The Genomic History Of Southeastern Europe

Edinburgh Research Explorer The genomic history of southeastern Europe Citation for published version: Mathieson, I, Alpaslan-Roodenberg, S, Posth, C, Szécsényi-Nagy, A, Rohland, N, Mallick, SK, Olalde, I, Broomandkhoshbacht, N, Candilio, F, Cheronet, O, Fernandes, DN, Ferry, M, Gamarra, B, GonzálezFortes, G, Haak, W, Harney, E, Jones, E, Keating, DJ, Krause-Kyora, B, Kucukkalipci, I, Michel, M, Mittnik, A, Nägele, K, NOVAK, M, Oppenheimer, J, Patterson, N, Pfrengle, S, Sirak, K, Stewardson, K, Vai, S, Alexandrov, S, Alt, KW, Andreescu, R, Antonovi, D, Ash, A, Atanassova, N, Bacvarov, K, Gusztáv, MB, Bocherens, H, Bolus, M, Boronean, A, Boyadzhiev, Y, Budnik, BA, Burmaz, J, Chohadzhiev, S, Conard, NJ, Cottiaux, R, uka, M, Cupillard, C, Drucker, DG, Elenski, N, Francken, M, Galabova, B, Ganetsovski, G, Gély, B, Hajdu, T, Handzhyiska, V, Harvati, K, Higham, T, Iliev, S, Jankovi, I, Karavani, I, Kennett, DJ, Komšo, D, Kozak, AM, Labuda, D, Ansari-Lari, MA, Lazar, C, Leppek, M, Leshtakov, K, Lo Vetro, D, de los Rios, A, Lozanov, I, Malina, M, Martini, F, McSweeney, K, Meller, H, Mentuši, M, Mirea, P, Moiseyev, V, Petrova, V, Douglas Price, T, Simalcsik, A, Sineo, L, Šlaus, M, Slavchev, V, Stanev, P, Starovi, A, Szeniczey, T, Talamo, S, Teschler-Nicola, M, Thevenet, C, Valchev, I, Valentin, F, Vasilyev, S, Veljanovska, F, Venelinova, S, Veselovskaya, E, Viola, B, Virag, C, Zaninovi, J, Zaüner, S, Stockhammer, PW, Catalano, G, Krauß, R, Caramelli, D, Zarin, G, Gaydarska, B, Lillie, M, Nikitin, AG, Potekhina, I, Papathanasiou, KA, Bori, D, Bonsall, C, Krause, J, Pinhasi, R & Reich, D 2018, 'The genomic history of southeastern Europe', Nature, vol. 555, no. 7695, pp. 197-203. https://doi.org/10.1038/nature25778 Digital Object Identifier (DOI): 10.1038/nature25778 Link: Link to publication record in Edinburgh Research Explorer Document Version: Peer reviewed version Published In: Nature General rights Copyright for the publications made accessible via the Edinburgh Research Explorer is retained by the author(s) and / or other copyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associated with these rights. Take down policy The University of Edinburgh has made every reasonable effort to ensure that Edinburgh Research Explorer content complies with UK legislation. If you believe that the public display of this file breaches copyright please contact openaccess@ed.ac.uk providing details, and we will remove access to the work immediately and investigate your claim. Download date: 17. Jun. 2020 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 The Genomic History of Southeastern Europe Iain Mathieson† (1), Songül Alpaslan Roodenberg (1), Cosimo Posth (2,3), Anna SzécsényiNagy (4), Nadin Rohland (1), Swapan Mallick (1,5), Iñigo Olalde (1), Nasreen Broomandkhoshbacht (1,5), Francesca Candilio (6), Olivia Cheronet (6,7), Daniel Fernandes (6,8), Matthew Ferry (1,5), Beatriz Gamarra (6), Gloria González Fortes (9), Wolfgang Haak (2,10), Eadaoin Harney (1,5), Eppie Jones (11,12), Denise Keating (6), Ben Krause-Kyora (2), Isil Kucukkalipci (3), Megan Michel (1,5), Alissa Mittnik (2,3), Kathrin Nägele (2), Mario Novak (6,13), Jonas Oppenheimer (1,5), Nick Patterson (14), Saskia Pfrengle (3), Kendra Sirak (6,15), Kristin Stewardson (1,5), Stefania Vai (16), Stefan Alexandrov (17), Kurt W. Alt (18,19,20), Radian Andreescu (21), Dragana Antonović (22), Abigail Ash (6), Nadezhda Atanassova (23), Krum Bacvarov (17), Mende Balázs Gusztáv (4), Hervé Bocherens (24,25), Michael Bolus (26), Adina Boroneanţ (27), Yavor Boyadzhiev (17), Alicja Budnik (28), Josip Burmaz (29), Stefan Chohadzhiev (30), Nicholas J. Conard (31,25), Richard Cottiaux (32), Maja Čuka (33), Christophe Cupillard (34,35), Dorothée G. Drucker (25), Nedko Elenski (36), Michael Francken (37), Borislava Galabova (38), Georgi Ganetsovski (39), Bernard Gély (40), Tamás Hajdu (41), Veneta Handzhyiska (42), Katerina Harvati (37,25), Thomas Higham (43), Stanislav Iliev (44), Ivor Janković (13,45), Ivor Karavanić (46,45), Douglas J. Kennett (47), Darko Komšo (33), Alexandra Kozak (48), Damian Labuda (49), Martina Lari (16), Catalin Lazar (50,51), Maleen Leppek (52), Krassimir Leshtakov (42), Domenico Lo Vetro (53,54), Dženi Los (29), Ivaylo Lozanov (42), Maria Malina (26), Fabio Martini (53,54), Kath McSweeney (55), Harald Meller (20), Marko Menđušić (56), Pavel Mirea (57), Vyacheslav Moiseyev (58), Vanya Petrova (42), T. Douglas Price (59), Angela Simalcsik (60), Luca Sineo (61), Mario Šlaus (62), Vladimir Slavchev (63), Petar Stanev (36), Andrej Starović (64), Tamás Szeniczey (41), Sahra Talamo (65), Maria Teschler-Nicola (66,7), Corinne Thevenet (67), Ivan Valchev (42), Frédérique Valentin (68), Sergey Vasilyev (69), Fanica Veljanovska (70), Svetlana Venelinova (71), Elizaveta Veselovskaya (69), Bence Viola (72,73), Cristian Virag (74), Joško Zaninović (75), Steve Zäuner (76), Philipp W. Stockhammer (52,2), Giulio Catalano (61), Raiko Krauß (77), David Caramelli (16), Gunita Zariņa (78), Bisserka Gaydarska (79), Malcolm Lillie (80), Alexey G. Nikitin (81), Inna Potekhina (48), Anastasia Papathanasiou (82), Dušan Borić (83), Clive Bonsall (55), Johannes Krause (2,3), Ron Pinhasi* (6,7), David Reich* (1,14,5) * These authors contributed equally to the manuscript Present address; Department of Genetics, Perelman School of Medicine, University of Pennsylvania, Philadelphia PA 19104. † (1) Department of Genetics, Harvard Medical School, Boston 02115 MA USA (2) Department of Archaeogenetics, Max Planck Institute for the Science of Human History, 07745 Jena, Germany (3) Institute for Archaeological Sciences, University of Tuebingen, Germany (4) Laboratory of Archaeogenetics, Institute of Archaeology, Research Centre for the Humanities, Hungarian Academy of Sciences, H-1097 Budapest, Hungary (5) Howard Hughes Medical Institute, Harvard Medical School, Boston 02115 MA USA (6) Earth Institute and School of Archaeology, University College Dublin, Belfield, Dublin 4, Republic of Ireland (7) Department of Anthropology, University of Vienna, Althanstrasse 14, 1090 Vienna, Austria (8) CIAS, Department of Life Sciences, University of Coimbra, 3000-456 Coimbra, Portugal (9) Department of Life Sciences and Biotechnology, University of Ferrara, Via L. Borsari 46. Ferrara 44100 Italy (10) Australian Centre for Ancient DNA, School of Biological Sciences, The University of Adelaide, SA-5005 Adelaide, Australia (11) Smurfit Institute of Genetics, Trinity College