EAA 2023 keynote: Tales from a small island: archaeological genomics on the Atlantic edge

[Music] foreign over the past 10 years ancient DNA and the science around it has revolutionized our understanding of different aspects of the past throughout the world I'm lying at the very westernmost point of Europe Ireland has and is yielding many fascinating insights about the movement of humans and and animals and at the very Forefront of This research have been Dan Bradley and Lara Cassidy both from the Smurfette Institute of genetics and Trinity College Dublin and I would like to now welcome Dan and Lara to the stage for their lecture entitled Tales From a small Ireland ancient genomics on the ancient Edge good evening everyone um yes that is a hard act to follow this is a difficult slot between the performance and the drink but I'll try and keep your attention anyway so um first of all thank you Eileen and the organizers it's a real honor to be asked to to who give this lecture and uh and it's fantastic to be here it's genetics and Archeology are such different disciplines and it's difficult to get the interplay between the two so it's great to be here and get the chance to to listen over the week to so much archeology so um what we're going to talk about are in in the in the spirit of weaving an art of our three Tales um this comes from work that we've done in the department of genetics in Trinity College over the last decade or or a little bit more uh in in the labs it's co-led by myself and Lara and uh what I'll do is I'll I'll tell you something of an introduction and something about migration and Lara will then move into uh the cousin's tale which is about kinship biological kinship and the survivor's tale because genetics of course is is very important for health and um has a central to Evolution and so so the first thing is to say whenever we talk about migration migration it really was a culture shock for me um some 10 years ago to find that migration was actually quite a valued Laden term um in in some parts of archeology that is not genetics it's just something you measure um but anyway how do we detect it how do we measure it and what is it about the genetics of ancient Europeans and what methods do we use that allow us to do that so there are three aspects really that give us the power with ancient DNA um to to measure an unfair migration and the first is the scaling up of data acquisition which is something that has happened uh in recent years it took the first human genome uh something like 300 million dollars in a decade to be sequenced now your Human Genome because of medical technology can be sequenced in an afternoon and costs less than a thousand dollars I'm not even sure how much it costs now because it costs keep on decreasing all the time that's a modern Human Genome it would cost more to sequence ancient human genomes secondly you have to type methods that's um ways of comp Computing and and fairing relationships from ancient data and modern data and thirdly the the material we're working with the Divergence that is there in prehistoric populations so well what is genetic data and it has it has quite quite a pedigree quite an ancient pedigree in in terms of Academia that on the on the left here is a slide from one of my predecessors in the department of genetics in Trinity and uh it's a a slide of Reese's negative frequency that's one genetic lockers and as you can see uh there's a gradient in measured in 1958 and drawn on this old glass slide across Ireland and that's because of History you know genes have a past and the patterns we see today are because of that past um on the right is uh the gradient of another Gene the old blood group um you know a single segregating variant and it has a geography across uh these two islands um and it's a recent geography because I've just put in a circle there and that's a sample that we sequent some years back of individuals of Britain's from New York um from the Roman era so it's a geography a genetic geography that's changed and these are interesting uh you know they tell us something but they're they're a bit Limited in that they're based on one locks and one Locus is not a lot of data so what do we work with now well in ancient and modern genomes we work with hundreds of thousands of genetic loci so we have gone up a factor of 10 to the five in sort of data we work with so it's been a huge change and what can you do with this sort of data well um what what you can do is you can plot individuals in genetic space so this is a map of European individuals all quoted by the two letter chords he asked for Espana PT for Portugal and as you can see uh people with Spanish and Portuguese ancestry clustered together down in here in this two-dimensional genetic space in a in in a sort of genetic Iberia um in a peninsula you can see another peninsula here um oops this is I.T Italians and here are some unusual Italians turns out they're Island Italians they're sardinians I'll page at the top we've got GB breaking away from Europe as you can see not quite managing it uh uh Ghana and I.E not quite breaking away from Britain either as you can see and why would it um but as you can see there's an East-West norocite genetic geography uh which which we can we can we can uh we can plot an unfair and um it has implications for uh for different things including