The Old Race

That puzzling variant of PLK4, that increases the miscarriage rate,  is very widespread, moderately common in almost all population – but it’s rare in Bushmen (frequency of 1 in 38 among the  Ju’/hoansi) and Pygmies,  enough to you suspect that it wasn’t there at all before the Bantu recently admixed with both groups.

The Bushmen (and probably African Pygmies as well)  apparently split off earlier than any other human population, something like 200,000 years ago. Some Bushmen among the Ju’/hoansi, show low or even zero admixture with other groups.  How they managed to have so little gene flow with other Africans for such a long time is a mystery to me, but that’s what the stats say.  There is more genetic distance between the Bushmen and Bantu than there is between Bantu and Koreans.

This long separation doesn’t necessarily mean that Bushmen ave the most divergent phenotypes (although thinking about it, they probably are, what with steatopygia , the tablier egyptienne, etc)  – strength and direction of selection are important, not just time.  But all else equal, more time allows greater changes.

This doesn’t mean that the Bushmen are what early homo sap was like – they’ve been evolving too – but we ought to be able to learn quite a bit about changes over the past couple of hundred thousand years by investigating  genetic differences between the Bushmen and everybody else.  For example, 200 k years ago, our ancestors didn’t have what it takes to out-compete Neanderthals and other archaics on their home grounds, judging by the fact that they didn’t manage it back then.  By 40k years ago we could and did – but that isn’t necessarily the case for Bushmen, a separate branch. Although, since Bushmen and Pygmies seem to have picked up a few percent of their genome from some very divergent group of archaic humans, perhaps they too developed the ability to kick archaic ass. But we don’t know if they did it in the same way, and it probably happened in fairly familiar African environments, instead of ice age Eurasia.

By the way, I have a very funny story about Bushman genetics, but there are not enough electrons in this margin.


Posted in Uncategorized | Tagged | 95 Comments

Roland Fryer wins the John Bates Clark medal

Roland Fryer has won the John Bates Clark medal for the top economist under 40.

I’ve paid attention to some of his papers that impinge on biology, even corresponded with him.  In one,  he argued that higher risk for cardiovascular disease among American blacks was due to genetic factors, those resulting from selection for salt-sensitivity in the Middle Passage, the risky journey to America.

There’s nothing wrong with looking for genetic factors as a possible explanation (or partial explanation) of such health differences, but it’s not necessarily easy to come up with correct answers, least of all when you don’t know anything about the subject – don’t know anything about biochemistry or medicine or natural selection.  Which is the default state of humanity, of course – and of economists.

There are black-white genetic differences in genes involved in holding onto salt and water, and that’s undoubtedly a product of selection – but you see those same variants in Africans.  It’s not something that showed up in 1750.

The death rate in the Middle Passage was about 15%,  during the period that most of the ancestors of American blacks came across. A lot of those deaths were due to diarrheal disease, but there is no reason to think salt-sensitivity is the overwhelmingly dominant factor that determines survival in such cases – obviously the immune system must also matter, for example. There is certainly no evidence that salt-sensitivity is even the main factor in determining survival in such cases, although it might be.  And since there were other causes of death there, such as scurvy and smallpox, it couldn’t have been all-important.

Moreover, it is almost impossible to cause a significant genetic change in one generation*, unless the selective factor wipes out something like the least resistant 95% of the population. I told Fryer this – worked out an example using the breeder’s equation, plausible values of heritability, a favorable (and unrealistic) estimate of the importance of salt retention, reasonable estimates of mortality on those ships – there was no way anything significant would happen.  He didn’t make use of the breeder’s equation, but  it can’t be that he never heard of it, because I told him.

He mentioned a case in which researchers successfully bred salt-sensitive rats in three generations, with noticeable change by the second generation –  but with this was done by  founding one line with the two most salt-sensitive rats (2 out of 47) and the other with the two least salt-sensitive rats.   That rat experiment was the equivalent of  95% casualties for three generations , all strictly due to salt sensitivity – pretty different from 15% casualties for one generation that are maybe kind-sorta influenced by salt sensitivity.  Consider also that they are comparing one line strongly selected for salt-sensitivity with another strongly selected for  salt-insensitivity, selection was a minimum of 60 times stronger in the rat experiment than in the Middle Passage.

There’s no chance that salt-sensitivity in African-Americans is a product of selection  in the Middle Passage.

In another paper, Fryer and Leavitt looked at intelligence in children aged 8 to 12 months. They find only minor racial differences in test outcomes.  which proves..???  At that point you’re testing motor skills or something –  anything like an IQ test is impossible.  Black kids are somewhat precocious in motor skills compared to whites: on average they walk about a month earlier, East Asians about a month later.  But chimps at age one will outperform all humans, while gazelles can run 15 minutes after birth.  All which proves nothing at all. That first paper is wrong.  This paper is … not even wrong.

