 Okay, well first of all I wanted to say how incredibly exciting it is to be here to help celebrate the 10th anniversary of the completion of the Human Genome Project, and I'm here to tell you a bit about how that's impacted my research personally, which is focused on human evolutionary genomics. So I thought I'd start out by talking about what I envision as some of the key challenges in human genomics research, now 10 years later after the completion of the genome project, and that includes better understanding of how the extent of the genetic variation and phenotypic variation that exists across a broad range of ethnically diverse people across the globe, and we're just at the beginning stages of understanding that. Secondly, to better understand what are the evolutionary processes generating and maintaining that variation, and lastly how to gene-gene, gene protein, and gene environment interactions contribute to phenotypic variability in humans. Now the focus of my research is on Africa, and there's a number of reasons for that. First of all it's thought that all anatomically modern humans originated in Africa about 200,000 years ago, and here these red dots are representing sites where there's fossil material for the earliest anatomically modern humans, and the oldest fossils are found in southern Ethiopia, dated to about 150,000 to 195,000 years ago. We also have evidence for very early modern behavior. This was an example of carved ochre found in southern Africa, dated to about 70,000 years ago, but there's been some more recent findings that are dating this back even further, maybe 120,000 years ago in South Africa. So it's thought that after anatomically modern humans arose in Africa about 200,000 years ago, there was a migration of very small numbers of people, was not too large, that then migrated across the rest of the globe. And so because it was a small number of groups of people leaving, you can imagine that outside of Africa there tends to be a subset of the genetic variation present in Africa, although of course, since this was approximately 50,000 to 100,000 years ago, there's been time for genetic variation to accumulate in other regions as well. So the pattern of genetic variation that we see today is going to be a reflection of our evolutionary and demographic history. If we were to look at identical twins, we don't see any differences. If we look at unrelated humans, there's about one nucleotide difference out of every 1,000 base pairs. And if we look at human versus our closest genetic relative, the chimpanzee, from whom we, our ancestors, diverged about 5 to 6 million years ago, it's about 1 out of 100. So since there are 3 billion DNA bases, that comes out to about 3 million differences between each pair of genomes. But there's a lot more to be learned, particularly about structural variation, like insertions and deletions, duplications and inversions. We are just at the beginning of characterizing that extent of variation, particularly across diverse populations. But all the studies have shown that generally, most of the variation that exists is within populations relative to between populations. And that simply reflects our very recent common ancestry within the past 100,000 or so years. So why are we particularly interested in studying Africa? Well one, if we want to learn about human origins, we need to go to the homeland, the birthplace of modern humans, which is Africa. If we want to learn more about the African diaspora and African-American ancestry, we need to be looking in Africa. Africa is also a region of very high prevalence of communicable diseases like HIV, malaria and TB, but increasingly there's a large prevalence of non-communicable diseases like hypertension and diabetes and so on. So if we want to understand the risk factors for these diseases, we need to be looking in Africa. And then lastly, as you heard from Dr. Green, there can be differences in how people respond to drugs. And those may be due to variation at drug metabolizing loci, and we need to have better characterization of variation in Africa to develop better treatments. Now despite the importance of studying African genomic variation, there has been relatively little work done in that region. And so for that reason, myself and my students and my collaborators in Africa have been going to Africa for over 12 years to do field work. We mainly focus on minority populations. These are groups that are otherwise not really being studied by biomedical researchers and they have very limited healthcare access. And for these we've collected now over 10,000 samples from blood. For many we have RNA, for many we have frozen plasma. And we get very detailed ethnographic information and information about things like diet and nutrition. I also want to point out that this can be exceedingly challenging work to do in the field. And we've actually spent a number of years developing these methods in our lab in the U.S. before trying to bring this to the field. Also equally important is training and capacity building in Africa. I've personally trained a number of African students in postdocs and one of the major goals of the H3Africa project, which is being spearheaded by NIH, is to do exactly this. And hopefully it's going to have a major impact on training and capacity building to do genomics research in Africa. And lastly, I can't emphasize enough how important I think it is to actually return results to participants. People rarely do and I can tell you they really, really appreciate it when we go back and they're genuinely interested in knowing. So while we're there, we're also not only collecting biospecimens, we're measuring phenotypic variation. Things like detailed anthropometric variants like height and weight, percent body fat, skin pigmentation. We're measuring cardiovascular lung and blood phenotypes. We're studying metabolic function and that will be informative for learning about adaptation to diet. And we're studying infectious disease status and immune response. So