 Thank you very much for that wonderful introduction. I'm really honored. I'm honored for that introduction. I don't know if I really earn it, but also I'm honored and very happy to have the opportunity to present my research here at Berkeley. I actually never really, I just made it to Berkeley once before for one of the Indian meetings and I'm happy that I made it here today, even though it was a little bit of a hassle with all the traffic that I had to face. So today, I mainly want to talk about my work in South America, especially the Central Andes, addressing some of the research questions that we've heard about earlier. So my main interest being an trained archaeologist and biologist or what here would be a biological anthropologist is in population history. And population history is right of a very broad term that came up several times in the history of our discipline, but it's not well defined. So I want to start to think about kind of what my definition or broader definition of population history is. Of course, paleogenomics today and ancient DNA research just recently in the last three or four years had a very large boost. There's big studies coming out nearly every week where we learn new things about populations, population histories, migration movements and interactions of past populations all over the time. And one critique that I sometimes have with regards to the work of my colleagues is that often enough some of these more population genetic studies kind of elect the expertise of anthropology. Not considering several other factors that would be necessary to actually formulate models to be tested that integrate the cultural record, the archaeological record, anthropological theory, ecological data in an equal way. So we often see very simplified models or theories that are tested. Of course the nature of models and the nature of the biological discipline is to operate in these simplifying manners. But on the other hand, since we deal with our own species, we know that we are more complex. So my overall goal is at one point to maybe find a way to address the complexity as good as possible to integrate all these interdisciplinary data points into my own research and also to create tools that allow other researchers to do that as well. So for me population history has to do with human diversity and the forces that shape human diversity. And that of course, especially for our species, is the interaction between culture and biology. Culture has become part of our biology. And besides earlier approaches of a biological anthropologist to see culture just as something that is an adaptation to specific environmental or whatever pressures, we now know that culture actually is a driving force in our own evolution. Because our decisions and our behavior has massive impact on the environment that we're living in. Meaning also that our behavior can actually alter the selective pressures that act on all populations. Overall meaning just that we have to address questions of human population history and human diversity from a more ecological and feedback perspective or from the perspective of human niche construction that tries to see the role of culture in shaping also the human biology. So there's different aspects of course that go into it. We have evolution, dispersal, one of the things that I will mostly talk about today, general diversity and of course it all relates to health and public health and past but also in contemporary and future population. So one of the main questions I tried to ask in my research over the years was how does ecological and socio-economic change influence human genetic diversity and demography? And demography is a good point to start with because demography is indeed and nobody will probably say something against it a good way to bring culture and biology together. Demography as in basic is of course biology as reproduction and death are biological events but also of course demography is formed and shaped by all kinds of socio-economic and cultural features that we have to consider. So of course there's many ways to address these things I can look and the bioarchaeological repertoire has a wonderful set of methods that we can use to learn about past populations and maybe also about contemporary populations related to their health, their diversity, their diets and all these factors. My specialization here of course is to look at the genetics part or the genomics part of our species. So many or for a long time the major approach to understand human genetic diversity has been by looking at contemporary populations and compare these populations in a synchronous way to each other by that it was possible to reconstruct major events in the overall population history of our species very well but on the other hand of course this approach is blind for events that might have altered or shaped population structures and patterns in the past so whatever we see now might be actually a product of quite recent events and might not actually reflect what was going on in the prehistory of those populations. So this is where actually we paleogeneticists come in because we can try a diachronic or a long durée approach to population history. By that I mean we get the ability to actually look at different points in time to see how genetic diversity was distributed in the populations that we observe on a geographic or global level and we also get the ability to actually monitor other things than just population history but also human genetic adaptation to specific environmental and physiological stress factors over time not relying on models that we need to kind of reconstruct and infer past events but actually maybe having the chance to directly look at populations at the time where something occurred. So the strengths of our approach that I see that cannot live without the modern genomics just doing ancient DNA alone will not give us the answer but adding up ancient DNA gives us that gives us the depth and the detail that we didn't have before to that extent. So as I said before modern DNA and first evolution and diversity from modern populations and tries to create models to understand what happened in the past. So we just have actually one point in time where we measure and we take this data that we get to try to reconstruct what might have happened there. And as I said before taking ancient DNA gives us the opportunity to actually look at different points in time and get an idea of if whatever we inferred before is actually true or if other events happened that shaped the population structure diversity or specific events that had selective impact on human genomes in the past. Of course only ancient DNA comes with some downsides we're limited normally in our sample size that we can work with since there's normally not so many skeletons around and then there's the additional problem that of course ancient DNA means dealing with degraded DNA. So even if we would have all the skeletons of the world it is possible that only a few of them actually contain suitable genetic material that is preserved for our analyses and that it is accessible for us meaning other factors have to be considered like contaminations that might derive from modern humans interacting with these samples and also other microorganisms that can lead to major problems analyzing the data that we do. So of course I speak here as if ancient DNA is a new discipline it isn't. It is around for about 30 years now. I think the first of course explorations into ancient DNA were done here at Berkeley and then the pioneers like Swan to Table and others brought the discipline all over the world and started to do their research also on questions that are relevant for us more archaeology interested people and the discipline grew and grew and grew and grew and lots of research was done but most of the research was done with technologies that we kind of abandoned in the last eight years. So even though that's not completely true we still rely on the old methods but when we have new approaches that give us fascinating insights that we couldn't have before. So I don't want to go too much into details because that will probably be boring for most of you since it is technology driven but just to give you an idea of what change in our discipline in the last years. So normally most of the ancient DNA studies that have to be have been done until recently just looked at small proportions of the human genomes that you were investigating. Most of that mitochondrial DNA so DNA that is only inherited in the maternal lineage that actually only gives us insights into maternal population dynamics. So of course if I only look at 50 percent of the population I might get a completely wrong picture of what might