 Yes, I'm going to continue on the theme of transatlantic slavery, but really what my talk is about is how we can use DNA to write history essentially, and of course the two talks after me will continue on that topic, talking using modern DNA but also using ancient DNA. I'm going to talk about it in the context of the transatlantic slavery. A lot of the stuff that I'm going to talk about is really not so much part of this project that we're here for, this citizen project, but was part of an earlier project called DuraTask, which was a Marie Curie training network, so a PhD training network that ran from 2011 to about 2015, and we had 10 partner institutions across Europe, and it produced 13 PhDs over the course of these four years, and all of them working on the theme of the transatlantic slavery, and it's contemporary legacy, but it was a very interdisciplinary project, so there were people, historians on board, archaeologists, and people working in population genetics. The website is still live, although it's not being updated very regularly on these days anymore, but if you're interested you can go and check it out. We worked around three major themes, and one of these themes is what I want to focus on for the rest of my talk today, which is origins and accession of ties, and I guess that's what brings you all together into this room today anyway. And opposed to the context of the slave trade, this is a topic that has interested historians for decades, and a lot of interesting work has been done. So for instance, what you have here, you might be familiar with images like it, this is from a database called the Transatlantic Slave Trade Database, which is really the culmination of several decades of documentary archival research, and this is available online, and it contains records of over 35,000 saving voyages across the Atlantic between about 1,500 and 1880, and the point here is though that it gives us a lot of information about the general volume of the trade, so how many people were enslaved, over 12 million during the course of the slave trade. It also gives us indications about their destinations, so where they were disembarked in the New World, in the Americas, so one thing that people often ignore is that to North America only about 500,000 slaves were imported into North America, thanks very much, whereas the great majority of slave Africans actually were converted to South America. Now when it comes to origins though, there are certain limitations with the kind of data that we have available, so out of these 35,000 records only about a third make any mention of origins at all, and when they mention origins they tend to mention postal shipping points as opposed to the people's actual ethnic or geographic origins, so this is why it's started here on the West African coast. This is a more precise image, giving you the same thing with numbers being transported over to the Americas. Now, this limitation of this type of data has been known to historians for a while, and here at the Dutch historian in already in the 70s simply concludes that this problem is too complex to be solved by the current resources and data that we have available. Another aspect of this is genealogical research, and here, especially in the context of African-American family history, there is the fact that people interested in their family history of African descent in the Americas, African-Americans also hit this wall of slavery as it's called. So the fact that before the 1870s census, there are essentially no records of enslaved peoples in the US in this case and in the Americas I guess more generally. In addition to this, the problem of trying to use historical records like those included in the San Francisco database to maybe make a connection back to Africa is an extremely complex problem. This is another aspect to this that you might be familiar with and in fact this has been mentioned already a number of times, but something that complicates the search for roots, origins in the context of African-American family history is the fact that after the abolition of slavery, slave people tended to take on the names of their owners, making it very difficult to trace the paper trails back further than that. And I just put this link in here because just to say that there are things happening also in this regard on the regard to the paper trail, but try to address these issues and make databases really available and if they can use in order to try and trace it in the Americas, particularly here in the case of African-Americans. So there is the, for instance, the Friedman Bureau project which includes over I think five million records of enslaved Africans who were reported after emancipation. And this type of data is really available and people can start to use it in order to trace their family history. But it will only go so far. So the other tool that has been publicized a lot through popular TV series like, you know, Finding Your Roots and so that Henry Dois Gates, you might have come across a person who has a show on PBS in the US, very much advertising DNA as a way for African-Americans in particular to try and go beyond the paper trail, go beyond the law of slavery and try to reconstruct their family history using DNA. And here, you know, just a couple of slides to illustrate the point that yes, it is very much true that especially in the last 10 years or so, we've, you know, undergone a revolution in terms of the way we generate genetic data. People talk about the sequencing revolution. We're able to sequence the entire human genome in about an hour or something like that. It