Assortative mating and the far tail of intelligence

A Fermi estimate of the increase in the tail population due to elitism, mobility and assortative mating. My conclusion is that there has been a significant increase in the number of kids in the far tail relative to, say, just one or two generations ago. I think my assumptions below are fairly realistic. (Note to non-scientists: this is only an order of magnitude calculation!)

My original estimate was off -- see erratum 2 below and this updated post

US birth cohort roughly 4 million.

Roughly few hundred kids at +4 SD ability (taking into account an excess over the normal distribution prediction; past assortative mating is probably one of the main causes of this excess ;-). What are these kids like? See here or here.

Number of +3 SD graduates per year from elite universities (average at these universities is above +2 SD): 10,000
(Note: a significant chunk of the roughly 40k +3 SD population!)

Plausible additional number of +3 SD / +3 SD marriages due to elitism and assortative mating, relative to one or two generations ago: + 1000 per annum

Number of additional +4 SD children produced, assuming, e.g., heritability of .8 (so, +4 SD kids are +1.6 SD from parental midpoint adjusted for regression): + 100 per annum

This ignores any environmental advantage to a child from having two +3 SD parents; but that effect might be small.

This enhancement of the tail population doesn't imply anything about the average intelligence of the overall population -- it is due to a concentration of good genes in a small sub-population that has been filtered by psychometric exams like the SAT, GRE and LSAT. I suspect that immigration (e.g., foreign graduate students at top programs, engineers coming to Silicon Valley, etc.) produces an effect of similar or even larger size.

Erratum 1: I kept screwing up the most important part of the calculation: what the mean IQ of children of +3/+3 SD parents would be. (See comments below from Carson Chow and Henry Harpending.)

Rather than continuing to try to do it in my head I looked up the result in Gillespie's Population Genetics (p.113) and the result is that the correlation between offspring and midparent is h^2 = heritability.

If the parental midpoint is +3 SD, then the children are distributed around an average of +3 h^2.

I had originally assumed h^2 = .8 which leads to +3 (.8) = +2.4, which requires a +1.6 fluctuation to produce a +4 kid. The odds of that are about 1/20.

If heritability is much lower, like .5, then it requires a +2.5 fluctuation to get a +4 SD kid and the effect of assortative mating is small.

Note, I have also been assuming that the residual variance among the +3/+3 kids is the full 15 point SD of the general population distribution. Probably the variance should be lower, but this requires us to go beyond the simplest models of additive genetic variance. ...


Erratum 2: After a bit more research, I would guess that the additive portion of genetic variance is no more than .5 or so. The high values of heritability typically quoted (e.g., .8) are broad sense heritabilities which include both additive and non-additive variance. The non-additive part results from dominance, gene interactions, etc. and does not allow a simple prediction of offspring qualities from the parental midpoint. So the result given above for h^2 = .5 is probably the largest effect we can expect from assortative mating, and it is not as large as I originally thought. A second point is that the residual variance once parental midpoint is fixed (i.e., exhibited by siblings) is probably smaller than 15 points. If narrow sense (additive) heritability is .5 then the remaining variability would have SD of about 11 points. This further reduces the chances of getting a +4 SD kid.

For parental midpoint n (in SD units), the probability of a child who exceeds this midpoint is

1 - erf( n sqrt ( 1 - h^2) )

where h^2 is the narrow sense or additive heritability. To see this, note that the mean among the offspring would be n h^2 , so the required upward fluctuation is n ( 1 - h^2 ) , but relative to the remaining variance, which has SD = sqrt ( 1- h^2 ). The probability of a child surpassing superior parents gets smaller the more exceptional the parents, however the probability is much higher than for a random child selected from the general population: 1 - erf(n). This would also be the result for any choice of parents if narrow sense heritability is zero. On the other hand if h^2 = 1 then all children would equal their parental midpoint: 1 - erf(0) = 1.

Note, throughout this discussion I have assumed no environmental boost from being raised by exceptional parents.

Humans interbred with Neanderthals

The papers from Paabo's group are available at the Science web site:

Special Feature

News focus

A Draft Sequence of the Neandertal Genome

Targeted Investigation of the Neandertal Genome by Array-Based Sequence Capture.

They describe the results of a shotgun sequencing of several Neanderthal genomes, which provide strong evidence for introgression of Neanderthal DNA into the human lineage from interbreeding. From the overview:

... Substantial controversy surrounds the question of whether Neandertals interbred with modern humans. To address this question, Green et al. tested whether Neandertals are more closely related to some present-day humans than to others.* Because modern humans are believed to have originated in Africa, if Neandertals diverged from modern humans before present-day populations began to differentiate, one would expect Neandertal sequences to match sequences from non-Africans and Africans to the same extent. Unexpectedly, the researchers found that Neandertals share more genetic variants with present-day non-Africans than with Africans. These results can be explained if gene flow occurred from Neandertals into the ancestors of non-Africans.

The observation that the Neandertal genome appears as closely related to the genome of a Chinese and a Papua New Guinean individual as to the genome of a French individual is particularly surprising as there is, to date, no fossil evidence that Neandertals existed in East Asia or Papua New Guinea. Green et al. thus suggest that gene flow between Neandertals and modern humans occurred prior to the divergence of European and Asian populations. Based on comparative genomic data, as well as a mathematical model of gene flow, the authors further estimate that between 1 and 4% of the genomes of people in Eurasia may be derived from Neandertals.

... Using this comparative approach, Green et al. came up with a list of 20 candidate regions that may have been affected by positive selection in ancestral modern humans. Five of these regions contain no protein-coding genes and may thus include structural or regulatory elements. Among the remaining 15 regions, the team identified genes involved in metabolism and cognitive and cranial development, which suggests that aspects of these processes may have been functionally important for the evolution of modern humans. [This is evidence for human evolution due to selection in the time since humans diverged from Neandertals about 300ky ago.]

The presence of Neanderthal DNA in some but not all modern human populations implies divergent evolution between groups. Needless to say, this is one of the biggest scientific results in human evolutionary history in some time.

More:

... The team measured the genetic proximity of Neandertals to pairs of modern humans from different continents, first using single-nucleotide polymorphisms (SNPs), or sites in the genome where a single nucleotide differs between individuals. When they compared a Neandertal with a European and an Asian, they found that the Neandertal always shared the same amount of derived (or more recently evolved) SNPs with each of them. But when they compared a Neandertal with an African and a European, or with an African and an Asian, the Neandertal always shared more SNPs with the European or Asian than with the African. "We've shown that Neandertals are significantly more closely related to non-Africans than Africans on average," says Reich.

Even though they looked at just two Africans for this part of the study, those two have a particularly ancient, diverse heritage, so they are a good proxy for much of the genetic diversity in Africa. But sequencing additional Africans would be a good idea, says Reich.

For now, it seems Neandertals interbred with the ancestors of Europeans and Asians, but not with the ancestors of Africans. At first, "we were baffled that this affinity with Neandertals was not only in Europe and West Asia [where it was most expected], but also in Papua New Guinea" where Neandertals never set foot, says Pääbo.

See earlier post for more background.

Also see John Hawks for in-depth analysis, including the following. In earlier work (see also here), Hawks, Wang, Cochran, Harpending and Moyzis argued that the rate of human evolution has sped up in the last 10ky or so. If their estimates are correct, the amount of change in the last 10ky may be greater than what occurred over most of the 300ky since modern humans diverged from Neanderthals.

... Green and colleagues did a similar exercise, except they went looking for "selective sweeps" in the ancestors of today's' humans. These are regions of the genome that have an unusually low amount of incomplete lineage sorting with Neandertals, and therefore represent shallow genealogies for all living people. They identify 212 regions that seem to be new selected genes present in humans and not in Neandertals. This number is probably fairly close to the real number of selected changes in the ancestry of modern humans, because it includes non-coding changes that might have been selected.