have happened in the past and then these limitations were not just because all the people that are involved in this science decided oh yeah that's everything we need but also because mitochondrial DNA is the most easiest target that we can aim for as a paleogeneticist since mitochondrial DNA which is found not in the nucleus of a cell but in the tiny mitochondria that make up kind of the energy center of our cells are there a much higher quantity meaning that in the living organism already there's a ratio of about 1000 mitochondria to one nucleus meaning 1000 times mitochondrial DNA to autosomal nuclear or chromosomal DNA and that also of course means that after death there's a higher likelihood that mitochondrial DNA is preserved than genomic nuclear DNA. So people were aiming for that and the technologies that we used in the past PCR Sanger sequencing and so on were well suited for this kind of approach but again we could only target specific regions of the genome that were normally quite small. Then around 2005 a new technology came up called next generation sequencing or there are several other names but this technology was quite rapidly integrated into the ancient DNA canon of methods and you all probably know about the first outcomes of the introduction of that technology into our field and that is the sequencing of the Neanderthal genome something that would have been impossible using the technologies on this side before. So now it was possible to use a technology that actually allows us to get the million fold amount of data out of the same sample which in the first moments sounds super amazing and I know that many colleagues use that kind of argument to just say like yay that is the solution but million fold more data of course doesn't always mean that all this data is usable for the end user. So there's complex biochemical and then computational approaches that are necessary to actually mine through the data to get what you are looking for but still now some years later enough development has flown into these technologies that we actually can use this method for all these fascinating things that we were thinking about not only looking at tiny parts of the mitochondrial genome but going for whole mitochondrial genomes and not only mitochondrial genomes but actually being able to sequence whole genomes of ancient specimen or to target large regions of the genome that are informative for whatever we're looking at no matter if it's population history or evolutionary biology. The other big benefit of these new technologies is that you can actually target DNA fragments that are much smaller that we could target before so one of the first observations in this technology shift was there are many samples that we gave up in the past because we were not able to access any genetic information and now this technology allows us to actually get information out of this sample. So it doesn't mean just because we didn't get DNA in the past that there was no DNA but that the DNA was broken down to so tiny pieces that we could simply not access the DNA and that is something that is possible also now using these new technologies and other factors like contamination risk and all these things can be lowered applying these new technologies. So of course for our discipline that was the huge array and this is also what kind of started this huge amount of paleogenomic work that came out in the recent years and is populating nature and science in other journals. Of course if you're working more in a field that is kind of interested not in who was the first or questions related to major regions of interest like Europe like me and you're more interested in later historic and prehistoric periods in the Americas you might not make it to nature and science with what you do but it's not about that it's about the accumulation of knowledge that is now possible that we didn't have before oh I didn't know that I have that feature okay so this is just actually what I just said so compared the classic approach to the new approach it simply allows us to access data that we were not able to access before we now have the chance to actually look at genomic information that gives us the same insights into maternal as paternal population dynamics what is of course crucial and also something that we couldn't really do before we can now look at subtle population dynamic processes as a mixture before that we kind of had to infer migration scenarios from seeing that maybe the diversity in one population changed and more lineages of another kind came up over time over time but that kind of always means that there must be kind of a significant demographic event that leads to an exchange in the population but it doesn't really show us if things normal human things happen like more people from one region go to another region and sexually interact or reproduce with the local populations there these are subtle genetic signals that we couldn't get before now we can actually get those things we don't have to wait for the major mass migration but we can actually see if there's a moment in time where maybe there is more gene flow between a region let's say the Central Andes and the Amazon region or something like that so that brings me to kind of my major field of research the population history of South America or more or less actually the Central Andes and what I have to say directly in the beginning is that whatever I present you here is work in progress and it's work that is in progress for now 10 or 11 years that simply means that or should give you an idea how complex the population history of the continent is and how hard it actually is to get into those subtle structures that I'm interested in if you don't just want to look at who came to the Americas first and at which point in time but if you really want to see if socioeconomic complexity is a driver for gene flow or not or is coincident with demographic changes if conflict or interaction between past populations led to also shifts in the populations that we observe then it gets a little bit more problematic especially if you work in fields like the Americas settled last by our species and also settled by a very restricted population and population size meaning we face a very low amount of overall genetic diversity so and then that whole thing got shut up nothing nobody really comes into the Americas until the European contact maybe there were Vikings or something like that before but we know that they never had a real actual impact on the Native American gene pool so that means we are working with a population where we don't really see like large scale shifts and changes as you see them in Europe where constantly new groups are coming in leaving signals that are very different from what we observed before and the same for mainland Asian so in the Americas we face the problem of low genetic diversity what makes it a very hard field but also kind of a poster boy example to develop new methods that are more sensitive and sensible to actually detect these kind of fine-grained interactions between human societies and populations so South America is quite well studied when it comes to the overall number of individuals that have been analyzed in the past when you look at this graph it doesn't really matter you see and it's actually hard to see I'm sorry for that that there's large numbers of individuals analyzed here that where the data already has been published that have been analyzed for mitochondrial DNA or for small fractions of mitochondrial DNA and this is kind of the main record that we have to talk about population history in South America for a long time there were not many people interested in South America maybe because it's so complicated maybe for other reasons maybe because most of the samples you get from the overall continent are very bad when it comes to the DNA preservation so there's actually only besides me for a long time three or four other people that did ancient DNA work there and most of the data are actually there I don't want to brag but it's simply because I didn't see a reason to shift to another continent is accumulated by me but it comes with the restrictions that I just mentioned and now the number of you know new data qualities that I just mentioned is growing so we have a large number of complete mitochondrial genomes now that give us completely new insights into the mitochondrial genetic diversity or maternal genetic diversity and there is also a large number now of genomes or genome-wide data that we generated in the last two to three years with my colleagues together from Boston David Reich that we are still working on to actually be able to publish so again here the results that I show are kind of previous