is, and that has led to large scale studies trying to match human genetic diversity as well. This is just a slide showing you, you know, that with this invention of new forms of sequencing, the cost of sequencing has also been reduced and of course that allows us to sequence more samples from different populations and to use it for various things. Now, this is a very famous example of the famous plot by now that probably all of you have seen. It's by John Horgray and his colleagues who use this kind of data, genome-wide SNP data from, in this case, about a thousand individuals from across Europe. And what they showed in the, you know, was quite incredible relation at the time, but that you have this genetic structure in human populations and the simple fact here is that it's, that it tends to mirror geography. So you can, you know, have here the genetic map of Europe with different European populations clustering, forming clusters and they tend to mirror the geographic map of Europe. So there's a close correspondence between genetic and geographic distances. And of course we can use this in, in ancestry research or when we're interested in population histories in order to trace our own ancestry. So you can sequence your own DNA and you can simply, you know, try to plot yourself in that, in that, in that cloud of data and see where you fall. But we can use it on, on other levels and other levels in order to reconstruct population histories. And that's, that's what I want to talk to you about for the next half an hour or so. So one more slide though, very important, is that there is, there is, when it comes to the history of African populations or African American or African Ascendant populations in the Americas is somewhat of a, of a discrepancy in terms of our imbalance in terms of how much data has been generated. So this is a, this is a plot from, or a graph from last year in Asia where they looked at how much, how many genetic studies have been done on human populations. So this is, this was the status in 2009 and this is 2016. And you can see that back in 2009, almost in a cloud of 96% were done on people of European ancestry. And only 4% were done on people of non-European ancestry. And out of those I think maybe half a percent or so were done on people of African descent. In the case, you know, in 2016 it looks a little better. So about 80% still, genetic studies are being done on people of European ancestry, about 20% of people of non-European ancestry. Out of those, maybe 2% are, have been done on people of African ancestry. And that's despite the fact that Africa is the place where the ultimate law originates from, the place that shows the greatest genetic diversity. And for that reason, that's also perhaps arguably the most interesting place to look at. There are various reasons for that that I don't have to go into. But just to say that if we want to use genetics in order to reconstruct population histories for African Americans, for instance, to trace their own history, it is extremely important to have reference data available. And because of how the field of, you know, how human genetics has been progressing, this is not necessarily the case. And so that complicates things a little bit. Having said that, there are various different projects undergoing at the moment that try to rectify this situation. And here I've just, you know, pulled up a few. There is the H3 Africa consortium, which is a big consortium of various different institutions in Africa and also elsewhere that try to map human genetic diversity in Africa. Another one is the Sangra Institute in Cambridge, the African Genome Variation Project. And then there were smaller projects like our own where we tried to essentially fill gaps in this genetic map in order to try and have the data available that we needed in order to try and answer the questions that we had. So bearing that in mind, I now want to give you just a few insights into the kind of work we have been doing both starting first with a modern DNA perspective and then going on to talk about ancient DNA. So in terms of modern DNA, we had one project which was led by a PhD student, César Fortes-Nima. And he looked at a population, modern day, present day, living population from French Guiana and Suriname. There were descendants of maroons, so maroons, runaway slaves who were independent communities in the early 1700s or in the 1700s after having essentially escaped from slavery, Dutch slavery in that case in that part of South America. And so what César was interested in was can we use modern genetics to try to add to the history of this particular population. So concerning the origins, concerning perhaps admittable patterns and so on and so forth. And I'm going to just show you a couple of those results. The study will be published any day now. So here the first thing that we found was that this population of descendants of maroons called the Norma Roe, they showed a very high level of African ancestry, much higher than other African descent populations in the Americas. So comparing it to African Brazilians or African Colombians, for example, the Norma Roe had almost 100% of African ancestry, whereas African Colombians and African Brazilians had higher levels of European and Native American ancestry in their genomes. And that simply reflects the history of relative isolation of the Norma Roe and basically a survival strategy, staying far away from Europeans as possible, and that led to this particular pattern. So then what we wanted to do was to see if we could use the DNA that we, the sequence data that we generated from