Again, that's really a small number. We have roughly 200,000-300,000 years for these to have occurred on the human lineage -- after the inferred population divergence with Neandertals, but early enough that one of these selected genes could reach fixation in the expanding and dispersing human population. That makes roughly one selected substitution per 1000 years.

Which is more or less the rate that we infer by comparing humans and chimpanzees. What this means is simple: The origin of modern humans was nothing special, in adaptive terms. To the extent that we can see adaptive genetic changes, they happened at the basic long-term rate that they happened during the rest of our evolution.

Now from my perspective, this means something even more interesting. In our earlier work, we inferred a recent acceleration of human evolution from living human populations. That is a measure of the number of new selected mutations that have arisen very recently, within the last 40,000 years. And most of those happened within the past 10,000 years.

In that short time period, more than a couple thousand selected changes arose in the different human populations we surveyed. We demonstrated that this was a genuine acceleration, because it is much higher than the rate that could have occurred across human evolution, from the human-chimpanzee ancestor.

What we now know is that this is a genuine acceleration compared to the evolution of modern humans, within the last couple hundred thousand years.

Our recent evolution, after the dispersal of human populations across the world, was much faster than the evolution of Late Pleistocene populations. In adaptive terms, it is really true -- we're more different from early "modern" humans today, than they were from Neandertals. Possibly many times more different.

Neanderthal-human introgression?

Rumors abound concerning results from the sequencing of Neanderthal DNA by Svante Paabo's group at MPI. More here and here from Nature. Will Paabo's group announce candidate alleles that may have introgressed into the human gene pool from Neanderthals? FOXP2?

Because Neanderthals lived mainly in Europe and Asia, a signal for introgression would be matching regions of DNA in Neanderthals and Eurasians that are not found in sub-Saharan Africans. While one commonly thinks of Neanderthals as primitive, they almost certainly had useful adaptations not already present in the small gene pool of modern humans, newly arrived from Africa. (Neanderthals had been in Europe and Asia for roughly 300k years before modern humans.) Adaptive or selectively neutral Neanderthal genes are more likely to have introgressed into humans than non-adaptive ones (see below).

2009 interview: ... We are analyzing this and it's a very difficult analysis, because if there are tiny parts of a genome that were contributed to humans, you have to assume that they are not due to some error in our analysis, or contamination of the DNA of modern humans, or any little bias in the algorithms we use.

In the public media we're pretty much depicted as saying there was absolutely no mixture. It's very hard to convey these subtle messages to the public. If you read our papers, we say very carefully that there is absolute proof that they didn't contribute mitochondrial DNA. That doesn't mean that they couldn't have contributed other parts of their genome.

... It's very clear from the genetic evidence that the big picture is the out of Africa one. That's not to say that there couldn't have been a small contribution from earlier archaic forms such as Neanderthals in Europe to present-day Europeans or erectus forms in Asia to Asians, but it has to be very small. ...

But the Neanderthal genome will be a chance to address that. And it depends on what you're interested in. As a geneticist I'm not so interested with who had sex with whom 30,000 years ago. The question is, as a geneticist today: did Neanderthals contribute significantly to our gene pool today? Did the Neanderthals have an impact on the variance we carry? And that's got to be small.

For a plausible argument for introgression of genes from Neanderthals to humans, see The 10,000 Year Explosion, by G. Cochran and H. Harpending:

... Logically, if admixture occurred at all, it had to happen somewhere in Neanderthal-occupied territory, which means Europe and western Asia. As modern humans expanded their territory, they must have encountered Neanderthal bands again and again. The two kinds of humans coexisted for a few thousand years before the Neanderthals disappeared, at least in some regions. This looks to be the case for the Châtelperronian culture of France and northern Spain, and there are traces of a similar culture in Italy. If there was trade, or if there was enough contact to transmit toolmaking techniques, there was sexual contact as well—depend on it. If in the future we look at very large genetic datasets from huge numbers of individuals, we might find a few traces of neutral Neanderthal genes.14

... Imagine that humans occasionally mated with Neanderthals, and that at least some of their offspring were incorporated into human populations. That process would have introduced new gene variants, new alleles, into the human population. Many, probably most, of those alleles would have done almost exactly the same thing as their equivalents in modern out-of-Africa humans; they would have been neither better nor worse than those equivalents—in other words, they would have been selectively neutral. Those neutral alleles from Neanderthals would have been rare, and they would probably have disappeared, the typical fate of rare neutral alleles.

... J. B. S. Haldane, the great British geneticist (1892–1964), found a systematic way of adding up all these probabilities, and his method yields a surprisingly simple answer. If the allele confers an advantage s, its chance of going all the way is 2s. In a stable population, a single copy of an allele with a 10 percent fitness advantage has a 20 percent chance of eventually becoming universal. The fate of one copy of a favorable allele is very much like that of a gambler who starts out with one chip and a roulette system—a way of beating the odds—that really works. ...

The price of eggs

Eggs purchased from female donors at elite universities are several times more expensive.



From Aaron D. Levine, "Self-Regulation, Compensation, and the Ethical Recruitment of Oocyte Donors," Hastings Center Report 40, no 2 (2010): 25-36. More.

Random microworlds: the mystery of nonshared environment

One of the most mysterious aspects of the nature-nurture question is the difficulty in characterizing the nurture component.

Turkheimer and Waldron: When genetic similarity is controlled, siblings often appear no more alike than individuals selected at random from the population. ... it has become widely accepted that the source of this dissimilarity is a variance component called nonshared environment.

... In what may have been the most influential article ever written in the field of developmental behavior genetics, Plomin and Daniels (1987) reviewed evidence that a substantial portion of the variability in behavioral outcomes could not be explained by the additive effects of genotype or the environmental influences of families. They suggested that this residual term, which they called the nonshared environment, had been neglected by environmentally oriented researchers who assumed that the most important mechanisms of environmental action involved familial variables, like socioeconomic status [SES] and parenting styles, that are shared by siblings raised in the same home and serve to make siblings more similar to each other. Indeed, Plomin and Daniels argued, once genetic relatedness has been taken into account, siblings seem to be hardly more similar than children chosen at random from the population.

In other words, despite a lifetime of proximity, your adopted child may bear no more similarity to you (in terms of, e.g., intelligence) than someone selected at random from the general population. The shared family environment that your children (biological or adopted) experience has little or no measurable effect on their cognitive development. While there are environmental effects on intelligence (the highest estimates of heritability for adult IQ are around .8, and some would argue for a lower value; see here for Turkheimer's work suggesting low heritability in the case of severe deprivation), they seem to be idiosyncratic factors that can't be characterized using observable parameters such as the parents' SES, parenting style, level of education, or IQ. It is as if each child experiences their own random micro-environment, independent of these parental or family characteristics.

The nonshared influences are by far the largest environmental (non-genetic) influences on intelligence -- in fact, they are the only detectable non-genetic influences. (Click figure for larger version; from a review by Plomin. More recent overview here.)



Identical twins, whether raised together or apart, turn out to be very similar, but one still finds differences in IQ and personality. The cause of those differences must be the different environments experienced by the twins, but can't be characterized by simple variables of the sort listed above: it is not the case that the twin raised by the higher SES family has, on average, a much higher IQ! In fact, twins raised in the same family are about as similar as those raised apart, so family shared environment does not produce a large measurable influence. See below for a plausible model that accounts for such outcomes.

By now these results are well understood and accepted by experts, but not by the general population or even policy makers. (See the work of Judith Rich Harris for popular exposition). The naive and still widely held expectation is that, e.g., high SES causes a good learning environment, leading to positive outcomes for children raised in such environments. However, the data suggests that what is really being passed on to the children is the genes of the parent, which are mainly responsible for, e.g., above average IQ outcomes in high SES homes (surprise! high SES parents actually have better genes, on average). Little or no positive effect can be traced to the SES variable for adopted children.

