this ucsd-tv program is presented by university of california television like what you learn visit our website or follow us on Facebook and Twitter to keep up with the latest programs you thank you very much and thank you for inviting me to be here for my first Carter meeting so these were the sorts of models that were around for the origins of modern humans in the 1990s so we have the extremes of multi-regional evolution a global evolution essentially of modern humans everywhere ancient humans lived they evolved through to modern humans with gene flow between the different regions at the other extreme the recent African origin model Out of Africa the replacement model arguing there's only one place where modern humans evolved and then modern humans spread out to replace these more archaic lineages outside of Africa and so about this diagram but this was prepared back in 1974 from me trying to summarize the results of my PhD thesis on skull variation in Pleistocene humans so I was using very primitive multivariate techniques and computing power to look at differences in skull shape and at this time then the antis halls were regarded probably by most workers as being direct ancestors or modern humans so I had the advantage of four hundred modern human skulls from Bill houses database and their skull measurements and I used that to show that a number of fossil skulls were within the modern human range this included the Upper Paleolithic people of Europe school five from the Middle East or mo1 from Africa and down here we had ancient skulls Homo erectus skulls from Java and China very different from modern humans in skull shape and two specimens one from Broken Hill in Africa and Petra Lana in Greece that seemed also very archaic and you can see the position of the Neanderthals so rather than the Neanderthals forming an intermediate between archaic you and modern humans as you'd expect if they were the ancestors of at least some modern humans in my data they seem to be going off in their own direction in terms of their cranial shape away from modern humans so this is the sort of view I've been putting forward recently for the last let's say five hundred thousand years of the story and I made the awful error here of combining a genetic tree with a fossil with fossil specimens so the background here is is a one version of a tree for mitochondrial DNA divergences based on complete mitochondrial sequences of recent humans and of Neanderthals and there's a coalescence estimated date around 400,000 years for the mitochondrial lineages from this data of modern humans and Neanderthals so if we map fossils onto these time periods we have of course modern humans and Neanderthals here and beyond four hundred thousand years in my view there was a common ancestral species for modern humans and Neanderthals which I call Homo heidelbergensis some in my interpretation heidelbergensis was a widespread species in Europe in Asia and in Africa it went in at least two eeveelution New Directions in Africa it gave rise to modern humans in western Eurasia it gave rise to Neanderthals anatomically modern humans are at the end of this part of the evolutionary sequence in Africa Neanderthals are at the end of this sequence in Eurasia but also there will be members of the lineage early on in Africa that are not anatomically modern humans that are on the sapiens lineage and there will be equally archaic Neanderthals that are on the Neanderthal lineage but are not yet showing all the features of the lake and tolls so when we look at Africa itself we've got Hydra against this specimen here this is Broken Hill CT scan of it and here's a modern human we've got a range of fossils from the Middle Pleistocene of Africa some of them we can take quite well some of them are not so well dated but we have to bear in mind that our sample of African fossils is actually quite limited geographically we know from the Middle Stone Age record the humans occupied pretty well at times the whole of the African continent but our fossil record is dominated by specimens from Morocco from the Rift Valley sites of East Africa and from southern african sites so there are whole areas of africa particularly Central and West Africa where we have no fossil human record at all for this period even though we know people were there from the stone tool record I think this is an important consideration but certainly for my interpretation and a number of not just my interpretation Africa is the only continent that shows transitional fossils between heidelbergensis and modern humans and here we've got a specimen from Ethiopia the her toes number one skull and a specimen from Morocco the ghibli hood number one specimen and these specimens do show mosaics of characteristics between modern humans and and heidelbergensis so in terms of looking at the pattern of evolution of modern humans in Africa I've certainly been changing my views through through the years one view of course is that there's really just a gradual change if we have a complete fossil record in Africa we will see a gradual transformation of hydaburg Ensis into modern humans as shown in this is the paper led by Tim white on the earth Oh specimen where hoto is interpreted as a as a straightforward transitional specimen in this sequence of heidelbergensis to modern humans but when we look at a single site such as all mocha bish in ethiopia these two skulls are dated within the limits of our methods to both about 195 thousand years ago they found a few kilometers apart but even in my PhD work in in 1974 these skulls had really quite different affinities although one to me seemed to be a modern human and we have parts of the skeleton that's indeed suggest it is in its main features but on o2 is a very distinct looking specimen which I can't call a modern human and yet there it is apparently alongside ah no one in Ethiopia 195,000 years ago so Phil right Myer has tended to say well this is variation perhaps within a single population I and some others have argued maybe variation is too great and it might instead indicate that we actually have separate lineages coexisting even in Africa of modern humans so in my original PhD analyses I really just had these African these ancient African specimens were in there jelly hood one omo 1 and amo 2 and we can take on my one out as a modern human but let's look at the rest of the Middle Stone Age sample in terms of skulls that's built up since I did my PhD so we have the single skull which was known before but not well studied we've got the younger loba specimen lightly hominid 18 we've got aleus springs the erode material it would - there - go with the old one her toe and and the bomb day specimen and the point is that for a while I continued to lump all these specimens together in a single Middle Stone Age human sample from Africa but as times gone on and Krishna I realized that was a major mistake because it obscured the variation that this material shows this really is showing a lot of variation in skull fall which makes me think about the pattern of evolution here and it's even more complicated than that because on my original heidelbergensis model hard up against this gives rise to Neanderthals in Eurasia and to modern humans in Africa but what if it doesn't just disappear when it evolves into these succeeding descended species what if it actually stays around alongside them that will complicate things - and and I've been working for a number of years on trying to date the Broken Hill fossil and the other associated bits and pieces of skeletons of several individuals from this site in what's now Zambia and using electron spin resonance and uranium series methods we're still working on this and also sediments associated with the skull the surprising results so far are that the material all seems to range from about 175,000 to 300,000 years old so whereas in the literature you'll see that Broken Hill is dated often at five 1,000 years the present evidence suggests that it's much younger than that so this might suggest that heidelbergensis is hanging on potentially alongside that evolving sapiens lineage in Africa worse than that it may have been hanging on in Eurasia as well but I haven't got time to talk about that and in North Africa georgette crew blends here and and there it can be demonstrated that there is a lineage of humans there potentially with some regional continuity here we've got Jeb Lee hood number one specimen maybe 160 thousand years old maybe even a bit older and here we've got Anna Tyrion skull from doris Sultan maybe a hundred thousand years old and it looks like there is perhaps morphological continuity between these two suggesting that there are regional trends going on in particular parts of Africa in terms of morphology and recently I've worked with a number of other people on an intriguing specimen from Nigeria that's only thirteen thousand years old from a site called Iwo le rue this is the oldest fossil human in the whole of West Africa it's only 13,000 years old and in my PhD it came out as a rather strange mixture of archaic and modern human features and I thought that was maybe just the way I'd studied it or measured it or analyzed it at that time in 1974 so I BRE visited the specimen where new dating work and in collaboration with people are Catarina Havarti geometric morphometrics and in the session the I APA meetings we presented the idea that this specimen actually was very distinct from from recent populations in West Africa and indeed showed some rather archaic features here it is compared with younger Loba specimen from nightly which is dated usually around 150,000 years old in Caterina's geometric morphometrics analyses this was the nearest neighbor in terms of cranial shape to iwo le rue and in the same session in which we presented our results at the PFAs anthem eating's is isabel Creve Coeur presented with co-authors including Allison Brooks the view that at a Shango in the Congo similarly in a late Stone Age context there were specimens from a Shango that showed archaic characteristics less than 20,000 years ago so a more complex picture here for african than we normally think of but i've been developing over the last 10 years a view that there isn't like a single center of evolution for modern humans africa review i used to have there isn't like a Garden of Eden in South Africa East Africa different parts of Africa contributed to what we call modern humans through movements of populations gene flow and exchanges of ideas so this is entirely a hypothetical model and we need to fit it in with the sort of detailed data that Allison and Ric have been putting together for the African climates but possibly with much of Africa very arid 150,000 years ago we would have small populations quite separate from each other isolated from each other then when North Africa certainly becomes more humid we perhaps have gene flow and maybe this is the time when the use of red ochre and the use of shale jewelry from Nosara shells spread over much of the range of modern humans at that time then when we come on to the last sixty thousand years we see again changes in aridity North Africa cut off and perhaps these populations arguably even who knows going extinct in some of these places and we have East Africa perhaps becoming a center for the eventual dispersal of modern humans Out of Africa so this is a up here is a diagram that Roger Lewin put together for multi regional evolution globally as it was in the 1990s and I'm suggesting now that this might represent the kind of pattern we've got for modern human evolution in Africa so the view I had 15 20 years ago was very much a punctuation of view of modern human origins that there would be one place in Africa where maybe an isolated population of modern humans really became modern physically and behaviorally genetically in quite a short period of time and then spread out from there I then moved away from that to the view that it was a much more gradual process in line with what I perceive to be the pattern from the I'll evolution in Europe now I think it's more like what I can called here a coalesce in African origin that different bits of the African populations are contributing to the behavior to the morphology to the genes of what we're going to call modern humans through time so a quick word on the spread out of africa and here we've got just a representative map from the genetics paper published in PNAS in 2012 though we can say that there are modern humans and atomically in china 40,000 years ago down in southeast asia 45,000 years ago reaching australia at least 42,000 years ago and at the same time they're over in Europe we can now place the arrival of modern humans even in Western Europe from fossil data and new dating work on those specimens to more than 40,000 years ago for Italy and for Great Britain so there's this quite early dispersal this brings modern humans potentially into the territory of these archaic humans that are living outside of Africa the implication that there was interbreeding when those populations were in some level of contact first of all the phase have been to breeding with Neanderthals maybe one phase maybe more we hear about that later and then over in maybe even over the Wallace