A few years ago I had a semi-serious post up making fun of Armand Leroi for broaching the topic of neo-eugenics. Now there are reports of elective pre-implantation screenings:

Genes determining sex, hair and eye colour can be identified, alongside any DNA red flags for diseases such as muscular dystrophy, cystic fibrosis and Down's Syndrome.

"Basically any genetic ailment, and there are thousands of them. We find the genetic error responsible for that in the embryo," Dr Steinberg says.

Only those embryos free of problem genetic markers and matching parental wishes, if stated, are then implanted in the mother.
...
"To deny them the ability to do that when the technology is there is to me unethical," Dr Steinberg says.

"You can say eye colour and hair colour are not diseases, no they're not, and there is a cosmetic element to it, but we fix crooked noses all the time.
...
He says he's concerned Australian women are risking their health by undertaking IVF overseas for "frivolous" reasons, using a process that raises the moral issue of "deliberate embryo loss".

"But the main issue is the idea of treating the child as an object, as product for which you are seeking quality control," Dr Tonti-Filippini says.

1) Part of this is publicity, you can get only so much information out of genetic tests right now (see Genetic Future). Take a look at Genetic determinants of hair, eye and skin pigmentation in Europeans, and note how much higher the odds ratio (20-30 vs. ~5) for OCA2 "blue-eye" markers are vs. the ones which might give some information about hair color. The same differences in effect size apply to disease loci. I suspect many people will balk at paying up when confronted with the provisionality of some of the inferences.

2) It isn't as if these fertility technologies aren't without downsides (not to mention the cost).
Note: The link above is to an Australian newspaper. So I don't take everything they report literally...perhaps they spiced up a quote here and there?

Labels: Genetics

Just watched the film Today's Man, which is about an individual, Nicky Gottlieb, with Asperger Syndrome. Near the end of the film he attends a meeting with others who are not "neurotypicals." Gottlieb has some weird ticks throughout the film which shows quite clearly that he's not "all there" (or, more precisely, no one else is there in his own mind). But it was really interesting to see a meeting of people with the same lack of normal social skills...they all seemed "out of sync" with each other (or, perhaps they were in sync in a different way which I wasn't able to perceive). The most peculiar aspect for me was that physically these were all human beings, but their manner, gestures (or lack of) and social fluidity was almost like that of alien species. I've met people who have major social skill deficits before, but I haven't observed dozens trying to interact. The closest thing I've seen in my own life are interactions with Singularitarians and some Perl Mongers. But these events & groups were tied together by a common theme or topic around which verbal exchanges invariably circled in a structured manner. Nicky Gottlieb going to be a meeting with other individuals suffering Asperger Syndrome and talking about their lack of social skills and attempting to grapple with the fact that most humans view them as abnormal freaks was different. It was like peering into the psychology of a species running radically different software.

Note: I'm not too interested in whether there really is unitary Asperger Syndrome rooted in specific neurological dysfunctions. Rather, it's clear that a minority of humans are noticeably socially retarded enough that when they interact together with others who lack normal social skills the communal synergy, or lack thereof, is definitely not a scene you see everyday.

Labels: Biological Psychology

One of my friends has built a Django site for a corp and wants to have a blog to accompany it. The blog will be pretty full featured with a lot of posts per day. His question: will he be better off using Wordpress + plugins, buying Expression Engine, or rolling his own in Django using some of the stub apps out there?

I ran into an interesting comment on the net the other day.. "for some, it is hard to determine what productive and ethical use society can make of genetic knowledge that certain individuals are predisposed to higher than average intelligence"

Perhaps others can think of some productive and ethical uses. Any suggestions?

Some people may already have a certain amount of such knowledge. For example, my high school geometry teacher thought I would probably do well - he had some strange rationale based on remembering how my mother had done in his class. I did do well, but maybe he was just a lucky guesser.

In lieu of a full review of Greg Cochran and Henry Harpending's new book, The 10,000 Year Explosion, I'll keep this short: this book is interesting, well-written, and probably mostly wrong.

The book reads as a series of historical narratives grounded around recent work by the authors in population genetics. In particular, they focus on claims very familiar to readers of this site: that Neandertal introgression was in fact probable, that the IQ of Ashkenazi Jews is a result of strong recent selection, and that thousands of new selected alleles are currently sweeping through human populations (along with novel theories, like that the evolution of lactose tolerance was a major force in population expansions). All of these things are, a priori, vaguely plausible, and are certainly fun to read about, but let's be honest: in a few years, most, if not all, of these are going to be in the dustbin (if any of them are true, my money is on Neandertal introgression).

In any case, this book is not intended to be "correct", so to speak--it seems to be more intended as an overarching frame of reference for viewing human history, acting as a counterpoint to that presented by authors like Jared Diamond. But if anthropology has a "Guns, Germs, and Steel problem", I just hope population genetics doesn't end up with a "10,000 Year Explosion problem".

Michael Blowhard is doing a 4-part interview with Greg Cochran on his new book, The 10,000 Year Explosion: How Civilization Accelerated Human Evolution. Part 1 is up.

Labels: Greg Cochran

The latest disease to be put under the scrutiny of a large genome-wide association study is psoriasis--see articles here, here, and here. These are mostly standard studies, but once again I'm struck by the effect of the MHC (HLA) region (see the figure). It was well-known, of course, that variation in HLA affects all manner of immunity-related phenotypes, but what's becoming clear is that this variation has much larger effects than other loci in the genome. I find this somewhat surprising--associations with MHC variability were identified because typing HLA was feasible many years ago; well before genome-wide association mapping with SNPs was even considered. This could be a case where, in the analogy to the man searching for the lost keys in the dark, looking under the lamppost was actually the best bet.