just to give you an example of how much phenotypic variation do we see? Well, this is from a study we published about a year ago and we're showing a strong correlation between hemoglobin levels and the altitude at which populations live in Ethiopia. So these are the high altitude Amharic populations and then these are lower altitude. We identified genes that are targets of selection and are associated with hemoglobin levels and they seem to play an important role in the HF1 pathway that is critical for hypoxia, response to hypoxia. Another example, people rarely realize how much variation there is in skin pigmentation in Africa but we see a tremendous amount that is not only interesting from an anthropological perspective, it's an adaptive trait but it also is going to influence things like vitamin D levels and perhaps folate levels as well. So there can be biomedical implications. These are principal components analysis of some cardiovascular traits and these were taken in Kenya and these are color coded by ethnicity basically or language family. And the point is you can see that people are clustering, groups are clustering by shared ethnicity generally with one exception showing that there are probably not only genetic but also environmental factors because these groups live near each other in regards to cardiovascular traits. We're also now starting to look at both directed and undirected metabolomics from plasma obtained from blood together with Charles Burant at University of Michigan. This is from 1,000 metabolites and we can see significant differences between populations that have distinct diets. So I want to tell you a bit about our study published a few years ago in which we characterize a broad range of genomic variable sites across the genome in over 2,500 Africans from 121 ethnic groups and in 98 African Americans from four regions of the US and then we had a comparative data set of more than 1,500 non-Africans. Here's the distribution of the populations we studied. It's pretty good but there are a lot of gaps as you can see and there's a lot more work to be done because there's over 2,000 ethnic groups in Africa. So first I'm showing you here the levels of genetic diversity in orange of the African populations and as every study has shown the highest levels of variation are in Africa and then we see decreasing variation as we move west to east across Eurasia into East Asia, Oceania and the Americas and that simply reflects our demographic history, series of founding events as we migrated out of Africa. This is looking at a principal components analysis based on individual variation. So each circle here actually represents a person and if they are genetically similar they are going to cluster together and the first principal component is distinguishing Africans from non-Africans. The second is distinguishing Native Americans, Eastern Eurasians and Oceaniens from the rest and the third principal component is distinguishing a very small hunter-gatherer tribe in Tanzania called the Hadza which we have been studying. We then use an approach which basically called structure which tries to infer genetically defined ancestral population clusters. Here we're looking globally and think of each color basically representing an inferred ancestral population. Each line is a person and for each person you can infer ancestry from these different clusters. So here are the non-Africans, they tend to cluster based on geographic location. Here are Europeans and Middle Eastern individuals, East Asian, Oceania, Native American, Southern Indian, Pakistani. Now here I just want to point out all the color, just showing you a lot of genetic diversity in Africa, a lot of population substructure. Some of the briefly major trends in orange are the people who come from western Africa, they speak what's called Niger-Cortifanian languages. Here the Eastern Africans in purple are people who come from northeast Africa speaking Khushidic languages and red are people who speak Nile-Saharan languages like the Maasai, they came from southern Sudan. And then you have these outlier groups like the Hadza hunter-gatherers and here are the Central African Pygmies or Southern African Saan. Now these three, we then repeated this analysis just in Africa and these three populations represent those that have been studied by the HapMap International HapMap Project and the Thousand Genomes Project. And I can say that without these projects we would know almost nothing about African diversity. It's made a major contribution. But as you can see, it represents just a fraction of the variation present in Africa and even in small geographic regions we're seeing quite a bit of variation. That's due both to their demographic history but also likely due to adaptation to very diverse environments in Africa and as you can see quite a bit of phenotypic variation as well. So what we saw in the African-American dataset that we studied, not surprisingly, is that most of the ancestry is from West Africa. The next largest component we see is variation in European ancestry ranging from about zero to greater than 30 percent and then we see very small components of ancestry from other populations in Africa or from Native Americans and other groups. Now, this is not a surprise based on what we know about the history of the slave trade but I want to point out a majority, a huge proportion of individuals who are part of that African diaspora originated from this part of West Africa which is near Angola. Did I get that right? Angola. And we have very little knowledge about variation in that region. In fact, we still have very little knowledge about variation across Western Africa. So if we want to be able to apply methods to study diseases prevalent in the African-American community, we need to be looking in that region. And lastly, I want to talk about our studies to identify targets of selection in the human genome and these are important because it's thought that mutations associated with disease and modern