There are HLA variants known to confer partial immunity. For example, people have studied sub-Saharan African prostitutes - possibly the highest risk group in the world, alongside gay sub-Saharan Africans - who were miraculously not HIV positive after several years as a sex worker.

more:

HLA Association with HIV-1 Transmission

Many of the earlier studies of HLA alleles associated with HIV-1 transmission included very small and diverse study populations with differing routes of exposure. Some studies contained prevalent HIV infected cases which could reflect progression of HIV-1 disease rather than susceptibility to infection, and most had little to no molecular confirmation of HLA alleles, resulting in no consistent HLA class I or II associations with HIV-1 infection (Just 1995). However, there is considerable evidence developing from a small number of individuals who have been exposed to HIV (some repeatedly exposed) but do not seroconvert or show any signs of HIV infection. These observations suggest that, in some cases, natural immunity may protect exposed individuals from HIV infection and that HLA-restricted CTLs may be responsible for the protective immunity (Shearer and Clerici 1996). Individuals who have been exposed but do not have HIV include prostitutes and others that engage in unprotected sex with HIV+ partners, infants born of HIV+ mothers, those exposed to contaminated blood products through transfusions, health care workers, and intravenous drug users (IVDU) with a history of needle sharing. Some of these individuals have been shown to exhibit HIV-specific HLA-restricted CTL responses, in the absence of HIV-specific antibodies. In fact, a strong T cell response, including but not limited to HLA class I &endash;restricted CTL responses, has long been invoked as a major factor in protective immunity against HIV infection and AIDS progression (Clerici and Shearer 1994).

In addition to HLA variation, deletions in CCR5 that are frequent in Northwest Europeans have also been shown to confer immunity:

Scientist have a theory as to why 10 percent of the European population is resistant to HIV. The epidemics of hemorrhagic fever that struck Europe during the Middle Ages, may have perpetuated a specific gene mutation that was resistant. Called delta-32 in the cellular receptor dubbed CCR5, it protects against HIV and may have also protected against the Black Death and smallpox.

more.

More recent examples include mutations in the CCR5 gene that appear to provide protection against AIDS. The CCR5 gene encodes a protein on the surface of human immune cells. HIV, the virus that causes AIDS, infects immune cells by binding to this protein and another protein on the surface of those cells. Mutations in the CCR5 gene that alter its level of expression or the structure of the resulting protein can decrease HIV infection. Early research on one genetic variant indicates that it may have risen to high frequency in Northern Europe about 700 years ago, at about the time of the European epidemic of bubonic plague. This finding has led some scientists to hypothesize that the CCR5 mutation may have provided protection against infection by Yersinia pestis, the bacterium that causes plague. The fact that HIV and Y. pestis both infect macrophages supports the argument for selective advantage of this genetic variant.

The sickle cell and AIDS/plague stories remind us that the biological significance of genetic variation depends on the environment in which genes are expressed. It also reminds us that differential selection and evolution would not proceed in the absence of genetic variation within a species.

Not to bang the GNXP drum overmuch, but given these kinds of nontrivial discrepancies between populations, it is premature to say that human evolution stopped at the neck or stopped after we dispersed from Africa.

One other point: given the demographics of those infected with AIDS, it seems behavior is the best inoculant. Berkeley scientists have shown that AIDS has an ultra-strong selective effect in Sub-Saharan Africa. As both sexual behavior and IQ are highly heritable, we can expect nontrivial behavioral selection in Sub-Saharan Africa within our lifetimes.

Also - while this sort of speculation is difficult to nail down without large scale sequence data, it's not out of the question that the CCR5 variants which provided immunity to the black plague also provided immunity to STDs that were common around that time...and the Renaissance started soon after the Black Plague ended. Perhaps the CCR5 variant is the tip of an iceberg of mutations that provided inoculation (both molecular and behavioral) against STDs. To take the speculation and run with it....because behavioral inoculation would have the effect of encouraging greater sexual restraint/planning/foresight/etc., the selective reaction to STDs may have promoted the development of civilization after the long European dark ages.