Labels: Genetics

Monoamine oxidase A gene (MAOA) predicts behavioral aggression following provocation:

Monoamine oxidase A gene (MAOA) has earned the nickname "warrior gene" because it has been linked to aggression in observational and survey-based studies. However, no controlled experimental studies have tested whether the warrior gene actually drives behavioral manifestations of these tendencies. We report an experiment, synthesizing work in psychology and behavioral economics, which demonstrates that aggression occurs with greater intensity and frequency as provocation is experimentally manipulated upwards, especially among low activity MAOA (MAOA-L) subjects. In this study, subjects paid to punish those they believed had taken money from them by administering varying amounts of unpleasantly hot (spicy) sauce to their opponent. There is some evidence of a main effect for genotype and some evidence for a gene by environment interaction, such that MAOA is less associated with the occurrence of aggression in a low provocation condition, but significantly predicts such behavior in a high provocation situation. This new evidence for genetic influences on aggression and punishment behavior complicates characterizations of humans as "altruistic" punishers and supports theories of cooperation that propose mixed strategies in the population. It also suggests important implications for the role of individual variance in genetic factors contributing to everyday behaviors and decisions.

Regularly readers know we've talked about MAOA in the past, it's one of those big-effect genes which keeps popping up in behavior genetic studies. It definitely increases my confidence in the reality of past associations to see it produce results in experimental situations which match our predictions. Along with the dopamine receptor loci this is likely a keeper. The issue of mixed strategies is something I've been mentioning periodically, the idea of evolutionary game theory has been around for nearly two generations, and frequency dependent selection is over a century old. But I think far too often people who are interested in the intersection between biology and behavior start conversations which assume that these are minor or trivial dynamics. At a certain point if "transient" states are more common than periods when there is an ESS, you need to reorient your frame. Human environments (that is, cultures) change a lot. Sometimes that change is exogenous (consider shifting climate), but in many cases instability in the background parameters upon which an ESS is conditional might result in periodic shifts in state.* Peter Turchin's work to some extent describes this on the macro-level, while Martin Nowak has developed models on a smaller scale. The main unfortunate byproduct of accepting this relatively more complex explanatory framework is that it makes glib assertions about natural selection on behavioral traits more difficult.

Here's the ScienceDaily summary.

* Basically, if you figure out an "unbeatable" strategy at winning a game, you are screwed when the rules change.

Labels: Behavior Genetics

Arnold Kling comments about my assertion that until recently cities were genetic black holes:

Today,. we think of cities as places where people come to thrive. Wealth is higher in cities than in small towns and rural areas. Richard Florida tells us that the creative class is to be found in cities.

...

I wonder: who came to cities? Was it people without land? Were cities like an awful lottery that people would play when they had no other choice? A bunch of landless people gathered together to prey on one another, with the winners thriving (moving to the country as soon as they could afford it) and the losers enduring a Hobbesian existence, where life was nasty, brutish and short?

My comment was a contention in relation to reproductive fitness; not quality of life or satisfaction (H/T to Greg Cochran for the observation). In the comments I cite a paper which suggests that city and rural divide in mortality favored the rural until around 1900 in the United States. The divide does not exist in large part due to proactive public health measures. Needless to say, though there were variations (e.g., compare 18th century London to Tokyo/Edo), in the pre-modern context public health was much more primitive.

So why move to the city? I think it is likely correct that city air makes one free. We can see this today as social change is occurring in Developing World megalopolises. In the ancient cities there were clear benefits for the poor, Rome and Constantinople had doles for the urban proletariat (though these doles were of course simply viable due to rents derived from their Empire). After the wars of the middle to late republic many impoverished peasants migrated to Rome to escape famine (the famine was exacerbated by the fact that many men were called up as soldiers to serve in foreign wars, and so their labor was missing). On the other hand, despite the dole there was often no regular employment. From the data I have seen modern urban-life worldwide tends to correlate with a lower fertility; and I see no reason this would not be so in the pre-modern world (I assume that the marginal return on "extra hands" provided by more children would likely be lower for a sporadically employed urban laborer than a farmer). But the main difference I suspect is disease load over time. Plagues regularly killed on the order of 50% of the population of ancient cities. After the population declines the cities would bounce back, but not through natural increase, but further waves of rural migrants. Rome's population declined to tens of thousands in the medieval period, from on an order of 1 million in antiquity. I am skeptical of the idea that most modern Romans are descended from a demographic expansion of the medieval deme as opposed to migrants from the hinterland.

All the population genetic negatives are not to deny that civilization and the city are to a large extent identical (Sumer). Who says that cultural creativity or innovation has to track Darwinian fitness? Look at our own modern societies, the least successful by accepted measures are often the most "fit" in Darwinian terms.

Addendum: A shorthand way of thinking what I'm asserting, imagine two brothers who are farmers. One moves to the city to get on the bread dole, while the other attempts to make do on the margins. The former might have a higher chance of surviving, but because of the greater power of disease in the urban context the city brother is likely to have far fewer descendants than the country brother as periodically all of his descendants come under threat of dying in a plague (again, I also believe that fertility will be lower for urban descendants).

Labels: History

At ScienceBlogs, I posted on a new paper on Jewish genetics. At Taki's Magazine I offer my opinion as to why labor and capital flows are qualitatively different, as well as some pedantic comments on points of Roman history. I wouldn't be such a stickler on Roman history...but people just love to make analogies based on presumed correspondences. I've been meaning to review Peter Turchin's new book, Secular Cycles, which you can read for free as a PDF. But if you recall how long my last post on his previous book was, it might be a little while in coming, so I encourage everyone to read Secular Cycles if the topic interests them. Finally, Ed Yong has a nice review of a new paper Language Phylogenies Reveal Expansion Pulses and Pauses in Pacific Settlement (the paper is in Science, but it doesn't seem online yet).

The Department of Children, Schools and Families in Great Britain has released its report for 2007/2008 breaking down nationwide educational attainment by pupil characteristics, including race. Actually, the report was released in November, but I was holding off on posting about it because I was under the erroneous impression that updated, race-specific data would be provided in mid-January.

DavidB has posted on previous reports for 2002, 2003/2004, 2005, and 2006/2007. Some readers might be interested in my summary of the 2007/2008 GCSE results for racial and ethnic groups in England, which I've posted here.

