← Back to Story

Transcript: “Rooting out Flawed Genetic Classifications—and the Racial Bias Behind Them”

Sarah Tishkoff and Dorothy Roberts Discuss Race and Genetic Research with Wharton Business Radio host Dan Loney.

Dan Loney: Race is playing a role in the process of genetic testing. Biases were included to a degree that could affect research. Now there’s research coming out of Penn about this, including the area of ancestry testing, which is a growing area right now…When we talked about this a year ago, it was about race being factored into this research. A year later, where are we with this?

Dorothy Roberts: Well this is a longstanding habit of scientists in the U.S. and Europe, to use race as if it were a biological characteristic in research. And what Sarah’s and my paper was about was criticizing how the variable of race continues to be used/treated as a biological category even in advanced genomic research and human genetic variation research and so since then one update is that Sarah and I have both participated in efforts by the NIH to address this problem of confusion, different meanings of race being used, misunderstandings, sloppiness, even again among very advanced genetic researchers in using the variable.

DL: Sarah?

Sarah Tishkoff: I think that we’ve come a long way but we still have a ways to go. I think particularly in the field of medical genetics or medical genomics, it’s very common in the literature for example that you’ll see people referring to what are thought of as biological racial classifications—caucasian for example—and what we and others have argued is that it’s much more informative to talk about ancestry rather than what people think of as biologically classified races, or with the new technology in genomics, maybe we don’t have to refer to the at all because you can make a lot of inferences just based on the genetic data that can be useful for the scientific research without categorizing people by ancestry for example.

ST: Everybody’s probably heard of 23 Me or Ancestry.com or other places that offer to do ancestry testing, but there are some issues with that, some ethical issues, some issues having to do with the science and how accurate it is and what they can and cannot tell people, and there are also some social issues that have arisen, so particularly in this political climate, I’ve been approached by some reporters who are interested in how the white nationalists are using this information about ancestry to make claims about whether you’re really white or not.

DR: I think ancestry testing is a great way of showing how the way scientists describe “race” versus “ancestry” affects the public imagination of what race means, so if companies are telling the public what race someone is or what percentage of different races they are, it reinforces the view that human beings really are divided into these four or five pure groups of race that are determined by genetic difference, and that has always reinforced a view of racial difference that is related to politics, and I think Sarah’s example of white supremicist groups using human genetic variation research done by scientists who did not intend at all to be promoting white supremacist views but the use of race as if it were a pure genetically-distinct grouping supports their views of white racial purity and superiority, and that’s another reason why scientists need to be clear about what is the meaning of the categories they’re using and to think about ancestry as opposed to race as the appropriate understanding of human difference that they’re looking for in the research.

DL: If memory serves, when we talked a year ago, also having the researchers have a much more exact idea of what they were bringing forth, and it’s a mindset among the researchers that needs to be adjusted.

DR: I think this is also a good example of the way in which Sarah and I have collaborated. I as a sociologist and legal scholar, Sarah as a genomic scientist, so from the social end and the biological end working together in an interdisciplinary way to really understand the full scope of what race means. It’s not just a biological concept, it’s also a social and political concept, and I think if biologists just focus on the biological implications, they sometimes miss the political ramifications of the research they’re doing, and conversely, if sociologists don’t understand the biology that biologists are investigating, we might not be able to explain to them our social perspective, so for the researchers, both perspectives are really important.

DL: I would think from the researcher’s perspective, as Dorothy just laid out, it’s not a natural thought process of thinking about something other than the research itself and the various implications that could play out outside of the research area in this case.

ST: We’ve learned a lot from each other and I’ve certainly been enlightened, particularly as we wrote together a commentary in Science, and there was a lot of back and forth because we were coming from very different perspectives and even using different language and different terminology and I think we learned a lot and I think there’s more room for that to happen in the field of genetics/genomics.

DL: So the idea is, in your conversations with the NIH, bring this information forward and bring it on to a variety of different researchers across the U.S. and across the world who may be to a degree are falling into this trap of not thinking along this boundary.

DR: Absolutely. That was a meeting of scholars and researchers from a variety of disciplines and one interesting aspect of it was a point where many of the genomic scientists were listening to a sociologist explain the various possible meanings of ancestry and race, you know, self-identification, how you identify yourself versus how a researcher might identify you. What an individual research subject might mean when they think about their ancestry versus what a biologist might mean. But if a biologist is using self-identified race or ancestry in their research, it may mean something different to the research participant than to the researcher. That’s problem, if they’re coming at it from different perspectives and so by having sociologists just in the room to explain the different ways people think about race socially, it was useful to the biologists who sometimes think about race as if it were this biological category that’s almost self-evident when in fact it can have many meanings to different people.