Quote from GNXP essay on IQ:

human genetic differences matter and they matter intensely

Second question: have experts studied the propensity for Alzheimer's in sub-Saharan Africans vs African-Americans? Those results would be interesting.

Lastly, what, if any, conclusions can we draw about people in general from studies such as these? (It's my roundabout way of saying that I hope this doesn't provide ammo for half-cocked "social psychologists")

Uh, let's keep the conversation on AIDS/HIV shall we? The business about intelligence and continued human evolution is distracting.

GNXP reader, you've given lots of data about the protective genes that, presumably, Indians lack (has anyone studied that to find out for sure?).

But what about the harmful genes mentioned by one of the researchers? Is that for real?

Another question: you refer to the gene that some sub-Saharan Africans have. But does that actually affect the statistical rate of conversion? If not, then the absence of the gene in India doesn't explain the difference in the rates.

And: don't lifestyle and nutrition potentially play a part in determining how likely people are to get full-blown AIDS?

Regardless of AIDS/HIV, Alzheimer's, IQ, etc., a big part of my above questioning goes towards the rate at which these genes can alter/mutate, if they do. How quickly can surroundings and related changes express themselves in genetics through adaptation and variation? Or is it a case of if your race has shitty genes for HIV, generations of you are stuck with it?

Should've listened to my mom and become a doctor so as not to ask all of these questions.

Wow, that's fascinating. Particularly because "Indian" is such a huge group. (Anyone know about the genetic diversity of India? I'm guessing it's big, since it's usually keyed to ethnic diversity, right?)--it would be interesting to see that result resolved by State/Country. Maybe it has to do with climate? I mean, totally extemporizing here, isn't HIV originally a simian virus? If a group traditionally spends a lot of time near monkeys might they have higher or lower immunity?

I read recently that some Europeans have better immunity, possibly conferred by the plague. Maybe the subcontinent somehow managed to avoid the plague?

Another Desi-genes thing I've heard lately is an increased probability of having narrow veins and arteries. Don't know if that's true or if it can be related in anyway.

Sorry, more on the plague right above me. Silly Saheli. Nice cite, gnxp.

...totally extemporizing here... If a group traditionally spends a lot of time near monkeys might they have higher or lower immunity?

Really? You were extemporizing? Wouldn't have guessed...but then again maybe there is some scientific tempor to that theory. I am just a whacko ignoramus.

But what about the harmful genes mentioned by one of the researchers? Is that for real?

don't get hung up on "harmful" vs. "protective." this isn't a function vs. loss of function thing, the "genes" in question are part of the broad HLA family, basically the they code for some very special functions within your immune system related to T cells.

humans are very diverse on the HLA loci, likely because there is selection for diversity in that pathogens mutate and evolve so fast that a genetic "monoculture" is really easy to wipe out quickly. this is one reason native americans had problems with eurasian pathogens, their HLA profiles are relatively homogenous so a really bad plague can fell them all in one swoop. the HLA loci are so diverse that your HLA profile might be closer to a chimpanzee than your cousin, because these alleles, forms of the gene, are older than the chimp-human line split.

all this is to go to the idea that "protective" and "harmful" are contextual terms. the "harmful" genes are probably "protective" in other contexts, and the "good" genes probably are "harmful" in other contexts. the hypothesis about CCR5 is that they were protective against bubonic plague, but seeing as how the frequency of the gene is modal in southern sweden, and that bubonic plague probably come from asia, i'm not so sure about that.

Maitri:

To what extent are these genetic differences mitigated by multiple generations of the same race successively enjoying a higher standard of living, better nutrition and a higher quality of education? At what point does nature stop being the lead?

Sometimes never. The Japanese, for example, have been first world for a very long time but are still not as tall as the Dinkas of the third world. Genetics do matter and put bounds on what can be achieved by environmental manipulation. Whether you're talking disease or IQ or height, the gaps between humans can be as big as that between a pygmy and Manute Bol...and about as unbridgeable by post-natal intervention.