Line in Australasia there was interbreeding with these other populations known as the Denisovans and signs of that gene flow showing up only in Australasia today so simple attempt to try to represent this and we'll have to leave out all these other species that we haven't been able to talk about but the suggestion now is that heidelbergensis doesn't didn't just go to ways to give rise to modern humans in Africa and Tony and O'Donnell's in western Eurasia it also gave rise to a third branch the Denisovans in Eastern Europe and then when modern humans came out of Africa they pick up some Neanderthal DNA perhaps in the Middle East or Asia a smaller group of them heading to Australasia pick up some Denisovan DNA on the way to Australia and even within Africa the possibility that an archaic lineage in Africa perhaps surviving members of hydaburg put in more archaic DNA back into the modern human gene pool in sub-saharan Africa so quite a complicated story and just to finish off with these different models of human evolution that I started with so in 1970 nobody thought there was a recent African origin people thought it was either a global pattern as in multi-regional evolution for model human origins or that maybe Europe or the Middle East or southeast asia were centers for the evolution of modern humans but the pendulum then started swinging strongly towards a recent African origin and I and many other people by about the year 2000 were confident enough to say that we can more or less talk about a hundred percent recent African origin for modern humans genetically behaviorally physically but now the pendulum has swung back a bit potentially and now we have to say mostly Out of Africa probably so on that note I will stop thank you I'm going to talk a little bit about genetics and particularly I have to touch on some of the highlights over the last couple decades just very briefly but it's well accepted that modern humans originated in Africa and spread out from there but what's really not well understood is the extent to which the ancestral population it gave rise to our species was isolated and to what extent different archaic forms may have contributed to modern diversity and I think answering these questions is going to be very important for understanding the origin of innovation how these innovations may have been shared or how they were uniquely adapted to one small area of Africa so I was thinking a lot like Ajit when I was want to take one step back and show how humans fit into the larger picture of primate phylogeny this is a tree a phylogenetic tree based on the latest whole genome sequence data and what you can see is that humans form this one branch most closely related to chimpanzees and certainly we are great apes but what I also wanted to point out there's something a little bit unusual about humans and that is if you notice the other great apes have subspecies or closely related forms that they ship they coexist with for example orangutans tzer's they're smashing Bornean orangutans gorillas there's three or four different subspecies of gorillas loose lowland eastern western lowland gorillas mountain gorillas which aren't shown and then of course in chimpanzees we bonobos and then four different subspecies chimpanzees and we stand alone here and so the question is where how did we become the last standing representative of the genus Homo and one possible answer is that or why are we alone one possible answer is that our success resulted in the demise of what might be considered subspecies I know Chris stays away from that word just closely related groups genetic genetic li differentiated but related morphologically groups and these could have been as we know very well from the fossil record sharing coexisting with us for long time but have since disappeared so let's let's take a closer look at the tip of that human or that a hominid branch and as has already been pointed out the fossil record is reconstructed and we can see the time at which presumably these different forms existed I've color-coded them to show that yellow is Eurasian form specifically and dark blue is African forms and the ones that are found in both places are our mixed colors and you can notice again the couple of major points that Homo sapiens is a very young species appearing only in the last couple hundred thousand years or and that the fact that there's overlapping this and these bars suggest that they were coexisting for quite some time so what happened to these other forms and how did we become the last standing representative of the genus well that brings us to the models also covered by Chris nicely and I do have Chris's picture up here and Milford Wolpaw who is known very well known for his putting forward the multi-regional evolution model which I find to be a very interesting model from a population genetics point of view where you have the gradual transition from archaic to modern form over the full range of all these different groups in Eurasia and Africa and this is seen to occur because they're connected they're all connected through this sort of trellis like pattern with these little blue arrows representing gene flow which is the steady process of migration of individuals from one group to the other interbreeding and then a very critical component to this is that there's natural selection so the traits that are favored are going to spread from one place to the next and the other traits that are locally adapted will stay local and so you'll have local differentiation as well as this idea that the beneficial traits that make us anatomically modern are eventually assembled in one morphological package through this process and so it's seen as you wouldn't even call these different species we'd call them just different forms of the gene of one species under that view now of course the extreme alternative view is this recent out of africa model or the complete replacement model which suggests quite differently that all of the traits that make us modern trace to one place a single population in Africa and whereas this model would predict that some of our genes would trace back to many of these different ancestors in the past all of our genes would trace back to one local group in Africa so those are predictions of the two models what kinds of genetic evidence has been produced in the past that made us so strongly go with the recent replacement model and kind of move away from the the multi-regional model well going back to I would call it a landmark paper in 1987 led by Alan Wilson and Rebecca Kahn and Mark stone King this is what is what is known as the indirect