Well, maybe the papers I have not read are high-quality.   There’s a rule, I think: when someone is egregiously wrong on all the things you know anything about, he must be damn good on everything else.

* the next person who talks about how much people living in slum conditions  in the US over the last couple of generations must have been selected for X, where X is anything at all, WILL BE BANNED


Posted in Uncategorized | Tagged | 44 Comments

The Lottery

Lotteries can be useful natural experiments; we can use them to test the accuracy of standard sociological theories, in which rich people buy their kids extra smarts, bigger brains, better health, etc.

David Cesarini, who I met at that Chicago meeting, has looked at the effect of winning the lottery in Sweden.  He found that the “effects of parental wealth on infant health, drug consumption, scholastic performance and cognitive and non-cognitive skills can be bounded to a tight interval around zero.”

As I once mentioned, there was an important land lottery in Georgia in 1832. The winners received an 160-acre farm. But by 1880, their descendants were no more literate, their occupational status no higher. The families in the top 2/3rds of income managed to hang on to some of their windfall, but lower-income families did not.

This remind of a story by Gerald Kersh, “Whatever Happened to Corporal Cuckoo? ” – about a medieval soldier who stumbled into immortality.  Someone asks him (in 1945) – why hadn’t he saved his pay?  With compound interest, yaddaa yadda.

” Why didn’t I save my pay? Because I’m what I am, you mug! Hell, once upon a time, if I’d stayed away from cards, I could’ve bought Manhattan Island for less than what I lost to a Dutchman called Bruncker drawing ace-high for English guineas!  Save my pay! If it wasn’t one thing it was another. I lay off liquor. Okay. So if it’s not liquor, it’s a woman. I lay off women. Okay. Then it’s cards or dice. I always meant to save my pay; but I never had it in me to save my pay!  Doctor Paré’s stuff fixed me–and when I say it fixed me, I mean, it fixed me, just like I was,  and am, and always will be. ”

Low leverage of wealth on your children’s  traits is something that exists in a particular society, with a particular kind of technology. Back in medieval times, a windfall could  have kept your kids alive in a famine, and that certainly had a long-term positive effect on their cognitive skills.  Dead men take no tests.  The most effective medical interventions today are cheap – everyone in Sweden and the US already has them – but there are places where those interventions are not universally available.  Some families in Mozambique can afford artemisin, some can’t – this must make a difference.

Suppose we had a method of dramatically improving a kid’s genetic potential for intelligence and success, one that cost five million dollars a pop: then wealth could  influence the next generation in ways that it can’t today.  In other words,  Cesarini’s conclusions are correct for Sweden-now  (but not for Sweden in 1700), probably correct for the US today, but maybe not true tomorrow.

It is not just wealth that has a small effect on your kid’s potential: playing Mozart doesn’t help either. Other than locking away the ball-peen hammers, it’s hard to think of any known approach that does have much effect – although we don’t know everything, and maybe there are undiscovered  effective approaches (other than genetic engineering). For example, iodine  supplements have a good effect in areas that are iodine-deficient.  We now know (since 2014)  that bromine is an essential trace element – maybe people in some parts of the world would benefit from bromine supplementation.

What about the social interventions that people are advocating, like Pre-K ?  Since shared family effects (family environment surely matters more than some external social program) are small by adulthood, I think they’re unlikely to have any lasting effect.  We might also note that the track record isn’t exactly encouraging. If there was a  known and feasible way of boosting academic performance, you’d think that those teachers in Atlanta would have tried it. Sure beats prison.

Maybe there’s an effective approach using fmri and biofeedback – wouldn’t hurt to take a look.  But even if it did work, it might simply boost everyone equally, and obviously nobody gives a shit about that.

Posted in Uncategorized | 104 Comments

Safety Dance

I understand that many college students are made to feel “unsafe” by certain speakers addressing particular subjects.  But although most people like crisp breakfast cereal, there are those that like it as soggy as possible. One would think that there must be a significant fraction of the university market that wants genuinely scary content, a fraction that prefers mega-aggression to microaggression.

I can do that.

Posted in Uncategorized | 44 Comments

He still has that hair

Steven Pinker was saying that greatly extending the human lifespan is hard, maybe impossible: ” I suspect that death will never be conquered (though our lifespans will continue to increase, at least for a while). Any cost-free longevity gene or easily tunable molecular pathway would have been low-hanging fruit for natural selection long ago. Senescence is baked into most of our genome because of the logic of evolution: since there’s a nonzero probability at any moment that an organism will die in an unpreventable accident, making genes for longevity moot, selection tends to sacrifice longevity for performance at every level of organization. This means we’d have to know how to tinker with thousands of genes or molecular pathways, each a tiny (and noisy) effect on longevity, to make the leap to immortality. The low-hanging fruit is in fact at the other end of the lifespan and income scale. We’ve made massive global progress in reducing maternal and infant mortality and premature death, but we’re not seeing a cohort of billionaire centagenarians. ” Centenarians.