Evidence of gender differences in the ability to inhibit brain activation elicited by food stimulation:

Although impaired inhibitory control is linked to a broad spectrum of health problems, including obesity, the brain mechanism(s) underlying voluntary control of hunger are not well understood. We assessed the brain circuits involved in voluntary inhibition of hunger during food stimulation in 23 fasted men and women using PET and 2-deoxy-2[18F]fluoro-D-glucose (18FDG). In men, but not in women, food stimulation with inhibition significantly decreased activation in amygdala, hippocampus, insula, orbitofrontal cortex, and striatum, which are regions involved in emotional regulation, conditioning, and motivation. The suppressed activation of the orbitofrontal cortex with inhibition in men was associated with decreases in self-reports of hunger, which corroborates the involvement of this region in processing the conscious awareness of the drive to eat. This finding suggests a mechanism by which cognitive inhibition decreases the desire for food and implicates lower ability to suppress hunger in women as a contributing factor to gender differences in obesity.

ScienceDaily has a lot more:

"The finding of a lack of response to inhibition in women is consistent with behavioral studies showing that women have a higher tendency than men to overeat when presented with palatable food or under emotional distress," Wang said. "This decreased inhibitory control in women could be a major factor contributing to the observed differences in the prevalence rates of obesity and eating disorders such as binge eating between the genders, and may also underlie women's lower success in losing weight while dieting when compared with men."

Here's a question: do the sexes differ in time preference?

Labels: sex differences

Just noticed that The BBC has a new site all about Darwin! God bless the British television owners! Also, our old friend Armand Leroi has a program this Monday, What Darwin Didn't Know. And Armand's been busy, apparently he's going to host a special for National Geographic, Darwin's Lost Voyage. Also, remember about Blogging For Darwin. I suspect I'll participate over at ScienceBlogs...but I'm not going to unless I reread Origin, which I haven't since I was about 12 years old when I didn't know anything about evolution.

Labels: Darwin

In the historical literature on Charles Darwin one of the commonest assertions (or assumptions) is that there was a long delay (of about twenty years) between Darwin's first formulation of the theory of natural selection, and his publication of that theory in 1858 (followed in 1859 by a fuller publication in the Origin of Species). Based on this assumption the historian or biographer then procedes to speculate on the psychological or social reasons for this extraordinary 'delay'.

I have long been sceptical about this approach. Of course, in one sense there was a delay: we know that Darwin formulated his first private version of natural selection in 1838, and didn't publish it until 1858. But to call this a 'delay' - with the implication that it requires some deep explanation - requires us to assume that everyone should publish the first idea that comes into their head. While this is undoubtedly a common practice - both now and in Victorian times - it isn't necessarily a good one, and it wasn't Darwin's. He liked to do things thoroughly, and most of his major works took years to prepare. As a well-known example, he took about 8 years over his study of barnacles. This isn't unduly long, when we consider that he described about 200 species in 30 genera, each requiring careful comparison and often dissection of many specimens, a reading of all the existing literature on the species, and careful writing up of the results. In fact, at an average of one species every two weeks, it might be considered a rushed job, though it was widely admired by experts at the time and since. (Added: for comparison, Thomas Davidson and Sydney Buckman each spent more than 40 years studying brachiopods and ammonites respectively, but at a higher rate of species per year than Darwin with his barnacles.)

I am therefore pleased to find that the common assumption of a 'delay' is strongly challenged in a recent article by John Van Wyhe, available here. Maybe he slightly protests too much, but in my view he is a great deal more right than wrong. At least there is now no excuse for glib repetition of the usual version.

Someone has suggested that the cover of our new book (the 10,000 year explosion) symbolizes the splitting of the human race into different species. I will award a metaphorical cigar to the first person who figures out what it _really_ means.



(Daddy's Skeleton Army is the alternate title suggested by my son Ben)

Labels: 10000 Year Explosion

From Toward Rational Exuberance: The Evolution of the Modern Stock Market:

What must have been most galling was a simple point Cowels often made that was never answered effectively by the investment advice practioners. As Cowels put it, "Market advice for a fee is a paradox. Anyone who really knew just wouldn't share his knowledge. Why should he? In five years, he could be the richest man in the world. Why pass the word on?"

In spite of the conclusions he reached, Cowels never doubted that investors would keep buying newsletters. As he put it, "Even if I did my negative surveys every five years, or others continued them when I'm gone, it wouldn't matter. People are still going to subscribe to these services. They want to believe that somebody really knows. A world in which nobody really knows can be frightening."

The quote is from Alfred Cowles, the early patron of econometrics. It's kind of like The Bachelor, people know it's not really going to work out, but they keep watching just in case there's a repeat of Trista & Ryan. This book was written in 2001, but a great deal of it will be of interest to those of us who live in 2009 (see this review).

Labels: Finance

Fortune is apparently digging....

Several press reports over the weekend and today in Britain have mentioned a new report on mixed ethnicity (or mixed race) families in Britain. The headline finding is that about 1 in 10 children are in mixed-ethnicity families. To be (slightly) more precise, 9 per cent of children are themselves of mixed ethnicity and/or living in families of mixed ethnicity. The report is credited to Lucinda Platt of the Institute for Social and Economic Research. The headline figure sounded on the high side to me, so I wanted to track down the report itself. As there was nothing about it on the website of the ISER I was beginning to suspect an odorous rodent, but I finally found a link to the text on the website of the Equality and Human Rights Commission, which sponsored the research. A copy can be downloaded here. I have only skimmed it, but the most relevant findings seem to be at pages 39-40. On a quick read I don't see any obvious flaw in the methods (such as double-counting), but at first sight there is one important area of uncertainty. The list of ethnic groups includes 'Other White', i.e. other than 'White British'. 'Other White' accounts for about 5 per cent of the population, and it is reaonable to assume that a large proportion - probably a majority [this is not correct: see Addendum below] - of them will be in mixed families with 'White British'. If these are counted among 'mixed ethnicity' families, they could account for up to half of the '1 in 10' headline figure, which is perhaps not quite what the headline writers had in mind. Perhaps a closer reading will clarify this. Kudos to the first reader to find the answer.