Nobel Prize Winner James Heckman has done a lot of work on this and has concluded that IQ (for example) cannot be substantially positively altered beyond the initial big jump that comes from living in an industrialized society.

Another continuing blind spot in the vision of most educational planners and policy makers is a preoccupation with achievement tests and measures of cognitive skill as indicators of the success of an educational intervention. By narrowly focusing on cognition, they ignore the full array of socially and economically valuable non-cognitive skills and motivation produced by schools, families and other institutions. This emphasis also critically affects the way certain early intervention programs have been evaluated. For example, while enriched early intervention programs do not substantially alter IQ, they do substantially raise the non-cognitive skills and social competence of participants.”... An important lesson to draw from the entire literature on successful early interventions is that it is the social skills and motivation of the child that are more easily altered— not IQ. These social and emotional skills affect performance in school and in the workplace. We too often have a bias toward believing that only cognitive skills are of fundamental importance to success in life.”

Other skills can be easily altered, but not IQ. The Flynn effect is puzzling in light of these experimental observations, but the new era of direct brain measurements pioneered by guys like Paul Thompson (cited above) and Richard Haier will render IQ obsolete:

Human intelligence determined by volume and location of gray matter tissue in brain

Single ‘intelligence center’ in brain unlikely, UCI study also finds

Irvine, Calif. , July 19, 2004
General human intelligence appears to be based on the volume of gray matter tissue in certain regions of the brain, UC Irvine College of Medicine researchers have found in the most comprehensive structural brain-scan study of intelligence to date...

The researchers used a technique called voxel-based morphometry to determine gray matter volume throughout the brain which they correlated to IQ scores. Study results appear on the online version of NeuroImage.

Previous research had shown that larger brains are weakly related to higher IQ, but this study is the first to demonstrate that gray matter in specific regions in the brain is more related to IQ than is overall size. Multiple brain areas are related to IQ, the UCI and UNM researchers have found, and various combinations of these areas can similarly account for IQ scores. Therefore, it is likely that a person’s mental strengths and weaknesses depend in large part on the individual pattern of gray matter across his or her brain.

“This may be why one person is quite good at mathematics and not so good at spelling, and another person, with the same IQ, has the opposite pattern of abilities,” Haier said.

This makes sense as similar things have been observed in primates and other animals, where the location of brain alterations makes a big difference.

Anyway, as these brain measurements predict IQ, which correlates with outcome measures like how much money you make, how likely you are to get a PhD, and how well you will do on the SAT, it will be only a few years before we can observe whether IQ-relevant time evolution in mean brain structure is in fact occurring. This might be the case - in OECD countries people are taller today than they were 100 years back, and they might also be smarter.

Of course there are exceptions to every statistical rule, but the gross statistics matter - just as the fact that there are *some* non-Asian minorities in academic positions does not stop people from using a statistical label to refer to them as "underrepresented".

Second question: have experts studied the propensity for Alzheimer's in sub-Saharan Africans vs African-Americans? Those results would be interesting.

Well the results wouldn't be comparable. With life expectancy having fallen off a cliff across Africa (see this shocking graph), the people who make it to old age in sub-Saharan Africa are exceptional. In Uganda and Botswana life expectancy is below 40.

don't lifestyle and nutrition potentially play a part in determining how likely people are to get full-blown AIDS?

I haven't seen much about nutrition, but behavior and lifestyle plays a big role. Studies consistently show that men who have sex with men and people who have high numbers of sexual partners have the highest infection rates. Thus gay males and African Americans (and to a lesser extent Americans of Hispanic descent) have much higher infection rates.