The implications are quite shocking, especially for two groups: high investment parents (because the ability of parents to influence their child's development appears limited) and egalitarians (because the importance of genes and the difficulty in controlling environmental effects seem to support the Social Darwinist position widely held in the previous century).

It is plausible to me that each child tends to create their own environment over time, by selectively seeking out or avoiding stimuli of various types. A bookish kid may end up at the library regardless of whether their father takes them there. An athletic kid may end up on the playground whether or not their mother takes them there. It has been argued that this effect is the reason that the heritability of IQ increases with age: over time, genetic influences assume greater importance as they cause the individual to create or seek out their preferred environment.

In a previous post I discussed individual cognitive profiles as described by an n-vector. Similarly, one could think of an individual's learning profile and learning environment as two more n-vectors. These n-vectors may or may not be well-matched, leading to outcomes with significant and hard to characterize variability. For example, one can imagine that both the environment (provided by parents, siblings, teachers and peers) and a particular child's reactions vary in each of the factors listed below.

Pressure and competition

Stimulation through stories and pretend play; flights of imagination

Ability to learn from repetition and drill / tendency to boredom

Isolated study vs group activities

Visual vs aural vs mechanical stimulation

Level of discipline or structure imposed

Close mentoring vs freedom of exploration

Abstraction vs experimentation

(One can think of many more.)

The factors listed are not intrinsically good or bad for learning -- what matters is whether the learning environment is matched to the nature of the individual child. Some react well to discipline or pressure or story telling, others do not. Further, none of the factors is obviously correlated with SES, parental education level or IQ. Even if they were, it's plausible that a child to some extent creates their learning environment outside the control of parents and teachers (e.g., through peer group or choice of play activities).

An individual whose learning vector (learning style) is well matched to their environment will thrive: the nonshared environmental component in their development will be large and positive. For others, the environment will have a smaller or even negative impact. Because both the learning vector and the environment vector vary in a many-dimensional space, and over time, prediction or control of the overall environmental effect on development is difficult.

Nonshared environmental contributions to development, which are the largest environmental contributions, are effectively random. They are not amenable to control, either by parents or policy makers. Note, this picture -- that each child creates their own environment, or experiences an effectively random one -- does not seem to support the hypothesis that observed group differences in cognitive ability are primarily of non-genetic origin. Nor does it suggest that any simple intervention (for example, equalizing average SES levels) will eliminate group differences. However, it's fair to say our understanding of these complex questions is limited.

Technical remark: if n is large, and factors uncorrelated, the observed environmental variation in a population will be suppressed as n^{-1/2} relative to the maximum environmental effect. That means that the best or worst case scenarios for environmental effect, although hard to achieve, could be surprisingly large. In other words, if the environment is perfectly suited to the child, there could be an anomalously large non-genetic effect, relative to the variance observed in the population as a whole. Of course, for large n these perfect conditions are also harder to arrange. (As a super-high investment parent I am actually involved in attempting to fine tune n-vectors ;-)

Environmental effects cause regression to the mean of a child relative to the parental midpoint. Parents who are well above average likely benefited from a good match between their environment and individual proclivities, as well as from good genes. This match is difficult to replicate for their children -- only genes are passed on with certainty.

Gene machines

Ever wonder how automated genetic sequencing works?

The blog post linked to below gives a detailed description of the Illumina platform, complete with illustrations.

Genetic-Inference: ... Second gen sequencing has allowed us to re-sequence many human individuals; the 1000 Genomes Project is using 454, SOLiD and Illumina machines to sequence hundreds of individuals. This sort of thing has allowed us to get an idea, not just of what a single genome looks like, but how the genome changes from person to person; we can look at how much variation there is in the genome, how different populations differ in their genome structure, and even what makes a cancer genome different from a healthy genome.

However, second gen sequencing is not without its flaws. While it has got cheap ($40-50 to sequence a human genome these days) [typo? I think he means $40-50k], it still requires a lot of reagents, a lot of work and a lot of cost (the RT-bases aren’t cheap, and neither are the enzymes used). The low read lengths are still a problem, as they make knowing precisely where you got the bit of DNA from hard to discover (especially if you want to know which of two paired chromosomes it came from), and it takes a very long time to run the machines to completion. In the third part of this series, I will take about the new technologies that are appearing on the horizon; the so-called Third Generation Sequencing machines, which promise to make sequencing an entire human genome cost a couple of hundred pounds and take a few hours.

Last summer at Scifoo I met Jonathan Rothberg, founder of 454 Life Sciences, which makes one of the 2nd generation machines. 454 machines did James Watson's genome and also the neanderthal sequencing for Svante Paabo's group at MPI. We had a pretty interesting conversation, ranging from technology and venture capital to genetics and human biodiversity :-)

Back in 2005 I posted the figure below showing the super-exponential growth rate of sequencing capability. It looks like progress has continued at a similar rate since then.

Geeky scientists and engineers fueled by government grants and venture capital have once again delivered promethean gains to society :-) So much of progress, scientific and otherwise, is driven by the nature of available tools; see related post Prometheus in the basement.

It's all in da gene: sleep

I could really use this mutation. It seems I naturally need 8+ hours of sleep -- very inconvenient!

Related posts: All in da gene , mutants

"Horses ain't like people, man, they can't make themselves better than they're born. See, with a horse, it's all in the gene. It's the fucking gene that does the running. The horse has got absolutely nothing to do with it." --- Paulie (Eric Roberts) in The Pope of Greenwich Village.

ScienceNOW:

Early Risers Are Mutants

Don't hate those people who are perky and efficient after only a few hours of sleep. They can't help it. New research suggests that a genetic mutation may explain why some people sleep less.

Researchers don't know exactly why some people do fine with as little as 4 hours of sleep a night, while others need 12. "We've believed for a long time that there's a genetic basis," says Paul Shaw, a neurobiologist at Washington University in St. Louis, Missouri. But scientists have only recently begun to ferret out which genes are responsible.

In 2001, geneticist Ying-Hui Fu and colleagues identified a mutation in a gene called Per2 that appeared to cause familial advanced sleep-phase syndrome (FASPS). People who have this condition sleep a normal 8 hours, but they go to bed earlier than most people, retiring at 6 or 7 in the evening and waking at 3 or 4 in the morning. "After that was published, a lot of these people [with unusual sleep schedules] came to us," says Fu, who is now at Stanford University in Palo Alto, California. "So we started to collect DNA samples." The team now has genetic information from more than 60 families.

Fu and her colleagues have spent the past several years mining this vast genetic storehouse for more mutations that might affect sleep patterns. In 2005, they uncovered another mutation associated with FASPS. And now they say they have found the first genetic mutation in humans that appears to affect sleep duration rather than sleep timing. The mutation lies in DEC2, a gene that codes for a protein that helps turn off expression of other genes, including some that control circadian rhythm, the internal clock that regulates a person's sleep-wake cycle. The mutation occurred in just two people, a mother and her daughter. The women sleep an average of only 6.25 hours, whereas the rest of the family members sleep a more typical 8 hours.

To confirm that this mutation shortens sleep, Fu and colleagues engineered mice to carry the mutant form of DEC2. The mutant mice slept about an hour less than normal mice, the team reports today in Science. The finding also held for fruit flies: Mutant flies slept about 2 hours less than normal flies.

DEC2 likely isn't the whole story when it comes to short sleep. "Genetic control of sleep is going to be complex and is going to include multiple types of genes," says Shaw, who was not affiliated with the study. But that doesn't diminish the importance of this paper, he notes. "It's really an amazing piece of work."

The findings, says Fu, could lead to better treatments for sleep disorders. If the mutated form of DEC2 were available in a pill, Fu says she'd take it, noting that she needs about 8 hours of shuteye a night to feel rested. "All my life I've wanted to be able to sleep less."