Addendum: On a slightly closer reading, I find I was wrong to assume that a majority of 'Other White' would be in families with 'White British'. Of those 'Other White' adults who are in partnerships, only about 1 in 3 such partnerships are 'mixed' (see pages 24-30). However, it does seem (unless I have missed something) that the 'headline' figure for children in 'mixed ethnicity' families would include 'White British - Other White' families. It also appears possible that 'mixed ethnicity' would include, say, 'Black Caribbean - Black African' or 'Indian - Pakistani'. If so, it seems distinctly misleading for the author to equate 'inter-ethnic' and 'mixed-race' (page 3 and elsewhere).

A long article by Steve Jones here from today's Guardian. (But ignore the rather feeble first paragraph. I doubt that 'every schoolchild' has as many ideas about Darwin - whether true or false - as this suggests.)

Portfolio has a story up, Biotech Decline:

Even the industry's lobbying group, the Biotechnology Industry Organization, says that 45 percent of publicly traded biotech companies will run out of cash in the next 6 to 12 months. A mere 10 percent of the 370 listed companies have a positive cash flow.

...

Much of the activity at the J.P. Morgan conference involved companies and investors that still have money shopping for deals. "We are being visited by a number of companies," said Jay Flatley, CEO of the genomic sequencing company Illumina, based in San Diego.

Illumina recently announced an $18 million development deal with Oxford Nanopore of Britain for its next-generation genetic sequencing technology. Illumina has remained profitable with a healthier-than-average stock price even during the downturn.

Dan MacArthur has more on Illumina & Oxford Nanopore.

Labels: Biotech

Some of you may know that California is in a budget crisis. A significant portion of the proposed stimulus package is going to go to the states which are in a financial crunch because of tax shortfalls. But did you know that the in the 19th century several American state went into default to European bankers? Below the fold for the list....

Arkansas (never paid in full)
Indiana (eventually paid)
Illinois (eventually paid)
Michigan (never paid in full)
Louisiana (never paid in full)
Mississipi (repudiated debt, never paid)
Florida (repudiated debt, never paid)
Maryland (eventually paid)
Pennsylvania (eventually paid)

(cite)

One of the justifiers of the repudiation of the debt in Mississipi was Jefferson Davis. This caused some issues when the Confederacy went looking for financing from those Europeans who Davis had argued a decade before should be stiffed.

Labels: History

See Greg Clark's paper In Defense of the Malthusian Interpretation of History if you haven't read Farewell to Alms. Commentary from Calculated Exuberance.

Labels: Economics

Huge Rise In Male Mortality Coincided With Move From Communism To Capitalism. In other news, A Trivers-Willard Effect in Contemporary Humans: Male-Biased Sex Ratios among Billionaires (H/T Dienekes). Economics might not take human biology into account, but it have an influence on human biological processes (recall that fertility collapsed in Russia after the fall of Communism).

Labels: Economics

Over at my other weblog I review a paper on the genetics of Koreans. The title is a shout-out to an old Korean American friend of mine who received a great deal of grief from his female Korean American peers for openly admitting that he was into a Hmong girl. She was very good-looking, but as they said, "But she's Hmong...."

Labels: Genetics, Koreans

Genetic Differences between the Determinants of Lipid Profile Phenotypes in African and European Americans: The Jackson Heart Study:

Single-base changes in DNA can affect biochemical measures, such as blood cholesterol or lipid levels. Such changes or ā€œvariantsā€ can be associated with a trait either because they cause the trait or because they are linked to other causal variants. In either case, the associated variant(s) may be useful in predicting the trait. The chromosomes in which DNA is packaged cross over and recombine with each other in each generation, so that in historically separate populations, such as Africans and Europeans, the patterns of genetic linkage between variants differ. In the current study, we analyzed a large group of African Americans, testing genetic variants that had been associated with cholesterol and lipid levels in European-derived populations to assess their predictive value on two different genetic backgrounds within the same cohort. The ability of some variants to predict cholesterol or lipid traits was strongly dependent on genetic background, indicating that they may be tightly linked to other causal variant(s) in European populations and may not, themselves, be directly responsible for trait variability. We conclude that the predictive value of specific variants for risk assessment can differ critically across populations.

Different populations have different genetic backgrounds.

Rebecca Skloot points me to a piece which claims that the idea that there are, on average, six degrees of separation is false:

It turned out, she told us, that 95% of the letters sent out had failed to reach the target.

Not only did they fail to get there in six steps, they failed to get there at all.

...

We like the idea of six degrees of separation, she says, because it makes the world feel more intimate. But there are barriers - like race and class - she argues, that can sometimes make separation real and deep.

You can read the longer more academic argument against the idea online. The idea that there were only six degrees struck me as weird because my own personal life seems to exhibit a lot of clustering based on affinities.

Recordkeeping alters economic history by promoting reciprocity:

We experimentally demonstrate a causal link between recordkeeping and reciprocal exchange. Recordkeeping improves memory of past interactions in a complex exchange environment, which promotes reputation formation and decision coordination. Economies with recordkeeping exhibit a beneficially altered economic history where the risks of exchanging with strangers are substantially lessened. Our findings are consistent with prior assertions that complex and extensive reciprocity requires sophisticated memory to store information on past transactions. We offer insights on this research by scientifically demonstrating that reciprocity can be facilitated by information storage external to the brain. This is consistent with the archaeological record, which suggests that prehistoric transaction records and the invention of writing for recordkeeping were linked to increased complexity in human interaction.

It's OA, so click through and check out the charts. The effect is striking. I think this sort of result relates to my post about The Transparent Society.

Serotonin Transporter Genotype Modulates Social Reward and Punishment in Rhesus Macaques (paper is OA, so click through for stats & charts):

Serotonin signaling influences social behavior in both human and nonhuman primates. In humans, variation upstream of the promoter region of the serotonin transporter gene (5-HTTLPR) has recently been shown to influence both behavioral measures of social anxiety and amygdala response to social threats. Here we show that length polymorphisms in 5-HTTLPR predict social reward and punishment in rhesus macaques, a species in which 5-HTTLPR variation is analogous to that of humans.

