutations and cancer risk

POLD1 and POLE mutations of the exonuclease (ie proofreading) domain lead to "a dominantly inherited syndrome that confers increased risk to colorectal cancer (CRC) and polyposis" (polymerase proofreading-associated polyposis, PPAP). Aoude et al [2016] found a POLE mutation segregating in one Queensland melanoma kindred. The pathogenic mutations are (the exonuclease domain of POLE is R311--L526):

Coding Changersnumcancer
POLE R446Q rs151273553CRC (exm1053970)
POLE A895T rs201115064CRC (exm1053889)

Likely disease-associated variants in POLD2, POLD3, POLD4, POLE2, POLE3, POLE4 have also been detected. Note that CRC with somatic POLE mutations are immunogenic with a good prognosis.

20170709: Redcliffe train line bike path

Left train station car park ~12:40
Scarborough waterfront near Newport Drive ~13:40
Mermaids by The Bay, Deception Bay ~15:00
Returned station car park ~17:00

Rode back through North Lakes
- very upmarket bike paths along golf course.

Just before North Lakes proper,
a suburban house with 4-5 peacocks!

Map annotated using MilerMeter

Had to use the "manual" option, as default "runner" option
followed some roads unsuccessfully.

20170704: Conversation article on electric guitar

Australian "new music specialist" Zane Banks - seems to also have a bent for C&W.

Glenn Branca - "Hallucination City: Symphony for 100 Guitars", "Lesson No. 1 for Electric Guitar" [1981] and other symphonic pieces; "no wave", "post-minimalist totalism" and "noise music", the band Theoretical Girls. "Lesson No. 1" very King Crimson slash Reich, Branca is "known for his use of volume, alternative guitar tunings, repetition, droning, and the harmonic series."

Danielle Haim (Haim); St Vincent; Orianthi Panagaris

King Gizzard and the Lizard Wizard (Australian psychedelic band).

Takaakira "Taka" Goto from Japanese band Mono; Markus Reuter

Jazz (Australia): Steve Magnusson and James Muller

Internationals: John Scofield, Bill Frisell, Ben Monder, Julian Lage, Kurt Rosenwinkel, Ralph Towner, John Abercrombie, Ulf Wakenius

A list (that includes many who toured with G3) from musicradar.com:

Al Di Meola, John Petrucci, Steve Vai, Ron Thal, Paul Gilbert, John McLaughlin, Yngwie Malmsteen, Allan Holdsworth, Zakk Wylde, Steve Morse, Jason Becker, Mattias Eklundh, Jeff Loomis, Shawn Lane, Marty Friedman, Buckethead (Brian Carroll), Rusty Cooley, Fredrik Thordendal, Michael Angelo Batio, Joe Satriani.

20170702: Mt Nebo stroll

Down Double Break from Forest Rd, and along pad through lawyer vine to "big strangler fig" from OSM (not that impressive compared to elsewhere in park).

20170626: Alex Callinicos

Alex Callinicos, a leader of the Socialist Workers Party in Britain,
Professor of European Studies at KCL.

20170623: A nice formula for recurrence risk ratio

Sun et al [2016]:

λ = (p*r2 + q)/(p*r + q)2

r is the per-allele risk ratio.

20170622: Doxastic norms

These are norms pertaining to what to believe (Feldman 1988; Miller 1995). Doxastic norms may be moral, rational or prudential, or epistemic. In the case of this application, questions of whether there are doxastic norms loom as large as those of what the content of particular norms ought be. Many epistemologists have thought there cannot be any doxastic norms because we do not or cannot have voluntary control over what we believe (Alston 1989). But this contradicts the social facts as Social Epistemology (Coady 1992; Goldman 1999) or an Economic Theory of Knowledge (Hardin 1997; 2002) make clear.

20170615: Phenomenology

Review of Phenomenology and Naturalism: Examining the Relationship between Human Experience and Nature

..."phenomenology," as understood in the volume, refers to the style and method of doing philosophy that originated with Edmund Husserl and was carried forward in different ways by Martin Heidegger, Jean-Paul Sartre, and Maurice Merleau-Ponty

Evan Thompson in Mind in life: Biology, phenomenology, and the sciences of mind [2010] has it that "life has its own interiority...mind is not pure interiority, but rather a form of structure of engagement with the world...experience is not an epiphenomenal side issue...". Thompson is an enactivist. Even when not attending, prereflective self-consciousness is active: sentience. So, after Heidegger and Merleau-Ponty, intentionality is bound up with self-consciousness as being-in-the-world. Thompson is also thus a neurophenomenalist, an approach founded by Francisco Varela [eg 1991, 1996]. This may or may not gel with his thinking about the selfhood of autopoietic systems.

Thompson [2004]:

Phenomenality corresponds not to qualitative properties of consciousness (qualia), but the qualitative appearance of the world for consciousness. As Merleau-Ponty puts it, qualia are not elements of consciousness, but qualities for consciousness...although any experience must always be given from a first-person perspective (otherwise it is not an experience), it can also be given from a second-person perspective in empathy. To deny this claim commits one to the view that we never experience the thoughts and feelings of another person in any sense, but can only infer their existence from physical behaviour. This view involves a highly problematic dichotomy between inner experience and outer behaviour. In face-to-face encounters, we are confronted neither with a mere physical body, not with a hidden psyche, but with a unified intentional whole.

...Experience involves not simply awareness of its object (noema), but tacit awareness of itself as process (noesis) [Husserlian terms]. In consciously seeing an object, one is at the same time tacitly and pre-reflectively aware of one's seeing. In visualizing a mental image, one is at the same time tacitly and pre-reflectively aware of one's visualizing... The roots of such pre-reflective bodily self-awareness sink to the involuntary and pre-cognitive level of operative intentionality [Merleau-Ponty], the dynamic interplay of affection and receptivity...

Welton in Bodily Intentionality, Affectivity, and Basic Affects, distinguishes intention-of-action and intention-in-action [ Kohler 2011]. Kohler is interested in Schwartz's CBT approach to OCD - how (unsurprisingly) brain PET responses to the compulsion-associated stimulus alters, and that it follows reflective re-valuation, which he interprets in a existentialist/phenomenological framework.

...Schwartz makes a distinction between the form and content of OCD. The form of OCD is what makes it a neuropsychiatric problem; the individual is overcome with an intrusion of obsessive-compulsive thoughts originating from the biological malfunctioning of the brain. The content, or why the individual has one symptom over another (i.e., the continual urge to wash one's hands as opposed to checking if the appliances are all turned off) has no biological explanation. Schwartz here is concerned with the form of OCD, which is biologically explained, and with his Four-Step Process he claims that one is able to manage and even overcome the obsessive-compulsive urges despite the content. In fact, Schwartz claims that understanding the psychological and emotional content will rarely make the obsessive-compulsive urges go away (Schwartz 1996, xxxi).

20170609: Framingham eQTL

Joehanes et al [2017] The Molecular QTL Browser allows access to their (and GTEx results).

20170607: Iterative Tower of Hanoi

Donald Arseneau on c.l.f:

1) move the smallest disk to the "next" peg (consistently clockwise, counter-clockwise or whatever suits your layout). In fortran p = modulo(p+1,3). 2) move (legally) another disk. repeat.

20170604: Consciousness

Boley et al [2013] Consciousness in humans and non-human animals: recent advances and future directions.

More obliquely, Patchev [2014] describes mouse models of PTSD and its effects on sleep. The Threat Simulation Theory (TST) of sleep [Revonsuo 2001] is the suggestion for another function of sleep. Oneirism is dream enactment, as in the classical cat peri-locus coeruleus alpha nucleus ablation model, but also as in human REM sleep behaviour disorder (eg as in prodromal alpha-nucleinopathies - Lewy Body Dementia etc), where one sees "rapid eye movement [or paradoxical] sleep without atonia" associated often with aggressive behaviours.

Satre and Jouvet [1979]:

Le répertoire des comportements oniriques correspond aux comportements présentés par l'animal vivant dans son milieu naturel (aggression prédatrice, attaque aggressive, colère, peur, échage et exploration). Le comportement complexe le plus fréquemment observé est l'attaque. Le comportement sexuel, en revanche, n'a jamais été observé. Les stimulations visuelles et tactiles ne modifient pas le déroulement des comportements oniriques.

Soh et al [1992] comment:

Most isolated REMs were related to orienting behavior, whereas most REM bursts were related to generalized body movements (jumping, attacking, etc.). Only isolated, high amplitude REMs had any possibility of corresponding to visual images in dreams.

And recall that there is "no effective thermoregulation in paradoxical sleep (PS), either to ambient or hypothalamic thermal stimuli" - part of Friston's thinking.

20170603: FAPI

FAPI: Fast and Accurate P-value Imputation for genetic association

20170603: Randomized PCA etc

Many recent papers on accurate randomized algorithms for matrix approximation (and completion): randomized numerical linear algebra. A low rank approximation of a larger matrix is a "sketch", see review by Woodruff [2011]. For example, Halko et al [2011]:

we compute PCAs of very large data sets via a randomized version of the block Lanczos method...proofs...show that this method requires only a couple of iterations to produce nearly optimal accuracy, with overwhelmingly high probability (the probability is independent of the data being analyzed, and is typically 1 - 10-15 or greater). The randomized algorithm has many advantages...the present article adapts the algorithm for use with data sets that are too large to be stored in the random-access memory (RAM) of a typical computer system.

Szlam et al [2014] provides a Fortran library (ID).

Erichson et al [2016] provides the R rsvd package based on Halko et al [2011].

Abraham and Inouye [2014] use a randomized approach to genetic PCA in flashpca and flashpcaR. This also implements univariate canonical correlation analysis and sparse canonical correlation analysis for allelic association to multiple phenotypes.

20170527: Kant and feminism


...in the essay "On the Common Saying: 'This May be True in Theory, but it Does not Apply in Practice,'" Kant states that the only requirement for being a citizen is "being one's own master," only to add, as if it were a trivial afterthought: "apart from the natural [requirement] (that one [es] is not a child, not a woman)"

Kant seems to think that women are incapable of Enlightenment, for God-given biological reasons (eg they lack the necessary virtue of courage).

20170525: Popular contemporary European art music

(mostly late works) by such composers as Preisner, Gorecki and Pärt, not to mention such Swedish representatives as Sven-David Sandström and Jan Sandström...Schnittke

20170524: relativistic thermodynamics

Requardt [2008]:

Entropy and pressure are invariants under Lorentz transformations.

The first and second law of thermodynamics to hold also for moving equilibrium systems. However, work is more complicated, as the volume of the system undergoes Lorentzian contraction.

Requardt [2013] extends this to "prove the reality" of Unruh radiation, using a model akin to the moving mirrors thought experiment: "It can be shown that mirrors which undergo a changing acceleration radiate. In particular, in the direction of increasing acceleration there is a negative flow of energy wh ile in the direction of decreasing acceleration there is a positive energy flow...".

20170524: The Hole problem in relativity - philosophy

John Stachel on some relevant philosophical concepts:

Quiddity: What characterizes all entities having the same nature, i.e., the same intrinsic properties.

Example: All electrons have the same quiddity, as do all goats.

Haecceity: What individuates entities of the same quiddity.

Example: Position of an electron.

We can now reformulate the earlier observation: Up until the last century, it was assumed that each entity of some quiddity always had an intrinsic haecceity; i.e., it could always be individuated independently of any relations, into which it entered. This assumption is often referred to as Leibniz's principle:

"For there are never two things in nature that are perfectly alike and in which it is impossible to find a difference that is internal, or founded on an intrinsic denomination" (G.W. Leibniz, The Monadology in Loemker, 1969).

However, the advent of quantum statistics led to the recognition that elementary particles indeed have quiddity but no inherent haecceity. As noted above, every electron has the same mass, charge and spin, which fix its quiddity; but an individual electron is only singled out (to the extent that can be) by its relations to other entities, such as its effect on the emulsion of a photographic plate.

Einstein's way of avoiding the hole argument can be similarly formulated: The points of spacetime have the same quiddity, but no intrinsic haecceity. A space-time point is only singled out (to the extent that can be) by the unique physical properties of the fields at that point.

So both relativity and quantum theory lead to the same conclusion: Leibniz's principle is not universally applicable. There is a category of entities with quiddity but no inherent haecceity. Given that both general relativity and quantum mechanics are based on such entities, it is difficult to believe that, in any theory purporting to underlie both relativity and quantum theory, inherent individuality would re-emerge in its fundamental entities, whatever they are.

Given the Minkowski metric and its associated flat inertial connection, such a[n inertial reference] frame may be defined by taking any time-like autoparallel ('straight') line, and constructing the family of such lines, one through each point of the manifold (i.e., a fibration of the space-time), each of which is parallel to the initial line.

20170517: c22orf9

Aliases are LSC3 and KIAA0930, and the protein is Q6ICG6. Protein is detectable in melanocytes. Interacts with tyrosine 3-monooxygenase activating proteins (YWHAB, YWHAH, YWHAZ etc), which are members of the 14-3-3 protein family, mediating signal transduction by binding to phosphoserine-containing proteins.

Cornell and Toyo-oka [2016] produced knock-out mice for YWHAZ (and YWHAE):

14-3-3epsfl/fl/zeta+/+/Cre+ mice did show white patches on their ventral region the 14-3-3eps+/fl/zeta/¿/Cre+ mice, and the 14-3-3epsfl/fl/zeta+/-/Cre+ mice have larger white patches than the 14-3-3epsfl/fl/zeta+/+/Cre+ mice. This indicates that neither14-3-3¿ or 14-3-3¿ is dominant in regulating the size of the white patches. Together, these data suggest the importance of the 14-3-3 proteins in melanocyte development.

20170516: Extending the evolutionary synthesis

Biofilm evolution

Niche construction

20170515: Public DNS

After repeated trouble getting onto PubMed, finally changed DNS from ISP to Google Public DNS.

20170505: Hemineglect of memories

Faw and Faw [2017]:

Bisiach and Luzzatti tested hemispatial neglect patients on their ability to mentally picture the Piazza del Duomo in Milan, which the patients were familiar with prior to their brain damage. If asked to imagine facing the cathedral and describe the scene, they could only bring to mind the details to their right. When asked to imagine the square from the other side of it, they could only bring to mind the opposite side of the square, the one they had just previously neglected.

Guariglia et al [2013] found representational neglect in 1/3 of 96 consecutive R hemisphere injured patients using the "Familiar Square Description Test" but not the O'Clock Test. Faw and Faw [2017] interpret these findings as that the engram is complete, but "broadcasting" back to the appropriate brain regions for interpretation has failed.

20170505: Pursuit and evasion

Chung et al

The lion-and-man game is a geometric version of the cops-and-robbers game. In the original version, the game takes place inside a circular arena of radius r. The players have the same maximum speed, which we will assume to be one. The objective of the lion (pursuer) is to capture the man by moving onto the his current lo- cation. Suppose the players move in turns. It is not too difficult to see that the lion can win the game with the following strategy: starting from the center of the circle, stay on the radius that passes through the man's current location and move as close as possible to the man. It can be shown that the capture time of this strategy is O(r2) using a similar analysis to the one given by Sgall (2001). It is interesting to note that when the players move in continuous time, this strategy does not reduce the distance between the players to zero in finite time. Littlewood (1953) provided a proof of how a similar argument can be made against any lion strategy. Therefore, if the game is played in continuous time, the man escapes. Alonso et al. (1992) showed that there exists a strategy with which the lion can get within a distance c of the man in time O(r log(r/c)). Recently, Bhadauria and Isler (2011) showed that three lions can capture the man in any polygon (possibly with holes).

On the open plane, it has been known that three lions [travelling at same speed as target] can capture the man if and only if the man's initial location is contained in their convex hull (Jankovic, 1978). Recently, Kopparty and Ravishankar (2005) generalized this result to arbitrary dimensions: d + 1 lions in Rd.

The continuous time case reminds me of the Axiom of Determinacy, and of hypercomputing, which I guess the lion and the man have to be doing.

20170427: Recombination

Bherer et al [2017]:

Used 3.3 million recombination events from 104000 individuals from Iceland, BTNS, NTR, 23andMe etc.

We found male and female recombination rate to be strongly associated with density of H3K4me3 marks (Fig. 6a, Supplementary Fig. 17), and that H3K4me3 marks containing at least one degenerate PRDM9 motif are more recombinogenic in both sexes than those lacking a motif

20170424: FTO

Key variant for obesity is rs1421085, according to Claussnitzer et al [2015], though several SNPs tag an extensive haplotype eg rs1121980, rs9939609 (the most commonly available), and rs8044769. These don't affect BMI change due to diet or exercise [Livingstone et al 2016]. Smemo et al [2014] showed that the obesity-associated FTO SNPs interact with the promotor region of IRX3 500 kpb distant, acting as IRX3 eQTLs. Mouse Irx3 knockouts are 30% lighter. Rask-Andersen et al [2015] and Hunt et al [2015] also demonstrate effects on IRX5, which is 1.2 Mbp away. RPGRIP1L is adjacent to FTO, and mouse knockouts gain weight. The whole region FTO-IRXB cluster (16:53.56-55.48Mbp) is a topologically associated domain [ Dixon et 2012], and is full of deeply conserved noncoding elements. Claussnitzer et al [2015] identified rs1421085 as disrupting an ARID5B motif, doubling IRX3 and IRX5 expression:

This results in a cell-autonomous developmental shift from energy-dissipating beige (brite) adipocytes to energy-storing white adipocytes, with a reduction in mitochondrial thermogenesis by a factor of 5, as well as an increase in lipid storage. Inhibition of Irx3 in adipose tissue in mice reduced body weight and increased energy dissipation without a change in physical activity or appetite. Knockdown of IRX3 or IRX5 in primary adipocytes from participants with the risk allele restored thermogenesis, increasing it by a factor of 7, and overexpression of these genes had the opposite effect in adipocytes from nonrisk-allele carriers. Repair of the ARID5B motif by CRISPR-Cas9 editing of rs1421085 in primary adipocytes from a patient with the risk allele restored IRX3 and IRX5 repression, activated browning expression programs, and restored thermogenesis, increasing it by a factor of 7.

20170414: 9q32 melanoma locus

rs10739221 appears in Perry et al [2014] as a locus associated with age at menarche (P=4e-41), with a stronger signal for rs10453225, which is only very weakly associated with nevi and melanoma (P ~ 1e-3). However, rs10453225 does lie within the proximal (9q31.1) melanoma peak, while rs10739221 is in the distal 9q31.2 peak.

20170411: Chromatin HMM colour codes

StateColour Meaning
1 Red TssA (Active_TSS)
2 OrangeRed TssAFlnk (Flanking_Active_TSS)
3 LimeGreen TxFlnk (Transcr_at_gene_5_and_3primer)
4 Green Tx (Strong_transcription)
5 DarkGreen TxWk (Weak_transcription)
6 GreenYellow EnhG (Genic_enhancers)
7 Yellow Enh (Enhancers)
8 MediumAquamarine ZNF/Rpts (ZNF_genes/repeats)
9 PaleTurquoise Het (Heterochromatin)
10 IndianRed TssBiv (Bivalent/Poised_TSS)
11 DarkSalmon BivFlnk (Flanking_Bivalent_TSS/Enh)
12 DarkKhaki EnhBiv (Bivalent_Enhancer)
13 Silver ReprPC (Repressed_PolyComb)
14 Gainsboro ReprPCWk (Weak_Repressed_PolyComb)
15 White Quies (Quiescent/Low)

20170406: Tim Maudlin's kinds of realism

Internal Realism: any acceptable account of the world should explain how language use helps people to prosper in their interactions with the world...Putnam uses this plus "the view that whether a theory has a unique intended interpretation has no absolute sense" ie antirealist, and implying some kind of verificationist semantics.

Metaphysical Realism: some significant parts of language have definite truth conditions such that it is possible for an operationally ideal theory, as stated in this part of the language, to be false

Mild Scientific Realism: most scientific theories meeting a certain standard of evidential support are true or approximately true

Optimistic Scientific Realism: every interesting fundamental question about the world can eventually be answered by an approximately true or true theory...

Piercian Realism: OSR plus truth = best possible scientific theory

Regarding Eddington's table consisting largely of empty space:

...[T]he more pedestrian reading is just to deny that tables are as one conceives them to be, that is to assert that one has some false beliefs about tables.

20170406: Proportion of genome that is functional

Lindblad-Toh et al [2011] say "at least 5.5% of the human genome has undergone purifying selection, and report constrained elements covering ~4.2% of the genome."

Smith et al [2013] report on a pipeline to specifically identify evolutionarily constrained RNA structures. Their "predictions comprise 13.6% of the human genome, 88% of which fall outside any known sequence-constrained element". These ECS predictions were usually within 50 kb of protein-coding genes. Iyer er al [2015] screened the human genome, finding a consensus transcriptome of 91013 genes, of which 58648 were lncRNAs, 48952 previously unannotated. Only 1% of lnRNAs were ultraconserved, and 7% harboured known GWAS SNPs, usually from "gene deserts".

lncRNA sequences vary greatly across species, but there is stronger selection on secondary structure. One window on this is the effects of SNPs on lncRNA function versus predicted folding effects. However "conservation in lncRNAs proven to be functional is also very weak [overall, so] ...selective constraints [seems to be] limited to the maintenance of a few clusters of positions, which may be involved in lncRNA function by participating in structure or binding motifs present in the mature transcript" [ Pegueroles and Gabaldon 2016].

NONCODE is a database of ncRNAs

The LincSNP database ( contains SNPs in lncRNAs and their TF binding sites.

20170404: Dichoptic colour mixing

Dichoptic or binocular colour mixing and fusion - an old question was regarding "binocular yellow" (green and red to one eye each). Such fusion can occur if smaller patches and low luminance.

Erkelens and van Ee [2002] present multiple examples of both mixing and induction (which will persist in one eye).

the strength of brown was maximal for stimuli of Munsell hue 5 YR ("orange"), low value (roughly "lightness"), and moderate chroma (roughly "saturation"). Importantly, across a range of yellow-red Munsell hues (2.5Y to 7.5 R), strong perception of brown was associated with dark samples, while orange was associated with samples of higher lightness. Subsequent research focused primarily on whether or not brown was an elemental or unique hue percept rather than studying the determinants of brown stimuli themselves. This debate proved inconclusive. However, Uchikawa, Uchikawa, and Boynton returned to research on the factors that influence the perception of brown, showing that proximity of a higher luminance surround had a great influence on whether a target stimulus was perceived as brown or orange. They found that a bright, contiguous surround, even if it was only 1/16 the size of the target stimulus, greatly increased the likelihood that a subject would call the target brown instead of orange.

20170331: xfam.org

xfam.org is a collection of databases and HMMs for RNA (Rfam), protein (Pfam) etc.

20170329: gllm

gllm is my package for generalized log-linear models eg missing or incomplete data.

Excellent agreement with poLCA::poLCA on (lca) test dataset k=2.

   A B C D Freq               Model LRTS Prop1    Prop2
1  0 0 0 0   29  gllm::gllm   13.15617   0.5148   0.4852
2  1 0 0 0   52  poLCA::poLCA 13.15617   0.5147   0.4853
3  0 1 0 0   43  e1071::lca   13.33003   0.507    0.493
4  1 1 0 0  238
5  0 0 1 0   55  gllm   A       B      C      D
6  1 0 1 0   21  Class1 0.2277 0.22014 0.7434 0.7881 
7  0 1 1 0   12  Class2 0.8651 0.83602 0.1779 0.1656 
8  1 1 1 0   64
9  0 0 0 1   57
10 1 0 0 1   28
11 0 1 0 1   30
12 1 1 0 1   54
13 0 0 1 1  178
14 1 0 1 1   61
15 0 1 1 1   59
16 1 1 1 1   19

20170326: Processing bottlenecks in brain

Three classic examples [Marois and Ivanoff 2005]:

Visual short term memory: 1. attentional blink is 0.5 s gap in which a second stimulus cannot be consciously attended to. The second stimulus elicits an N400 ERP (semantic processing) and tertiary visual cortex, but not P300 (working memory update) and fronto-parietal cortex. 2. Limits on capacity to ~4 objects that are not too complex.

3. In tasks requiring selection of an appropriate response to a stimulus, the psychological refractory period (~0.7 s) is where a decision is not possible for a second stimulus, even if in a different modality. This can be reduced by practice at the task. "Electrophysiological studies have established the temporal boundary of the PRP bottleneck to a stage between stimulus consolidation in working memory and motor preparation."

20170325: Brain energetics

20% of total energy consumed metabolically is in resting brain (~25 watts). This increases ~5% for evoked activities ("perceptual and motor activity, task performance and similar cognitive functions") eg Xie et al [2016].

Merckle [1989] estimated this was roughly equivalent to 1013 to 1016 operations per second. I would very roughly estimate 1012 operations per gram cortex per second, which decreases 20% in non-REM sleep, and increases 5-10% when undertaking a cognitive task involving that brain region.

Tim Detter thinks the above is an underestimate given more recent results, and suggests more like 1021 FLOPS. The AI Impacts project (aiimpacts.org) review five estimates ranging 1013 to 1025, the latter including "metabolome" contributions.

Anders Sandberg [2016] reviews much of what I cover below - though also mentions cooling requirements as another constraint. He mentions the Landauer limit at 37C is 1.3x10-21 J, so the upper physically possible computational rate is ~1022 irreversible operations per second.

Atwell and Laughlin [2001]:

Each neuron (plus associated glial cell) is estimated to use 4.4e8 ATP/s/neuron, at rest, ie 2.4e-11 W. Assuming signalling at 4kHz, this rises to 1.7e-10 W.

"...an increase in activity of 1 action potential/cortical neuron/s will raise oxygen consumption by 145 mL/100 g [rat] grey matter/h", that is 6.5 umol ATP/g/min or 21 umol glucose/100 g/min. An increase of 1 spike/neuron/s is a 16% increase in energy consumption over baseline. Extrapolating to humans gave 14 umol/g/min or 308 ml oxygen/100 g/min, which is pretty close, see below.

In human neocortex, 11000-50000 neurons/mm3 (mean ~44000). This is 10-fold less than in the rat, but synaptic density is unchanged. That is 9-13e8 synapses/mm3

Cerebral blood flow is ~11 ml/s (1350 g brain) ie 0.9 ml/s/100 g tissue
Cerebral O2 consumption 0.06 ml/s/100 g
Cerebral ATP total consumption ~200 umol/s
1 ml/s blood ~ 2.3 W

Hall et al [2012]

Oxidative phosphorylation is the main mechanism initially providing energy to power neuronal activity, that the main subcellular mechanisms underlying information processing, i.e., presynaptic action potentials, neurotransmitter release, postsynaptic currents, and postsynaptic action potentials, all consume oxygen...11% of O2 use was on presynaptic action potentials, 17% was on presynaptic Ca2+ entry and transmitter release, 46% was on postsynaptic glutamate receptors, and 26% was on postsynaptic action potentials.

Therefore 6 O2 + 1 glucose => 26 ATP. This is confirmed by metabolic studies of human brain where oxygen:glucose ratio estimated 4.9-6.2, and O2 consumption of 1.56 umol/g/min (4 mW/g tissue - grey and white; 0.35 ml O2/g/min). Hydrolysis of one mole of ATP provides 28-33.5 kJ. Alternatively, one ATP = 4.6-5.6e-20 J.

The visual task used by Lin et al [2010] consumed 11.1 umol/g/minute of ATP at rest (5-6 mW/g), increasing by 2 umol/g/minute (1 mW/g) in the active condition (98% aerobic). Riedl et al [2016]:

most energy required for neuronal communication is consumed postsynaptically, i.e., at the target neurons.

Dorr et al [2012] show that arteriolar diameter decreases and vessel tortuosity increases in a transgenic mouse Alzheimer's model.

Microvascular flow rates are probably controlled by diffusion of neurotransmitters, notably glutamate, from local neurones (neurovascular coupling).

Human cortices contain ~1.5e10 neurons (chimps ~6.2e9, elephant ~1.1e10). Human cortical synapses 1-3e4 per neuron, and greater myelination than elephants, so ~5x faster.

Rodriguez and Granger [2016] point out that sizes of human brain regions are strongly proportional to other species', so that novel human abilities arise from an expansion of a standard architecture.

Shulman et al [2014] review the energetics, mentioning that astroglia are using only 15-20% of resting metabolism. Anesthesia studies show nice linear relations with calibrated BOLD, PET, EEG 13C MR spectroscopy. In deep anesthesia, use ~50% O2; in NREM sleep ~80%.

Seymour et al [2017] present cerebral blood flow data based on carotid foramen area:

Species total ICA blood flow ml/s Mean ml/sNotional W
H. sapiens 6.0 7.3 8.1 7.0 7.6 7.2 25
H. neander 4.9 6.1 4.7 6.9 6.2 5.8 20
H. heidel 3.2 5.6 4.4 16
H. erectus 4.5 3.0 3.9 3.8 5.0 4.0 14
H. habilis 2.4 2.4 9
H. rudolfensis 1.8 1.8 6

Alexandre et al [2018] discuss costs of cognition. And Pontzer et al [2016] show that "human total energy expenditure (TEE) exceeded that of chimpanzees and bonobos, gorillas and orangutans by approximately 400, 635 and 820 kcal/day, respectively, readily accommodating the cost of humans' greater brain size and reproductive output. Much of the increase in TEE is attributable to humans' greater basal metabolic rate...[traditional human hunter-gatherers] have far greater percentage body fat"

20170325: Stichometry

Since scribes paid by the line (usually 35 characters, as in hexametric poetry) , counts in documents important. Commonly, find documents organised along poetic or musical or Pythagorean lines eg section lengths are in twelfths of the total, the key argument is halfway, and the totals for Plato's writings are 1200 (the Apology), 12000 (the Republic), 3600 (Gorgias), 14400 (Laws) lines [Kennedy [2010]].

Stichometric measurements bring a new datum to this debate over the Divided Line. The numerical value of the Golden Mean is, to three places, 0.618, and thus a unit length divided at the Golden Mean will be divided at 61.8 percent. Surprisingly, the Republic's discussion of the Divided Line begins at 61.7 percent of the way through the text. By itself, this could be a coincidence, but the other dialogues typically contain allusions to the Golden Mean near 61.8%.

Moore [2012] takes up these ideas to analyse Plato's Philebus, which he thinks full of Pythagorean allusions:

The people of old [the Pythagoreans], superior to us and living in closer proximity to the gods, have bequeathed us this tale, that whatever is said to exist consists of One and Many, having in its nature Limit (peiras and the Unlimited (apeiron).

These are male and female, continuous and discrete, and so forth. In music, harmony is the combination of these two principles.

The tetractys is the symbol ascribed by the (later) neo-Pythagoreans to Pythagoras (triangle of the 1-4 dots) in the Pythagorean oath, and is associated with Platonic ideas about the importance of the decad (ten gifts of the Demiurge in Timaeus). The Pythagorean oath, as recorded by Sextus Empiricus, is:

No, I swear by him who gave the tetractys to our head,
which has the source and root of everlasting nature.

Each row is one level cum emanative hypostasis, through which apeiron is filtered in its undiminished bestowal into form.

OneThe Onestatic
TwoThe female, who proceeds from the onemovement and epithesis
ThreeThe male
(Five) (Marriage 2+3)
(Seven) (Athena, Reason - underivable from 1..8,10)
(Nine)Also Justice
(Ten) (return to Unity, Counter-Earth)

Valentinus held the Tetractys as mother of the universe, as did Proclus, Theon, Syrian etc

Hamilton wrote a sonnet to the tetractys, and it is suggested he thought quaternions might partake of these properties. His friend Coleridge frequently mentioned "the adorable tetractys".
Non-commutative algebra of the quaternions

Only reals, complex, and quaternions obey division algebra.

20170323: Nevi and ANKRD's and NFIC

Two lesser peaks on 4 and 10 are intergenic, but there are ANKRD17 and ANKRD16 in the region of each. The 4q13.3 intergenic ADAMTS3--COX18-ANKRD17 has many other GWAS SNPs eg rs10518107 is a waist-circumference SNP 5 kbp away.

There is a very narrow peak for rs34466956 on 19p13.3 5kbp upstream from NFIC (NFIC nuclear factor IC). The SKIN CHROMATIN ACCESS track shows an equally narrow peak straight over this SNP, in agreement with the Roadmap ChromHMM for foreskin melanocytes. This coincides with a broader DNaseI cluster.

NFIC is a CCAAT-binding transcription factor. It regulates development in multiple tissues (eg teeth, fibroblasts, liver, prostate), and is expressed in melanocytes. A NFIC-BRAF fusion has been reported in 1 of 46 mucosal melanoma cases. In one Alzheimer's disease GWAS, rs9749589 was reported to affect risk on a APOE*4 background.

Mice homozygous for a targeted null allele have abnormal incisor and molar root development, show reduced alveolar bone formation, and exhibit impaired feeding leading to severe runting and premature death when reared on standard laboratory chow.

Pjanic et al [2013]:

NFI family of proteins displays the unusual property of regulating not only the initiation of transcription but also of mediating DNA replication. NFI recognition sequence were found in the promoter sequences of many cellular genes, where they may act as activator or repressor of transcription. Recently, it has been proposed that NFI may be involved in a long range regulation of gene expression, through the formation of chromatin barrier and by blocking the propagation of a heterochromatic structure

It has a predilection for target genes that are highly expressed.

Recall NFIB [ Chang et al 2013 ] coregulates epithelial-melanocyte stem cell behaviour in hair follicles. In Aaron's paper, NFIC and NFIA have opposite effects to NFIB.

http://www.exsnp.org/eQTL lists rs131031 (chr22) and rs3731951 (chr2) as NFIC level eQTLs (P ~ 10-8).

20170322: Petra Haden and her sisters (triplets)

Children of Charlie Haden. Petra sings with the Decemberists, eg their cover of Wuthering Heights, and does a capella versons of King Crimson, The Who etc, folk, and jazz standards (with Bill Frisell).

20170320: Van Vogt's Supermind

Research Alpha
The Proxy Intelligence

20170316: "Meaning is normative"

Following Kripke's Wittgenstein...Gibbard gives:

X ought to accept "snow is white" iff snow is white.

This is an objective ought, so that meaning supervenes on natural facts, as the subjective ought would be:

X ought to accept "snow is white" iff X ought to accept snow is white.

20170315: Testicular Germ Cell Tumour GWAS

Chung et al [2013] Genome-wide significant SNPs associated with TGCT
SNP Gene Region
rs4657482UCK2 1q24.1
rs210138 BAK1 6p21.31
rs995030 KITLG12q21.32
rs2736100TERT 5p15.33
rs4635969TERT-CLPTM1L 5p15.33
rs755383 DMRT1 9p24.3
rs2900333ATF7IP 12p13.1

20170314: Quantum simulation

Quantum simulation promise to be one of the primary applications of quantum computers, should one be constructed.

Continuous-variable quantum computing can be used to simulate processes such as those in Quantum Field Theory.

20170309: Annotation tools

From Danielle Posthuma's group:


cf RegulomeDB

and Cancer Regulome Explorer

20170308: GHR

Common splice variant d3-GHR polymorphism is deletion of entire exon 3 (frequency ~25%), which increases GH responsiveness by 30%, leading to increased height in individuals with GH deficiency, and possibly central adiposity and low birth weight in the general population. In the GIANT consortium study, no effect on height or BMI in the general population. Is tagged by rs6873545 (B37 5:42631264), and can be measured biochemically by ELISA on serum GH.

Tagging SNPs (rs6873545 is absent) on the CoreExome (1KG European):
Marker 1 Marker 2 Nmean D' r2
rs6873545 rs4590183 2504 0.9990.901
rs6873545 rs7701605 2504 0.8180.588
rs6873545 rs6413428 2504 0.8160.356
rs6873545 rs3877899 2504 0.8200.315
rs6873545 rs13154178 2504 0.6860.328

No association of rs4590183 with CMM, nevus count, or height.

20170307: Other renal disease genes

Zhang et al [2008] reported rs1800860 to be associated with reduced kidney size. Kaczmarczyk et al (2016) give G/G total kidney volume 103 (23) v. 110 (19) ml/m2 (P=0.034) in 188 newborns. This is variously a splice site or synonymous change in the RET proto-oncogene. The A allele is 0.31 in Europeans, 0.14 in Africans.

Quinlan et al [2007] report PAX2 SNPs rs1800898, rs11599825, rs11190688, and rs11190702 had a similar effect size (10%). Zhang et al 2011] report rs12329305 (MAF 3%) where the T allele produces "no identifiable transcript" with again an allelic effect size of an 11.8% decrease in renal volume:
Genotype N Kidney volume ml/m2 Cystatin C
PAX2 rs11599825(A/A or A/G), OSR1 rs12329305 (T/C) 2 100.64 (9.65) 2.19 (0.32)
PAX2 rs11599825(G/G), OSR1 rs12329305 (C/C) 81 137.72 (29.33) 1.89 (0.36)
RET rs1800860(A/A or A/G), OSR1 rs12329305(T/C) 5 108.49 (9.57)* 2.13 (0.18)**
RET rs1800860(G/G), OSR1 rs12329305(C/C) 24 138.23 (31.26)

PAX2 rs11599825101025201750.23 0.98
OSR1 rs12329305 219552913 0.06 0.90
RET rs1800860 1043606687 0.31 0.84
BMPR1A rs7922846 1088641251 0.50 - Obesity; on Affy GenomeWide V6

20170228: KITLG

Now genome-wide significant for naevus count (best SNP rs7313352), but looks like no overlap with melanoma risk. Rick points out has well characterized effects on melanocyte growth. Guenther et al [2014] found that rs12821256 alters the binding site for the lymphoid enhancer-binding factor (LEF) transcription factor, so reducing LEF responsiveness and enhancer activity by 20% in keratinocytes. Mice carrying ancestral or derived variants of the human KITLG enhancer exhibited significant differences in hair pigmentation. This SNP was associated with fair hair in Iceland and the Netherlands, and is 350 kbp from KITLG.

Aside from rs12821256, a top SNP from our big hair colour meta-analysis is rs151086684, which is ~100 kbp away from KITLG. Effects of KITLG variants on skin colour are undetectable.

By contrast, the functional SNP from Zeron-Medina et al [2013] is rs4590952, which lies within a p53-binding site, and is in strong LD with known testicular cancer SNPs rs995030 and rs1508595 [ Kanetsky et al 2009, Chung et al 2013 ]. It is a nevus gene, which is not the case for rs12821256:

SNPChr BP37 BP38 All MAF Effect SE Nevus PMela PNote
rs995030 12 88890671 88496894g/a 0.36+0.051 0.01061.805e-06 0.2915TGCT SNP
rs7313352 12 88949124 88555347g/a 0.22+0.060 0.01051.197e-080.6611Strongest nevus SNP
rs4590952 12 88953659 88559882g/a 0.20+0.055 0.01014.526e-080.7231Cancer functional SNP
rs1508595 12 88986016 88592239g/a 0.36-0.001 0.00175.536e-01 0.4810TGCT SNP
rs151086684 12 89097290 88703513c/t 0.07-0.015 0.02690.5877 0.3102Hair colour
rs12821256 12 89328335 88934558t/c 0.12-0.011 0.01640.5038 0.7354Icelandic functional variant

Hair Colour

SNPChr BP37 BP38 All Hair Colour Note
rs7313352 12 88949124 88555347a/g 0.08 0.11 0.064 7.83e-05 Strongest nevus SNP
rs4590952 12 88953659 88559882g/a 0.39 0.107 8.99e-05 Cancer functional SNP
rs151086684 12 89097290 88703513c/t 0.01 4.16e-07 2.48e-17
rs12821256 12 89328335 88934558t/c 4e-7 1e-7 5.05e-10 2.29e-23 Icelandic functional variant

Linkage disequilibrium (r2):

rs995030 1
rs7313352 0.873 1
rs4590952 0.861 0.946 1
rs1508595 0.793 0.875 0.927 1
rs151086680.007 0.008 0.008 0.008 1
rs128212560.016 0.017 0.017 0.016 0.360 1

p53 in the acute sunburn response, especially within keratinocytes, is an important factor. Keratinocytes stimulate melanocytes by release of KITLG, FGF2, EDNs and alpha-MSH. So p53 transcriptionally upregulates the pro-opiomelanocortin (POMC) gene after UV exposure, the precursor to alpha-MSH [ Walker and Box 2008].

Somatic mutations of KIT are present in melanoma [ Curtin et al 2006], but are not common (2%) [ Handolios et al 2010], occurring on sunexposed sites.

The previously reported KITLG testicular germ cell tumour SNPs (assoc P=9e-43) are:
rs995030 g/a 0.43 9.6e-43
rs1508595 c/t 0.38 8.4e-41
Wang ZQ, Si L, Tang Q, Lin D, Fu Z, Zhang J et al. Gain-of-function mutation of KIT ligand on melanin synthesis causes familial progressive hyperpigmentation. Am J Hum Genet 2009; 84:672¿77.

Wehrle-Haller B. The role of Kit-ligand in melanocyte development and epidermal homeostasis. Pigment Cell Res 2003;16:287-96.

Zazo Seco C, Serrão de Castro L, van Nierop JW, Morín M, Jhangiani S, Verver EJ et al. Allelic Mutations of KITLG, Encoding KIT Ligand, Cause Asymmetric and Unilateral Hearing Loss and Waardenburg Syndrome Type 2. Am J Hum Genet 2015;97:647-60.

White spotting genes have been associated with hypopigmentary disorders and deafness in neurocristopathies resulting from NC stem cell-derived melanocyte deficiency during development. These include PAX3, SOX10, MITF, SNAI2, EDNRB, EDN3, KIT, and KITL

Fisher et al [1998] is the original paper linking MITF and KITLG: "Activation of c-Kit by KITL ("Steel factor") activates MITF via the MAPK pathway".

Yang et al [2015] found that MIR34C (11q23.1, within c11orf88, b37:111513439..111513515, miR-34c) knocks down its targets KITLG, PLCXD3, E2F3, MET, YY1, SERPINE1, PDGFRs. MIR29C regulates MIR34C levels

Jonnalagadda et al [2016] found that in South Asians, rs1426654 is associated with skin melanin index.

Tebaldi et al [2015] report "whole-genome cartography of p53 response elements ranked on transactivation potential":

we examined expression changes of thirteen genes as a function of different p53-inducing conditions, providing further evidence for KITLG p53-dependent regulation.
Mendel-From et al [2009] rs17084733 eye colour P=0.03

In the Scottish population, the three SNPs, rs1492354 (p = 0.0009), rs1907702 (p = 0.018) and rs10777129 (p = 0.007) located in intron 1 of the KITLG gene were significantly associated with hair colour. Of the three SNPs, the rs1492354 AG genotype showed the highest contribution to both the red hair colour dimension a* values (P < 0.001, R2 = 0.119) and the yellow hair colour dimension b* values (p = 0.001, R2 = 0.103), the latter value was significant when uncorrected by permutations. When retyped in the Danish population, the SNPs rs10777129 (p = 0.02, OR = 3.0) and rs1492354 A (p = 0.04, OR = 2.5) were significantly associated with light hair colour. In both the Scottish and the Danish populations, the alleles associated with lighter colour had similar low frequency (rs1492354: 0.08 and rs10777129: 0.07), and given the sample sizes only one homozygote rs10777129 GG was observed, in a Dane.

20170228: HDAC4

New peak (rs3791540) on 2q37.3 for nevus count is intronic to HDAC4 (histone deacetylase 4). HDAC4 "does not bind DNA directly, but through transcription factors MEF2C and MEF2D. It seems to interact in a multiprotein complex with RbAp48 and HDAC3". It is widely expressed, including in the skin. This is distant from the other 2 chr2 SNPs, that lie in LOC101929926, which expresses only in testis, according to the Human Protein Atlas's RNAseq [ Fagerberg et al 2014].