the SIP data actually, that we generated for these individuals to try and see whether we can look at their origins. And this is like the other plot that I showed you earlier, again the PCA plot, the principle component plot, but showing you in different colored triangles here, that's the genetic variation in Africa. So different populations, I'm sorry I don't have the legend here, but here you basically have populations from West Africa and populations, yeah, populations from West Africa, populations from then going around the Barabinine and the Barabiapra, populations from West Central Africa and then populations from Southern Africa. And although it's not particularly clear on this plot, the Norma Roe are these dots here in black, and they tended to cluster with populations from the Barabinine, the historical Barabinine, so that's the area of Western Nigeria essentially, and both those, so that would be present day Ghana. In contrast to this, the African Brazilians, you really can't see that very clearly here, but they're in gray, and they cluster with populations from Angola and West Central Africa. So while we're not able to pinpoint exactly a particular population that the source population, we can narrow it down considerably using the kind of data we have available. So the other thing that we wanted to do, and this now goes back to kind of like a little biology, and you probably all know this, what we wanted to do was to try and see if we could say anything else about the demographic history of these populations. And to try and get there, you have to understand the ways in which we inherit our DNA, and we all know this, you get a copy from your mom and a copy from your dad. If you take this and consider a case where you have a local population, and then a population of migrants that comes in, and you have some sort of admixture event, you can use essentially the different bits of or stretches of DNA from each, and you can use the length of these segments essentially in order to get some sort of estimate of when that particular admixture event happened. So essentially like the longer time away it is in the past, the shorter the segment. So to illustrate this, you can take the case of European civilization of the New World, you have a European migrant population coming to the Caribbean, Puerto Rico for instance, and you're to mix in there with locals, and you will end up with an admixture population like Puerto Rico. And what we can do is to not only estimate how much of European ancestry is present in that particular population, but we can try and estimate at what point in time in the past that ancestral components arrived in this case in the Caribbean. So this is what we did for this particular population, the normal. And even though the challenging bit here was that there was, as we said earlier, there was only very little European and Native American ancestry present actually in the genome. So there were 98% African. But even so, we got some results which suggest different pulses of first Native American ancestry and then later on a pulse of European ancestry into essentially African genomes. Interestingly, the dates or the estimates that we got correspond quite well with the formation of the first immigrant communities around in around the 17th and 40th, 1750s. So that was quite fascinating to see. So then lastly, another thing that we did was to look at essentially patterns of sex bias gene flow within these genomes. And this, you know, already discussed it in the case with the history of the European father and an African mother. Well, here you see this on a population level once again. So in every case of African descending populations in the Americas, you see a dominant African female gene flow and the male dominated European gene flow. You know, reflecting basically these patterns of extra. Interestingly, one result that popped up, not quite sure how to explain, but apparently in African Columbians, there is a stronger male Native American input as well, which is a bit more hard to explain. So those are the kind of things that, you know, you can read from just looking at modern DNA. What I want to do for the next few minutes is to give you an insight into some of the things that we have done and can do using ancient DNA. Of course, the limitation with modern DNA is that it's basically a talent test of what, you know, what we're dealing with now. But if you want to look further back in time and dance and have this talk after mine, we'll go further into this. You really need to understand samples from the way that they're passed. So here, I'll give you a few examples of this. So the first point to make that, you know, like other areas of research, ancient DNA has been impacted majorly by this event of this new sequence in technology. So this is, you know, I just put it into Google Scholar, you know, type in ancient DNA and you see that after this technology was introduced, you have, you know, hundreds if not thousands of studies being conducted just because we're A, able to generate the data and B, we're able to believe in the data that we actually generate. So in a few bullet points, the main impact for that is, or the reasons for that is the extremely high sequencing output that we have is much more efficient, is less time consuming, you don't have to spend as many hours in the lab trying to generate the data using toning, you know, PCR and stuff. And you use much less sample material. And then this important point, which is the ability to assess data authenticity and to estimate consummation rates. So the first study that I want to talk to