um forensics but uh the second thing I was going to mention is haplotype matching so what this is this is a uh for illustration it's a it's a detection of plagiarism from a student and how do you detect plagiarism you don't do it by single words you do it by strings of words in sentences and sentences have genealogies and if you submit a paper and the genealogy of your sentence is from someone else it's detectable because you can detect in strings and genes are the same they are current chromosomes and you can unfair genealogical relationships very securely using this method which is identity by descent um uh based on strings of matches so genealogical relationships strings of variance in DNA is a very secure way of of of detecting relationship and then thirdly uh whenever we now know from ancient DNA and some of this is done by us but of course much done by other excellent research groups in the field whenever humans emerge from the ice in in Europe they emerge in very Divergent populations and why is that it's because these populations overwintered in different places they follow different trajectories through time they were separated and they're different and uh they they mix together in modern Europeans I've just um I mean Emily Breslin one of our group is giving a talk on on some of the or upper Paleolithic work uh on Saturday this one's an interesting one these are early European farmers with the Advent of the Neolithic they spread through Europe um and have a profound effect in European populations and later these other two to the right they um they mix in in the third millennium BC again altering European genetics so uh the the question I'm going to ask and try and answer in my little bit of the talk is if one looks at Ireland we're just going to look at Ireland was the population of Ireland uh was its history Through Time a smooth one a rugged one we had change everywhere has changed but where the changes gradual or in a temporal sense episodic where they're Cliff edges that's the question whether Cliff edges in the population history population prehistory of the island of Ireland so I'm gonna use three samples for illustration um the first is a Mesolithic genome sample the third the second is from Ballina Hadi quite close to here um a neolith example and the third is from Rafael Island an early Bronze Age sample and what we're going to do is plot them on our map of Europe so where do the Mesolithic hunter-gatherer samples not just the Irish sample but Western hunting others where do they fall they fall above they fall northwards in the plot of Europe um up here at the top in a sort of genetic Valhalla it's a it's a genetic place that doesn't exist anymore um because their descendants to exist but their descendants have don't exist in the form they wear they've mixed through into the population uh cool that is Europe today um secondly what happens uh whenever you look at a Neolithic sample and this is just to point out that uh the first genomes from the British Isles were sampled and uh sequenced by ourselves in a collaboration which was a trinity uh Queen's University Belfast collaboration so we're very proud of that actually and what's the story they told this is about anahati um she's from a Neolithic tomb a few miles away from here very few miles away and she's famous she's not a facial reconstruction and um she falls down here close to Sardinia and in fact this is typical of Neolithic samples from from Western Europe this is where they fall and this distance between the earlier Mesolithic genome and the later um genome it's too far for this to have been just from indigenous drift this indicates a severe change in composition which comes from migration and WhatsApp migration um was it was that typical that's only one genome well in fact whenever we look at a whole range of genomes 42 Neolithic genomes from um from Ireland sampled from different terms you find that yes it is typical um with one exception or one and one individual from um uh Carlton Jones's excavations in park nabinia and County Clare and this is an interesting individual because uh using the genealogical detection technique like plagiarism detection technique um we were able to unfair that one individual out of the 42 had a Mesolithic great great grandparent um and that great muzolithic great great grandparent was an Irish Mesolithic because it was more similar to the Irish sample than to other Mesolithic samples now that's interesting because first of all it shows that there wasn't a full population replacement at the Neolithic so there was some some inheritance genetically came through from the Mesolithic but the other side of it shows it shows we have the power to detect great Greg grandparent um ancestors and only one out of 42 showed that so therefore there are 672 ancestral genomes that were surveyed and only one of those well was an iris Mesolithic so it was a fairly overwhelming genetic process this transition to farming secondly um now all our earliest genomes were from Northern Ireland it's not because I'm from Northern Ireland is because we have this collaboration from Queens um but uh we were fortunate enough to get the chance to sample some early Bronze Age samples from Russian Island this is a kiss burial they're from and um where do they fall well this is an example it falls up here not so far actually