results because it turned out even though we have this wonderful new tool now and we can look at the whole genome and not just mitochondrial DNA that it is the genetic differentiation between the populations is so low that we actually have a hard time to differentiate and to distinguish on the level that would be of any kind of interest for archaeology so the oldest samples that have been analyzed so far are about 10 000 years old and we have a pretty uneven spatial distribution meaning that most of the human remains that have been studied come from the central Andes and this is something that is mainly driven by preservation for the overall region of the Amazon basin there's not many human remains that are available and on the other hand we have access to samples from Brazil mostly the overall environmental conditions we're not favorable to preserve any kind of organic or genetic material so it's just a mineral matrix and then now there's a growing body of data coming from the southern cone of South America there's simply population size is probably not as much not so many archaeological sites that are known with human remains and so that of course is a limiting factor and everyone who knows in the who works in the Andes knows that there's no place in the world that is probably that crowded with bioarchaeologists like the central Andes there's no chance that you go from one little hamlet to the next without meeting a German or US archaeology crew and a bioarchaeologist and so on so we have a pretty good idea of the overall record what also is favorable for me because it means that we have a rich amount of data from other disciplines that I can use to build my models that I'm working with so over the years I gathered a rather large collection of South American samples we have around 1800 samples in our collection now at UCSC coming from different regions in South America reaching from around 10 000 years to the European contact when you look at this map those tiny blue things here I did not remember diamonds are samples that are actually like early hollow scene maybe like Pleistocene that we already analyzed we have sites iconic sites like La Ricancha Arroyo Seco and other sites there and then a lot of other sites all the red dots and actually that's not enough dots but it was impossible to show them all that represent places that we've sampled and as you see of course there is this major dominance of the West of Western South America so in the last years besides all my other work we started to use these new technologies to kind of restart or redo what I've done in the past to see if the higher data resolution can tell us something that we didn't know before and so we were able to as you saw before now generate a large number of complete mitre genomes individuals for which we have whole genomic data and actually there's an arrow because you see that I just managed to put two 1 500 years more into my own timeline and actually that is right I just remember because we just got back new radiocarbon dates for a sample that we have from from Patagonia from close to Arroyo Seco that actually dates 11 500 years old and had some DNA preserved and we are hoping that we are able to publish that soon so what did we do is all the data that we got so what you see here in this map is simply a distribution of those samples for which we were able to get genomic genome white data not only mitochondrial data the color code shows that of course there's a dominance of later periods the middle horizon the late intermediate where we have the most samples from and it gets a little bit thinner when we go into the early periods but that again is driven by the archaeological record and all the probe but also by the preservation that we find so that paper that was already mentioned earlier where we took the mitochondrial data that we generated to look at the initial peopling of the continent even though I was saying like oh I'm not interested in those questions if you have the data you of course also try to contribute to those questions and so my colleagues at the Australian Center of Ancient DNA Research and I looked into the maternal mitochondrial diversity of the initial Native American populations and if we can use that information to get a better understanding of the initial peopling process and the subsequent demography of the overall Native American population and that is possible because we actually can use mitochondrial DNA in phylogenetic ways so we can generate trees and when we look at these trees and see specific events of diversification we can actually correlate these other events with demographic change because let's say as a simply very basic approach you can say that more or intense diversification also means that the population sizes have to increase and loss of diversity often correlates with the overall loss of population size in a population in a population so we use that data and manage to get a much better calibration of the initial peopling event than what people got before using only modern DNA because we can use our ancient samples actually at different points in time to calibrate what we call molecular dating and so what we found is that there was indeed of course an initial population that lived on in Beringia before it entered the Americas and there was also something that has been called the Beringian standstill before meaning that we had populations isolated for amount of time in Beringia before they were able to enter the Americas before that the time range of that event was between 20,000 and 30,000 years what is a very long time period what we figured out in a parallel another group from Copenhagen looking at another quality of DNA is that this event actually didn't take that long but that we can narrow it down to a period of 3,000 and 8,000 years what is kind of crucial information when you think of the overall ecology of that specific area the possibility to sustain populations there for a while and so on so this overall event was much shorter also another thing that we found is that the so-called founding lineages that make up the Native American mitochondrial gene pool actually all kind of diverged or evolved around the same time before we used the calibration that we used there was a large gap between some of these lineages when they might have been involved actually giving many people a hard time to believe in one initial wave of immigration or a very let's say short and rapid mode of migration into the Americas knowing now that we have this narrow time frame and also all these lineages occurring at the same time means that we can really kind of reconstruct the initial founding population that existed the people that lived together in a specific amount of population structure isolated from the rest of the Asian continent and at one point as soon as it was possible made its way into the Americas most likely along the coastal route because we also see that there's a diversification event that happens before the inner route actually opens so we see a massive increase in genetic diversity that we can correlate with a massive increase in demography and meaning or in our interpretation meaning due to the rapid threat of the populations throughout the Americas actually a chance is given for populations to occupy their own niches establish their populations establish new amounts of genetic diversity and by that also demographically go up to larger sizes much more as if they would have remained on the in Beringia so one other big question that normally comes up when it comes to South the people in South America is which routes people took and there's been lots of discussions if people just moved along the Pacific route or actually if they split it up when entering South America going straight into Amazonia and some taking the Pacific route or maybe just a complete inland's route so what we did is we took all our very old or the oldest individuals that we had and generated full genomes of these individuals and then compared the genetic data and tested several scenarios how individuals in the past related to each other and one wonderful thing that we could rely on was that we had all these very old individuals along the Pacific route and we also have some ancient individuals from the Brazilian highlands that we could look at plus of course all the modern genetic data that is available and so what we did and was actually not even I can read it here I'm sorry for that is that we modeled around and looked what scenarios the most likely scenario for a route and what we found is that populations in Arroyo Seco in the early Holocene are more closely related to populations in the central Andes than they are to let's say Brazilian populations meaning that