The unfortunate consequences of misunderstanding race

Share on FacebookShare on Google+Email this to someoneTweet about this on Twitter

Reading the links that come in to GNXP, I happened upon this post on what the author referrs to as “scientific racism”. This bit caught my eye:

I sat on a grant review committee recently for a national-level competition for multi-million dollar grants of an agency I won’t name. The review committee was quite large, probably 25 or more scholars from around the U.S. One of the grant applications that the other reviewers (mostly from the biological sciences) rated the highest was one that proposed to look at the “genetic racial differences among Blacks and whites” to different kinds of treatment for HIV/AIDS. I rated this grant proposal among the lowest I had reviewed because of the methodology: all of the participants in the study would be sorted into the supposedly self-evident categories “Black” and “white” based on self-identification. When I raised this objection among my colleagues in the biological and health sciences, they all blinked hard, and looked at me as if I’d committed some sort of unpleasant faux pas. The chair of the committee finally acquiesced that this was a methodological flaw in the proposal, but the grant was nevertheless awarded millions of dollars.

Biomedical researchers are caught between a rock and a hard place here–none of them enjoy being referred to as scientific racists by their colleagues, I’m sure, but they’re also interested in real phenomena.

It’s well-known that minorities are less likely to participate in biomedical research (though recent studies suggest this is not because they’re less willing). From a geneticist’s perspective, the discomfiting implication of this is that tests of a drug’s efficacy and safety are done on a range of genetic backgrounds that are strongly biased towards the European mean. That is, drugs are accepted or rejected largely based on how well they perform in a sample of individuals of European descent. This is obviously a problem for the applicability of any results, and the NIH is indeed making minority inclusion a requirement for funding certain projects.

Clearly, none of this would be an issue if everyone responded identically to drugs (or if the correlation between drug response and race were zero). It is, however, an issue. Now, ancestry could be related to drug response through any number of mechanisms, either directly (through genetics) or indirectly (through socioeconomic status, etc). Teasing apart those influences means looking at both of them. But then you have someone like the author, who simply dismisses the correlation altogether! And who evidently has some say in the funding of these studies!

It’s worth pointing out that HIV progression does indeed have a genetic component and that certain alleles like CCR5-delta32, which is strongly protective against HIV infection, show marked geographic differences in frequency. A priori, looking for genetic components to differential drug response between populations seems entirely reasonable. Lastly, her main point seems to be that genetic ancestry and self-identified race might not match up. They do.

I’m well-aware that positions like the author’s are made possible by people at the other end of the spectrum, who see races as the embodiment of some Platonic ideal. But rejecting idiocy certainly does not require one to embrace blindness!

Related: Cancer and Race

Labels: ,

26 Comments

  1. who see races as the embodiment of some Platonic ideal. 
     
    and those who reject race because of its conception as a platonic ideal are caught in their own dichotomous constructs.

  2. There are differences in populations based on geography and likewise the effectivness of medical treatments may vary. If using race is good at exploring these differences, I say go for it. But it’s no different than looking at geographic populations or ethnicities and drug effectiveness based on that. 
     
    Sick-cell is found in all races where malaria is/was present, and is not found everywhere in all races. Personal geographic history may even be a more precise way of measuring.  
     
    Is an Eskimo going to react the same as Mexican just because they both belong to the “Asian” race? Maybe, maybe not. 
     
    But I see no harm in using race as a loose indices because it does bring more precision to the table.

  3. The author of the post (Jessie Daniels ) is a sociologist.  
    There doesn’t seem to be any point in discussing these matters with sociologists or cultural anthropologists. They don’t get it and they don’t want to get it , at all . Most biologists and other life scientists seem pretty ignorant about race right now, but they might not remain so. The biologists and life scientists on that funding board probably let the sociologist have her way because they just didn’t know how to articulate a defense of the race concept.

  4. Your quote of her statement strongly suggests that the self-identification was her problem. If self-identification is indeed sufficiently rigorous for a genetic study of this kind, then the grant applicants should have explicitly stated such in their proposal. It sounds like the proposal was flawed on grantsmanship if not methodology.

  5. Your quote of her statement strongly suggests that the self-identification was her problem. If self-identification is indeed sufficiently rigorous for a genetic study of this kind, then the grant applicants should have explicitly stated such in their proposal. It sounds like the proposal was flawed on grantsmanship if not methodology. 
     
    depending on your aim/study population doing ancestral informative genotyping might not be a good return-on-investment. funds are finite.