Science:

The Transcriptional Repressor DEC2 Regulates Sleep Length in Mammals

Ying He,1 Christopher R. Jones,2 Nobuhiro Fujiki,3 Ying Xu,1,* Bin Guo,4 Jimmy L. Holder, Jr.,1,{dagger} Moritz J. Rossner,5 Seiji Nishino,3 Ying-Hui Fu1,{ddagger}

Sleep deprivation can impair human health and performance. Habitual total sleep time and homeostatic sleep response to sleep deprivation are quantitative traits in humans. Genetic loci for these traits have been identified in model organisms, but none of these potential animal models have a corresponding human genotype and phenotype. We have identified a mutation in a transcriptional repressor (hDEC2-P385R) that is associated with a human short sleep phenotype. Activity profiles and sleep recordings of transgenic mice carrying this mutation showed increased vigilance time and less sleep time than control mice in a zeitgeber time– and sleep deprivation–dependent manner. These mice represent a model of human sleep homeostasis that provides an opportunity to probe the effect of sleep on human physical and mental health.

Thomas Bouchard against group think and political correctness

Psychologist Thomas Bouchard, a leader in the field of twin and adoption studies, is retiring from the University of Minnesota. (Some Bouchard papers from Google Scholar.) He gave this farewell interview to Science. (See also this 2004 video; I think the interviewer is David Lubinski.)

Important components of today's conventional wisdom will turn out to be completely wrong. Analyze any past period and you'll find that this statement turned out to be true.

Q: What got you into twin studies?

TB: I was teaching the psychology of individual differences, and in 1979, two different people put a copy in my mailbox of a story about twins reared apart and their similarities when they met. [These were the "Jim twins," Jim Springer and Jim Lewis, who had been separated at birth and reunited at age 39. Both married women named Linda, divorced, and remarried women named Betty. They named their sons James Allan and James Alan, respectively, and both had dogs named Toy.] They sounded interesting, so I asked a few of my colleagues to help me study them. We ended up studying twins reared apart—126 pairs including 74 pairs of identical twins—for 20 years. [The twin study wound down in 2000.] I found that I loved working with twins. They're still amazing and a major mystery to me.

Q: What were attitudes toward behavioral genetics in the early years of your career?

TB: In graduate school at UC [the University of California] Berkeley, I was reading a book edited by psychiatrist D. D. Jackson on the etiology of schizophrenia. The first chapter, by a geneticist, was on twin studies. Then Jackson refuted it all with just the kind of crap you hear now against twin studies. He said families are the cause of schizophrenia. I remember saying in a graduate seminar, "Most of this stuff [in Jackson's argument] is junk"—I crawled out of the seminar room a bloody pulp. The reaction [from seminar members] was my first absolutely clear-cut demonstration that psychologists believed correlation is causation, ... and many still do.

In the '70s, when I was teaching research by [IQ researcher Arthur] Jensen and [twin researcher Francis] Galton, people picketed me, called me a racist, tried to get me fired. The progressive student association sent members in to ask hostile questions. ... So I put a tape recorder on the podium and said: "I'm going to tape my lectures." I never heard from them again. They knew what they were saying was nonsense and I would be able to prove it.

Q: Do you think perceptions have changed dramatically since the '70s now that twin research has revealed genetic bases for many disorders, such as autism (which had been blamed on cold mothers) and ADHD (for which many blamed food dyes)?

TB: Within the university—at least at U. Minnesota—the cumulative impact of behavioral genetics findings has had a lot of effect. There's a lot more tolerance for the idea of genetic influences in individual differences.

But we still have whole domains we can't talk about. One of the great dangers in the psychology of individual differences is self-censorship. For example, when I was a student, it was widely accepted that black self-esteem was much lower than white self-esteem, and that was a cause of differences in achievement between the two groups. Now that's been completely overturned—there is virtually no racial difference in self-esteem. But people had enormous amounts of data [showing this] that they didn't publish because it did not fit the prevailing belief system. How much wasted effort was generated by the flawed self-esteem work as an explanation of the black-white IQ difference? Now a days, I'm sure there are people who are not publishing stuff on sex differences. Look what happened to Larry Summers [who resigned as president of Harvard University after suggesting that discrimina tion alone doesn't account for women's lower representation in math-based disciplines]. I talk about those things in my class all the time—that males and females have different interests; ... in a sense, females have a broader and richer view of life. There are a lot of people who simply won't talk about those things. Academics, like teenagers, sometimes don't have any sense regarding the degree to which they are conformists.

Nicholas Wade also writes about this in a guest post on the NYTimes TiernyLab blog:

... You’re an expert because all your peers recognize you as such. But if you start to get too far out of line with what your peers believe, they will look at you askance and start to withdraw the informal title of “expert” they have implicitly bestowed on you. Then you’ll bear the less comfortable label of “maverick,” which is only a few stops short of “scapegoat” or “pariah.”

A remarkable first-hand description of this phenomenon was provided a few months ago by the economist Robert Shiller, co-inventor of the Case-Shiller house price index. Dr. Shiller was concerned about what he saw as an impending house price bubble when he served as an adviser to the Federal Reserve Bank of New York up until 2004.

So why didn’t he burst his lungs warning about the impending collapse of the housing market? “In my position on the panel, I felt the need to use restraint,” he relates. “While I warned about the bubbles I believed were developing in the stock and housing markets, I did so very gently, and felt vulnerable expressing such quirky views. Deviating too far from consensus leaves one feeling potentially ostracized from the group, with the risk that one may be terminated.”

... If the brightest minds on Wall Street got suckered by group-think into believing house prices would never fall, what other policies founded on consensus wisdom could be waiting to come unraveled? Global warming, you say? You mean it might be harder to model climate change 20 years ahead than house prices 5 years ahead? Surely not – how could so many climatologists be wrong?

What’s wrong with consensuses is not the establishment of a majority view, which is necessary and legitimate, but the silencing of skeptics. “We still have whole domains we can’t talk about,” Dr. Bouchard said, referring to the psychology of differences between races and sexes.

Look for Wade to take a huge beating from climate conformists.

Genetic clustering: 40 years of progress

Represent each individual human by their DNA sequence. When aggregated, they cluster into readily identifiable groups. This has been known for 40 years now, although the technology and methods of analysis continue to improve. Below are results from 1966, 1978 and 2008.

If this seems counterintuitive to you, it might be because the space of genetic variation is of very high dimension. See here for more discussion and an illustration.

(Click images for larger version.)

Population Structure and Human Evolution
L. L. Cavalli-Sforza

Proceedings of the Royal Society of London. Series B, Biological Sciences, Vol. 164, No.995, A Symposium from Mendel's Factors to the Genetic Code (Mar. 22, 1966), pp. 362-379
http://www.jstor.org/stable/pdfplus/75457.pdf




Measurement of Differentiation: Reply to Lewontin, Powell, and Taylor
Jeffry B. Mitton

The American Naturalist, Vol. 112, No. 988 (Nov. - Dec., 1978), pp. 1142-1144
http://www.jstor.org/stable/2460361?origin=JSTOR-
pdf




Current state of the art, as discussed here. Figure: The three clusters shown below are European (top, green + red), Nigerian (light blue) and E. Asian (purple + blue).



According to the mathematical analysis given in this paper, populations with FST as low as .0001 can be resolved with current technology. (Typical FST between northern and southern Europe is about .006, between Europe and E. Asia about .1 and between Europe and Nigeria about .14 .)

Heritability of brain structure

The following paper has some beautiful figures, including the one below illustrating the heritability of physical brain structure based on MRI studies of twins.

Neurobiology of Intelligence: Science and Ethics

Jeremy Gray1 and Paul Thompson2
1Yale University, and
2University of California, Los Angeles

Review Paper, Nature Reviews Neuroscience 5:1-13, June 2004, published online, May 19 2004.

Human mental abilities, such as intelligence, are complex and profoundly important, both in a practical sense and for what they imply about the human condition. Understanding these abilities in mechanistic terms has the potential to facilitate their enhancement. There is strong evidence that the lateral prefrontal cortex supports intelligent behaviour. Variations in intelligence and brain structure are heritable, but are also impacted by factors such as education, and prenatal and family environments. The empirical convergence of cognitive, social, psychometric, genetic, and neuroimaging studies of intelligence is scientifically exciting, but raises important ethical questions. If these are not addressed, further empirical advances might be compromised.