DL: And some of the social impact from pieces outside of the actual race. These are things that play at times a huge role in terms of whatever the research may be.

ST: Right, well as geneticists, we’re looking at both genetic and environmental factors whatever might be influencing common diseases. Hypertension, for example, is more common among African Americans in the U.S. than any other ethnic group. Why is that? It could be a genetic factor; there are probably also environmental and socio-economic factors that are playing a role as well, and as geneticists we tend to be very focused on the genetic component and I think people like Dorothy tend to be really focused on the socio economic and we need to have more of an exchange, but I also want to point out that at that meeting, the other thing we talked about was the gross underrepresentation of minority populations in human genomics research. So studies of genetic factors influencing disease, and environmental factors. I mean one to two percent of all genomics research – I mean, it was shocking. So we want also to promote the inclusion of diverse populations in human genetics research.

DL: I’m guessing to a degree there’s been a distressing historical trend, why isn’t there more minority representation?

DR: It’s complicated. Part of it is not paying enough attention to what the significance of inclusion is. Why include people from diverse backgrounds. Also, a long history of including people of color in harmful experimentation. We have this history in the U.S. where black people, Latinos, Native Americans and other people of color have been the victims of medical experimentation. And at the same time, people of color being excluded from helpful clinical trials being conducted in an ethical way where they actually benefit from the clinical trial.

DL: If you think about the testing that has been done and the testing that hasn’t, what the impact could be on disease, lifestyle, a long list of areas where it could be significantly different now if steps A,B, and C were taken over the last 20-30 years.

DR: Absolutely. So we do have to ask Why is it that despite attention, finally, tor acial disparities in health, there hasn’t been more improvement. And part of it is because the kinds of questions that Sarah and I have been talking about just haven’t been grappled with. The old ways of doing science have continued: the old conceptions about race and predisposition with regard to racial difference that haven’t taken into account fully in an interdisciplinary way the way in which genes and environments—and that includes a broad range of social environments—interact and also without a real grappling of inequalities of health and healthcare and living conditions at the same time in the US.

DL: The impact of something like the soda tax has on the people of the city of Phila – which will actually come out in a few years probably.

DR: We could also look at the fact that Philly has the highest rate of poverty of any big city in the US, there are differences in education, access to schools, access to healthcare, just so many differences. I think this is another area whether there could be such productive collaboration between biologists and social scientists because biologists, even genomic researchers, are looking at gene-environmental interactions – what variables are included in the social environment. And sociologists are better at identifying that whereas the genomic scientists are better with the research on genes.

DL: Going back to the ancestry piece for a second – that is something that all of a sudden has become this issue that people want to think about, but it’s almost like it’s become a monopolized business that this point, it does provide a little bit of a benefit but it does provide a negative as well.

DR: Yeah, well I can give an example where many people see it as a huge benefit to understanding their history and their identity, and that’s African Americans who use ancestry testing to try to trace their roots back to some part of Africa, and they see it as making up for the loss caused by the slave trade. There are no records left. Slave traders and owners did not keep good records of where the enslaved Africans’ heritage was. So most African American’s can’t trace their history back. There’s a disconnect, a rupture caused by slavery, so for many African Americans being able to turn to a ancestry testing company to give them some clue, even if they know it’s not absolutely accurate. Many think that it gives them some clue. For my book Fatal Invention I interviewed Rick Kittels who founded the first of these companies, which still exists, African Ancestry, and he said to me, we can’t give people exactly definitively the place where their ancestors came from maybe 200 years ago in Africa, but right now we’re in darkness and I’m providing a little bit of light on this. And a little bit of light is better than none. And I have criticized the scientific flaws on this technology, and myself, I personally believe that African American’s identity is rooted more in a common struggle for racial equality; cultural, social, and political factors. But I do understand how many people see these technologies as providing a lost part of their identities and somehow reconciling them, to use Allendra Nelson and other sociologists’ term, she sees it as a reconciliation project, with this rupture from the past.