  6. to be clear, black americans are about 80% african and 20% european. this isn’t equally distributed, some black americans are mostly white, while many are far less than 20% european. that being said, if you are interested in doing an association study to figure out differences between the white and black self identified populations there is likely still going to be significant value added from an 80% african descended population toward figuring out between group differences. in a world with infinite money you’d do an analysis of the ancestry of these black individuals and throw out those who were genetically not so african. but in a world with finite money you might weight the cost of this vs. checking out more candiate genes or increasing the number of participants in the study.

  7. and those who reject race because of its conception as a platonic ideal are caught in their own dichotomous constructs. 
     
    That’s interesting, Razib. Rejectors of race interpret the statistical world of humankind as being just a reflection of a more real, higher truth that humankind has a completely unified heritage. Is that what you mean?

  8. The reaction to what she said was uniformly negative, according to her report. That isn’t what I’d expect based on what I read at GNXP. 
     
    In general the scientific study of human biodiversity would be easier to promote if it were completely disconnected from contemporary political issues and from trivial sterotypes, esthetic judgements, and gut feelings. Seriously, and not to nag, GNXP collectively has been pretty careless in these respects.

  9. The reaction to what she said was uniformly negative, according to her report. That isn’t what I’d expect based on what I read at GNXP. 
     
    why do you say that? the science is pretty standard and well-accepted in biomedical circles. it’s largely (though not entirely) outside of those circles (like in sociology, or politics) that it’s controversial.  
     
    Seriously, and not to nag, GNXP collectively has been pretty careless in these respects. 
     
    you have the open thread if you want to complain about GNXP collectively. if you have a criticism of something I’ve written, I’m happy to hear it.  
     
    Le GNXP, c’est pas moi.

  10. Apres gnxp, le deluge.

  11. “depending on your aim/study population, doing ancestral informative genotyping might not be a good return-on-investment. funds are finite.” 
     
    But in the US, more than half of new HIV infections are among African Americans (as defined by themselves and their medical practitioners). In the world at large, the huge majority of those infected with HIV are Africans. 
     
    Therefore, even if African ancestry has only a small genetic impact on treatment, the multiplying effect is huge. In fact, it would be better to recruit ONLY people of African ancestry than just to sign up cases in the US without regard to race, which might produce 13 or fewer percent African ancestry people. But best of all, recruit a representative genetic variety of candidates, as this study did. 
     
    By the way, the facts that a majority of new HIV infections in America are among African ancestral people, and that Africa has such a huge burden, suggests that there may indeed be a difference in susceptibility and progression to disease. I happen to be a gay man living in NYC, and I’m here to tell you that loads of white and Asian men are fucking around all the time unsafely, yet they’re not being infected to nearly the same degree.  
     
    If anything, I think it’s racist to ASSUME that there’s no genetic difference; as if, “sure THEY’RE getting sick; they’re not careful.”  
     
    “Racist”- one who wishes ill or persecutes another based on ancestry. “Near racist”- one who refuses to abandon an ideological or religious notion even at the cost of great fatality among another ancestral group, especially a disadvantaged or persecuted group. “Anti-racist”- one who acts to help an individual or group that is persecuted or disadvantaged, e.g. by doing relevant science, by giving poor people medicine. 
     
    A scientist who deals with these issues without any emotions or practical aims one way or the other doesn’t necessarily belong to any of these camps. S/he should be left alone to do the work that will help anti-racists do THEIR work. If they are clear-headed and decent, their findings will NEVER be racist. Racists have never needed scientific findings to back them up; they’re about something else altogether. 
     
    How to educated near racists? Rationally, explain how their attachment to anti-scientific notions actually undermines their more important humanitarian goals.

  12. The biological anthropologist Frank Livingstone correctly noted that there are no races, there are only clines. Human genetic variation follows a pattern of isolation by distance. Genetic distance increases continuously with geographic distance. When you divide a continuous trait into discrete categories you will always have problems at the margins of the division. Adjacent individuals on either side of the divide will be more closely related to each other than either will be to individuals on the far side of their own division. This is why the biological concept of race fails.  
     
    Anthropological geneticists who work with geographically diverse populations have known this for a long, long time. Biomedical geneticists who only work with samples drawn from US diversity never seem to understand this fundamental aspect of human genetic diversity. The reason is that their “African” samples are all drawn from a small region of West Africa, their “Asian” samples are all drawn from Japan, Korea and Eastern China, and their “Europeans” are all drawn from Northwestern Eurasia. It is not difficult to make discrete groupings when your samples are only drawn from the extremes of the cline. Fill in the points in between and the discrete categories don’t look very discrete any longer. 
     
    How well would it serve Barrack Obama to lump him in with “sub-Saharan Africans” when the category was created based upon the genetic diversity in West African Bantu populations when half his ancestry is East African Nilotic (Luo). East Africans are probably as genetically different from West Africans as West Africans are from Europeans.