HDAC4 is deleted in melanoma cell lines (3/76), according to Stark and Hayward [2007]. Cohen et al [2014]:

Histone acetylation is mediated by a balance between histone acetyltransferases (HATs) and the three classes of histone deacetylases (HDACs): Class 1 (HDAC1-3,8), class 2 (HDAC4-7,9-11), and class 3 (Sirt1-7).

The HDAC4 signature was validated by showing significantly increased HDAC4 activity in cells treated with interferon gamma, a known upstream activator of HDAC4, in GSE3920.

Cancer types with high EZH2 activation consistently also have low HDAC4 activation (p < 0.000001). Melanoma [was in the set with] the highest EZH2 activation and lowest HDAC4 activation.

Top SNP rs55875066 lies in MITF binding region of intron.

HDAC inhibitors (eg panobinostat, vorinostat, citarinostat, SB939) represent the first generation of approved epigenetic therapy agents, although their use as monotherapy in advanced sarcomas has had disappointing results beyond some prolongation of stable disease. Like other drugs tested in advanced tumors, HDAC inhibitors may be more effective in combination with other agents and/or when tested against time-to-event rather than objective response endpoints. Beyond the HDAC inhibitors, phase I and II trials are ongoing for at least three EZH2 and seven bromodomain inhibitors in a wide range of cancer types

HDAC1 and HDAC2 also have key roles in brain development. Jacob et al [2014] show that in "neural crest cells, HDAC1/2 induces expression of the transcription factor Pax3 by binding and activating the Pax3 promoter. In turn, Pax3 maintain[s] high Sox10 levels..."

Mechanistically, "DNA is negatively charged and histones are positively charged. The addition of acetyl groups to histone tails neutralizes their positive charges. This leads to less interaction between DNA and histones, thereby resulting in a more relaxed chromatin...In addition to their effect at the chromatin level, HDACs are also able to deacetylate nonhistone targets, including several transcription factors, to modulate their activity. "

20170224: Intercellular Mitochondrial Transfer

After a throw-away remark in a talk, I chased up:

Toralba et al [2015]

Diverse structures mediate intercellular mitochondrial transfer. These include tunneling nanotubes (TNTs), microvesicles, mitochondrial ejection or cytoplasmic fusion.

Cancer cells opportunistically acquire mitochondria to optimize or repair their metabolic machinery. In this regard, horizontal transfer of mtDNA from host cells to tumor cells with compromised respiratory function has been described [Tan et al 2015]. Tumor cells devoid of mtDNA, a model of extreme mtDNA damage, acquire mtDNA from surrounding donor healthy cells, repairing the transcriptional and translational activity of mtDNA-encoded genes and triggering a substantial recovery of mitochondrial respiration, as well as tumor initiation efficacy and metastasis capacity. A major source of healthy mitochondria in these contexts are stromal cells...

Tunneling nanotubes (TNTs) [Rustom Science 2004] form between different cell types in vitro and in vivo (Figure 1). Tunneling nanotubes facilitate the selective exchange of organelle or membrane vesicles and small soluble cytoplasmic and membrane molecules. The establishment of a nanotube begins with the formation of a filopodium-like membrane protrusion that retracts after reaching the receptor cell, leaving an ultrafine structure that is separated from the substrate. [Increased in stress]

20170223i: Arrow Impossibility Theorem and theory choice

Theory choice and social choice: Kuhn versus Arrow S Okasha - Mind, 2011 - research-information.bristol.ac.uk

Kuhn's famous thesis that there is 'no unique algorithm' for choosing between rival scientific theories is analysed using the machinery of social choice theory. It is shown that the problem of theory choice as posed by Kuhn is formally identical to a standard social choice [setup].

20170223: pqR v. R

Radford Neal's hotted up version of the R interpreter.

Merging multiple files (then one plot, not shown):

> x <- read.delim("deafness.assoc.txt")
> map <- read.table("~/Genetics/Maps/Dogs/dog131.map")
> vcf <- read.csv("~/Genetics/Maps/Dogs/Canis_familiaris_incl_consequences20140205.vcf.gz", sep="\t", skip=16)
> map$idx <- paste(map$V1,map$V2)
> x$idx <- paste(x$chr,x$ps)
> x <- merge(x, map, by="idx", all.x=T)
> names(x)[names(x)=="V3"] <- "snp"
> idx <- is.na(x$snp)
> x$snp <- as.character(x$snp)
> x$snp[idx] <- x$rs[idx]
> x <- merge(x, vcf, by.x="snp", by.y="ID", all.x=T)
program user system elapsed
R3.1.1 171.312 0.372171.866
pqR2.15.1159.220 0.380159.733

20170222: TPCN2

This gene encodes a putative cation-selective ion channel with two repeats of a six-transmembrane-domain. The protein (TPC2) localizes to lysosomal membranes and enables nicotinic acid adenine dinucleotide phosphate (NAADP) -induced calcium ion release from lysosome-related stores. This ubiquitously expressed gene has elevated expression in liver and kidney. Two common nonsynonymous SNPs in this gene strongly associate with blond versus brown hair pigmentation.

Ambrosio et al [2016]

TPC2 is expressed in melanocytes and localizes to the melanosome-limiting membrane and, to a lesser extent, to endolysosomal compartments by confocal fluorescence and immunogold electron microscopy. Immunomagnetic isolation of TPC2-containing organelles confirmed its coresidence with melanosomal markers. TPCN2 knockout by means of clustered regularly interspaced short palindromic repeat/CRISPR-associated 9 gene editing elicited a dramatic increase in pigment content in MNT-1 melanocytic cells. This effect was rescued by transient expression of TPC2-GFP. Consistently, siRNA-mediated knockdown of TPC2 also caused a substantial increase in melanin content in both MNT-1 cells and primary human melanocytes. Using a newly developed genetically encoded pH sensor targeted to melanosomes, we determined that the melanosome lumen in TPC2-KO MNT-1 cells and primary melanocytes subjected to TPC2 knockdown is less acidic than in control cells. Fluorescence and electron microscopy analysis revealed that TPC2-KO MNT-1 cells have significantly larger melanosomes than control cells, but the number of organelles is unchanged. TPC2 likely regulates melanosomes pH and size by mediating Ca2+ release from the organelle, which is decreased in TPC2-KO MNT-1 cells, as determined with the Ca2+ sensor tyrosinase-GCaMP6.

Bellono et al [2016] report that patch-clamped melanosomes have "currents outwardly rectifying with a negative reversal potential (Erev) characteristic of OCA2-mediated melanosomal Cl- currents. Cytoplasmic treatment with PI(3,5)P2 activated a large inward whole-melanosome current (via TPC2). Verapamil blocks this TPC2 inward current and there is one case report of hair colour changing from grey to black in a patient treated with verapamil (Lancet letter 1991).

20170221: BASP1

The 23andme melanoma GWAS paper [Ransohoff et al 2017] claims rs187843643 in BASP1 is significantly associated with melanoma risk. Law et al meta-analysis [2015] obtained P=0.19 (seen in 2 studies). Note that European MAF is 0.003.

20170213: Immanentizing the eschaton

Due to Eric Voegelin.

20170124: AKT

Abeyrathna and Su [2015] :

Akt family kinases (also known as protein kinase B /PKB) are serine/threonine kinases that belongs to the general class of AGC kinases (AMP/GMP kinase and PKC subfamily of proteins) which has 518 members in humans...Akt mediates cell survival induced by growth factors.

The PI3k-AKT-mTOR pathway is one major role, but there are nonclassical pathways leading to AKT activation. Different AKTs may be tissue specific (eg AKT3 KO mice have reduced brain size), and phosphorylation is a key regulator of function.

AKT activation is obligatory in several cancers (eg important in BRAF inhibitor resistant melanoma). Post-zygotic mutation of AKT1 causes Proteus Syndrome (mosaic tissue overgrowth, epidermal nevi). Extracellular HSP90 signals to AKT1/2 to promote skin cell migration and wound healing.

20170110: Enoggera Gorge with G.

Down taped track from Scrub Road gate, leaving ~10:30. Passed hardest section of gorge by ascending N bank. Followed S trending branch to where meets BS Break, orienteering control point marker still there. Returned S. Boundary and Scrub Rd. Saw several wompoo pigeons hanging around on road and creek. Back to car 19:00. Both of us bites on trunk, possibly from rather dark pool we swam in, which also contained a large leech.

20170106: Art of Eric Heller

Abbott and James Lawrence Professor of Chemistry and Professor of Physics at Harvard University working on few body quantum mechanics, scattering theory, quantum chaos, and quantum correspondence to classical chaos.




20170103: On the concept of species

JF Blumenbach (1826) A Manual of the Elements of Natural History is the English translation of the 1820 German edition. Uses species and genus to describe types of animals, plants and minerals. On life, or "Organised Bodies":

Some of them [Organised rather than Inorganic] are invariably produced by other bodies of the same form and kind; so that their existence in an unbroken series, up to the first Creation*, presupposes other similar bodies, to which they owe their being.

* Or at least to their first progenitors; for in the first part of my Contributions to Natural History (Beytrage zur NaturGeschichte), I have adduced facts rendering it more than probable, that even in the present Creation, new species of organized bodies arise. To such subsequent operations the first origin of many simple and microscopic beings appears referable; such, for instance, as the greater number of the Animalculae Infusoriae.

Blumenbach cites Bomare [1769] Universal Dictionary of Natural History. On crosses between races of horses (from volume 2 of 12):

Une observation des plus essentielles, & absolument nécessitire dans les Haras [stud], c'est le soin de croiser [intercross] les races, pour les empêcher [prevent] de dégénérer. Il y a dans la nature, dit M. de Buffon, un prototype général dans chaque espece, sur lequel chaque individu est modele, mais qui semble en se réalisant s'altérer ou se perfectionner par les circonstances; ensorte que relativement a des certaines qualités, il y a une variation bizarre en apparence dans la succession des individus, & en meme-temps une constance admirable dans toute l'espece. Le premier animal, le premier cheval, par exemple, a été le modele extérieur ou le moule intérieur sur lequel tous les chevaux qui sont nés; tous ceux qui existent, & tous ceux qui naitront ont été formés; mais ce modele a pu s'altérer & se perfectionner en communiquant sa forme & se multipliant.

George Bentham (1827) Outline of a New System of Logic: With a Critical Examination of Dr. Whately's Elements of Logic simply says:

"As to the particular sense in which naturalists make use of the word species, it is very different from the logical sense of the word, the only one in which it should be made use of on the present occasion."


"..the naturalist must always be assured of the all-comprehensiveness of his classes and subclasses...and his divisions must therefore have been performed in the contradictory bifurcate mode [ie hierarchical successive dichotomization]...that species must be compared with all the others, and a property must be mentioned by which it is distinguished from each of them. The species is then constituted by the coexistence of these properties, and all the other species are mentally combined into one second species, characterized by the absence of any one or more of these properties...It must be carefully remembered that these observations apply to the logical division of collective entities only, not to the analytic division of individuals."

Added 20200920: Linnaeus says:

Nowhere has the Great Creator placed so many objects before the human senses as in the vegetable kingdom, which covers this whole globe that we inhabit. Thus, if a pure method is of use anywhere, it is here, if we shall hope to gain a clear idea of Vegetables. Thus Cesalpino: Unless plants are reduced to orders, and distributed into their classes like the squadrons of an army, everything is bound to fluctuate.

...[E]veryone proceeded according to his own method and divided those genera into orders and classes: Cesalpino, Hermann, Ray, Knauth the elder according to the fruit; Tournefort according to the shape of the corolla; Rivinus according to the number and equality of the petals; Magnol according to the calyx. And none of these methods did any harm.

Andrea Cesalpino's 1583 de Plantis Libri XVI divides plants into 32 groups using Aristotelian "binary logic". He was professor of botany and medicine at the University of Pisa in 1560, and physician to Cosimo I de Medici and Pope Clemens III. And John Ray (1727-1605) "was one of the most innovative naturalists before Linnaeus. He experimented with a number of literary formats in natural history, and proposed a similar species concept to that of Linnaeus. ...His most famous work in botany was Historia plantarum [1686]".

JEH Smith points to a remark of Cesalpino's highlighted by Atran 1991:

In 1571 the Italian botanist writes that "Eternity can only arise from the eternal: since the proper work of the vegetative soul is to engender its like, which makes for the eternity of the species, it is necessary that its substance not be corruptible. The reason for the eternal lies neither in corruptible existences taken individually, nor in their totality"..."[F]or like everywhere engenders like," as Cesalpino puts it, "according to nature and of the same species" (Cesalpino 1583, p. 26; cited in Atran 1991, p. 142).

...Whereas for traditional biology, including that of Aristotle, the primary interest had always been in determining and grouping species according to morphological aspect and ecological proclivity, natural history after Cesalpino, and then more clearly after Linnaeus, gradually came to focus on determining species' genealogically-related affinities (Atran 1991, p. 80). For Cesalpino, essence is communicated through reproduction..., and it is this process that ensures the eternity of species, rather than merely approximating to eternity...

Recall Swammerdam was active at this time.

20161206: Doug Bates on misunderstood aspects of mixed modelling


The first notebook is on singular estimates of the covariance of vector-valued random effects.

It can be a bit hit-or-miss for some of the plots in the display directly from github.com. If you have jupyter installed (see https://jupyter.org) you should be able to view the notebooks in a Jupyter session.

20161129: BayesMendel risk calculator


20161128: Best ever MITF eQTL

rs717669 (monocytes) P=7e-5 between MITF and FOXP1.

20161121: BioPlex


The overarching project goal is to determine protein interactions for every member of the collection. A first paper in Cell reports the first ~2,500 experiments (~23,000 interactions). Our current release with more than 5,000 human proteins as baits (~50,000 interactions) is also now available.

20161114: Regional heritability, BLUPs etc

Speed and Balding [2014] MultiBLUP approach

Hoffman et al [2014] lrgpr R package

Widmer et al [2014] and Heckerman et al [2016] discuss improvements to the FaST-LMM suite, claiming in passing that "much" missing heritability is fictional ;).

FaST-LMM, which stands for Factored Spectrally Transformed Linear Mixed Models, is a program for performing both single-SNP and SNP-set genome-wide association studies (GWAS) on extremely large data sets. This release contains the improvements described in Widmer et al., Scientific Reports 2014, and tests for epistasis.

This version of FaST-LMM is designed for use with randomly ascertained data with Gaussian residuals. If you have case-control data with substantial ascertainment bias, you should first transform your phenotype(s) using LEAP (Weissbrod et al., arXiv 2014), which is available at https://github.com/omerwe/LEAP. If you are analyzing continuous phenotypes with non-Gaussian residuals, you should first transform your phenotype(s) using Warped-LMM (Fusi et al., Nature Commun 2014), available at https://github.com/MicrosoftGenomics/WarpedLMM.


DISSECT was able to fit, after eight iterations, an MLM to a sample of 470,000 individuals and 590,004 SNPs in less than 4 h using the aggregated power of 8,400 cores and a total of ~16TB of memory (~2GB of memory per core)

20161107: Lisa Barrett on neural basis of emotion - constructionalism

The brain basis of emotion: A meta-analytic review

Argues that emotions are actually high level, and do not map onto distinct brain regions. Rather, they mix concepts, executive function and interoceptive affect, the latter representing anticipatory changes supporting the "best" response to an event.

20161031: Category theory applied to language and cognition

Phillips and Wilson [2012] is the third in a series of papers.

20161026: Harmonizome


20161026: Terry Speed on normalisation

Terry Speed's talk on normalisation methods included an example dataset from L1000.

L1000 is custom (Luminex) Broad mRNA profiling assay, that has been used as part of the NIH NIH LINCS (Library of Integrated Cellular Signatures) transcriptional network project. For example, one subproject involves 23 cells being treated with 311 chemical perturbagens and CRISPR reagents. The expression level for 978 representative genes is measured. The cells tested include two melanoma cell lines (A375 and SK-MEL-1).

20161024: UCSC RefSeq Genes with Gene Names

A version of genePred that associates the gene name with the gene prediction information. In alternative-splicing situations, each transcript has a row in this table.

table refFlat
"A gene prediction with additional geneName field."
    string  geneName;           "Name of gene as it appears in Genome Browser."
    string  name;               "Name of gene"
    string  chrom;              "Chromosome name"
    char[1] strand;             "+ or - for strand"
    uint    txStart;            "Transcription start position"
    uint    txEnd;              "Transcription end position"
    uint    cdsStart;           "Coding region start"
    uint    cdsEnd;             "Coding region end"
    uint    exonCount;          "Number of exons"
    uint[exonCount] exonStarts; "Exon start positions"
    uint[exonCount] exonEnds;   "Exon end positions"

20161013: Spontaneous collapse models for QM

Tumulka [2007]:

Ghirardi, Rimini and Weber [Phys Rev D 1986] have proposed a nonlinear, stochastic evolution law for quantum mechanical wave functions that deviates from the unitary Schroedinger evolution by implementing spontaneous collapses of the wave function. Two primitive ontologies have been proposed for use with the GRW wave function: a matter density ontology and a flash ontology, leading to two collapse theories denoted GRWm and GRWf.
Adler [2015]:

Current formulations of objective reduction models use not the discrete localizations of the original GRW paper but rather a nonlinear coupling of the Schrodinger equation to a stochastic noise variable, as introduced in the continuous spontaneous localization (CSL) model of Ghirardi, Pearle, and Rimini [Phys Rev A 1990]). The structure of this model, as I noted in my book on emergent quantum theory, is uniquely fixed by two natural physical requirements, which might be expected to arise in an integral way from a more fundamental physical theory. The first is the requirement of state vector normalization - the unit norm of the state vector should be maintained in time. The second is the requirement that there should be no faster than light signaling - the density matrix averaged over the noise should satisfy a linear evolution equation of Lindblad form. The form of the stochastic equation that is fixed by imposing these two requirements has the satisfying feature that reduction to definite outcomes, with probabilities obeying the Born rule, can be proved.

The Diosi-Penrose model posits objective reduction precipitated by gravitational interactions between states, and is derived from the GRW. "For a nucleon in a spatial superposition of two locations separated by a distance much greater than rCSL [~100 nm], lamdba is the average localization rate. The DP model adds stochastic fluctuations to the gravitational field..." [Helou 2016, vide infra]. Smirne and Bassi [2015]:

The continuous spontaneous localization (CSL) model is the most refined collapse model, as it also applies to identical particles. The mass density of a quantum system is coupled with a white-noise field, which can be interpreted as a classical random field filling space. Different speculations on the origin of the noise field have been developed, tracing it back, e.g., to gravity or to cosmological particles.

The localizing action of the collapse noise has to increase with size of the system, which guarantees a consistent description of microscopic and macroscopic systems.

A relevant drawback of the original CSL model, as well as of most collapse models, is that the average kinetic energy of the quantum system diverges on the long time scale...[We]...formulate a new, dissipative CSL model.

Dissipative effects are expected to play an important role also in the experimental investigation of collapse models via optomechanical systems, where proper signatures could be visible in the density noise spectrum of the mechanical oscillator, or via the spontaneous photon emission from electrically charged particles.

The LISA Pathfinder experiment has recently put strong bounds on these models [Helou et al 2016]

20161012: Genome coordinates

GRCh = Genome Reference Consortium Human Reference


chrY:10,000-2,781,479 and chrY:56,887,902-57,217,415
chrX:10,000-2,781,479 and chrX:155,701,382-156,030,895

Dec 2013GRCh38hg38
Feb 2009GRCh37hg19
Mar 2006NCBI36hg18
May 2004NCBI35hg17
Jul 2003NCBI34hg16

20161007: Analogical reasoning

The ability to perceive and use relational similarity between two situations or events.

Emmanuel Sander psychologist working in area - Sander and Hofstadter (2013) wrote Surfaces and Essences: Analogy as the Fuel and Fire of Thinking.

Sander highlights brain imaging studies of Gallant [eg 2012, 2013]

Projection of the [PCA] recovered semantic space onto cortical flat maps shows that semantic selectivity is organized into smooth gradients that cover much of visual and nonvisual cortex. Furthermore, both the recovered semantic space and the cortical organization of the space are shared across different individuals.

First, much of anterior visual cortex represents scene categories that reflect the co-occurrence statistics of objects in natural scenes. Second, voxels located within and beyond the boundaries of many well-established functional ROIs (regions of interest) in anterior visual cortex are tuned to mixtures of these scene categories. Third, scene categories and the specific objects that occur in novel scenes can be accurately decoded from evoked brain activity alone. Taken together, these results suggest that scene categories represented in the human brain capture the statistical relationships between objects in the natural world. [Previous studies suggest] the parahippocampal place area is selective for presence of buildings ...and we find that voxels within PPA have large weights for the "Urban/Street" category

20160928: RN Breakfast Album of the Week

Arbina by Noura Mint Seymali - Mauretanian.

Tzenni live.

20160925: Springbrook Pinnacle

From Canyon Lookout - Warrie Circuit with V and K. Left ~1300, back ~1730. Track junction really obvious, and pad moderately well used (also tape). Discovered true summit, about 100 m away from viewpoint G. and I climbed a couple of months ago, and climbed N. end ~12 m grade 4-5. Much easier to descend razorback ridgetop southward. Still walk behind Rainbow Falls on circuit track.


Jeffrey Shallitt notes this observation from The Elements of Eloquence:

Multiple adjectives in English order as opinion-size-age-shape-colour- material-purpose eg little old lady.

20160914: expression STRs

Gymrek et al [2016] scanned 15000 lymphocyte transcripts versus 13 cis-STRs per gene. The eSTRs were enriched in 5'UTRs, and replicated in the 1KG expression dataset. Claim that "eSTRs contribute about 12% of the genetic variance attributed to common cis polymorphisms... after controlling for all variants that are well tagged by common bi-allelic markers in the cis region".

Of the 216 genes with P<2e-12, the ones of interest include:

20160908: Dudbridge and Newcombe 2015

Hum Hered 2015;80:178-186

LD-pruned genetic risk scores tend to have slightly lower R2 than
jointly modelled scores, but the differences are small at sample sizes
up to 100,000. If the proportion of correlated variants is high, joint
modelling can obtain modest improvements asymptotically.

20160905: If forticlientsslvpn session drops out

Loss of internet connection leads to this not cleaning up properly. It leaves an altered resolv.conf file behind in /etc. These delay DNS lookups by 5 s, until presumably the originals in /var/run/resolvconf/resolv.conf are used (this normally is symlinked to /etc/resolv.conf). In the most recent episode, lost DNS completely.

One fix:

sudo dpkg-reconfigure resolvconf

20160903: Dennett on reasons

drawing a quite sharp distinction between doing things for reasons and having reasons for doing things. There are reasons aplenty in nature (I call them free-floating rationales): trees do things for reasons, fungi do things for reasons...but they do not (usually) have reasons for doing these things.

20160811: Popper's ideas about the Enlightenment

...the main ideas of the European Enlightenment. Among these were the ideas of fighting for truth and against obscurity; of talking and writing lucidly and modestly; of practising irony and especially self-irony; of avoiding the pose of a deep thinker; of looking critically at society; and of looking upon the world with wonder, and with an infectious curiosity.

Should make everyone want to be enlightened ;)

20160806: Martin-Lof number-theoretic randomness [1965]

From Bailly and Longo Phenomenology of Incompleteness: from Formal Deductions to Mathematics and Physics:

This is a "Goedelian" notion of randomness, as it is based on Recursion Theory and yields a strong form of (strong) undecidability for infinite 0-1 sequences (an infinite sequence is random if it passes all effective statistical tests. On the side of physical dynamics, its mathematical counterpart can be found in reference to Birkhoff's notion of ergodicity, which refers to infinite trajectories, a purely mathematical approach at the infinite limit, with no need to refer to the interval of physical measure to engender randomness. That is, mathematical dynamical systems, in their (Lesbesgue) measurable spaces, allow to define generic points and infinite random trajectories, in the ergodic sense. Recently, M. Hoyrup and C. Rojas, under Galatolo's and the second author's supervision, proved that dynamic randomness (a la Poincare, thus, but asymptotically, following Birkhoff), in suitable effectively given measurable dynamical systems, is equivalent to (a generalization of) Martin-Lof randomness... relating undecidability and Poincare's unpredictability.

There are also Hippocratic, Solovay and Demuth randomness.

Shen (2006) and Bienvenu et al [2013] mention:

The famous Goedel incompleteness theorem says that for every sufficiently rich formal theory there exist true unprovable statements. Such statements would be natural candidates for being added as axioms, but how can we obtain them? One classical (and well studied) approach is to add (to some theory T) an axiom that claims the consistency of T. Here we discuss another approach (motivated by Chaitin's version of the Goedel theorem) where axioms claiming randomness (incompressibility) of some strings are added, and show that it is not really useful (in the sense that it does not help us to prove new interesting theorems). [However,] randomly chosen axioms may help to make proofs much shorter (unless NP=PSPACE).

20160705: Horizon

D = 3.86 h0.5 ,
where h in m, D km, approximating:

(2(7R/6)h+h2)0.5 ,
with radius of earth R=6371 km, and refraction correction (7/6).

20160704: Human genetic diversity

AWF Edwards "Human genetic diversity: Lewontin's fallacy

That multivariate discrimination is powerful. Bodmer [2015] gives a historical reviewi of his work.

20160704: GSEA


20160630: The Decemberists

From Wikipedia:

An American indie folk rock band from Portland, Oregon The band stages whimsical reenactments of sea battles and other centuries-old events, typically of regional interest, or acts out songs with members of the crowd. In 2011, the track "Down By the Water" from the album The King Is Dead was nominated for Best Rock Song at the 54th Grammy Awards

The Tain is a song cycle of the stories of Cuchulain.

20160628: GNAS

GNAS (20q13.3) GRCh38.p7 20:58839681..58911196

This locus has a highly complex imprinted expression pattern. It gives rise to maternally, paternally, and biallelically expressed transcripts that are derived from four alternative promoters and 5' exons. Some transcripts contain a differentially methylated region (DMR) at their 5' exons, and this DMR is commonly found in imprinted genes and correlates with transcript expression. An antisense transcript is produced from an overlapping locus on the opposite strand. One of the transcripts produced from this locus, and the antisense transcript, are paternally expressed noncoding RNAs, and may regulate imprinting in this region. In addition, one of the transcripts contains a second overlapping ORF, which encodes a structurally unrelated protein - Alex. Alternative splicing of downstream exons is also observed, which results in different forms of the stimulatory G-protein alpha subunit, a key element of the classical signal transduction pathway linking receptor-ligand interactions with the activation of adenylyl cyclase and a variety of cellular reponses. Multiple transcript variants encoding different isoforms have been found for this gene. Mutations in this gene result in pseudohypoparathyroidism type 1a, pseudohypoparathyroidism type 1b, Albright hereditary osteodystrophy, pseudopseudohypoparathyroidism, McCune-Albright syndrome, progressive osseus heteroplasia, polyostotic fibrous dysplasia of bone, and some pituitary tumors.

Galpha and XLalpha (extra-large Galpha) are involved in G-protein coupled receptor transduction, eg beta-2-adrenergic receptor. Turan and Bastepe [2015]:

Paternal Gs-alpha expression is silenced in some tissues, including proximal renal tubules, neonatal brown adipose tissue (BAT), thyroid, gonads, paraventricular nucleus of the hypothalamus and pituitary... The most extensively investigated effector stimulated by Gs-alpha is adenyl cylase, which catalyzes the conversion of ATP into the second messenger cAMP.

McCune-Albright Syndrome is characterized by the classic triad of polyostatic fibrous dysplasia, café-au lait skin pigmentation with highly irregular borders, and peripheral precocious puberty...[and is due to]... missense mutations at Arg201 or Gln227,...important to GTP hydrolase activity.

20160627: Timing of UV exposure

Ransohoff et al [2016]:

Sun exposure is a major risk factor for skin cancer; however, the relative contribution of ultraviolet (UV) exposure during childhood versus adulthood on skin cancer risk remains unclear.Our goal was to determine the impact of residential UV, measured by AVerage daily total GLObal solar radiation (AVGLO), exposure during childhood (birth, 15 years) versus adulthood (35, 50 years, and present) on incident non-melanoma skin cancer (NMSC) and malignant melanoma (MM) in postmenopausal women.Women were followed with yearly surveys throughout the duration of their participation in the Women's Health Initiative Observational study, a multicenter study from 1993 to 2005. A total of 56,557 women had data on all observations and were included in the baseline characteristics. The main exposure, residential UV (as measured by AVGLO), was measured by geographic residence during childhood and adulthood. Outcome was risk of incident NMSC and MM.Over 11.9 years (median follow-up), there were 9,195 (16.3 %) cases of NMSC and 518 (0.92 %) cases of MM. Compared with the reference group (women with low childhood and low adulthood UV), women with low childhood and high adulthood UV had a 21 % increased risk of NMSC (odds ratio 1.21, 95 % confidence interval 1.12, 1.31). Women with high childhood and high adulthood UV had a 19 % increased risk of NMSC (odds ratio 1.19, 95 % confidence interval 1.11, 1.27). Surprisingly, women with high childhood UV and low adulthood UV did not have a significant increase in NMSC risk compared with the reference group (odds ratio 1.08, 95 % confidence interval 0.91, 1.28) in multivariable models. Residential UV exposure in childhood or adulthood was not associated with increased melanoma risk.This study reveals an increase in NMSC risk associated with adulthood residential UV exposure, with no effect for childhood UV exposure.

Wu et al [2014]:

Background: Few prospective studies have examined the relationship between sun exposure, other potential risk factors, and risk of different skin cancers [including basal cell carcinoma (BCC), squamous cell carcinoma (SCC), and melanoma] simultaneously. Methods: We evaluated the association between a number of potential risk factors and skin cancer risk in a cohort of 108,916 US women, the Nurses' Health Study II (1989-2009). Results: During 2.05 million years of follow-up, we identified 6,955, 880, and 779 diagnoses of BCC, SCC, and melanoma, respectively. Compared with participants in the lowest quintile of cumulative ultraviolet flux in adulthood, participants in the highest quintile had multivariable-adjusted relative risks (RR) of 2.35 (Ptrend < 0.0001) for BCC, 2.53 (Ptrend = 0.009) for SCC, and 0.68 (Ptrend = 0.38) for melanoma. In contrast, the RRs were 1.68 (95% CI, 1.55-1.82) for BCC, 1.68 (1.34-2.11) for SCC, and 1.80 (1.42-2.28) for melanoma for participants with 5+ blistering sunburns when compared with participants without sunburn between ages 15 and 20 years. We found significant interactions between family history of melanoma, number of blistering sunburns between ages 15 and 20 years and BCC risk, and between sunburn reaction as a child/adolescent and SCC risk (all Pinteraction < 0.05). Conclusion: In a cohort of U.S. women, we found that sun exposures in both early life and adulthood were predictive of BCC and SCC risks, whereas melanoma risk was predominantly associated with sun exposure in early life.
Cain et al [2014]:

Vitamin D is formed mainly in the skin upon exposure to sunlight and can as well be taken orally with food or through supplements. While sun exposure is a known risk factor for skin cancer development, vitamin D exerts anti-proliferative and pro-apoptotic effects on melanocytes and keratinocytes in vitro. To clarify the role of vitamin D in skin carcinogenesis, we performed a review of the literature and meta-analysis to evaluate the association of vitamin D serum levels and dietary intake with cutaneous melanoma (CM) and non-melanoma skin cancer (NMSC) risk and melanoma prognostic factors. Twenty papers were included for an overall 1420 CM and 2317 NMSC. The summary relative risks (SRRs) from random effects models for the association of highest versus lowest vitamin D serum levels was 1.46 (95% confidence interval (CI) 0.60-3.53) and 1.64 (1.02-2.65) for CM and NMSC, respectively. The SRR for the highest versus lowest quintile of vitamin D intake was 0.86 (0.63-1.13) for CM and 1.03 (0.95-1.13) for NMSC. Data were suggestive of an inverse association between vitamin D blood levels and CM thickness at diagnosis. Further research is needed to investigate the effect of vitamin D on skin cancer risk in populations with different exposure to sunlight and dietary habits, and to evaluate whether vitamin D supplementation is effective in improving CM survival.

Colantonio et al [2014]:

OBJECTIVE: We sought to update the evidence of the association of melanoma and indoor tanning focusing on frequency of use and exposure to newer tanning beds. METHODS: We searched Scopus, MEDLINE, and Cumulative Index to Nursing and Allied Health Literature on August 14, 2013. We included all observational studies that included patients with melanoma who had indoor tanned. Odds ratios (OR) with 95% confidence intervals (CI) were extracted and combined using generic inverse variance methods assuming a random effects model. RESULTS: In all, 31 studies were included with data available on 14,956 melanoma cases and 233,106 controls. Compared with never using, the OR for melanoma associated with ever using indoor tanning beds was 1.16 (95% CI 1.05-1.28). Similar findings were identified in recent studies with enrollment occurring in the year 2000 onward (OR 1.22, 95% CI 1.03-1.45) and in subjects attending more than 10 tanning sessions (OR 1.34, 95% CI 1.05-1.71). LIMITATIONS: The quality of evidence contributing to review results ranges from poor to mediocre. CONCLUSION: Using tanning beds is associated with a subsequent melanoma diagnosis. Exposure from more than 10 tanning sessions is most strongly associated and there was no statistically significant difference in this association before and after 2000, suggesting that newer tanning technology is not safer than older models.

Ferrucci et al [2014]: "2 US case-control studies of melanoma (1161 cases, 1083 controls, ages 25-59 years)... Persons who used indoor tanning exclusively in businesses were at increased risk of melanoma (OR=1.82, 1.47-2.26)..."

Guy et al [2015]: Among female high school students during 2013, a total of 20.2% engaged in indoor tanning and 10.3% engaged in frequent indoor tanning. Among male high school students, 5.3% engaged in indoor tanning and 2.0% engaged in frequent indoor tanning...From 2009 to 2013, indoor tanning significantly decreased among female students from 25.4%.

20160623: Stress-induced brain pathology

Lopes et al [2016]

Exposure to chronic stress is frequently accompanied by cognitive and affective disorders in association with neurostructural adaptations. Chronic stress was previously shown to trigger Alzheimer's-like neuropathology, which is characterized by Tau hyperphosphorylation and missorting into dendritic spines followed by memory deficits. Here, we demonstrate that stress-driven hippocampal deficits in wild-type mice are accompanied by synaptic missorting of Tau and enhanced Fyn/GluN2B-driven synaptic signaling. In contrast, mice lacking Tau [Tau knockout (Tau-KO) mice] do not exhibit stress-induced pathological behaviors and atrophy of hippocampal dendrites or deficits of hippocampal connectivity. These findings implicate Tau as an essential mediator of the adverse effects of stress on brain structure and function.

20160621: Yin-Yang-1

YY1 is a transcription factor intimately involved with cell cycle control eg via AKT1. It interacts with MITF.

https://elifesciences.org/content/4/e06857 look at these interactions. They mention that MITF cofactor is CTNNM1, heterodimerisation partners are TFEB, TFE3, TFEC. BRG1 (SMARCA4) interacts with MITF. SOX10 regulates MITF expression.

20160620: Putnam on Davidson etc

From his blog. His most recent (2015) paper

...argues, inter alia, that Tarski's results do not support deflationism about truth, contrary to what is often assumed... Tarski's formal methods intuitively draw on and presuppose not just one property of truth, the T-Schema, or Disquotation, but on that property AND the further property that the extension of "true" depends on the extension of "refers". The concepts of truth and of reference are intimately related, and his entire procedure exploits the relation...I also briefly describe and endorse Tyler Burge's claim that reference is psychologically more primitive and more ubiquitous than language use.

Burge writes

The reference relation holds between a psychological state or event, or a piece or use of language, on one hand, and a subject matter, on the other. When I say "x refers to y", I intend the reference relation. The relational use entails some state or activity with a referential function. But not all states or activities with referential functions establish a reference relation. When a reference relation is established by a person or animal - when the person or animal refers to something, some entity - the relation is always established by way of some thought, cognition, perception, or other psychological state or event, or by way of some piece or use of language, or other symbolism.

Davidson, in a Nice Derangement of Epitaphs argues for "language skepticism", in that rule following (shared conventions) is not sufficient. The "passing theory" is Davidson's idea that mutual understanding of speaker and listener requires active interpretation and interactions causing both to converge to a temporary facon a parler or translation schema. One example, quoted by Bar-On [2015], is malapropism:

If a fellow speaker were to tell us: "I came across many allegories in my visit to the Everglades" , we would understand her to mean alligators...If she later were to say alligators, we would have no trouble re-adjusting our interpretation of her.

20160609: Turner syndrome and cancer

Schoemaker et al 2008:

Women with Turner syndrome seem to be at increased risk of gonadoblastoma, meningioma and childhood brain tumours, and possibly bladder cancer, melanoma, and corpus uteri cancer, but are at a decreased risk for breast cancer.

However, breast cancer is certainly not unknown.

20160609: Bee electroreception

Lihoreau and Raine (2013)

The global atmospheric circuit generates a permanent electric field between the Earth surface and outer atmosphere (Rycroft et al., 2000). The ground and plants conductively linked to it are negatively charged (Bowker and Crenshaw, 2007), whereas animals build up positive charge as they move in contact with air molecules (Jackson and McGonigle, 2005). Electric fields emanating from plants and pollinators, such as bees, are believed to promote pollination by enabling pollen grains to "jump" from flowers to pollinators and vice versa (Corbet et al., 1982).

[A] bee visit temporarily modifies the electric charge of (Petunia) flowers, suggesting that floral electric properties could be used by future visitors to assess the reward value.

Greggers et al. (2013) [showed] that the sequence of wing beats and abdomen movements performed by dancers generates a specific modulated electric field composed of pulses of several hundred volts. To test whether bees can detect and discriminate such electric cues, the team stimulated tethered bees with either constant or modulated electric fields. Stimuli mimicking those of a waggle dancer (among others) triggered both antennal movements and walking activity...

Sutton et al (2016) find the hairs on the bee detect electric fields. Antennae in other insects eg crickets, may perform this role.

20160605: Top T1D SNPs (Oram 2015)

SNP Gene OR WeightEffect Allele
DR3/DR4-DQ8 48.18 3.87
DR3/DR3 21.12 3.05
rs2187668, DR4- 21.98 3.09
rs7454108 DQ8/DR4-DQ8
DR4-DQ8/X 7.03 1.95
DR3/X 4.53 1.51
rs1264813 HLA_A_24 1.54 0.43 T
rs2395029 HLA_B_5701 2.5 0.92 T
rs3129889 HLA_DRB1_15 14.88 2.70 A
rs2476601 PTPN22 1.96 0.67 A
rs689 INS 1.75 0.56 T
rs12722495 IL2RA 1.58 0.46 T
rs2292239 ERBB3 1.35 0.30 T
rs10509540 C10orf59 1.33 0.29 T
rs4948088 COBL 1.3 0.26 C
rs7202877 1.28 0.25 G
rs12708716 CLEC16A 1.23 0.21 A
rs3087243 CTLA4 1.22 0.20 G
rs1893217 PTPN2 1.2 0.18 G
rs11594656 IL2RA 1.19 0.17 T
rs3024505 IL10 1.19 0.17 G
rs9388489 C6orf173 1.17 0.16 G
rs1465788 1.16 0.15 C
rs1990760 IFIH1 1.16 0.15 T
rs3825932 CTSH 1.16 0.15 C
rs425105 1.16 0.15 T
rs763361 CD226 1.160.15 T
rs4788084 IL27 1.16 0.15 C
rs17574546 1.14 0.13 C
rs11755527 BACH2 1.13 0.12 G
rs3788013 UBASH3A 1.13 0.12 A
rs2069762 IL2 1.12 0.11 A
rs2281808 1.11 0.10 C
rs5753037 1.1 0.10 T

20160605: Reverse mathematics


Reverse mathematics is a program in mathematical logic that seeks to determine which axioms are required to prove theorems of mathematics. Its defining method can briefly be described as "going backwards from the theorems to the axioms", in contrast to the ordinary mathematical practice of deriving theorems from axioms. It can be conceptualized as sculpting out necessary conditions from sufficient ones.

The reverse mathematics program was foreshadowed by results in set theory such as the classical theorem that the axiom of choice and Zorn's lemma are equivalent over ZF set theory. The goal of reverse mathematics, however, is to study possible axioms of ordinary theorems of mathematics rather than possible axioms for set theory.

Reverse mathematics is usually carried out using subsystems of second-order arithmetic, where many of its definitions and methods are inspired by previous work in constructive analysis and proof theory. The use of second-order arithmetic also allows many techniques from recursion theory to be employed; many results in reverse mathematics have corresponding results in computable analysis.

The program was founded by Harvey Friedman (1975, 1976) and brought forward by Steve Simpson.

For each theorem that can be stated in the (weak) base system but is not provable in the base system, the goal is to determine the particular axiom system (stronger than the base system) that is necessary to prove that theorem.

20160603: intension

From http://www.britannica.com/topic/intension:

..."intension" indicates the internal content of a term or concept that constitutes its formal definition; and "extension" indicates its range of applicability by naming the particular objects that it denotes. For instance, the intension of "ship" as a substantive is "vehicle for conveyance on water," whereas its extension embraces such things as cargo ships, passenger ships, battleships, and sailing ships.

And from the SEP article on intensional transitive verbs:

A verb is intensional if the verb phrase (VP) it forms with its complement is anomalous...[or] ambiguous [:] substitution-resistance, the availability of unspecific readings, and existence-neutrality...For example, "Oedipus is seeking a member of his family" could be true because Oedipus is seeking one specific member, or no specific individual...(relational/notional ambiguity).`

One interest is whether it is possible to create a formal language semantics that can capture these properties eg in higher order type theory used in (the non-propositionalist) Montague Semantics, where composition is application of a function eg wooly() can be applied to sweater.

20160527: Laws of Nature

Emma Tobin argues that there are two distinct kinds of ceteris paribus laws in chemistry; idealized and approximate chemical laws. See also: bibliographies of the metaphysics of science

20160523: ASIP revisited

Schiaffino [2010] review mentions:

"In the mouse, Agouti is a paracrine signaling polypeptide of 131 amino acids that is normally produced by the dermis at the base of hair follicles...Agouti appears to function as an endogenous inverse agonist of MC1R, both by competing with and inhibiting the effect of melanocortins, and by stabilizing the inactive conformation of the receptor, thus decreasing both agonist-dependent and agonist-independent signaling.

Beta-Defensin 103 is a secreted molecule structurally similar to Agouti, highly expressed in the skin and able to modulate the activity of MC1R, by acting as high-affinity neutral agonist.

Attractin (ATRN) is a large type I transmembrane protein of 1428 amino acids, In melanocytes, attractin is expressed as a transmembrane form and functions at the cell surface as a low-affinity yet essential co-receptor for Agouti protein, which interacts with attractin with its N-terminus, while the C-terminus is engaged in MC1R binding.

Mahogunin has been identified as an additional accessory protein required for efficacious inhibition of MC1R signaling by Agouti. In fact, its mutation suppresses pheomelanin synthesis even in the presence of hypermorphic Agouti variants, while has not effects in the presence of MC1R loss-of-function...All Mahogunin isoforms decrease cAMP accumulation mediated by agonist-stimulated MC1R, but not beta-2-adrenergic receptor, by a mechanism independent from receptor downregulation and ubiquitylation, and most likely involving competition with Galphas for receptor binding.

