Welcome back this is module H in our unit on Schizophrenia. This is a supplementary module on race, genetics, and ancestry. And I'm sure you've recognize, I certainly recognize that race has been a very, has a very problematic history in the, in, in genetics. We've talked about the eugenics movement in this course I'm sure most, all of you know that genetics has been used by various people over the centuries to try to justify as a rationale for racism. And it, so it's a potentially a very controversial topic. In this module, I'm, I'm not going to try to embrace that controversy. I'm, I'm really, what I'd like to do, race is, it although as you'll see, geneticists don't like the term race necessarily, but the, the concept of, of, of race is something that is important in current genetic research. And so what I've liked to do is really try to explain to you primarily at the conceptual level, although we're going to definitely look at some studies, why it is very important? And there's, and there's two things that I'm going to touch on. There's first of all, a methodological issue here with quote, endquote race. And that is, that in doing a genetic analysis, geneticists have to pay attention to population differences. We've seen that in the studies, the large scale genetic studies, that we've talked about in Schizophrenia. The Sanders' case case control study in the large scale GWAS study. In both cases the way they dealt with this, is they only looked at individuals of European ancestry. And I'm going to explain why they, why that's important to do, first off. Secondly, we're awash with new genetic data now, and we're learning more and more about genetic differences among human populations, and so I want to just highlight some of I think the more important findings that are emerging from recent human genetic research. And actually, you'll see I'm going to talk mostly about non-behavioral phenotypes, mostly about physical phenotypes in medical disorders. But le, let's first talk about the methodological issue here. Why in the Sanders study or the GWAS, are they restricting their samples to European individuals only? And to illustrate this, I just randomly, almost randomly, it's not completely randomly, I took a particular genetic vary in a snip called rs4680. That's a snip, and a gene called COMT. And again it's a snip. It has two alleles, and I think the two alleles are the A allele, and the G allele. And actually, COMT is a, is a, is a gene that's studied a lot in psychiatric genetics. And some of you will know that this particular snip is what's called the, the, the val Met COMT polymorphism. But I'm, I'm not particularly interested in that here. It turns out that I could look at a database, and I did recently I just went into what's called The 1000 Genomes Project, and I just looked at what the frequency of this particular A allele was in three large population groups, individuals of African, East Asian, or European Ancestry. And what you can see is that it's actually, not a typical, if you begin to explore snips in that database is there, are differences in the frequency of the allele. In this case, the A allele is much more common among Europeans than among African, and East Asian. Now given that, it turns out that given that difference, that if we didn't do any, if we didn't try to control for population background, the A allele would be associated with any phenotype for which Africans, East Asians, and Europeans differ on the average. If we didn't control for anything, and this is the methological issue. So is the A allele, for example, a skin's cancer risk factor? Because Europeans are much more likely to develop skin cancer than Africans or East Asians. Or is A allele an obesity risk factor? Europeans are more likely to be obese than east Asians or Africans, or maybe it's an eye color allele. Well, in fact, of course, it's not any of those things. It's, It's something that, that's involved in neurotransmission. It's not a skin cancer gene. It's not an obesity gene. It's not an eye color gene. It's only being associated with those things because it's associated with whether or not you're European, versus in this case, non-European. That issue is what geneticist called the population stratification problem, sometimes called ethnic stratification. And what pop, what they mean by population stratification is. If population groups differ in the phenotype, what do we mean by the phenotype? It could be rate of disease, rate of cancer, rate of hypertension, it could be a phenotypic trait, how tall you are, the, the color of your skin. If they differ in the phenotype, and they differ in the frequency of a genetic variant, for example a snip. If those two conditions are met, then that genetic variant will be associated with the all those phenotypic differences among the population. Necessarily, so unless we control for the population difference, even if the differences in disease or phenotype are entirely environmental in origin. So for example, prostate cancer rates differ across major ethnic groups. So do, so we just saw the COMT allele differed. Does that mean an, an, and as well as other snips will differ in their frequency across those groups, does that mean that those alleles help explain, why some populations are at higher risk for prostate cancer than others? Well, no. They don't explain it at all. And in fact, the differences in prostate cancer, I don't know this, but the differences in prostate cancer