In contrast to monkeys with two copies of the long allele (L/L), monkeys with one copy of the short allele of this gene (S/L) spent less time gazing at face than non-face images, less time looking in the eye region of faces, and had larger pupil diameters when gazing at photos of a high versus low status male macaques. Moreover, in a novel primed gambling task, presentation of photos of high status male macaques promoted risk-aversion in S/L monkeys but promoted risk-seeking in L/L monkeys. Finally, as measured by a "pay-per-view" task, S/L monkeys required juice payment to view photos of high status males, whereas L/L monkeys sacrificed fluid to see the same photos.

ScienceDaily has an interesting tidbit:

In a series of experiments, the S version of the gene in monkeys was found to influence their risk-taking when faced with particular social stimuli.

"Based on work in humans, we interpreted this to reflect an induction of a fearful emotional state, which often leads people to become risk averse," said Karli Watson, Ph.D., of the Duke Department of Neurobiology, lead author on the paper.

In human populations of European ancestry, 48% are S/L and 36% are L/L. The rest are S/S. The S allele is more common in Asian populations, Watson noted.

More on 5-HTTLPR.

Labels: human biodiversity

Everyone is talking about Steven Pinker's article in the NY Times on his experiences toying with results from personal genomics companies Counsyl and 23andme. The article is, like most of Pinker's writing, entertaining and scientifically legit; well worth a read.

One interesting point to notice is that the different companies involved in this nascent field (I find Razib's analogy to the early computer industry apt) are taking very different angles; there's been something of an adaptive radiation into these new niches (ie. 23andme going for the social networking approach, Counsyl a more medical one, etc.) My impression is this is likely to be overall a good thing for people who, of course, will have varied ideas of the type of data that piques their interest.

Labels: Genetics

Matt Yglesias has a post up where he says:

I tried to explicitly say that I didn't want the story to be read as an analogy since I don't believe in trying to conduct arguments by analogy.

Analogies are sketchy, though useful when used well. There are two primary issues I think. First, analogies are a precise phenomenon, not a willy-nilly appeal to ad hoc affinities as they are used in many contexts. For example, people actually need to know something about one of the elements to obtain more insight about the other element. That is, you need to know something about the source to map information onto the target. So, for example, the analogy between Catholics:Protestant and Shia:Sunni is close useless because most people don't even have a detailed knowledge of Christianity and the differences between Catholicism and Protestantism to translate that structure to the case of Islam. Secondly, widespread recourse to analogies tends to breed sloppy thinking because of their imprecision. Analogies are good for "quick & dirty," but nothing really beats sharing a common base of data with those whom you are attempting to communicate with.

I have a piece up at Taki's Magazine, The Limits of Certitude. It might be read along with a post at ScienceBlogs, Science is rational; scientists are not.

Dan MacArthur points to a paper claiming that large allele frequency differences between populations are due to demographic effects. The data the authors are working with is a set of a few thousand markers (SNPs and others) genotyped on 53 populations from across the world. Their main points boil down to two things:

1. "Large" allele frequency differences are "surprisingly" common between human populations.

2. Such allele frequency differences are not enriched near genes (as would be expected if genes are more likely targets of positive selection than non-genic regions).

This work can essentially be seen as a push back against the trend towards finding "evidence" of positive selection in the human genome in any gene one finds interesting, and the authors cite a number of papers that fully or partially base their claims for selection on allele frequency differences between populations. As a warning about the caveats in such types of analysis, this is a useful paper, but it's important not to overstate what the data actually say:

1. When the authors say that large allele frequency differences are common, it's important to define "large". In this case, they're talking about things with an allele frequency difference of 0.3 or above. That is, if an allele is at 30% frequency in Africa and 60% frequency in Asia, that counts. How you define large is obviously subjective, and personally I wouldn't have chosen that threshold. But in any case, the authors are right to say that if you see an allele frequency difference of 0.3 and 0.4 between continental populations in your favorite gene, that alone is not strong evidence for selection.

2. The enrichment (or lack thereof) of large allele differences near genes was more comprehensively studied in a paper from about a year ago. The authors there found that there is indeed such an enrichment, but that it occurs at a more stringent definition of "large" than the one considered here. So the fact that allele frequency differences of 0.3 are not enriched near genes is not all that surprising.

To summarize, this paper shows that many claims about selection on individual loci based entirely on modest (what the authors call large) allele frequency differences between populations are massively overstating their evidence. But then again, you already knew that.

Labels: Population genetics

[Updated]

After declining pretty steadily from 1991 to 2005, in 2006 teen birth rates showed a slight uptick. Rather than swallow what the mass media and doomsaying blogosphere infers, read the report for yourself -- what you want to know is contained in the first 5 to 10 pages. Since most people worry about the long-term trend, and where things are going, I've taken data from the report's tables and put them into easy to understand time-series graphs, broken down by race and ethnicity. I'll then address a few of the larger issues.

All birth rates are live births per 1000 women in a given group. I'll only look at births to 15 - 17 year-olds because mothers younger than that are even rarer, and people freak out less about mothers at or above the age of majority. The 18 - 19 graphs look similar, and you can create them yourselves using the NCHS' report and Excel. Update: see the end of the post for the one 18 - 19 year-old graph that is different, which shows birth rates among Hispanic 18 - 19 year-olds increasing since 2000. [End update]

Most of the recent increase is due to 18 - 19 year-old births, so that's another reason not to care about an increase in "teen pregnancy" -- 18 and 19 year-olds are adults. Moreover, there is an increase across all age groups, especially 20 - 24. So, there's nothing special about teens of any age -- the 15 - 17 year-olds increased a bit, while the 18 and 19 year-olds appear to really be part of a larger group of 18 to 24 year-olds. (Nature doesn't adhere to our numbering system, where there's a bright line between 19 and 20.) Births are just up overall, and the closer we get to the female fecundity peak in the early 20s, the stronger the signal is.

First, the NCHS has data going back to 1970, although it is not broken down by race. Still, here is that graph:


There is a downward trend throughout, with a steady oscillation around that trend. So, the rate will probably continue to decline into the following decades, and we shouldn't be fooled by a temporary increase. For the near future, it looks like the rate will remain pretty flat for about 5 years, then start to increase again, with a decrease again after that, all on a downward trend.