populations, common diseases may have been selectively advantageous in the past. So for example, hypertension and diabetes, obesity and asthma. So here's just a set of some of the populations we've studied and they have very different diets like hunter-gatherers, pastoralists and agriculturalists. They have lived in very diverse climates and they have very different infectious disease exposures. So likely they've undergone local adaptation. So the first example I'm going to tell you about is our study of the evolution of lactose tolerance in East Africa. It's known that in people who can digest milk as adults, they have expression of the enzyme lactase fluorizine hydrolase or lactase expressed in the small intestine, it can break down the complex sugar lactose into glucose and galactose taken up in the blood. But in the most mammals in the majority of the world, this enzyme is shut down shortly after weaning. And then we're not able, people who have this enzyme shut down, if they drink milk as adults, they can't digest this. And it can cause severe intestinal distress. Now anthropologists have noted for many years a strong correlation between the prevalence of lactose tolerance and the prevalence on practice of daring or cattle domestication. So the highest prevalence is in Northern Europe and then it decreases in frequency as we go into Southern Europe and the Middle East. Very low frequency in East Asia, very low frequency in Native Americans, very low frequency in most West Africans and hence it's very common in African American population. But in East African groups that have cattle, they have a high prevalence of this trait. Now a mutation was identified in 2002 that is associated with lactose tolerance in Europeans but the African populations we're studying didn't have it. So we did what's called a lactose tolerance test where you basically give the sugar lactose, you add some water to it. You have, this is a time test so it can be kind of challenging in the field. Everybody has to line up, has to drink it at one time. This is actually a group of pastoralists from Ethiopia that we just recently studied. And then you're going to simply use like a diabetes monitoring kit where you're going to take a prick of blood and you're going to do this at baseline. And then you do this every 20 minutes and you're going to measure glucose levels over about an hour. And you're going to look at the maximum rise in glucose. And from that you can define whether someone is lactose tolerant or has the lactase-persistent trait in light blue or if they're intolerant here shown in dark blue, summer intermediate. So doing that and sequencing a candidate region, we were able to identify three novel mutations that are associated with the lactose tolerance trait in East Africans. And those are shown here. They were located at about 14,000 base pairs upstream of the lactase gene in a non-coding region of a neighboring gene. They arose independently from the mutation associated with lactose tolerance in Europe due to what's called convergent evolution. And interestingly, although these are very common, they are very geographically restricted in their pattern, their distribution in Africa. So that's demonstrating that you can have a functionally important variant common but very, very geographically restricted. And we would miss it entirely if we looked in West Africa, for example. And then when we looked at signatures of selection, we saw pretty whopping signature in that we looked at genetic markers going up 3 million nucleotides. And here in red are individuals who are homozygous for the C variant associated with lactose tolerance. And they're homozygous for millions of nucleotides flanking. So that's consistent with very recent strong selection, increasing this to high frequency and dragging with it the neighboring variation. And we can estimate the age to be about 3,000 to 7,000 years old. And that corresponds with the introduction of cattle domestication into sub-Saharan Africa. So a great example of gene culture co-evolution. But perhaps one could argue that Mendelian traits are the low-hanging fruit. So what we want to look at now are complex traits. And perhaps one of the classic complex traits is height. It's highly heritable, genome-wide association studies, and thousands of Europeans has identified hundreds of loci, each explaining a tiny percent of the variation in height. And I should point out that most are not part of the growth hormone IGF-1 pathway. Now in Africa, we see some of the broadest ranges in height, from short-statured pygmies in Central Africa to these tall and thin individuals in southern Sudan and in Kenya. And it's thought that this variation may be due to adaptation to different environments. Now I'm going to tell you about our studies of the genetic basis of short stature and pygmies. And whenever I do this to audiences who are used to doing genome-wide association studies in tens of thousands of individuals, I'm going to show you why that's not possible. So to get to the pygmies, we have to cross this river. There's a guy with a hand crank, basically, who's going to take us in the cars. And because I'm a woman, I get some shade, so that's good. And a few other hazards we run into, but I'm smiling because the head is cut off here. But actually, I want to recognize Dr. Alain Framont, who actually is from currently at the Musee de Lome, who's been working with the pygmies for over 30 years and did much of the sample collections. So the groups that we're focusing on, and I should mention that human geneticists have been interested in this for over 50 years and with not much success in finding genetic variants associated with this interesting phenotypic variable trait. So we studied three groups in Cameroon whose mean male height is 152 centimeters. And then we studied the neighboring Bantu-speaking people whose mean height is 170 centimeters. Now, short people of short stature commonly referred to as, quote, pygmies, unquote, exist in global populations, primarily