  13. The biological anthropologist Frank Livingstone correctly noted that there are no races, there are only clines. Human genetic variation follows a pattern of isolation by distance. Genetic distance increases continuously with geographic distance. 
     
    this is true to a large extent. but there are discontinuities in the linear increase of genetic distance with geographic distance. for example, genetic distance across the himalayas is much greater than would be expected by geographic distance (here’s a citation). there are other such barriers; this is supported by the genetic evidence. see here:Examination of the relationship between genetic and geographic distance supports a view in which the clusters arise not as an artifact of the sampling scheme, but from small discontinuous jumps in genetic distance for most population pairs on opposite sides of geographic barriers, in comparison with genetic distance for pairs on the same side 
     
    Biomedical geneticists who only work with samples drawn from US diversity never seem to understand this fundamental aspect of human genetic diversity. The reason is that their “African” samples are all drawn from a small region of West Africa, their “Asian” samples are all drawn from Japan, Korea and Eastern China, and their “Europeans” are all drawn from Northwestern Eurasia 
     
    Conveniently, a paper on the genetic structure of European-Americans was mentioned here a couple days ago. Long story short: European-Americans are not all from Northwestern Eurasia. Your point about the sampling of Americans from disparate points on the globe is correct; this actually makes the social mapping of race more genetically relevant in the US than in other places.

  14. this is true to a large extent. but there are discontinuities in the linear increase of genetic distance with geographic distance. for example, genetic distance across the himalayas is much greater than would be expected by geographic distance 
     
    I don’t think that’s really to the point. The point is to abandon platonic categories.  
     
    For example, tall people have an increased risk of back problems. Does it make sense to say that that’s false because we can’t draw a line between short people and tall people?

  15. I don’t think that’s really to the point. The point is to abandon platonic categories. 
     
    but it isn’t a platonic category, it is a statistical or instrumental one. what geneticists want to do in these association studies is eliminate confounding background noise, right? how do you do that? you can genotype people and assign them to their appropriate clusters, or have them self-report. the former is probably better, but does it add enough value for you to justify the cost? self-reporting isn’t as accurate, but it is cheap, as in free. you basically want to remove the genetic correlation substructure from your pool, right? that’s the instrumental aim. self-identification works to do that, to an extent.

  16. but it isn’t a platonic category, it is a statistical or instrumental one 
     
    That’s my point. 
     
    you can genotype people and assign them to their appropriate clusters, or have them self-report 
     
    I have no problem with self-reporting. It might even be better than genotyping, if that’s what you want to correlate to for medical purposes.

  17. I don’t think that’s really to the point. The point is to abandon platonic categories. 
     
    sure, ok. but human population structure is more complicated than a perfect linear relationship between genetic and geographic distance. It is not true that there are “only clines”.

  18. There is not a perfect linear relationship between genetic and geographic distance, but there is a pretty good imperfect relationship. The point is that genetic variation is far more clinal than it is racial.  
     
    The most structured populations seem to be in areas of the world where there are geographic barriers to gene flow (for example Island populations in Oceania). This is not surprising. Interestingly, these structured populations would, a lot of times, be classified as one “race” in the traditional sense. 
     
    There is a paper just out on population structure in Native Americans in PLoS Genetics. ….haven’t read it yet.

  19. The point is that genetic variation is far more clinal than it is racial. 
     
    yes, absolutely. but to make the leap from that to “study X is flawed because it assumes races are different genetically” is not valid (not that you’ve made that leap), especially in the us, where, as you point out, ancestral populations come from very different geographic locations. more precision is always better, but as a first-order approximation, if you want to understand how ancestry and disease are related, self-reported race isn’t a bad place to start.  
     
    There is a paper just out on population structure in Native Americans in PLoS Genetics. ….haven’t read it yet. 
     
    just saw that in my inbox myself; I’ll give it a read this weekend.

  20. I think, p-ter, you and I are largely in agreement.  
     
    “but to make the leap from that to “study X is flawed because it assumes races are different genetically” is not valid (not that you’ve made that leap), especially in the us, where, as you point out, ancestral populations come from very different geographic locations” 
     
    Quite right. Especially about the substructure within the US. 
     
    My purpose in bringing up the largely clinal nature of genetic variation in human populations was to point out that the classical conception of human races does not actually work for our species. Because of the nasty political history of race I think scientists, and human geneticists in particular, need to be careful how they deal with the subject. This means, to some extent, choosing language carefully. 
     
    My impression reading the discussion was that human races were accepted by the participants as being something real. …probably more sloppy language and convenience than anything. …or a backlash against anti-science, anti-reality social scientists. …all the more reason for scientists to be more precise I think.