Next, the data with race broken down begins in 1980 and includes White, Black, Asian / Pacific Islander, and American Indian / Alaskan Native. The graph with Hispanic ethnicity is further down. Here are the birth rates for Blacks and AIANs, and for Whites and Asians:




Since the graphs look so similar to the all-race graph, we can assume that from 1970 - '79, the birth rate declined across all races. Again, we see a downward trend with a steady oscillation around it. In no case is the 2006 uptick dramatic, and it looks just like it does in the all-races graph. So again, the rates will probably remain mostly flat for the next 5 years, increase, then decrease, following the overall downward trend.

Finally, the data on Hispanic and non-Hispanic groups starts only in 1990. Here are the trends:




The patterns are the same as in the other graphs, so it wouldn't be too risky to assume that, if we had the actual data, they'd look like they do in the first graph back to 1970. There are clear race and ethnic differences -- the Black line is always above the White line -- but the downward trend and presence of oscillations does not have to do with race or ethnicity. Whatever causes them is at a societal level. The only interesting difference in these Hispanic graphs is that non-Hispanic Blacks and Hispanics have switched in the rank order: we now have more of a Hispanic problem than a Black problem.

That should give people a clearer idea of what is going on. It's silly to forecast based on two data-points -- last year and this year -- when there is plenty of other data to help out. The reason people do this is because it allows them to indulge their desire to imagine the end of the world. The slightest aberration or reversal of a reassuring trend is greeted with exhilaration by the declinist junkies -- "So there's hope that the world is doomed after all!"

Furthermore, these data are all on birth rates, not pregnancy rates. We won't know what the teen pregnancy rate was until we have abortion data, and the NCHS says that won't be for awhile. (The most recent data are from 2004.) For all we know, pregnancy rates are the same or lower than before, but more of those who do get pregnant may opt to keep the kid.

I suspect something like that is true based on adolescent sex behavior data that I've already written about. All measures of sluttiness among teenage girls -- having sex at all, having 4+ partners, not using a condom, etc. -- have declined from 1991 to 2007. Indeed, there is no change in any of the measures from 2005 to 2007. See here for the data from the YRBS. This suggets that the 2006 uptick in birth rates was not due to greater rates of bad behavior among teenagers, but to a greater aversion to abortion among today's young people.

A statement by the National Campaign to Prevent Teen and Unplanned Pregnancy hopes for the worst about these YRBS data:

For teens in school, YRBS data for 2005 and 2007 reflected small increases in sexual activity and decreases in contraceptive use that were not statistically significant, but the changes are likely large enough to account for the 3% increase in the teen birth rate (and likely a similar increase in the teen pregnancy rate) between 2005 and 2006.

The part that I've boldfaced is mostly a lie. If you look at the YRBS data in the link above, there is an apparent but non-significant increase of 1.0% in the percent who have had sex, from 2005 to 2007. However, there was also an apparent increase of 1.1% from 2001 to 2003, and of 1.5% from 1997 to 1999 -- greater than the most recent apparent increase, yet which resulted in no uptick in teen births. As for percent who are currently sexually active, there is an apparent but non-significant increase of 1.1% from 2005 to 2007 -- but there were similar increases from 2001 to 2003 (up 0.9%) and 1997 to 1999 (up 1.5%). The same is true for percent who use birth control pills. Only the percent who used a condom the last time they had sex shows an apparent and non-significant decrease that isn't matched by similar apparent decreases in previous years. Again, though, the change is not significant.

Here's what the CEO of the National Campaign told USA Today:

Sarah Brown, CEO of the National Campaign to Prevent Teen and Unplanned Pregnancy, says she is less inclined to believe abortion is driving higher teen birth rates and suggests that increases in high-profile unmarried births in Hollywood, movies and even politics is a significant factor for impressionable teens.

"In the last couple of years, we had Jamie Lynn Spears. We had Juno and we had Bristol Palin. Those three were in 2007 and 2008 and not in 2005 to 2006, but they point to that phenomenon," she says.

You have to admire her candor and pity her desperation: she admits that the purported causes came years after the effect. So, rather than Hollywood influencing a social trend, the already existing social trend began to reach even into Hollywood. After that happens, there may be positive feedback, but let's be clear about who started it. In general, Hollywood doesn't want to influence any social trend -- they want to figure out what the existing trends are and pander to them to get rich. They're greedy capitalists, not mad scientists.

When I am elected dictator, my first campaign to improve the people's mental health will be to censor anyone who, in a mainstream forum, argues that 1) Hollywood is responsible for degrading our morals, 2) poverty causes crime, 3) "it's the parents' fault," or 4) these kids these days don't know what good music is. Ah, to read the newspaper without feeling the urge to strangle 5 of the 6 people quoted -- that oughtta make everyone feel a bit more sane.

Note: here's the graph for non-Hispanic Black and Hispanic 18 - 19 year-olds:

So as with the 15 - 17 year-old case, Hispanics have overtaken Blacks, and also that their rate has been increasing since 2000. This is the only worrisome data -- and yet another reason to slam our borders shut to anyone who didn't graduate college.

Labels: babes and hunks, Demographics, previous generations were more depraved

Of late I have come to believe that something close to the Transparent Society is inevitable in the near future. One of the primary complaints about the proliferation of surveillance and distribution of personal information is that it curtails privacy. But it came to my mind that perhaps this is another case of the the future looking much more like the past? After all, there is a common perception that in small villages and hunter-gatherer bands "everyone knew everyone else," and that individual privacy is an expectation only in the avenues toward anonymity which emerge in mass societies. The curve below shows is a stylized conception of what I have in mind.

The 10,000 Year Explosion: How Civilization Accelerated Human Evolution, has a website up, http://the10000yearexplosion.com. If there is an accelerating wave of media coverage that would probably be the place to track it....