In the United States, nearly 406,000 people were living with AIDS at the end of 2003, and African Americans accounted for half of these AIDS cases. Among women, minorities—particularly African Americans—are hit by the vast majority of AIDS cases. Rates of HIV/AIDS diagnoses in African-American women are 19 times higher than those of white women and 5 times higher than those of Hispanic women in the 32 states with stable HIV/AIDS reporting. African Americans also suffer the vast majority of deaths caused by AIDS, accounting for more than half of all U.S. AIDS-related deaths in 2003.

How quickly can surroundings and related changes express themselves in genetics through adaptation and variation? Or is it a case of if your race has shitty genes for HIV, generations of you are stuck with it?

Selection has to be very strong to have a rapid effect on the whole distribution. For a single locus the equations have been worked out for a number of different selection regimes. See here for a textbook example with a dichotomous Mendelian trait and (scroll down) here for an example with a continuous trait.

This picture may also be helpful. What we are talking about with AIDS is something like the threshold selection shown on that page, where those with a small number of sexual partners live and those with a large number die. You can model that with a hard step function or a softer sigmoid.

Insofar as genetics influences sexual behavior at all (expressed by a parent-child correlation aka heritability coefficient), those who pass down sexually restrained behavior (whether biologically or culturally) will be more likely to have grandchildren.

You can also come up with an explicit multilocus model. Assume that sexual behavior as measured by number of partners over lifetime is the (pseudo) continuous trait which ranges from 0 to Wilt-the-Stilt numbers and beyond. The long tail is required to model the large number of partners of prostitutes and gay males, who are crucial hubs in most epidemiological graph-theoretical models of infection.

The center of your sigmoidal threshold function will be more lenient if you possess genes which give you a molecular inoculant. For example, some of the prostitutes in the above references required dozens or hundreds of partners before seroconversion was observed. So you can model the presence of an inoculant by a shift in the sigmoidal death function.

The sad part is that rapid selection means mass death of the sort that is currently happening in Botswana, where 40% of the adult population has HIV/AIDS.

razib:

the hypothesis about CCR5 is that they were protective against bubonic plague, but seeing as how the frequency of the gene is modal in southern sweden, and that bubonic plague probably come from asia, i'm not so sure about that.

There is molecular evidence too, though. The CCR5 variant impedes the ability of Yersinia pestis to get into the cell. The variant protein provides less of a foothold for infection. The same has been observed at the cellular level for HIV - CCR5 delta-32 cells are less vulnerable for seroconversion.

To bring it back to Indian susceptibility to AIDS, here are the premises:

1) As far as I'm aware of, no study has shown widespread molecular immunity to HIV. As the CDC says in its page on the CCR5 delta-32 variant:

The allele’s prevalence varies by ethnicity, being as high as 10%-15% in Caucasians, ~2% in African Americans, and virtually absent in native Africans and East Asians...

To date, no recommendation on population testing has been issued. Since the major factor in HIV infection is exposure to the virus, protective measures against the exposure continue to be recommended as public health policy.

Thus molecular inoculation is (as far as we know) currently less of a factor than behavioral inoculation. Within the European population, this may be tentatively illustrated by the difference in infection rates between gay males and heterosexual males. Of course, it may be that the (behavioral) genetic basis of homosexuality covaries with the (molecular) genetic susceptibility to HIV infection, so this is why I say "tentatively illustrated".

2) For this reason, talk of "susceptibility genes" or resistance genes cannot omit talk of genes which influence behavioral properties, particularly behavioral attitudes towards sex. Thousands of such genes exist, of course. It is incoherent to believe that genes may determine sexual orientation but not sexual desire/appetite/restraint/etc.