the first idea of a route that follows along the Pacific is the model that is supported by the genetic data but we also see that even in Arroyo Seco and especially if we go a little bit further to the north we find signatures that must derive or that are more closely related to modern Amazonian populations saying that there must have been a distinction between those populations and actually there must have been two routes that at one point or down here kind of meet each other and start to admit with each other again so this is kind of at least from the genomic side what we see as the kind final confirmation of the two route scenario meaning with one initial population of Native Americans maybe there was somebody before we don't know that yet but at least for this timeframe we have a highly related population that must have diverged from each other around 10,000 to 12,000 years ago was isolated from each other for a while only down here actually these populations meet each other so we took also this genetic information of course to learn a little bit more about our later populations that we analyzed especially from the central Andes and what we find and what we see in this tree is actually confirming what I mentioned before so Bodigiria York, Wari Highlands, Lauri Cacha what I cut here are all central Andean populations dating to different populations Lauri Cacha is relatively old about 8,000 years old and the Middle Horizon in the late intermediate period are relatively recent in the Andean prehistory so we see that all these populations branch together here in this tree meaning they are highly related to each other meaning that it doesn't seem like there have been any large-scale population replacements in the Andes at least from the time when Lauri Cacha started to exist until the European contact and we again also see that they are more closely related to Patagonian populations than they are to Amazonian populations that we have here so what we thought is very interesting is the question of genetic continuity in the central Andes and so we took all our individuals coming from different time points the oldest being 8,000 at this point where we did the analysis and the youngest being about 900 years old and what you see in this heat map is the amount of sheer genetic drift with the modern indigenous populations in this region that kind of simply means the genetic distance to those populations red being the lowest distance that we can observe and what we see is no matter if it's 8,000 years ago or 900 years ago the closest related populations that we find in South America are always the central Andean populations so that speaks for a long period of genetic continuity and also with to a very restricted gene flow into that region from other regions in South America what is very interesting I think at least and then we find and this is not in this graph that was the colonial period actually at mixtures promoted and this overall distinction that we see here gets a little bit lower and there seems to be more gene flow between regions another thing I don't want I probably already go over time is and this is a hard to see is with all these diversifications that I mentioned and this is one of these trees that I mentioned before showing different lineages mitochondrial lineages and you see that there's specific times here in the tree where we have like many of these new branches that start to exist and then we have times where there's where it's more quiet and not so much happens one thing that we also observed in our studies is that actually there are very distinct patterns of genetic diversity that we find all over South America but also especially in the central Andes meaning that we find lineages that are only represented in the northern Peruvian Andes and we find lineages that we only or mostly only find at the central coast of the Andes that of course points to a limited gene flow between those very adjacent regions and also points to the fact that there needed to be some time in an event to establish these distinctions between the populations again pointing to a rapid mode of the peopling of the Americas with groups settling at specific niches and niches and establishing there for a while without having major interactions with other populations at least when it comes to reproduction that of course doesn't mean that no trade happened and then at the end of the pre-Columbian era we actually can use the same technologies that I mentioned before and the same techniques to see that there is something that has a massive impact on the Native American demography so of course we all expect that because when we look into the historical records we see all these things that happen but before we did that in 2011 actually other studies just using modern mitochondrial DNA of Native American populations never found anything that could correlate with those historical records talking about mass mortality of Native American populations coincidentally correlated with the European contact so if you use the same calibration technique and throw in ancient data here you actually see the demography this is the demographic development how the populations enter the continent the population sizes go rapidly up in a very fast mode until they reach some kind of plateau and then here where I for whatever reason covered the timeline with this text field but believe me it's about 500 to 600 years ago so it's coincident with the European contact we have a massive decline of the population size of the female effective population size we're only looking at mitochondrial DNA here that means that there was a population decline of about 50 percent of the overall female Native American population coincident with the European contact so that supports many of the things that we can know from the historical record gives us also the chance to quantify the impact and one other thing that we learn when we look at this blue thing up here is that whatever caused this demographic impact was actually quite fast acting so population sizes were lost pretty quickly and also that the population the regular population diversity and demography kind of catches up or regenerates pretty quickly afterwards that means not that of course the actual census sizes directly went up and we know from the historical records that that would not be true but that means that there was a chance for new diversification happening meaning that whatever led to the population decline that period that major effect is not acting anymore that means we have to look at something that is fast acting and inflicts mass mortality and that brings us back to the scenarios of disease transmission that we of course all heard about before but we of course also know that the classical scenarios of immune avity and so on don't hold true that's not how our immune system works but when we combine new pathogens with of course socioeconomic and political struggles that come of course with the European contact and the European colonization and also internal conflicts if we just think of the Andes and the civil war raging between the two Inca emperors around this time we have enough conflict that weakens the population and makes it kind of prone or more vulnerable to these kinds of effects so as I mentioned before of course normally I try to concentrate more on the central Andes in general and even though we see Machu Picchu here and I have a project together with my colleague Richard Berger at Yale where we're looking at the burial population found at Machu Picchu I will not really talk about Machu Picchu here there's far more interesting things than Machu Picchu so why are the central Andes so wonderful to do my kind of research the central Andes are wonderful first of all because they're so overpopulated with anthropologists archaeologists linguists and so on that you have this rich data that you can rely on and then also simply because they give you unique environmental features that you hardly find anywhere else in the world you have these closely adjacent extremely different environment different environments the coast the high altitude mountains and of course the Amazonian rainforest that comes with completely different environmental pressures and factors acting at the human populations living in those regions of course this has major impact on the archaeological record and what we can observe in the development of socio-economic complexity and subsistence strategies but of course also has impact on selective factors acting on the population here so we have the high altitude Andes where of course their physiological stresses acting on the populations over since the first people started to permanently settle at high altitude that Andean population