  21.  
    My purpose in bringing up the largely clinal nature of genetic variation in human populations was to point out that the classical conception of human races does not actually work for our species.
     
     
    what would you say the classical conception is? 
     
    My impression reading the discussion was that human races were accepted by the participants as being something real. 
     
    can you elaborate on this? the ‘real’ part is what i’m curious about. do you think species are ‘real’ for example? why?

  22. p-ter said: 
     
    Lastly, her main point seems to be that genetic ancestry and self-identified race might not match up. They do.  
     
    Razib said just a bit above: 
     
    but it isn’t a platonic category, it is a statistical or instrumental one. what geneticists want to do in these association studies is eliminate confounding background noise, right? how do you do that? you can genotype people and assign them to their appropriate clusters, or have them self-report. the former is probably better, but does it add enough value for you to justify the cost? self-reporting isn’t as accurate, but it is cheap, as in free. you basically want to remove the genetic correlation substructure from your pool, right? that’s the instrumental aim. self-identification works to do that, to an extent. 
     
    It seems to me that researchers can get far more accurate self reporting of race or ethnicity if they ask subjects that question in the ?right? sort of way. I think, in other words, that most people have a pretty good idea of their racial and ethnic background ? or anyway, a much better idea than they might reveal if the are simply asked their race or ethnicity. Politics and cultural stance gets involved; as well many or most people conflate genetics (ancestry) and culture if they aren?t specifically asked not to do so. 
     
    For example, if one asks Haley Barry what her race is she has and will under most circumstances say she?s black. Period, end of story. (For one thing she?s a trail blazer as a black actress, and not at all as a white one, and a lot less so as an amorphously mixed race one.) However she certainly knows her mother is 100% white / Caucasian (or anyway nearly so), and almost certainly knows that her father probably or did have some (perhaps considerable) white ancestry as well. She?s probably seen pictures of him for example. 
     
    If researchers looking for more accurate racial self identification by study participants, had a preamble that said that a great many Americans come from racially and or ethnically mixed, or partly mixed backgrounds, and asked them to self report not what group they most identify with, or feel the world regards them as belonging to, but rather how much of their ancestry was e.g. black versus white or some other racial group (e.g. Amerindian or Hispanic), I think one would get self reported results which are FAR closer to what DNA testing might reveal. Not all the way there certainly, and some particularly politically/ culturally adamant types might refuse to give what the researchers are looking for but instead stick with their platonicly absolutist identification, and some dimmer subjects may be unable to make the distinction as well particularly when e.g. their white admixture isn?t as personally close as their mother but instead one of their great grandparents, but I do think these virtually cost free self reported results would be a lot closer to DNA reported ones than when simply asking for race or percentage of different races w/out a preamble.  
     
    No, I don?t have a study demonstrating this; it?s an hypothesis based on personal experience.

  23. How much would that affect things? I guess the question is, what fraction of American blacks have a mixed-race set of parents or grandparents. You’re almost certainly going to know the apparent race of your parents.

  24. I think the classical conception of race is that members of a race are more closely related to each other than they are to any member of a separate race. I think it would be impossible to group all populations into such categories even if you excluded all recently admixed individuals. 
     
    No one has been able to provide a clear and universally applicable definition of a species since the publication of the Origin and the advent of modern biology. That is pretty clear evidence to me that “species” are little more than a useful way of talking about population groups. ….in practice, I tend to prefer the phylogenetic species concept.

  25. I guess the question is, what fraction of American blacks have a mixed-race set of parents or grandparents. You’re almost certainly going to know the apparent race of your parents. 
     
    I think the great majority of notably light skinned blacks are aware that they have a substantial percentage of white ancestry, even when it entered quite a ways back, i.e. even when both of their parents would be and were classified by overall American society and most blacks as being “black”, and even when all four of their grandparents were as well. (Obviously if all four grandparents of a light skinned black are “black”, then all or most of them would most likely have had to be light skinned blacks.) This was not an uncommon pattern among more upper class blacks through the 1930′s and beyond. See e.g. the book “Our Kind of People”.

  26. i would say many black americans, especially light-skinned ones, are aware of their white ancestry. 
     
    That is pretty clear evidence to me that “species” are little more than a useful way of talking about population groups 
     
    i have the same attitude toward race. and species. all of these are just instrumental categories. from an evolutionary genetic perspective i’m just interested in generating short-hand for patterns of genetic variation, and these categories allow us to derive an understanding of that variation without having to characterize each individual. the phylogenetic species concept is fine for systematics of course, but sometimes it is not too useful when you are interested in specific genetic questions.

a