Click for larger image, or look at the actual paper. From the figure caption, which is quite long:

Figure 3 |Linking genes, brain structure and intelligence. a| Genetic influences on intelligence have been assessed directly (top arrow). The consensus of many studies is that at least 40% of the variability in general cognitive ability (g) can be attributed to genetic factors89. Gene effects on brain structure can be assessed by collecting MRI brain scans (left) from twins or extended families, and comparing volumes of grey matter (green), white matter (red) or cerebrospinal fluid (blue). Overall brain volume is 85% heritable34and correlates with psychometric intelligence (0.33) (REF.30). ...The volume of grey matter in each lobe is genetically influenced to different degrees (the volume of grey matter in the frontal lobe, shown at right in yellow and pink, is highly heritable). b| Genetic influences on brain structure can be assessed using statistical maps. In the classical twin design, a feature is heritable if within-pair correlations (typically called intraclass correlations) are higher for pairs of identical twins (who share all their genes, except for rare somatic mutations) and lower for same-sex fraternal twin pairs (who, on average, share half their genes). To better understand genetic influences on brain structure, correlations are shown for regional grey matter volumes in sets of identical (monozygotic (MZ)) and fraternal (dizygotic (DZ)) twins. These correlations vary across the brain surface (red, highly correlated; blue, less well correlated). The structure of the brains of identical twins is more similar than that of fraternal twins. ...Variations in grey matter volumes are strongly influenced by genetic factors, especially in frontal brain regions (for example, the dorsolateral prefrontal cortex). A subsequent study in a larger, independent sample34found that variations in total grey matter volume were almost entirely attributable to genetic factors... These genetically mediated differences in brain structure explain a proportion of the variation in general cognitive ability. This ability is also influenced by non-genetic factors such as education and nutrition97,151, prenatal and family environments, training35and environmental hazards such as lead poisoning.

Race and IQ in Nature

I wouldn't touch this with a 10 foot pole, but Nature did here (sort of), as part of its Darwin 200 celebration. (Via Razib.)

In this, the second of two opposing commentaries, Stephen Ceci and Wendy M. Williams argue that such research is both morally defensible and important for the pursuit of truth. In the first, Steven Rose argues that studies investigating possible links between race, gender and intelligence do no good.

The ensuing online debate is as polarized as you might expect. Read James Flynn's comment, which utterly demolishes Steven Rose.

Flynn: ...In Rose’s original paper [commentary], he asserts that the trait in question (intelligence) leaves aside other desirable traits and argues that the groups in question can be divided into subgroups that are more biologically coherent. He concludes that the hypothesis is not subject to scientific treatment; and therefore, no useful social policy will emerge. In his response to Ceci and Williams, he says something very different, namely, that by about 1975, it had been definitively shown that genes had no place in explaining group differences. So from that date, Jensen and everybody else had no excuse to persist.

To assert both that a hypothesis is not scientifically testable and that it has been conclusively falsified is incoherent. The only way to reconcile them is to assume that Rose does not really mean Jensen had been refuted by 1975, but is saying that by that date, it should have been clear to everyone that the question was indeed unanswerable.

I predict most readers will find about half the remarks in the debate incredibly stupid. But which half? :-)

Incidentally, one of the commentary authors and I were both on the faculty at Yale at the same time and I have known her for many years.

My Darwin 200 post is here, but see also this ;-)

Designer babies

Enhanced or unenhanced? You be the judge... From the episode Space Seed (trailer); full video.

McCoy: The Eugenics Wars of the 1990s.
Spock: Your attempt to improve the race by selective breeding.
McCoy: Oh no not our attempt, Mr. Spock. A group of ambitious scientists.





Soon to be outlawed? Probably not in Asia...

WSJ: Want a daughter with blond hair, green eyes and pale skin?

A Los Angeles clinic says it will soon help couples select both gender and physical traits in a baby when they undergo a form of fertility treatment. The clinic, Fertility Institutes, says it has received "half a dozen" requests for the service, which is based on a procedure called pre-implantation genetic diagnosis, or PGD.

While PGD has long been used for the medical purpose of averting life-threatening diseases in children, the science behind it has quietly progressed to the point that it could potentially be used to create designer babies. It isn't clear that Fertility Institutes can yet deliver on its claims of trait selection. But the growth of PGD, unfettered by any state or federal regulations in the U.S., has accelerated genetic knowledge swiftly enough that pre-selecting cosmetic traits in a baby is no longer the stuff of science fiction.

"It's technically feasible and it can be done," says Mark Hughes, a pioneer of the PGD process and director of Genesis Genetics Institute, a large fertility laboratory in Detroit. However, he adds that "no legitimate lab would get into it and, if they did, they'd be ostracized."

But Fertility Institutes disagrees. "This is cosmetic medicine," says Jeff Steinberg, director of the clinic that is advertising gender and physical trait selection on its Web site. "Others are frightened by the criticism but we have no problems with it."

PGD is a technique whereby a three-day-old embryo, consisting of about six cells, is tested in a lab to see if it carries a particular genetic disease. Embryos free of that disease are implanted in the mother's womb. Introduced in the 1990s, it has allowed thousands of parents to avoid passing on deadly disorders to their children.

...In a recent U.S. survey of 999 people who sought genetic counseling, a majority said they supported prenatal genetic tests for the elimination of certain serious diseases. The survey found that 56% supported using them to counter blindness and 75% for mental retardation.

More provocatively, about 10% of respondents said they would want genetic testing for athletic ability, while another 10% voted for improved height. Nearly 13% backed the approach to select for superior intelligence, according to the survey conducted by researchers at the New York University School of Medicine.

There are significant hurdles to any form of genetic enhancement. Most human traits are controlled by multiple genetic factors, and knowledge about their complex workings, though accelerating, is incomplete. And traits such as athleticism and intelligence are affected not just by DNA, but by environmental factors that cannot be controlled in a lab.

While many countries have banned the use of PGD for gender selection, it is permitted in the U.S. In 2006, a survey by the Genetics and Public Policy Center at Johns Hopkins University found that 42% of 137 PGD clinics offered a gender-selection service.

The science of PGD has steadily expanded its scope, often in contentious ways. Embryo screening, for example, is sometimes used to create a genetically matched "savior sibling" -- a younger sister or brother whose healthy cells can be harvested to treat an older sibling with a serious illness.

It also is increasingly used to weed out embryos at risk of genetic diseases -- such as breast cancer -- that could be treated, or that might not strike a person later in life. In 2007, the Bridge Centre fertility clinic in London screened embryos so that a baby wouldn't suffer from a serious squint that afflicted the father.

...For trait selection, a big hurdle is getting enough useful DNA material from the embryo. In a typical PGD procedure, a single cell is removed from a six-cell embryo and tested for the relevant genes or SNPs. It's relatively easy to check and eliminate diseases such as cystic fibrosis that are linked to a single malfunctioning gene. But to read the larger number of SNP markers associated with complex ailments such as diabetes, or traits like hair color, there often isn't enough high-quality genetic material.

William Kearns, a medical geneticist and director of the Shady Grove Center for Preimplantation Genetics in Rockville, Md., says he has made headway in cracking the problem. In a presentation made at a November meeting of the American Society of Human Genetics in Philadelphia, he described how he had managed to amplify the DNA available from a single embryonic cell to identify complex diseases and also certain physical traits.

Of 42 embryos tested, Dr. Kearns said he had enough data to identify SNPs that relate to northern European skin, hair and eye pigmentation in 80% of the samples. (A patent for Dr. Kearn's technique is pending; the test data are unpublished and have yet to be reviewed by other scientists.)