approach and that is were you serving variation in natural populations and then you make inferences about the past by looking at the shape of the gene tree that you construct from the genetic variation in this particular case the in the 80s they were looking at restriction fragment length polymorphism they reconstructed a tree I'm actually showing a tree from a paper in 2000 by alpha l'instant where they used whole mitochondrial genomes but the result was the same there the reason that they came down on the side of a recent African origin was that you can see three different reasons one is that the African lineages budget which I've shown in red are longer and they're more mutation 'le diverse which suggests there's been more evolutionary time in Africa for these lineages to evolve also the root of the tree is found among Africans suggesting that the mitochondrial DNA traces back to a single point in Africa a single woman at one point in the past that lived in a population in Africa and the final point that's just as important is that the time for all of this evolutionary change leading back to the root of this tree was very recent within the last 200,000 years which is very much within the timeframe of the period in which Homo sapiens originated so I've shown the mitochondrial tree now against the reconstructed fossil record and you can see that the all the lineages of contemporary humans we have Africans and on one side of the root and Asian Africans and others on the other side all tracing back to a common ancestor within the last 200,000 years there is no other lineages present in the modern pool that would represent perhaps what might be variation found in these other forms these archaic forms so this was a strong signal to coming coming from the genetic side that made such an impact in the way we think it's quite interesting to think what if we had found a different tree or looked at a different region in the genome first but this was the first it made a large impact some of you are old enough to remember 1988 when it would made the cover of Newsweek now another nail in the coffin of multi-regional evolution came ten years later Sante Pablo had also been working in Alan Wilson's lab trying to figure out ways to recover ancient DNA from fossil material and in 1997 svante and his group were successful in isolating DNA from three Neanderthal bones they sequenced a small portion of the mitochondria this called the D loop and they reconstructed a gene tree and they found that that the Neanderthals formed a clade or a group that was quite distinct from everything we knew that was segregating in modern humans suggesting there is no mixing no interbreeding again supporting the idea of this complete replacement at least with respect to Neanderthals now move forward another 10 years and certainly the technology had improved and so on today's group continued to work on this ancient DNA and particulars own here mitochondrial sequencing of the entire six thousand five hundred nucleotides of the mitochondrial genome in five different Neanderthals a more resolved tree for modern humans you can see again the same pattern very distinct from variation in modern humans as you know so unequivocally establishing that the anatole mitochondrial DNA falls outside variation in in modern humans now we take it forward on just one more year so as recently 2009 we were still talking about complete replacement with respect to mitochondrial DNA the technology in DNA sequencing has gotten to the point where it becomes more feasible to sequence whole genomes and in particular genomes of extinct forms and the complete genome sequences were obtained in draft form from Neanderthals as well as this mystery form called denise eva from a molar and a finger bone and i don't want to go into the details as i said but the the bottom line here is that non-africans and a group from Australasia or Oceania contain a small amount of their genome that comes that directly matches the Neanderthal in the denisa genome respectively and in order to explain that interesting pattern of sharing it was suggested that there were two interbreeding events one as anatomically modern humans first got out of Africa into the Middle East mixing with Neanderthals as second interbreeding events somewhere in Southeast Asia just before anatomically modern humans made it into New Guinea and Australia some forty thousand years ago and then a second migration from the Middle East to East Asia without inter without interbreeding with denisa now I these draft genomes provided compelling evidence for interbreeding but it was not the first evidence we had for interview for interbreeding going back to 2005 Dan Garrigan in my lab a postdoc in my lab was leading a project to look at genetic variation and an interesting locus on the X chromosome and when we reconstructed the gene tree for this region on the X chromosome we found a gene tree that was quite distinct in many ways from the mitochondrial gene tree for example there are two major clades one clade is entirely restricted to Asians and the other clade shows ever is in Asians and everyone else including Africans most of the most of the low-side we look at look just like this part of the clade where there's an African out group and everybody else on the rest of the branches but in this case it was Asians as the out group the route was in Asia and the time to the route was about one and a half million years so it's interesting to think if we discovered this kind of pattern before we discovered the mitochondrial pattern how we might think differently about about human origins from a genetic perspective but in this case nevertheless we hypothesized that this pattern arose through a process of interbreeding as anatomically modern humans got into East Asia perhaps mixing with Homo erectus and this little piece of the genome surviving in modern humans today but interestingly as many years later as the Neandertal draft genome came out we compared the sequences this divergent lineage to the NIA to the inner gel genome and it matched the Neanderthal genome so we were correct in saying that there was introgression of this divergent branch but we were not correct in that it turns out it was likely in the and earth all that was the offered this this branched into the genome of modern humans now I also have a grad student who is finishing up who is in his graduate career discovered a few more low-side that looked very much like this one which had an Oceania an out-group and another one that looks something like this he had suggested to me that this was these are other examples of interbreeding