20160502: Reading N900 addressbook backup

From https://talk.maemo.org/showthread.php?t=59384

Is a Berkeley DB:

db5.1_dump.exe -p addressbook.db | sed -e '9,$ !d' \
-e '/^ pas-id/d' \
-e 's/\\0d\\0a/\n/g' \
-e 's/\\\\, / /g' \
-e 's/^ //' -e 's/\\00//' \
-e 's/\\c3\\89/É/g' \
-e 's/\\c3\\8a/È/g' \
-e 's/\\c3\\a9/é/g' \
-e 's/\\c3\\a8/è/g' \
-e 's/\\c3\\aa/ê/g' \
-e 's/\\c3\\a0/à/g' \
-e 's/\\c3\\a1/á/g' \
-e 's/\\c3\\a2/â/g' \
-e 's/\\c3\\a7/ç/g' \
-e 's/\\c3\\b4/ô/g' \
-e 's/\\c3\\af/ï/g' \
-e 's/\\c3\\ab/ë/g' \
> mycontact.vcf  

20160428: Intermittent freezing of Xfce

This is where the cursor turns to an arrowed cross, and all windows stop responding.

Killing the xfwm4 process (it restarts automatically) seems to leave everything OK.

20160427: Leiter Poll moral philosophers

1. Bernard Williams
2. Derek Parfit
3. Philippa Foot
4. Thomas Nagel
5. T.M. Scanlon
6. G.E.M. Anscombe
7. Christine Korsgaard
8. Peter Singer
9. Alasdair MacIntyre
10. R.M. Hare
11. Martha Nussbaum
12. Judith Jarvis Thomson
13. Harry Frankfurt
14. Allan Gibbard
15. J.L. Mackie
16. Stephen Darwall
17. Peter Railton
18. Simon Blackburn
19. Samuel Scheffler
20. John McDowel

Leiter commented:

My top ten would include Stevenson, Brandt, Williams, Foot, Mackie, Railton, Scanlon, probably Nagel, probably Hare, probably Gibbard.

And see Annette and Kurt Baier, Julia Annas, Wiggins ("A Sensible Subjectivism?").

20160426: Immune status and nevus counts

Goldman and Richfield [1951]: nevus eruptions with corticosteroid therapy.

Hughes and Bailey [1989]: eruptions in children treated for lymphoma.

20160413: APOE etc

APOE E Allelers429358 rs7412
2 T T
3 T C
4 C C

The SNP rs7412 (B37 19:45412079) is on the CoreExome array. rs429358 (B37 19:45411941) is APOE*C130R and is imputable eg:

rs429358-rs7412 Prop 95%.CL D D'
C C 0.1467 0.1173--0.1835 0.0127 0.9999
C T 0.0000 0.0000--1.0000 -0.0127 -0.9999
T C 0.7665 0.7077--0.8303 -0.0127 -1.0000
T T 0.0868 0.0646--0.1166 0.0127 0.9999

Joshi et al [2016] confirm a -1.24 year of life effect per E4 allele (in women). Song et al [2012] report results from the Sydney Memory and Aging Study

At Wave 1, MCI (minimal cognitive impairment) subjects had lower levels of ApoA1, ApoA2 and ApoH, and higher levels of ApoE and ApoJ, and a higher ApoB/ApoA1 ratio. Carriers of the apolipoprotein E E4 allele had significantly lower levels of plasma ApoE, ApoC3 and ApoH and a significantly higher level of ApoB. Global cognitive scores were correlated positively with ApoH and negatively with ApoJ levels. ApoJ and ApoE levels were correlated negatively with grey matter volume and positively with cerebrospinal fluid (CSF) volume on MRI. Lower ApoA1, ApoA2 and ApoH levels, and higher ApoB/ApoA1 ratio, increased the risk of cognitive decline over two years in cognitively normal individuals. ApoA1 was the most significant predictor of decline. These associations remained after statistically controlling for lipid profile. Higher ApoJ levels predicted white matter atrophy over two years.
In Mather et al [2016], they test ApoH QTLs for association to cognitive phenotypes in the MAS and OATS:

several APOH SNPs were nominally associated with diabetes and cognitive performance...rs2873966 was associated with the cognitive domains of attention/processing speed (p=0.001), executive function (p=0.014) and global cognition (p=0.001). Two other SNPs also showed evidence of association with attention/processing speed (rs2010251, rs758767).

In Mather et al [2015] they list SNPs previously examined for association to hippocampal volume and late onset Alzheimer's Disease. Below I add AD SNPs from the I-GAP study [ Lambert et al [2013]]:
rs6265 BDNF
rs821616 DISC1
rs1754606 DISC1
rs1011313 DTNBP1
rs1018381 DTNBP1
rs2075650 TOMM40
rs4680 COMT
rs6581612 CHARGE
rs7294919 CHARGE
Alzheimer SNPs (CHARGE)
rs744373 BIN1
rs3764650 ABCA7
rs3818361 CR1
rs3851179 PICALM
rs610932 MS4A6A
rs3865444 CD33
rs670139 MS4A4E
rs9349407 CD2AP
Alzheimer SNPs (IMAP)
rs6656401 CR1
rs6733839 BIN1
rs9331896 CLU
rs983392 MS4A6A
rs4147929 ABCA7
rs9271192 HLA-DRB1/5
rs8093731 DSG2
rs190982 MEF2C
rs2718058 NME8
rs1476679 ZCWPW1
rs7274581 CASS4

Herold et al [2016] analysed data from 3500 Ss (1070 families), and add:
SNP Annotation
rs7609954 PTPRG
rs1347297 OSBPL6
rs1513625 PDCL3

Early-onset familial form of AD, onset age <65 years of age, <5% of diagnosed AD cases: APP, PSEN1 and PSEN2. Late-onset form explained heritability (20+ loci) is up to 50% already.

20160412: Emotion and opioids

Panksepp [2015] mentions, among many other interesting findings, that "ultra-low-dose" buprenorphine might be useful for suicidality.

Yovell Y, Bar G, Mashiah M, et al . Ultra-low-dose buprenorphine as a time-limited treatment for severe suicidal ideation: a randomized, controlled trial. Am J Psychiatry. In press.

Fava et al [2016] Opioid Modulation With Buprenorphine/Samidorphan as Adjunctive Treatment...

From this perspective, it is no surprise that prior to the modern era of psychopharmacology, the only highly effective medicinal ways to reduce depressive affect were opioids, and hence morphine was commonly used to treat depression in the 19th and first half of the 20th century. [46] Indeed, Emil Kraepelin provided explicit advice on how to use morphine for depression, with substantial care taken to minimize addiction. Indeed depression and low positive affect are, in part, low-brain-opioid states. [26] The likelihood of traditional full-agonist opioids, all of which are highly addictive, ever being part of modern psychiatric practice is remote. However, being a partial agonist of mu receptors, buprenorphine is more likely to receive acceptance, especially since it also has useful kappa-opioid-receptor antagonist properties - a major target for current antidepressant drug development

But the main thrust of the paper is cross-species affective neuroscience.

20160402: Kieran Setiya and Sam Scheffler

Kieran Setiya is at MIT. Samuel Scheffler is at NYU, and is a nonconsequentalist (wrote intro to What Matters).

Scheffler's most recent book is Death and the Afterlife, on the importance people place on a continuation of humanity.

20160331: Speculative Realism

Brassiers's description: "actor-network theory spiced with pan-psychist metaphysics and morsels of process philosophy".

A review of Grattan's book

Correlationism, as a word, is distilled directly from the work of Kant who talks of it as follows: "What we call external objects are nothing but mere presentations of our sensibility ... its true correlate, i.e., the thing in itself, is not cognized at all through these presentations and cannot be."

Meillassoux [After Finitude] cites the "ancestral statements" that scientists make (e.g., scientists have no problem concluding to the existence of a fossil, its 'reality' or 'event', that they have never seen alive and that even lived before humanity as such (so before the phenomenal existed)). For the true correlationist this should be impossible. As Gratton puts it, "Correlationists, thus, end up reducing everything, including the ancestral, to its appearance to conscious beings, yet the ancestral is precisely that which is not given to any consciousness or language"

The key is of course the idea that, contrary to (strong) correlationists such as Merleau-Ponty, time is not born of my relation to things.

the naturalism of Iain Grant, where nature is the starting point of the analysis, is discussed along with the neo-Darwinism of Elizabeth Grosz and the materialism of Jane Bennett [both posthumanists].

20160328: Proof-theoretic semantics

Schroeder-Heister's SEP article:

Inferentialism (see Brandom, 2000) [holds that] inferences and the rules of inference establish the meaning of expressions, in contradistinction to denotationalism, according to which denotations are the primary sort of meaning. Inferentialism and the "meaning-as-use" view of semantics is the broad philosophical framework of proof-theoretic semantics.

20160327: Albany Creek

Went for lovely walk in Dawn Road Reserve - 51 hectares of bushland. Eastern side of circuit road within follows Albany Creek, which is pretty, with palms - has a small section of remnant rainforest. Saw several wallabies, and apparently koalas are to be seen. Backs onto crematorium grounds on NE.

20160324: Jurgen Schmidhuber on interestingness

[His] Formal Theory of Fun and Creativity (1990-2010): a creative agent - one that never stops generating non-trivial, novel, and surprising behaviours and data - must have two learning components: a general reward optimiser or reinforcement learner, and an adaptive encoder of the agent's growing data history (the record of the agent's interaction with its environment).
Driven by Compression Progress: A Simple Principle Explains Essential Aspects of Subjective Beauty, Novelty, Surprise, Interestingness, Attention, Curiosity, Creativity, Art, Science, Music, Jokes

I argue that data becomes temporarily interesting by itself to some self-improving, but computationally limited, subjective observer once he learns to predict or compress the data in a better way, thus making it subjectively simpler and more beautiful. Curiosity is the desire to create or discover more non-random, non-arbitrary, regular data that is novel and surprising not in the traditional sense of Boltzmann and Shannon but in the sense that it allows for compression progress because its regularity was not yet known. This drive maximizes interestingness, the first derivative of subjective beauty or compressibility, that is, the steepness of the learning curve. It motivates exploring infants, pure mathematicians, composers, artists, dancers, comedians, yourself, and (since 1990) artificial systems.

20160316: We have glowworms

There are two under the old leaves at the R of the steps going up to the top terrace. Didn't see them on 17th or anytime since. However, a colleague recalls them in a garden in Windsor, so they are seen in Brisbane.

Rigby and Merritt [2011]:

The glowworm Arachnocampa flava is a carnivorous fly larva (Diptera) that uses light to attract prey into its web [and mucus balls] ["Harrison" is the SE Queensland (sub)species]. The light organ is derived from cells of the Malpighian tubules, representing a bioluminescence system that is unique to the genus. Bioluminescence is modulated through the night although light levels change quite slowly compared with the flashing of the better-known fireflies (Coleoptera).

In Arachnocampa, the larval light organ is derived from the distal segments of the four Malpighian tubules...[and] is located in a swollen posterior segment. Externally, a tracheal reflector surrounding the light organ is visible through the cuticle.

Larvae douse in response to stimuli such as light and loud noise, usually over a period of 1-2 min (Gatenby, 1959; Broadley and Stringer, 2009). When stimulated by the presence of prey, they brighten (Broadley and Stringer, 2009). The bioluminescence output comes under circadian regulation as well, with A. flava larvae maintaining rhythmic bioluminescence under constant darkness (Merritt and Aotani, 2008). Larvae in forest habitats tend to glow most brightly after dusk and slowly diminish towards dawn.

[They]...capture prey every 5-37 days...[F]ree water could be a limited resource in Arachnocampa: first, larval immobility and confinement to the snare in sheltered environments means that larvae may not have access to water droplets for ingestion; second, the carnivorous habit means that food is sporadically available and nutrient-rich but water-poor; and third, the production of large amounts of mucus and use of mucous droplets to capture prey exposes larvae to water loss through evaporation. Arachnocampa shows some water-conserving adaptations: larvae re-ingest droplets from their fishing lines as they carry out web maintenance (D.J.M., unpublished observations); also, the mucous droplets may be hygroscopic (Stringer, 1967), allowing the extraction of atmospheric water.

20160316: Interoception

Craig coined the term "interoception" in 2002. Craig [2009]:

A model of awareness based on interoceptive salience is described, which has an endogenous time base that might provide a basis for the human capacity to perceive and estimate time intervals in the range of seconds to subseconds. The model posits that the neural substrate for awareness across time is located in the anterior insular cortex, which fits with recent functional imaging evidence relevant to awareness and time perception. The time base in this model is adaptive and emotional, and thus it offers an explanation for some aspects of the subjective nature of time perception. This model does not describe the mechanism of the time base, but it suggests a possible relationship with interoceptive afferent activity, such as heartbeat-related inputs.

[S]ingle-cell recordings and high-resolution tracing ... revealed a phylogenetically novel ascending sensory pathway from lamina I (in the spinal cord) that joined with another novel pathway from the solitary nucleus (in the brainstem) in primates to form projections not only to all autonomic cardiorespiratory integration regions in the spinal cord and brainstem, but also to a specific thalamo-cortical relay nucleus that is modality specific and topographically organized in a unique manner. The subsequent recognition that this pathway fundamentally represents the long-missing homeostatic afferent side of the autonomic nervous system led to the insight that the terminus of this novel pathway in the posterior insular cortex of primates provides the basis for the sense of the physiological condition of the entire body, which I termed "interoception".

In this model, the cortical representation of the sentient self in the AIC is based on the integration of salience across all conditions in the individual's body and in the physical and emotional environment at each moment of time. The salience of any factor is determined by its significance for the maintenance and advancement of the individual and the species.

The phylogenetically novel homeostatic afferent pathway from lamina I and the solitary nucleus in primates provides the basis for the sense of the physiological condition of the entire body in the posterior insular cortex. This engenders distinct feelings from the body that have a fundamental homeostatic basis in their concomitant autonomic (cardiorespiratory) sequelae. The neural construction of these feelings seems to provide an important template for the construction of all feelings. The primary neural representations of the state of the body are re-represented and integrated in the mid-insula and, again, in the anterior insula...

An anatomical correlate of these hypothetical meta-representations of global emotional moments/behaviours might be provided by the so-called Von Economo neurons (VENs) that are selectively located in the anterior insula and the anterior cingulate cortices of humanoid primates; they are especially numerous only in adult humans but are not present at all in monkeys (Nimchinsky et al. 1999; Allman et al. 2005).

...the AIC is selectively activated when a subject identifies himself/herself in a photo, consistent with the idea that the feeling of identifying with the self-image (conditioned by feelings of ownership of movements and emotional expressions in a mirror) requires a mental representation of the sentient self... Interestingly, a study of subjective awareness of mental processes (the "feeling-of-knowing" inherent in memory recall; Kikyo et al. 2002) reported a selective and graded activation of the AIC and the ACC. Finally, this model is strongly supported by recent clinical studies which reported that degeneration of VENs in the AIC and the ACC is selectively associated with the loss of emotional awareness and self-conscious behaviours in patients with fronto-temporal dementia.

The ACC is intimately interconnected with the AIC and is co-activated in most, but not all, such studies. I view the AIC and the ACC as complementary sensory and motor limbic cortices that work together and provide the substrates for the feelings and motivations, respectively, that make up all emotions and emotional behaviours

[A] lesion of the anterior insula can eliminate emotional involvement with music (Griffiths et al. 2004). A positron emission tomography (PET) study showed selective activation of the left AIC associated with rhythm by using occasional mistimed intervals in well-known short melodies ...recognition that the AIC is uniquely associated with emotion, movement, music and time suggested the formulation of this model.

Liberati et al [2016] have shown via direct intracerebral recordings that the local field potentials at any site in the insula occur for not just nociceptive, but also vibrotactile, auditory and visual stimuli, in keeping with Craig's hypotheses.

Davey et al [2015] contrast the insular cortex hypothesis with their evidence for the default mode network as central to the self:

The posterior insular cortex, as a zone of convergence for interoceptive information, has complex functions (Craig, 2002). In representing the internal somatic state, it has a role in detecting salience (Liberati et al., 2016); and the salience network has been demonstrated to operate independently of the DMN (Sridharan et al., 2008; Menon and Uddin, 2010).

DMN is default mode network of posterior cingulate cortex, medial prefrontal cortex, and inferior parietal lobule that is active at rest, and when engaged in "self-reference" tasks, such as assessing appropriateness of descriptors for one's own personality.

Sokol-Hessner et al (2015). Interoceptive ability predicts aversion to losses

Participants performed both a heartbeat-detection task indexing interoception and a risky monetary decision-making task, from which loss aversion, risk attitudes and choice consistency were parametrically measured. Interoceptive ability correlated selectively with loss aversion and was unrelated to the other value parameters.
Terasawa et al [2014] Interoceptive sensitivity predicts sensitivity to the emotions of others, but Ainley et al [2015] Heartfelt empathy? No association between interoceptive awareness, questionnaire measures of empathy, reading the mind in the eyes task or the director task.0

And two odd studies on propranolol:

Sokol-Hessner et al (2015). Determinants of Propranolol's Selective Effect on Loss Aversion

Propranolol selectively reduced loss aversion in a baseline- and dose-dependent manner (i.e., as a function of initial loss aversion and body mass index), and did not affect risk attitudes or choice consistency.
Replicating (?) Rogers et al [2004].

20160315: Sarcoma viruses in the dog (and cat)

Waugh et al [2015] exclude an EBV-like gammaherpesvirus as a major cause of lymphoma in dogs.

The Kaposi's sarcoma virus (HHV-8) is a good model in that it is vascular-endothelium-tropic. Note the Merkel cell polyoma virus was only recognised in 2008 (80% of Merkel cell carcinomas).

Linkage analysis [ Pedergnana et al 2012] in 40 large Cameroonian families suggests a susceptibility locus on 3p22 (LOD score=3.83, P=2e-5). The 95% confidence interval (corresponding to a 1.5 LOD decrease from the chromosomal position with the peak LOD score) defined a 5.35-Mb chromosomal distance (from SNP rs4796 to SNP rs2226462). This region contains 36 genes, NC1IL1 to MEK26A1

Feline sarcoma viruses are three recombinant replication-defective versions of the feline leukemia virus that have incorporated oncogenes (often fusion proteins) from host genomes. They are associated with fibrosarcomas.

Felis catus gammaherpesvirus 1 is a newly recognized virus (~16% domestic cats, increased if carry feline immunodeficiency virus) that clusters with the percaviruses (the new subfamily). It is not associated with any pathology.

Vaccine-associated feline sarcoma (VAFS) is a fibrosarcoma arising at the site of vaccination. Banerji et al [2007] claim a strong association with p53 genotype (P=10-5).

The Lawrence Livermore Microbial Detection Array is a Nimblegen array containing 388000 probes for all sequenced viruses (6000) and bacteria (15000), which is used with a Phi29 polymerase-based whole transcriptome amplification (WTA) kit (Qiagen).

20160311: banausic, hypocorism, paranymph

Practical, utilitarian (banausoi are Greek artisans or labourers). The stereotypic antibanausic rhetoric of philosophers of some eras.

A hypocorism is a pet name ("child-talk").

Paranymph is Bridesmaid/groomsman, or attendant of a doctoral student.

20160310: deLong recommends

Everybody needs to read, think about and teach the lessons of two of the very best books of the past two years: Martin Wolf's 'The Shifts and the Shocks' and Barry Eichengreen's 'Hall of Mirrors.'

20160307: Vitamin D

The Cochrane review of Vitamin D supplementation and cancer prevention [2014] reviewed 18 RCTs - "Trial sequential analysis (TSA) of the 18 vitamin D trials shows that the futility area is reached after the 10th trial, allowing us to conclude that a possible intervention effect, if any, is lower than a 5% relative risk reduction". There was a small effect on total mortality, confirmed (P=0.02) in an extended analysis of 56 RCTs (N=95000) with a NNT of "150 people treated over five years to prevent one additional death". The overall RR was 0.97 (0.94-0.99), but for D3 was 0.94 (95% CI 0.91 to 0.98).

The Vitamin D/Calcium Polyp Prevention Study is a double-blind, placebo-controlled factorial trial of vitamin D and/or calcium supplementation for the prevention of large bowel adenomas, that ran 2004-2013.

Baron et al [2015] report:

We recruited patients with recently diagnosed adenomas and no known colorectal polyps remaining after complete colonoscopy. We randomly assigned 2259 participants to receive daily vitamin D3 (1000 IU), calcium as carbonate (1200 mg), both, or neither in a partial 2×2 factorial design. Women could elect to receive calcium plus random assignment to vitamin D or placebo. Follow-up colonoscopy was anticipated to be performed 3 or 5 years after the baseline examinations, according to the endoscopist's recommendation. The primary end point was adenomas diagnosed in the interval from randomization through the anticipated surveillance colonoscopy.

Participants who were randomly assigned to receive vitamin D had a mean net increase in serum 25-hydroxyvitamin D levels of 7.83 ng per milliliter, relative to participants given placebo. Overall, 43% of participants had one or more adenomas diagnosed during follow-up. The adjusted risk ratios for recurrent adenomas were 0.99 (95% confidence interval [CI], 0.89 to 1.09) with vitamin D versus no vitamin D, 0.95 (95% CI, 0.85 to 1.06) with calcium versus no calcium, and 0.93 (95% CI, 0.80 to 1.08) with both agents versus neither agent. The findings for advanced adenomas were similar. There were few serious adverse events.

Daily supplementation with vitamin D3 (1000 IU), calcium (1200 mg), or both after removal of colorectal adenomas did not significantly reduce the risk of recurrent colorectal adenomas over a period of 3 to 5 years.

The Women's Health Initiative randomized 400 IU D3 - no effect on CRC risk; Trivedi et al [2003] used 100000 IU every 4 months. The VITAL study (N=26000) is giving 2000 IU D3/d, and started in 2010. D-Health gives 60000 IU monthly and has 22000 participants, and started the intervention in 2014. Jacobs et al [2016] point out that:

the association between serum 25(OH)D and colorectal adenoma have consistently demonstrated a statistically significant inverse relationship for incidence, but not recurrence...

Williams et al [2016] found Vit D not associated telomere length in a Finnish birth cohort (5096 Ss age 31).

Beveridge et al [2015] meta-analyse supplementation trials and found no effects on BP.

20160301: Google safe browsing


ht Nasser Abbasi on c.l.f.

20160218: Humour

Humor - International Journal of Humor Research.

Strohminger et al [2012]:

...mirth promotes tolerance of deontological violations in moral dilemmas, whereas elevation promotes rejection of such violations. Contrary to the negativity-diminishment hypothesis, our results cannot be explained simply in terms of positive affect. Instead, the present study supports the view that positive emotions are functionally distinct, and influence decision-making according to emotion-specific appraisals.

20160216: The Magnetic Fields


Their best-known work is the 1999 three-volume concept album 69 Love Songs.

20160211L CNV estimation

In the Illumina SNP genotyping assay two probes are used to detect the presence of the two different alleles at each SNP. The alleles measured by the X channel (Cy5 dye) are called the A alleles, whereas the alleles measured by the Y channel (Cy3 dye) are called the B alleles. Two values are automatically generated from Illumina BeadStudio software from the X and Y values, the Log R Ratio and the B allele frequency. The Log R Ratio (LRR) is a measure proportional to the total signal intensity (and therefore the copy number) . The B Allele Frequency (BAF) is a measure of the relative contribution to the total signal from the B allele (and therefore reflects the allelic composition).

LRR = log2(X + Y)
BAF = Y/(X + Y)

20160209: Ascertaining version of Debian or Ubuntu

lsb_release -a

20160201: Antimatter storage

...fabrication of very high aspect ratio (1000:1) micro-Penning-Malmberg trap arrays designed to store antimatter. The structure consists of thousands of 10um diameter tubes etched by deep reactive ion etching through Si wafers

One can consider a nanotrap (as small as a cyclotron radius of positron) containing only one positron which avoids all plasma complications and pushes the density over the Brillouin limit, and permits confinement times limited only by vacuum conditions.

20160130: The internal model principle

John Baez on the topic

The internal model principle (IMP) states that ... regulation can only be achieved if the system contains an "internal model" of the input being regulated


Conant and Ashby [1970]

Every good regulator of a system must be a model of that system

under very broad conditions, ... any regulator that is maximally both successful and simple must be isomorphic with the system being regulated.

...the living brain, so far as it is to be successful and efficient as a regulator for survival, must proceed, in learning, by the formation of a model (or models) of its environment

...looks like an extension of W. Ross Ashby's Law of Requisite Variety from his Introduction to Cybernetics (1956).

Sontag [2003]. Adaptation and Regulation with Signal Detection Implies Internal Model:

...a simple result showing, under suitable technical assumptions, that if a system S adapts to a class of external signals U, in the sense of regulation against disturbances or tracking signals in U, then S must necessarily contain a subsystem which is capable of generating all the signals in U. It is not assumed that regulation is robust, nor is there a prior requirement for the system to be partitioned into separate plant and controller components.

Andrews et al [2008]. An approximate internal model principle: Applications to nonlinear models of biological systems

20160122: Yablo paradox [1986, 1993] and Berry Paradox

Yablo paradox is an infinite sequence of sentences:

S1 : for all m > 1, Sm is false.
S2 : for all m > 2, Sm is false...

Unlike the liar paradox, seems non-circular (ie without self-reference). A typed logic possibly does not escape it, but proof that it is a paradox requires a jump from S1 is false, S2 is false, to:

For all n, Sn is false

which requires additions to standard first-order arithmetic and logic. Priest further has argued that there is a fixed-point and thus circular representation:

there exists no number greater than x that satisfies the predicate
Y(x) = (for all k>x not(Sat(G[Y(z)],k)))

where G[] is a Goedelization of the function.

The Berry Paradox was described by Bertrand Russell, and is a more usual self-referential paradox:

"The smallest positive integer not definable in under sixty letters."

Boolos [1989] uses a formal version to prove Goedel's first incompleteness theorem:

"m is the first number not definable in less than k symbols"

And also:

The Bad Company problem - an exact demarcation of the line between the good abstraction principles and the bad. But the criteria that we end up with are surprisingly complex, in both a philosophical and a mathematical sense. And this is going to cause philosophical troubles, because the distinction between good and bad principles is, it turns out, massively undecidable.

20160120: DASS

The Depression Anxiety Stress Scales (DASS) are 21 and 42 item questionnaires [ Lovibond and Lovibond 1995] that can be summarized as three subscales - Depression, Anxiety, Stress-Tension. The Anxiety scale correlates 0.81 with the Beck Anxiety Inventory, and the Depression scale 0.74 with the Beck Depression Inventory. The lower correlation with the BDI was blamed on the latter including items tapping weight loss, insomnia, somatic preoccupation and irritability that were thought to overlap with anxiety as well as tension.

Table 1 from Osman et al [2012]

Item F1 F2 F3
Factor 1 (F1; Depression)
3. no positive feeling .74-.03 .05
5. no initiatives .32 .07 .24
10. had nothing .76-.03 .01
13. down-hearted .58-.10 .15
16. unable about things .60-.03 .10
17. wasn't worth much .82 .02-.12
21. life is meaningless .79 .03-.18
Factor 2 (F2; Anxiety)
2. dryness of mouth -.09 .57-.04
4. breathing problem -.06 .80-.11
7. trembling .04 .57 .09
9. worried .11 .35 .27
15. close to panic .15 .49 .23
19. action of my heart -.01 .71-.12
20. felt scared .06 .45 .23
Factor 3 (F3; Stress)
1. hard to wind down -.16-.03 .83
6. over-react .11-.10 .69
8. nervous energy -.10 .05 .80
11. getting agitated .06-.03 .71
12. difficult to relax -.08-.03 .85
14. intolerant .07 .10 .53
18. rather touchy .16-.11 .66

Henry and Crawford [2005] settled on:

a quadripartite structure, [that...] consisted of a general factor of psychological distress plus orthogonal specific factors of depression, anxiety, and stress. This model was a significantly better fit than a competing model that tested the possibility that the Stress scale simply measures Negative Affectivity.

The DASS Depression Scale exhibits ceiling effects in clinically depressed populations [ Page et al 2007], though it is "moderately sensitive" to changes within individuals over time. Scores do not covary greatly with demographic variables [ Crawford and Henry 2003].

In a sample of 750 twin pairs from the ATR-based Twin study in Wellbeing using Integrative Neuroscience of Emotion (TWIN-E) [Burton et al 2015], results for the DASS-42 were:

SubscaleMean SD RangerMZ rDZ
Depression2.991 4.4630-41 0.3720.043
Anxiety 1.876 2.7070-24 0.3600.172
Stress 5.951 5.1360-35 0.3620.156

Davey et al report similar findings for the DASS-21 (within-twin, below diagonal; MZ cross-twin, above diagonal):

Subscale rMZ Depression Anxiety Stress
Depression0.31 - 0.24 0.27
Anxiety 0.23 0.51 - 0.25
Stress 0.35 0.64 0.58 -

From the AE common factor model of Burton et al, the expected correlations (total sample) are:

Subscale rMZ Depression Stress
Depression0.31 - 0.07
Stress 0.34 0.37 -

20160120: MKL and gfortran

mecej4 on c.l.f. mentions:

MKL is available for use with Gfortran. The Community License for MKL would let you build the updated Lapack95 using Gfortran. In fact, I did that just two days ago, on Windows/Cygwin. All that you have to do is

make FC=gfortran libia32 INSTAL_DIR=<directory_name>

after you have unpacked the sources.

20160113: Epistemology

Goldreich (2004):

Complexity Theory offers its own perspective on the concept of knowledge (and distinguishes it from information). It views knowledge as the result of a hard computation. Thus, whatever can be efficiently done by anyone is not considered knowledge. In particular, the result of an easy computation applied to publicly available information is not considered knowledge. In contrast, the value of a hard to compute function applied to publicly available information is knowledge, and if somebody provides you with such a value then they have provided you with knowledge.

And two types of quasi-realist ethical standpoints that arise from Hume, responding to Lazari-Radek and Singer's [2014] summary of expressivism, and also that by Parfitt in On What Matters:

Existence internalism [eg Bernard Williams]:

a consideration is a reason for an agent to P if and only if there is a sound deliberative route for the agent from her pre-existing motivational set to P-ing via that consideration. On this view then, it is true that what reasons an agent has at a given moment is a function of her desires (plus other items in the motivational set, plus sound deliberation, plus imagination)

Expressivism [eg Simon Blackburn]:

Expressivism is a semantic (or better still a metasemantic) view about the meaning of normative sentences/utterances and the content of normative judgments. This means that it tells a story of what normative sentences mean on the basis of the desire-like, world-to-mind direction of fit mental states, which these judgments express.

20160112: Hao Wang (1921-1995)

Logician cum philosopher cum theoretical computer scientist who worked with Goedel. His paradox for intuitionists Floyd [2011]:

Some numbers are, intuitively, small [or accessible or feasible], and others large. If a natural number n is small, then so is n+1. But then all natural numbers n are small, by induction.

The paradox suggests "a gray area between the analytic and the synthetic"... The idea that we could single out the operation of adding 1 as a clear basis for getting larger and larger numbers is itself "an act of abstraction, to give form to a range of nebulous relations of order"

Dummett extended this to other domains eg redness.

20160112: Barcoo Fever

In the The Bone is Pointed (1947), Bony diagnoses himself as suffering from Barcoo sickness while working in the Channel country.

In another section, the sacred stones used while pointing the bone are "maralingalitha", the British nuclear tests were not until 1956-7.

Hayman [1992] mentions:

A disease known variously as "Barcoo fever, Barcoo spews, Barcoo sickness", or simply "the Barcoo", once prevalent in outback northern and central Australia.

The disease had features of a toxic rather than an infectious illness and had the characteristics of poisoning by toxic cyanobacteria (blue-green algae). In particular the symptoms were similar to those shown to be due to the hepatotoxin of the tropical cyanobacterium Cylindrospermopsis raciborskii (Woloszynska).

Poisoning by cyanobacterial toxins was once widespread in outback northern Australia and toxic cyanobacteria must still be present in many areas. Although not reported and probably not diagnosed as such, the disease still occurs in mild form. Widespread illness does not occur but individuals still experience symptoms similar to those described a century ago. Precautions are necessary to prevent algal contamination or proliferation in domestic or communal water supplies.

Griffiths and Saker [2003]:

Poisoning of humans resulting from consumption of water affected by the toxic cyanobacterium Cylindrospermopsis raciborskii was first reported almost 20 years ago from Palm Island, northern Queensland, Australia.

20160111: Entropy and information

Parrondo et al [2015]:

Shannon entropy coincides with the equilibrium entropy for canonical equilibrium states

p0(x) = e-bH0(x)/Z0,

where H0(x) is the Hamiltonian,
b = 1/(kT), (with T the temperature) and
Z0 is the partition function.

In this case, we recover the thermodynamic relation

F = E - TS

between free energy F(p0) = -kT ln Z0, average internal energy E = <H0> p0, and Shannon entropy S(p0).

One can go further and define a non-equilibrium free energy for a generic statistical state p of a system in contact with a thermal bath and with Hamiltonian H0 (refs 22-26):

F(p; H0) = <H0> p - TS(p)

It can be shown that the Shannon entropy and the associated non-equilibrium free energy are analogous to their equilibrium counterparts in non-equilibrium isothermal processes. Here, isothermal implies that the system is in contact with a thermal reservoir at temperature T , although the system itself may not have a well-defined temperature. As illustrated in Box 1, the minimal work, on average, necessary to isothermally drive the system from one arbitrary state to another is simply the difference, deltaF, between the non-equilibrium free energy in each state. The excess work with respect to this minimum is the dissipated or irreversible work, Wdiss .

We can also identify an entropy production as the increase of Shannon entropy in the system plus the increase in thermodynamic entropy in the reservoir (refs 23,26-28)

T delta Stot = Wdiss = W - deltaF ≥ 0

This is the extension of the second law to isothermal processes connecting non-equilibrium states, and forms a basic tool to develop the thermodynamics of information. [...]

To turn this observation into a statement about free energy, we focus on measurements where neither the Hamiltonian nor the microstate of the system is affected. The result is that the measurement does not change the average energy, and the non- equilibrium free energy increases by

deltaFmeas = -T deltaSmeas = kT I(X;M)

[where I() is the mutual information].

Because the mutual information is positive, measurement (or information acquisition) always increases the free energy, which in turn increases the work that can be extracted isothermally. This increase, as suggested by equation (3), must be accompanied by work.


This means that we can extract an amount of work, on average, proportional to the information acquired in the measurement. For error-free measurements, the value of M is unequivocally determined by X , and the mutual information is the Shannon entropy of the measurement I (X(tms);M) = H(M). Remarkably, the Szilárd engine reaches this bound, because M is left or right with probability 1/2, giving H(M)=ln2.

22. Gaveau, B. & Schulman, L. A general framework for non-equilibrium phenomena: The master equation and its formal consequences. Phys. Lett. A 229, 347­353 (1997).

23. Esposito, M. & Van den Broeck, C. Second law and Landauer principle far from equilibrium. Europhys. Lett. 95, 40004 (2011).

24. Still, S., Sivak, D. A., Bell, A. J. & Crooks, G. E. Thermodynamics of prediction. Phys. Rev. Lett. 109, 120604 (2012).

25. Horowitz, J. M., Sagawa, T. & Parrondo, J. M. R. Imitating chemical motors with optimal information motors. Phys. Rev. Lett. 111, 010602 (2013).

26. Deffner, S. & Lutz, E. Information free energy for nonequilibrium states. Preprint at http://arxiv.org/abs/1201.3888 (2012).

27. Hasegawa, H-H., Ishikawa, J., Takara, K. & Driebe, D. J. Generalization of the second law for a nonequilibrium initial state. Phys. Lett. A 374, 1001­1004 (2010).

28. Takara, K., Hasegawa, H-H. & Driebe, D. J. Generalization of the second law for a transition between nonequilibrium states. Phys. Lett. A 375, 88­92 (2010).

The maximum work formulation of the second law of thermodynamics is generalized for a transition between nonequilibrium states. The relative entropy, the Kullback-Leibler divergence between the nonequilibrium states and the canonical distribution, determines the maximum ability to work. The difference between the final and the initial relative entropies with an effective temperature gives the maximum dissipative work for both adiabatic and isothermal processes. Our formulation reduces to both the Vaikuntanathan-Jarzynski relation and the nonequilibrium Clausius relation in certain situations. By applying our formulation to a heat engine the Carnot cycle is generalized to a circulation among nonequilibrium states.

Ito and Sagawa [2015] on thermodynamics of sensing and feedback eg in E. coli.

Weilenmann et al [2015] use the framework of Lieb and Yngvason to axiomatize a relationship between thermodynamic and information-theoretic entropies. Lieb and Yngason's model is of the system being able to adiabatically move between states by interacting with an idealized device that stores energy (?) using a "weight" that may be left higher or lower after the interaction (has stabilized). Better, I think, is Faist et al [2015] who "determine the minimal work required to carry out any logical process [to be] given by the entropy of the discarded information conditional to the output of the computation.

We emphasize that the phenomenon observed here is fundamentally different from the notion of thermodynamic irreversibility. Here we always consider the optimal procedure for implementing the logical process, whereas a thermodynamically irreversible process is in fact an "inefficient" physical process that could be replaced by a more efficient, reversible one. In our framework, the thermodynamically irreversibile processes are those physical implementations that do not achieve the bound.
Phil Trans Roy Soc A (Feb 2016) Theme issue "DNA as information"

Added 20180314:

Astumian et al [2016] The Physics and Physical Chemistry of Molecular Machines

And from Chang et al [2017] Design and Synthesis of Nonequilibrium Systems (again Astumian is an author):

Energy ratchet mechanism

A mechanism by which directed motion and pumping is driven by external or light-driven modulation of both the depths of energy wells (relative state energies) and the heights of energy barriers (relative transition state energies)

Information ratchet mechanism

Mechanism by which directed motion and pumping is controlled by modulation of energy barrier heights in selected states where allosteric feedback is used to determine in which states the energy barriers are raised and lowered. All chemically activated biomolecular motors and pumps driven by catalysis of an exergonic chemical reaction (ATP hydrolysis or ion transport) operate as information ratchets.

Added 20190726: Harrington et al [ https://arxiv.org/abs/1811.07708] couple a qubit (pseudospin-half system) to a microwave cavity, so that the cavity resonance shifts depending on the qubit state (semiclassically). The measurement is acting as the environment, and can exhibit back action, where the measurement can reverse the qubit state, allowing time-reversal and testing of a generalised (quantum) Jarzynski fluctuation theorem. Continued monitoring of the system leads to greater and greater probability of irreversibility:

Using experimental data we have shown that a statistical arrow of time emerges fundamentally in quantum measurement, where information and backaction arise from entanglement with a fluctuating environment.

Added 20200724:

Gavrilov et al [2018]:

Our system is a micron-scale colloidal particle in water, in a virtual double-well potential created by a feedback trap. We measure the work to erase a fraction of a bit of information and show that it is bounded by the Shannon entropy for a two-state system...

[Previously,] the functional form of entropy has not been directly measured but is rather inferred from other quantities, such as the integral of the specific heat divided by temperature. Here, we measure the work required to erase a fraction of a bit of information, isolating directly the change in entropy for a system that is initially or finally in local but not global equilibrium (i.e., "weakly" nonequilibrium)...

[We can test] explicitly aspects of the continuum version of the entropy. For example, our data are consistent with a conditioned version of Crooks relation (SI Appendix). Further, the measurements presented here also confirm the identification between the total entropy production Stot and the relative entropy between the forward and backward path measures

Muschik [2016] tries to generate phenomenological (ie macroscopic) thermodynamic axioms from stochastic thermodynamics, especially Jarzynski's inequality.

(I) the exponential mean process work obeys the second law
(II) the non-regular processes have an extent as great as Lemma I allows

Non-regular processes are sometimes confusingly called "violations" of the second law of thermodynamics.

If Jarzynski's equality holds,

non-stochastic processes are always reversible, and irreversible processes generate stochasticity.

Beyond Jarzynski's equality,

(III) The extent of the non-regular processes (the non-regular admixture) is not greater than that of the regular processes

(IV) In equilibrium and consequently for reversible "processes" the fractions of non-regular and regular processes are equal.

20160108: Teleonomy

Teleonomy is the term introduced by Pittendrigh [1958] to refer to processes having the appearance of being purposive. Monod [1972] thought the genesis of "projective" [having a project or purpose] systems from an objective universe is "the central problem of biology" [ Pross 2005].

Weber [Biopoetics 2016, Ch 4]:

Since Kant the old Aristotelian principle of an organismic teleology has been regarded as illusion (Weber and Varela 2002). Biologists in Kant's wake have agreed to call organic phenomena "teleonomic" - a rather weak compromise to avoid the strong connotations of teleology.

20160108: JC Harsanyi

Economist (1994 Nobel with John Nash) and political and moral philosopher working on game theory, Bayesian reasoning and utilitarianism. Was at UQ for two years.

Bayesian rationality postulates are absolutely inescapable criteria of rationality for policy decisions [... and] together with a Pareto optimality requirement, entail utilitarian ethics as a matter of mathematical necessity.

I will describe the criteria guiding an individual when he is honestly trying to make an impartial and impersonal moral value judgement...any evaluative judgment based on biased, partial and personal criteria will be a mere judgment of personal preference.

Pareto optimality is a rather noncontroversial moral principle: ...at least one of n individuals personally prefers social situation A over social situation B, and that none of the other individuals personally prefers B over A.

Specific argument with Rawls was that the "original position" is more compatible with the average utility principle, rather than the maximin "difference" principle that Rawls derives from it. Moehler [2012] argues that Harsanyi's weak point is the requirement for impersonality.

While contractarianism assumes natural equality among the involved Bayesian agents (a la Hobbes), contractualism assumes only moral equality, as in Kant, Rawls, and Scanlon. As well as Nash bargaining, David Gauthier in Morals by Agreement suggest Kalai-Smorodinsky bargaining.

20160107: Gachter and Herrman [2015]

Gloria Origgi summarised this:

Simon Gächter and Benedikt Herrmann Reciprocity, culture and human cooperation: previous insights and a new cross-cultural experiment: "in some cultures, when people were tested in cooperative games (such as the public good game), the people who cooperated were punished, rather than the free-riders."

20160106: Gilbert Syndrome

rs8175347 is the 5-8 repeats TA STR in the TATA box in the UGT1A1 promoter. The other common functional SNP rs4124874 lies in a phenobarbital enhancer module (r2=0.67 with rs817534).

In the UK 1946 birth cohort, homozygosity of 7 repeats was associated with higher FEV1, especially in heavy smokers [ Horsfall et al 2014].

Kundur et al [2015]:

A benign, non-hemolytic, unconjugated hyperbilirubinemia [...] associated with mutation in the Uridine Glucuronosyl Transferase 1A1 (UGT1A1) gene promoter, reducing UGT1A1 activity, which normally conjugates bilirubin allowing its elimination from the blood. Individuals with GS demonstrate mildly elevated plasma antioxidant capacity caused by elevated levels of unconjugated bilirubin (UCB), reduced thiols and glutathione. Interestingly, the development of, and risk of mortality from, cardiovascular disease is remarkably reduced in GS individuals.
Bulmer et al [2013]:

Recent reports show clear protection from cardiovascular disease in this population. Protection of lipids, proteins and other macromolecules from oxidation by bilirubin represents the most commonly accepted mechanism contributing to protection in this group. However, a recent meta-analysis estimated that bilirubin only accounts for ~34% of the cardioprotective effects within analysed studies. To reveal the additional contributing variables we have explored circulating cholesterol and triacylglycerol concentrations, which appear to be decreased in hyperbilirubinemic individuals/animals, and are accompanied by lower body mass index in highly powered studies. These results suggest that bilirubin could be responsible for the development of a lean and hypolipidemic state in GS.