could be purely environmental. But nonetheless, we could get an association between a genetic factor, and risk for prostate cancer. Because these two conditions are made, met. Prostate cancer rates vary by group, and the frequency of a genetic variant varies by group. It's the problem of population stratification that leads geneticists to often restrict their sample to one population group. Unfortunately, that's often Europeans geneticists are trying to redress that limitation now, by looking at more groups. Often they will do things, in addition to just restricting their sample. But in any good genetic analysis there will be some attention paid to this problem because the problem is thought to be perverse. Perverse, pervasive. Sorry. Why is it pervasive? Well, we gets back to this notion here. Are the genetic variants in our genomes, do they vary across, in their frequency across the population? So the second thing I want to begin is, is to, in, in this module, is to begin to look at some recent human genetic research, and tell us, and, and talk about what it tells us about population differences. And the extent to which genetics are associated with those population differences. And I actually have five conclusion, I think that have emerged out of this recent research that I'm going to go through here. First of all, if we look at common genetic variants, the types of things that you look at in a GWAS. Those variants exist, almost always exist in all populations groups. Although, they probably exist at different frequencies. They're kind of like that COMT allele that I gave you in a, a couple previous slides. Here's an illustration of this from a, 2008 study. They took a large population. Of individuals from all of the major continental groups, and then they classified them by their continent of origin. Africa, East Asia, so on and so forth. And then they genotyped them on over 500,000 snips. And they identified whether or not the, the, the polymorphisms in those snips, the variants, did they exist in all five populations? In only four of the populations, in three, in two, in one? It turned out that over 80% of those snips, the variants, existed in all five populations. Common genetic variants typically exist in all population groups. They don't tend to exist in one and not the others. In fact, only less than 2% only existed in one population groups. So the genetic, those common genetic variants we carry, they're carried in virtually every population group in the world. They're just carried at different frequencies. Some populations of, of a variant might be more common than in others. That's the first conclusion. The second conclusion is actually kind of a derivative of the first. For those common genetic variants, the vast majority of genetic variants is within rather than between populations. And this might be kind of a, a, a subtle statistical point. If we took the totality of genetic variants in a population, we can divide that genetic variants between what exists between individuals in the same population, individuals both from Europe, let's say, or individuals both from Africa. And the portion that's due to differences between population, differences between Africa and East Asia for example. And this studies actual been done multiple times and the results are very, very consistent. If we do that, most the overwhelming majority of variants, in genetics exist within populations not between populations. Most of the reasons we differ from one another aren't because we our ancestors came from different continents. That only accounts for 10% of those differences. Most of the reasons we differ genetically exists within continents. Differences between Europeans, between, between Africans, between East Asians. That's the second conclusion. The third conclusion is, well, there are differences, right? I'd said that there are differences. Otherwise, the whole issue of population stratification is a non-issue, but they definitely exist. They're just a minority of the total genetic variants. But why do they exist, why did they ever come about in the first place? Well to summarize a whole lot of research on human evolution and migration there are two major factors that contribute to genetic differences among populations in the world. The first is Chance and the second is Selection to local circumstances. And to understand this we need to understand a little bit about human migration, human evolution. Here's a schematic of human migration I think you should take the years here with a, with a big grain of salt they're just a very approximate. Humans evolved, all the evidence now points to human evolving as a species, homosapiens emerging in East Africa about 2000 years ago. And so all human existed in Africa until sometime within the last 100,000 years, when a land bridge, and actually the land bridge is down here, when a land bridge emerged between the Arabian peninsula in East Africa and some humans migrated out and began to populate the rest of the world. Up until that time we all lived in Africa, but then Europe, Europe was populated, Asia, the Americas, Oceana. When we look today right the whole world is populated with humans. Why might humans in different region of the world differ genetically? First, chance. What are the major chance mechanisms that could lead to genetic differences? One is that when humans migrated out of Africa, by chance, a subset of all humans, right, not all the humans migrated, some stayed. In