and introgression and I said well you've got you've got to do a lot to prove that but meanwhile the draft genomes came out he was proven correct in fact one case it was Neanderthal admixture in the other case it was denisa Denis van admixture but using these kinds of examples where we could identify based on certain characteristics of the DNA sequence what was likely to have interests we decided to pursue a pro an approach where we would do this in a computational way and screen larger regions of the genome looking for interests fragments and the theory and behind this is simple in that if we have a model of divergence where two groups diverge in the deep past they stay separate into separate lineages such as one leading to archaic so one leading to moderns and there's a period of short period of recent interbreeding where some of the DNA is now exchanged what the what the genetic pattern might look like in that kind of model is for example if we have these two bars representing chromosomes and modern humans and these in arc and archaic form in the first generation you have a hybrid where each chromosome one coming from a modern parent one coming from an archaic parent are form distinct full chromosomes but in the process of recombination and many generations later these orange chunks these archaic chunks get broken down into smaller and smaller pieces and so our was to look for these small little chunks of DNA by the signatures that were specific to the interbreeding process which I don't have time to go into but given that we didn't have a reference sequence for fossil DNA from Africa we decided to use genomics and this computational approach to scour the genome for these kind of inter dressed fragments so we applied our approach to a data set of sequences we gathered from a bunch of different African populations we found evidence for several integral haplotypes in different parts of Africa but the frequencies of the introgression part appala types were highest in the population the pygmy populations from central Africa using extensive computer simulations we first showed that the data were not consistent with a model of no admixture at all so we rejected the null model of no admixture and using a likelihood approach we were able to make inferences about the model that involved introgression particular contemporary African populations seem to contain about 2% contribution of genetic material from an archaic form that interests approximately 35 to 40 thousand years ago from a group that split perhaps as long ago 700 thousand years and the divin the distribution of these different fragments we suggested that the interbreeding may have been centered somewhere in central Africa now I just want to digress for one second here about it very interesting this story that came up this year that involved the Y chromosome and that is the discovery of a very rare and ancient Y chromosome that didn't fit into the known picture of Y chromosome diversity many many years of research on the Y chromosome told us it was very much like mitochondrial DNA although Y chromosome variation today traced back to a very recent single ancestor that lived in Africa about a hundred to one hundred forty thousand years ago this was this was the picture that we had at the time and it made a lot of sense because the most divergent lineages on the Y tree were found in hunter-gatherer populations like the koi song the pygmies which was very similar to autosomal patterns well this particularly lineage was discovered in an african-american man from South Carolina who happened to Smith this DNA to the National Geographic Genographic project which sees thousands and thousands of samples was very unusual did not fit on the tree and when we rate when we dated it we found that the tmrc a pushed was pushed back over three hundred thousand years ago and interestingly when we did extensive database searching to see if we could find where else in the world a chromosome like this might exist we were fortunate to find in Africa out of thousands and thousands of chromosomes that we searched through there were eleven that seemed to be very closely related to this chromosome and it turns out all eleven come from one little tiny region of western Cameroon from a farming group called the embo now I just highlight that because the embo live very very close to this site that Christopher was talking about the EO Lehrer site where there's intermediate or mosaic forms a half sort of intermediate between archaic and modern modern groups well so just to update where we stand today now we have to add another admixture process in in Africa along with the two outside of Africa and I just wanted to end by saying something about what kinds of models are supported by the genetic evidence certainly we can no longer believe that the right replacement model is correct even after 25 years of dominating as a paradigm it seems to have fallen and the new models that seem to be relevant our models that were also proposed many years ago based on fossil data I will talk I think that this is probably the most reasonable model it's a it's a replacement model with hybridization but I also I'll skip over that is there doesn't seem to be a single location for transition in Africa there's no single Garden of Eden there's a transitional forms that are spread very vastly across the continent therefore there's certainly the opportunity for gene flow among these morphologically diverged groups over a long period of time and I really like the idea Chris already described it where you have the multi-regional model now restricted to Africa so a different geographic scale but these traits that make us anatomically modern could have been flowing along this trellis network of gene flow among these morphologically diverged groups for quite some time until the anatomical form was assembled in a single package somewhere in Africa and I really like the idea and I think before we really fully be able to address this from a genetic perspective we need to find the genes that encode these anatomically modern traits and look at their evolutionary histories so I'll leave it at that I'm gonna tell you a little bit about the technology used to sequence the genome of Neanderthal and this Denis of an individual and what we have learned from doing that so before I start I just show the pictures and names of many of the key players and that's the collecting these data has really been kind of the lifelong project of this visionary Svante Paabo who was one of the first to imagine that DNA might stick around and very very old fossil material and invented many of the