Buch et al [2010] report from a GWAS that cholesterol stones are increased in Gilbert's, possibly via a role for bilirubin in nucleation. The population-attributable fraction of the common ABCG8 and UGT1A1 variants in European subjects is 15% to 20%, approximating the total genetic gallstone risk:

UGT1A1 (rs6742078; P = .018) was associated with overall gallstone risk. In a sex-stratified analysis, only male carriers of rs6742078 had an increased risk for gallstone disease (P = 2.1×10-7; odds ratio(recessive), 2.34; P(women) = .47). The sex-specific association of rs6742078 was confirmed in samples from South America (P(men) = .046; odds ratio(recessive), 2.19; P(women) = .96).

Marschall et al [2013] studied MZ twins with and without Gilbert's in the Swedish Twin Registry. Gall stone disease was increases in carriers (OR=1.6).

20160106: Vivax malaria

Bertolino and Bowen [2015] review the liver stage.

During the pre-erythrocytic asymptomatic phase of malarial infection, sporozoites passing through the hepatic sinusoids recognize a Kupffer cell, bypassing it to enter nearby hepatocytes. In the average attack, schizonts will develop within parasitophorous vacuoles in less than 100 hepatocytes over 7-10 days, leading to a release of 10000-30000 exoerythrocytic merozoites.

In P. vivax infection, hypnozoites are the dormant cell type within the liver that give rise to relapse (probably ~90% of episodes in endemic areas are relapse rather than reinfection). Dormancy in the Finnish hibernans strain was at least 9 years [ Hulden and Hulden 2011]. Primaquine is the current treatment that eradicates this stage (toxic to those with G6PD deficiency).

Pigment gallstones are increased in liver pathologies such as cirrhosis, but Gilbert's increases cholesterol stones (as well as pigment stones in those with cooccurrent thalassemia etc).

20160105: Neologism

Nonce words are coinings for a single occasion - may then become institutionalized. "hapax logomena": types with one token ie "something said only once" ie makes only one appearance in corpus (flother, sassigassity).

"Neologisms - New Words since 1960" lists 2700 words coined in British English 1960-1990.

Oxford Dictionary of New words 2000 words 1980-1990.

orthoepically: pronounce an acronym

20151223: EQUATOR Publishing guidelines


Enhancing the QUAlity and Transparency Of health iResearch...

20151218: SBTC and inequality

Skill-Biased Technological Change: "[is] technological change is the primary source of wage inequality in recent decades (as opposed to policy and institutions)", specifically deunionization.

20151209: Physics, Gravity

Current theories for quantisation of gravity:

Alain Connes - derivation of the Standard Model from "one of the simplest and most natural finite noncommutative geometries of the right dimension (6 modulo 8)". Unfortunately, predicted (2007) Higgs mass would be 170 GeV, rather than 125, subsequently repaired. Can be applied to supersymmetry models.

20151207: Effectiveness of psychotherapy

Allegiance effects are significant in RCTs of psychotherapy (OR=1.3, i2=0.5) except for CBT, even though trial quality not related. Publication bias also inflates effect size estimates in depression RCTs by 25%.

Current therapies over a variety of conditions with some supportive evidence, reviewed:

Linde et al [2015] review 30 studies of interventions for depression in the primary care setting:

The overall effect size was -0.3 SD: 0.3 for face-to-face CBT, -0.15 for PST, -0.24 for face-to-face psychotherapy, -0.28 other face-to-face Rx, -0.43 for remote therapist-led CBT, -0.56 for remote therapist-led problem-solving therapy, -0.40 for guided self-help CBT, and -0.27 for no or minimal contact CBT.

Our analyses suggest that the differences between different types of psychological treatments are minor, and remote therapist-led, guided self-help, and minimal-contact approaches can yield effects similar to personalized face-to-face therapies. There are also hints that psychological therapies might be less effective for patients having minor depression and dysthymia than for patients with major depression.

20151125: ImpG

ImpG is the program described in Pasaniuc et al [2014] for imputation of association signals for untyped SNPs using summary statistics.

We assume that summary association statistics consist of z-scores known to be normally distributed with mean 0 and variance 1 under the null model of no association. LD between SNPs i and j induces a covariance between their observed z-scores, according to the correlation rij between the two SNPs. Thus, under null data, the vector Z of z-scores at all SNPs in a locus is approximately distributed as a Gaussian distribution, Z~N(0,S), with S being the correlation matrix among all pairs of SNPs induced by LD (Sij=rij). In a given study, we only observe z-scores at the typed SNPs (Zt), with no information about untyped SNPs. We estimate S using reference panels of haplotypes (e.g. 1000G) and analytically derive the posterior mean of z-scores at unobserved SNPs (Zi) given Zt and S (ImpG-Summary). We use the conditional variance to estimate the imputation accuracy (r2pred), in a manner similar to the r2hat estimator in HMM-based imputation.

20151123: Nucleotide ambiguity

IUPAC-IUB Symbols for Nucleotide Nomenclature: [Cornish-Bowden Nucl. Acids Res 1985, 13:3021-3030] lists nucleotide symbols, including those used for ambiguity:

Symbol Meaning Nucleic Acid
A A Adenine
C C Cytosine
G G Guanine
T T Thymine
U U Uracil
M A or C
R A or G Purine
W A or T
S C or G
Y C or T Pyrimidine
K G or T
V A or C or G
H A or C or T
D A or G or T
B C or G or T
X G or A or T or C
N G or A or T or C

20151120: Broad caninearray FAQ


20151118: Recording from input stream

Use audacity and pavucontrol (PulseAudio Volume Control) - switching the latter to the recording tab and turning on monitoring (of the playback stream). This then passes the output to audacity which can monitor or record. One can export to multiple formats from audacity.

20151117: Addison's disease

Soleimanpour et al [2014] show that CLEC16A is a membrane-associated endolysosomal protein that regulates mitophagy:

[Mitophagy is] a selective form of autophagy necessary for mitochondrial quality control...Clec16a has been identified as a disease susceptibility gene for type 1 diabetes, multiple sclerosis, and adrenal dysfunction... Islets from mice with pancreas-specific deletion of Clec16a have abnormal mitochondria with reduced oxygen consumption and ATP concentration, both of which are required for normal beta-cell function. Indeed, pancreatic Clec16a is required for normal glucose-stimulated insulin release. Moreover, patients harboring a diabetogenic SNP in the Clec16a gene have reduced islet Clec16a expression and reduced insulin secretion.

The human diabetes-associated CLEC16A SNP rs12708716 is an eQTL for CLEC16A in islet cells, with G/G having half the expression level of A/A.

In peripheral blood CEU data, no such association (Kruskal-Wallis H=1.1, df=2, P=0.5696).

PH_hs_0033019 is Phalanx HOA probe for CLEC16A

Dog Leukocyte Antigen region is CFA12:3.32-5.83 Mbp. A number of diseases are reported associated, especially:

hypoadrenocorticism in the Nova Scotia duck tolling retriever [Hughes 2010]

20151117: Creek Track at Bunyaville

Rode the Creek track, circling back along the Jinker Track, and back down to Bunker Road near the dam. Very pleasant.

20151115: JSFS

The joint site frequency spectrum counts the number of single-nucleotide polymorphisms (SNPs) in a sequence for which the derived allele occurs in each population. For example, J[a,b] = 5 means that there are 5 positions in the sequence at which the derived allele is found in exactly a individuals of Population 1 and in b individuals of Population 2. This can binned, and composite likelihood methods used to estimate population genetic model parameters.

Mathew and Jensen [2015] discuss using the JFSF of multiple populations to estimate both demographic model parameters and selection coefficients. They describe an extension of the jaatha package (available in R). The publically available package currently only implements the neutral model approach.

20151115: BCL2L12 and skin cancer

BCL2L12 19:49665142..49673916 (b37: 50168399..50177173)

BCL2-like 12 (proline rich): encodes a member of a family of proteins containing a Bcl-2 homology domain 2 (BH2). The encoded protein is an anti-apoptotic factor that acts as an inhibitor of caspases 3 and 7 in the cytoplasm. In the nucleus, it binds to the p53 tumor suppressor protein, preventing its association with target genes. Overexpression of this gene has been detected in a number of different cancers.

Gartner et al [2013] found a recurrent F17F somatic mutation in 12 of 285 melanomas. This mutation led to increased BCL2L12 mRNA and protein levels because of differential targeting of WT and mutant BCL2L12 by hsa-miR-671-5p. Protein made from mutant BCL2L12 transcript bound p53, inhibited UV-induced apoptosis more efficiently than WT BCL2L12, and reduced endogenous p53 target gene transcription.

MIR671 7:151238421..151238538 (b37: 150935507..150935624)

20151113: Strand for HumanCoreExome

Note that alleles in HumanCoreExome-12v1-0_B.annotated.txt are [ALT/REF].

Also that there are least 6 versions as per http://www.well.ox.ac.uk/~wrayner/strand/:


20151110: CTLA-4 inhibitors for melanoma

Model fitting to overall survival in the Hodi et al [2010] ipilimumab trial for advanced melanoma.
Study MedianOS1Y 2Y 3Y 10Y
Schadendorf11.4mo0.47 0.29 0.22 (0.2-0.24)
Zimmer . 0.38 (0.27-0.49)0.22 (0.13-0.33).
Ahmad 6.1mo 0.31 0.15 .
Larkin 10.0m 0.44 0.22 0.18 0.044
Normal Care
Larkin 6mo 0.25 0.13 0.07 0.004
Hersh 10.5mo0.53 0.27 0.20 .
Selby 2.2mo 0.42
Robert 9.1mo 0.36 0.18 0.12
Huncharek 7.4mo
Sasse 0.35

Jiang et al [2014] meta-analyze 9 studies (N=2481)

Hersh et al [2015] compare nab-paclitaxel to dacarbazine (nab denotes albumin bound nanoparticles).

Robert et al [2011] test ipilimumab/dacarbazine to dacarbazine. Reanalysis in Sherrill et al [2013].

Larkin et al [NEJM 2015] compare anti-CTLA4, anti-PD1, combination in the Checkmate 067 trial: "OS data are at present insufficiently mature"

Huncharek [2001] meta-analyse trials of dacarbazine (DTIC)

Sasse et al [2007] carried out a Cochrane review:

1 year survival:

20151110: R NIMBLE

NIMBLE is a package that compiles an extended version of the BUGS language to R-callable C++ (Duncan Temple Lang and co.).

The GLMM example http://r-nimble.org/examples-2.

## define the model
code <- nimbleCode({
    beta0 ~ dnorm(0, sd = 10000)
    beta1 ~ dnorm(0, sd = 10000)
    sigma_RE ~ dunif(0, 1000)
    for(i in 1:N) {
        beta2[i] ~ dnorm(0, sd = sigma_RE)
        logit(p[i]) <- beta0 + beta1 * x[i] + beta2[i]
        r[i] ~ dbin(p[i], n[i])

## constants, data, and initial values
constants <- list(N = 10)

data <- list(
    r = c(10, 23, 23, 26, 17, 5, 53, 55, 32, 46),
    n = c(39, 62, 81, 51, 39, 6, 74, 72, 51, 79),
    x = c(0,  0,  0,  0,  0,  1, 1,  1,  1,  1)

inits <- list(beta0 = 0, beta1 = 0, sigma_RE = 1)

## create the model object
Rmodel <- nimbleModel(code=code, constants=constants, data=data, inits=inits, check = FALSE)
Rmcmc <- buildMCMC(Rmodel)

## Compile the model and MCMC algorithm

Cmodel <- compileNimble(Rmodel)
Cmcmc <- compileNimble(Rmcmc, project = Rmodel)

samples <- as.matrix(Cmcmc$mvSamples)

20151031: brms

The brms package provides an interface to fit bayesian generalized linear mixed models using Stan. Syntax is very similar to lme4:

fit <- brm(count ~ log_Age_c + log_Base4_c * Trt_c + (1|patient) + (1|visit) + (1|obs),
           data = epilepsy, family = "poisson")
#> Compiling the C++ model

20151029: Recent papers in hemangiosarcoma

SymbolEnsembl geneCFAStartEnd
TP53 ENSCAFG000000167145 3256059832574109
NF1 ENSCAFG000000185929 4146716541679382
PTPRB ENSCAFG00000000446101227335412384706
MYC ENSCAFG00000001086132520039225205878
POT1 ENSCAFG00000001754141098687111070651
BRAF ENSCAFG00000003907168222323 8316807
NRAS ENSCAFG00000009532175241276952418204
HRAS ENSCAFG00000031029182564228125645415
CXCR4 ENSCAFG00000005109193887399238877889
PLCG1 ENSCAFG00000009082242919855529212172
CRKL ENSCAFG00000015188263054199930581964
MAPK1 ENSCAFG00000015421263101360731132998
KRAS ENSCAFG00000011428272226160222292402
CXCL12ENSCAFG00000007026282895915 2899207

MicroRNA-214 Promotes Apoptosis in Canine Hemangiosarcoma

CXCR4 and CXCL12 upregulated in hemangiosarcoma

Median survival time of dogs with splenic hemangiosarcoma is 1.6 months post-splenectomy, and slightly improved by chemotherapy.

Serum big ET-1 levels in dogs with HSA were significantly (P <0.01) higher than in other dogs: 17 healthy controls, 11 benign splenic lesions, 10 non-HSA splenic tumours.

In one human hepatic angiosarcoma, sequencing found amplification of KIT, PDGFRA and VEGFR2. Treatment with sunitinib lead to appearance of a PLCG1*R707Q mutation in metastases - possibly bypassing VEGFR2 dependence.

Murali et al 2015:

Angiosarcomas are rare malignant mesenchymal tumors of endothelial differentiation. The clinical behavior is usually aggressive and the prognosis for patients with advanced disease is poor with no effective therapies. The genetic bases of these tumors have been partially revealed in recent studies reporting genetic alterations such as amplifications of MYC (primarily in radiation-associated angiosarcomas), inactivating mutations in PTPRB and R707Q hotspot mutations of PLCG1. Here, we performed a comprehensive genomic analysis of 34 angiosarcomas using a clinically-approved, hybridization-based targeted next-generation sequencing assay for 341 well-established oncogenes and tumor suppressor genes. Over half of the angiosarcomas (n = 18, 53%) harbored genetic alterations affecting the MAPK pathway, involving mutations in KRAS, HRAS, NRAS, BRAF, MAPK1 and NF1, or amplifications in MAPK1/CRKL, CRAF or BRAF. The most frequently detected genetic aberrations were mutations in TP53 in 12 tumors(35%) and losses of CDKN2A in9 tumors (26%). MYC amplifications were generally mutually exclusive of TP53 alterations and CDKN2A loss and were identified in 8 tumors (24%), most of which (n = 7, 88%) arose post-irradiation. Previously reported mutations in PTPRB (n = 10, 29%) and one (3%) PLCG1 R707Q mutation were also identified.

Calvete et al [2015] found R117C in the POT1 gene is responsible for cardiac angiosarcoma in TP53-negative Li-Fraumeni-like families, as well as in a breast angiosarcoma family. This mutant appears once in the ExAC collection (121324 alleles).

shelterin complex: POT1, TRF1, TRF2, TIN2, TPP1 and RAP1.

20151021: Airway branching

Bower et al [2014] show

In the vertebrate lung, deleting enhanced green fluorescent protein (eGFP)-labeled intrinsic neurons using a two-photon laser leaves adjacent cells intact, but abolishes branching.

Several important genes in airway branching morphogenesis: SHH, Yy1 (a ZFTF), Spry2.

20151016: Testing Sib-pair zlib related speed again

>> file wc All_hair5.assoc.txt.gz

Field counts for "All_hair5.assoc.txt.gz":

L 1 Len 253 NFields 22: "chr     rs      ps      n_miss  allele1 allele0 af"

Number of lines               = 9130946
Length of longest line        = 296750 chars
Total number of words         = 200880812
Maximum words per line        = 22
Constant word count per line? = T
Length of longest word        = 32745 chars
[   69.75 s]

$ time zcat All_hair5.assoc.txt.gz | wc
9130946 200880812 2177418095

real	0m40.135s
user	1m3.215s
sys	0m1.002s

20151014: Formal concept analysis

The approach of Wille applying lattice and ordered sets to concepts and concept hierarchies as graph of binary cross-table of concepts (incidence matrix, formal context). Current applications in computing - dependency graphs, database conceptual schemas, text mining, semantic web, diagnostic coding, linguistics, finance. Extended to rough and fuzzy sets. Some relation to Boolean factor analysis, and can be basis for cluster analysis. Temporal Concept Analysis incorporates ordering and trajectory information.

Programs for this are LatDraw http://latdraw.org/ and ToscanaJ (see the Tockit website.

20151013: Primitive roots

mupad has numlib::prevprime and numlib::primroot. At moment, only runs on orpheus!

20151012: Thermodynamics of replication etc

England [2013] discusses the statistical physics of self-replication

We often think of the main entropic hurdle that must be overcome by biological self-organization as being the cost of assembling the components of the living thing in the appropriate way. Here, however, we have evidence that this cost for aerobic bacterial respiration is relatively small, and is substantially outstripped by the sheer irreversibility of the self-replication reaction as it churns out copies that do not easily disintegrate into their constituent parts.

More significantly, these calculations also establish that the E. coli bacterium produces an amount of heat less than six times (220npep/42npep) as large as the absolute physical lower bound dictated by its growth rate, internal entropy production, and durability.

This was cited by a related paper Bartolotta et al [2015] on the "Bayesian Second Law of Thermodynamics", which incorporates measurement, thus removing the inconsistency between Shannon/Gibbs entropy and Boltzmannian entropy. GE Crooks, Phys Rev E 60:2721 (1999).

And by Gharibyan and Tegmark [2013] who provide QM "sharpening":

Observation: When an object is probed by the subject, its entropy on average decreases.

Decoherence: When an object is probed by the environment, its entropy increases.

20151005: Boltzmann Brains

Coined by Allbrecht and Sorbo [2004], but concept explored earlier by Martin Rees [Before the Beginning 1997], according to Vilenkin [2006]. If pocket universes arise within "our" eternally inflating universe, "real" observers arising in these may outnumber "freak" observers, plus if brains, why not planets that also give rise to life and intelligence [Gott 2008].

Gott also discusses observing a BB, especially those that might arise out of Unruh radiation that is seen accelerating relativistically (so is observer dependent). He echoes Hume in suggesting the "freak" appearance of a BB is more likely than an actual BB forming, and further comments they will be usually be short-lived - so the Copernican principle excludes me being a BB.

20151003: Origins of musical cognition

Human locomotion more rhythmic than other primates, and need to suppress sound of this when listening to the environment. Humans tend to walk in step:

Larsson [2013]:

According to Changizi (2011), the most informative sounds of moving individuals are the basic building blocks of music. Four properties of moving individuals correspond directly to four fundamental ingredients of music: (1) the distance to the sound source (i.e., the moving individual) corresponds to loudness in music, (2) directionality influ ences pitch through the Doppler effect, (3) the moving individual's speed corresponds to tempo in music, and (4) the moving individual's gait pattern corresponds to the rhythm in music.

Antonovic points out that descriptions of music always require metaphorical language from other domains eg visuospatial

20151003: The prelinguistic origins of concepts

Mandler [2010] thinks the basic concepts acquired by the developing infant are visuospatial, the primitives (image schema) being:


One important expansion is analogization of these into other domains, eg time.

Spatial information is readily available even to very young infants; it is the most continuous, observable, and structured information about the world that they have. Bodily feelings are not only more intermittent, they are often unstructured, and even for adults difficult to describe. They add experiential richness to action and event concepts, but that is not the function that I believe image-schemas are used for, which is to put information in a form that can stand for (symbolize, if you will) observations about the world, thus enabling both thought and language understanding.

Vandervert [2017] argues "that Mandler's conceptual primitives are encoded as cerebellar internal models in the infant in accordance with...[the] cerebellar sequence detection process...[M]athematics and physical laws are products of the computational mechanisms of the human cerebellum, and that physical laws are, indeed, the result of the cerebellum's distillation of the features objects moving in space". He differentiates between cerebellar dynamics models (paravermis-interpositus) from inverse dynamics models (hemisphere-dentatus) that automatize cortical tasks [Ito 1993; Movement and thought: identical control mechanisms by the cerebellum]. Vandervert claims number sense and mathematical imagination depend on higher order cerebellar inverse dynamic models, and so are cerebro-cerebellar. Cerebellar sequence detection is supposed central to unconscious goal directed thought. Specifically, Marvel and Desmond [2015] claim inner speech is involved in verbal working memory ("Medial regions of the anterior cerebellum support overt speech, consistent with other forms of motor execution such as finger tapping, whereas lateral portions of the superior cerebellum support speech planning and preparation e.g., covert speech"), that may not be accessible to conscious recall.

Igloi et al [2015]:

[C]ognitive aspects of navigation mainly induce activity in ... cerebellar lobule (VIIA Crus I). Our results demonstrate a functional link between cerebellum and hippocampus in humans and identify specific functional circuits linking lobule VIIA Crus I of the cerebellum to medial parietal, medial prefrontal, and hippocampal cortices in nonmotor aspects of navigation. They further suggest that Crus I belongs to 2 nonmotor loops, involved in different strategies: place-based navigation is supported by coherent activity between left cerebellar lobule VIIA Crus I and medial parietal cortex along with right hippocampus activity, while sequence-based navigation is supported by coherent activity between right lobule VIIA Crus I, medial prefrontal cortex, and left hippocampus.

20151003: Poetics and linguistics

Dingemanse et al [2015] discuss nonarbitrariness of vocabulary:

Iconicity, "in which aspects of the form and meaning of words are related by means of perceptuomotor analogies", eg onomatopoeia. Relative or diagrammatic iconicity is where relations between multiple forms resemble logical relations between meanings.

Systematicity, where statistical features such as vowel quality, syllable duration and stress differentiate different lexical classes (eg nouns v. verbs) or semantic features such as concreteness, magnitude and spatial demonstratives. Morphological complexity eg "oak tree" indicated relationships, but again not cross-language - relative arbitrariness.

Arbitrariness allows flexibility of concepts.

Note that Dingemanse took the IgNobel in 2015 for universality of "huh".

20151001: Dancing with Pixies (JM Bishop)

Refers to reductio where any possible computation will be performed "randomly" in a sufficiently complex physical system.

One solution is panpsychism or some type of emergentist position, another is Tononi Integrated Information Theory.

Henry Stapp is another physicist (LBNL, Berkeley) with QM explanations of mind and consciousness. Points out that some past events are indeterminate until present choice made.

20150929: Interbreed distances

Boyko et al [2010] paper on dog breed morphology:

GoldenGSD 0.02810.33440.3152(N=100220)
GSD Pugs0.04340.39050.3628(N=92563)

20150917: Word of the day

Pogonotomy - beard removal.

20150913: Tibro

Line of Credit

Abseiled off Z in 3 pitches - straightforward.

Line of Credit is really nice - P1 high angle slab with lots of crimpers. P2 sidles up steep headwall - traverse L along nice ledge then up aretey bit. All well bolted.

20150912: Walk Brisbane Forest Park

England Creek, Right Branch.

Left car at Mt Nebo Refuse Transfer Station ~0930
Returned ~1610

Very pleasant after initial dense lantana forced us to leave the creek for 50 m or so. Further down, nice rainforest - picabeens palms, a couple of deep swimming holes (lots of big tadpoles), waterfalls (dry) and little gorges requiring bypassing a couple of times.

On Side Break, four 4WDs arrived at creek crossing as we did. Ray was looking forward to climbing Side Break, as ascends 300 m in 1000 m. Was 3.5 km then along ridgetop road.

20150822: Epithelial migration in the EAC

Haven't you wondered how it works?

Revadi et al [2011]: Epithelial migration on the external ear canal wall in normal and pathologic ears

Twenty-five subjects with normal ears and 4 with keratosis obturans were recruited for the study. Colored ink dots were placed around the tympanic annulus at the 12, 3, 6, and 9 o'clock positions. Migration patterns and the rate of travel of these ink dots were examined and photographed until the ink dots reached the bony cartilaginous junction.

Fifteen healthy subjects and 1 with bilateral KO completed the study. The ink dots migrated laterally, with a rate of migration in normal ears between 42 and 205 um/d. The mean rates for each quadrant, measured clockwise from the 12 o'clock position, were 104.93, 89.80, 72.67, and 109.93 um/d, respectively.

20150822: SETDB1

Elsasser et al [2015] discuss the role of SETDB1 (alias ESET, KMT1E), where it carries out the trimethylation of H3K9me3 silencing the associated genes, and most importantly, endogenous retroviruses. The latter are characterized by the presence of H3K9me3 and of H3.3, though H3K9me3 seems important only in embryonic development - in adult tissues methylation may be more critical. That is, it establishes (long term) silencing. Knocking H3.3 out leads to diminished H3K9me3, and increased expression not just of ERVs/transposons but of nearby genes. Transposon activity might be a mechanism of genomic instability.

So Collins et al [2015] report SETDB1 knockout increases ERV activity in murine B lymphocytes.

Wu et al [2014] find upregulating SETDB1 in lung cancer suppresses filopodia formation, migration, and invasion. By contrast, Sun et al [2015] found upregulation enhanced non-small cell lung cancer tumour size. Kostaki et al [2014] found SETDB1 expression to be higher in more advanced melanomas, and associated with p16(INK)(4A) methylation.

20150817: Various population genetic results

Sampling variance of heritability:

The proportion of variance explained by all SNPs, h2G, has sampling variance approximately 105/N2, so ASE ~ 316/N from unrelated samples [Visscher et al 2014]

Kemper et al [2014] test various selection signature tests at cattle loci underlying traits under strong selection. Aside from megaphenic loci such as coat colour (PMEL) and muscle doubling (MSTN), loci such as DGAT1, MSTN, PLAG1, GHR, ABCG2 gave ambiguous results for "a modified version of Depaulis-Veuille's H-test (referred to as haplotype homozygosity, HAPH), the integrated haplotype score (iHS), and Wright's measure of population differentiation (FST)". If the favourable allele was already relatively common in the originating population, the selection signal will not be strong.

20150807: Quantum Darwinism

Wojciech Zurek's concept:

Hilbert space is a big place, exponentially larger than the arena of classical physics. The Hilbert space of macroscopic systems is dominated by states that have no classical counterparts. Yet the world observed by macroscopic observers exhibits powerful regularities that make it amenable to classical interpretations on a broad range of scales.

How do we explain this? The answer, of course, is that Hilbert space is not sampled uniformly; rather, the initial state and the Hamiltonian governing evolution are both very special. Quantum Darwinism is a framework for describing and quantifying what distinguishes quasi- classical states awash in the enormous sea of Hilbert space.

This insight captures the essence of Quantum Darwinism: Only states that produce multiple informational offspring - multiple imprints on the environment can be found out from small fragments of E [the environment]. The origin of the emergent classicality is then not just survival of the fittest states (the idea already captured by einselection), but their ability to "procreate", to deposit multiple records - copies of themselves - throughout E.

Proliferation of records allows information about S to be extracted from many fragments of E (in the example above, photon E). Thus, E acquires redundant records of S . Now, many observers can find out the state of S independently, and without perturbing it. This is how preferred states of S become objective. Objective existence - hallmark of classicality - emerges from the quantum substrate as a consequence of redundancy.

20150801: Epistasis

Cheng et al (2011)

For example, Brockmann et al. (2000) estimated that about 33% of the phenotypic variation of body weight in mice could be attributed to epistasis. Another study indicated that there might be a time-dependent pattern for these genetic effects and epistasis might be more important for early growth traits (Carlborg et al., 2003).

20150724: writing a CSV in Fortran


20150708: Power et al [2015], Martin et al [2015]

Polygenic risk scores for schizophrenia and bipolar disorder predict creativity

We tested whether polygenic risk scores for schizophrenia and bipolar disorder would predict creativity. Higher scores were associated with artistic society membership or creative profession in both Icelandic (P = 5.2 × 10 and 3.8 × 10-6 for schizophrenia and bipolar disorder scores, respectively) and replication cohorts (P = 0.0021 and 0.00086). This could not be accounted for by increased relatedness between creative individuals and those with psychoses, indicating that creativity and psychosis share genetic roots.

The relative contribution of common and rare genetic variants to ADHD

Price et al [2015] used PRS for ADHD [ Hamshere et al 2013], and CNV burden. A 1 SD increase in PRS gave OR=1.15, the presence of any rare CNV >500 kbp, OR=1.63.

20150626: SCRT (sucrose transporter)

Bin et al [2015] mention:

SLC45A1, SLC45A2 (MATP), SLC45A3, and SLC45A4 [18], [have] high similarity to a recently identified functional animal sucrose transporter (SCRT) in Drosophila melanogaster. SCRT is similar to plant sugar uptake transporters (SUTs) containing a typical sucrose transporter sequence (RxGRR). Interestingly, the expression of SCRT is enriched in melanin-containing organelles as well as in the gastrointestinal tract, suggesting a possible role for sucrose transporters in melanin synthesis [Meyer et al 2011].

20150624: On automation

leif.roar@dimnakorr.com on r.a.sf.w writes:

I believe the traditional phrasing is "Go away or I will replace you with a very small shell script."

20150618L Folate and UV

Branda RF, Eaton JW. Skin Colour and Nutrient Photolysis: An Evolutionary Hypothesis. Science 1978;201(4356):625 - 626.

Akhtar MJ, Khan MA, Ahmad I. Photodegradation of folic acid in aqueous solution. J Pharmaceut Biomed 2009;25:269 - 275.

Off MK, Steindal AE, Porojnicu AC, et al. Ultraviolet photodegradation of folic acid. Photochem Photobiol 2005;80:47 - 55.

Juzienne A, Tam TTT, Iani V, et al. Methyltetrahydrofolate can be photodegraded by endogenous photosensitisers. Free Rad Bio Med 2009;47:1199 - 1204


Serum and erythrocyte folate status in patients with psoriasis and healthy volunteers was measured before and after exposure to solar radiation, broadband UVB or use of sunbeds. In some cases plasma homocysteine and serum 25-hydroxyvitamin D (25(OH)D) were also measured. Serum and erythrocyte folate levels in healthy volunteers and in psoriasis patients were not influenced to any statistically significant extent after exposure to solar radiation, to single or to multiple UV treatments. However, a slight decay of blood folates and an increase of plasma homocysteine levels were observed in psoriasis patients after exposure to UV radiation. Exposure to sun or sunbeds does not have any significant effect on the levels of blood folate of healthy humans.

http://www.ncbi.nlm.nih.gov/pubmed/24509071 Borradale D, Isenring E, Hacker E, et al. Exposure to solar ultraviolet radiation is associated with a decreased folate status in women of childbearing age. J Photochem Photobiol B: Biology. 2014 Feb 5;131:90-5.

20150526: Iterated Prisoner's Dilemma

Press and Dyson [2012]

The two-player Iterated Prisoner's Dilemma game is a model for both sentient and evolutionary behaviors, especially including the emergence of cooperation. It is generally assumed that there exists no simple ultimatum strategy whereby one player can enforce a unilateral claim to an unfair share of rewards. Here, we show that such strategies unexpectedly do exist. In particular, a player X who is witting of these strategies can (i) deterministically set her opponent Y's score, independently of his strategy or response, or (ii) enforce an extortionate linear relation between her and his scores [Zero Determinant strategies]. Against such a player, an evolutionary player's best response is to accede to the extortion. Only a player with a theory of mind about his opponent can do better, in which case Iterated Prisoner's Dilemma is an Ultimatum Game.
Lee et al [2015]
We show that the history of play in a population game contains exploitable information that can be successfully used by sophisticated strategies to defeat memory-one opponents, including zero determinant strategies. The history allows a player to label opponents by their strategies, enabling a player to determine the population distribution and to act differentially based on the opponent¿s strategy in each pairwise interaction. For the Prisoner¿s Dilemma, these advantages lead to the natural formation of cooperative coalitions among similarly behaving players and eventually to unilateral defection against opposing player types. We show analytically and empirically that optimal play in population games depends strongly on the population distribution. For example, the optimal strategy for a minority player type against a resident TFT population is ALLC, while for a majority player type the optimal strategy versus TFT players is ALLD. Such behaviors are not accessible to memory-one strategies. Drawing inspiration from Sun Tzu¿s the Art of War, we implemented a non-memory-one strategy for population games based on techniques from machine learning and statistical inference that can exploit the history of play in this manner. Via simulation we find that this strategy is essentially uninvadable and can successfully invade (significantly more likely than a neutral mutant) essentially all known memory-one strategies for the Prisoner¿s Dilemma, including ALLC (always cooperate), ALLD (always defect), tit-for-tat (TFT), win-stay-lose-shift (WSLS), and zero determinant (ZD) strategies, including extortionate and generous strategies.

Some zero determinant (ZD) strategies are very successful; Adami and Hintze have shown that ZD strategies are evolutionarily unstable in general, but can be effective if opponents can be identified and play can depend on the opponent's type (including versus itself). In particular, how a strategy fares against itself becomes crucial in population games.

20150518: philosophical bits and pieces


A-series [and] B-series time concepts whereby the former refers to the culturally harnessed experience of past'present'future (yesterday-today-tomorrow) while the latter refers to the be before-and-after row of typically measurable events

IAI Talk by Peter Hacker on the concept of mind. Wittgensteinian and Analytic approach based on english uses of "mind".

"have a mind and have a body".

memory, thought, opinion, intention, cleverness and rationality, narrow or broad

term "mind" itself is eliminable, replacable with these psychological properties, and not really definable. Mind is so broad, cannot be identical with brain. Rationality and reasons are central to his thought.

Don't like science fiction on philosophy. "We are not identical with our own brain". Replace metempsychosis with brain transplantation.

ASSERTS: Error to think of brain as if is an animal - essential organ but not identical

Contra this:


20150506: 100 Year return value of 5 day total extreme precipitation

Ye and Li (2011) give

Table 1: The 100 year return value of 5 day total extreme precipitation derived from GEV distribution and observed total precipitation that caused 1974 and 2011 floods, plus my values for Toombul

Station Name ID Latitude Longitude Extreme precipitation (mm)
100 ARI 1974 flood 2011 flood
Mt Nebo Post Office 40147 -27.40 152.79 852 1118 519
Alderley 40224 -27.42 153.00 552 553 329
Toombul Bowls Club 40237 - - - 617 269

20150430: Data Recovery on Linux

Last year, was forced to use foremost (/usr/bin/foremost) to recover several documents that had not been backed up, after I stupidly ran "make_project" in the Qobischeme directory (starts with 'rm -rf "$*"'). Ended up with lots of reasonably anonymous files sorted by header type, with the Open Office files I wanted tucked under .jar (IIRC).


A review article is in LWN.

20150428: GPX to KML

gpsprune is an application for viewing, editing and converting coordinate data from GPS systems.

20150425: Mystery Track

0615Departed home
0745Binna Burra
0830Started Walking
0905Coomera Circuit turnoff
0940Start of Mystery Track
~1040Bottom of Mystery Track
1305Base of Coomera Falls
1330Departed Falls
1500Back to circuit
1700Binna Burra
1930Arrived home

Mystery Track drops down along creek where official track first comes onto the edge of the gorge (17 min walk before the Falls Lookout). Well used, with yellow tape. Pretty, crossing the creek above a falls (after 200 m or so), contours N. then drops steeply (one move to cross into a gully) 50 m, followed by a little (wet that day) scramble, down and then contours back S. in rainforest below lower falls on creek, then down to 8 m "organ pipe" clifflet/bench in rainforest at bottom. Where hits river, is pink tape on tree.

Slow for me rockhopping as some rocks slimy, though travelled faster on return, stayed more in water as shoes already wet due compulsory wade below falls. Went to Coomera Falls lookout after topping out.

20150420: Rouson et al Scientific Software Design (OO Fortran)

Fortran C++
Derived Type Class
Component Dynamic Polymorphism
select type emulate using dynamic_cast
Type-bound procedure Virtual Member function
Parent Type Base Class
Extended Type Subclass
Module Namespace
Generic interface Function overloading
Final procedure Destructor
Defined operator Overloaded operator
Defined assignment Overloaded assignment op
Deferred procedure binding Pure virtual member function
Procedure interface Function prototype
Intrinsic type/procedure Primitive type/procedure

Unified Modelling Language includes case diagram as a useful tool: [Object Oriented Analysis]

Actor, System, Uses

[Object Oriented Design]

Decompose into classes ie Abstract Data Types. The Class Diagram is a three part box:
Class Name
Attributes: data representing state of this instance,
"+"/"-" designate public/private
constructor [<<sterotype name>>]
other methods

20150416: On the word "Taoist"

Nathan Sivin (1978).  On the Word "Taoist" as a Source of Perplexity. With
Special Reference to the Relations of Science and Religion in Traditional
China History of Religions.
Vol. 17, No. 3/4, Current Perspectives in the Study of Chinese Religions
(Feb-May, 1978), pp. 303-330.

Kubo Noritada

Taoism...was founded on a variety of ancient popular beliefs; was centred about immortality lore; incorporated the lore of philosophical Taoism, the Book of Changes, yin-yang, the Five Phases, divination and apocalyptic prognostication, astrology and so on, as well a shamanistic beliefs; and was given religious forms modelled on the style and organisation of Buddhism. Taoism is a religion of benefit in this world, the chief goal of which is eternal life, exempt from ageing."

Ofuchi Ninji:

Taoism, in my view, cannot be explained except by reference to the timelessly unchanging, realistic and optimistic sentiments of the Chinese people, who, concerned solely with reality and placing their trust entirely in man's present life, unashamedly regard man's happiness in the world as the ultimate value.

Russell Kirkland [1995] traces much in the Tao te ching to an older text Nei Yeh, which concentrates more on cultivation of qi, ching and shen, but does mention the tao and te ("inner power").

20150414: Nonclassical logics

Ian Rumfitt. The Boundary Stones of Thought: An Essay in the Philosophy of Logic

Giovanni Sambin. Steps toward a dynamic constructivism

Sequent calculi, also called Gentzen systems, [...generalize] the form of a natural deduction judgement (can be seen as a formulation of the predicate calculus):

Antecedent Yields [turnstile] Succedent
Ai, ...An |- Bi, ...Bk

The LK (klassische Prädikatenlogik) system [1934] consists of a number of inference rules eg Cut G |- D,A; A,S|-P implies G,S |- D,P. The LJ system is the intuitionist version, where only one formula per suceedent is allowed. The cut-elimation theorem (of Gentzen) shows that the Cut rule is derivable from other rules, particularly useful for automatic theorem provers.

The standard semantics of a sequent is an assertion that whenever every A_i is true, at least one B_i will also be true. One way to express this is that a comma to the left of the turnstile should be thought of as an "and", and a comma to the right of the turnstile should be thought of as an (inclusive) "or". The sequents

Ai, ...An |- Bi, ...Bk


|- (Ai & ... & An) -> (Bi | .. |Bk)

are equivalent in the strong sense that a proof of either one may be extended to a proof of the other. Structural rules include weakening, contraction and permutation.

Weakening (W) allows the addition of arbitrary elements to a sequence. Intuitively, this is allowed in the antecedent because we can always restrict the scope of our proof (if all cars have wheels, then it's safe to say that all black cars have wheels); and in the succedent because we can always allow for alternative conclusions (if all cars have wheels, then it's safe to say that all cars have either wheels or wings).

Contraction (C) and Permutation (P) assure that neither the order (P) nor the multiplicity of occurrences (C) of elements of the sequences matters. Thus, one could instead of sequences also consider sets.

[O]nly two rules in LK need to be restricted to give LJ.

Isaac et al [2013]:

Leitgeb has shown that any dynamical system performing calculations over distributed representations may be interpreted as a symbolic system performing non-monotonic inference. This result appears to show that there is no substantive difference in the representational power of symbolic systems and that of neural networks.

20150413: PRIMOS

Steve Lionel writes (c.l.f):

Absolutely it was done. Large sections of PRIMOS were written in FORTRAN IV (plus some PR1ME extensions). In one of my first jobs (pre-DEC), I had a PR1ME engineer dictate Fortran source patches for the PRIMOS file system to me over the phone.

Brian Taylor adds:

CTSS, developed at Lawrence Livermore National Laboratory and run on early Cray supercomputers, is another operating system written in Fortran.

Michael Metcalf also notes the IBM FORTRAN Q compiler was in Fortran (Scarborough and Kolsky, 1980).




From Avigad and Morris. Character and object (2014) :
As Michael Detlefsen has pointed out to us, one common view is that
[a mathematical] "object" is what is remains invariant under all the
different representations of a thing; in other words, what is left over
when one has "squeezed out" all the features that are contingent on
particular representations.

20150322: Music

Empyrean Atlas (NY prog rock band)

Steve Reich
Electric Counterpoint (1989) original soloist Pat Metheny
Music for 18 Musicians (1978)

20150322: CMM SNPs from US

Kocarnik et al [2014] :

2131 invasive melanoma cases and 20353 melanoma-free controls from Multiethnic Cohort (MEC), Women's Health Initiative (WHI); and Epidemiological Architecture for Genes Linked to Environment (EAGLE), BioVU, the Vanderbilt biorepository linked to de-identified electronic medical records, Nurses' Health Study (NHS) and Health Professionals Follow-up Study (HPFS).

SNP GeneChromosome Risk allelen# Studies OR95% CIP Het P
rs258322 CDK10 16 A 22,08251.551.41 - 1.708.54E-190.62
rs4785763 AFG3L1P (near MC1R)16 A 21,99351.311.22 - 1.401.01E-140.73
rs16891982SLC45A2 (MATP) 5 G 15,94933.112.31 - 4.187.39E-140.43
rs1393350 TYR 11 A 22,00951.251.17 - 1.356.21E-100.80
rs4636294 MTAP (near CDKN2A) 9 A 22,05351.181.11 - 1.275.51E-070.18
rs7023329 MTAP (near CDKN2A) 9 A 22,11451.171.10 - 1.251.93E-060.36
rs910873 PIGU (near ASIP) 20 A 15,93731.311.15 - 1.482.46E-051.00
rs2284063 PLA2G6 22 G 22,08751.091.01 - 1.160.019 0.27

20150313: Exporting from orcid.org and manipulating BibTeX files

Former is not straightforward!

Bib-Tools is Perl for manipulating and making bibliographies in Bib::CrossRef.

my $refs = Bib::Tools->new();
$refs->add_pubmed('duffy d*[author]');
print $refs->print;

orcid-python is a similar Python wrapper, but not up to date:

Easiest is from Martin Fenner ( http://blog.martinfenner.org/), who has made:


This can be accessed as:

curl -LH "Accept: text/x-bibliography; style=apa" http://feed.labs.orcid-eu.org/0000-0002-1825-0097

curl http://feed.labs.orcid-eu.org/0000-0002-1825-0097?format=json
curl http://feed.labs.orcid-eu.org/0000-0002-1825-0097?format=bib

This worked nicely, but highlighted that imports from EuropePMC do not include a full (BibTeX) record. Instead, I appear as sole author. And DOIs do not seem to be always recorded.

A couple of R packages "bibtex" and "RefManageR". RefManageR is very useful eg

dld <- ReadBib("~/Txt/Bibliographies/Duffy.bib")
rgms <- read.csv("~/Desktop/RGMS_Publication_List.csv") 
dld$title <- gsub("  "," ", gsub("[\t\n]"," ", dld$title)
dld[!(dld$title %in% rgms$Article.Title)]

20150312: AIXI and a measure of intelligence (Algorithmic IQ)

Shane Legg (Deep Mind), did PhD with Marcus Hutter, and Joel Veness.

20150312: PPARGC1B and alcoholism

...13 genes previously identified as associated with alcoholism or related traits in one or more GWAS, including KCNA3, DICER1, ZNF415, Catalase and PPARGC1B.