from where uh populations from the western parts of the British Isles Falls today um so and again this is a radical change it's a radical change now that's only one genome clotted here is that a radical change that occurred uh instantaneously or sharply not instantaneously but sharply or is it is it part of a Continuum you know it's only one data point so what we need are more data points from the Bronze Age and Beyond to answer this question was the transition into Metallurgy um roughly 2400 2500 BC on the island was it one that was linked to genome change or not was it a cliff Edge so if you plot the percentage of Anatolian ancestry roughly for um individuals through time so this is just a plot of a percentage of the Anatolian ancestry through time which starts out um low zero because these are just two Mesolithic genomes then it rises sharply uh just after 4000 BC and then you have a plateau of Neolithic ancestry early farmer ancestry and a truly ancestry whereas the majority component and then there is yes there is a cliff in ancestry which occurs around about two thousand five hundred two thousand four hundred BC is pretty sharp now there is complexity because if you color those samples um these are colored according to age apart from the ones from 2400 on where some of these burials are from reuse of megaliths in the copper age stroke Bronze Age um and then some are not and you find there's a difference there's less of a change in those from the megalithic context than there is a nailers uh so it is with complexity but it's still profound massive and I would argue in in archaeological time pretty sharp as a cliff Edge lastly uh just a a shout out for Victoria who's giving a talk again on Saturday and she's looking at the same period with cattle and cattle cattle don't show a clefage there's a spoiler but there's more to our talk than that so hello okay so I am going to be moving us on to the second of the three tales and this is going to be the cousin's Tale and in this um I I just want to consider a little about orange and DNA can tell us about kinship and obviously there's a lot of things that cannot uh tell us but one thing DNA does provide is direct information on on shared descent uh which Dan touched on earlier it allows us to construct true genealogies and map genetic relatedness between members of a community and those patterns they are going to be inevitably shaped by the kinship systems and the marriage customs of those communities and um I won't say I am not saying here uh that genetic relatedness alone determines kinship uh it's it's almost actually the reverse we're more leveraging the fact that culture and a kinship system can't determine genetics in the archaeological record and that's how we can learn um about kinship so how do we detect shared descent in the genetic record Dan touched on this but uh people who share a recent ancestor are going to share long chunks of identical DNA sequence inherited from that ancestors who say we've got two cousins here you can see they share this orange chunk inheritance from their shared grandparent and just for example some of you might have done a 23andMe test and if you did you might have gotten a result that looks something like this telling you that they've discovered your second cousin who are you are now morally obliged to invite to your wedding yeah because that's how kinship works and this is all down to you sharing these IBD segments so we can detect from the chromosomes you can see them here but what's really exciting is that we're getting to the stage in ancient genomics where we can actually match this level of resolution if we generate good enough quality data we can apply the exact same type of analyzes uh to ancients as we do with moderns and as a matter of fact um I've kind of cheated the result I'm showing you here it's not from 23andMe it's actually from our lab and this is not to Modern individuals this is a result for two adult males from Neolithic Ireland and we found that these men they're approximately fifth degree relatives so they too could be second cousins but their median radio carbon dates are about 200 years apart so that's something 23andMe doesn't really have to worry about at least yet so we could be uh sampling something closer here to a direct ancestor descendant relationship but there's something actually quite unusual uh about these two relatives and that is that they weren't found um in the same location uh not even close they were actually and teared over 100 kilometers apart uh and two of them was a spectacular Monumental landscapes in Ireland uh some would say in Europe uh the first came from the mega passage to Newgrange which I think some of you have already visited today or yesterday um the other site um some of you might not know as well but I think deserves just as much love this is karaokeel in County Sligo and this is another really incredible passage complex where the monuments are perched on top of various hilltops so then our question here is how to relatives come to their final resting places uh first in such auspicious locations but also so far away from one another and the most parsimonious explanation we have here is that we are sampling members of an elite a long distance kinship Network between Monumental tombs is something we'd expect for quite a complex uh Society with Elite