adapted to over time and then we have the chance to actually kind of compare these populations to populations that only live 15 miles down the valley and show completely different patterns in their genomic structure even though they're so closely related to each other so we have a wonderful situation actually to look into in C2 evolution in our human population and then of course when it comes to socio-economic complexity and all those things I don't have to tell you how interesting is to see like how different size styles of subsistence and so on inter correlate with demographic development then we also have a very good climate record again you probably cannot read what is there but this is more kind of a slide to say like we have so many climate records that we can use that is cool we don't really have that for many other regions and the number is growing and growing so we also can throw that into our models and look at specific things and then of course we have an amazing cultural diversity that we see through the sea through the Colombian period we have a very kind of rapid mode of increasing socio-economic complexity in the Andes much more rapid than in many other regions of the world ranging from a kayak and maybe peri-american periods to early states in the middle horizon to the Inca empire so we have all these kind of rapid socio-economic and political changes that of course will have had influence on the demography and since the periods in between are also very short that allows us to get an idea of the impact on demography and genetic diversity that we can hardly find in other regions so as I said I'm working in the Andes over 10 years now and got quite a bit of data to look at and of course most of this is mitochondrial data so the first question is even though I said before we see overall genetic continuity in the Andes is there something happening when we just look at the Andes and the wonderful thing that you might see in this map is yes something is going on so what this map is simply showing us is Peru first and then the genetic distance of the past populations at a specific point in time from the modern Peruvian population and what you see is that here at the oldest point in time here marked at 2500 BCE so where we have large populations that actually cover all of Peru not of course populations were there but where we have large sets of samples where we can actually model in space because we have data points all over the country we see that low distances to the modern populations I actually just found in the south while populations up in the north in the central Andes were actually quite distant from modern populations and then we see that these things shift over time and actually what we read out of this development is there's a process of genetic homogenization or loss of genetic structure that happens over time that might of course correlate with human activities migration and social political changes that correlate with the onset of the warrior empire or the incursor something like that we also see that kind of or what we this graph here suggests is that whatever kind of is the dominant pattern of genetic diversity that is dominant today kind of derives from the south from south central highlands and so on so again we took our genomic data and I'm sorry my graphs are all so bad just to show we thought like okay that is only mitochondria data so something might be there that we oversee so we looked at the nuclear data again showing us the demography of and genetic diversity of male and female population components and what we find if we throw some genomes in the pool is actually kind of a comparable picture we see here all the older ancient DNA samples that we have branching closely together in the tree and then we have our youngest ancient DNA sample that we had in the mix that dates into the late intermediate period around 1100 AD actually being closer to modern Bolivian Ketcher and Amara populations than those ancient in general but still this tree shows us that all these populations closely branched together meaning yes there is continuity in the Andes there's not much going coming in from the outside for a long time but there is a substructure in the Andes so there is population dynamics happening that are visible in the genetic structure of those populations over the period of 8 000 years that we're looking at so we did another test where we tried to model whatever is happening and when you look at this weird tree here you actually see here in this red dot all our oldest genomes that we have from the central Andes and over here you have the modern Amara and Ketcher populations and right here in the middle again we have our late intermediate Peruvian population that we're looking at and we get a signal we get an estimate of it mixtures saying that at the late intermediate period at this point of time actually the population had about 40% ancestry of these older populations here that come from all over Peru all over the central Andes and 60% of the ancestors of the Amara and Ketcher were me being more the southern Peruvian southern central Andean events so again showing that there's something going on people are migrating through the Andes and it seems like the dominant source of this process actually comes from the ancestors of what are the Amara and Ketcher populations speaking populations today I don't want to do the same mistake as any other in many other populations geneticists to directly refer linguistic groups to actual populations so actually before that but it fits good to what we observe to get an idea of what might have been happening in the past we had this wonderful Palpa Valley time transit project where we were able to look into a very wonderful archaeological record over 5000 to 6000 years and where we also had the possibility to study populations from the high altitude Andes populations from the lower valleys at the coastal desert and populations deriving directly from the coast and to see over time in a diachronic manner how these populations interact with each other and to correlate that with the archaeological and the climatological record so one thing that you have to know about this region in southern Peru the most iconic cultures that we know there in the pre-colonial time or the pyracus culture from about 800 BC to 200 AD and then the Nazca culture most famous for their gigantic geoglyphs and of course because people thought that these geoglyphs could just have been created by aliens and not by humans being dominant in these lower valleys there and then in the upper valleys we have the influence later of the early early state societies like the Wari empire and so on so the first thing I can say is we found no alien DNA and it's good to say that because there's lots of YouTube videos and so out now by I think the guy's name Brian Bauer I forgot his name who studied some or still thinks that these wonderful pyracus skulls that have these intentional skull deformations that are pretty amazing cannot be actual human beings but must be either aliens or nephilim meaning the children of angels inter sexually interacting or reproducing as humans and so this guy is around since a while and always tries to find people that do this genetic work for him to support his claim actually quite recently lab took his samples and analyzed it and got no results and he took that first no first time he got no results and he used that to support his claim that it must be aliens because if they're aliens we cannot read the DNA so we can think you know when we teach our undergrads how to approach the scientific methods that yeah and now they found European DNA on a sample which came from the United States from a collection and has been there for 100 years so lots of things could have happened to the sample and they found a European haplotype and now he has this new idea that actually these are kind of early Europeans that were brought over by aliens or the angels or something like that so I can try to show you tell you I analyzed about 150 paracas individuals with wonderful skulls and when I got DNA the DNA preservation is very bad it was always native American DNA so what I found looking at this time transact is that actually we were able to reconstruct the genetic relationships of those populations over time to each other don't worry so much about the numbers the main observation that we actually made and that is interesting is that at the coast and at the footholds we see population continuity from the earliest times that we were able to sample until the middle horizon meaning that there have been no larger migrations that came into that region from regions that are genetically distant that are shaped or were kind of part