20150306: Distances between populations

Reyes-Centenoa et al [2014]:

For each pair of SNPs separated by less than 0.25 cM, we quantified LD as r2LD. All of the observed r2LD values were then binned into 50 recombination distance classes, from 0.005 to 0.25 cM, with incremental upper boundaries of 0.005 cM. Pairs of SNPs separated by less than 0.005 cM were not considered because at such short distances gene conversion may mimic the effects of recombination (44). We also adjusted the r2LD value for the sample size using r2LD-(1/n). We estimated Ne for each population in each recombination distance class as Ne=(1/4c)[1/r2LD - 2], corresponding to the effective population size 1/2c generations ago, where c is the distance between loci, expressed in Morgans. Finally, the long-term Ne for each population was calculated as the harmonic mean of Ne over all recombination distance classes up to 0.25 cM. At this point, based on the independently estimated values of Ne (Table S2), we calculated the separation time between populations in generations as T=log(1-Fst)/log(1-(1/2Ne)).

Hayes BJ, Visscher PM, McPartlan HC, Goddard ME (2003) Novel multilocus measure of linkage disequilibrium to estimate past effective population size. Genome Res 13(4):635'643.

McVean GAT (2002) A genealogical interpretation of linkage disequilibrium. Genetics 162(2):987'991

Holsinger KE, Weir BS (2009) Genetics in geographically structured populations: defining, estimating and interpreting F(ST) Nat Rev Genet 10(9):639'650 Software on Barbujani laboratory website

20150305: Automated Theorem Provers

Review in

Avigad and Harrison Formally Verified Mathematics

and specifically on higher order logic:

Benzmuller and others recently implemented Goedel's ontological proof in a higher order logic theorem prover (modal logic can be represented as HOL aka simple type theory) to confirm his axioms were too strong, cf various modifications: "computational metaphysics" or "computer assisted theoretical philosophy".

20150212: Better than the 2015 Hugo list

Brian Scott thought:

Catherine Asaro,  Undercity
Ann Leckie,  Ancillary Sword
R.C. Lewis,  Stitching Snow
Joel Shepherd,  Operation Shield

Katherine Addison (Sarah Monette),  The Goblin Emperor 
James A. Hetley,  Ghost Point 
Tanya Huff,  The Future Falls 
Sharon Lee,  Carousel Sun 

20150205: Testing fixed effects in lme4

Karl Ove Hufthammer lists:

The multcomp package
estimable() in the gmodels package
the lsmeans package
the contrast package

Multcomp uses asymptotic tests («infinite» degrees of freedom, z-tests)
for everything except normal linear (non mixed-effects models), where
it uses ordinary t-tests. (It even uses z-tests for gls models, though I
think this shouldn't be necessary.) It has excellent support for adjusting
for multiple comparisons, in a variety of ways.

estimable() uses the *minimum* of the degrees of freedom of the parameters
of the chosen contract when testing and constructing confidence intervals.

lsmeans uses asymptotic tests for lme(r) models, but calculates the
correct (?) degrees of freedom for gls and lm objects. (In a simple
gls example I tested, it seemed to sum the degrees of freedom.) It also
supports the multcomp package.

contrast calculates degrees of freedom for both lme and gls
models. However, the calculated degrees of freedom seems to differ between
what I believe should basically be equivalent lme and gls models (e.g.,
a two time-point longitudinal gls model with compound symmetry and a
mixed effects model with a random intercept for subject).

20150204: Browser hijack on a windows laptop

Since I spend most of my time on linux, this was a novelty to me.

When I recently updated filezilla from Sourceforge, it looks like it installed Omiga-Plus, an obnoxious ad-ware ("Possibly Unwanted Program") program that sets the IE, Firefox and Chrome start pages and search menus to their search page.

Eventually installed AdwCleaner, developer (Xplode) is a member (?) at http://general-changelog-team.fr/en/. Download available at a number of places, including https://www.bleepingcomputer.com .

I then ran the AVAST related rootkit detection and cleaning program aswMBR, which downloaded a 150 MB (!) virus definition database. This is suggested as a backup to clean up residual files.

Other freeware packages recommended for this type of problem are Malwarebytes Anti-Malware (MBAM) and Farbar Recovery Scan Tool (FRST).

20150204: %in% for Fortran

Harald Anlauf (anlauf.2008@arcor.de) offers:

module in
  implicit none
  public :: operator (.in.)
  interface operator (.in.)
     module procedure in_set_int
  end interface
  function in_set_int (val, set)
    logical             :: in_set_int
    integer, intent(in) :: val
    integer, intent(in) :: set(:)
    if (size (set) > 0) then
       in_set_int = any (set == val)
       in_set_int = .false.
    end if
  end function in_set_int
end module in

  use in
  print *, 2 .in. [ integer:: ]
  print *, 3 .in. [ 1, 2, 3, 4 ]

20150202: Predicting SNP effects on splicing (Hui et al Science)

Xiong et al (2015). The human splicing code reveals new insights into the genetic determinants of disease. Science DOI: 10.1126/science.1254806. Published Online December 18 2014.

http://tools.genes.toronto.edu/ The tool reports the maximum mutation-induced change in PSI (percent of transcripts with the exon spliced in) across 16 tissues, a regulatory score obtained by aggregating the change in PSI across the tissues, and how these values compare to those for common SNPs, in the form of percentiles.

Testing it using that ASIP SNP (runtime 4min 32s) :

JobVariant ID Transcript dPSI dPSI_percentile log(reg_score) reg_score_percentile PSI_WT
1 rs1205312 chr20:NFS1:-:NM_001198989:Exon10 0.42 83.00 -2.73 35.00 80.84
chr20:NFS1:-:NM_021100:Exon11 0.42 83.00 -2.73 35.00 80.84

20150123: Speech synthesis and 2001

The musician Max Neuhaus mentions:

In the speech research department of Bell Telephone Laboratories, Max Mathews had gotten their computer to sing Bicycle Built for Two in a nasal 'voice'.

20150118: Strong Church-Turing Thesis

The "strong" or "physical" Church-Turing Thesis, which I think Deutsch first stated (1985):

'every finitely realizible physical system can be perfectly simulated by a universal model computing machine operating by finite means'. Classical physics and the universal Turing machine, because the former is continuous and the latter discrete, do not obey the principle, at least in the strong form above. A class of model computing machines that is the quantum generalization of the class of Turing machines is described, and it is shown that quantum theory and the 'universal quantum computer' are compatible with the principle.

Zenil (2013) A Computable Universe: Understanding and Exploring Nature as Computation.

Yao et al [2003] gives the example of the Newtonian 4-body problem, which has non-collision singularities ie breaks the Extended Church-Turing Thesis (that hypercomputing is not possible), but requires infinite precision and an infinite system!

Arrighi and Dowek [2011] talk about Gandy's model of physical computation and how it fits in with the (physical) Church-Turing Thesis: computation requires and is bounded by:

  1. homogeneity of space
  2. homogeneity of time
  3. finite velocity of propagation
  4. quiescence (canonical state) exists for any region of space

and these can be extended into a quantum version.

Added 20161216:

Isaac et al [2013] on the Cobham-Edwards thesis:

The following thesis was formulated independently by Cobham (1965) and Edmonds (1965):

The class of practically computable problems is identical to the PTIME class, that is, the class of problems which can be computed by a deterministic Turing machine in a number of steps bounded by a polynomial function of the length of a query

...There is wide agreement that a problem has not been "well-solved" until a polynomial time algorithm is known for it. Hence, we shall refer to a problem as intractable, if it is so hard that no polynomial time algorithm can possibly solve it...

The Invariance Thesis (see e.g. Van Emde Boas, 1990) states that: Given a "reasonable encoding" of the input and two "reasonable machines," the complexity of the computation performed by these machines on that input will differ by at most a polynomial amount. Here, "reasonable machine" means any computing device that may be realistically implemented in the physical world. The situation here is very similar to that of the Church-Turing Thesis: although we cannot prove the Invariance Thesis, the fact that it holds for all known physically implementable computational devices provides powerful support for it.

20150112: Removing <New Page> after table in LibreOffice

Problems I commonly experience are inserting text after a table that has had a new page inserted following it, or rejoining such a new page to the table above it. Deleting or backspacing skips across to the new page.

Ctrl-Shift-Del is (usually) set to give delete-to-eol, and being used from the end of the table (last cell on R), clears the obnoxious new-page.

And (added 2017-04-22), to rotate one page, insert a manual (page) break, using the Page Break Style option to choose landscape, then insert a following break with the default. One can also set up a custom landscape Style, and switch into that style in the same way.

20150111: On a reading list, from comment at http://robertpaulwolff.blogspot.com.au/

Daniel Muñoz said...

If you're looking for a reading list in analytic philosophy, I'd suggest Sinan Dogramaci's (though it's mostly papers, not books):

Some additions you'll want to make if you're doing analytic ethics (like me!)

  1. The Right and the Good (ch. 1-2) - W.D Ross
  2. Does Moral Philosophy Rest on a Mistake? - H.A. Prichard (bonus: his letter to Ross, published in Moral Writings, which is one of the first stabs at the modern theory of reasons!)
  3. Principia Ethica (ch. 1) - G.E. Moore
  4. The Methods of Ethics - Sidgwick
  5. Groundwork... - Kant (as RPW says)
  6. The Sources of Normativity - Korsgaard
  7. Reasons and Persons (parts 3 & 4) - Derek Parfit
  8. Intention - Anscombe (for phil. action & ethics)
  9. The Notional Category of Modality - Kratzer (for phil. language & ethics)
  10. The Possibility of Altruism - Thomas Nagel
  11. An Enquiry Concerning the Principles of Morals, A Treatise of Human Nature (2.3.3) - David Hume

And a bonus: Practical Reality - Jonathan Dancy

You can tell I'm biased toward early 20th-century British moral philosophy.

20141218: Leopard spotting in the horse

Philos Trans R Soc Lond B Biol Sci. 2015 Jan 19;370(1660). pii:
20130386. doi: 10.1098/rstb.2013.0386.  Twenty-five thousand years of
fluctuating selection on leopard complex spotting and congenital night
blindness in horses.  Ludwig A1, Reissmann M2, Benecke N3, Bellone R4,
Sandoval-Castellanos E5, Cieslak M6, Fortes GG7, Morales-Muñiz A8,
Hofreiter M9, Pruvost M10.  Author information Abstract

Leopard complex spotting is inherited by the incompletely dominant locus,
LP, which also causes congenital stationary night blindness in homozygous
horses. We investigated an associated single nucleotide polymorphism in
the TRPM1 gene in 96 archaeological bones from 31 localities from Late
Pleistocene (approx. 17 000 YBP) to medieval times.  

Transient receptor potential melastatin-1 (TRPM1) is essential for dim light vision (gene is HSA 15q13.3). In the human melanocyte, metabotropic glutamate receptor 6 signaling enhances TRPM1 calcium channel function and increases melanin content [Devi et al Pigment Cell Melanoma Res. 2013 26:348-56]. Gene variants may be associated with Scz.

20141217: Woodward's interventionist theory of causation

"no causal difference without a difference in manipulability relations, and no difference in manipulability relations without a causal difference"

X is a cause if X can act as an intermediate variable in a causal chain.

A necessary and sufficient condition for X to be a (type-level) direct cause of Y with respect to a variable set V is that there be a possible intervention on X that will change Y or the probability distribution of Y when one holds fixed at some value all other variables Zi in V. A necessary and sufficient condition for X to be a (type-level) contributing cause of Y with respect to variable set V is that (i) there be a directed path from X to Y such that each link in this path is a direct causal relationship ... ; and that (ii) there be some intervention on X that will change Y when all other variables in V that are not on this path are fixed at some value. (Woodward 2003, 59)

An intervention variable I for X w.r.t. Y is a cause of X proximal to all causes of X [?], X is the intervening variable in the I-Y relationship, and I is uncorrelated with all other causes of X.

So, causation is relative to a particular variable set, analysis is counterfactual. Baumgartner (2010) suggests there are necessary conditions:

There must be a possible intervention
Which is independent of other causes
The other relevant variables must be able to be fixed [?]

Has been applied to the "exclusion argument", which states that higher order "macroscopic" entities that are supervenient on microscopic mechanisms are not causative of each other (ie are epiphenomena).

Applicable to reasoning about causation of nonintervention, cf physics based models involving transfers of momentum, energy etc. Also applicable to pseudocausation, where correlation induced by common cause.

Does the EA applies equally well to a lesser form of top-down causation, that of implementation of a computer program? WJ Rapaport (1998) specifically discusses the relationship between supervenience and implementation (which seems to have a sizeable philosophical literature).

I posted:

The IEP article [on panpsychism] seemed a nice summary, listing the Emergence, Not Testable, and Combination objections. The various QM based theories might be testable. I believe the "Dancing with Pixies" argument against computationalism is seen as pro-panpsychism, but it seems to me must have a formal physics based refutation along the lines of those against Maxwell's Demon ie such panpsychic consciousnesses will be too expensive thermodynamically to select out to interact with other minds or the environment. The other line of argument there is stability/repeatability.

In a related vein, Rapaport [1998] discusses the question of *implementation* of an algorithm or program as a physical process: he decides that implementation is not reduction (down to physics), instantiation, or supervenience (I found his discussion of these three terms very useful).

Another approach linking computation and emergence in physics is James Crutchfield's (Santa Fe Institute) work on computational mechanics and statistical complexity. He claims that the the description of the intrinsic computational capacity of a given physical system, for example undergoing a phase change, requires moving up one level in a hierarchy of representation.

Zenil (2013) A Computable Universe: Understanding and Exploring Nature as Computation.

20141212: Vitamin D and mortality

Afzal et al (2014) carried out a mendelian randomization analysis of DHCR7 (rs7944926 and rs11234027) and CYP2R1 (rs10741657 and rs12794714) genotype, serum 25OHD, and mortality within the Copenhagen General Population Study, Copenhagen Ischemic Heart Disease Study, and Copenhagen City Heart Study (N=95766).

Pooled, the median VitD level was 49 nmol/l (IQR 33-67). A 20 nM lower VitD increased all cause mortality 1.19 (1.14-1.25), 1.12 (1.03-1.22) for cancer mortality. The distance between the most extreme genotypes was 8.4 nM.

The hazard ratio per one DHCR7/CYP2R1 allele score increase was 1.02 (1.00 to 1.03; P=0.03) for all cause mortality, 0.98 (0.96 to 1.01) for cardiovascular mortality, 1.03 (1.00 to 1.06; P=0.04) for cancer mortality, and 1.03 (1.00 to 1.06; P=0.07) for other mortality

20141117: Lindblad-Toh et al on canine hematological cancers

Broad angiosarcoma patent (2013)

Cancer-associated germ-line and somatic markers and uses thereof WO 2013181367 A1

20141117: Papers from Social Cognitive & Affective Neuroscience (and PRSB)

Way and Lieberman (2010) discuss population level associations between 5-HTTLPR, MAOA-uVNTR and OPRM1 A118G genotypes and the relative degree of individualism/collectivism.

Panksepp (1998) found that low, nonsedative doses of the physical pain-killer morphine quelled the distress of separation from the caregiver in infants of multiple mammalian species (Herman and Panksepp, 1978; Panksepp et al., 1978a, 1978b). This data indicates that the endogenous opioid system is involved in signaling the distress of separation from conspecifics. According to recent neuroimaging data in humans, this role for the opioid system appears to apply to the distress associated with the severance of a social bond in adulthood as well. In a positron emission tomography study, women exhibited decreased µ-opioid mediated neurotransmission when recalling the death of a loved one or the breakup of a romantic relationship (Zubieta et al., 2003). Hence, it appears that the level of µ-opioid receptor dependent signaling within the brain is a reflection of an individual's current inclusion status. In line with this hypothesis, a functional polymorphism (A118G) in the µ-opioid receptor gene was recently associated with self-reported dispositional sensitivity to rejection (Way et al., 2009), as measured by Mehrabian's (1976) Sensitivity to Rejection Scale.

Eisenberg and Hayes [2010] suggest the SERT-collectivism results are too confounded by culture, reanalysing the data of Chiao and Blizinsky [2010]:

Nation5-HTTLPR %S from Chiao and BlizinksyIndividualism (SCSI_rev) from Chiao and Blizinksy PC1 from Novembre et alPC2 from Novembre et al EuropeanAsianAsian (w/o Indian & Russia)Europe/Asia
Argentina 51.0454000
Australia 45.9110000
Austria 43.65450.01830.01941001
Brazil 46.9662000
Denmark 40.80260.03290.007961001
Estonia 34.81401001
Finland 42.45370.0560.02411001
France 43.18290.00219-0.01381001
Germany 43.03330.02470.008521001
Hungary 41.71200.020.03651001
India 58.85520100
Israel 49.2646000
Italy 48.5424-0.0340.02341001
Japan 80.25540110
Korea 79.45820110
Mexico 51.9670000
New Zealand43.0321000
Poland 36.96400.04420.0421001
PR China 75.20800110
Russia 43.91610.04210.04320100
S. Africa 27.7935000
Singapore 71.24800110
Slovenia 42.520.02260.04491001
Spain 46.7549-0.0243-0.03211001
Sweden 43.63290.04490.008251001
Taiwan 70.57830110
Turkey 54.2963-0.03260.08071001
UK 43.98110.0266-0.01141001
USA 44.539000

Troisi et al [2012]

The A118G polymorphism of the mu-opioid receptor gene (OPRM1) may moderate the impact of early maternal care on fearful attachment. In a sample of 112 Italian psychiatric patients (major depressive disorder, 31%; eating disorders, 25%; anxiety disorders, 21%; bipolar disorder, 14%; cluster B personality disorders, 9%), those expressing the minor 118-G allele had similar and relatively high scores on fearful attachment regardless of the quality of maternal care. By contrast, early experience made a major difference for participants carrying the A/A genotype.

20141113: Natalie Zemon Davis

From the Wikipedia article:

Her book, Fiction in the Archives: Pardon Tales and their Tellers in 16th-century France (1987), [is] a study of the stories people of all classes told to the king to get pardoned for homicide in the days before manslaughter was a possible plea. In her Women on the Margins (1995), she looked at the autobiographical accounts of three 17th-century women ' the Jewish merchant Glikl Hamel, the Catholic nun Marie de l'Incarnation, who came to New France, and the Protestant entomologist-artist Maria Sibylla Merian'and discussed the role of religion in their lives.

In Trickster Travels (2006), she describes how the early 16th-century North African Muslim "Leo Africanus" (Hasan al-Wazzan) managed to live as a Christian in Italy after he was kidnapped by Christian pirates

20141112: Dog OCD

Tang et al (2014)

We find 119 variants in evolutionarily conserved sites that are specific to dogs with OCD. These case-only variants are significantly more common in high OCD risk breeds compared to breeds with no known psychiatric problems. Four genes, all with synaptic function, have the most case-only variation: neuronal cadherin (CDH2), catenin alpha2 (CTNNA2), ataxin-1 (ATXN1), and plasma glutamate carboxypeptidase (PGCP). In the 2 Mb gene desert between the cadherin genes CDH2 and DSC3, we find two different variants found only in dogs with OCD that disrupt the same highly conserved regulatory element. These variants cause significant changes in gene expression in a human neuroblastoma cell line, likely due to disrupted transcription factor binding.

Cao et al (2014)

20140827: SLC14A2

Expression in multiple tissues

Expression in melanocyes and melanomas

Induced pluripotent and embryonic stem cells, including melanocytes

20140818: utility of an integer square root routine

Ron Shephard commented on c.l.f that:

When people mention allowing integer arguments to sqrt, this is always 
what I think of first. The reason is that integer square roots are kind 
of a tricky operation. The value you want for sqrt(M) is the largest 
integer K such that K*K <= M. If you convert to real, then take the square 
root, and then convert back to integer, you have to watch out for 
rounding and arithmetic errors in the least significant bits that end up 
giving you a result that is off by one.

An example of where this operation is necessary is the computation of 
the column and row indices of an array stored lower triangle packed by 
rows.  The linear index is given by the expression

    IJ = (I*(I-1))/2 + J     ; for I >= J > 0

So given the IJ value, solving for I and J involves a quadratic 
equations solution which in turn requires a sqrt. The off-by-one error 
in the computation of I means that J does not satisfy its bounds, so 
with the current situation, the programmer needs to check and adjust the 
values if necessary. Another example of integer square roots involve the 
computation of the smallest radius that includes a specific number of 
discrete grid points.

20140817: "Readonly content cannot be changed"

In libreoffice, as per here, files from MS Word may have protected text (or boxes etc). These are only editable after:


and uncheck "Cursor in Protected areas".

20140806-20160628: more blog posts

"genes as a program": Down in the RNA world, we already have "genes" that can replicate themselves, or more accurately cooperate by cross-replicating each other:



"When such cooperative networks are competed directly against selfish autocatalytic cycles, the former grow faster, indicating an intrinsic ability of RNA populations to evolve greater complexity through cooperation. We can observe the evolvability of networks through in vitro selection. Our experiments highlight the advantages of cooperative behaviour even at the molecular stages of nascent life."

The basic physics of maintaining a stable dynamic structure (retaining identity over time either by self-regulation or copying) is the thermodynamics of information - folks will have seen papers by Jeremy England and others discussing this point eg



That is to say, even though the same molecule can be both gene and "phene", at the basic level these are two different processes thermodynamically.

"[there is] no prior requirement for the system to be partitioned into separate plant and controller components". The latter is referring to regulation and adaptation, but it is equally true of replication.

On Davidson on conceptual relativism.

What are we to make of this? It has a few amusing shots eg "The trouble is, as so often in philosophy, it is hard to improve intelligibility while retaining the excitement" or "Since there is at most one world, these pluralities are metaphorical or merely imagined", or "How would you organize the Pacific Ocean? Straighten out its shores, perhaps, or relocate its islands,or destroy its fish", or "As philosophers we are peculiarly tolerant of systematic malapropism, and practised at interpreting the result". I do like the ambition of giving solid meaning to the idea.

But "unmediated touch?" I'm afraid the "direct perception" movement in psychology is generally recognised as having failed. And conceptual schema are either true or false? Surely they are either more or less successful in predicting what the world throws up ("all models are wrong, but some are useful"). And "My skin is warm" has so much baggage (and so many facts) attached to it, that are essential to me to understand it. In his yawl/ketch example, in practice we would have a dialogue where we test our own understanding of the other's concepts.

As for translation: "He who translates is a heretic but he who refuses to translate is a blasphemer". ;)

Having sat on NHMRC grant review panels on several occasions, I am sure there is no political interference of that sort, for project grants at least (the main way these things are funded). Project grants are assessed and ranked using averaged scores from multiple scientists from the Australia-wide medical research community using carefully standardised criteria, not by bureaucrats. The problem is simply that there is one pot of money, and only 14% of project grants were funded last year. So many really worthwhile public health projects miss out.

What "politics" there are, are in the ranking of project significance, where the panel members have to decide whether work on diabetes or cancer or infectious diseases in the third world, say, is more or less important than climate change.

I see a tension between the emotional associates of an aesthetic experience, and the intellectual understanding of the causative elements in the perception involved, as a critic or an artist (where there are the technical questions). In the case of human facial beauty, most people have aesthetic concepts that revolve around high level features our brains extract viz X's cheekbones are too wide to be truly attractive, Y's face is too symmetrical, Z's hair colour exacerbates their unfortunate complexion. Many of Sibley's terms (handsome, comely, graceful, dainty) are used in judgements of physical attractiveness.

This is also true of natural scenes - one can think of those aesthetic concepts applicable to landscape painting and photography that also apply to actual landscape:

"The Minto Crags ring dyke consists of fine grained rhyolites which have intruded the Walloon Coal Measures. It is of particular aesthetic appeal."

If asked why I agree with that geologist that the Minto Crags are attractive, I would mention rock colour and how it contrasts the surrounding vegetation (leading to visual surprise), size relative to rest of landscape and to the small trees growing on it ("human sized"), and would compare the effect on me to similarly appealing bonsai/penjing rock-and-vegetation arrangements. In the latter case, artists have attempted to develop design principles that maximize aesthetic pleasure. I don't think it is reductionist to say that those principles are mapping high level features of the human perceptual system. I further don't think the concepts I would use in this example are in a different category from Sibley's examples.

As to more abstract concepts such as "unity", they may be rather empirical - paying customers don't like versions of the Marriage of Figaro that lack certain arias, because the cumulative pleasure of the finale is reduced, even though it is not technically possible to work out how the effect occurs. Another abstract principle is surprise, where a novelty that is just congruent enough with preceding material to cause pleasure (I have previously cited the example of repeatable "chills" listening to particular musical passages, but this is just as true of children hearing Little Red Riding Hood for the 100th time). Speaking reductionistically again, I think this is a very general but abstract feature of how we process information, and is not unique to humans. So it cannot be summarized as any simple rule, but is still comprehensible to naturalists thinking in terms of higher and higher levels in a perceptual neural network.

A fourth way way might involve the dialogue of ideas between multiple knowers in a Darwinian way (no, I won't use the m word). There can be no sensible dialogue with, for example, solipsists and Ideally Coherent Caligulas, in the sense of rational argumentation (though there are various appeals to emotion/empathy that might change those individual's intuitions). So although these might arise every generation de novo, they won't usually spread widely, as most of us aren't particularly interested in talking with such people. In the population of knowers, certain other ideas propagate and outlive individuals. And certain sets of ideas can gang up on others, eg utilitarian and deontologist ideas can agree on a large number of practical cases and coexist in law and politics. Warrant is then the pragmatic measure that idea X is still around, and norm takes on a statistical sense.

I have been surprised that there is so little written on the philosophy in Anathem - the more "technical" reviews I can find have been by mathematicians or physicists. The idea of a MWI quantum physicalist platonism struck me as hilariously good, and I did wonder if he was riffing off someone modern other than Penrose and Deutsch and Tegmark etc eg the argument about what we could share with a bat is everything in the Platonic Realm. Another feature is that the standard SFnal transcendence here simultaneously old-school platonic and vanVogtian. Finally, I can see why you might characterize it as Science Fantasy, but so much of the pleasure is how well he fits in (speculative) physics and cosmology - for example, how time-travel by rocket is allowed in the Godelian rotating universe https://en.wikipedia.org/wiki/G%C3%B6del_metric

From my go-to man for levels, Bunge [2010] defines ontological and epistemic levels:

To uniquely characterize the notion of a level of reality we also need that of relation between two adjoining levels. We stipulate that level Ln-1 precedes level Ln if every object in Ln is composed of objects in Ln-1. Thus, the part-whole relation defines the level precedence relation. [...] The concept of a level allows us to refine that of emergence, namely thus. Instead of saying that property P is an emergent, we should say that P emerges on level Ln, which suggests that P was not possessed by any entities on the preceding level Ln-1. For example, presumably the property of being valuable emerged along with the life level; and that of acting purposefully emerged along with the level composed of animals endowed with a prefrontal cortex - roughly mammals and birds. [...]

The levels examined [above] are ontic, not epistemic: their constituents are material things, not ideas. The epistemic counterpart of the ontic ladder is shown... Whereas the ontic levels are ordered by the ontic relation of precedence, the epistemic levels are ordered by the relation of reducibility, as when physicists explain magnets in terms of the spin and the associated magnetism of their constituent atoms, and social scientists claim that all social facts emerge from individual actions.

Classical and quantum phenomena have a (minor) levels relationship in this schema. So (from Google Scholar) "physics in a wide range of regimes both at the classical and at the quantum level", "relevant differences are brought out already at the classical level", "thorough investigation of 1 + 1 gravity on the classical and quantum level", "spatial correlation at the quantum level", "renormalisable at the quantum level" are used. I guess decoherence might count as emergence of novel properties of level n entities by interaction of Ln-1 entities.

Bunge's schema kind of hangs together, but I think has problems with how it interacts with logical and mathematical (eg geometric) relations between objects, which are universal across different levels. Also, his dependence of ontology on scientific knowledge implies that we might remove distinctions between levels as understanding changes. He also gives many examples of downward causation eg purposeful behaviour leads to maintenance of hydration of an organism. In his ontology, the value of hydration to the organism, and caring about hydration, exist at a different level of organisation from the brute fact of osmotic pressure across a membrane.

So would this really alter scientific models? A lot of science is limited to one level at a time, so levels talks won't appear in Cell, but might well in Behavioral and Brain Sciences.

Hi Philip. Busemayor and colleagues carefully avoid any idea that cognition is non-classical in its physics eg
One perspective is that the brain does not instantiate any quantum computation at all. Rather, interference effects in the brain can occur if neuronal membrane potentials have wave-like properties...
merely that non-classical probability may more nicely model how different beliefs interfere with one another in the irrational process of human decision making eg the conjunction fallacy (Linda the feminist bank teller), overextension of category conjunctions ("chess is a sport that is a game").

You may also like the work of Arto Annila




Emergence can be understood as a natural process when entities of nature are described as actions that all are composed of some integral number of quanta. Then, the central connection between the symmetry of action and the qualities of a system, given in terms of conserved quantities and motional modes, can be analyzed mathematically to conclude that novel characteristics will emerge due to quanta that are either acquired or lost at a dissipative step of evolution. Conversely, no new qualities can appear in an isolated system or in a system that has attained a thermodynamic stationary state in its respective surroundings. The evolving system demands a holistic description whereas the stationary system suffices with a reductionist account

Smilansky ("Utilitarianism and 'punishment' of the innocent: the general problem". Analysis 1991, 256-61) brings up the settings for false positive and false negative rates in the judicial system. He looks at the case for increasing the sensitivity of the test (increasing FPR and decreasing the FNR) to improve the law and order situation. A corollary is that transparency into the procedures that would cause indignation and alteration in preferences should be avoided as it would lead to a net increase in aggregate suffering, so returning to the old conspiracy of consequentalist do-gooders against the deontologists ;)

There are a few issues here. Firstly, a majority of studies find that individuals carrying the low-MAOA-activity genotypes are (roughly) a modest 1.1-fold more likely to be aggressive. This is a "marginal effect", in that it averages over the levels of adversity in the particular sample studied. Most of these studies are limited to ethnically homogenous samples, and the meta-analysis of Ficks and Waldman (2014) suggests that ethnic makeup of the sample is not a significant modifier of effect size. Skeptics of this finding have to find a mechanism by which the investigator's rating of the aggressiveness of an individual can be biased by that individual's genotype. The only plausible cause of confounding of this type is that genotype is associated with membership of a particular social group.

One recent study Kuepper et al (2013) is quite nice. It is carried out in 239 university students of "Caucasian ethnicity", who performed a competitive reaction-time test that is designed to be irritating, and measures aggression by the individual to an unknown (actually computer) opponent (volume of noise to be played in the loser's ears). There are 56 low-activity individuals, 69 mixed (female, MAOA is on the X chromosome), and 114 high-activity individuals. The low group (male or female) responded 30% more aggressively. So numbers in each genotype group in both observational and experimental studies are quite adequate (a total of 13600 individuals in the Ficks and Waldman meta-analysis).

The second point is that the differences in genotype frequencies between different ethnic groups are also robust. Therefore, in the absence of interactions (different genetic backgrounds on which this particular polymorphism acts, and gene by environment interactions), we would predict a 1.02-fold difference in rates of aggression between a population with 60% as opposed to 40% frequency of the low-allele (3-repeat). Obviously, this is not a particularly easy to detect, or particularly important cause of inter-group differences ie

Pop1: P(3R)=0.4, male risk of aggression 3R=11%, 4R=10%: Pop1 rate=10.4%
Pop2: P(3R)=0.6, Pop2 rate=10.6%)

This may be amplified slightly by the adversity-genotype interaction if there are significant between population differences in adversity. In terms of population attributable risk (what difference would we see if a wonder drug changed all "3R"'s to "4R"): population rates only drop from 10.4% to 10%, or 10.6% to 10%.

So does this alter my understanding of human nature - that some people have shorter fuses than others, and may have to work harder to contain this tendency...

"proper Negative Capability" - Lem, Harrison, and maybe Wolfe, though I usually feel I just haven't tried hard enough. And Haldeman too. Maybe more a fantasy/horror thing - it doesn't really fit in the SFnal tendency to scientism (a very popular word on the intertubes at the moment).

I rather feared "marginal effect" would be confusing, in that it referred to averaging over one margin of a cross-classified table (a la "main effect"). The random-effects model P-value for the overall effect of MAOA was 1.37x10-6, difficult to explain as due to just to "moving somewhere else".

With respect to Genotype x Treatment interaction, the meta-analysis of Byrd and Manuck (2014) found that the low-activity MAOA genotype heightened risk for antisocial behavior among males exposed to maltreatment specifically (p = .00000082) combining across 13 different studies, a result they found robust to several sensitivity analyses. There was no such effect seen in 11 female cohorts.

So my original point stands, that the associations have been found within particular ethnic groups on multiple occasions, and it only when this is combined with the equally strongly supported fact that genotype frequencies differ between ethnic groups that we arrive at the hypothesis that you find so odious. This has nothing to do with any particular models of evolution, they are straightforward observations.

However, there are good examples of interactions between ethnic background and what might seem to be straightforward associations. My favourite example is Apolipoprotein E (APOE) and Alzheimer's Disease (AD). The APOE*4 allele is very robustly associated with increased risk of AD in European and Asian populations. Since the APOE*4 allele is the ancestral allele, it is much more common in African populations. However in African and to a lesser extent African-American populations (bearing in mind there is significant European admixture in the AA population), the APOE*4 allele does not predict AD. I suspect that there are environmental differences or differences in genetic background on which APOE variation is acting that explain this observation. One could further speculate that the more recently derived APOE*3 and APOE*2 alleles have been selected for after the loss of either these protective genetic or environmental factors by the move out of Africa. As I noted previously, Ficks and Waldeman tested to see if study participant ethnicity was an explanation for the considerable between-study heterogeneity they observed, and found it not statistically significant.


The variance components approach has been extended to include a variety of interactions and correlations. An obvious point is that one's parents share genes with you, and if these genes alter parental behaviour and treatment of you, then we get gene-environment covariation. The reciprocal interactions between siblings are also commonly modelled - the classic example of this is where identical twins become less similar by "rubbing" against each other - one becomes more extroverted and the other more introverted as a result.

I don't think there is data or statistical power to test complex population level models a la Maynard Smith (frequency-dependent selection and interaction etc). MAOA is supposed to exhibit some signals of selection in humans but not other species in a couple of older studies, but doesn't seem to pop out in the more recent genome-wide selection scans. The actual VNTR is only in humans and macaques.

I am reminded firstly of Sokal and his distrust of these kinds of analogies, and secondly of computable and uncomputable numbers. I am not sure which type of interpretations corresponds to computable: uninformed or informed...

"or a mathematical structure at all (think of the theory of natural selection, or continental drift in geology)" I don't think either of these is a good example. "Origin of Species" doesn't have any equations, but it did present a model and an algorithm, which have since been mathematized in evolutionary genetics - the recent arguments about group selection seem to swing between mathematical models and observational data. In the case of continental drift, we have straight geometry (fit of coastlines), species distributions, and arguments about the physics. I have some sympathy with Mario Bunge's pragmatic way of thinking - his ontological "levels ladder" (physical, chemical, living, thinking, social, artificial) looks sensible, and highlights emergence (P being a property of an object not possessed by any entities on the preceding level, although every object in the current level is composed of objects from the previous level). One does wonder whether these levels will proliferate indefinitely. And it doesn't make much of "universal" mathematical models can apply to phenomena

It seems to me that the EA applies equally well to a lesser form of top-down causation, that of implementation of a computer program. One paper I found, by WJ Rapaport (1998) http://philpapers.org/archive/RAPIIS.1.pdf specifically discusses the relationship between supervenience and implementation (which seems to have a sizeable philosophical literature). To me, it seems a fact that the same program can be implemented (as a physical system) to cause the same output on a variety of different substrates, or even on the same substrate using a different encoding. Then, low level descriptions of causation that exclude (!) discussion of the high level similarities between implementations are not very interesting.


Genome Workbench is the NCBI sequence analysis program. The tutorial is extensive: search the public databases, perform BLAST searches, download SNP tables.

20140804: Hair and Eye Colour Classifications

Martin R (1914). Lehrbuch der Anthropologia.
Martin R (1928). Lehrbuch der Anthropologia. 2nd ed.
Saller K, Martin R (1956). Lehrbuch der Anthropologia. 3rd ed.
Schultz BK (1930). Eine verbesserte Augenfarbentafel Anthropologischer Anzeiger Jahrg. 6, H. 4 (1930), pp. 331-335

Miller, Thomas R (2004). Seeing Eyes, Reading Bodies: Visuality, Race and Color Perception or a Threshold in the History of Human Sciences. Amsterdamer Beiträge zur neueren Germanistik, Colors 1800/1900/2000: Signs of Ethnic Difference. Edited by Birgit Tautz, pp. 123-141(19)

... Based on a color ranking developed by Rudolf Martin at the University of Zurich, the Augenfarbentafel or eye-color table was sold at cost to scientists and researchers. It was first marketed in 1903 with claims of unprecedented scientific accuracy in typologizing eye color.


1-2 blue iris (1a, 1b, 1c, 2a: light blue; 2b : darker blue iris)
3 blue-gray iris
4 gray iris
5 blue-gray iris with yellow/brown spots
6 gray-green iris with yellow/brown spots
7 green iris
8 green iris with yellow/brown spots
9-11 light-brown iris
10 hazel iris
12-13medium brown iris
14-16dark-brown and black iris

Piquet-Thepot [1968] on eye colour classifications
Colour Martin Martin-Schultz Martin-SallerTopinard
Blue or Grey 13-16 1-2b 1-2 bleu
Mixed Light (bl-gy) 7-12 3-6 3-4 clair
Mixed Dark (br-gr/ye) 7-11 5-6 moyenne
Browns 1-6 12-16 7-8 fonce

In Corsica (N=475), blue ~12%
Piquet-Thepot [1968] on eye colour from different samples


Eye colour in Swedish families [ Beckman et al 1960] and Swedish gypsies [ Beckman and Takman 1965]
215 0
348 0
419 1
521 4
610 3
720 27
812 80

The visual assessment of the eye color was carried out using the traditional scale of Martin, based on 16 ocular prostheses. The iris color was determined according to the Martin/Schulz`s table of the iris-color. The four-category grading system (groups of the iris color) was used to determine the variations in some anthropometrical parameters: blue (group I), grey (group II), green (group III) and brown (group IV).

20140802: Eye Colour Review

Iris Anatomy: 0.3-0.4 mm thick, stroma ~0.25 mm thick Irises with more crypts are thinner; irises with more extensive furrows and darker color are thicker peripherally. Posterior Iris Pigmentary Epithelium cuboid double layer

Alexandria, Egypt: Brown or hazel, 120; Green or blue, 25.

Vonderach et al [2006] provides data on a variety of physical anthropological measures based on a large survey in Westphalia by Schwidetzky and Walter (1967) across children of multiple ethnic origins.

RegionN BlueBrown
Mecklenburg-Vorpommem 29 0.6900.0
N.Saxony-Anhalt 33 0.5150.091
Brandenburg 34 0.3310.021
S.Saxony-Anhalt 38 0.4740.063
Thuringia 36 0.3890.111
Saxony 82 0.5520.083
Lusitia 44 0.5910.023
W.Pomerania 88 0.5110.023
E.Pomerania 42 0.4760.024
W.East Prussia 155 0.5100.032
E.East Prussia 29 0.5520.034
NE.East Prussia 40 0.5250.050
Masuria 140 0.5000.050
E.West Prussia 102 0.5000.069
W.West Prussia 23 0.2170.043
Danzig 38 0.4740.079
East Brandenburg21 0.3610.098
N.Lower Silesia 256 0.5000.082
S.Lower Silesia 365 0.4220.093
Upper Silesia 318 0.4250.082
W.Sudetenland 90 0.5400.098
E.Sudetenland 33 0.4200.132
Baltic Germans 16 0.4380.063
Abroad II 160 0.4500.100
Abroad III 36 0.6110.028
Abroad IV 35 0.3140.171
Münsterland 41800.5440.041
Ruhr 47120.4890.062
Sauerland 45470.5690.051
East Westphalia 31530.5790.037
Einbeck 278 0.5410.068

Brown is defined as Martin classes 12-16, and blue 1a-2b.

Homogeneity chi-squared test for homogeneity of the first 15 studies is not significant X2=33.7, df=28, P=0.21.

Gronkiewicz and Gronkiewicz [1995] provide data on eye colour in 10500 Polish adults collected from 1960-1990. Colour was scored using Martin scale, along with details of the structure of the iris of eyes and distribution of the pigment on slit lamp examination: light-coloured (Jasne) was Martin 12-16,9,10 (blue or grey), and dark (Ciemne) was 1-6.

Study Population N Grey/Blue Brown Other
Vonderach et al [2006] East Prussia 224 0.518 0.036 0.446
Vonderach et al [2006] Ruhr 4712 0.489 0.062 0.449
Kazoka and Vetra [2011] Latvia 873 0.680 0.071 0.250
Sulem et al [2007] Iceland 5704 0.731 0.093 0.176
Beckman et al 1960 Sweden 300 0.567 0.107 0.327
Gronkiewicz and Gronkiewicz 1995 Poland 10504 0.399 0.132 0.469
Walsh et al [2012] Estonia 579 0.788 0.155 0.057
Vinding [1990] Denmark 906 0.506 0.159 0.334
Walsh et al [2012] Norway 547 0.762 0.170 0.068
Walsh et al [2012] Ireland 498 0.715 0.209 0.076
Forsius [2010] W. Finland - - 0.21 -
Schmidt-Pokrzywniak et al [2009] Germany 827 0.607 0.227 0.166
Maung [1941] Scotland 22361 - 0.231 -
Kastelic et al [2013] Croatia 105 0.447 0.296 0.257
Walsh et al [2012] N. Italy 542 0.292 0.592 0.116
Forsius [2010] Saami 592 - 0.59 -
Czoma et al [2008] Hungary 905 (0.388) 0.611 -
Giuffrè [2005] Sicily 141 - 0.688 -
Ulivi et al [2013] Georgia (WSR)496 0.093 0.696 0.211
Walsh et al [2012] S. Spain 511 0.160 0.775 0.065
Lasithiotakis et al [2004] Crete 110 - 0.800 -
de-Misa et al [2004] Canary Islands5116 0.168 0.832 -
Walsh et al [2012] Greece 547 0.125 0.846 0.029
Ulivi et al [2013] Kyrgyzstan (ESR)519 0.062 0.871 0.067
Beckman and Takman [1965] Swedish Gypsies 116 0.008 0.922 0.069
Hashemi et al [2005] Iran 3450 0.019 0.954 0.026
Chia et al [2012] Singapore 266 0.0 1.000 0.000

Several studies above used the 8-point Martin-Saller scale.

Wilde et al [2014]

Ancient DNA was retrieved from 63 out of 150 Eneolithic (ca. 6,500-5,000 y ago) and Bronze Age (ca. 5,000-4,000 y ago) samples from the Pontic-Caspian steppe, mainly from modern-day Ukraine.

...[W]e compared the 60 mtDNA HVR1 sequences obtained from our ancient sample to 246 homologous modern sequences from the same geographic region and found low genetic differentiation (FST = 0.00551; P = 0.0663).

Gene SNP Polymorphism Europe Asia Africa Modern Ukrainian sample Ancient sample
HERC2 rs12913832 A > G 0.7100.0020.0000.651 (0.546-0.744)0.160 (0.099-0.247)
SLC45A2 rs16891982 Leu374Phe C > G 0.9700.0070.0000.927 (0.849-0.965)0.432 (0.296-0.578)
TYR rs1042602 Ser192Tyr C > A 0.3680.0020.0000.367 (0.279-0.466)0.043 (0.018-0.106)

We assumed codominance for both SLC45A2 rs16891982 G and TYR rs1042602 A alleles (22, 35) and that the derived HERC2 rs12913832 G allele is recessive (36). Using these simulations, neutrality (S = 0) was rejected under all assumed ancestral effective population sizes - ranging from 103 to 105 at the time of the ancient sample (SLC45A2 P < 1e-5, TYR P < 2e-5, and HERC2 P < 1e-5). The values of selection acting on the SLC45A2 rs16891982 G allele, the TYR rs1042602 A allele, and the HERC2 rs12913832 G allele that best explained the observed derived allele frequency changes were 0.030, 0.026, and 0.036, respectively.