you about is now already a couple of years old. But this was really, one of the first studies, the other one was done by Dan and the student of his actually, where we tried to use genetic data in order to trace the origins of enslaved Africans. And so historical, using historical studies. So this particular study was done on individuals that came from the island of St. Martin. That was just flattened by a hurricane. In St. Martin, it's better known for, I think that's pretty important, where you have planes coming in above your head. But in about 2010, Jay Haviser, an archaeologist who works on the island, during construction work, discovered three burials on the island. And two of them were manals and one was a female, aged between 25 and 30 years old. And judging by the artifacts, our associations like puff shirts and such that were in the burials, he estimated about a mid-17th century date for those, for those new burials. Now interestingly, all three of them had this clear sign of dental modification, which was, clear sign that they were probably from Africa, given that that was a very common practice in Africa at the time. And then also was an indication that they were probably born in Africa as opposed to in the new world. And what we wanted to do was to try and see if we could use the genetic techniques that I've been describing in order to try and trace their origins in Africa. Before we did that, we tried to also look at historical records. And so this is, when you look at the transatlantic slave trade database that I showed you earlier, you actually find one record of one particular slave investor that arrived in St. Martin in 1665. So it fits with the date that we estimated for those, for those three burials. You can even get the name of the vessel, the captain's name. It tells you where the voyage started, and it tells you also where slaves were purchased, but brought over to St. Martin. The first place was in Goree, in Senegal, and the second place was, you know, Mina on the, on the Gold Coast. What it doesn't tell you though is, is where these individuals came from. And so this is where the DNA comes in. This just gives you a summary of the kind of data we were able to generate, it wasn't much. So no coverage genome, so that means we randomly sample bits of their genomes at, at, at, at lower coverage, as we call it. We estimated the continuation rate. So, you know, trying to establish that the data that we generated was actually authentic, real deal. We've got mitochondrial genomes out of them. So we've got their mitochondrial haplogroups that you will be familiar with, and also for one of them, a white chromosome, white chromosome haplogroup. Now, and this is, this is, you know, the main result really, and again, having a background of genetic variation for West Africa, sorry, for, for, for Africa in general, going from Southern Africa here, in this case, to Central Africa, so Angola and Cameroon in these regions, West Africa, and then this, this further into the area of challenge in these kind of regions. And so what was interesting for, for these three individuals, that they essentially, you know, pop up with different, with different populations. So whereas the first one was a greater affinity with populations from Cameroon, particularly the Balmoun, the two other individuals, you know, further closer to, or probably with populations from the, from Ghana and Nigeria. So that, you know, again, like with the result for the normal, we're not able to pinpoint exactly where these individuals come from, but we're able to distinguish between different regions in Africa, given the kind of data that they have available. For the one individual that we got this white chromosome haplogroup for, this was actually quite, quite, quite interesting because it showed this particular haplogroup has a very restricted distribution in South Saharan Africa. And it's almost at, here in this part of Northern Cameroon, it's, it's present a very, very high frequency. So it actually fits with genome white data results that we got for this particular individual. So at least for this one individual, you know, we, we had to better, you know, we could fairly certain that this individual came perhaps when it comes to some of that, some of that particular area. Now, the second state study that I want to show you is work that I'm doing together with an archaeologist working in the state called Julie Javitski. She works on 18th century plantation sites in Maryland. And this is a, this is a community led project. And it is interesting because they chose to focus not on the grand plantation house or other aspects of the plantation, but they focused on the state forces. And one of the most common finds on these plantations are, are, are these type of artifacts. Does anybody know what this is? Probably. Any idea? This is about, this is about an inch. So this is a, you might not be able to get it just from that picture. It is part of a pipe, that's right. Galen pipe, plate pipe, that's right. It is a plate pipe, at least part of it. And you'll be able to recognize it much better if I show you the other part of the pipe. That, that will be the, you know, top of the pipe. And so what Julie wanted to know is if we could extract DNA from the pipe stems and say anything about the person who spoke the pipe. And so we did. And it turns out we can't. So we extracted DNA from this pipe stem, human DNA. And the first thing that we did, we weren't able to generate a whole lot of data. It's coming from the pipe stem, but, and the first thing that, that we found was that the human DNA that we