of major cultural transitions that we observed there but then in the in the late intermediate period the latest period that we're looking at actually the genetic distance increases rapidly so something must have happened the populations from the middle horizon are absolutely distant from the late intermediate periods speaking for or suggesting population discontinuities and the other thing when we look at the highlands is that the coastal populations are actually quite distant from the highland populations even though we talk about geographic distance of 50 miles but going up 3000 to 4000 meters into the upper valleys and then in the late intermediate period actually there is no genetic distance between the highlands and the coast anymore meaning that or suggesting that maybe a major migration or population replacement might have happened from the highlands to the coast so we modeled a little bit around and took our climate data that we have and the settlement density data and what the settlement density data first chose is here blue coast green highlands is that we have kind of growing population sizes or settlement sizes into the nascar period and then at the end of the nascar period actually there's less settlements the region is nearly abandoned in the middle horizon so the time of the barry state and then in the late intermediate period there's a huge amount of new settlements that are there so of course that kind of fitted with the observations that we made in the genetic data but that of course just is coincidence and doesn't say much and also what we know from the climatic record and that's not shown here is that these coincident with these major settlement periods here there's shifts in the overall precipitation that we can observe so we are at the desert fringe of the desert and the desert margin shifts up the Andes and back through to the coast over time meaning that there's periods where the lower valleys are more favorable to occupy and periods where they're less favorable this gap here actually falls into a period of nearly complete densification meaning there's no direct precipitation at all and also the precipitation in the highlands so what runs down the valleys is pretty low simply showing that that might have made it hard for people to survive there and to sustain populations and then in the short period where our population sizes go up there's actually a 250 year period where precipitation becomes increases massively and we actually have direct precipitation down to the lower valleys meaning availability of water that of course reminds us very much of climate determinism a little bit differentiated view on that but let's say when you live on the fringe or on the desert margin the availability of water is of course a factor that you have to consider especially if you want to sustain growing populations in more complex societies what we did we took our genetic data took the settlement density data and all the other data that we had and generated several models that we then tested using Bayesian statistics to see if there's something happening and if we can actually see that migration happened and what we found is that there's actually a massive migration percent where about 25 percent of the coastal population was exchanged by highland populations 25 doesn't sound much but in genetic terms that is huge that is a mass migration event and not a subtle trade diaspora or whatever event and what we also find is that there is an earlier migration event that is much smaller happening around the end of the Nazca culture so people are leaving the area we expected that and it seems they are moving actually up the mountains but don't have enough impact at this time to actually leave a signal in the high altitude populations of a mixture that kind of lowers the genetic distance that we observe but then this large amount of new settlements seems to derive from larger populations that come from high altitude going down to the coast and what we also see is also what I thought was very exciting is we get a date where actually the highland populations and the coastal populations split from each other so what we see is that it's not something that results from the initial peopling of the region so some people moved in the highland some people moved to the coast but actually that the split that causes the genetic distance that we observe is pretty recent in terms of the prehistory and happens around 1400 to 1180 BCE so in a period that we call the initial period actually the period where all over the central Andes peoples for the first time shift to larger scale agriculture and agriculture has been around for long and the bad thing with South America is all these things that we have in Europe with the neolithic transition don't work there because it's actually a subsequent and slow process but here's the first time where people actually rely on staples mace and so on on a large scale and so I don't have an explanation for that and that is actually something for the archaeologists but what I think based on my understanding of migration theory and human behavior what we also have around this time is the formation of more distinct cultural features and all that stuff so there is differentiation that is visible in the subsistence strategies of people between the highlands and the coast of course but also differentiation in the cultural features and identity that might be related to that is often one of the major factors that limits the possibility for people to actually reproduce with each other so limits gene flow so maybe what we see here is kind of what borders do to people separating them from each other even in a genetic way but of course I know that there's no strict borders around this time and not that the archaeologists just like get the pitchforks and get me out of the building here but you know it's interesting to think about what could have initiated that split around this time so I'm just talking about coincidence and coincidences of course nothing more than coincidence so these nice slides here are just what I said before in a more graphical way I should show that before so we have the split that happens between the populations and then when we take together the climate record we see that an initial migration from the coast to the highlands happens in a period where we have actually have prolonged droughts around the coast and probably more favorable weather on the eastern sides of the Andes at a time where actually also the first states are established there so we have Wari and Tiwanaku in those regions increasing political policies that of course also attract people to go there because there's labor there is resources so it must not be the climate that drives the people but actually changes in general political structure and subsistence that of course to some point affected also by the climate but actually maybe living up there was more tempting around this time and then several hundred years later we see the reverse movement and that kind of correlates again with this time of precipitation at the coast and the climatic record also shows that we have prolonged droughts actually in those regions occupied by the Wari and the Tiwanaku so we have large accumulated populations living in these states relying on very complex systems of agriculture to sustain the population surplus economic systems that have to shift in their activities because the environmental factors alter and we know if it's driven by the climate or not that of course the Wari Empire and the Tiwanaku Empire collapse and those political entities don't exist anymore and they're centralizing factors also not so all these factors might be triggers for populations moving from A to B moving to other regions and then if you have regions that actually turn out to be quite beneficial to sustain a population because there's lots of water and in general it's nicer to live at low altitude than a higher altitude these kind of mixture of climatic changes and social political changes might have driven massive migrations from the southern highlands into the rest of Peru and by that we come to some kind of event that might also explain all the things that we observed before this homogenization that we observe all over Peru that is driven by a southern Indian genetic signal so the interesting thing behind it is it is pre-inca so it is not the large empire that seems to drive this homogenization and it also seems to be post-Wari and Tiwanaku so it's also not those empires driving that so it's actually the collapse of the centralized systems and the advent of small-scale polities and completely changing interaction between regions here that are not controlled by larger powers anymore that might be