Labels: 10000 Year Explosion

...well. I took a bit longer than I had intended in getting to work on this next post, but it's okay. When I left you hanging a few weeks ago, I mentioned that the idea of reinforcement (that is, selection for increased pre-mating isolation in the presence of post-mating isolation) was out of favor for a while, but it was brought back in a big way by a certain paper. I'll spend some time talking about that paper here.


So, of course, the paper I'm talking about is Coyne and Orr 1989 (open access). Maybe it's just because of the way I was taught speciation, but this paper arguably opened the flood gates for a lot of modern research on speciation. First of all, Coyne and Orr came up with a useful way to quantify reproductive isolation. This allowed them to undertake one of the first meta-analyses of speciation ever, and they came up with some rather interesting data. The most important stuff is in figure 5:



Figure 5 shows that among both "allopatric" and "sympatric" taxa, pre-zygotic isolation increases with time, but among sympatic taxa, it increases faster! This is in contrast to the rates of increase of post-zygotic isolation, which are the same between allopatric and sympatric taxa. This definitely argues in favor of a reinforcement hypothesis.

Anyway, there are many other important points to this paper, primarily that there is a correlation between genetic distance and reproductive isolation. This is precisely what we expect under the current model of speciation, which I will hopefully talk about in the future.

As an administrative note, I figured initially that I would do a series of posts, which would build on top of each other, but I'll probably put that off. I'll just be blogging about things I find that are interesting, which can come a lot quicker! Hopefully you'll see more of me in the near future.

This year is the bicentenary of the birth of Charles Darwin, and the 150th anniversary of the Origin of Species. Predictably, there will be a large number of commemorative events and publications. I found a useful list here. I suspect that by March we will all be heartily sick of CD, but I am still looking forward to the re-opening of Down House.

My immediate reason for posting is to draw attention to a series of programmes on BBC Radio 4 in the coming week, including four special editions of Razib's favourite 'In Our Time'. I assume that these will be available worldwide online. Incidentally, while checking the Radio 4 schedules I noticed that on Tuesday there is a documentary on the singer Nico, so it is a good week for the discerning minority who are fans of Charles Darwin and the Velvets.

This note concludes my discussion of R. A. Fisher's neglected treatment of human evolution in The Genetical Theory of Natural Selection. In Part 1, here, I pointed out that Fisher believed that human evolution was continuing rapidly in the present day. He also argued that natural selection among humans now operated mainly through differences in fertility rather than mortality. It was therefore important for Fisher to show that such differences in fertility were heritable.....


The Inheritance of Human Fertility

Chapter 9 of GTNS deals with this subject. Fisher first provides data to show that the variance in human fertility is far higher than would be expected by chance, that is, if offspring were allocated to parents by a random process (p.207-9. As previously, I will give page references to the Dover edition). Some individuals have fewer, and some more, offspring than can be expected on a chance basis. He remarks that the non-random element in human reproductive variance appears to be much higher than in most other organisms (apart from the social insects, where some individuals specialise in reproduction). Among humans differences in fertility are also more important than differences in mortality (p.213). Fisher then discusses the factors leading to such large differences in fertility, and suggests that they are mainly differences of temperament rather than physiology (p.209-13). Fisher believes these differences largely innate (p.210-11), making it plausible that they are to some extent heritable. He then considers direct evidence of the inheritance of fertility, and concludes that over 40% of the observed variance is due to heritable causes (p.217). Unfortunately, his evidence for this rests entirely on studies of the British aristocracy. Subsequent studies of wider samples have usually produced lower estimates of heritability, sometimes effectively zero. However, a recent survey of the literature by M. Murphy suggests that the correlation between fertility of parents and offspring is not negligible, and may have been increasing over time. But little is known about the extent to which this correlation is due to genetic rather than cultural factors.

Social Differences in Fertility

Chapter 10 produces evidence from several countries that under modern conditions fertility is inversely related to social class: the higher the class, the lower the fertility. His evidence is reasonably strong, so I need not elaborate on it. He also makes a number of interesting comments:

a) he disputes the common belief that the class differential in fertility is due to the earlier adoption of birth control by the higher classes, and that the differential will disappear as the use of birth control spreads (p.239). Here Fisher has been proved partly right (in Britain, at least): the differential has not disappeared, but it has narrowed with the availability of modern contraceptives, and especially the Pill. Moreover, for much of the 20th century there was a 'U-shaped' fertility pattern, with the highest and lowest social classes more fertile than the lower-middles.

b) Fisher believes that the inverse correlation of class and fertility is comparatively recent in the modern West, but that it also existed in ancient Greece and Rome (p.241).

c) the inverse correlation is 'unnatural' in the sense that it has to overcome natural obstacles. Ordinarily, we would expect the wealthier classes to be more fertile, because they could delegate much of the burden of parental care (p.242).

d) the fertility differential means that a large amount of upward social movement is needed just to maintain the proportions of different occupational groups. Fisher has some interesting discussion of the effects of this on social attitudes and values (p.243-5). But he probably overestimates the importance of differential fertility as compared with economic and technological development. There has been a huge increase in the proportion of non-manual jobs in modern economies, which cannot be explained by differential fertility.


The Social Selection of Fertility

To explain the social class fertility differential Fisher appeals to what he calls the 'social selection' of fertility (p.250-6). (He gives credit for the basic theory to the little-known eugenist J. A. Cobb.) The key point of the theory is that in modern societies social class is influenced both by natural ability and by the resources provided by parents to their offspring, such as paid education, jobs in family businesses, capital gifts and loans, influential social contacts, etc. Since the amount of resources available per child is greater when there are fewer children in the family, there is a social advantage in relative infertility. Given equal natural ability, children from small families are more likely to rise in social status (or avoid a decline) than those from large ones. The higher social classes will therefore become on average less fertile. Since marriage occurs mainly within social classes, the qualities correlated within each class (such as high abilities and low fertility in the upper classes) will become statistically and genetically linked. Fisher then proposes this as the main factor behind the decline of ruling classes and of civilisations (p.256-61). He discusses but rejects alternative explanations, and in particular Gobineau's theory that decline is due to racial mixture. Fisher points out that racial mixture increases genetic variance and therefore increases the intensity of natural selection, but whether this helps or harms the quality of the population (in 'virtue and ability') will depend on the prevailing conditions of selection. If they are unfavourable (dysgenic) racial mixture will accelerate the decline, but if they are favourable its long-term effects will be beneficial (p.257) despite possible short-term drawbacks.