Mobility and we have two additional pieces of evidence for that unleash the first is that we found that other pairs of Passage team samples from other sites they also show uh some accesses of genome sharing causing them to clusterogenetically which you can see in pink here the second is that the individual from Newgrange uh he was not only exceptional in his sampling location so his remains were uh sampled from the right hand recess which is a focal point in Irish passage tombs and then obviously Newgrange itself is a focal point within the wider ritual landscape of Bruno buonia but this uh man's genome it was also exceptional in that he was the son of first degree relatives which is an incredibly rare observation uh in both modern and ancient genomic data sets and that makes sense because we know that the first degree incest taboo uh that's one of the few cultural universals probably for underlying uh biological reasons and we only have a handful of uh confirmed historical cases where this Behavior was sanctioned by society and all of those are sibling marriages uh among uh dynastic leads uh we also see that the practice strongly Associates monumentality and a degree of political deification so this is interesting uh that potentially observing such a rare marriage custom could tell us something about politics or even the religion of a past Society if uh uh ethnographic analogy can be considered a useful tool in archeology which I think it can and I actually think it's one of the most exciting things about genetic data at the moment that it's opening up a whole new Avenue for cross-cultural uh comparisons uh for understanding how human societies vary through space and time and even maybe why they vary okay so we're now going to jump forward in time to the Bronze Age and here I want to talk about a related application of IBD analysis and this is the potential to produce detailed maps of connectivity between archaeological sites so this isn't about the flow of trade necessarily this is about substantial movement of people and Marriage Partners so we're talking about people for once not pots we know this is uh influenced primarily by geography right it's it's way easier to marry somebody down the road from you but culture is obviously going to play a very important role as well political boundaries ethnic differences uh economy and Technology can all uh influence uh and shape choice of marriage partner and as an example of this I want to look at the Bronze Age population from Rutland Island that Dan mentioned earlier uh since 2016 we sampled a lot more genomes from bronze age uh Ireland and other ancient DNA Labs have been very uh very productive as well so we now have something like 132 genomes from across Britain and Ireland that are amenable to IBD analysis and now what we can do is put the Wrath and population into much better context the question being which sites showed the most shared ancestry with the wratham population and we can visualize this with a nice interpolated map here where more red is more IBD shared with the wratham population and what we're seeing is that rocklin's IBD neighborhood uh it's quite a maritime one right its top hits are along the North Coast of Ireland and also uh the west coast of Scotland so this is a really interesting pattern because it suggests that these populations may have had a relatively strong seafaring culture uh perhaps not unlike the one we see Millennia later in the medieval period when we know that there were close political ties between these two seaboards so in some societies depending on culture depending on technology the scene might not act as a barrier uh Geographic barrier to move them but could even uh potentially be a unifier in some cases I just uh to contrast here is another Bronze Age population we've sampled this time from outside of Dublin uh a keynote and here what you're seeing is that the highest hits are further south actually in and around the modern day province of Leinster uh polar Rockland population in contrast is showing very little sharing with these sites on the eastern state board so what we're basically catching here is different spheres of mobility and potentially even different spheres of influence and at that resolution that's only going to sharpen uh the more we sample and so this is one of the really I think exciting kind of Horizons in ancient DNA right so this is the third and the final part of the cousins tale we like doing things in threes and uh this is going to bring us to the early medieval period so moving forward in time and what I want to do now is look at kinship within a single size and this size it's an enclosed rural settlement and a secular Cemetery at ranola uh in County Roscommon and this is part of a bigger multi-disciplinary project with tii and again with Queens and this type of size this is very common and a very well studied uh site type in early medieval Ireland it's been argued that most probably represent familial burial grounds although it's possible that some serve larger communities and this is something that we can now start directly testing that with ancient DNA and ranola is also uh interesting in this regard because it grew into something of a regional uh Hub uh through time and that could have potentially changed how the cemetery was being used so the first thing we can ask then is okay