in regions that are in tropical forests. And for that reason, it's thought to be an adaptive trait. Now, there's a number of theories about why that's so, but I'm going to focus on this last one. And that has to do with the remarkably short lifespan of pygmies. The chance of living to age 15 is only 40%. If they make it to age 15, the expected lifespan is about 25 years of age. So because of that, there is selection for early reproduction. And in fact, pygmies, at least this particular group, are reproducing earlier and they're reaching puberty significantly earlier than other Africans. So the thought is, and then the growth trajectory seems to stop at the point of puberty. So it's thought that you're sort of trading off growth for earlier reproduction and earlier puberty. Now, there have been just a handful of physiologic and metabolic studies done. Many of these were done by David Ramoyne who sadly passed away in the last year and Luca Cavalli-Sforza over the past 30 or so years. But there's just a handful of these. So I want to tell you first about a study in which we genotyped a million single nucleotide polymorphisms. And here's where you're going to be really impressed by my sample size. In 67 pygmies and 58 of the neighboring Bantu individuals. And here I'm showing a structure plot in orange of the Bantu individuals and in green of the pygmies. And you can see that they're highly admixt. And so in fact, the degree of admixture is directly significantly positively correlated with height. So the more Bantu ancestry the taller they are, consistent with this being a genetic trait. The other thing we did with collaborators at Cornell is we developed a method. This is actually really tricky with pygmies because there's no ancestral pygmy population. They're all admixt. And admixture has been undergoing admixture with the neighboring Bantu for probably thousands of years. So we developed a method to scan along the chromosome and determine tracks of Bantu ancestry in red and pygmy in blue. And the point is these tracks of Bantu are really, really tiny. So there's been a lot of admixture. And you can see this region of chromosome 3 where there's less recombination occurring at a position 45 MB. And the next question we wanted to know is how do the genomes of pygmy hunter gathers differ from the genomes of the Bantu agricultural list or of the mausipostralis, for example? And we did genome-wide scans of selection. And these are representing places where we saw evidence for selection. But there was a cluster right in that same region of chromosome 3, about a 15 million base per region. And we have no power to do a genome-wide association study. But under the hypothesis that this is adaptive, we focused on those regions that are targets of selection and then looked for association with height. And we found a number of significant associations. And this was one of the top hits within that region of chromosome 3 and encompassed several genes, one of which is DOC3, which had been associated with height in non-African populations. So we've replicated that. And nearby is a gene called SISH. It plays a critical role in regulating cytokine signaling. At the time we found this study came out showing that it's associated with resistance to bacteria infection, malaria, and TB in Africa. And interestingly, it directly inhibits human growth hormone receptor action by blocking the stat 5-phosphorylation pathway, mice that overexpress this gene are short. So that raised the question to us. Could selection for something like immune function directly result in short stature and pygmies? The last thing I want to tell you about is a recent paper we published in Cell in which we sequenced, we did high coverage whole genome sequencing using complete genomics technology in 15 diverse hunter-gatherers, including five pygmies and then five each from 200 gatherer groups in Tanzania. We identified over 13 million variants. And over 3 million of these had never been described, not in thousand genomes, not in DB SNP. So that represents a big fraction of diversity just from 15 people. Many of these are known regulatory sites. Now combining the studies, we wanted to say what pathways are enriched for genes near targets of selection in pygmies. These include genes that play a role in neuroendocrine signaling, reproduction, metabolism, and immune function. And we also see an enrichment for genes in pygmies that play a role in pituitary function. Indeed, one of our top hits is thyrotropin releasing hormone receptor. And that was interesting because it plays an important role in regulation of the hypothalamic pituitary thyroid axis. It influences a number of potentially adaptive traits. And anthropologists have noted for years that pygmies have a significantly lower level of goiter compared to the neighboring Bantu groups. And thus, it's been suggested that they may possess a biological adaptation to a low iodine environment. So here, again, we have something like adaptation to diet, which is influencing these traits, these other genes that have pyotropic effects. And lastly, the last question we asked for the pygmies is we look for SNPs, or variants shown in green, that were specific to the pygmies and at very high frequency. We're going to call these ancestry and formative markers. We found about 25 clusters across the genome. But the most striking cluster was right here exactly in that same region of chromosome 3 that we had previously identified. And what was striking is that included 44 pygmy-specific SNPs in 100% linkage disequilibrium, over 170,000 base pairs. It contained a very interesting candidate gene called Hessex1, which plays a critical role in development of the interior pituitary site of expression of growth hormone and other reproductive hormones. We found a non-synonymous variant associated with idiopathic short stature in the literature, but that was present in other Africans. What is unique to pygmies is that haplotype. So my hypothesis is that there's something about that