It is also empirically unsound - high parent-child correlations have been found for an enormous number of behavioral traits, even among twins raised apart. MIT's Stephen Pinker goes into great detail in The Blank Slate. But here's one example:

There have now been several studies showing a correlation between an individual's sexual orientation and his or her genotype. In one, a sample of 115 gay men who had male twins, 52% of identical twin brothers were also gay compared with only 22% of fraternal twin brothers and 11% of adopted brothers (Bailey & Pillard, 1991). In a comparable sample of 115 lesbians, 48% of identical twin sisters were also lesbians compared with only 16% of fraternal twin sisters and 6% of adopted sisters (Bailey, Pillard, Neale, & Agyei, 1993). A subsequent study of nearly 5,000 twins who had been systematically drawn from a twin registry confirmed the significant heritability of sexual orientation for men but not for women (Bailey & Martin, 1995). Finally, an analysis of families in which there were two gay brothers suggested a correlation between a homosexual orientation and the inheritance of genetic markers on the X chromosome (Hamer & Copeland, 1994; Hamer, Hu, Magnuson, Hu, & Pattatucci, 1993).

and another

Genetic or biological variables also can be conceptualized as family influences on adolescents' sexual behavior. For example, androgen hormone levels and the timing of pubertal development are partially hereditary, and they affect adolescent sexual behavior (Morris, 1992; Udry & Campbell, 1994). Researchers have reported positive correlations between age of menarche among mothers, daughters, and sisters (Garn, 1980; Newcomer & Udry, 1984). Further, mothers' young age of first intercourse predicts sons and daughters also having sex before age 14 (Mott, Fondell, Hu, Kowaleski-Jones, & Menaghan, 1996). Genetic, shared environmental, and nonshared environmental models show heritability for the timing of sexual intercourse among siblings, although this varies by racial and gender pairs (Rodgers, Rowe, & Buster, 1998). There also is preliminary evidence for a relationship between dopamine receptor genes and age of first sexual intercourse (Miller et al., 1999).

and one more

We also assessed smoking, drinking, drug use, sex, driving and gambling on separate risk-taking scales related to each particular kind of risky behavior. The driving-risk scale asked about typical driving speeds, response to traffic signals and following distances at high speeds, among other factors. To assess risky sexual behavior, we asked about the number of sexual partners and whether or not a condom is used, and if so, how consistently...

The first question we sought to answer was whether the six arenas of risk are interrelated, pointing to a concept of generalized risk-taking.

As it turned out, smoking, drinking, sex and drugs work in tandem with each other. Among both males and females, students who did one tended to do the others. (We know from other studies that they also listen to rock and roll.) Reckless driving, however, was related to only one other arena of risk: drinking. Unfortunately, this connection is often deadly. Among males, gambling was related to drinking and sex. But among women, it was not related to any other kinds of risk-taking. With the single exception of gambling among women, we felt justified in computing a generalized risk-taking score based on all six kinds of risk-taking....

Studies of the heritability of sensation-seeking in humans have used classical twin-comparison methods. Comparisons of identical and fraternal twins in which both siblings were raised in the same families show that sensation-seeking is about 60% genetic. That is a high degree of heritability for a personality trait; most range from 30% to 50%...

Genes play yet another role in risk-taking: They influence two other personality traits associated with general risk-taking, including the traits of aggression, or its obverse, agreeableness, and for sociability, the main component of extroversion.

The new science of molecular genetics has made it possible to identify major genes influencing personality and forms of psychopathology. A group of scientists in Israel were the first to find an association between novelty-seeking (a trait very highly correlated with impulsive sensation-seeking) and a gene that codes for a class of dopamine receptor, the dopamine receptor-4 (DRD4) gene.

Anyway. Point is that behavior is the strongest inoculant against AIDS. Both biology and culture influences behavior. And for both biological and cultural reasons, Indians do not tend to have large numbers of sexual partners (fecundity is not the same as promiscuity). Thus statements like this are terribly misleading:

India says more than five million of its citizens are infected with the HIV virus, second only to South Africa.

India has more than 1 billion people, while South Africa has 25 million. For actual AIDS infection rates, go to AVERT. India has prevalence rates around 1% in most states according to 2003 data. It is highly unlikely that an outbreak will be seen there (or in China) which is on the scope of that in Sub-Saharan Africa.

South Africa has 25 million

Sorry, it's 42 million. Doesn't change the point though, which is that rates of infection differ by about 20 fold.