Selection coefficients for other loci:

Locus Selection coefficient Age Effect*
LCT 0.008-0.018, 0.026-0.080
G6PD 0.018-0.049, 0.014-0.049
SLC45A20.04-0.05 [Beleza 2013] 13700 5, 3.4**
SLC24A50.08-0.16 [Beleza 2013] 11400 7.6-11.4, 5.0**
KITLG 0.02 [Beleza 2013] 31100 2.8-3.8
TYRP1 0.02-0.03 [Beleza 2013] 14000
HERC2 1.1**

* average melanin units in admixed populations ** Marcheco-Teruel et al [2014] Zhang et al [2013] comment:

...[D]espite its core importance, it is almost impossible to measure small fitness differences among genotypes in natural populations (Orr, 2009). In this current study, we used estimated FST (global) and branch length (total) as surrogates to reflect the strength of directional selection. It can be shown that, when a directional selection with no dominance is assumed, the estimated FST and branch length should have an approximately linear relationship with the expected allele frequency change delta q = sq(1 - q)/2, where q is the allele frequency before selection, s is the selection coefficient, and the fitness for each genotype AA, Aa and aa are 1, 1 - s/2, and 1 - s, respectively.

In the Cuban population, Marcheco-Teruel et al [2014] suggest that rs35395 (SLC45A2) no effect on skin colour, after adjusting for rs16891982.

Beleza et al [2013] note:

Because of [the] limiting effect [of effective population size], it has been proposed that the rate of adaptive evolution in humans before 50,000 years was relatively low, starting to increase only with the onset of population growth in the Upper Paleolithic and its subsequent magnification during the Neolithic (Hawks et al 2007). In Eurasia, it is likely that demographic limits to adap tation were prolonged by bottlenecks affecting population migrating out of Africa (Schaffner et al. 2005; Voight et al 2005; Laval et al. 2010), and this could explain, in combination with weaker selection pressures related to UVR availability in the ancestral Eurasian population, the presence of very few skin-lightening alleles shared between Europeans and East Asians (McEvoy et al. 2006), of which the derived KITLG lineage is the best-known example.

...there is some evidence that lighter iris colors, because of their recessive mode of inheritance, may be preferred by males in assortative mating regimes to improve paternity confidence [Laeng et al 2006]. Consistent with positive assortative mating, an exact test of Hardy-Weinberg equilibrium reveals an excess of HERC2 rs12913832 homozygotes in both the modern (P = 0.0543) and ancient (P = 0.0084) East European samples genotyped here.

deGruijter et al 2011 give estimates for selection

Gene Population under positive selection Time of selection (95% CI S (95% CI) h (95% CI)
TYRP1 Europe 50,250 (13,500 to 97,250) 361.94 (55.6 to 984.79) 1.70 (0.71 to 4.28)
OCA2 Europe, East Asia 45,875 (12,750 to 95,006) 383.42 (36.07 to 992.04) 2.04 (0.78 to 4.56)
KITLG Europe, East Asia 46,000 (11,750 to 93,269) 385.49 (37.68 to 1036.32) 2.01 (0.74 to 4.41)
DCT East Asia 48,500 (11,250 to 97,006) 789.24 (55.21 to 1584.98) 1.90 (0.67 to 4.11)
SLC45A2 Europe 27,500 (10,500 to 73,294) 676.53 (127.73 to 1026.06) 1.10 (0.53 to 2.60)

here S = 4 * Ne * s, Ne is the harmonic mean Ne through time, s the usual selection coefficient, and h is the overdominance parameter. The median of the posterior distributions of Ne are: Ne(YRI) = 5070 (2400-8930), Ne(CEU)=1630 (130-6900), Ne(CHB)=4720 (1920-8890), Ne(CEU+CHB)=4390 (1420-8890) so the estimated s for OCA2 is 383.42/(4*4390) ~ 0.022 (0.01-0.07)

The Altai Neanderthal samples (ERR229910-ERR229914) are consensus (brown) at rs12913832.

Nicolas et al [2003] found participants with light iris colour to have twofold increased risk of AMD progression.

Vercellini et al [2014] study of 223 deep endometriosis cases, 247 endometriomas, and 301 controls found 30%, 16%, 15% were blue eyed.

Kumanov et al [2014] report 16% of varicoele cases and 62% of other Bulgarian andrology clinic patients had light (blue or green) eyes.

Di Stasio et al [2011] report blue eyes predicts risk of T1D in both Lazio and Sardinia (OR=2). Worldwide incidence of T1D varies by at least 100-fold, being highest in Finland and Sardinia (Italy) and lowest in Venezuela and China.

David Mackey's 9 point grading system.

Eye and skin darkening occur in subordinate tilapia, and lightening in dominant fish.

Hiramoto et al [2003] reported that UVB irradiation of the mouse eye leads to increased pituitary production of a-MSH, and subsequent increases in the number of epidermal Dopa-positive melanocytes in the skin. The effect was blocked by ciliary ganglionectomy. Similar effects were found [Hiramoto et al 2012] for UVA irradiation of the eyes of hairless mice: increases in plasma a-MSH, TNF-a, PGE2, and increased MC1R expression on fibroblasts. Hiramoto et al [2013] similarly report UVB irradiation of the eye increased the number of epidermal Dopa-positive melanocytes. Stress has similar effects in the mouse.

Reversal of mydriasis due to phenylephrine by thymoxamine is slower in brown eyes - mass effect due to binding to melanin.

CTNNA2 is mentioned as associated with eye colour along the Silk Road [Ulivi et al 2013]. The www.marcopolo2010.it scientific expedition studied several populations:

Country N Blue Eyes
Crimea 1020.093
Georgia 14
Armenia 174

GPNMB is an important melanosomal structural glycoprotein, upregulated by UVB, and GPNMB-siRNA treatment reduces melanosome numbers, Tyr, Trp1, Pmel17 and OA1. GPNMB belongs to the vertebrate Pmel17/NMB family, which encompasses GPNMB, Pmel17 (melanocyte protein 17), and their orthologues. Pmel17 is the main structural component of melanosomes. GPNMB can mediate melanocyte adhesion to keratinocytes through its RGD domain and is thought to be involved in the transport of late melanosomes to keratinocytes

Zhang et al [2012]

DBA/2J mouse iris pigment dispersion (IPD) is due to a premature stop codon mutation Gpnmb*R150X [Anderson et al 2002]. IPD (and pigmentary glaucoma) is diminished in D2 eyes with decreased pigment production, suggesting melanin pathway intermediates are toxic agents. Swaminathan et al 2103 explore epistasis in IPD using the B6 x D2 RI strain BXD. In the D2 mouse, IPD and iris atrophy are due to Tyrp1 mutation. Along with Gpnmb and Tyrp1, Oca2, Myo5a, Prkcz, and Zbtb20 were found to be phenotype modifiers, with the interactions varying by age.

Tyrp1 is expressed in the iris stroma, while Gpnmb is expressed by the iris pigment epithelium and Myo5a regulates Tyrp1 gene expression. The primary signaling pathway for lysophosphatidylcholine-based dendrite formation in human melanocytes involves the activation of PRKCZ. ZBTB20 plays a role in metal ion binding to melanosomes especially that of zinc. In xiplasmaB cells, Zbtb20 induction depends directly on Irf4, acting at a Zbtb20 promoter starting two novel exons. [Chevrier et al 2014].

Jiang et al 2010 find DHICA-melanin to more potent hydroxyl radical scavenger than DHI-melanin.

Genetic variations in mice with regard to the color coat have had a significant influence on the discovery of genes influencing pigmentation (Bennett and Lamoreux, 2003) and on the evolution of mouse genetics (Paigen, 2003). Studies using these strains have been responsible in revealing a wide array of important genetic pathways related to pigmentation. Recently, many groups have reported various approaches to examine these important pigment-related pathways using the iris (Anderson et al., 2008; Brooks et al., 2007; Trantow et al., 2009, 2010). The mouse iris is unique and represents a powerful opportunity for studying pigment cell biology due to its dense pigmentation. The iris and iridial melanin play an important role in influencing vision (Iwata et al., 2000; Summers, 1996; Wong et al., 2005). Moreover, several diseases, including forms of oculocutaneous albinism, Horner's syndrome, Fuchs heterochromic iridocyclitis, Waardenburg syndrome, Hermansky-Pudlak syndrome, and pigment dispersion, involve the pigmentation of the iris. Despite its biological significance, there is a definite knowledge gap concerning the basic molecular and cellular processes influencing pigmentation of the iris. Thus, gone are the days, when iris had only cosmetic influence.

Li et al [2006] cultured human uveal melanocytes and concluded that they do not proliferate in response to a-MSH, or increase levels of tyrosine hydroxylase, DOPA oxidase, and protein levels for tyrosinase, TRP-1, and TRP-2. MC1R was not detectable on Northern blot analysis.

Wagner et al [2013] report expression array results for 10 ocular tissues. The most specific (Z-score compared to all eye) iris genes were GSTA3, AREG, CHST9, C21orf94, HPDL, GALNTL5, CITED1, DLST, METTL6, C6orf124, SLC45A2, TMEM161B, RP9, GPR180, MNAT1, C1orf129, RASSF9, FAM92A1, PPIL5, LOC100130232. Their data are available here.

Probe Logarithmic Intensity Error values for ten tissues measured by the Affymetrix Human Exon 1.0 ST Arrays on pools from six human eyes [Wagner et al 2013].
Expression level (PLIER)
Choroid RPE 35.2724 43.038424.5202 1127.3 63.2947 29.1767 62.3873649.064
Ciliary Body 33.7753 39.029929.8349 653.09828.6567 24.7626 42.8304622.374
Cornea 21.2993 62.836120.7583 16.7598 13.7744 27.6684 31.6619443.525
Iris 118.57935.576723.6287 851.387188.077 29.0646 54.4613896.021
Lens 27.4626 63.536125.1454 61.3076 11.275 27.8404 31.988596.9705
Optic Nerve 28.0041 59.436230.8925 11.5371 11.7035 33.7576 42.6598213.963
Optic Nerve Head 30.1094 55.516933.9434 51.55 10.1055 27.9031 34.4655361.326
Retina 18.774 59.468432.7491 17.4515 11.5321 50.8542 34.113726.983
Sclera 34.2883 57.527933.1721 316.31120.7723 25.1094 46.6958562.464
Trabecular Meshwork26.4513 61.094127.7272 302.24838.8142 25.976 51.5084742.47

SNPs tagged by rs12896399 near SLC24A4 - none are eQTLs using MuTHER:

SNP r2 All Position ENSEMBL Annotation
rs12896471 1.000 C/G 14:92307559
rs4904866 1.000 C/T 14:92302159
rs61977801 0.9946 C/A 14:92299996
rs746586 0.9893 C/T 14:92309623
rs17809188 0.9893 C/T 14:92294769
rs35983729 0.9839 T/G 14:92321417 upstream_gene_variant T/G ENSG00000140090
rs1885194 0.9839 T/C 14:92311118
rs17184180 0.9839 T/A 14:92314043
rs4900109 0.9787 G/T 14:92297047
rs12883151 0.9786 A/G 14:92323733 intron_variant, 5_prime_UTR_variant A/G ENSG00000140090 regulatory_region_variant
rs56137766 0.9680 T/A 14:92319597 upstream_gene_variant T/A ENSG00000140090
rs941799 0.9626 C/T 14:92310481
rs4144266 0.9257 A/G 14:92322861 upstream_gene_variant A/G ENSG00000140090
rs4904871 0.9154 A/G 14:92329568 intron_variant A/G ENSG00000140090

MITF is transactivated by PAX3, LEF-1, CREB, PPARG, SOX10, Onecut-2, Brn2.

Melanosome-specific proteins

OA1 ocular albinism type 1 protein/GPR143
MART-1 melanoma-associated antigen recognized by T cellsKO mouse colour dilution
Pmel17/gp100KO decreased eumelanin in hair,
VAT-1 vesicle amine transport protein 1 homolog
CHCHD3 coiled-coil-helix-coiled-coil-helix domain containing 3
GPNMB glycoprotein (transmembrane) non-metastatic melanoma protein b

OA1 (GPR143) is located at X:9,725,413-9,765,965. Mutations in GPR143 give rise to X-linked ocular mystagmus and Nettleship-Falls type ocular albinism. In the mouse, it is a coat and skin colour gene. Expression is increased in darker colours, and overexpression upregulates MITF etc.

Peles and Simon 2010:

Uveal melanosomes from the iridal stroma contain both eumelanin and pheomelanin, the ratio of which varies with iris color. Herein, we report the absorption coefficient at lambda = 244 nm for individual human iridal stroma melanosomes from dark brown and blue-green irides. The melanosomes are nearly identical in size, but differ in the relative concentration composition, ranging from a eumelanin/pheomelanin ratio of 14.8:1 (dark brown) to 1.3:1 (blue-green or hazel). The absorption coefficient of the melanosome decreases as its pheomelanin content increases. The origin of this decrease is attributed to a corresponding decrease in the number of UV-absorbing chromophores, reflecting the different molecular volumes of the monomeric building blocks of the two pigments. In agreement with reported data on synthetic pigments, the absorption coefficient of pheomelanin is found to be slightly larger than that for eumelanin at lambda = 244 nm (by a factor of 1.2). On the basis of the reported optical properties of synthetic models, this result suggests that the absorption of pheomelanin is less than eumelanin at wavelengths of biological relevance ( approximately 315-400 nm).

BACE2 is most highly expressed of its family in mouse pigmented cells, and the knockout mimics the PMEL17 knockout - spherical rather than elliptical melanosomes and silver coat [Rochin et al 2013] .

In the Ocular Tissue Database, levels of BACE1 and BACE2 seem about the same.

Okuda et al [2014] report that the main contributors to DHI-melanin are tetramers and pentamers.

rs10809808 near TYRP1 was reported as part of the UV GeoGWAS of [Hsu et al 2013]. These authors also notes that SNPs in Vitamin D pathway genes (CYP24A1, CYP27A1, CYP27B1, CYP2R1, DHCR7, GC) also followed UV exposure data for the populations: p=3×10-4 "against [a] set of background SNPs in other genes" [the method is not given].

Melanosomes like other lysosome dervied organelles maintain and sense internal pH using proton pumps of the vacuolar V-ATPase family. NCBI gene search suggests that ATP6V1B2 (8:20.1-20.2Mbp), ATP6V0A1 (17:42.45-42.53Mbp), and ATP6V1G2 (6:31.54-31.55Mbp) are melanosome associated.

In zebrafish, Dooley et al [2013] suggest SLC45A2 and v-ATPase both can influence pigmentation:

...rescue the melanization potential of the albino melanosomes through genetic and chemical inhibition of V-ATPase, thereby increasing melanosome pH.

Assortment on eye colour:

Beckman [1963] gives:

Elston (1961) Swedish Lapps


Beckman (1963) Swedish Lapps


Husband Wife (1533 pairs) Cohen's Kappa=0.001 (JSE=0.023)
Husband 1=blue 264268150

Irisplex SNPs:

SNP Gene Consequence
rs12913832 HERC2 distant enhancer
rs1800407 OCA2 R419Q
rs12896399 SLC24A4
rs16891921 SLC45A2
rs1393350 TYR
rs12203592 IRF4 TF for MITF
(rs1800414)OCA2 H615R (Asian)
Kong et al [2014] :

Brown fat can reduce obesity through the dissipation of calories as heat. Control of thermogenic gene expression occurs via the induction of various coactivators, most notably PGC-1a. In contrast, the transcription factor partner(s) of these cofactors are poorly described. Here, we identify interferon regulatory factor 4 (IRF4) as a dominant transcriptional effector of thermogenesis. IRF4 is induced by cold and cAMP in adipocytes and is sufficient to promote increased thermogenic gene expression, energy expenditure, and cold tolerance. Conversely, knockout of IRF4 in UCP1+ cells causes reduced thermogenic gene expression and energy expenditure, obesity, and cold intolerance. IRF4 also induces the expression of PGC-1a and PRDM16 and interacts with PGC-1a, driving Ucp1 expression. Finally, cold, beta-agonists, or forced expression of PGC-1a are unable to cause thermogenic gene expression in the absence of IRF4. These studies establish IRF4 as a transcriptional driver of a program of thermogenic gene expression and energy expenditure.

20140620: Polycystic Ovary Syndrome

4-7% of women of reproductive age

2 of the following, excluding CAH, androgen tumour, hyperPRL, Cushings:

Two subgroups - lean and obese, where latter more Rx resistant.

Clomiphene citrate
Ovarian ablative therapy
Laparoscopic ovarian drilling

Lots of meta-analyses (18 in 2014, 7 genetic):

Zhang et al [2014] (6 RCTs) found metformin and acarbose equieffective with respect to improvements in ovulation rate, menstrual patterns, or changes in serum levels of testosterone, adverse events, or BMI. Significant heterogeneity of effects and not great study quality.

Cassina et al [2014] combined 9 studies to show metformin did not increase birth defects. Lin et al [2014] concluded that "with respect to CPR, a GnRH antagonist protocol is similar to a GnRH agonist long protocol. However, for severe OHSS, a GnRH antagonist protocol is significantly better in PCOS patients".

Palomba et al [2014] concluded "metformin administration increases the live-birth and pregnancy rate in PCOS patients who receive gonadotropins for ovulation induction". Haqq et al [2014] found " lifestyle (diet and exercise) intervention improves levels of FSH, SHBG, total testosterone, androstenedione and FAI, and FG score".

20140618: PPARA, PPARG and melanocytes

Agonists for both increase melanocyte proliferation, as well as melanin synthesis via "stimulation of the tyrosinase activity without an increase in the expression of tyrosinase" [Kang et al 2004], while PPARB may inhibit proliferation.

20140618: Diabetes and osteoarthitis

Some evidence that diabetic changes in joint might predispose to OA above indirect effects of overweight in T2D.

Kim et al [2014] compared OA to OA+DM chondrocytes, finding downregulated PEX-16 (a peroxisome component) and miR-223 induction. In knockdown studies, this increased chrondrocyte apoptosis and cartilage destruction in db/db mice.

A number of different miRNAs have been implicated in OA (miRs 21, 337, 181b, 199a).

Metabolic syndrome may link CVD and rheumatoid diseases via proinflammatory adipokines [ Abella et al, 2014]. RA sufferers experience a 50% increase in CVD, and Crowson et al [2011] report higher prevalence of MetS in RA subjects (33%) compared with non-RA subjects (25%). Rheumatoid cachexia leads to increased fat mass and decreased muscle mass.

The increase of fat mass is also related to the occurrence of OA and to the plethora of cardiovascular comorbidities. Regarding the influence of fat mass on OA, it is evident that biomechanical aspects are of weight in the pathogenesis of diseases. Biomechanical loading is necessary for the maintenance of cartilage homeostasis. However, abnormal loading is associated with inflammatory and metabolic imbalances, in part because it triggers the same signaling pathways as those induced by inflammatory cytokines [24]. Chondrocytes sense mechanical stress through ion channels, integrin-mediated connections to the extracellular matrix, and intracellular or membrane deformation [25]. However, OA is more common in women [26, 27] and exists in non-weight-bearing joints [28], indicating that a metabolic component is also present [29]. Of late, OA and MetS have been related to each other. The prevalence of MetS is higher in OA subjects [30, 31]. Recently, Zhuo et al. proposed a theory focused on inflammation, oxidative stress, common metabolites, and endothelial dysfunction to link metabolic OA aetiologically to MetS. They suggest that metabolic OA should be a new facet of the definition of MetS, supported by its strong associations and shared mechanisms with MetS components. However, further research is needed to define the reciprocal influence of OA on the currently accepted components of MetS [32].
Rahman et al [2013] used data from the Canadian Community Health Survey (CCHS B W1=113323, W2=115568, W3=115915) for 41000 self-reported OA sufferers and 41000 matched controls. Heart disease was increased in OA OR=1.4, after adjusting for DM, BMI, HT.

Prior et al [2014] Abella et al [2014]:

Adiponectin is also implicated in the pathogenesis of OA. It has been reported that plasma and serum levels of adiponectin were significantly higher in OA patients than in healthy controls [107, 108] and correlate with disease severity [109]. In addition, adiponectin induces VCAM-1 expression [74] and IL-8 secretion [75] in human primary chondrocytes, contributing to, respectively, the leukocyte and monocyte infiltration and the chemotactic gradient seen in inflamed joints.

Both types 1 and 2 diabetes and some medications used to treat it are associated with osteoporotic fractures. Recent meta-analyses indicate an increased relative risk for hip fracture in type 2 diabetics of 1.4 to 1.7 and in type 1 diabetics of 6.3 to 6.9. In the GLOW study, obesity, although thought to be protective against all fractures, increased fractures in the ankle and lower leg.

Yerges-Armstrong et al [2014] mention higher systemic bone mineral density (BMD) is associated with an increased risk of radiographic knee osteoarthritis. Zhang et al [2000], for example, found knee OA in women two-fold higher in those above the lowest BMD quartile in Framingham, but that progression was slowest in denser bone.

Radin et al. [39] proposed a mechanical theory of OA initiation in which increased subchondral bone stiffness may result in shear stresses within the overlying articular cartilage, leading to cartilage damage. However, more recent observations suggest rather that OA periarticular bone is hypomineralized with reduced material density [40]. Other studies examining the presence of OA in skeletons have associated osteophytes with enthesophytes (bony outgrowths at the entheses) and eburnation [41]. These features were observed to be widespread in affected skeletons, suggesting a generalized tendency to form new bone; OA may therefore represent a systemic disorder of bone response [41]. In theory, this bone-forming phenotype could result in both acquisition of a higher peak bone mass (HBM) and abnormal response of periarticular bone to later mechanical stress leading to premature OA, particularly radiographic OA.

Puenpatom and Victor [2009] used NHANES III data to study OA and metabolic syndrome. Metabolic syndrome was prevalent in 59% of the OA cases and 23% of control. Each of the 5 cardiovascular risk factors that comprise MetS was more prevalent in the OA population versus the population without OA: hypertension (75% vs 38%), abdominal obesity (63% vs 38%), hyperglycemia (30% vs 13%), elevated triglycerides (47% vs 32%), and low high-density lipoprotein cholesterol (44% vs 38%). The association between OA and MetS was greater in younger subjects and diminished with increasing age. Having OA at age 43.8 years (mean age of the general population) was associated with a 5.3-fold increased risk of MetS. Ong et al [2013] expand on this, and report OA to predispose to CVD (OR=1.5), especially angina (OR=2.2), lower than the associations betwen RA and these conditions. Yoshimura et al 2011 report on a sample of 1690 Japanese, where MetS also predicted OA with a linear relationship with number of components (overweight, HT, dyslipidemia, IGT). Nüesch et al [2011] report 1163 patients with (knee or hip) osteoarthritis (total cohort 22978) had excess all cause mortality compared with the general population (standardised mortality ratio 1.55, 95% confidence interval 1.41 to 1.70). Excess mortality was observed for all disease specific causes of death but was particularly pronounced for cardiovascular (standardised mortality ratio 1.71, 1.49 to 1.98) and dementia associated mortality (1.99, 1.22 to 3.25) - an effect mediated by diabetes history, but not by overweight. "The most striking finding, however, was the strong relation between excess mortality and walking disability. Our analysis suggested that the more severe the restriction in walking ability the more likely a person was to die early. Most of the excess mortality associated with walking problems was due to cardiovascular causes". These findings were confirmed by Hawker et al [2014] in the Ontario Hip and Knee Study, where HAQ walking score and diabetes independently predicted increased mortality, while overweight was protective.

Dahaghin et al 2007 suggested hand OA is a better test of the effects of diabetes and obesity, given it is non-weight-bearing. In a cross-sectional analysis of the Rotterdam Study, BMI>27.4 was associated with HOA (OR 1.4, 95% CI 1.2 to 1.7). The association between diabetes and HOA was only present in people aged 55-62 years (OR 1.9, 95% CI 1.0 to 3.8). Similar findings were described by Nieves-Plaza et al [2013] in cases and controls from Puerto Rico: HOA OR for diabetes, 2.2.

In the Rotterdam study, Hoeven et al [2012] found knee OA was associated with carotid IMT thickness after adjustment for diabetes, BMI, Chol and smoking. Hoeven et al [2014] in a longitudinal analysis in the RS, the best baseline OA-related predictor of incident CVD was HAQ score (over 8 ADLs).

Hardcastle et al [2012] found joint replacement 2.4-fold higher in those with BMD Z-score > 3.2. They showed four BMD SNPs to be associated with prevalent radiographic knee OA. The strongest signal (p=0.0009; OR=1.22) was on 12q3, due to SNPs in SP7. Additional loci mapped to 7p14.1 (TXNDC3), 11q13.2 (LRP5), and 11p14.1 (LIN7C). For all four loci the allele associated with higher BMD was associated with higher odds of OA.

Diabetes mellitus (DM) and osteoporotic fractures are two of the most important causes of mortality and morbidity in older subjects. Recent data report a close association between fragility fracture risk and DM of both type 1 (DM1) and type 2 (DM2). However, DM1 is associated with reduced bone mineral density (BMD), whereas patients with DM2 generally have normal or increased BMD. This apparent paradox may be explained by the fact that, at a given level of BMD, diabetic patients present lower bone quality with respect to non-diabetics, as shown by several studies reporting that diabetes may affect bone tissue by means of various mechanisms, including hyperinsulinemia, deposition of advanced glycosylation endproducts (AGEs) in collagen, reduced serum levels of IGF-1, hypercalciuria, renal failure, microangiopathy and inflammation. In addition, the propensity to fall and several comorbidities may further explain the higher fracture incidence in DM patients with respect to the general population. It is reasonable to expect that close metabolic control of diabetes may improve bone status, although its effect on reduction of fracture risk has not yet been demonstrated. However, metformin has a direct effect on bone tissue by reducing AGE accumulation, whereas insulin acts directly on osteoclast activity, and thiazolidinediones (TZD) may have a negative effect by switching mesenchymal progenitor cells to adipose rather than bone tissue. New prospects include the incretins, a class of antidiabetic drugs which may play a role linking nutrition and bone metabolism. Better knowledge on how diabetes and its treatments influence bone tissue may lie at the basis of effective prevention of bone fracture in diabetic patients. Thus, close glycemic control, adequate intake of calcium and vitamin D, screening for low BMD, and prevention and treatment of diabetic complications are key elements in the management of osteoporosis in both DM1 and DM2. Attention should be paid to treating diabetes with TZD in women with DM2, particularly if elderly. Lastly, patients with osteoporosis and diabetes should be offered the same pharmacological treatments as non-diabetics, although specific trials on the effects of anti-osteoporotic drugs in the diabetic population are lacking.

The infrapatellar fat pad (Hoffa's fat pad) lies within the joint capsule, and produces a wide variety of cytokines which may be involved in OA.

Note that PPARG KO mouse develops spontaneous OA, and agonists of PPARg inhibit inflammation and reduce synthesis of cartilage degradation products both in vitro and in vivo, and reduce the development/progression of cartilage lesions in OA animal models. In humans, PPAR agonists, have been reported to affect bone cell function and fracture risk. PPARG action may be to increase adipogenesis at the expense of osteoblast formation, while PPARA agonists may prevent bone loss (metformin seems neutral). The two commonly used agents rosiglitazone (now withdrawn because of adverse events) and pioglitazone

Framingham osteoarthritis study:

Hart, Doyle and Spector [1995] found pG, sChol, to be associated with radiographic knee OA in 1003 women from the Chingford Study.

Mork et al [2012] confirmed the BMI and knee (but not hip) OA relationship using the Norwegian HUNT study

Kadam et al [2010] and Clockaerts et al 2012] find statin usage reduces progression of OA.

Panoutsopoulou et al [2013] carried out a Mendelian Randomization analysis of FTO, BMI and OA (arcOGEN N=15000). They found rs8044769 to predict OA risk, an effect fully attenuated by adjusting for BMI. This variant is associated with BMI, but not diabetes.

Haara et al 2003: screened 7217 Finns (cluster sampled population study, the Mini-Finland Health Survey) in 1979, X-raying hands of 3600 Ss oversampled for musculoskeletal disease. Prevalence of Heberden's nodes 10%; The subjects with BMI > 35 kg/m2 had an about twofold risk of OA in any finger joint or symmetrical DIP OA as compared with subjects with normal BMI (20.0-24.9 kg/m2). Male smokers had reduced OA rates: 20+ cig/d OR=0.4 (0.2-0.9). Total mortality increased in OA females (RR=1.2), not in males (RR=0.9).

Visser et al 2014 report on the Netherlands Epidemiology of Obesity (N=5315) where fat percent and WHR both predicted male and female OA (OR ~ 1.2-1.4), with visceral adipose tissue a stronger predictor in males (OR=1.5)

Jonsson et al 2011 studied THR and TKR in 6100 Icelanders (AGES-Reykjavik Study). Of 5170 with hand CT and photos, 508 had undergone JR. There was a significant association between JR and hand OA to coronary calcium, carotid plaques and periventricular white matter hyperintensities.

20140616: Metaphysics

Jeffrey Ketland epigrammatizes:

...for the metaphysician, there also is a more or less canonical list of the kinds of things one are interested in: properties, relations, quantities, abstract entities and structures, formal systems, moral values, propositions, pieces of discourse, possible worlds and fictional entities.

Added 20180609:

Peter Wolfendale has it as:

Provisionally, let us say that natural science is the inquiry concerning what is, and that metaphysics is the inquiry concerning what "what is" is. This is just what it is for them to share the world as their object. The former is the inquiry into the contents of the world, while the latter is the inquiry into the structure of the world as distinct from its contents.

...the most abstract theorising that is still subject to the trial in experience...a delicate balance of a priori and a posteriori.

...[Further], we will distinguish between those discourses that aim at objective truth (mathematical and natural science) and those that aim at non-objective truth (ethics, politics, aesthetics, etc.).

It is only insofar as the claims of the natural sciences provide reasons that can be deployed in metaphysical debate that metaphysics is objective. The only way to avoid this conclusion is to either abandon the objectivity of metaphysics (e.g., transcendental idealism), to hold that there are special objects of metaphysics distinct from those of natural science (e.g., theology), or to hold that there is a special mode of access to objects different from that of the natural sciences (e.g., classical rationalism). I take it that there are good reasons to avoid all three.

...a few exemplary [metaphysical questions]: What are entities? What are properties? What is essence? What is modality (possibility, necessity, actuality, etc.)? What is causality? What are part-whole relations? What is time/space?

20140527: Twin rating contrast effects

They are indirect genetic effects. The model is a reciprocal phenotypic interaction between siblings (or MZ twins), but in this case, the term "contrast" is used because the interaction is only in the mind of the rater (who "contrasts" the phenotype of sib1 with that of sib2). The same statistical model applies to both.

The coefficient on the effect of sib1 on sib2 (and on the reciprocal path) is negative. The observed effects are a change in the phenotypic variance for members of groups where this is acting, and a shift in the covariance (all varying according to the degree of relationship). A higher MZ than DZ phenotypic variance has been reported for the ADHD ratings in multiple studies.

20140509: On the Mendelian-biometrics dispute

Well, I was pleasantly impressed by Cox's line of argument, recalling that Galton too was somewhat of a saltationist (transilience), and Huxley as you note. Galton gets as much a serve from Wallace in his 1895 article as Bateson.

By 1911, it's obvious that the science has moved on - Fisher (Mendelism and Biometry) completely accepts Bateson's work despite being in the biometrical camp. In later writing he implies his thoughts on selection were "immature" even in the early 1920's.

I agree that Bateson later thinks of himself as an evolutionist rather than one of those overzealous adaptationists ("Not till knowledge of the genetic properties of organisms has attained to far greater completeness can evolutionary speculations have more than a suggestive value"), and quotes Samuel Butler approvingly in his wonderful 1909 piece. Bateson's thoughts about the genetics of speciation are pretty modern, but those of Weissman in the preceding chapter are just as good.

20140508: Askenazi genetics

Bray et al 2010

The strongest signal of selection unique to the AJ population was on chromosome 12 (110.6-111.72 Mb hg18). This locus was not found to be under selection in either the EA population or among reported selected regions of the HGDP Middle Eastern populations (40), making it an apparent Jewish-specific selected locus. Interestingly, this locus is also within a region that was recently identified as having high IBD across many Jewish Diaspora populations (13). This region contains 18 genes, including the mitochondrial aldehyde dehydrogenase gene, ALDH2, [as well as] TRAFD1/FLN29.

20140501: On scientism, and other blog comments

I am most bemused by the idea that we can split off science (and scientific methods of inquiry) from mathematics and logic. Scientisms of all flavours suppose the existence of scientific reasoning, which is rational and mathematical in nature. Not having read the Lange paper cited, I would still strongly argue that a logical argument showing hypothesis X cannot be true is just as scientific as an experimental disconfirmation.

Joe - re testimony, the historical sciences and history. I suppose it depends on whether you think anthropology, sociology and psychology are sciences or not, and whether we can bring insights to bear from the social sciences onto any particular historical question. Here's a lovely scientistic bit from

Martin "Towards a scientific history of religions" (Ch 2 of "Theorizing Religions Past: Archaeology, History, and Cognition") via Google Books,

For historiography to be a reliable way of knowing, then, its methods must at least approach the standards for verification that exist within science, especially the nonexperimental "historical" sciences like astronomy, geology, paleontology and evolutionary biology...historiographic interpretation must be subject to certain general standards and tests - of human behaviour, of logical antecedents and consequences, of statistical or mass trends...

There is or was something called psychohistory, and Erik Erikson (the adolescent psychiatrist) did write Young Man Luther: A Study in Psychoanalysis and History, where he tried to use contemporaneous observations of human behaviour and his ideas about "the crisis of identity" to interpret Luther's life and works. I'm guessing he would suggest Luther would have said something like that (or if he hadn't, he later remembered saying it), because it would fit his theoretical model nicely.

I think they also mostly subscribe to a minimal ethic that characterizes the scientific endeavour as a humanistic institution: knowledge is a good that should be increased, for both intrinsic and extrinsic reasons; all are free to think and question, and ideas should be exchanged freely; participants in the great work of increasing knowledge are all on an equal footing. And a pragmatic belief that institutions not respecting these maxims will be outcompeted by institutions that do.

I am sympathetic to the the presence of overlap and interplay between philosophical and scientific thought (why else would I reading this blog), but...

Doesn't logic look more and more like a branch of mathematics? The mathematicization of science has been discussed here previously, and the arguments about taxonomy are more and more statistical ones - what are useful criteria for natural categories, where nature is what scientists specialize in studying. Even the various problems of induction that philosophers have elucidated have particularly mathematical and statistical solutions in scientific practice viz Neyman-Pearson, Fisherian fiducial inference, Edwards' likelihood principle, various flavours of Bayes, and less mainstream approaches like Solomonoff induction (which do look attractive to me).

It is unclear what people use the phrase "scientific logic" to mean, but it is, I think, informal logic that includes a number of heuristics enriched by philosophical thought eg Koch and Bradford-Hill criteria for causation in epidemiology, Popper (whether or not well digested), a vaguer smattering of logical positivism (perhaps just the spirit), and the statistical schools mentioned above. It is just rigorous enough to avoid getting into trouble in the long run.

So were IAU discussions about the definition of a planet philosophical? They certainly involved reasoning and dialogue and rhetoric, and I don't think they involve any particular hypotheses that can be tested. They are probably not particularly important to the future direction of the science either.

I was looking for discussions of "scientific logic" and can't resist sharing

Nothing could have been contrived; every thing owes its existence to the remarkable course of scientific logic. It is not false modesty when I say that I feel myself to be merely the instrument of this logic.

From, Wilhelm Reich The function of the orgasm

[On science and sexual orientation]

A topical example is the recent Academy of Science of South Africa (in collaboration with the Uganda National Academy of Sciences) report


- apparently an answer to the Ugandan president's claim that "experts prove there is no connection between biology and being gay".

...I think the 70's science wars (against say Jensen and later Wilson) were not driven by rigidity on the part of the scientists, but by a willingness to follow where the data seemed to lead. The self-correcting wandering course of sciencific knowledge runs a lot slower than social and political forces. As the OP points out, the scientific idea that sexual orientation seems to have strong biological features and is resistant to change by cultural or psychological interventions, is just as unpalatable to some people as the idea that differences in individual intellectual ability might persist, no matter how rich and equal the environment, absent biology-based interventions of some kind.

Actually, I'll give a quote from my favourite rigid scientistic philospher of the moment (Mario Bunge, Matter and mind: A philosophical inquiry):

My reason for advocating a scientistic approach to ethics is that I take this discipline to be concerned with moral dilemmas, all of which are social problems - the domain of social science. Indeed, I suggest that moral problems arise when scarce resources are handled by persons with unequal power. This is why the more divided a society, the more severe the moral problems that arise in it. I also suggest that the best way of resolving such conflicts in a fair and peaceful way is to discuss and bargain in the light of what is known about the social systems in question, so that the stronger party may compensate the weaker one, e.g., by offering it a share in a bunch of desirable items. Obviously, the scientistic approach to morality opposes moral relativism as well as moral absolutism (or dogmatism). Contrary to the fashionable opinion, that we should tolerate all values and norms because they are only rooted in gut feelings or in custom, a science-oriented moral philosophy will promote fairness and reciprocity, for they favor individual welfare and social progress. The same approach will also favor close contacts between ethics and ontology, because moral behavior is a form of social behavior, the justification of which calls for a social ontology. Indeed, different social ontologies suggest different moral philosophies...

Not only scientistic determinism but ontological determinism as well! I'll mention he promotes a "scientia" where philosophy is central to scientific practice, but science strongly constrains what kinds of metaphysics, for example, are sensible in a Pragmatic kind of way.

There might be two issues here - the utility of the openness of publicly funded research, and the long term utility of basic research. My acquaintance with corporate funded research gives me the impression that it is often a little secretive (publication in the open literature can be delayed by several years - after all it may suddenly become profitable if some additional information comes along), and short-termist (investors can make better returns elsewhere).

The Impact of the Social Sciences: How Academics and their Research Make a Difference By Simon Bastow, Patrick Dunleavy, Jane Tinkler (2014) documents economic impacts of UK university research in that area.

Further to my comments


Since the seminal papers by Arrow and Nelson, it has been accepted by most economists that a private enterprise economy fails to provide adequate incentives for investment in knowledge production. There are several reasons for this assertion. First, there is inherently a high degree of uninsurable risk and uncertainty that increases as we move along the basic research end of the research spectrum. Secondly - and this is specific to knowledge as a commodity - it is believed that knowledge, once produced, is in some meaningful sense "on the shelf". As a consequence, neoclassical/mainstream economics held that, once produced, knowledge was freely available to all, including those firms that may have made no contribution whatever to the production of the knowledge: A classic "free rider" situation.

[On axiomatization of Darwinism]

Warren Ewens' commentary on the "Outline" was good value, and I found Grafen's response to the commentaries more useful than the target article.

In the spirit of some the comments in the thread:


Informal models have always been used in biology to guide thinking and devise experiments. In recent years, formal mathematical models have also been widely introduced. It is sometimes suggested that formal models are inherently superior to informal ones and that biology should develop along the lines of physics or economics by replacing the latter with the former. Here I suggest to the contrary that progress in biology requires a better integration of the formal with the informal.

[Specifically on determinism and causality]

Re Marko's comments above re causality - all the causal models I deal with involve injection of randomness (it doesn't matter - except metaphysically ;) - if this is quantum or classical). Just consider the transmission of a gene from a diploid parent. There can be an iron law that the distribution of outcomes follows even if an individual event can't be predicted.

Hi Liam U - my own sympathy lies with formal models as far as empirical data can currently support them. Big chunks of genetics are one of the more mathematical areas eg population and evolutionary genetics, so I think the prospects for a formal Darwinism of some kind are good, and there are lots of contributions from statistical physics as well as game theory. For example, I couldn't resist this paper from Physical Review Letters that uses "Glauber's formula and self-consistent equations of the Schwinger-Dyson type" to solve for probabilities in recombinant inbred lines.

Nevertheless, we usually end up with some kind of linear approximation to truth at the moment in biology overall. This is true even where the model may not be a good statistical fit to data (because the sample is large), but in terms of understanding might be the right complexity. A slightly different phenomenon is where a physically accurate model might be a "worse" fit in likelihood terms than a linear model that includes higher order (polynomial) terms added in that "soak up" noisy features of data. You may be aware of some to-and-fro about this in climate modelling, for example.

There are a few recent papers that show the stochastic optimization that evolution does can get stuck in local maxima (well we knew that) or even in between. One I came upon discusses one tiny domain, RNA sequence and secondary structure, where for 20-base RNA molecules there are 1e12 possible genotypes but only 11219 possible phenotypes. In this situation you get "survival of the flattest", where a less advantageous phenotype that is produced by a wider range of genotypes will be found more easily, and then fixed. This paper gives a model in which a deleterious mutation will often go to fixation if the landscape is relatively flat. The point is that the shape of the fitness landscape is set by the constraints of physics, chemistry (eg the "choice" of nucleotides RNA has ended up with - contingent or inevitable?), and then the ecology (!)

[On science and aesthetics]

Dear SG - "at the level of particular individuals' interest, it will be basic psychology, combined with personal life history. For different groups' interest, it will be sociology, perhaps with a touch of economics.."

So beauty is a completely socially determined concept? Where does basic psychology come from? And art that stands the test of time does so purely due to extrinsic factors? You don't feel the power of art objects from other cultures, say Australian indigenous art, where you have no idea of the underlying cultural associations? Your first query to Dan is I think answered by the fact that the act of reporting generates additional brain activity over that just associated with perceiving and cogitating about X. Re "flow", I think suppression refers again to activity in a particular region (the effects on consciousness may be positive). Dear ejwinner - my point about the beauty of Abstract Expressionism was that it runs contrary to embedded cultural expectations precisely because there is no obvious meaning, but some people saw something in it (and curators paid money for it that the ordinary voter thought was ridiculous). Either it is purely cultural (a rather crude sociology of taste, status signalling etc), or there is something that different observers that share a basic neurological organisation agree on as pleasing to the eye, which is why it has been designated as a store of value different from the work of "lesser" artists. Anyway, this drifts a bit from the target article - I'll stop here.

I can't resist one last response on the aesthetics discussion. I certainly don't dispute the cultural aspects of art, and I use terms like "power" because that is what I experience looking at or listening to various cultural artifacts. And I am perfectly aware of and am fond of the insights of mathematical theories of aesthetics. I'll just echo Robin and Coel in saying I think useful contributions are possible at low and high levels. Coming back to the OP via consciousness and aesthetics, which is kind of what started me on this sidetrack


Although they often occur at moments where something unpredictable happens, the incidence of chills increases with exposure to a piece of music, at least for moderate amounts of exposure, as though learning the musical structure helps the listener recognize the critical chill-evoking deviations. These findings underscore the importance and paradoxical nature of expectation in our experience of music. Our aesthetic response seems to hinge on violations of the expectations induced by our knowledge of musical rules, yet repeated exposure enhances the response. It is as though the aesthetic response is driven by something that lacks direct access to our explicit memory (because the response to the expectation violation is enhanced even though consciously we know in advance what will happen), but that benefits from the enhanced structural representations attained from repeated exposure.

