My only question: what's wrong with Figure 2?

Not a big SF/fantasy reader since high school, but I make exceptions for Neal Stephenson and for GRRM. No one does twisty turny backstabbing politics better than him. General premise is that if you're in line for a throne, your life expectancy follows a heavy-tailed distribution with a mean of 50 pages, and you will likely be surprised at who kills you. Everyone has a flaw and it's often hard to figure out who to root for. Nice and dark.

I've taken Modafinil a couple times -- it works! -- and routinely take atenolol, a beta-blocker, for stage fright. The latter in particular makes a huge difference for big talks. Works on first dates, too!

I am very sympathetic to the noise argument. In discussions of the proportion of variance explained in [important variable X], the statement that Y explains 40% of the variance is, in my experience, frequently greeted with the automatic assumption that some more-important causal factor must explain the remaining 60%. Instead, the first question should be what proportion of the variance is explainable at all. If 50% is due to noise, then 40% actually covers 80% of the explainable variance. 
 
Note that estimating the proportion of explainable variance is difficult, often (seemingly) impossible. Reliability coefficients are useless, because variance also arises from non-identity between the measurement and the construct being measured. If one can account for 100% of the variance, that's the end of the story, but one must not be working in population genetics. A more typical situation is to have factors that cover a minority of the variance. Then...what? I wonder whether a principled way to decide between noise and as-yet-undetected causal factors exists, even in some useful subset of situations.

Wake up, people. Added value? Useful? Our financial markets are riddled with (and in some cases based entirely on) implementations of theories (like Black-Scholes) devised by social scientists. They are as quantitative as any physical theory -- in fact, if you want to learn about Borel sigma algebras and measure theory, you'd do worse than to learn it from a derivatives trader.

Major problem: no control for expression level, the best determinant of evolutionary rate in bacteria, in their analyses. Shapiro & Alm find that genes with similar function evolve at rates that are more similar than expected if the genes were sampled randomly. They conclude that these patterns are due to selection on gene function. 
 
The obvious question is whether such genes have more-similar expression levels. At least one creditable hypothesis for why highly expressed proteins evolve slowly suggests that expression level constrains evolution for reasons having nothing to do with function (instead, selection to withstand ribosome errors to avoid toxic protein misfolding imposes the major constraint). 
 
Another major issue: they use dN/dS everywhere, presumably under the assumption that dS is the neutral fixation rate. But in bacteria, it's been known for decades that synonymous sites are under selection in bacteria. dN/dS has no simple interpretation in this case. 
 
Personally, I take a dim view of these broad bioinformatics papers that make sweeping claims on the basis of relative evolutionary rates. If a gene evolves slowly, it is said to be functionally constrained. If a gene evolves rapidly, it is under positive selection. Yet biochemical follow-up on these claims is never done. Where is the mutagenesis study that shows that a slower-evolving protein tolerates fewer substitutions than, say, a faster-evolving paralog? Where is the study on any one of the supposedly 40% of substitutions in Drosophila that are supposedly adaptive? Contrast with stuff out of Joe Thornton's lab, and you can see why Dean and Thornton just go off on Shapiro & Alm-type analyses.

Ugh. First, deciding whether the ancestor of humans and apes was an ape or a human seems silly. Does he mean to imply that we modern humans could interbreed with the common ancestor of both (and therefore are conspecific), but modern chimps could not? Yikes. 
 
An upright ancestor, ok. That ain't a human by a longshot. If his title was, "A bipedal ancestor for the apes?" then I wouldn't be twirling my finger by my temple.

hairless individual less likelyExcuse -- more likely.

A request to be gentle naturally provokes the urge to slaughter. At least the NSFW posts don't take themselves seriously. The risk of coming down on newbies is that they will think thrice before posting their personal theories again. God, I hope that's true. 
 
1) If you want to post a new sweeping hypothesis based (however loosely) on pre-existing data, provide references. In "many other species", conquering males "kill the existing pups." I'm sure they do. Please provide a link to something credible. And so on, throughout. 
 
2) "...much of human culture evolved around allowing fathers to identify their specific offspring" This is your theory, I take it? Could you perhaps provide examples to defend this idea about "much of human culture?" You mean, like storytelling, religion, art, and food preparation? Do tell. 
 
3) When presenting a novel theory, kindly present falsifiable hypotheses (in principle if not practice) which distinguish it from other theories. If other theories do not exist, or are lame, as asserted, kindly begin by convincing us that there is indeed a problem to be solved. 
 
To the theory. First, skin color. Let us speculate that most invading males during the course of human evolution were from nearby rival groups and had the same skin color. How does skin color help in recognizing your kids? 
 
Second, hairlessness. Why is a hairless individual less likely to be recognizable as your progeny than a hairy one? Hair (distribution, texture, thickness) would seem to be among the most recognizable of human traits. 
 
Third, what part of this theory do chimps not satisfy? They have largely hairless faces. Are you theorizing about hairless bodies (at which point the theory's merit revolves around our ability to recognize each other's lumbar vertebra, I suppose) or hairless heads? Please fill in the basics for us. 
 
Happy holidays.