fact, probably the vast majority stayed. It was probably a small group migrated out. So you, we took a subset, and by chance, the subset is not going to look identical to the ones that remained. There'll be some genetic differences because we took a sample from a larger population. That's, geneticists call that phenomena drift. But as we, humans migrated around, a certain subset would migrate and they would have a, a subset of the genetic variance that existed in the population they migrated from. That's a chance mechanism that leads the differences among us. Another factor that's important, another chance factor, or new mutations, right? If, if we have a mutation that occurred in the world population in the last 50,000 years, then it occurred in East Asia or Europe or Africa, or maybe maybe even in the Americas. So that can lead, these new mutations that occurred in only one region of the world, could also lead, it's chance event, could also lead to differences. Non chance differences of among world populations are due to selections associated with local conditions. There's not a lot of example of selection, but there's certainly multiple examples, I'm just going to highlight one that your probably familiar with, which is skin color. Humans before they migrated out of Africa would all, in all likelihood have darker skin because dark color skin protected us from the sun that we were exposed to in the Savannah, but as we migrated out into the northern climes we weren't exposed to sun so much. And so that our skin could lighten if we inherited mutations that were associated with light skin. And in fact, there was an advantage to people living up in Scandinavia to have light colored skin because they didn't, weren't exposed to the sun much not like they, if they lived on the equator. They weren't exposed to the sun much, they weren't getting a lot of vitamin D, for example, from the sun. If they had dark color skin, but if they have a light color skin then they can absorb more vitamin D, so there was an advantage. So as we migrated away, there was an advantage to having light colored skin in the northern climes because of vitamin D relatively equated, so that began to contribute to differences in population. But they're not, there are other examples of that. They tend to relate, these, these examples tend to relate to things like diet or very localized weather patterns and those types things, infections maybe. [SOUND] The fourth thing, and this I think, for me one of [LAUGH] the more remarkable things, about the genetic differences among us. And I think it's actually very important. It's a little complicated, you'll see in the figure I'm going to show you here in just a couple seconds. The genetic indifference's between any two individuals we would randomly select from the population. Is really proportional to how distant their relatives live from one another. Here's, it's a very complicated graph, but here's a study that I think is a beautiful illustration of this point. This is a study by Novembere et all published in 2008 in Nature. And what they did is they took a large sample. In this case it's only Europeans. It's about 1400 Europeans and they genotype them, I think on 500,000 different snips. And each point here represents a different individual. So there's 1400 points here in this two dimensional space here. This is just a map of Europe where they're sampling the individuals from. So each point is an individual and what they did is they measured how genetically different. On these 500,000 snips. It's kind of elaborate statistical procedure. We don't need to understand that. But they just measured how genetically different each two individuals were. And then they plotted them on this two-dimensional space such that points that were close to one another were very genetically similar. Points that were distant were genetically quite different from one another. So, the distance between points here is proportional to how genetically different the two individuals are. And then the last thing is that they colored the points here, in this two-dimensional space, depending upon where the individuals were sampled from. So the, if they were sampled from the Iberian Peninsula, they're color-coded purple. And look at, they're all mapping here. If they're if they're sampled from Great Britain they're, I guess, I don't know, a pink or red color and they're all here. The Scandinavians here, the Slovak people here, the Greek people here, the Italians there. The remarkable thing here is they're actually able to recover the map of Europe, by plotting out the genetic differences here. How similar you are here in Europe is really a function, how genetically similar to how close your ancestors lived apart. If you're, if they're close then you're going to be more genetically similar than if you're coming from Spain versus somebody coming from Russia. And you're actually able to actually just recover the whole map of Europe. That leads to the last point. And, and maybe this is, maybe one of the most important points. And it's why geneticists don't like the term race. Race has a connotation, has a lot of connotations. But, one connotation is that gen, the genetic differences among populations exist in discrete groups. Individuals from Africa are genetically similar and different from individuals from East Asia, who are genetically similar and different from individuals from Europe. And it's, it, it suggest that there's these kind