Dr. Kearns' talk attracted the attention of Dr. Steinberg, the head of Fertility Institutes, which already offers PGD for gender selection. The clinic had hoped to collaborate with Dr. Kearns to offer trait selection as well. In December, the clinic's Web site announced that couples who signed up for embryo screening would soon be able to make "a pre-selected choice of gender, eye color, hair color and complexion, along with screening for potentially lethal diseases."

Genetic substructure of Ashkenazim

More principal components results from the huge space of human genetic variation.







Figures from the paper below (additional labeling by Razib).

A genome-wide genetic signature of Jewish ancestry perfectly separates individuals with and without full Jewish ancestry in a large random sample of European Americans

Anna C Need, Dalia Kasperaviciute, Elizabeth T Cirulli and David B Goldstein

Genome Biology 2009, 10:R7doi:10.1186/gb-2009-10-1-r7

Conclusions: Here we show that within Americans of European ancestry there is a perfect genetic corollary of Jewish ancestry which, in pinciple, would permit near perfect genetic inference of Ashkenazi Jewish ancestry. In fact, even subjects with a single Jewish grandparent can be statistically distinguished from those without Jewish ancestry. We also found that subjects with Jewish ancestry were slightly more heterozygous than the subjects with no Jewish ancestry, suggesting that the genetic distinction between Jews and non-Jews may be more attributable to Near-Eastern origin for Jewish popuations than to population bottlenecks.

Pinker on personal genomics

Read the whole thing in the Times magazine. More on Eric Turkheimer's work here.

My Genome, My Self: ...To study something scientifically, you first have to measure it, and psychologists have developed tests for many mental traits. And contrary to popular opinion, the tests work pretty well: they give a similar measurement of a person every time they are administered, and they statistically predict life outcomes like school and job performance, psychiatric diagnoses and marital stability. Tests for intelligence might ask people to recite a string of digits backward, define a word like “predicament,” identify what an egg and a seed have in common or assemble four triangles into a square. Personality tests ask people to agree or disagree with statements like “Often I cross the street in order not to meet someone I know,” “I often was in trouble in school,” “Before I do something I try to consider how my friends will react to it” and “People say insulting and vulgar things about me.” People’s answers to a large set of these questions tend to vary in five major ways: openness to experience, conscientiousness, extraversion, agreeableness (as opposed to antagonism) and neuroticism. The scores can then be compared with those of relatives who vary in relatedness and family backgrounds.

The most prominent finding of behavioral genetics has been summarized by the psychologist Eric Turkheimer: “The nature-nurture debate is over. . . . All human behavioral traits are heritable.” By this he meant that a substantial fraction of the variation among individuals within a culture can be linked to variation in their genes. Whether you measure intelligence or personality, religiosity or political orientation, television watching or cigarette smoking, the outcome is the same. Identical twins (who share all their genes) are more similar than fraternal twins (who share half their genes that vary among people). Biological siblings (who share half those genes too) are more similar than adopted siblings (who share no more genes than do strangers). And identical twins separated at birth and raised in different adoptive homes (who share their genes but not their environments) are uncannily similar.

...Behavioral genetics has repeatedly found that the “shared environment” — everything that siblings growing up in the same home have in common, including their parents, their neighborhood, their home, their peer group and their school — has less of an influence on the way they turn out than their genes. In many studies, the shared environment has no measurable influence on the adult at all. Siblings reared together end up no more similar than siblings reared apart, and adoptive siblings reared in the same family end up not similar at all. A large chunk of the variation among people in intelligence and personality is not predictable from any obvious feature of the world of their childhood.

Think of a pair of identical twins you know. They are probably highly similar, but they are certainly not indistinguishable. They clearly have their own personalities, and in some cases one twin can be gay and the other straight, or one schizophrenic and the other not. But where could these differences have come from? Not from their genes, which are identical. And not from their parents or siblings or neighborhood or school either, which were also, in most cases, identical. Behavioral geneticists attribute this mysterious variation to the “nonshared” or “unique” environment, but that is just a fudge factor introduced to make the numbers add up to 100 percent.

No one knows what the nongenetic causes of individuality are. Perhaps people are shaped by modifications of genes that take place after conception, or by haphazard fluctuations in the chemical soup in the womb or the wiring up of the brain or the expression of the genes themselves. Even in the simplest organisms, genes are not turned on and off like clockwork but are subject to a lot of random noise, which is why genetically identical fruit flies bred in controlled laboratory conditions can end up with unpredictable differences in their anatomy. This genetic roulette must be even more significant in an organism as complex as a human, and it tells us that the two traditional shapers of a person, nature and nurture, must be augmented by a third one, brute chance.

The discoveries of behavioral genetics call for another adjustment to our traditional conception of a nature-nurture cocktail. A common finding is that the effects of being brought up in a given family are sometimes detectable in childhood, but that they tend to peter out by the time the child has grown up. That is, the reach of the genes appears to get stronger as we age, not weaker. Perhaps our genes affect our environments, which in turn affect ourselves. Young children are at the mercy of parents and have to adapt to a world that is not of their choosing. As they get older, however, they can gravitate to the microenvironments that best suit their natures. Some children naturally lose themselves in the library or the local woods or the nearest computer; others ingratiate themselves with the jocks or the goths or the church youth group. Whatever genetic quirks incline a youth toward one niche or another will be magnified over time as they develop the parts of themselves that allow them to flourish in their chosen worlds. Also magnified are the accidents of life (catching or dropping a ball, acing or flubbing a test), which, according to the psychologist Judith Rich Harris, may help explain the seemingly random component of personality variation. The environment, then, is not a stamping machine that pounds us into a shape but a cafeteria of options from which our genes and our histories incline us to choose.

Recent evolution in humans

Did evolution stop once modern humans emerged in Africa? Or, to the contrary, did it speed up?

This question is addressed in the forthcoming book by Greg Cochran and Henry Harpending: The 10,000 Year Explosion. Harpending is an anthropogist and Cochran a physicist. Together they have produced a number of interesting research ideas in the area of human evolution (see below). I've read a pre-release draft of the book and recommend it highly. If you enjoyed Guns, Germs and Steel by Jared Diamond, then you owe it to yourself to read this book, which directly engages Diamond's thesis that geography (not DNA) is destiny.





I discussed research supporting accelerated recent human evolution by Cochran, Harpending and collaborators in an earlier post: We are all mutants now. The figure below is from a Times article by Nicholas Wade.

We are all mutants now: Some interesting new science suggests that human evolution has accelerated in the last tens of thousands of years. The study by Hawks, Wang, Cochran, Harpending and Moyzis (of UW Madison, Affymetrix, U Utah and UC Irvine) uses linkage disequilibrium tests on hapmap SNP data to determine that roughly 7% of all genes have undergone strong selection recently. The method looks for regions of DNA with similar SNP patterns. If an advantageous gene swept through a population in a relatively short time, replacing other variants, then the pattern of nucleotide polymorphisms in that area of the chromosome will be particularly uniform throughout the group. The results imply that we are all descended from mutants who, relatively recently, out-competed and replaced their contemporaries. The distribution of mutations is not uniform in different geographical populations (i.e., races). Recent evolution is causing genetic divergence, not convergence.

There is a good theoretical argument for why evolution may speed up due to population growth. Given a particular probability distribution for producing beneficial mutations, a large population implies a faster rate of incidence of such mutations. Because reproductive dynamics leads to exponential solutions (i.e., a slight increase in expected number of offspring compounds rapidly), the time required for an advantageous allele to sweep through a population only grows logarithmically with the population, while the rate of incidence grows linearly.