are we finding then a lot of genetic relatives at granola and how does that relate to the different phases of the site's usage so here you are seeing the number of relatives identified for each individual plotted against their median date and what you see is that during the main phases of occupation here uh nearly everyone is related you can see geological links as lines and and they connect nearly all individuals over the course of about three centuries so that's really nice that's confirming what has long been hypothesized about these types of sites and suggesting that even in round is Heyday it seems to be acting primarily as the burial ground for one family another thing we also see is that a single Y chromosome lineage dominates at the size which you can see in pink here and that fits uh very well with what we know from the historical record about Gaelic land ownership and how that was organized around patrilineal descent groups called durvania so that's something we're actually quite lucky with in Ireland that we have this very old and very rich vernacular history um including Gaelic inheritance laws and we're going to be able to compare them to to the genetic record going forward and see how they relate but also see if they always agree with each other uh and uh differences between sites and then finally the last point I want to want to make here um is that we do have also individuals with no relatives yet detected at the site so you could ask the question then if this is a familial Cemetery who are they and importantly uh they still could be family members they could be spouses who married in with no descendants they could be fostered children one thing we also really need to remember is that we have a limit of resolution here at about the degree of third cousins so we're not going to be able to detect detect uh individuals from from more distant branches of a clan so yeah what I'm saying uh again is DNA can't alone give us a full picture of kinship at any site and we're always going to have to consider other forms of data and one of the most important is of course going to be the burial context I'm just touching on this but one example we could look at is spatial distribution so here at ranola I've highlighted burials in light blue who don't have any detectable genealogical ties to the main family and you can see that they tend to fall at the edges of the main area of burial activity this is the pattern so far so that could imply a different status either within the family or maybe these are not family members they could be dependents of family and kind of going maybe in that thing we can then ask another question is okay well are these individuals still then from the local area around ranola and we can do that using Isotopes or DNA or if you're very lucky both uh both is usually best and just one example around Allah we have actually found uh two of these unrelated individuals are also outliers with respect to their Geographic Origins so this is a woman an infant who were buried together but they are also not related to each other and their genetic profiles you see here they are not typical uh from medieval populations from the west of Ireland which you see down here instead we're seeing a substantial amount of uh British type ancestry so I could go into a lot more detail here I'm not going to but if you do want to hear more about sort of these blind skill approaches to identifying non-locals please do check out uh um one of our groups talk Maeve uh on Saturday where she looks into this more final then uh of the three Tales uh this is going to be the survivors Tale so genomes are genomes they don't just hold a record of our ancestry they also tell stories of life and death uh we know malnourishment infectious disease and high rates of child mortality were all the norm uh for populations like the one living at ranola and many in the past but every one of you here today is the descendant of survivors uh people who lived long enough to have children of Their Own and we would expect any traits that increased those odds of survival these would increase in frequency uh over time in the population that is natural selection and it's incredibly powerful evolutionary Force and I I just want to touch on one of the most textbook Classics classic examples of natural selection in humans such as lactase persistence this ability to drink milk into adulthood and Ireland actually has the highest frequency of this trait in the world but we want to then ask the question well when did this trade actually undergo selection when did it rise in frequency in the population and this is actually a question we've only been able to definitively answer uh relatively recently with ancient genomes so we now know that the associated mutation was uh rare in the early Bronze Age but it has risen rapidly and relatively steadily through time maybe Pluto in a little in recent centuries so that uh that's really nice that answers the question of when but uh the more difficult and more interesting question is also why why was milk such a matter of life and death in Ireland in the past and um I should mention recently Nick Patterson and colleagues um looking at a similar data set in Britain they suggested this might be to do with differences in Dairy economies between the islands of Britain and Ireland and those on the continent