[On continental drift]

I thought this well written too. A comparison with Einstein, as pointed out above is not so useful, but a comparison of scientific acceptance of Darwin's and Wallace's ideas might be interesting, especially w.r.t. mechanism. Ernest Mayr claims that "during the first 60 years...natural selection was accepted virtually only by naturalists. I do not know of a single experimental biologist prior to 1920 who was a consistent selectionist" (as opposed to evolutionist), and looks to the settling in of genetics and then the Modern Synthesis as late as the 1940s as the final conclusive evidence.

Continental drift was initially welcomed by many European (Swiss, Dutch, German) and British geologists, but it just didn't sustain its initial impetus. Trumpy [2001], "a life-long drifter" (b. 1921) http://www.mantleplumes.org/WebDocuments/Trumpy2001.pdf sees a mixture of the American hostility mentioned plus increased parochiality in European geology. But it is telling that in both cases, mechanism was the final arbiter, and took a similar amount of time to be settled. Lake [1928] says "some sort of continental drift [is] a matter not of interpretation, but of observation...The Himalayas were probably formed by Central Asia being pushed over by India. But these movements are very different from those imagined by Wegener. The great difficulty in Wegener's theory is to explain lateral movement at all...".

[On reduction]

Regarding reductionism, there are plenty of philosophers who are reductionists. Here's Patricia Churchland,

Churchlamd [1998]

I am convinced that the right strategy for understanding psychological capacities is essentially reductionist, by which I mean, broadly, that understanding the neurobiological mechanisms is not a frill but a necessity... Insofar as I am trying to discover macro-to-micro explanations, I am a reductionist...[M]any philosophers who agree with me on the brain-based nature of the soul nonetheless rail against reductionism as ridiculous if not downright pitiful...

Specifically on Wittgensteinian type arguments:

Much has already been said on this matter elsewhere, and I shall bypass a lengthy discussion of philosophy of language with three brief points. (1) It is rather far-fetched to suppose that intuitions in the philosophy of language can be a reliable guide to what science can and cannot discover about the nature of the universe. (2) Meanings change as science makes discoveries about what some macro phenomenon is in terms of its composition and the dynamics of the underlying structure. (3) Scientists are unlikely to halt their research when informed that their hypotheses and theories "sound funny" relative to current usage. More likely, they will say this: "the theories might sound funny to you, but let me teach the background science that makes us think the theory is true. Then it will sound less funny."

Hi EJ, re Churchland's remarks being "somewhat unpleasant" - remarks of McGinn's in past reviews over 20 years have been frequently vigorous (eg), and I suspect Churchland's tone reflects this history. I quoted Churchland to address the idea it is scientists who use "commonsense arguments" against particular philosophical positions, and who just should be told off as ignorant. Indeed this is the line that McGinn seems to take (though the bit of the original review I can see ends at "Churchland's reasoning is as unsound..."). If people can get emotional about geological theories, they can get emotional about theories of consciousness - but the inevitable advance of neuroscience will soon end all these squabbles ;)

I thought the target paper was a nice transect through a whole lot of areas. And I have learnt a new word, ostension ;) I think the argument in the paper against this fails against the objection that science is all about pointing as exactly as possible.

Mario Bunge pushes for a Pragmatic metaphysics, with the term realism reserved for the epistemology, and ontology separately "as general science", "in the very marrow of science", as "all generalizations about facts...involve ontological categories". This implies our ontology will be changing all the time as we learn more, so I guess we shouldn't get hung up on the changeable nature of scientific truth and underdetermination. We just provisionally accept a naturalistic ontology, devise tests for it, and discard or revise it as necessary. In his way of thinking, the realist epistemological assumptions are that reality is knowable, and logic/maths an appropriate tool, so we assume we will continue to progress. Metaphysical realism as defined by Putnam is that there is a single true representation that our theories can converge towards, but I don't think such a strong assumption is necessary.

Objects are central to his ontology, and Bunge makes much of a distinction between objects and processes that act like objects: his example is photons and phonons, where I presume a structuralist view might make phonons first-class objects.

Unfortunately, he then points out that epistemology depends on ontology: "epistemological realism without materialism is vulnerable and useless to science, because one may adopt Plato's or Hegel's objective realism", which is a bad thing!

"Physicists are philosophers too":

We do not have to take physicists at their word when they make philosophical forays: we can avail ourselves of certain formal tools to find out what their theories are really about. For example, around 1930 Bohr and Heisenberg claimed that the task of physics is not to find out what nature is, but what we can say about nature. Any realist philosopher could have told them that the people who analyze what physicists say are philosophers and historians of science, whereas physicists like themselves study atoms, stars, and things, as shown by their formulas and experiments.

I don't want to be accused of be meta-ethical, but I don't think language that implies blameworthiness and perhaps indirectly the discussion of appropriate sanctions for those who have failed in their duties but have some excuses is really the way to go. Tough love? A lot of this seems to be more social etiquette - being polite and appreciative that others are being helpful. I too have worked in care of the terminally ill, though in a professional capacity. If people are grumpy because they are in chronic pain, or just feel crappy, I don't think carers should give in to feelings of resentment, or alternatively praise people to their faces for being such a good patient - it just applies inappropriate moral pressure. Relatives of a patient might be in more of a position to say something judgmental, and if they don't, too bad. Maybe working through pain and distress in a nonconfrontational reflective (in the psychotherapeutic sense) way might be an option. I know you are putting forward the idea that we are respecting the patient by making such demands, but I can see it just as easily backfiring, if only because it will not always be clear what someone actually has the capacity to do. I mean, exhorting someone to offer up their suffering to God may also be useful to them on some occasions too...

As students, we were brought up on Talcott Parson's and the sociology of the "sick role". It isn't quite so useful for the hospice situation, as its main duties for the patient include seeking out appropriate help and working hard to get better. Nevertheless, there are papers in the medical sociology literature that might be relevant...

eg Illness as moral occasion: restoring agency to ill people http://hea.sagepub.com/content/1/2/131.short

Most of the writing on the "dying role" seems to be limited to intrinsic or family-centred moral issues, eg accept the necessity of care by others and not fight against it. The "good death" is just as much for the relatives and onlookers to "the drama" in that sense.

20140422: On Print-On-Demand

Lawrence Watt-Evans lwe@sff.net writes:

On 2014-04-18 16:50:49 -0400, Lynn McGuire said:
An interesting side note. This book (and its other series members) were printed on demand. The dates of printing are on the last page. The POD date of the second book is 04/01/2014 and the third book is 04/06/2014. I did not know that they could publish and sell POD books for under $10. We get our POD software manuals printed at www.lulu.com and they are not that cheap. Of course, we are printing in the 8.5x11 format.
The large format is significantly more expensive, but also Lulu is significantly more expensive than CreateSpace (part of Amazon) or Lightning Source. It's got far more options -- you can't GET 8.5 x 11" at CreateSpace, for example -- but it's definitely a high-cost service.
Lightning Source is complicated. On a direct comparison copy-by-copy basis, yes, it's significantly cheaper, but then there's set-up and catalog fees, so it depends on exactly what you're doing.

20140417: OSX gfortran -static

Izaak B Beekman (zbeekman@gmail.com) comments:

In the event that anyone else wants to try this, I have found a working
solution, however it is ineligant and better solutions may exist. The
link line with gfortran48 on Mac needs to look something like this:

gfortran48 -static-libgfortran -lgfortran -lgcc -lSystem -nodefaultlibs
-mmacosx-version-min=10.6  -fbacktrace -Wall -Wextra -pedantic
-std=f2008 -Wno-unused-parameter -Wno-maybe-uninitialized -O3
 -o  /opt/local/lib/gcc48/libquadmath.a

This creates a redistributable binary which will work on OS X 10.6 and
newer, and does not require XCode or GCC/GFortran to be installed. My
sources are in C and Fortran, and the main program is in Fortran. The
critical steps are to specify -nodefaults -static-libgfortran
-lgfortran -lgcc -lSystem and pass the full path of the static
libquadmath.a library.

More generally, from Thomas Koenig on c.l.f (20170906):

gcc -static-libgcc hello.o -Wl,-Bstatic -lgfortran -lquadmath -lm -lc


gcc -static-libgcc -o foo omp_parallel_1.o -Wl,-Bstatic \
    -lgfortran -lquadmath -lgomp  -Wl,-Bdynamic -lm -lpthread -ldl

20140414: Getting my Alcatel-Lucent IP Touch phone into DTMF mode


20140407: Multivariate normal integration by different algorithms

Mendell-Elston and Genz MFT log-likelihood versus heritability for Weil et al (1970) dataset: 32 litters (size 5-13), of which 16 received treatment).

Mendell-Elston v. Genz MFT likelihoods

Tests for the fixed effect are more comparable:

Model M-E -2LL Genz -2LL LMM -2LL df Naive GLM
~ trt 237.8613 238.6081 -346.9372 3 259.1418
~ 1 241.5857 242.3306 -343.3822 2 268.1579
LRTS 3.7244 3.7225 3.5551 1 9.0160

Using 32 litters as 2 large pedigrees or 32 separate pedigrees. Used Genz's mvndst with either MAXPTS=50*NDIM or MAXPTS=1*NDIM gave exactly the same results and time.

PedigreesMVN algorithmMaximizerLL h2Beta(treatment)Time (sec)
2 M-E BOBYQA -118.9310.5 0.487175 1.28
2 M-E VARMET - - - 0.07
2 Genz BOBYQA -119.3040.5 0.473954 179.94
32 M-E BOBYQA -118.9310.5 0.487175 0.02
32 M-E VARMET -118.0670.79 0.460134 0.03
32 Genz BOBYQA -119.2300.5 0.502297 94.2

20140329: Books recommendationf from other people

Thorne Smith, "TOPPER" and "THE NIGHT LIFE OF THE GODS." are on gutenberg.au.

Brian Scott (again):

Sherry Ramsey, One's Aspect to the Sun; not quite thriller, not quite space opera, and really quite good. Another good one that doesn't quite fit but might still appeal is Becky Chambers, The Long Way to a Small, Angry Planet, and a third is Jacqueline Koyanagi, Ascension.

Nina Allen on the Clarke awards:

Emily St John Mandel Station Eleven 2015 - post-apocalypse.

Sarah Hall The Carhullan Army

And the movie Frozen (2005), a ghost story.

S.K. Dunstall Linesman (2015)

G. Willow Wilson, Alif the Unseen (2012) [WFA winner]

Charles L Harness Redworld (1986)

Robert Jackson Bennett, City of Stairs (2014)

John Darnielle, Wolf In White Van

Ariel Djanikian, The Office of Mercy

The Insult by Rupert Thomson

Dinosaur Beach by Keith Laumer

The Deaths of Tao by Wesley Chu

The Martian by Andy Weir

Thomas Piketty CAPITAL in the Twenty-First Century

Charles L Harness Redworld (1986)

Jo Walton The Just City

Adam Roberts liked:

David Ramirez, The Forever Watch
Nnedi Okorafor, Lagoon - I thought OK, but not brilliant.
E J Swift, Osiris (2013) Cataveiro (2014)

Abigail Nussbaum:

A Stranger in Olondria by Sofia Samatar

Europe in Autumn by Dave Hutchinson.

Jim Hines (and others) liked:

Delia's Shadow by Jamie Lee Moyers

Dirk van dem Boom liked:

Gary Gibson, Stealing Light, Nova War, Empire of Light

James Nicoll recommends:

The Language of Power, Rosemary Kirstein
The Steersman, Rosemary Kirstein
The Lost Steersman, Rosemary Kirstein
The Steerswoman's Road, Rosemary Kirstein

A Door Into Ocean by Joan Slonczewski
A Voice Out of Ramah by Lee Killough
Don't Bite the Sun by Tanith Lee
Ninefox Gambit by Yoon Ha Lee
The Next Continent by Issui Ogawa

Tony Nance's counteroffers:

Addison - The Goblin Emperor
Brackett - The Book of Skaith (if I'm allowed to cheat)
McKinley - Sunshine
Liz Williams - Snake Agent

Max Gladstone _Three Parts Dead_ et seq (fantasy police procedural)
(several others agreed).

Lin Carter _Tara of the Twilight_ ?

Michael Cisco _The Narrator_, _The Divinity Student_, _Member_.

Some of his other books, particularly _Celebrant_ seem deliberately uninviting if you don't know what you're getting in to. That said, the payoff is huge.

Edgar Pangborn on Gutenberg.

A r.a.sf.w thread on fantasy recommendations:

[Brain Scott's] shortlist, based partly on how strongly [he] was affected
on first reading, would certainly include: [removing ones I have read]

_The Blue Sword_ (McKinley)

_The Dark Is Rising_ (Cooper)

_The Door into Fire_, _The Door into Shadow_, _The Door
into Sunset_ (Duane)

In addition, the entire Essalieyan corpus by Michelle
Sagara (writing as Michelle West), though I can't pick out
any one book, and the same for her slightly less ambitious
Elantra series.

Leslie Barringer _Shy Leopardess_ 

_Between the Forest and the Hills_ (Ann Lawrence)

_Chase the Morning_, _The Gates of Noon_, _Cloud Castles_
(Michael Scott Rohan)

_The Deeds of Paksenarrion_ (Moon)

_The Grey Mane of Morning_ (Chant)

_The King of Attolia_ (Megan Whalen Turner)

_The Orphan's Tales: In the Night Garden_, _The Orphan's
Tales: In the Cities of Coin and Spice_ (Valente)

_A Rumor of Gems_ (Ellen Steiber)

_Sunshine_ (McKinley)

Chris Buckley <alan@sabir.com>

Other very, very good additions might be
_One for the Morning Glory_ (Barnes)

More personal favorites
_A College of Magics_ (Stevermer)
_Juniper, Gentian, and Rosemary_ (Dean)
Kencyrath series (Hodgell)
Abhorsen series (Nix)

More from Brian Scott:

Ilona Andrews's Kate Daniels series is another automatic
purchase; it began in 2007 with the authors' first novel.

Elizabeth Bear's first novel came out in 2005 and was
science fiction.  Since then she's written much more
fantasy than science fiction; I especially recommend the
Edda of Burdens trilogy, and most especially its first
volume (2008), _All the Windwracked Stars_.  She also has a
couple of good novels written with Sarah Monette, who (as
Katherine Addison) recently wrote the excellent _The Goblin

Robert Jackson Bennett's _City of Stairs_ (2014), his fifth
novel, was rather different and quite good; I've not read
his first four (starting in 2010).

Marie Brennan's first novel came out in 2006; my favorite
of hers is _Midnight Never Come_ (2008), the first volume
of her Onyx Court tetralogy.

Kevin Hearne's Iron Druid Chonicles are quite good; his
first novel is from 2011.

Helen Lowe's first novel was the excellent YA _Thornspell_
in 2009; she has since written _The Heir of Night_ and its
sequel _The Gathering of the Lost_, and I understand that
the third volume, _Daughter of Blood_, has been written;
the series is very good high fantasy.

Melissa Marr started in 2007 with _Wicked Lovely_, the
first of a decent YA series.  The unrelated _Untamed City:
Carnival of Secrets_ is nominally also YA and is, I think,
better; I'd like to see a sequel.  Her adult _The Arrivals_
is good.

Seanan McGuire's Toby Daye series, which started in 2009
and currently stands at eight books with five more under
contract, is excellent.  I'm less taken with her InCryptid
fantasy series, but her novel _Indexing_, originally an
Amazon serial, is quite good.

Henry H. Neff's first novel was _The Hound of Rowan_
(2007), first of the Tapestry pentalogy.  Don't be fooled
by the opening, which has a slightly-younger-than-YA feel:
the protagonist of the pentalogy starts young but is forced
to grow up very quickly.  It's an ambitious effort, and
Neff pulls it off well.

Rachel Neumeier's first book, _The City in the Lake_, came
out in 2008; it has some of the same feel as the
Riddle-Master trilogy.  She has since written a complex
high fantasy trilogy (the Griffin Mage trilogy), the
excellent _House of Shadows_, and the YA fantasies _The
Floating Islands_ and _Black Dog_.  I can recommend all of

Gaie Sebold's _Babylon Steel_ (2011), her first novel, and
its sequel _Dangerous Gifts_ are solid fantasy entries in
an interesting setting -- three settings altogether,

Nalini Singh's Guild Hunter and Psy/Changeling series are
definitely paranormal romance, but each has a long-term
story line that advances significantly in each book, and
she's a good storyteller.  If you don't mind the requisite
romance elements, both series are enjoyable.

Dianne Sylvan's first novel was her first Shadow World
novel, _Queen of Shadows_, in 2010; there have been five
more in the series since then.  Not exceptional, but solid
urban fantasy.

Eileen Wilks's Lupi series started as decent urban fantasy
with a romance element and has steadily improved and moved
even further away from romance; the series is now one of my
handful of automatic purchases.  It started in 2004, with
her first novel.

All of these have been published traditionally.  There are
also some self-published writers whose fantasy I've
enjoyed, among them the following:

D.K. Holmberg's Cloud Warrior Saga, currently at five
books, is readable high fantasy.  His Painter Mage series,
currently at two books with a third due in a month, is
readable urban fantasy.  Nothing special, but as good as
much that is published traditionally.

Andrea K Höst has been self-publishing fantasy with a bit
of science fantasy since 2010; her most recent book, _The
Pyramids of London_, is the start of a fantasy alternate
history series and is very good.  Her Touchstone trilogyd it.
fantasy than straight fantasy, but I also greatly enjoyed

L. Shelby has been self-publishing novels since 2014; she
writes very stylish fantasy emphasizing culture rather than
traditional typical fantasy elements.

Frank Tuttle started self-publishing the Markhat Files in
2008; these feature the finder (think private detective)
Markhat in a world of vampires, ogres, trolls, etc.
Somewhat in the vein of Glen Cook's Garr Garrett P.I.
novels.  _All the Paths of Shadow_, in a very different
setting, is even better.

Several people recommend Edward Ormondroyd _David and the Phoenix_ now on Gutenberg.

Sea Wasp commented "In the 80s, I'd also mention Doomfarers of Coramonde and Starfollowers of Coramonde, the duology by the sadly no-longer-with-us Brian Daley."

Jo Walton likes:

M.J. Locke Up Against It

Denis Diderot Jacques the Fatalist and His Master "...tops and questions itself, plays complex games with narrative and time and nested stories, has the world's most intrusive narrator, who at one point starts to tell us a story while the characters sleep then decides he is sleepy too and gets into bed between the characters."

Jim Hetley, writing as James A. Burton: Powers. Dominions.

Katarzyna de Lazari-Radek, Peter Singer [2014]: The Point of View of the Universe: Sidgwick and Contemporary Ethics

Beth Revis' unsettling but oh-so-compelling Across the Universe, in which 100 VIPs from Earth are cryogenically frozen on the generation ship Godspeed. Multiple generations spool out during Godspeed's voyage, but their real purpose is to ensure that these cryo-pods stay perfectly preserved. Once unfrozen, these Earthlings will be the first to step onto their new planet. [Sequels are A Million Suns, Shades of Earth]

2016 Hugo inspired lists (Brian Scott's):

James A Burton,  Dominions
Ann Leckie,  Ancillary Mercy
Michelle Sagara,  Cast in Honor
Michelle Sagara (as Michelle West),  Oracle
Graydon Saunders, _A Succession of Bad Days_.
Elizabeth Bear,  Karen Memory
Karen Lord,  The Galaxy Game
Joel Shepherd,  Originator

Several people recommend the Saunders, and Chris Buckley mentioned Ninefox Gambit by Yoon Ha Lee as similarly good.

M Suddain's Hunters & Collectors: a space opera about a food critic.

Scott Hawkins The Library at Mount Char (urban fantasy).

20140321: population genetics

Pool et al (2010). Population genetic inference from genomic sequence variation

20140320: Some more compiler timings

Compiler Code Time Comment
sunf95 12.3 weil.in 2m35.933s Wrong answer!
gfortran 4.6.3weil.in 3m33.351s

20140319: Anemia as a presentation for colorectal carcinoma

Hamilton et al (2008) estimated likelihood ratios and PPVs for anemia and subsequent diagnosis of CRC (6442 cases, 45066 controls). For females (227/3021 cases, 58/21138 controls) the LR for Hb < 9 g dl-1 was 41 (30-61).

Age (y) UK Incidence 2003< 9 g/dl 9-9.9 g/dl
70-79 0.0021 8.6% (5.4-14)3.6% (2.1-6.0)
80- 0.0027 7.1% (4.5-11)2.2% (1.5-3.1)

If anemia was accompanied by microcytosis or other signs of iron deficiency, then the PPV was 10% for females aged over 70 years.

Ramos et al 2007 reviewed the literature on diagnostic delay in CRC and concluded "that there is no association between diagnostic and therapeutic delay and survival in colorectal cancer patients".

Simunovic et al (2008) used SEER data (7989 cases) and found:

The risk of death was slightly greater if the consult-to-admission interval was 22 or more days versus 1-7 days (HR 1.1, 95% CI 1.0-1.2, p = 0.013) and if the test-to-admission interval was 43 days or more versus 1-14 days (HR 1.2, 95% CI 1.1-1.3, p = 0.003).

20140315: Xfce crash

Gave message about being unable to log out. When restarted, setup was corrupted - only one work space (unable to change), and loss of window decorations (margins etc) and most windows unable to be dragged. This is a known problem, and one solution is to delete the session data:

xfwm4 is not restored by the session manager. You can either start xfwm4 with Alt+f2 or clear the session cache (rm -r ~/.cache/sessions).

20140314: OO Fortran is faster than C++

Arabas et al (2013) Object-oriented implementations of the MPDATA advection equation solver in C++, Python and Fortran

The Fortran was twice as fast as the C++ Blitz based code.

20140313: UK, the progressive rock band

After King Crimson broke up in the late 70's, John Wetton (sings on Starless) went on to form UK, with Eddie Jobson (keyboards, electric violin), Allan Holdsworth, and Bill Bruford, then later Terry Bozzio. Jobson and Bozzio both had played with Frank Zappa.

20140313: More book recommendations

Jeff Vandermeer's Annihilation

Anybody read "No Return" by Zachary Jeningan? Many people compare this to the book of the new sun, Elizabeth Hand wrote : "It has the sweep of Frank Herbert's *Dune* and the intoxicatingly strange grandeur of Gene Wolfe's *Book of the New Sun*, with a decadent, beautifully rendered vision all its own."

James Nicoll thought:

Adam Rakunas' Windswept and the sequel Like a Boss, which is about a mid-level union official who keeps being drawn (very unwillingly: she would just like to own and operate a distillery) into planetary and galactic scale crises.

I thought Windswept entertaining.

20140228: Mutational spectrum of cancers

Alexandrov et al 2013 report that particular mutational signatures characterize different tumour types within TCGA. In the case of melanoma, C>T, predominates with TCG>TTG being the most triplet most likely to be affected (signatures 7, 10, 1B).

20140217: NextGen sequencing prices

Nicole Cloonan writes:

...there's a recently launched website that can help you find sequencing providers based on your specific needs. Prices are displayed for most providers instantaneously.


20140207: Simeon ten Holt

His Canto Ostinato for multiple (four or more) pianos is highly regarded.

A minimalist, he

generally used consonant, tonal materials and his works are organized in numerous cells, made up of a few measures each, which are repeated ad libitum according to the player's preference. Many of his works are for piano or ensembles of multiple pianos.

20140206: Consequences Argument(s)

L is deterministic law:
  a property of the universe

S is deterministic signalling rule using a Causal Decision Theory rule:
  if I believe L is true then I signal L is true
  S is determined by prior history of the universe

P is hypothesis that prior Hx + L entails S->T to be truth tracking

if emit S->F, then Lewis argues that this is a true statement
in that P is falsified or L actually is false

AND from Balaguer 2014, wrt determinism and free will, the Consequences Argument is roughly:

state of universe at time t0 P0, 
state of universe at time t2 P, when a given decision was made
N(X) is operator "X is true, and no one has ever had any choice about 
it being true"

Premise 1: N(P0 and L) ie prior Hx + L are determined
Premise 2: hypothesis that prior Hx + L entails P
P is determined.

20140203: Music of Iannis Xénakis

A modernist composer, music theorist, architect and engineer (1922-2001), worked for Corbusier, studied with Messiaen, producing musique concrète, and computer generated music, with numerous pieces based on mathematics, algorithms.

Psappha is reminiscent of gamelan - percussion piece composed using boolean operations on sequences or processes (called sieves by Xenakis).

Pleiades is another percussion piece (microtonal, quite nice).

20140201: Mt Nebo Walk

Circuit walk from Boombana section of NP:

circuit walk at Mt Nebo

Taped track down ridge to Cabbage Tree Creek where meet fire trail. Nissen et al (1999) thesis:

The permanent westem branch of Cabbage Tree Creek, and the intermittent eastem branch and Branch Creek, drain the northern part of the [Mt Crosby] area, feeding into Lake Manchester (formed in 1916) and from there into the Brisbane River above Mount Crosby.

The original road (the Butcher's Track) up the Brisbane Valley ran along Mt Crosby and Lake Manchester Rds to Cabbage Tree Run and then continued north of the river (actually 1-2 km north of World's End Pocket).

20140129: Beagley-Gibson scale for microtopographic assessment of photoaging

Beagley J, Gibson IM. Changes in skin condition in relation to degree of exposure to ultraviolet light. Perth: Western Australia Institute of Technology, School of Biology, 1980

Fritschi L, Green A. Sun damage in teenagers' skin. Aust J Public Health 1995;19: 383-6.

English DR, Armstrong BK, Kricker A. Reproducibility of reported measurements of sun exposure in a case-control study. Cancer Epidemiol Biomarkers Prev 1998;7: 857-63.

Green AC. Premature ageing of the skin in a Queensland population. Med J Aust 1991;155: 473-4, 477-8.

Holman CD, Evans PR, Lumsden GJ, Armstrong BK. The determinants of actinic skin damage: problems of confounding among environmental and constitutional variables. Am J Epidemiol 1984;120: 414-22.

dos Santos Silva et al 2009

Green et al 2011 Factors Associated with Premature Skin Aging (Photoaging) before the Age of 55: A Population-Based Study http://www.karger.com/Article/FullText/322623

Hughes M.C. Bredoux C. Salas F., Lombard D., Strutton G.M., Fourtanier A., Green A.C. Comparison of Histological Measures of Skin Photoaging

Hughes MC et al. Validation of skin surface microtopography as a measure of skin photoaging in a subtropical population aged 40 and over Photodermatology, Photoimmunology & Photomedicine 2012.

Maria Celia B. Hughes, MMedSci; Gail M. Williams, PhD; Peter Baker, PhD; and Adèle C. Green, MBBS, PhD Sunscreen and Prevention of Skin Aging: A Randomized Trial Ann Int Med 2013

Michelle R. Iannacone1, Maria Celia B. Hughes1, Adèle C. Gree Effects of sunscreen on skin cancer and photoaging Photodermatology, Photoimmunology & Photomedicine 2012.

20140116: MITF*E318K rs149617956

On CoreExome Array, "exm329674".

Frequency is fairly consistent world wide (a bit lower in Africa).

rs149617956 (MITF*E318K) in multiple studies
Pop Group NCarrierMAF
Mixed_GEMControl 2430530.0109
Mixed_GEMCMM 1194440.0184
AustraliaControl 1953140.0072
AustraliaCMM 2059340.0165
UK Control 2115180.0085
UK CMM 1929340.0176
Italy CMM 667120.0090
Italy Sporadic524110.010
Italy Familial14310.003
Italy Control 2205140.003
France Control 1659100.003
France CMM 603170.014
France RCC 16450.015
France CMM+RCC 6250.040
Poland Control 211440.0009
Poland CMM 74820.0013
Poland Breast 68310.0007
Poland Prostate75300.0000
Poland Colon 72910.0007
Poland Lung 73700.0000
Poland Kidney 57610.0009
EurAmESP Control 4300270.0031
AfrAmESP Control 220320.0005
CSAgilentControl 66210.0008
AFR Control 24600.0000
AMR Control 18110.0028
ASN Control 28600.0000
ASW Control 6100.0000
CEU Control 8500.0000
CHB Control 9700.0000
CHS Control 10000.0000
CLM Control 6010.0083
EUR Control 37910.0013
FIN Control 9300.0000
GBR Control 8910.0056
IBS Control 1400.0000
JPT Control 8900.0000
LWK Control 9700.0000
MXL Control 6600.0000
PUR Control 5500.0000
TSI Control 9800.0000
YRI Control 8800.0000

20140114: PLINK2

Very fast implementation from BGI (Cognitive Genomics):


The notes on algorithms are very useful - playing with Fortran 2008 popcnt().

20140109: SCSI card again

The installed card is an Adaptec AHA-2910C (99).

It does seem to be recognized by the system:

lspci -v 

08:01.0 SCSI storage controller: Adaptec AIC-7850 (rev 03)
	Subsystem: Adaptec AHA-2904/Integrated AIC-7850
	Flags: bus master, medium devsel, latency 64, IRQ 17
	I/O ports at e400 [disabled] [size=256]
	Memory at fbefe000 (32-bit, non-prefetchable) [size=4K]
	Capabilities: <access denied>
	Kernel driver in use: aic7xxx
	Kernel modules: aic7xxx

The documentation for the aic7xxx driver (aic7xxx.txt) says the AHA-2910C is supported, though giving slightly different details:

1. Supported Hardware

   The following Adaptec SCSI Chips and Host Adapters are supported by
   the aic7xxx driver.

 Controller Chip     Host-Bus Int-Connectors Ext-Connectors Notes
   AHA-2910C aic7860 PCI/32   SE-50M

   Chip    MIPS Host Bus  MaxSync MaxWidth SCBs Notes
   aic7860 10   PCI/32    20MHz   8Bit     3

Anyway the slide scanner worked with xsane, though a bit cranky about resolution, and I still need to find the feeder.

20140107: Primary aldosteronism

Biggest recent finding is that 40% of adrenal aldosterone-producing adenomas harbour somatic (or germline) mutations in KCNJ5 (11q24, 128.761313-128.791060 Mbp).

Scholl et al [2013], Azizan et al [2013] and Williams et al [2014] document similar patterns for CACNA1D (3p14, 53.529076-53.847179 Mbp), ATP1A1 (1p21,116.914802-116.948069 Mbp), and ATP2B3 (Xq28, 152.783905-152.848387 Mbp).

20140103: APOBEC3B

APOBEC3B encodes an APOB mRNA editing enzyme (causes C>T mutation) that is an antiviral component of the innate immune system. Overexpressed in cancers: cervix, bladder, lung, head and neck, breast. Action is as a cytidine deaminase that targets TCA or TCT triplets: C->U, U then converted to A by DNA polymerase, or U excised by uracil DNA glycosylase and A inserted.

APOBEC1 and APOBE3A are also likely to be activated. The cytidine deaminase signature is frequently seen in tumour somatic mutation patterns (signature 2 of Alexandrov et al 2013).

Poon et al (2013) describe the mutational signature of Aristolochic Acid

UV C->T untranscribed strand (NER)
UV CC->TT untranscribed strand (NER)
5-methyl cytosine deaminationC->T (CpG islands)
Tobacco carcinogensG->T
Tobacco carcinogensC->A (transcribed strand)
Tobacco carcinogensCC->AA (transcribed strand)
DNA polymerase errorsG->T
DNA polymerase errorsT->C
Aristolochic AcidA->T (A[C|T]AGG untranscribed strand)


20140101: Bunya Crossing

Pretty reserve with swimming hole on South Pine River - some riparian rainforest, including up around the Kumbartcho Sanctuary (formerly Bunya Park Sanctuary).

Couple more visits since this. A paddler told us that Samford Bowls Club (near bridge on Dayboro road) to Bunya Crossing when the river is high is very pleasant, though one should be wary of snags and the flotsam traps. Need 1 m at the Drapers Crossing gauge (142202A) about 1 km above Kumbartcho.

20131216: Back to MQLS and ROADTRIPS

R implementation and Sib-pair now agree perfectly for corrected chi-square

read.data <- function(pedfil="phenofile", kinfil="pedinfo", 
                      genofile="genofile") {
  x <- read.table(pedfil)
  idx <- paste(x[,1], x[,2])
  n <- nrow(x)
  kin <- matrix(0, nr=n, nc=n)
  kinvals <- read.table(kinfil)
  idx1 <- paste(kinvals[,1], kinvals[,2])
  idx2 <- paste(kinvals[,1], kinvals[,3])
  pos1 <- match(idx1, idx) 
  pos2 <- match(idx2, idx) 
  kin[n*(pos2-1)+pos1] <- kin[n*(pos1-1)+pos2] <- 2*kinvals[,4]
  kin <- kin + diag(1,n)
  genos <- read.table(genofile)
  list(ntot=n, ntyp=n, 
markervar <- function(dta, geno=1, filter="all") {
  if (filter == "all") {
    sel <- seq(1, dta$ntyp)
    sel <- filter
  ainv <- solve(dta$kin[sel,sel])
  ones <- matrix(1, nr=dta$ntyp, nc=1)[sel,]
  Y <-  as.matrix(dta$geno[1:dta$ntyp, geno], nc=1)[sel,]
  (t(Y) %*% ainv %*% Y - 
   solve(t(ones) %*% ainv %*% ones) %*% (t(ones) %*% ainv %*% Y)^2) / (dta$ntyp-1)

assoc.stat <- function(dta, geno=1, typ="correctedX2") {
  sel <- dta$pheno %in% c(1,2)
  n <- sum(sel)
  onec <- as.matrix(dta$pheno[sel]-1, nc=1)
  ones <- matrix(1, nr=n, nc=1)
  nc <- sum(onec)
  if (typ == "correctedX2") {
    V <- onec - nc/n * ones
  Y <-  as.matrix(dta$geno[sel, geno], nc=1)
  psi <- dta$kin[sel,sel]
  s2 <- markervar(dta, geno=geno, sel) 
  (t(V) %*% Y)^2/(s2 %*% t(V) %*% psi %*% V)

20131213: "GWAS to Therapy by Genome Edits?"

Hardison and Blobel [ Science 2013; 342:206] discuss the paper of Bauer et al in that issue. GWAS for persistent fetal Hb found two loci in 2007-8. The peak SNPs in intron 2 of BCL11A were shown to lie in an transcriptional enhancer (CTG(n9)GATA). Although the effect sizes for the SNPs were small (1.7-fold expression change), deletion of BCL11A in erythroid lineages is an effective intervention in murine hemoglobinopathy models. Disruption of the enhancer element greatly reduced expression in erythroid cells but not B-lymphocytes.

[T]he modest phenotypic effects of the common variants mapped in GWAS need not be interpreted as indicative as indicators of modest effects of the regulatory region or locus...

And Bauer et al

We propose the GWAS-identified enhancer of BCL11A as a particularly promising therapeutic target for genome engineering...sparing nonerythroid lineages.

20131211: The Saragossa Manuscript

Jan Potocki Manuscrit trouvé à Saragosse (1804; usually the The Saragossa manuscript in English) is a collection of stories modelled on the 1001 Nights set within the framing tale of a French army officer. Influenced Robert Irwin's The Arabian Nightmare (he discusses it in his book on the Arabian Nights), and made into a cult surrealist film in 1965.

20131209 and 20160913: webcomics revisited


2D Goggles (Babbage & Lovelace)
9 Chickweed Lane
Abstruse Goose
A Girl and Her Fed
Basic Instructions
Bob the Angry Flower
Breaking Cat News
Bug Martini
But I'm a Cat Person
Camp Weedonwantcha
Cat And Girl
Dark Legacy Comics
Darths and Droids
diesel sweeties
Doc Rate
Dork Tower
Dresden Codak
Dumbing of Age
Evil Inc.
Exterminatus Now!
Full Frontal Nerdity
General Protection Fault
Girl Genius
girls with slingshots
(and I've pulled the Goats site back out for when it starts up again)
GU Comics
Guilded Age
Gunnerkrigg Court
Help Desk Comic (Ubersoft.Net)
Homestuck (MS Paint Comics)
In His Likeness
Irregular Webcomic! (now in Sundays-only new explain-o-post material mode)
Kevin and Kell
Last Kiss Comics
Least I Could Do
Leftover Soup
Looking for Group
Manly Guys Doing Manly Things
Nodwick (temporarily halted)
Order of the Stick
Partially Clips
Penny Arcade
Piled Higher and Deeper
Punch an' Pie
Quantum Vibe
Questionable Content
Real Life Comics
Red Meat
Rich's comics blog (a variety of things, it's just started up a 50th-anniversary Doctor Who strip)
Rusty & Company
Saturday Morning Breakfast Cereal
Scandinavia and the World
Scary Go Round
Scenes from a Multiverse 
Schlock Mercenary
Skin Horse
Slowpoke Comics/Jen Sorensen
Sluggy Freelance
Something Positive
Supernormal Step
The Abominable Charles Christopher
The Daily Blink
The Dreamland Chronicles
The Joy of Tech
the noob
The System
the three other Aaron comics -
The Whiteboard
Three Panel Soul
This Modern World/Tom Tomorrow
Tom The Dancing Bug
Two Lumps
Use Sword on Monster
What-If xkcd (the spinoff site)
Yet Another Fantasy Gamer Comic

And see:

Essay: sauron, the great prince of middle-earth


20131207: Book recommendations

Peter Cannon suggests:

Martin Gardner's autobiographical novel The Flight of Peter Fromm

and Michael Coffey:

J.F. Powers's 1963 novel Morte D'Urban...a supremely funny tale about Father Urban, a Midwestern priest who believes in golf, baseball, scotch, and the Roman Catholic Church, perhaps in that order. Fr. Urban's struggles with the management of a parish are rendered with a comic's timing, and Powers's prose and dialogue are as sharp as anything in Flannery O'Connor. Is anyone as funny today? Maybe David Gates; maybe Tom Drury.

20140129 Lawrence Watt-Evans:

Whereas my list would be mostly from the 1960s -- The Sentinel Stars by Louis Charbonneau, The Whole Man by John Brunner, etc.


And speaking of "pulpy", George O. Smith's Highways in Hiding (1956, abridged as The Space Plague , 1957) counts as one of my "obscure favorites". It wasn't as good as The Fourth "R" (1959, aka _The Brain Machine_), by far his best book, but it had a certain pulpy charm.

Robert Carnegie: Kenneth Bulmer To Outrun Doomsday

LWE: Phyllis Gotlieb Sunburst

Brian Scott: T.L. Sherred Alien Island

Several people remembered Colin Kapp. Two I haven't read:

Colin Kapp The Wizard Of Anharitte

Colin Kapp The Ion War is available online.

20131205: The other David Duffy

David Cameron Duffy
Professor at University of Hawaii, Ph.D. Princeton
Professor of Botany, Ecology, Evolution And Conservation Biology


20131122: TFAP2A (transcription factor AP-2 alpha

Location 6p24.3

From http://www.ncbi.nlm.nih.gov/gene/7020

The protein encoded by this gene is a transcription factor that binds the consensus sequence 5'-GCCNNNGGC-3' [...and] functions as either a homodimer or as a heterodimer with similar family members...Defects in this gene are a cause of branchiooculofacial syndrome (BOFS).
Stenou et al [2013]:

microphthalmia-associated transcription factor, SOX10, and AP2-alpha, which are master actors of melanoma development,

In branchiooculofacial syndrome adults sufferers, Premature graying of hair occurs. One case report of an infant with BOF syndrome together with preaxial polydactyly and a white forelock at birth [OMIM 113620].

20131113: Animal welfare

Clare Palmer [2013] gives:

...[A]ct consequentialism [holds] an act is morally right if and only if the total amount of good'in this case pleasure'minus the total amount of pain, is greater than the net amount for any incompatible act available to the agent on that occasion (Sinnott-Armstrong, 2011).

Tyler Cowen [ Policing Nature 2001] writes:

Scruton (1998) treats the policing nature argument as a reductio ad absurdum on vegetarianism and animal right. The most serious treatment, by far, is found in Sapontzis (1987, chapter 13).

Holmes Rolston III (1988). Environmental Ethics: Duties and Values in the Natural World. Regarded as a key thinker in this area.

Scruton, Roger. On Hunting. London: Yellow Jersey Press, 1998.

Sapontzis, S.F (1987). Morals, Reason, and Animals. Philadelphia: Temple University Press,

20131103: renal disease genes

Marker ChrPos (bp) Annotation
rs1061170 1196659237CFH*Y402H (?strep disease)
rs4664308 2160917497PLA2R1 idiopathic membranous nephropathy (European)
rs13177732 5176856925(European)
rs2187668 6 32605884HLA-DQA1 idiopathic membranous nephropathy (European)
rs1801239 10 16919052CUBN*I2984V (European)
rs60910145 22 36662034APOL1 (African)
rs73885319 22 36661906APOL1 (African)
rs71785313 22 36662046APOL1 (African)

Cubilin: note that cubilin expression is monoallelic [Aseem et al 2013], and is essential for albumin reabsorption in the renal proximal tubule. [Amsellem et al 2010]. A European haplotype comprising three missense mutations (p.Leu2153Phe, p.Ile2984Val, p.Glu3002Gly) is a risk factor for albuminuria [Boger et al 2011], "associated with a 41% increased risk for the development of persistent microalbuminuria during 20 years of follow-up among 1304 participants with type 1 diabetes in the prospective DCCT/EDIC Study". Other cubulin mutations cause hereditary megaloblastic anemia secondary to vitamin B12 malabsorbtion. The common coding variant rs1801239 (I2984V) has a frequency of ~0.15 across much of Africa and Europe, and in PNG, though low or absent in the Americas and many Asian groups.

20131029: APOL1 a major cause of African (American) renal disease

Pollak et al [2012], Ko et al [2013]:

People of recent African ancestry develop kidney disease at rates 4-5 times higher than most other groups (lifetime risk ESRD AA 7.5% v. EA 2.1%): hypertensive kidney disease, focal segmental glomerulosclerosis (FSGS), and HIV-associated nephropathy (HIVAN). Most of this risk is conferred by two coding sequence variants in the serum-resistance-associated (SRA)-interacting-domain-encoding region of the APOL1 gene (22q12.3) that act in a recessive manner. Frequency of these alleles is high, and thought to arise from selection for trypanosomiasis resistance. There is no association with type 2 diabetic ESRD, but these alleles do strongly predict lupus nephritis (25% of cases and 12% of controls are homozygous, PAR=26%) in African Americans [ Freedman et al [2012]].

G1 0.21 rs60910145, rs73885319
G2 0.13 rs71785313 (indel)

Genovese 2010Focal segmental glomerulosclerosis0.660.125
Genovese 2010Hypertensive end-stage renal disease0.4650.12
Kopp 2011Focal segmental glomerulosclerosis0.720.13
Kopp 2011HIV-associated glomerulopathy0.720.08
Kanji 2011Chronic haemodialysis 0.37
Friedman 2011 0.13

20131029: Gerard O'Brien (University of Adelaide)

Process theories currently dominate the theoretical landscape. According to the most popular of these, informational contents enter consciousness when they gain competitive access to a global workspace, permitting them to be broadcast throughout the brain (Baars 1997]: Global Workspace Theory. On the alternative higher-order thought (HOT) theory, the content of a representing vehicle becomes conscious when it is the target of a second-order representation (Rosenthal 1997). The common thread here is that, of all the representing vehicles present in the brain, only those caught up in specific kinds of computational processes - such as global broadcasting or internal monitoring - contribute to conscious experience.

20131028: Andromeda at Ngun-Ngun

Is at bottom of slabs below main cliffs, 300 m W traverse of furthest W switchback of tourist track. Did five climbs (13-18, all 17-20 m). Sun hit by 11, but trees shaded in afternoon.

20131026: Jan Swammerdam and others

The Book of Nature in Early Modern and Modern History edited by Klaas van Berkel, Arie Johan Vanderjagt.

Voetius ("Pope of Utrecht") 1589-1676.

Matthew Cobb comments:

"All animals hatch from eggs that are laid by a female of the same species". Swammerdam thus showed that all animals - including insects - bred true. The "fixity" of species was a prerequisite for the subsequent recognition of evolution.