To elaborate on the last point, consider the set of mutations that are sufficiently advantageous that they would sweep through a population of N humans (i.e. reach fixation) in some specified period of time, such as 5000 years. If the probability of such a mutation is p, the rate of occurrence in the population is proportional to pN. Now imagine the population of the group increases to 100N. The rate of mutations is then much higher -- 100pN -- but the time necessary for fixation has only increased by the logarithm of 100 since selective advantage works exponentially: the population fraction with the mutant gene grows as exp( r t ), where r is the reproductive advantage and t is time. This rather obvious point -- that linear beats log -- suggests that the rate of evolution will speed up as population size increases. (A possible loophole is if the probability of mutations as a function of relative advantage is itself an exponential function, and falls off rapidly with increasing advantage.) If the Hawks et al. results are any guide, as many as 7% of all genes have been under intense selection in the last 10-50,000 years. (See here for another summary of the research with a nice illustration of how linkage disequilibrium arises due to favorable mutations.) Importantly, the variants that reached fixation over this period are different in different geographical regions.

Thus civilization, with its consequently larger populations supported by agriculture, enhanced rather than suppressed the rate of human evolution.

A related question is whether selection pressure remained strong after the development of civilization. Perhaps reproductive success became largely decoupled from genetic influences once humans became civilized? Not only is this implausible, but it seems to be directly contradicted by evidence. The graph below, based on English inheritance records, shows that the rich gradually out-reproduced the poor: the wealthy had more than twice as many surviving children as the poor. (Note the range of inheritances in the graph covers the middle class to moderately wealthy; the poor and very rich are not shown.) Thus, in this period of history wealth was a good proxy for reproductive success. Genes which were beneficial for the accrual of wealth (e.g., for intelligence, self-discipline, delayal of gratification, etc.) would have become more prevalent over time. In a simple population model, any lineage that remained consistently poor over a few hundred year period would contribute almost zero to today's population of Britons.



The graph is taken from this paper:

Survival of the Richest: The Malthusian Mechanism in Pre-Industrial England

GREGORY CLARK AND GILLIAN HAMILTON

Fundamental to the Malthusian model of pre-industrial society is the assumption that higher income increased reproductive success. Despite the seemingly inescapable logic of this model, its empirical support is weak. We examine the link between income and net fertility using data from wills on reproductive success, social status and income for England 1585–1638. We find that for this society, close to a Malthusian equilibrium, wealth robustly predicted reproductive success. The richest testators left twice as many children as the poorest. Consequently, in this static economy, social mobility was predominantly downwards. The result extends back to at least 1250 in England.

See also my review of Clark's A Farewell to Alms, and this video of a talk by Clark. When Clark wrote the book he wasn't sure whether it was genetic change or cultural change that led to the industrial revolution in England. In the video lecture he comments that he has since become convinced it was largely genetic. That doesn't jibe with the back of the envelope calculation I give below -- even in the optimistic case (largest effect) it would seem to take a thousand years to have a big shift in overall population characteristics.

Here's a very crude back of the envelope calculation: if, in a brutal Malthusian setting, the top 10% in wealth were to out-reproduce the average by 20% per generation, then after only 10 generations or so (say 2-300 years), essentially everyone in the population would trace their heritage in some way to this group. In our population the average IQ of the high income group is about +.5 SD relative to the average. If the heritability of IQ is .5, then in an ideal case we could see a selection-driven increase of +.25 SD every 2-300 years, or +1 SD per millenium. This is highly speculative, of course, and oversimplified, but it shows that there is (plausibly) no shortage of selection pressure to drive noticeable, even dramatic, change. If the estimate is too high by an order of magnitude (the rich group doesn't directly replace the others; there is inevitably a lot of intermarriage between descendants of the rich and non-rich), a change of +1 SD per 10,000 years would still be possible. There's clearly no shortage in genetic variation affecting intelligence: we see 1 SD variations not just within populations but commonly in individual families!


So where does this leave us?

1) The rate of positive mutations went up due to population growth. More importantly, the rate of mutations that were likely to sweep the entire population in a fixed period of time probably went up.

2) Natural selection did not abate: there is evidence for differential reproductive rates that are impacted by genes.

3) Humans living today are possibly quite different from our ancestors of 50,000 years ago. I would guess we are smarter and better suited to living in a complex society that requires cooperation and planning. We are also probably more likely to be lactose tolerant, nearsighted and bad at hunting ;-)

Cochran and Harpending's new book deserves wide attention and serious discussion.

More genetic substructure

Via Genetic Future, two more principal components analyses of SNP data. Click the genetics label for related posts.

European, Nigerian and East Asian samples from HapMap and 100 African-American samples clustered based on data from ~600,000 genetic markers. The bend toward the East Asian cluster is probably due to Native American admixture.

Alkes L. Price, Nick Patterson, et al. (2008). Effects of cis and trans Genetic Ancestry on Gene Expression in African Americans, PLoS Genetics, 4 (12) DOI: 10.1371/journal.pgen.1000294.





Resolving Finns and Swedes. Jakkula et al. (2008) The Genome-wide Patterns of Variation Expose Significant Substructure in a Founder Population, The American Journal of Human Genetics, 83 (6), 787-794 DOI: 10.1016/j.ajhg.2008.11.005.

Resolution of population genetic structure

How much can we resolve the substructure of a population with a given amount of data? The paper below gives a quantitative answer. With current technology, we should have no problem resolving even small national populations (see italicized text in quote below), with nearest neighbor FST as small as .0001 (i.e., 99.99 percent of variation is within-group and only .01 percent between groups)! According to this Table of European, Nigerian and East Asian FSTs, the FST between France and Spain is .0008, whereas between Nigeria and Japan it is about .19 .

Within the European + HapMap sample analyzed here, over 100 statistically significant PCA vectors were identified. That is, there is a >100 dimensional space within which structure can be teased out. (However, the largest single vector accounts for only a percent of total variation, and the integral over all 100 vectors is probably only a few percent.) Norwegians and Swedes could be resolved with 90 percent accuracy. Note the Patterson et al. paper was written before this recent analysis, which confirms their theoretical predictions of sensitivity. (Figure below.)



The first author, Nick Patterson (profiled here), is a mathematician turned cryptographer turned quant (Renaissance) turned bioinformaticist.

Population Structure and Eigenanalysis

Nick Patterson et al. (Broad Institute of Harvard and MIT)

Abstract
Current methods for inferring population structure from genetic data do not provide formal significance tests for population differentiation. We discuss an approach to studying population structure (principal components analysis) that was first applied to genetic data by Cavalli-Sforza and colleagues. We place the method on a solid statistical footing, using results from modern statistics to develop formal significance tests. We also uncover a general “phase change” phenomenon about the ability to detect structure in genetic data, which emerges from the statistical theory we use, and has an important implication for the ability to discover structure in genetic data: for a fixed but large dataset size, divergence between two populations (as measured, for example, by a statistic like FST) below a threshold is essentially undetectable, but a little above threshold, detection will be easy. This means that we can predict the dataset size needed to detect structure.



...Another implication is that these methods are sensitive. For example, given a 100,000 marker array and a sample size of 1,000, then the BBP threshold for two equal subpopulations, each of size 500, is FST = .0001. An FST value of .001 will thus be trivial to detect. To put this into context, we note that a typical value of FST between human populations in Northern and Southern Europe is about .006 [15]. Thus, we predict: most large genetic datasets with human data will show some detectable population structure.

Human genetic variation, Fst and Lewontin's fallacy in pictures

In an earlier post European genetic substructure, I displayed the following graphic, illustrating the genetic clustering of human populations.




Figure: The three clusters shown above are European (top, green + red), Nigerian (light blue) and E. Asian (purple + blue).

The figure seems to contradict an often stated observation about human genetic diversity, which has become known among experts as Lewontin's fallacy: genetic variation between two random individuals in a given population accounts for 80% or more of the total variation within the entire human population. Therefore, according to the fallacy, any classification of humans into groups ("races") based on genetic information is impossible. ("More variation within groups than between groups.")