were able to generate was likely from a female. So we're dealing with a female pipe. Now, of course you could say, well, that could be Julie's DNA, you know, after all, like excavated those remains. But I'm not showing you this, but we have ways of authenticating our DNA by looking at characteristic damage patterns, for example. And I can tell you that we've had these characteristic damage patterns. Also, we were able to get enough DNA out of it in order to trace the ancestry of the DNA that was in the pipe stem. And so in here, in this case, you can see, although I have to tell you it was not a lot of data. So here you have, again, a background of African populations. And in this case, the pipe stem, DNA plotted close to this population cluster, which happened to be demanded from Sierra Leone. So here, you know, just to show you that one can use this type of data from artifacts. And there are reasons why, you know, we might give, I mean, there are reasons why it's not too crazy to think that you might get DNA from a pipe stem and it's made of silica. There are reasons why that's not impossible. And here, you know, we've shown that it might well be possible. And that, you know, tells you something about the individual background and so on and so forth. Right, so the last one in just a few minutes, this is a project that was also done as part of this Eurotest project by the student, Marcella, that's in the Royal Velasco. She worked on a little bit of a different context. So we're not talking here about the Americas, we're not talking about the Caribbean. So context that you want, you know, traditionally associated with Foundation slavery and so on. But it's working in a context with a small island in the South Atlantic called St. Helena and you will maybe have heard about it recently in the news because they got an airport, which is a term most used as an airport in the world. But during construction work, in that, oh, it was also Napoleon II Exile, so we will go back probably, what the island is not known for is that during the period, in the later period of the Transylvanian slave trade, over 30,000 in South Africans or liberated Africans at that point were brought to the island. So this is at a point after the abolition of the slave trade in 1907. And the British Navy was essentially rolling off the coast of West Africa and intercepting slavers that were still engaged in the trade, liberating the individuals on board. But instead of repatriating them to where they came from in Africa, a lot of them were brought to St. Helena. Now, this is a tiny island. And some of those were then repatriated to other parts of Africa, like Sierra Leone, Liberia, but others also ended up in Jamaica, British Dya, and Trinidad. A large number actually died on the island because of poor sanitary conditions and so on and so forth. And during the work that was done in preparation for the airport, British archaeologists excavated over 300 burials at the site where now the airport is essentially. And those 300 are probably only a small fraction of the burials that are actually there, the estimates of several thousand burials being there. So what Marcella did, she was able to generate about 20 mitochondrial genomes, or about 20 individuals, and also some genome data that I'm going to show you in a minute. So looking at the mitochondrial data, we can look at essentially the frequency distribution of the mitochondrial capital groups that we have there, and we can compare it to a database of the same capital group frequency distributions from West Africa. And although this is, you know, this is not a particularly good way of doing it, but you can see that if you just compare these frequency distributions, you can see that there might well be a greater affinity to West Central Africa than to Southern Africa, simply based on the distribution of those capital groups. Much more powerful way of doing this though is again to look at genome white data rather than just unique parental markers. So what we've got here then is again a principle component plot where we, you know, try to plot these 20 individuals from St Helena on, you know, against the background of Afro-African genetic variations. And here, you know, the result is that they do cluster all with populations from West Central Africa. Population from Angola and Cameroon here in the North. Indicating, you know, suggesting that at least this sample of 20 individuals are more likely to have come from that part of Africa as opposed to West Africa or Southern Africa. So that, you know, with that I just conclude my talk just to sum up to say that DNA can can be used to trace ancestral origins where structural and genealogy information is missing. But I think it's important to bear in mind that at least as the reference data is at large at the moment, we're not able to precise the pinpoints of populations but only to narrow it down to regions perhaps. And again, it's important to remember that the accuracy of these results depends on very much the available reference data. So I'd just like to acknowledge and that this research was funded through the European Union and the American Reactions. And I'd like to thank all the members of your test network. Thanks for your attention. Thank you. Thanks very much, Alice. That's part of this presentation. Where are we going to go in the future? So if we kind of project ourselves five years into the future, where do you think we'll be in terms of either the amount of DNA that we're going to be collecting from African populations but also what does that mean in terms of actually pinpointing where