the trigger for these kind of population movements that we observe and that also of course means that this is pre-European context so that actually here before the income for the Europeans we actually find the genetic diversity in central Indian populations that we still find today so the modern genetic diversity of these populations is shaped by these pre-Columbian late prehistoric events sorry that I used the term prehistoric I know that's not good when I talk about South America but as a trained or also trained European archaeologists that kind of the appealing term for me to talk about that so I'm coming to an I see that I'm already 15 minutes over time so I just that is the main thing that I wanted to talk about before and then I just wanted to say that of course there's all these other things that we can look at it's not only population history and archaeology of course but also this wonderful chance to now where we have this genome wide data to look at human evolution and human adaptation and one thing that I'm actually interested in as I said before how biology and culture interact and shaping the diversity of our species nowadays something that relates of course to the concept of niche construction the feedback effect rather than the linear or just evolutionary perspective on how diversity is generated and so actually just have it as teasers here you already heard that I did research into high altitude adaptation and hypoxia and that is of course and so far interesting because the idea that populations established quite early at high altitude around 10,000 to 12,000 years ago quite shortly after people entered the Americas is quite stunning for many people especially when you have been to high altitude on your own you know that you have all these rather negative physiological reactions to being in at a high altitude mainly driven by hypoxia so you might suffer from mountain sickness cerebral indema edema and so on and on the long term it actually also has an impact on the demography of the populations because hypoxia leads to reduce birth weights and increases the chance of still births in those populations so the question is if populations actually manage to establish their 10,000 years ago and when we think of all these well-known cultures and groups and archaeological groups from South America they're actually all high altitude populations something must have happened that allowed them to survive there something must have happened to enable them to survive in this more less hostile environment for humans and so of course one thing can be genetic adaptation but genetic adaptation is rather slow so what we found is that actually genetic selection and adaptation is happening over a long time frame but to reach the genetic diversity or the genetic patterns and adaptations that we find in modern Indian populations it actually takes a very long time we see the same frequencies for specific genetic markers that we analyzed in the ancient populations in modern populations popping up around yeah also the late prehistoric period meaning that there must have been something else that allowed those people to survive and that brings us back to niche construction so there must be cultural adaptations economic subsistence adaptations that actually facilitate genetic adaptation over time so human activity and behavior gives time for the genetics to actually act slowly and adapt and of course there might be also other features that we're just starting to explore there's of course not just the genome and the culture we all know now that there's things like the epigenome that we can look at as a direct mechanism that modulates the way how genetic information is translated and stands and isn't is directly informed by environmental factors and also microbiome so all these organisms living in us and on us that have a major impact on population on the environmental fitness of an individual and the survivability of an individual so there's lots of things to come but what we can say is of course that there must have been things majorly driven by economic adaptations and cultural adaptations that have facilitated genetic evolution and that is interesting because that is the wonderful example for niche construction showing how our behavior and how our culture has become a part of our own biological diversity and then of course there's the question of disease transfer I told you that we found that we have these massive decreases of population sizes coincident with the European contact but we actually what we want to figure out next is if it was really just the diseases which diseases were it and also if there was a differential mortality in Native American populations so if let's say some populations or individuals had specific genetic markers that made them less susceptible to these introduced to these introduced pathogens meaning if there was some kind of selective impact that acted on the populations too so not only an undirected loss or mass mortality but actually directed impact on the populations that massively shaped the gene pool of those populations towards a direction of individuals that have a reduced susceptibility to these kind of diseases and that of course is something that is also of interest when we think of public health and especially of public health of Native American populations today because it can help us maybe to understand some symptoms that we observe in a much better way than we did in the past concluding everything I hope I managed to show you that actually South America and especially the Central Andes are very dynamic when it comes to their population history and that we still are far from really understanding everything but that we that I probably have enough work for the rest of my career to work on this topic and and that what we observe there and what observe the genetic diversity and the general human diversity in Native American populations of South America and of course also subsequently in mixed populations is driven by social and natural environments meaning that there is not simply a scenario where one specific genetic makeup pops up and is continuous until the next tops up that but that our behavior in our interaction with the environment actively shapes the genetic diversity that that actually these processes also shape the central Indian gene pool and so far that it gets very homogeneous and that this is not related to the major empires or the European contact and yeah and that there's lots of works to do to actually dig through those patterns of low genetic diversity and a relatively short population history to design tools that help us to understand that better and then ultimately actually also design tools that allow a transdisciplinary approach to all these things which then would in my eyes finally allow to actually do something that is worse being called population history thank you very much and I of course want to thank all these wonderful people that take part in my research that is of course not just a one person effort it's my own lap it's the Max Planck Institute for the science of human history in Vienna Germany the Pontifikat Universität Katolica de Peru the right lap in Harvard and the Australian Center for Ancient DNA thank you sure you'll take a few questions before we move in and then the rest yeah I'm very sorry that I shortened the time for questions of course thank you for an interesting talk I'm interested in the response to hypoxia because in the Himalayas there has been noted a set of mutations in the prolilla hydroxylases that are related to hemoglobin production in terms of hypoxia sensors my understanding is that those sets of isoforms don't exist in the Andes what genes do you use as an index of response to hypoxia given that the hypoxia sensor genes don't seem to have mutated in the way that they did in the Himalayas no that's really a fascinating question especially one it points to the fascinating aspect of altitude adaptation so that we have convergent evolution we have these distinct populations that adapted to high altitude in different ways and of course also have different genes all the same genes involved but different polymorphisms that control that and of course Rasmus Nielsen who's here at Berkeley did some fascinating studies on populations from the Himalaya and I think he also works in some Andean populations and he probably would even be the much better colleague to talk with in this point so what we did is we looked at some genes that have been identified before in genome-wide scans to probably be under the influence of selection yes so it is EGLN-1 E-PAS so things that are involved in the broader metabolism and then another gene that is not directly included there that is NOS-3 