ways of getting that DNA out and sequencing it and many of the other people who helped in analyses along the way so Neanderthals by now need no introduction they were morphologically very similar to us yet distinct enough that from the early days it was known that they were something different they have many many distinguishing characteristics that other experts here could go over in great detail I just want to put this in some context that we look morphologically at the crania of Neanderthals and nearly contemporaneous modern human compared to our closest living relative the chimpanzee one can probably find many more similarities here than differences if you put it in context with this out-group the chimpanzee so Neanderthals show up classical Neanderthals that everyone would agree is in the and earth all maybe a few hundred thousand years ago and they disappear mysteriously from the fossil record about thirty thousand years ago they are known mainly from their bones and their stone tool technology from Europe in the Middle East but we know their range extends into Asia and I agree that it is really an open question how far to the east that they went perhaps one day we will find Neanderthals from China I would not at all be surprised by that sequencing their genomes has largely been a success story that follows on the heels of advances in high-throughput sequencing there are several companies now that will sell you machines that will sequence D a hundreds of thousands of times faster and cheaper than what we could do ten years ago one of these companies Illumina is kind of the leader in the field now they are based here in San Diego and are pretty much the the world leader in this DNA sequencing technology and it's what we largely used a sequence DNA that comes out of old bones so the first really genome scale dataset from Neanderthals came from these three bones that were dated to about 38,000 years ago that were excavated from this cave the Vindhya cave in Croatia in 1980 they were chosen from amongst hundreds of bones that sponte and his co-workers had screened through to look for the presence of some surviving DNA and the absence of contaminating modern human DNA which is a large problem with bones that have after all been touched by many people who go and dig them out of the ground and museum curators who handle them so these were found to be largely free of contaminating modern human DNA and have lots of DNA to can be sequence so a few years ago we accumulated a little over 1 billion base pairs from each of those 3 bones and a little bit of DNA from some other bones so that on average we had a little more than one full coverage of the genome of the Neanderthal one immediate surprise from this data set was that when one sequences all of the DNA that comes out of an old bone very little of it is actually from Neanderthal at all most of it identified by sequence similarity is likely from soil living microbes that have colonized the bone in the time that has been sitting in the ground and in fact most of those microbes we've never sequenced so we don't see any similarity in terms of DNA sequence so we can just assume these are some microbes that live long ago in the past and we focus really on the things that look like primate look like a Neanderthal so another thing that we have known for some time is that DNA as it's sitting around for tens of thousands of years it sustains chemical damage and this is the main chem damage that happens cytosine will spontaneously deaminate to uracil so it's chemically different and then the polymerase is the enzyme that we use to read the DNA will read this uracil as a thymidine so where this Neanderthal when he was alive walking around had a cytosine we will read this often as thymidine this process is happening all the time it happens in your cells it's happening right now but we have energy dependent repair mechanisms that will detect this cytosine deaminase inand fix it they're not perfect but they're good enough to keep us alive or lifetime but as soon as we die this process goes on unchecked and cytosine accumulates damage so that the longer the DNA has been sitting in the ground the more we will see this c-to-t difference we were able to see early on as we started accumulating data that this c to t damage pattern was concentrated on the ends of reads and learned a lot about the process of diagenesis how the DNA changes over time the spatial pattern wound up being very important to learn about and so once we had learned what we could learn from that we started doing data analysis and this is a figure of a broad scale view of the genetic difference between Neanderthals human and chimpanzee so we have a reference human genome we know there's DNA sequence variation amongst human but the reference sequence the one that we have known and in some form for about ten years now is what we all use to compare other humans that we might sequence to see where they are different from this reference this one instance of the human genome we have something similar for the chimpanzee so if we just align the Neanderthal the human the reference human and the reference chimpanzee we can ask for each of the three bones where do we see some difference where is the chimpanzee difference and the Neanderthal and the human are the same or where is the Neanderthal different and the human and the chimpanzee are the same or where's the human difference where the Neanderthal and the chimp are the same and one thing that kind of stands out and in each of these bar charts I'm just showing what the difference is so for example here the Neanderthal in the human have a G but the chimpanzee has an A so doing this kind of analysis you can see that these types of differences happen faster than these types these are transitions these are trans versions this is a description of molecular evolution these things happen faster than these things the same pattern is here over in the differences that are just specific to human but the neanderthal pattern is very very different we see this huge excess of G 2 a and c 2 t as we know to expect from DNA that has been damaged so in this first data set that we published a little over two years ago now it was the case that what we could learn about human evolution from Neanderthal DNA was nothing really about this lineage here nothing about Neanderthals themselves and that's because we know that most of the differences that are specific to Neanderthals are in fact errors they're errors in the DNA that have accumulated while the DNA was sitting in the ground or they're the machine error the background sequencing error that