To understand this statement better, consider the F statistic of population genetics, introduced by Sewall Wright:

Fst = 1 - Dw / Db

Db and Dw represent the average number of pairwise differences between two individuals sampled from different populations (Db = "difference between") or the same population (Dw = "difference within"). Even in the most widely separated human populations Fst < .2 so Dw / Db > .8 (roughly). This may not sound like very much genetic diversity, but it is more than in many other animal species. See here for recent high statistics Fst values by nationality.

Dw / Db > .8 means that the average genetic distance measured in number of base pair differences between two members of a group (e.g., two randomly selected Europeans) is at least 80 percent of the average distance between distant groups (e.g., Europeans and Asians or Africans). In other words, if two individuals from very distant groups (e.g., a Japanese and a Nigerian) have on average N base pair differences, then two from the same group (e.g., two Nigerians or two Japanese) will on average have roughly .8 N base pair differences.

How can the Fst result ("more variation within groups than between groups") be consistent with the clusters shown in the figure? I've had to explain this on numerous occasions, always with great difficulty because the explanation requires a little mathematics. In order to make the point more accessible, I've created the figures below, which show two population clusters, each represented by an ellipsoid (blob). The different figures depict the same pair of objects, just viewed from different angles.

The blobs are constructed and arranged so that the average distance between two points (individuals) within the same cluster is almost as big as the average distance between two points (individuals) in different clusters. This is easy to achieve if the ellipsoids are big and flat (like pancakes) and placed close to each other along the flat directions. The figure is meant to show how one can have small Fst, as in humans, yet easily resolved clusters. The direction in which the gap between the clusters appears is one of the principal components in the space of human genetic variation, as recently found by bioinformaticists. The figure at the top of this post plots individuals as points in the space generated by the two largest principal components extracted from the combination of data from HapMap and from large statistics sampling of Europeans. Exhibited this way, isolated clusters ("races") are readily apparent.

The real space of genetic variation has many more than 3 dimensions, so it can't be easily visualized. But some aspects of the figures below still apply: there will be particular directions of variation over which different populations are more or less identical (orthogonal to the principal component; i.e. along the flat directions of each pancake), and there will be directions in which different populations differ radically and have little or no overlap. Note, however, that we are specifically referring to genetic variation, which may or may not translate into phenotypic variation.







Related posts: "no scientific basis for race" , metric on the space of genomes.

The existence of this clustering has been known for 40 years.

East Asian genetic substructure

Below are some results on East and Southeast Asian genetic substructure. As you can see, Koreans are (sort of) midway between Chinese and Japanese. It will almost certainly be possible to differentiate between different regional origins based on DNA, once larger statistics studies become available. (See European results.)

Figures: Each point is an individual, and the axes are two principal components in the space of genetic variation. Colors correspond to individuals of different Asian ancestry.






Thanks to Chao Tian of UC Davis for sending me an early draft of the paper.

Analysis of East Asia Genetic Substructure: Population Differentiation and PCA Clusters Correlate with Geographic Distribution.

C. Tian1, R. Kosoy1, A. Lee2, P. Gregersen2, J. Belmont2, M. Seldin1

1) Rowe Program Human Genetics, Univ California Sch Medicine, Davis, CA; 2) North Shore-LIJ Res Inst, Manhasset, NY, Baylor Col Med., Houston TX.

Accounting for genetic substructure within European populations has been important in reducing type 1 errors in genetic studies of complex disease. As efforts to understand complex genetic disease are expanded to other continental populations an understanding of genetic substructure within these continents will be useful in design and execution of association tests. In this study, population differentiation(Fst) and Principal Components Analyses(PCA) are examined using >200K genotypes from multiple populations of East Asian ancestry(total 298 subjects). The population groups included those from the Human Genome Diversity Panel[Cambodian(CAMB), Yi, Daur, Mongolian(MGL), Lahu, Dai, Hezhen, Miaozu, Naxi, Oroqen, She, Tu, Tujia, Naxi, and Xibo], HapMap(CHB and JPT), and East Asian or East Asian American subjects of Vietnamese(VIET), Korean(KOR), Filipino(FIL) and Chinese ancestry. Paired Fst(Wei and Cockerham) showed close relationships between CHB and several large East Asian population groups(CHB/KOR, 0.0019; CHB/JPT, 00651; CHB/VIET, 0.0065) with larger separation with FIL(CHB/FIL, 0.014). Low levels of differentiation were also observed between DAI and VIET(0.0045) and between VIET and CAMB(0.0062). Similarly, small Fst's were observed among different presumed Han Chinese populations originating in different regions of mainland of China and Taiwan(Fst < 0.0025 with CHB). For PCA, the first two PC's showed a pattern of relationships that closely followed the geographic distribution of the different East Asian populations. For example, the four "corner" groups were JPT, FIL, CAMB and MGL with the CHB forming the center group, and KOR was between CHB and JPT. Other small ethnic groups were also in rough geographic correlation with their putative origins. These studies have also enabled the selection of a subset of East Asian substructure ancestry informative markers(EASTASAIMS) that may be useful for future genetic association studies in reducing type 1 errors and in identifying homogeneous groups.

Related posts: "no scientific basis for race" , metric on the space of genomes

European genetic substructure

Figure: Each point is an individual, and the axes are two principal components in the space of genetic variation. Colors correspond to individuals of different European ancestry.

The figure above is from the Nature paper: European Journal of Human Genetics (2008) 16, 1413–1429; doi:10.1038/ejhg.2008.210

Abstract: An investigation into fine-scale European population structure was carried out using high-density genetic variation on nearly 6000 individuals originating from across Europe. The individuals were collected as control samples and were genotyped with more than 300 000 SNPs in genome-wide association studies using the Illumina Infinium platform. A major East–West gradient from Russian (Moscow) samples to Spanish samples was identified as the first principal component (PC) of the genetic diversity. The second PC identified a North–South gradient from Norway and Sweden to Romania and Spain. ...

Some interesting points:

1) Significant East-West and North-South substructure is apparent already from the figure. The resolution of the study is sufficiently high that Swedes and Norwegians can be distinguished with 90 percent accuracy (Table 4). Crime scene forensics will never be the same -- "the Swede did it!" ;-)

In conclusion, we have shown that using PCA techniques it is possible to detect fine-level genetic variation in European samples. The genetic and geographic distances between samples are highly correlated, resulting in a striking concordance between the scatter plot of the first two components from a PCA of European samples and a geographic map of sample origins. We have shown how this information can be used to predict the origin of unknown samples in a rapid, precise and robust manner, and that this prediction can be performed without requiring access to the individual genotype data on the original samples of known origin. ...

2) Genetic distances between population clusters are roughly as follows: the distance between two neighboring western European populations is of order one in units of standard deviations and the distance to the Russian cluster is several times larger than that -- say, 3 or 4. From HapMap data, the distance from Russian to Chinese and Japanese clusters is about 18, and the distance of southern Europeans to the Nigerian cluster is about 19. The chance of mis-identifying a European as an African or E. Asian is exponentially small! (Table 5)

...The distance measure is a measure of the distance in standard deviations from a sample to the center of the closest matching population.

...For the other HapMap populations, the classification procedure assigned 100% of the YRI [Yoruban = Nigerian] samples to France, and almost 100% of the CHB and JPT [Chinese and Japanese] samples to Russia. However, the distribution of the distance measure for the four populations was quite different. For the CEU [HapMap European] samples, the median and 95% CI of the distance measure were 0.41 (0.11–1.01), whereas for the YRI, CHB and JPT populations, the median and 95% CIs were 19.3 (18.0–20.6), 17.7 (15.9–19.3) and 18.0 (15.4–19.6), respectively.

...The Yoruban [Nigerian] and Asian samples were identified as belonging to the countries on the south and east edges, respectively, of the European cluster, and the distance measure clearly indicates that they do not fit well into any of the proposed populations. ...

Figure: The three clusters shown above are European (top, green + red), Nigerian (light blue) and E. Asian (purple + blue).


See additional discussion at gnxp (the modified figure is from Razib), Dienekes

Related posts: "no scientific basis for race" , metric on the space of genomes