of clusters, these discreet clusters. But what this suggest is really that I, the genetic differences among us are really continuous. And they're continuous as a function of how far apart our ancestors lived. So, geneticists prefer the term ancestry which is really a term that reflects these dimensional, these quantitative differences. So that genetic differences between populations aren't discreet. They're continuous, ancestry, rather than race. The last thing I'm going to touch on before finishing this is so, we've seen and talked a little bit, we've, we've covered the methodological issue. But, have talked a little bit about the extent to which genetic differences exist among world population groups. And and the nature of that genetic difference, well does it matter? Does it matter for phenotypes? The answer to that is I think we know a little bit, but we don't know much at this point in time. We know for rare Mendelian diseases it definitely matters, and I think you probably know this before coming into this course. There are certain diseases like Tay-Sachs disease, these are just a couple examples. Tay-Sachs disease exists in, in multiple population groups, but has a higher frequency in Ashkenazi, IE East European Jews. Sickle Cell Anemia is an example of a disorder, a genetic disorder that exists only in individuals of African ancestry, because it was selected for, because in the heterozygote, it's a recessive disorder. In the heterozygote state, it's advantageous or protective against malaria. Cystic fibrosis is more common in Europeans, apparently for genetic reasons. So, for rare Mendelian disorders, and this is just a couple examples. There are many, many examples of, of this. These are probably mutations that occurred in the last 50 or 60,000 years, as people have migrated out of Africa. What about common disease? The types of things that are coming out of GWAS. Do those also differ across world population groups? This is a real complicated slide, but the, it, basically what this slide illustrates is the answer to that question is yes. Frequency of the risk alleles identified through re, GWAS. Some of them do appear to vary across different population groups. Here's, I'm, I'm only going to look at, I've illustrated I guess six different ones. I'll just look at hypertension for the sake of time here. In hypertension, what's plotted here is the frequency of, I guess there's about seven or eight risk alleles, each line here is a different risk allele. And the balls on it are how frequent the, the, the risk allele is in these populations. Some of the populations are af, from Africa, some from Europe, some from East Asia. Some are admixed like individuals from South America or Central America, I guess. What you see here is that, for example this, this third or fourth risk allele here it's hard to see but the risk allele is, is more common in individuals coming from an from African groups than from other groups. Other risk alleles might be more common in Europeans than others. Does this mean that for example, this particular allele here, does it mean that African's as a group, are genetically more predisposed to hyper tension than European's or East Asian's? Well, no. Again, remember coming out of a GWAS the effect is very small. And what's more, maybe on this allele, people from Africa have a lower frequency of allele and the European's don't. What's more is most of the heritability is missing, so we really don't know. So even though there's maybe higher risk here, we really don't know the whole genetic architecture. So we can't say if different groups are higher genetic risk for these common disorders at this time, but right, geneticists are learning more about this everyday. The point being, that I want to emphasize here is that the, the risk alleles are likely to vary, as we look across different population groups. This is not a, a, a course on medical genetics. It's a course on behavioral genetics. What about behavior, you may be, some of you are asking what about behavioral phenotypes? I could say with 100% certainty at this particular moment in time, we really don't know enough to say anything about whether or not genetic factors, common or rare, really contribute to population differences in behavioral phenotypes. [SOUND] So, summarizing this supplemental. Genetic variance, it exists mostly within, not between populations. 90% is within 10% with between. It's better characterized as by what geneticists call ancestry continuous rather than race discreet. Genetic differences do exist among populations. If you're talking about common variance, we're talking about differences in allele frequencies. Because of that, in a GWA study, there's always need to be some control for ancestral background or population stratification, there always will be some discussion in that, in that type of study. What do we know today about genetic contributions, to what in the US is called health disparities. The fact that different population groups do have different phenotypic risk for different medical disorders. We don't know a lot. For rare Mendelian disorders, we do know that different groups are at heightened risks for those disorders and that actually has important preventive implications. For common inherited disorder, as of today, we're only beginning to understand this as the GWA studies become larger in sample. Not only Europeans but hopefully Africans, East Asian, and other world populations. Thank you. [SOUND]