during the Iron Age and that that could indeed be one factor but for the pressure to remain constant for so many Millennia in Ireland that suggests some more fixed underlying environmental pressure and like so many issues in Ireland the problem could possibly be the weather so other genes we found with signatures of selection in Ireland these paint a picture of of a sun starved population and we need uh UV light from the sun to synthesized vitamin d and vitamin D is essential for efficient calcium absorption healthy bones and a healthy immune system and we saw this even during the covid-19 pandemic where vitman D deficiency in Calcium deficiency were strongly correlated with the disease severity and poorer prognosis and actually the same is true for tuberculosis so lactose increases calcium absorption in the gut which could have given an ancient immune systems at the Boost they basically needed when they were constantly being battered by infections and it should also be noted that ancient DNA allows us to actually study the evolution of those infectious agents themselves and don't worry I'm not going to talk about that but if you are interested in that topic do again I'm going to do another plug check out um Issa Jackson's talk uh on on Saturday on Dental uh patho violence in the oral microbiome okay and then to finish I've been talking a lot about how long-term hardships that were endured by our ancestors that these can shape genomes over many Millennia but ancient DNA also provides snapshots of population and individual health and at particular points in time and there's an old kind of cliche adage that mended bones are the earliest manifestations of human compassion in the archaeological record but DNA also lets us move beyond the bone and understand the experience of individuals with genetic diseases and disorders that that may be quite invisible in the archaeological record and as an example of this I'm just going to bring a full circle back to the Irish Neolithic to pool nebron portal tomb uh where we sequence the Genome of a male infant with three copies of uh chromosome 21 which results in Down syndrome and we also found that this infant uh he had a isotopic signature consistent uh with being breastfed so that could suggest that this infant then uh was cared for in his short life and his internment also suggests that his death was of some significance uh to his community so as you've seen uh during this talk uh an ancient genome can tell stories that span centuries and Millennia even but at the most base level um a genome belonged to an individual and it can also help humanize that person and tell us something of their own uh life story and uh I'm Gonna Leave You there I just quickly like to acknowledge all the members of our research teams at Trinity um our amazing wonderful collaborators too many to mention or funders the organizers of the EAA for giving us the honor to talk to you all today and uh you all uh for listening I hope everyone has a fantastic year uh Gary Miller Mark of thank you Laura and thank you Dan and and um it's like the de Cameron of genomics with all those tales and um I'm pleased that Lara's connected it with with sessions over the next few days as well so but you know who they are so you can you can go pester them over a pint of milk outside now and um but we're very fortunate to have the quality of academic Excellence shown by Laura and Dan and did by Eileen sitting next to them and others in this audience here in Ireland you are changing the narrative we leave of Ireland's past in in in such a transformative way it it's quite remarkable and it's very very exciting so thank you Laura and Dan we're almost before the milk at the end of the opening ceremony um and as I mentioned earlier we're in for a bit of a treat now um we're going to be led out of the ceremony and into the reception by um our archaeological colleague and friend James McKee who will be playing a fourth Century a replica I have to say to museum colleagues a replica of a fourth Century um horned instrument discovered at lochner shade in County Arma um James will then be accompanied by his fellow performers from the world-renowned RMA rhymers who will keep that entertainment going in the reception area the rhymers wear masks they're crafted from flax Willow and straw and their activities are based on the traditional practice of mumming and this masked tradition of rhyming with its unique blend of Music Drama song and dance dates back over two and a half thousand years and is a very important part of Irish folk traditions and with that it's a wrap The Narrative we have I feel chosen to weave this evening is is one of partnership it's one of friendship it's one of solidarity it's one of recognizing cultural diversity here in Northern Ireland and elsewhere is one of recognizing academic Excellence of recognizing the talent and and hope of Youth and and of the hope and ambition of this place and indeed for archeology so put your hands together for this place for Belfast for Conference Center for all the Prize winners and all those involved this evening and all the organizers in the EA everybody too far too few to mention all those wonderful artists who've entertained us this evening and thank you for attending and on behalf of all involved may you have a super super time here in Belfast [Music] foreign