Stages of Human History

Fisher argues that a negative correlation of social class and fertility will occur whenever social conditions are similar to those in modern western society; broadly, whenever society is based on 'individuals co-operating for mutual advantage in a state of law and order' (p.261). But many societies are not of this kind. Fisher particularly discusses what he calls (in a non-pejorative sense) 'barbarian' societies, such as those of the Homeric poems, where there is little central government and law, and social life is based on kinship and the institution of the blood feud. In such societies fertility is a positive social advantage, and infertility a drawback (p.261-4). The qualities recognised by the society as valuable therefore become positively correlated with fertility (p.264). Fisher considers this form of 'social selection' far more powerful in promoting 'the higher human faculties', such as aesthetic appreciation, than either individual or group selection (p.264). He then has a fascinating section on 'Heroism and the higher human faculties', in which he give a major role to sexual selection. Unlike some more recent writers, such as Geoffrey Miller, Fisher recognises that marriage choices in such a society are made not by individuals but by families: 'The prestige of the contracting parties is all-important, and while this is partly personal, it is also largely tribal' (p.266). Sexual selection therefore reinforces the advantages of such socially valued attributes as heroism, even beyond the point at which they are directly beneficial (p.266); an example of Fisher's famous 'runaway' process. Fisher himself summarises his theory as follows:

To summarize the points of anthropological importance: (i) a barbarian people organized in kindred groups and recognizing the blood feud as the principle of social cohesion, can scarcely fail to experience a selection in favour of two qualities on which the success of the kindred group principally depends (a) the public spirited, patriotic, or heroic disposition (b) fertility. (ii) The stratification of society in these two qualities implies a selective advantage of the heroic temperament beyond the optimum advantage ascribable to prudent boldness, by reason of the social advantage of fame or heroic reputation. (iii) The power of recognizing the heroic qualities, and of conscious choice in intermarriage, introduces the dual effect of sexual selection in intensifying both the qualities selected and the communal recognition and appreciation of such qualities. (iv) This selection of the popular emotional response to the heroic qualities has the important effects of (a) stabilizing the foundations of the system by strengthening the existing basis of social cohesion, (b) intensifying the selective advantage ascribable to fame or prestige, (c) increasing the selective advantage of all qualities consciously envisaged in sexual selection, (d) exaggerating the realities of natural inequality by the development of an extreme aristocratic doctrine of hereditary nobility.(p.268)

Overall, Fisher's theory of human evolution is subtle and ingenious, but often speculative. The evidence for some of his key propositions, such as the high heritability of fertility, is painfully thin. Nevertheless, Fisher's ideas are always intriguing, and even his wildest speculations are well worth reading. Indeed, although much of this part of GTNS inevitably seems dated, in some respects it still compares favourably with more recent treatments of human evolution. Along with Darwin's Descent of Man, and Westermarck's Origin and Development of the Moral Ideas, I believe it should have a place among those few 'classics' that are still capable of stimulating modern research on the subject.

One of the established cool examples of convergent evolution (which for my purposes here I'll define loosely here as the evolution of different populations to the same phenotype via different mutations) has been the repeated loss of pigmentation (and eyes) in fish that have adapted to life in light-poor, nutrient-poor caves. In 2006, a group reported that albinism (panel J in the picture) in several of these caves was due to mutations in OCA2 (a SNP in a regulatory region of this gene also causes blue eyes in humans).

Not all cavefish however, are fully albino--in some populations, there also exists a "brown" phenotype (panel "G" in the picture) with reduced pigmentation. In a new paper, the gene underlying this phenotype is shown to be MC1R (this gene, of course influences pigmentation in all sorts of species), and, similarly to OCA2, two different mutations have arisen in different populations.

One might imagine that light pigmentation in cavefish could just be due to simple drift--a random mutation that knocks out pigmentation is no longer selected against in a place where there's little light, and so could drift up to high frequency. But the fact that this phenotype has arisen so many times, and reached high frequency in the presumably short time period that these fish populations have been isolated (I say presumably short because I can't find any numbers on this, but the different populations can interbreed freely) suggests a role for strong positive selection for this phenotype in adaptation to the cave environment.

Labels: Evolution, Genetics

Everyone & their mother is emailing me about Jonathan Haidt's new commentary in Edge, FASTER EVOLUTION MEANS MORE ETHNIC DIFFERENCES:

I believe that the "Bell Curve" wars of the 1990s, over race differences in intelligence, will seem genteel and short-lived compared to the coming arguments over ethnic differences in moralized traits. I predict that this "war" will break out between 2012 and 2017.

There are reasons to hope that we'll ultimately reach a consensus that does not aid and abet racism. I expect that dozens or hundreds of ethnic differences will be found, so that any group - like any person -Ā can be said to have many strengths and a few weaknesses, all of which are context-dependent. Furthermore, these cross-group differences are likely to be small when compared to the enormous variation within ethnic groups and the enormous and obvious effects of cultural learning. But whatever consensus we ultimately reach, the ways in which we now think about genes, groups, evolution and ethnicity will be radically changed by the unstoppable progress of the human genome project.

Yes, psychopathy might have adaptive "strengths" in a frequency dependent context, but I don't think that's what Haidt meant! One difference with the IQ wars when it comes to personality is that it seems every single dopamine receptor has already been implicated in behavior genetic variation, while we're still a long way from IQ loci results which have been reproduced, though one might double-check on the details of the statistical analyses on suggestive findings from behavior genetics. In any case, since the heritability of behavior in economic games has already been established, it would be interesting if GWAs found some loci which tracked the variation. My own hunch is that personality variation is less continuous than IQ (characterized by a few morphs hanging around fitness peaks), with an underlying architecture of larger effect QTLs. Perhaps altruism is just way simpler to modulate than general intelligence?

Labels: human biodiversity