20131024: math rock

Math rock is post-rock post-progrock, so instrumental - heavily guitar driven, odd time signatures, rhythmic with fast repeated riffs, jazzy (but "no solos"). Very reminiscent of late Crimson and Sufyan Stevens. Both Clever Girl and TTNG use brass/sax.

Bands: Piglet, Hella, This Town Needs Guns, Tera Melos, Palvo, Rooftops, Giraffes? Giraffes!, Clever Girl.

From http://www.progarchives.com/

Although there are Math Rock bands in different countries around the world, most reside in the United States, the Midwest in particular, and tend to be divided by regions: Pittsburgh bands (Don Caballero, Six Horse) Chicago bands (Shellac, U.S. Maple), Ohio bands (Keelhaul, Craw) Louisville bands (June 44, Rodan, The For Carnation, Crain), and San Diego bands (Drive Like Jehu, Tristeza) among others on both coasts. Japan was also an important country in the Math Rock genre with bands like Ruins and Zeni Geva.

Battles, Storm and Stress (Ian Williams from Don Cab)

Unconscious Collective

A brief guide to Japanese Math Rock mentions Ruins, Toe, Tricot (almost poppy, TTNG like, toured with).

20131016: John Hollander

Rhyme's Reason: A Guide to English Verse. Eg:

Here is a later form of the roundel that Algernon Charles Swinburne wrote so many of:

The roundel ends as it begins; we take
The first words from the first line, where it bends
Easily, and with the refrain we make
The roundel ends.

- But not just yet: its rhyming still extends
Through six new lines before they come awake,
Again, those last few words, those sleeping friends

We started out with and will not foresake;
What though the weary journey's way one wends,
When it is finally time to take a break
The roundel ends.

20131015: mTOR

MTOR 1p36.2 11166588-1132261 bp.

Mechanistic Target Of Rapamycin serine/threonine kinase (mTOR) and AMP-activated protein kinase (AMPK) cross-regulate cellular metabolism and autophagy. Activated AMPK stimulates glycolytic flux, increases glucose uptake, induces fatty acid oxidation, and antagonizes mTOR kinase. AMPK stimulates autophagic breakdown of macromolecular complexes in the cell, producing energy and nutrients, while mTOR suppresses autophagy to promote cell growth and proliferation [ Brunton et al 2013]. mTOR is the key downstream target for AKT etc leading to cell cycle arrest, immunosuppression, on sIL6R production in immune cells. The effects of rapamycin on laying down of memory are also known to be via mTOR.

mTOR acts as part of two complexes mTORC1 and mTORC2 with different substrates.

Harrison et al [2009] demonstrated that rapamycin fed late in life extended lifespan in mice, a finding replicated in several other studies. Neff et al [2013] further confirmed this, but claimed many effects were age independent (eg improved learning), and specific effects on aging per se of particular organ systems were few: metabolic rate, T cell subclass counts, all cancers. Zhang et al [2013] have shown i.p. rapamycin to preserve the follicle pool reserve and prolong the ovarian lifespan of rats. Rapamycin has renal, gonadal and hepatic toxicities.

Wu et al [2013] found that mice with two hypomorphic (mTOR d/d) alleles are viable, expressing mTOR at approximately 25% of wild-type levels (with reduced mTORC1 and mTORC2 activity). They exhibit an approximately 20% increase in median survival. While smaller than wild-type mice, these animals had no alterations in normalized food intake, glucose homeostasis, or metabolic rate, but had reduced aging tissue biomarkers. Many but not all organ systems preserved function with age.

Rapamycin-derived drugs sirolimus and everolimus are mTOR inhibitors used against multiple cancers and tuberous sclerosis (normally, TSC protein modulates mTOR). Metformin also inhibits mTOR. One pharmacogenomic GWAS published.

Kimura et al [2011] report on sirolimus-eluting stents for cardiac revascularization (5092 individuals compared to 5405 with bare metal stents): reduced all-causes death hazard ratio HR=0.72 (95%CI 0.59-0.87), mainly driven by a reduction in non-cardiac death HR=0.64 (0.48-0.85).

Several papers on effects of metformin and phenformin, eg Dilman & Anisimov [1980]:

The chronic treatment of female C3H/Sn mice with phenformin (2 mg/day) and diphenylhydantoin (2 mg/day) prolonged mean life span by 23 and 25%, respectively, and decreased spontaneous tumour incidence by 4.0 and 2.3 times, respectively.

And in the WHI, Gong et al [2015]:

Diabetes was associated with higher risk of total invasive cancer (HR, 1.13; p<0.001), and of several site-specific cancers (HR, 1.2-1.4, and up to over 2-fold). Diabetes was also associated with higher risk of death from cancer (HR, 1.46; p<0.001). There was no overall difference in cancer incidence by diabetes therapy (p=0.66). However, there was a lower risk of death from cancer for metformin users, compared to users of other medications, relative to women without diabetes, overall (HRs, 1.08 versus 1.45; p=0.007) and for breast cancer (HRs, 0.50 versus 1.29; p=0.05).

20131008: Magpie map

Quest newspapers annually collate a map for cyclists.

20131002: Checking hard disk problems

The palimpsest (GNOME) and smartctl (commandline) utilities allow one to check the SMART logs etc for the drives. Currently that drive passes the short and long self-tests

SMART Self-test log structure revision number 1
Num  Test_Description    Status                  Remaining  LifeTime(hours)  LBA_of_first_error
# 1  Short offline       Completed without error       00%     13069         -
# 2  Short offline       Completed without error       00%     13068         -

And (VALUE is a normalized value of RAW_VALUE),

SMART Attributes Data Structure revision number: 10
Vendor Specific SMART Attributes with Thresholds:
  1 Raw_Read_Error_Rate     0x000f   120   099   006    Pre-fail  Always       -       1818159
  3 Spin_Up_Time            0x0003   095   094   000    Pre-fail  Always       -       0

20130927: MC1R in orangutans

Haitini et al [2007] comment there are fixed differences at 13 residues of MC1R between human and orangutan:

A27V, R34W, S41P, V57A, S90G, V107A, L113V, V165I, V174L, V186A, E269Q, G274S, H301R

However, MSH binding and cAMP production don't differ greatly between these species.

If want a good candidate SNP, may need "ancestral" human allele to be associated with red hair. So rs910873 flanking sequence matches well in orangutan, but they carry the G allele.

20130926: On-line journal on gaming etc


20130806: UV induced immunosuppression in the mouse

Prue Hart (in seminar today) mentioned that this abrogates OVA asthma model, as well as DTH.

PGE2 mediated, via the prostaglandin E receptor subtype 4 (EP4), leading to DNA methylation mediated changes in marrow (DCs, but also Treg). Effects are systemic and prolonged (28 days or more following single dose). Blockable by indomethacin [ Ng et al 2010, Soontrapa et al 2011].

Apparently, phototherapy has been used for atopic dermatitis.

It is known that green tea polyphenols [Katiyar et al 2010] and antioxidants reduce UV-induced immunosuppression, presumably via reduction of DNA damage and enhancement of nucleotide excision repair.

20130729: Prospection and free will

Martin Seligman et al discuss the concept of prospection: the interior experience of future-oriented cognition.

From Flickers

...[P]rospection models the event of one's choice in a given episode of deliberation as exogenous, and thus as carved off from its antecedent causes and allowed to vary freely across a range of values. If an agent experiences deliberation in this way, then she likely experiences her choice as not having antecedent sufficient causes. And a choice that is experienced as not having antecedent sufficient causes is, a fortiori, experienced as not having antecedent deterministic causes.

20130720: Erasmus Rucphen Family Study

The Erasmus Rucphen Family Study (Pardo, MacKay et al. 2005) (ERF; the Netherlands) is a family-based study that investigates the genetic or igins of complex diseases in an isolated community in the south west of the Netherlands. The study population essentially consists of one extended family of descendants from 20 related couples who lived in the isolate between 1850 and 1900.

20130629: h-index

From here

At time of promotion to Professor in the JHU School of Medicine, mean (median) number of papers is 68 (61), first/last authored is 32 (30), total citations 2974 (1710), first/last author citations 1431 (826), h-index 25 (23).

For election to the US NAS in 2005, median biomedical and biology h-index was 57 (from Hirsch's original paper 2006).

20130629: NFIB is a governor of epithelial-melanocyte behaviour in a shared niche

Nuclear Factor I/B (NFIB) is a transcription factor that coordinates melanotype proliferation and diffentiation in the hair follicle during the hair cycle. NFIB is produced by hair follicle stem cells. The target motif TTGGCANNNTGCCAA was present in the mouse EDN2 promoter, and acts to increase responsiveness of the second population of stem cells in the follicle, the melanocyte stem cells, to KITL. The hair follicle stem cells are activated at anagen by Wnt and BMP/TGFB.

Human EDN2 lies at chr1:41944446-41950344

20130626: Wide Range Assessment of Memory and Learning (WRAML)

Does not discriminate ADHD as well as CBCL etc [Dewey et al 2001], but ADHD associated with lower verbal memory index scores.

20130419: Melanin index

The M index is defined as 100 log10(1/proportion red reflectance at 655nm) [Shriver 2000]. Beleza et al [2013] further square root transform this (MM).

Other approach is slope of absorption curve from 620-720 nm as per Kollias and Baqer [1985].

Dwyer et al [1998] is a third approach, using skin reflectance 400-420 nm.

MD400 = 100(0.035307 + 0.009974(R420 - R400))

20130417: Condamine River Road

Goes up and back down quite quickly.

Gauge at BROSNANS BARN (422341A Condamine River at Brosnans Barn) can be read at http://watermonitoring.derm.qld.gov.au/host.htm.

Brosnans barn on the DERM Water monitoring site which I have been watching all week was down to .612m and is currently .597 @ 1600hrs today. I wanted to see for myself so I headed down there today and as I was entering the Head, a Subaru and a Pathfinder had just come through the Gorge with no issues. I drove all 14 crossing in and then back out & no crossing was deeper than the bottom of my side steps (about 380mm), with the average crossing only around 250mm.

20130410: Red Rock Gorge, Red Rock Falls, Sundown homestead

Started 08:25, Falls at 13:15, back at car 16:15

Lots of nice rockholes in lower gorge, with campsites galore. Last km up to falls involved scrambling up big boulders etc in creek. Turned out J had previously been there with K2 Extreme. Straightforward ascent to top of falls up pad to north, but then missed any track up to road and campsite, scrub bashing up from creek to road on ridge top. G. saw interesting fungi in scrub.

20130409: Red Rock Gorge, Waterhole ("rest day")

Moved camp 600 m into nice spot just inside Red Rock Gorge. In pm, strolled down to Red Rock Waterhole (a few pigs) for swim. Couldn't find any obvious track ascending from there (as implied by NPWS map).

20130408: Nundubbermere Falls - Severn River

After investigating downstream, where no easy passage on our side (too hard to cross with highish water), followed creek South up to a saddle. Then along ridge above river to C. Creek, where little grassy plateau with water (3 deer). Down to river, where very slow going on slippery rocks (showered for an hour).

20130407: Sundown homestead - Nundubbermere Falls

Started ~11:30. Up to Mt Lofty, then along track following boundary fence North. Above river, it turned West, terminating at old yards. We couldn't cross to the campsite on the Northern bank, so camped up on a knoll above the falls.

20130309: Analysis of case-control data utilizing known prevalence

Usual approach is:

intercept = log( q0/(1-q0) )

where q0 = P(Y=1) ie marginal probability of caseness.

RR = OR/[(1-r0) + r0*OR]

where r0 = P(Y=1|unexposed).

20130307: Krugman quotes Keynes

"Words ought to be a little wild", said John Maynard Keynes, "for they are the assault of thoughts on the unthinking."

20130302: AUSS2 and human-neanderthal sweep

Known to be contain one of the most rapidly evolving regions Human Accelerated Region 31 (i4). The human-Neanderthal differences span:

exon 1-4; chr7:68,662,946-69,274,862 (hg18)

involved in the autism spectrum disorder deletion (Pinto et al 2010).

20130301: Holographic Principle

Ng [2010]:

The Wigner-Salecker experiment (clock-mirror) suggests that quantum fluctuations in distance l, dl ~ l1/3 lP2/3 (where lP is the Planck length 10-35 m). One empirical test is smearing of images of very distant quasars [Perlman et al 2012], but the VLTI would be needed test that alpha=2/3.

According to the holographic principle, the number of degrees of freedom in a region of space is bounded not by the volume but by the surrounding surface. This suggests that the physical degrees of freedom are not independent but, considered at the Planck scale, they must be infinitely correlated, with the result that the spacetime location of an event may lose its invariant significance. If we take the point of view that holography has its origin in spacetime foam (as we have argued above), then we can argue that space-time foam gives rise to nonlocality.

Heuristically, the partitioning of a spatial region l x l x l, into small cubes can go no smaller than dl, so that the degrees of freedom ~ (l/lP), that is proportional to the surface area rather than the volume of the space.

Bekenstein, Hawking,'t Hooft, Smolin, Verlinde

Extension of calculations for a black hole event horizon to an (arbitrary) holographic screen dividing two spatial regions.

Known not to hold in some cases.

Allows one to derive Newtonian gravity as an entropic force. Can also produce MOND and GR and QLG. The expression for entropy is modified by the use of infinite statistics: statistics of particles with an infinite number of internal degrees of freedom,so losing the Gibbs 1/N! term.

Can also give rise to warm inflation, and is consistent with measured cosmological constant. Can also include dark matter.

E. P. Verlinde, On the Origin of Gravity and the Laws of Newton, arXiv:1001.0785 [hep-th]

T. Padmanabhan, Equipartition of energy in the hori- zon degrees of freedom and the emergence of gravity, arXiv:0912.3165 [gr-qc]

L. Smolin, Newtonian gravity in loop quantum gravity, arXiv:1001.3668 [gr-qc]

20121017: Fortran stuff

Rafik Zurob writes:

If you're using a recent enough version of gfortran, use -M -cpp (and
possibly -MF) to generate a dependency file that can be used with make.  You
can find the details in the GCC option documentation.

Paul van Delst thought:

I use makedepf90, but others exist. I like makedepf90 because it's small
and portable (written in C! :o)

Deepak Eachempati mentioned:

I just wanted to notify those who are interested in a free compiler supporting Fortran coarrays that this is available in the OpenUH compiler. OpenUH is developed by the HPCTools group at the Univ ersity of Houston. It is not a full Fortran 2008 implementation. It is based on the Open64 compiler, which support Fortran 95 and some Fortran 2003 features. We have implemented the coarray features that were added to Fortran 2008.


FX coudert@alussinan.org writes:

-Bstatic is a linker option, while -static is also understood by the compiler (the compiler's driver, whose job is then to call the linker).

You may also want to explore the combination "-static-libgfortran -static-libgcc", which should alleviate any linking to GCC support libraries, yet dynamically link to OS libraries (like C library).

Tim Prince writes (20150921):

-march=native will be more important now that you are testing on an AVX platform and giving other compilers the benefit of it, if you can upgrade to a more current gfortran.

-funroll-loops unrolls too aggressively for recent Intel platforms. It should be accompanied by --param max-unroll-times=2 (or 4, as has been set for PGI). On these platforms, ifort usually improves with -Qunroll4.

gfortran -ffast-math -funroll-loops --param max-unroll-times=2 -Ofast -march=native steve kargl thought:

My personal choice of options are '-O2 -march=native -mtune=native -funroll-loops -ftree-vectorize'. After seeing Tim's post, I need to re-evaluate the use of -funroll-loops.

which he updated 2016-01-28:

.ifdef DEBUG
FFLAGS = -g -pipe -O -fmax-errors=1 -Werror -Wall -Wno-conversion -fcheck=all
FFLAGS+= -ffpe-trap=invalid -Wno-maybe-uninitialized
FFLAGS = -O2 -pipe -march=native -mtune=native
FFLAGS += -funroll-loops --param max-unroll-times=4
FFLAGS += -ftree-vectorize -Wall -Wno-maybe-uninitialized

edmondo.giovannozzi (2017-09-06) recommends the gfortran ASAN library:

gfortran -g -fsanitize=address -static-libasan ...

The compiled program will stop if and when encounters a memory error, and can be launched within gdb.

Michael Larabel

While born originally at Google as projects for LLVM, AddressSanitizer and ThreadSanitizer have been ported to GCC and will be part of the forthcoming GCC 4.8 compiler release. Back at Google, they're onto developing MemorySanitizer for LLVM.

AddressSanitizer (ASAN) was developed at Google to be a fast memory error detector. ASAN is capable of finding use-after-free and heap, stack, and global buffer overflow bugs within C and C++ programs. AddressSanitizer was merged into LLVM 3.1 earlier this year and running an ASAN-enabled program leads to only about a 2x slowdown for developers wishing to easily spot these memory errors within their programs. The GCC port of ASAN is currently supported on IA-32, x86-64, and x32 architectures under Linux.

ThreadSanitizer (TSAN) is another newer Google project for detecting data race within C/C++ programs. ThreadSanitizier is based upon the popular Valgrind program and is similar to Helgrind. The slowdown of running a ThreadSanitizier-enabled LLVM program is around 5~15x slower and also requires a run-time library aside from a supported compiler. Within LLVM, ThreadSanitizer is much more primitive and in an alpha state compared to ASAN.

With GCC 4.8, both AddressSanitizer and ThreadSanitizer are available for detecting memory errors and data races, respectively. For tapping AddressSanitizer in GCC, the -fsanitize=address switch must be passed to GCC when building a program. The -fno-omit-frame-pointer switch can also be optionally passed for nicer stack-traces. If you wish to utilize ThreadSanitizer in GCC 4.8, the -fsanitize=thread switch must be thrown, but this port is only working on 64-bit Linux.

20121012: Robert Aickman The Model

Wolfe says about it: (via Urth, jbarach@aol.com)

The late Robert Aickman was up to something, and quite consciously. He did not title his little book The Model for nothing. What he has done (and knew he had done) is discover a new wonderland. I would bet that the number of people who have done that is less than one tenth the number who have discovered actual islands, inhabited or treasure; we must now add Aickman to that short list. He has told us what I think none of us had realized before -- that czarist Russia can be just as viable a wonderland as, say, Baghdad under the Caliph. And he has not merely told us; he has proved it by leaving us a model.

20120927: Mario Bunge

Philosopher of science and physicist (to 1966, with an interest in quantum realism):

scientific realism or hylorealism, "systemism", emergentism

His ontological categories: thing, property, state, event, process, causation; change (instantaneous event or drawn-out process), fact (a thing with a given property, a thing in a given state, a thing undergoing a given change); space, time; system (things plus relational properties).

So, things are not "bundles of properties", "sets of processes", "snapshots of processes": phonons are processes rather than things, even though it is convenient to treat them as such - no "free phonons".

Properties are intrinsic or relational, changeable or invariant.

I am a materialist but not a physicalist because, as a physicist, I learned that physics can explain neither life nor mind nor society. Physics cannot even explain phenomena (appearances), because these occur in brains...; nor can it fully explain machines, as these embody ideas, such as those of value, goal and safety...

...a mechanism [is] a process (or sequence of states, or pathway) in a concrete system, natural or social. ...[M]ost mechanisms are concealed, so that they have got to be conjectured. This suggests the plan ... : system, mechanism, mechanism guessing, and explanation.

The coexistence of parallel mechanisms is particularly noticeable in biosystems and social systems. Think, for example, of the various mechanisms that operate in a scientific community, such as original research and criticism (the truth-and-falsity-finding mechanisms); peer-review (the social quality-control mechanism); and a combination of cooperation in the search for truth and the detection of falsity, with competition in the allocation of credits, jobs, and resources.

...to explain is to exhibit or assume a (lawful) mechanism. This is the process - whether causal, random, or mixed - that makes the system work the way it does.

The days of phenomenalism and descriptivism are over in science... No law, no possible mechanism; and no mechanism, no explanation. No wonder then that the hallmark of modern science is the search for lawful mechanisms behind the observed facts...

Strong methodological naturalism is identical with scientism, or the thesis that the scientific method is applicable in all research fields, the social sciences among them. This version of naturalism, often called scientistic, has been championed by the positivists, logical positivists, and Quine among many others. By contrast, Putnam, Davidson, Rorty and others have defended what may be called nonscientistic naturalism, which is little more than a perfunctionary nod to ontological naturalism (see De Caro and Macarthur 2004). Ultrastrong methodological naturalism is the program of reducing the social sciences to the natural ones...

20120925: tongue twister

They that thee thoroughly thou, thoust thou thoroughly too.

20120924: Distance correlation

Szekely, Rizzo and Bakirov [2007] developed a variety of new measures of multivariate dependence that are better than the approach described in Science 2011.

The distance covariance V is a measure of the distance between the characteristic function of X,Y fX,Y and the product of the marginal characteristic functions fX fY

Simplified R code from energy library.

    x <- as.matrix(dist(x))
    y <- as.matrix(dist(y))
    n <- nrow(x)
    m <- nrow(y)
    stat <- 0
    dims <- c(n, ncol(x), ncol(y))
    Akl <- function(x) {
        d <- x
        m <- rowMeans(d)
        M <- mean(d)
        a <- sweep(d, 1, m)
        b <- sweep(a, 2, m)
        return(b + M)
    A <- Akl(x)
    B <- Akl(y)
    dCov <- sqrt(mean(A * B))
    dVarX <- sqrt(mean(A * A))
    dVarY <- sqrt(mean(B * B))
    V <- sqrt(dVarX * dVarY)

20120920: from mp4 to Ogg Theora

ffmpeg -i  file.mp4 [-vn] file.ogg

And to Vorbis:

ffmpeg -i example.flv  -acodec libvorbis -vn example.ogg


avconv -i input.mp4 -c:a libvorbis -vn output.ogg

20120910: List of most influential 20th century philosophical articles

Adams Robert phenomenalism and corporeal substance in leibniz
Anderson Elizabeth What is the Point of Equality (1999)
Andreasens O. Robin A New Perspective on the Race Debate (1998)
Annas Platos Republic and Feminism
Anscombe GEM Modern Moral Philosophy(1958)
Anscombe War and Murder
Appiahs Anthony The Uncompleted Argument: Du Bois and the Illusion of Race (1985)
Arrow Kenneth A Difficulty in the Concept of Social WelfareTHE JOURNAL OF POLITICAL ECONOMY Vol. 58 No. 4. (Aug. 1950) pp. 328-346
Benjamin Walter L'oeuvre d'art à l'époque de sa reproduction mécanisée Zeitschrift für Soziaforschung (1936)
Berlins Isaiah Two Concepts of Liberty
Block On a confusion about a function of consciousness
Blumenberg Hans Paradigmen zu einer Metaphorologie Archiv für Begriffsgeschichte (1960)
Boxills Bernard The Morality of Reparations (1972)
Brouwer On the significance of the Principle of Excluded Middle in Mathematics Especially in Function Theory
Burge Individualism and the Mental
Burnyeat Idealism and Greek Philosophy
Carnap Die Überwindung der Metaphysik durch logische Analyse der Sprache
Carnap Empiricism Semantics and Ontology;
Cassirer Some Remarks on the Question of the Originality of the Renaissance
Cavell Stanley The Availability of Wittgensteins Later Philosophy (1962)
Chalmers Facing up to the Problem of Consciousness
Chalmers: The Conscious Mind
Chisholm The Problem of the Speckled Hen
Chomsky A Review of B. F. Skinners Verbal Behavior
Churchland: Eliminative Materialism and the Propositional Attitudes
Code Lorraine Is the Sex of the Knower Epistemically Significant?
Cohen GA The Structure of Proletarian Unfreedom
Cohen AG On the Currency of Egalitarian Justice
Darwall Gibbard and Railton Towards Fin de Siecle Ethics; Some Trends.
David Lewis Adverbs of Quantification
David Lewis Dispositional Theories of Value
Davidson Donald Actions Reasons and Causes
Davidson Donald On the Very Idea of a Conceptual Scheme
Davidson Mental Events
Davidson Truth and Meaning
Dennett Dan Intentional Systems
Derek Parfits 1971 Personal Identity
Derrida Jacques La Différance Bulletin de la société française de philosophie (1968)
Dodds Plato and the Irrational
Dodds The Parmenides of Plato and the origin of the Neoplatonic One
Donnellan Keith Reference and Definite Descriptions
Dretske Epistemic Operators
Dummett Truth
Dworkin Ronald What is Equality? (Parts 1 and 2)
Dworkins Ronald The Model of Rules (1967).
Field articles as collected in *Realism Mathematics and Modality*.
Fodor & Pylyshyn - Connectionism and Cognitive Architecture: A Critical Analysis(1988)
Fodor Jerry Special Sciences
Fodor Propositional Attitudes
Foot articles as collected in *Virtues and Vices*
Foot Philippa Utilitarianism and the virtues
Foot Phillippa Morality as a System of Hypothetical Imperatives
Foucault Nietzsche Genealogy History
Frankena The Naturalistic Fallacy
Frankfurt Freedom of the Will and the Concept of a Person
Frankfurt Harry Equality as a Moral Ideal
Frankfurt Harry Alternate Possibilities and Moral Responsibility.
Frege Gottlob (1892) On Sense and Reference
Frege The Thought
Garcias J.L.A. The Heart of Racism (1996)
Gödel On Formally Undecidable Propositions of Principia Mathematica and Related Systems I
Geach Ascriptivism
Geach Assertion
Geach Good and Evil
Gettier Edmund Is Justified True Belief Knowledge?(1963)
Ghiselin Michael A Radical Solution to the Species Problem
Goldman Alvin A Causal Theory of Knowing
Goldman What is Justified Belief
Grice H.P. Meaning(1957)
Grice H.P. Logic and Conversation(1976)
Hare The Promising Game
Harman Inference to the Best Explanation
Hart H.L.A. Are there any natural rights?
Hartmann Klaus Hegel: A Non-Metaphysical View
Harts H.L.A. Positivism and the Separation of Law and Morals (1958)
Heidegger Origin of the Work of Art
Heidegger Question Concerning Technology
Hempel Aspects of Scientific Explanation
Hempel Problems and Changes ... .
Hempel Studies in the Logic of Confirmation
Henrich Dieter Fichtes Original Insight
Henrich Dieter Kants Notion of a Deduction and the Methodological Background of the First Critique
Henrich The Proof-Structure of Kants Transcendental Deduction
Hesses Mary Models in Physics(1953)
Hilbert On the Infinite
Hintikka Cogito ergo sum: Inference or Performance?
Hoffmans Paul the unity of descartes man
Hohfelds Wesley Some Fundamental Legal Conceptions as Applied in Judicial Reasoning (1913)
Horkheimer Max Traditionelle und kritische Theorie Zeitschrift für Soziaforschung (1937)
Hull David Are Species Really Individuals?
Jackson Frank Epiphenomenal Qualia
Jackson What Mary didn't Know
Jaegwon Kims in Supervenience and Mind.
Johnston Mark Dispositional Theories of Value
Judith Jarvis Thompson A defense of abortion
Kaplan David Quantifying In
Kaplan David Demonstratives
Waters Kenneth C Genes Made Molecular:
Kitcher Philip and All That: A Tale of Two Sciences(1953)
Kitcher Philip and Kim Sterelny The Return of the Gene
Knobe Intentional Action and Side Effects in Ordinary Language
Korsgaard Skepticism about Practical Reasoning
Korsgaard Two Distinctions in Goodness
Kripkes a puzzle about belief
Kripke Saul Naming and Necessity(1972)
Kripke Saul A Completeness Theorem for Modal Logic
Kripke Saul Semantical Considerations on Modal Logic
Lakatos Proofs and Refutations
Lewis David Counterpart theory and quantified modal logic
Lewis David How to Define Theoretical Terms
Lewis David New Work for a Theory of Universals
Lewontin Richard The Analysis of Variance and the Analysis of Causes
Lewontin Richard The Units of Selection
Longino Helen Can There Be a Feminist Science?
Machamer Peter Lindley Darden and Carl Craver Thinking about Mechanisms
Mackie JL Causes and Conditions
Mackie the Refutation of Morals (1946)
Mallons Ron Race: Normative Not Metaphysical or Semantic (2006)
Marcus Moral Dilemmas and Consistency
Marcus Ruth Modalities and Intentional Languages
McDowell Virtue and Reason
Mills Susan and John Beatty The Propensity Interpretation of Fitness
Montague The Proper Treatment of Quantification in Ordinary English
Moore A Defense of Common Sense
Moore Proof of an External World
Moore The Refutation of Idealism
Nagel Thomas Moral Luck
Nagel Thomas What is it Like to be a Bat?(1974)
Nagel Thomas War and Massacre
Ogden and Richards The meaning of meaning
Owen The place of the Timaeus in Platos dialogues
Parfit Derek Equality or Priority? Ratio
Parfit Derek Personal Identity
Perry Frege on Demonstratives
Perry The Problem of the Essential Indexical
Winchs Peter Understanding a Primitive Society
Strawson PF Freedom and Resentment
Popper Science: Conjectures and Refutations
Prichard Does Moral Philosophy Rest on a Mistake?
Priest Logic of Paradox (1979)
Prior The Autonomy of Ethics (1960)
Prior The Runabout Inference Ticket
Putnam Hilary The Meaning of "Meaning" (1975)
Putnam Minds and Machines
Quine Truth By Convention
Quine W.V.O. On What There Is (1948)
Quine W.V.O. Two Dogmas of Empiricism (1951)
Rachels James Active and Passive Euthanasia
Railton Peter Alienation Consequentialism and the Demands of Morality
Ramsey Frank Truth and Probability.
Ramsey The Foundations of mathematics
Rawls John Two Concepts of Rules Philosophical Review (1955) 64 3-32.
Rawls John Justice As Fairness
Rawls Justice and Fairness: Political not Metaphysical
Rawls Phil Review article Justice As Fairness
Ridings Laura Jocasta (1928)
Goodin Robert E Whats so special about our fellow countrymen?
Russell Bertrand On Denoting (1905)
Russell Mathematical Logic as Based on the Theory of Types
Russell On Vagueness
Russell The Philosophy of Logical Atomism
Ryle Gilbert Knowing How and Knowing That
Sartre Existentialism is a Humanism
Scanlon Preference and Urgency
Searle JohnMinds, Brains, and Programs(1980)
Searle John How to derive Ought from Is.
Sellars Philosophy and the Scientific Image of Man
Sellars Wilfrid Empiricism and the Philosophy of Mind(1956)
Sen Amartya Equality of What
Shannon A Mathematical Theory of Communication
Shelbys Tommie Foundations of Black Solidarity: Collective Identity or Common Oppression? (2002)
Singer Peter Famine Affluence and Morality
Smart Extreme and restricted utilitarianism
Smart JJC Sensations and Brain Processes
Smith Michael Dispositional Theories of Value
Sober Elliott and Richard Lewontin Artifact Cause and Genic selection
Stalnaker Assertion
Star Susan and James R Griesemer Institutional Ecology Translations and Boundary Objects: Amateurs and Professionals in Berkeleys Museum of Vertebrate Zoology 1907-39
Stevenson The Emotive Meaning of Ethical Terms
Strawson Peter Freedom and Resentment (1960)
Strawson P.F. On Referring(1950)
Tarski Alfred The Semantic Conception of Truth and the Foundations of Semantics(1944)
Tarski The Concept of Logical Consequence
Taurek John Should the Numbers Count?
Thomson JJ The Trolley Problem The Yale Law Journal (1958) 94 1395-1415.
Thomson Judith Jarvis A Defense of Abortion
Turing A.M. Computing Machinery and Intelligence(1950)
Turing On Computable Numbers With an Application to the Entscheidungsproblem
Urmson J The Interpretation of the Moral Philosophy of J.S. Mill Philosophical Quarterly (1953) 3.
van Inwagen Peter the doctrine of arbitrary undetached parts
van Inwagen Peter The Incompatibility of Free Will and Determinism
Vlastos The third man argument in the Parmenides
Watson James and Francis Crick Molecular Structure of Nucleic Acids: A Structure for Deoxyribose Nucleic Acid
Wiggins Truth Invention and the Meaning of Life
Williams Bernard Moral Luck
Williams BernardPersons, Character, and Morality(1977)
Williams Internal and External Reasons
Williams Internal Reasons
Williams The analogy of city and soul in Platos Republic
Wilson Margaret Dauler Descartes: the epistemological argument for mind-body distinctness
Wittgenstein L A lecture on ethics first published in The Philosophical Review 1965
Wolf Susan Moral Saints
Wollheim Richard A paradox in the theory of democracy
Young IM Polity and Group Difference: A Critique of the Ideal of Universal Citizenship Ethics (1989) 99:250-74.

20120808: George Neuner recommendations on Garbage Collectors

Rick Jones's biblio/collection: http://www.cs.kent.ac.uk/people/staff/rej/gc.html

Henry Baker's biblio/collection: http://home.pipeline.com/~hbaker1/

Edsger Dijkstra's collection: http://www.cs.utexas.edu/users/EWD/

the biblio at: http://www.memorymanagement.org/bib/

and poke around on Citeseer http://citeseerx.ist.psu.edu http://citeseer.ist.psu.edu/stats/articles

And Voitaly Magerya suggested:

* "Uniprocessor Garbage Collection Techniques" by Paul R. Wilson [1]. [1] ftp://ftp.cs.utexas.edu/pub/garbage/bigsurv.ps

* "Dynamic Storage Allocation: A Survey and Critical Review" by Paul R. Wilson, Mark S. Johnstone, Michael Neely and David Boles [2]. [2] ftp://ftp.cs.utexas.edu/pub/garbage/allocsrv.ps

* "Representing Type Information in Dynamically Typed Languages" by David Gudeman ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/typeinfo.ps.gz

Boehm's Multiprocessor Memory Allocation and Garbage Collection" http://www.hpl.hp.com/techreports/2000/HPL-2000-165.pdf and "Reducing Garbage Collector Cache Misses" http://www.hpl.hp.com/techreports/2000/HPL-2000-99.pdf

20120726: Jurgen


P. Taine (r.a.sf.w) opined:

With notes about most of the obscure (and many of the not so obscure) references. It also has the complete text, and displays the notes and text side-by-side.

20120720: Robert Fripp and the League of Crafty Guitarists

Guitar Craft was his teaching approach cum musical movement, with a novel tuning.

Erich Tamm wrote in Robert Fripp - From Crimson King to Crafty Master:

The third GC album, Get Crafty I, was recorded by Fripp and a twenty-six-member incarnation of the League of Crafty Guitarists in October, 1988, in Wessex. Some of the selections were taped at concerts, others during rehearsals. To the best of my knowledge, Get Crafty was never distributed to record stores, but exists solely as a cassette available by mail order through Guitar Craft Services. Which is too bad, because it is far and away the best of the three Guitar Craft recordings to date.

Fripp has faced the same dilemma as John McLaughlin: now that I can play anything I want to, just exactly what do I want to play, and why? And I suppose it must be said that at times both musicians have succeeded in answering such questions to the satisfaction of themselves and their audiences, and at other times they have failed. In Fripp's case, the success stories include, for instance, the "Sailor's Tale" solo on Islands and the twenty-minute rhapsody of "The Heavenly Music Corporation" on No Pussyfooting - both pieces requiring, when you think about it, not so much in the way of technical virtuosity as sheer inspiration and the ability to listen.

20120710: Richard Pettigrew

Epistemic utility theory, probabilism

An Objective Justification of Bayesianism I: Measuring Inaccuracy
An Objective Justification of Bayesianism II: The Consequences of Minimizing Inaccuracy*


20120706: Genetics of moral judgements


20120628: Scream for Jeeves

by Peter Cannon

Sea Wasp and David Shallcross both recommend (rawsf).

20120511: CR Shalizi on smooth tests

Neyman's original smooth test, which seems to have originated from the problem of how to combine p-values from independent experiments, which should all be uniformly distributed under the null hypothesis. One nice feature of this test is that if we reject the null, we immediately have an alternative, namely our maximum likelihood estimate of theta, for what the actual distribution is --- it tells us not just that the null model is wrong, but how, and what a better one would be like.

The R ddst package performs data-driven smooth tests.

20120511: book recommendations from CR Shalizi

Nathan Long, Jane Carver of Waar
James S. A. Corey (i.e., Daniel Abraham and Ty Franck), Leviathan Wakes
Matthew Johnson, Fall from Earth 
Bernard Williams, Truth and Truthfulness: An Essay in Genealogy

20120507: Road wear and tear

Road wear can be approximated as the fourth power of vehicle weight divided by the number of axles.

20120430: Apert

Les jumeaux: étude biologique, physiologique et médicale.

Publisher: E. Flammarion 1923.

Cited by von Verschuer 1939 as one of the earliest descriptions of the twin method.

20120330: t-distribution and Tsallis entropy

Tsallis entropy is an extension of the usual Boltzmann-Gibbs entropy that allows for long range interactions, so that total energy of the system is no longer proportional to the system size. The Tsallis type extension of the Gaussian distribution (q-gaussian) turns out be the Student's t-distribution with q=(df+3)/(df+1). at least for a range of df > 0. Borland [2008] derives it as a times series with a feedback process for the Black-Scholes model.

20120127: Markov blanket

Every set of nodes in the network is conditionally independent of A when conditioned on the set MB(A).

20120118: Darwin on the evolution of morality

from Soshichi Uchii, Darwin on the Evolution of Morality

(written in October, 1838) in his Notebooks:

Two classes of moralists: one says our rule of life is what will produce the greatest happiness.---The other says we have a moral sense.---But my view unites both & shows them to be almost identical. What has produced the greatest good or rather what was necessary for good at all is the instinctive moral senses: (& this alone explains why our moral sense points to revenge). In judging of the rule of happiness we must look far forward & to the general action---certainly because it is the result of what has generally been best for our good far back.---(much further than we can look forward: hence our rule may sometimes be hard to tell). Society could not go on except for the moral sense, any more than a hive of Bees without their instincts. (Old & Useless Notes 30, Barrett et al., 1987, 609.)

20110802: Slatkin, Hill on human Ne

With fertility and mortality rates typical of humans, the effective size is ~ 1/3 the census size (WG Hill, Effective size of populations with overlapping generations, Theor Popul Biol 3 (1972), pp. 278-289).

20110720: Cuckoo Hashing

Reinhard Kutzelnigg introduces this with:

Cuckoo Hashing is a hashing algorithm with the advantage of constant worst case search time, contrary to standard hashing algorithms as Open Addressing or Hashing with Chaining. (See Knuth (1973) for surveys on hashing.) The algorithm was introduced by Pagh and Rodler (2001) and a further analysis was done by Devroye and Morin (2003). The main idea of Cuckoo Hashing is to split up the available amount of memory. Therefore, it uses two tables of size m, and two independent hash functions h1 and h2, to store the n data points. Every key x might be stored at exactly two positions, namely h1(x) and h2(x). So, one has to inspect at most two positions during the search for x. Further, we allow at most one element to be stored at any position. It may of course still happen, that both possible storage places of a given key x are already occupied. We solve this problem imitating the nesting habits of the cuckoo, and allow x to throw out the key (say y) occupying the storage position. Next, we insert the y at its alternative position, which is also possibly occupied. We repeat this steps, until we find an empty position or conjecture that we have entered an endless loop. In the the latter case, we have to choose new hash functions and rebuild the data structure.

The Wikipedia article comments that three hash, or cuckoo-bucket hybrids are slower but allow a fuller table (cuckoo works best <50% occupancy).

20110708: comic novels (bookcritics.org)

Richard Russo Straight Man
Nabokov Pnin
Jennifer Egan A Visit from the Goon Squad

The Egan is good.

20110708: numerics

The Institute for Reliable Computing: http://www.ti3.tu-harburg.de/~rump/intlab/

BARON global optimizer: http://archimedes.cheme.cmu.edu/baron/baron.html

20110119: Solar power for the green man

200-650 Calories given maximal efficiency of photosynthesis at 11%: 4730 * 1.8 * (0.2,0.7) * 0.11 = 187, 655

This allows him to walk (50 kcal/km)

field metabolic rate FMR (kJ/d) = a (mass [g])^b reptilian 0.196 M^(0.889) varanidae 0.208 M^(0.915)

Insolation for most people is from 150 to 300 W/m2 or 3.5 to 7.0 kWh/m2/day == 3009 kcal/m2/day. For example, typical insolation in Hawaii is closer to 5.5 kW-hrs/square meter/day (4729 kcal/m2/day.

Exposure ratios for most body sites in the standing position are 0.14 to 0.39, and 0 to 0.75 in the sitting position.

BSA = 1.6-1.8 m2

For actual sunlight, where only 45 percent of the light is photosynthetically active, the theoretical maximum efficiency of solar energy conversion is approximately 11 percent. In actuality, however, plants do not absorb all incoming sunlight (due to reflection, respiration requirements of photosynthesis and the need for optimal solar radiation levels) and do not convert all harvested energy into biomass, which results in an overall photosynthetic efficiency of 3 to 6 percent of total solar radiation.

Bolton et al (1991): Within the ideal assumptions: (1) two Photosystems for photosynthetic fixation of CO2, (2) all solar photons >700 nm are absorbed, (3) the photon requirement is 8 for each CO2 molecule fixed and O2 molecule evolved and (4) the principal stable product of photosynthesis is d-glucose, the theoretical maximum efficiency of conversion of light to stored chemical energy in green-plant type (oxygen-evolving) photosynthesis in bright sunlight is calculated to be 13.0%. Thermodynamic arguments are presented which indicate that a photosynthetic system with one Photosystem would be highly unlikely to be able to drive each electron from water to evolve O2 and reduce CO2. The practical maximum efficiency of photosynthesis under optimum conditions is estimated to be 8-9%.

The highest solar energy conversion efficiency reported for C3 crops is about 2.4% and about 3.7% for C4 crops across a full growing season based on solar radiation intercepted by the leaf canopy. Higher short- term conversion efficiencies are observed for brief periods during the life of a crop reaching 3.5% for C3 and 4.3% for C4. A major factor that lowers the conversion efficiency below this theoretical value even for healthy canopies under optimal conditions is insufficient capacity to utilize all radiation incident on a leaf.

For C3 plants, overcoming photorespiration is an obvious target. This may be achieved by converting C3 crops to C4 or by improving the specificity of Rubisco for CO2. Conversion of a C3 to C4 crop would theoretically raise the maximum solar energy conversion efficiency at 30 8C from 4.6% to 6%. If Rubisco can be engineered to completely eliminate the oxygenation reaction, this would raise efficiency to 8.8%.

Algal photosynthetic efficiency has been doubled by reducing the chlorophyll antenna size.

Pirt et al (1980): The photosynthetic efficiency (PE) of a growing algal culture was determined from the growth yield (Y), that is, biomass produced/light absorbed and the calorific value of the biomass (k); PE = kY. To obtain the maximum photosynthetic efficiency the algae were grown in light-limited chemostat cultures in urea-mineral salts media plus CO2 and steady-states were obtained at different specific growth rates. Maximum growth yields (YG, g dry weight kJ-1) were 0.0153 for the Sorokin Chlorella strain 211/8k and 0.0206 for a newly selected mixed culture MA003 which consisted of an alga and three species of heterotrophic bacteria. The maintenance energy (m) of the mixed culture MA003 was in the range 0-0.32 kJ g-1 dry weight h-1 and the specific maintenance rate (mYG) was in the range 0-0.0066 h-1. In Chlorella strain 211/8k the maximum PE was 34.7% which corresponds to a quantum demand (n) of 6.6 per O2 molecule evolved. In the mixed culture MA003 the maximum PE was 46.8% with 95% confidence limits, 42.7-51.5. From our results with culture MA003 it is deduced that the maximum practicable storage of total solar energy by algal biomass growth in vitro is 18%.

sib-pair code