haplotype that is altering gene expression in that region. Now, nearby, we found another cluster upstream of a gene called POU1F1, or PIT1 in mouse. Again, plays a role in anterior pituitary development and is a transcription factor playing a major role regulating growth hormone expression. So these are great candidates, and I also want to point out that these are the most differentiated regions of the entire genome in pygmies compared to other groups. We then genotype SNPs at these regions in a broader range of Western and Eastern pygmies, and we found a statistical association with short stature. And the next step is we're going to be collaborating with other investigators to make genetically modified mouse models so that we can see what the effect of these are in vivo. So they lead to some interesting hypotheses. One is that alterations in growth hormone, IGF1 pathway play a role in short stature in pygmies. Secondly, for the first time, we suggest that development and expression of hormones produced by the anterior pituitary play a central role in the pygmy phenotype influencing many traits like growth, reproduction, metabolism, and immunity. And third, that short stature could be a byproduct of selection acting on pleiotropic loci. Indeed, if you look at the function of growth hormone, which is expressed by the pituitary, it influences IGF1 expression in the liver and that influences bone growth, that influences height, but it's also influencing insulin metabolism and fat metabolism. And infectious disease is altering immune response and those cytokines are altering gene expression of many of the genes in this pathway. And here in blue are the phenotypes that we're going to actually study in the field. So my point is you have to take an integrative approach and to look at this as a complex adaptive trait. So lastly, I want to also say what I think is another future challenge is to use integrative evolutionary genomic approaches to study complex traits. We ultimately want to identify genetic, epigenetic, and environmental factors influencing complex traits and looking at this complex web of interaction. We can identify naturally occurring genetic variants that influence that web, and we can look at how environmental factors like diet and pathogens and climate influence this web of interactions. And lastly, if we compare individuals who have similar genetic ancestry, but they live in an urban versus a rural environment, we can see what some of the environmental factors are in urban environment and how they differ in terms of these physiological traits that might be leading to some of these common diseases in urban environments. And then lastly, we can have the opposite. We can have groups that have very different genetic ancestries, but living in a similar environment and have different phenotypes. So in this case, the Fulani and Cameron seem to have an innate resistance to malaria susceptibility relative to the neighboring group. And the Saan and Botswana seem to have an innate susceptibility to TB relative to the neighboring group. So I think this approach will be informative for identifying those risk factors. And I'll just end by concluding, Africans have the highest level of variation. The demographic history of Africans and local adaptation to diverse environments has resulted in population and regional specific genetic variation. And there's an urgent need to include ethnically diverse Africans in genomic studies to identify both unique rare and common variants which may be a functional significance including those associated with disease risk. And I will just thank the many, many people who contributed to the study and the funding agencies and most importantly, the many Africans who contributed. And if there's any time, I'm happy to answer any questions. It was time for a quick question. There's somebody at a microphone, we'll go with this. Hi, thanks for the terrific talk. I wonder if you could comment on the practice that is emerging or in the last few years seems to be gathering steam in the field that I work in which is sort of psychiatric genetics of consortium studies that depend in order to get their ends up on collecting samples from all over. I mean, so the one I'm thinking of in particular collect samples from all over the United States and Europe arguably a sort of a Caucasian European ancestry sample but very diverse geographically. And I just wonder in light of your work in Africa what you think about that. I think there's an urgent need to be studying psychiatric disorder in Africa and there's so little being done. I can't even tell you. I mean, there are major urban centers. I think it's possible to form collaborations and to start getting information on this but certainly in the more rural areas, no information. I mean, you just, you don't hear about it. It's not being cared for, not being, they're being cared for locally, usually in the village. So I think that's really important and we should be looking throughout Africa and forming consortiums. Can I answer the one other, okay. Yes, I was astonished by the very abbreviated lifespan of the pygmies and I was wondering if you have a list of the five primary causes of death within that population. Well, I think the primary cause of death is in very high infectious disease burden. So we know they have a very high parasite load, for example. Some of the other common causes, they just have a very harsh lifestyle living in this dense tropical forest. For example, one major source of nutrition is honey that's way up in the trees. There's a lot of falls from high areas, believe it or not. I mean, it's things like that. I think I should stop here. I'm just trying to figure out whose phone it is. Let's bring it up. I think it's yours. Oh, it's not mine. All right, thank you very much. All right, thank you so much. That's terrific. Okay, our next speaker is David Kingsley from Stanford University. We'll talk about the molecular basis of evolutionary change, genomic insights from fish and humans.