ST: So I understand that but- also I’m concerned about the limitations of the science. And the expectations that people have. Because I’ve gotten a lot of emails from people in the African American community who’ve written to me and said “I did this test, and I went to one company and it told me I was from this ethnic group. And then I went to another company and they told me I was from a different ethnic group. Which one of them is true? The truth is -- my research is focused on characterizing genetic variation in Africa, across many diverse SubSaharan populations. This hasn’t been easy. It’s taken 15 years where my lab actually goes to Africa, works with African collaborators, goes to the field, goes to really remote areas, to collect DNA samples and information about variable traits and disease, brings them back to the lab and characterizes genomic diversity and what we’ve seen is that there’s a huge amount of genetic variation in Africa. You can have more variation amongst two ethnic groups in Africa that you have between someone from East Asia and someone from Europe. That’s how diverse the populations are there. So number one, that kind of blows apart any idea of an African race, there’s a huge amount of variation even amongst African populations. We’ve shown that African American ancestry, and this is no big surprise, based on what we know from the slave trade records, is mainly from West African populations. But because we’re still lacking in knowledge of diversity in Africa, how can you tell anybody their ancestry when you haven’t even characterized what those ancestral populations might be, and secondly, the differences are so subtle amongst these populations it’s very difficult to say with certainty, and furthermore, the reality is, once we start using whole genome sequencing, and now there’s technology that’s advanced to the point where the costs are coming way down to the point where we may actually be able to do that, that’s going to be informative, but what we’re going to find is that if you were actually going to look at an African American person’s genome, you’re going to see parts of the genome coming from many different ethnic groups and regions in Africa, based on the history of the slave trade and the fact that there has not been assortative mating since they were brought over as slaves.

DR: I think the claim that any of these companies can trace someone’s genetic or ancestral roots to a specific tribe in Africa is subject to all of the flaws and criticisms that Sarah’s laid out. However, and I absolutely agree with that, I am not a promoter at all of using these to determine what your ancestry is, and I also have a political-social view on it, and that is that genes are not what make up someone’s identity, but I’m just expressing though, for many people it’s not just a biological question; it’s a matter of being able to have some grip on a part of their identity that they see as having been stolen from them by the slave trade.

DL: And for some people, they may see it as a last resort in terms of understanding their history, even to a low degree. And going off something Sarah that you said a second ago, and you’ve been doing this work for quite some time, and it seems like you are even still the tip of the iceberg on this. So for you to be in that position and have some of these companies saying they can definitively say yes absolutely, it’s very hard to put a lot of credence behind it.

ST: Yes, but I do agree with Dorothy in that sometimes having even just a little bit of knowledge is better than having no knowledge, and I do believe the technology is going to advance in the future such that we’re going to get better answers at least. I think it will be possible in the future.

DL: but the important thing is also to have the understanding you have be multiplied by all researchers across the U.S. and world to change that mindset so you take in a lot more factors that have been being taken in over the last several decades.

DR: Yeah, and also for that to reach the public because Sarah’s research along with the sociological research and historical research that shows the race, the concept of biological race is (was) invented by scientists. Sarah’s research as she says is exploding the idea that there is one biological race of Africans. Her research shows, as she has just said so eloquently, that it’s been known that Africa is the most genetically diverse place in the world, and now her research is showing that it’s even more genetically diverse than we thought. That’s an example of how combining good genomic research and sociological understandings of what we mean by “race” can help to, finally, do away with these false concepts of discreet biological groupings of human beings that always tend to be placed in a hierarchy and end up supporting these very very dangerous views about human equality: who’s the same, who’s different, who’s superior, who’s inferior – it’s important for the public to understand the science of human genetic variation as well as the politics of it.

DL: And that may be the toughest part of it, is having them understand because of so many other venues and paths that are out there by various people right now.

ST: And you also have to put this in perspective. Because if we look at the entire genome, there are 3 billion what we call nucleotides, or letters, that form our genome, and out of, if you put it all together there’s let’s see, less than .1 percent difference among all modern humans. And that’s reflecting the fact that all modern humans are thought to have arisen in Africa within the past 200,000 years, and then migrated within the past 80,000 years or so. From an evolutionary perspective, that’s really really recent. So as a species, we’re really similar, and that does not mean that differences don’t exist—they do. The fact is that people migrate to new regions and mutations still occur, so you do get differences accumulating, but relative to the entire genome, it’s a very small amount.

Transcript provided according to WCAG 2.0 Requirements, Guideline 1.1 Text Alternatives, University of Pennsylvania 2017