He’s got the argument backward: sure, natural selection has not favored perfect repair, so says the evolutionary theory of  of senescence.  If it had, then we could perhaps conclude that perfect repair was very hard to achieve, since we don’t see it, at least not in complex animals.*  But since it was not favored, since natural selection never even tried, it may not be that difficult.

Any cost-free longevity gene that made you live to be 120 would have had a small payoff, since various hazards were fairly likely to get you by then anyway…  And even if it would have been favored, a similar gene that cost a nickel would not have been.  Yet we can afford a nickel.

There are useful natural examples: we don’t have to start from scratch. Bowhead whales live over 200 years: I’m not too proud to learn from them.

Lastly , this would take a lot of work.  So what?

*Although we can invent things that evolution can’t – we don’t insist that all the intermediate stages be viable.

Posted in Uncategorized | 75 Comments

Eight thousand years of natural selection in Europe

They get a selective advantage of 1.5% for the European lactose tolerance allele: “We estimated the selection coefficient on the derived allele to be 0.015 (95% confidence interval; CI=0.010-0.034) using a method that fits a hidden Markov model to the population allele frequencies as they change over time. ”

That’s way lower than any other result I’ve ever seen, including runs I’ve done myself. I’ve seen estimates a full ten times higher.  I don’t get it.

Posted in Uncategorized | 13 Comments

Born that way

In the Atlantic Monthly , which no longer has any reason for existence, since there’s really no point in placing a computer screen on the bottom of a birdcage, Jason Silverstein  – a doctoral candidate in anthropology at Harvard –  has a piece explaining that genes don’t cause racial health disparities – society does! Moreover,  it’s immoral to even look for such genetic explanations.

I guess he should have gotten this published earlier, since we’ve already found some of those naughty genes – pretty important ones. So it’s too late.

Probably the biggest well-understood case involves two common variants of APOL1, a gene that mostly transports lipids, but also zaps trypanosomes – the cause of sleeping sickness.  Humans with the standard form of APOL1 are immune to most trypanosomal infections,  but two strains have evolved resistance to  the standard form of APOL1 – they’re the ones that cause sleeping sickness in humans   In response, two variants of APOL1 that block one of the strains of sleeping sickness that infects humans (Trypanosoma brucei rhodesiense) have become common in central and west African populations, and their diaspora.

Unfortunately, these  protective variants are hell on your kidneys. And they’re common:    > 30% of African-Americans carry a risk variant.  It increases the risk of kidney failure ( by a lot) in several conditions  (focal segmental glomerulosis(FSGS),  HIV-associated nephropathy (HIVAN), and diabetes-associated nephropathy).  Altogether, African-Americans  develop kidney disease at a rate 4-5 times higher than other groups.  It is worth noting that African-Americans that don’t have these APOL1 variants appear to have  a risk similar to that of whites.  You have to give the Man credit for focusing the bad health effects of racism in a way that could easily fool people into thinking that genes matter – even more so considering that nobody discovered these APOL1 variants until 2010.  I think we’re talking secret underground labs: probably the same guys that taught chimps to make war and brainwashed chimpettes into liking dolls instead of trucks.

There are of course diseases in which one or a few genes of major effect  play a major role in disease among African Americans, like sickle-cell anemia.  There are other diseases that have very different incidence and outcomes in blacks and whites, but have not been pinned to a single locus. In some of those cases,  the difference may be caused by  several genes, or perhaps many genes of small effect rather than a couple of major-effect variants.  For example, hypertension goes up strongly with the amount of African ancestry: some of that is probably caused by the APOL1 variants, but there is also evidence that differences in how blacks handle salt and water conservation also contribute – they have high-risk angiotensin variants,  and a more active version of NKCC2, a NA,K, Cl cotransporter.  Blacks often have a salt-sensitive form of hypertension. They also have lower urine volume, higher urine concentrations, 30% lower Ca excretion, expanded plasma volume, etc.

Age-adjusted sarcoidosis mortality is 12 times higher in blacks than whites: of course we don’t even know what causes that one.  Hate, maybe.  Or possibly the luminiferous ether.

Historically, we know that many infectious diseases had different courses in blacks and whites – for example, blacks were considerably more likely to survive yellow fever than whites, presumably because of oppression, but there’s been less work on that subject in recent years, partly because many of those diseases are well-controlled and partly because most researchers have enough sense to avoid working with anything so dangerous.

To  sum up, race is indeed a factor in health differences.  We know quite a lot about it , and are learning more.  Sometimes that new knowledge helps you figure out better treatments. I don’t think finding out things is immoral, with the possible exception of an easy method of inducing vacuum decay in the comfort of your own home.

Which means that Silverstein is a jackass. Nothing stopped him from digging into biomedical research to  see if his thesis was substantially true: he didn’t bother.  But why is he a jackass?  Born that way, probably.

Posted in Uncategorized | 57 Comments