which is part of course of the nitric oxide metabolism and now we expanded our panels actually on a much wider scale to include far more genes because the first approach that we had was relatively naive we just took whatever was identified in the modern record and looked if we see a signal in the ancient record but of course that limits us very much in really understanding the dynamics because as you mentioned there the the genes of the SNPs that we identified the polymorphisms that we identified to be under selection don't correlate with any direct physiological reactions so they're not neither directly translated in a way nor are they kind of structurally involved in any of these metabolisms so actually what we see is yeah they're selection acting but and we assume that it is correlated with altitude adaptation but it could also be something different and that is one of the major problems that you have with ancient DNA that of course we cannot look at phenotypes I cannot ask the measure nitric oxide content in all these things in one of my skeletons so I'm I have to look at modern population studies and take that and transfer that to the ancient populations and as you might know that even though anthropologists have been working on altitude adaptation since I think biological anthropology is around just in the recent years actually geneticists started to focus on a large scale level on this topic and they are still kind of a divide between those classic physiological studies and the genetic studies where those people don't really kind of interact with each other over don't read the literature of each other to bring together these things so I have to say right now I don't have a satisfying answer to that I can say that the genes that I analyze are broadly involved in these processes but actually the polymorphisms that we found to be under selection are in no way directly involved in the translation on the structure of the translation process so I know that there's selection happening I hope that I know more and as I said we now have a more widespread approach where we target all the genes that are part of these metabolic pathways we also have modern individuals that we integrate in our study indigenous populations that get a little girl a little bit give us some more diversity than what because one of the problems is that whenever you talk about genomics in the central Andes you actually normally just have like a handful of individuals that normally just derive from either something that was termed Aymara or something that was termed Quechua there's a little bit more diversity in the altitude studies but we don't have a good understanding actually and so now we have the benefit of interacting with Peruvian colleagues who actually have wonderful collections of samples from people where they have the whole biography of the individuals all the anthropological information that you want so you can actually be sure that these individuals come from these regions see their ancestry in these regions and so on and it's not just a diffuse thing that has been termed Quechua just following a major description of a language group and I hope that this will allow us to actually then also go deeper into these processes because there will also be diversity you know you will not only find differences between Taibetian populations and Andean populations you will find differences between populations in the northern Andes and in the southern Andes especially if these divides that we observed in the genetic data the population data mean that there was not much gene flow between these regions still they are all adapted to that thing so there will be also convergent evolution on the small scale that act is I assume any other questions yes it does to some extent as far as we can see it now as I said the problem with homogenization and the nuclear genome right now is that everything looks very homogenous because we have this kind of lack of diversity and actually it turns out that right now mitochondria DNA with what we have is a much better marker to distinguish because there's more mitochondria diversity that we can deal with between the regions than than nuclear diversity but this process that or this model here actually points towards what we observe in the homogenization so we see that there is some kind of genetic structure in the past and that these archaic and older ancient individuals are to some extent distinct from these modern populations and we also see that in the same period where we see this kind of mitochondrial homogenization process we have this period where actually an admixture event happens that brings in the ancestral that admixtures the ancestral populations of these Amados and Quechua with those older northern and central Andean population and northern and central central Andean populations that we observed before so this has the same direction that we observe from the mitochondrial data so we are talking about populations found in the southern central Andes here and as we saw it in all the mitochondria stuff and also points towards even though that's a mixture not homogenization but it also shows that this dominant that this signal from the south actually becomes dominant taking up 60% of the genomic diversity that we observe in these individuals so it seems like also in the nuclear data we see this homogenization process but we cannot completely and satisfyingly validate this hypothesis because we simply have to find better ways to mine the data and actually get those kind of subtle processes better pinned down in the data but it looks like it all fits at this point but again that is not of course the final verification of my hypothesis Can I add to that? Linguistics, Lev Michael who teaches here in the linguistics department suggests from his very detailed Pan-South American study and I heartbroken and I didn't think to invite him I should have you should talk to him he proposes linguistically that people are migrating from the south out of the Chaco out of this sort of Argentine eastern slope area that there seem to be a long-term relationship with the southern highlands and that area that they're moving north and that would reflect definitely seeing genetically and he's taking that from the language of relationships, the amount of relationships and that that would be his suggestion of where Imara actually is coming from that is very interesting because my linguist that I work with so Paul Hagerty and so of course have a complete difference then and even though we're in a think tank together by the Max Planck and where we try to actually have this transdisciplinary approach so we're linguist geneticist and so on there is this constant struggle between me and the linguist and our linguist or especially with Paul I love him but just scientifically love because you might have more and what he his models don't fit with the genetic data not at all he sees that the major empires are the major factors that distribute languages through the Andes and also of course says that Imara most likely has been distributed in the context of the Shaheen Pan-Andion development meaning of course with Shaheen we don't expect that there's kind of a state-like structure that majorly expands but at least that we see that there's some kind of mentality exactly and that this was one of the major driving factors and then that Kitch was actually distributed by the Wari Empire and meaning by that that Kitch was actually kind of distributed more in a political and military manner as in whatever organizing language in the Wari Empire and that of course if you really want to correlate language with population movement that wouldn't fit with what we see but again I think that sometimes of course there might be actually other effects that trigger the dispersal of language completely differently from population dispersals he's not alone the french linguists also have this southern no and that's very interesting actually because that would make sense when we observe that and when there is a correlation between population movement and language that would be a fascinating fascinating scenario one wonderful thing now that I know that is that I just last week was offered new samples not directly from the Chaco but at least I wanted the map here but I don't find a map here from the eastern andy so from the argentinian regions eastern slopes of the argentinian andy's also kind of a time transact ranging from at least 6000 before to inca period that would of course be a wonderful thing to integrate into this thing and see how these populations relate to our southern central andy populations and so on that's really fascinating I really need to read his work thank you for reminding me of that you