we have when we are sequencing DNA and having only one phone coverage of the genome it's rife with errors so we see something like a hundred and thirty thousand transversions on this lineage and only thirty thousand on the human lineage and we know that we should accumulate basically the same amount of differences yet there's this huge excess over on the Neanderthal side so at that point what we could do with the Neanderthal genome is really define this point here that is find the positions where the Neanderthal matches the chimpanzee but the human is different that is asked what changes have happened in the human genome since we diverged from Neanderthal but all of the questions about specific changes in the and earth all would have to wait better technology which skipping ahead a little bit has come online and we'll get to that in a minute but before I start talking about the contrast that we want to make between Neanderthals and humans and answering the questions we can answer with Neanderthal DNA I'm going to do a very brief primer on genetics and sexually reproducing diploid like us if you imagine yourself as the current generation which I used to do less and less I do but you're in the current generation here you have you know you get your genome two copies of the genome one copy you get from your father and one copy you get from your mother so of each of your chromosome pairs one of those pairs came directly from the sperm then your dad and one from the individual egg and your mom that would eventually yield you what you may not know or think about all the time is that that chromosome that your dad bequeathed to you or your mom bequeathed to you did not exist in your dad or your mom it was a special version of the two chromosomes that existed in your dad or your mom that were recombined together stitched together and put in that single sperm that would become you or that single egg that would become you and each sperm and each egg are different so my brothers and sisters are different so this recombination denoted by the little X here shows that this chromosome inherited from the father has the front tip here from the chromosome that he got from his mother and the back part from the chromosome that he got from his father so in you the current generation is the physically melded manifestation not of your parents but of your grandparents so this is the father's chromosome which is the father's mother stitched to the father's father in a single chromosome so this happens this process happens more or less in a random place along the chromosome in every generation so if one is to trace one's ancestry back in time there is in fact a different ancestry at every place in the genome every place in your genome traces back a different path through your ancestors so if you are to ask for example where did I say there's a gene here for a very wide nose like I have so this wide nose gene is here um maybe I got this from my father who got it if it's in this yellow region from his father who got it from his mother if the female is on bottom here well where in my genome is that gene from my father's mother it's not there I didn't get anything in my genome there from my father's mother my father's mother is missing in my ancestry for that gene maybe some other gene I did get something from my father's mother so this makes reconstructing ancestry very very difficult it also means that if you compare any two people who are alive today and ask where and when in the past did they have a common ancestor who was that individual who gave DNA to the two people who were comparing now there is always a person who was that common ancestor but as you move across the genome from place to place that person who had the DNA that would eventually be inherited in the two people you're comparing now that will be a different person who may have lived in a completely different part of the world hundreds of thousands of years different in time so these genealogies go back in time sometimes very shallow sometimes very very deep that's just the reality of the population genetics within humans or any sexually reproducing diploid species okay so with this in mind we can ask how long ago in the past on average do two humans have a common ancestor to random place in the genome and this turns out to be four humans something like 450,000 years ago in the past okay so it's 450,000 years ago plus or minus about 450,000 years ago which is funny but it's also true this is a Poisson process the mean and the variance are the same so it really could be much much more recent it is much more recent in some places and much much deeper in other places and we can say now with Neanderthal genome the average coalescent between a human and a Neanderthal was something over 800,000 years ago in the past okay but we know from various lines of evidence that the population split that would lead to humans on one hand and Neanderthals on the other hand from this ancestral population that the Chris stringer might call Homo heidelbergensis something like this this common ancestor population it wasn't a single individual it was a population that had its own genetic variation and something like 300,000 years ago this variation would be split off into one population that would become us and into another population that would become Neanderthals so what this means is these random genealogies that go back in time they often predate the population split that would lead to humans and Neanderthals so the variation that's alive today within us was already alive in that common ancestor population and was alive within the and Earth all's up until the time that they went extinct so one implication of this is that there are many places in the genome where you may have a common ancestor with a Neanderthal more recently than you have a common ancestor with another human in fact we think this is true for about eighty-five percent of the genome that Neanderthals fall within the variation and through about eighty-five percent of the genome okay so that means that we share variation with Neanderthals for a seemingly uninteresting reason that they fall within our gene trees we're that closely related to them that we might share I might have an a where a Neanderthal has an a because that mutation happened in a common ancestor and someone else doesn't have an a because their common ancestor was longer go in the past didn't have that mutation somewhere else in the genome maybe they are more closely related to a Neanderthal than me okay so this is the backdrop against which we must compare human diversity to understand how we are different from