people might come from in Africa? Yeah. You saw the projects that I put up and although the efforts in Africa are not comparable to what is happening in Asia or what we have available for Europe and you can, you know, it's really fast and show me some plots earlier for Europe and for the British Isles and it is fascinating to see the kind of resolution that we can get. But, you know, given the work that is being done, you know, at some point we'll get there as well. Now, the other thing to say in that context though, it's the so in order to do this type of work, yes, you need reference data but you also need for variation to be present. So if you've had historical processes, migrations and so on that led to the homogenization of populations so in the context of Africa and people always talk about the Bantu migrations for instance that have, you know, kind of muddled up the signal if you want, then, you know, the idea of trying to have enough resolution and pinpoint origins will always be different. I think the other big thing to remember of course is that there's 200 times more genetic diversity in Africa than there is in the rest of the world so in practical terms if an African person wants to get as many DNA matches in their list of results for many of the commercial databases we'd almost have to test 200 times more Africans than we currently have tested Europeans. Well, that's a calculation that I can't follow that quickly but I think that it would maybe the fact that there is more genetic variation in Africa would actually make the exercise easier because you have more variation present. Would you get as many matches? Quite a difference. I believe that if you're able to generate data and you have more genetic variation you'll be able to acquire as many questions perhaps. We have Debbie here in the front. Do you think there's only shots having much more ancient DNA data? There's only a handful of studies so far that have produced any DNA. It was in European rock there were probably more in South Africa but there's so much there must be so much knowledge in all the DNA samples from Africa, if one would have some ancient African DNA. Thank you Debbie. That's a very, very interesting question and of course as I said earlier it's the period of time and our common origin and now the history of anatomical humans going back 300,000 years it's a fascinating place to look at and to work on and I'm very much hoping that it will be available. There are things that I think last week or a couple weeks ago there was a paper by colleagues from Upsilap working on ancient DNA from South Africa. The challenge of course is preservation so a lot of areas are tropical or subtropical that leads to a rapid decay of ancient DNA but having said that we were also able to recover data that goes back thousands of years so it's not impossible that I'm sure that there will be studies coming up. Great, so we have a question here from Peter. As the Swedish has had to ask far less people in Denmark are taking genetic genealogy tests than in any way in Scandinavia. Do you have a theory why and how do you get your veins to test more? It's like a white spot on the map. That's a very very personal question there. Interesting. I didn't know that but maybe the genes are just comfortable with who they are. I don't know. It's an interesting question though but I wouldn't even want to venture this in my life. Other questions from Alice? Yeah, we have a question over here. Can I ask you to come over here because that loudspeaker is so close I'm afraid we're going to get some interference. I read a book about slaves who were taken from Baltimore down the park. They were sold piracy they were sold somewhere in the world. Is there any data on that? There was a pirate ship carrying it. It just took the whole village. This was the Barbary Pirates that actually came into Baltimore which is a little village in the Barbary. It was a corp. Martin D. That was a sin. But they ended up in North Africa rather than in the Caribbean where they ended up in North Africa. They were sold into the slave trade on the North African coast and two of them survived and made it home and kind of told the story. Is that anything that you've come across on us or is that part of your research about slavery? I'm familiar with that but the practice of slavery wasn't limited to the Transylvanian slave trade. It's an ancient practice that has been going on for millennia really. What stood out with the Transylvanian slave trade is the scale. We're talking about over 12 million people being enslaved and transported against the will to another part of the world. But to answer your question I'm not aware of any of this type of work being done and I also don't there are books that have been written about it but in terms of historical research that takes on that scale I'm not aware of that. In terms of your task as a project that ran from when was it? It was a five year project but is it still it's not active at the moment but what plans might there be for a re-activation or a re-visitation in the future? That depends on the funding and so much else does but we're happy that the 13 PhDs all finished their PhDs one of them is actually in the room here and they also carry on working on some of those topics so that work continues and yeah I mean there might well be further down the line in another grant application and a new project Any final questions for Hannes? Well it just remains for me to say thank you very much for a fascinating talk on a fascinating topic and thank you for explaining it so well Ladies and gentlemen, Hannes Schroeder