*nod*. damn, concurrence is so boring. ;)

what hypothesis-generating mechanism would lead you to test gene deserts for association with crohn's disease, for example? or a gene of unknown function for association with obesity? 
I think we may be in violent agreement. I'm saying our previous way of generating hypotheses -- namely, thinking hard and extrapolating from anecdotal clinical results -- has largely failed. However, does anyone doubt that there is a good biological reason for some gene-desert regions to be associated with Crohn's disease? The value of GWASs are to provide unbiased evaluation of the previously dominant approach, and they have found that approach wanting. Instead of abandoning the approach, why not use GWASs to fix it? Unless you believe that *there are no new predictive principles to be learned*. The goal should be to obviate expensive GWASs where the vast majority of results are "no effect", and replace them with something more efficient in cost per positive result. 
 
And how will GWASs fare for traits involving dose-dependent epistatic interactions between 5 loci? There aren't enough humans in the world to sequence. Delight in brute-force while it lasts -- I'm with you -- but I'd be leery of letting that glee lead to devaluation of genuine insight. GWASs and their ilk should be seen as ways to build insight, not as insight's alternative.

The "hypothesis-free" approach can be justified in part by the failure of candidate-gene approaches to predict the major loci found by replicable GWASs. And the author's caveats are well-taken. I'm reacting to the somewhat sensationalist title, which should have instead said, "Hypotheses are overrated for genome-wide association studies." 
 
One might ask whether there is any generality to the claim that once you have sufficient scale, you might as well throw out insight and do brute-force attack. If we could (if we had enough people, that is), would it be better to uncover the biology of diabetes and extract prevention/cure from that knowledge, or throw the Sigma-Aldrich catalog at a sufficiently large sample and feed everyone else whatever worked? 
 
So long as you have infinite resources, brute-force generally wins. But it's expensive, and that, I think, is the major failing of hypothesis-free research (although compared to pursuing irrelevant hypotheses, the cost-per-useful result may be much higher!). 
 
Roughly this large-scale, insight-free approach was tried by Big Pharma, and they are still reeling from its failure. It was the great robotic hope: combinatorial chemistry. A nice quote from a lead Bristol-Myers scientist: 
 
You end up making things that you can make, rather than what you should make. 
 
The same is likely true for GWAS -- we end up testing what we can, rather than what we should. Given our failures of insight, that's a good and realistic approach ("temporarily, in this field"). But rather than accepting that hypotheses are overrated, a point of view which naturally leads people to abandon hypothesis-generation, I believe the proper response is to figure out what's wrong with present methods of hypothesis-generation so that we can fix/replace them. The editorial I would have written would bear the title, "Why are our hypotheses so crappy?"

Oh, goodness. How moronic. If your goal is to publish replicable association studies with vanishingly small P-values, perhaps the hypothesis is dead -- and was never alive to begin with. 
 
If your goal is to cure/prevent diabetes, or understand its biology, may I humbly submit that hypotheses will be needed?

I wonder...the larger a society is, the more third parties there are. Perhaps this effect is simply stochastic. ;)

A prediction of regression to the mean is that any extremes will regress -- that is, taking high-ability, high-SES kids or low-ability, low-SES kids and retesting them should show similar regression. And Fig. 4 shows that this regression does occur, but it's less pronounced than in the high/low or low/high cases. That's one control for regression to the mean. 
 
A second check I'd like to see is for reliability. It may be that, if you are poor, when you score in the 90th percentile it's more likely to be by good fortune than raw talent. And vice-versa, with rich kids who score in the 10th percentile being more likely to have just had a bad day. (Please don't start screaming, this is just being careful.) Such an effect is trivially predicted if the mean test scores of the high-SES are higher than that of the low-SES group. (The reason you expect asymmetric regression is that the pressure to regress, if you want to think of it that way, grows stronger the farther you are from your group's mean. High-mean-group kids who score low are much farther from their mean than low-mean-group kids who score low, and vice versa, ergo the expected regression depends on which group you start in.) And Figure 1 shows that, indeed, the high-SES group scores substantially higher than the low-SES group. 
 
All told, regression to the mean seems a quite plausible explanation for all of Fig. 4.

Neither conditions (1) or (2) are generally true. If a mutation is beneficial, then it will tend to rise in frequency, not be diluted out. (This depends on population size, which is a major point of the present work -- the larger the population, the more likely a beneficial mutation is to rise up and take over.) While demography can have all sorts of interesting effects, isolation of an evolving population is certainly not required. Second, the normal functioning of natural selection is not to eliminate mutations, but to modulate the frequencies of mutations in accordance with the selective (dis)advantages they confer. That is, deleterious mutations tend to be eliminated, and beneficial mutations tend to fix in the population.

Strange -- PNAS gives me an "Article not found" message when I follow the link. Even when I sign in to my august institution's subscription.

see, even my sentence construction reflects my crappy genes! i'm sure you smart ones will figure out what i meant.

indeed i do. but as a less-able whitey, you'll forgive me for being an asian charity case. :)

Can we move on already? Let's treat people as individuals, and get on with the show. 
You're so right. No more referring to these fuzzy sets like "male" and "female", because there are exceptions, and you can't tell by looking. If someone asks, "Where is the restroom?", you would have us unfailingly inquire, "Which one?"  
 
slap! 
 
A world without groups is not only practically impossible, it's a complete waste of time. What's needed are informative group designations that are predictive -- albeit imperfectly! -- of important underlying traits. Race, for right now, is such a designation. (That said, I do advocate rampant interbreeding as a solution to the whole mess.)