Neanderthals so this process being random sometimes I'm more closely related to a Neanderthal than you sometimes you more than me it is random and the the expectation is that if this completely random process is going on then over the entire genome I will be more closely related to enander tall than you exactly as often as you are more closely related to a Neanderthal than me as long as they are a clean out group to all humans we're all dis the equally distantly related to Neanderthals so to test this hypothesis we sequence the genome of five individuals from around the world - from Africa three from outside of Africa from the places shown there one can ask along the genome locally along the genome how different are the three Neanderthals and orange and yellow here or the five humans that we sequence to the reference human genome this one instance of the human genome and this difference and the units here are as a percentage of the way back to chimpanzee this difference has a distribution because of that random coalescent process sometimes two people are very different and you get a segment that's way out here sometimes they're very similar but on average the Neanderthals are more distant to the reference human than any human is to the reference human but that summary turns out hides a lot of very important and interesting details so in an analysis devised by Nick Patterson and David Reich this very simple comparison we can ask this question are any two people that we compare equally dissimilar to a Neanderthal so if we find a place all the places in fact genome-wide where the West African and the French guy have a genetic difference maybe this guy the West African has a T and the French guy has a G what does the Neanderthal have if Neanderthals are a clean out group it should be 50/50 sometimes matching the West African and sometimes matching the French guy the the observation was that when one compares two Africans the result is statistically indistinguishable from 50/50 furthermore if you compare two of the non-africans the result is also statistically indistinguishable from 50/50 but any comparison between an African and a non African showed excess allele matching of the non African to the Neanderthal and this then is a summary of this all of these guys are basically the same and increase in the end earth all allele matching and these guys are the same but any comparison across them shows excess here so from this we devised the the parsimonious model that when humans migrated Out of Africa maybe 70 or 80 thousand years ago and first came into the Neanderthal territory there was an episode of admixture and this group of individuals would then colonize the rest of Eurasia and bring one to four percent of this Neanderthal ancestry with them so that was interesting story and kind of how things stood for a few months until a bone from this cave was discovered with remarkably well-preserved DNA there was some technological improvements that removed this base damage at the end so that from the Denisova compared to the Vindhya and the and earth all's we get really really clean data and could do a lot more with it it looks very similar to the Vindi in terms of divergence do you make a tree the Denyce of us it's in a really weird place it's a sister group to the Neanderthals and out-group to humans but it's genetically more distinct from Neanderthals in any to human are today and there's this really really odd signal of increased allele matching from individuals in Papua New Guinea this has been investigated more and this signal seems to locate east of the Wallace Line its present in of Australians and people from this region and largely absent from mainland Asia technology keeps getting better this guy Matthias Meier had the imagination to think that DNA might actually be single-stranded from these bones and devised a method of pulling down single-stranded DNA not just double-stranded DNA and was able to increase the yield from this Deniz abou bone many many times over and get DNA genome from the Denisova of about 30 full-coverage this allows one to pretty much wring out all the errors every base you get to see on average 30 times and now instead of having an excess branch leading to the archaic group you have a shorter branch which is in fact what one would anticipate if this individual has been dead for long enough the evolution that continues through every generation these accumulation of mutations the clock stopped on this individual and he's missing we estimate somewhere between 74 and 80 thousand years of molecular evolution which presents a new way to date bones just from the missing DNA sequence mutation within it I'm running out of time but I just want to finish up with this one last thing as I mentioned you have two copies of the genome the one that mother and father gave and every place in the genome tells a different evolutionary story a different tmrc a this means that within the genome of a single individual one can answer lots of questions about the whole of human evolutionary history and in fact you can look at the local density of heterozygous sites places where your mom was different from your dad and infer how long ago their common ancestor was across the genome and make these plots in a very creative use of genetic data that say through the past how big was the population size that led to various individuals here including the Denisova and one outcome of this was that the Denice event sample seemed to indicate that not only did they eventually go extinct but their effective population size the amount of diversity that they were carrying had been low for a long long time so this population the population size of whatever group this individual belonged to was low and likely struggling for quite some time up until the point when they eventually disappeared so I will just end with some questions there are a lot of great answers from you know ancient DNA but they propose new questions and one of them that is really a hot open question is this one episode of admixture that we have here with the rather crude method asking how much allele matching is there does that hide details that might be interesting might there have been subsequent episode of admixture in a long time that humans coexisted with Neanderthals who were these Denice events we have a single pinkie bone and a molar we know very little about them morphologically we don't know their range we know that their DNA winds up in people who live very very far away from the cave where this bone was found and finally were there other archaic that were involved in ancient admixture events thank you