Dublin, Dublin 2, Ireland (12) Department of Zoology, University of Cambridge, Downing Street, Cambridge CB2 3EJ, UK (13) Institute for Anthropological Research, Ljudevita Gaja 32, 10000 Zagreb, Croatia (14) Broad Institute of Harvard and MIT, Cambridge MA (15) Department of Anthropology, Emory University, Atlanta, Georgia 30322, USA (16) Dipartimento di Biologia, Università di Firenze, 50122 Florence, Italy (17) National Institute of Archaeology and Museum, Bulgarian Academy of Sciences, 2 Saborna Str., BG-1000 Sofia, Bulgaria (18) Danube Private University, A-3500 Krems, Austria (19) Department of Biomedical Engineering and Integrative Prehistory and Archaeological Science, CH-4123 Basel-Allschwil, Switzerland (20) State Office for Heritage Management and Archaeology Saxony-Anhalt and State Museum of Prehistory, D-06114 Halle, Germany (21) Romanian National History Museum, Bucharest, Romania (22) Institute of Archaeology, Belgrade, 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 Serbia (23) Institute of Experimental Morphology, Pathology and Anthropology with Museum, Bulgarian Academy of Sciences, Sofia, Bulgaria (24) Department of Geosciences, Biogeology, Universität Tübingen, Hölderlinstr. 12, 72074 Tübingen, Germany (25) Senckenberg Centre for Human Evolution and Palaeoenvironment, University of Tuebingen, 72072 Tuebingen, Germany (26) Heidelberg Academy of Sciences and Humanities, Research Center ‘‘The Role of Culture in Early Expansions of Humans’’ at the University of Tuebingen, Rümelinstraße 23, 72070 Tuebingen, Germany (27) ‘Vasile Pârvan’ Institute of Archaeology, Romanian Academy (28) Human Biology Department, Cardinal Stefan Wyszyński University, Warsaw, Poland (29) KADUCEJ d.o.o Papandopulova 27, 21000 Split, Croatia (30) St. Cyril and Methodius University, Veliko Turnovo, Bulgaria (31) Department of Early Prehistory and Quaternary Ecology, University of Tuebingen, Schloss Hohentübingen, 72070 Tuebingen, Germany (32) INRAP/UMR 8215 Trajectoires, 21 Alleé de l’Université, 92023 Nanterre, France (33) Archaeological Museum of Istria, Carrarina 3, 52100 Pula, Croatia (34) Service Régional de l'Archéologie de Bourgogne-Franche-Comté, 7 rue Charles Nodier, 25043 Besançon Cedex, France (35) Laboratoire Chronoenvironnement, UMR 6249 du CNRS, UFR des Sciences et Techniques, 16 route de Gray, 25030 Besançon Cedex, France (36) Regional Museum of History Veliko Tarnovo, Veliko Tarnovo, Bulgaria (37) Institute for Archaeological Sciences, Paleoanthropology, University of Tuebingen, Rümelinstraße 23, 72070 Tuebingen, Germany (38) Laboratory for human bio-archaeology, Bulgaria, 1202 Sofia, 42, George Washington str (39) Regional Museum of History, Vratsa, Bulgaria (40) DRAC Auvergne - Rhône Alpes, Ministère de la Culture, Le Grenier d'abondance 6, quai Saint Vincent 69283 LYON cedex 01 (41) Eötvös Loránd University, Faculty of Science, Institute of Biology, Department of Biological Anthropology, H-1117 Pázmány Péter sétány 1/c. Budapest, Hungary (42) Department of Archaeology, Sofia University St. Kliment Ohridski, Bulgaria (43) Oxford Radiocarbon Accelerator Unit, Research Laboratory for Archaeology and the History of Art, University of Oxford, Dyson Perrins Building, South Parks Road, OX1 3QY Oxford, UK (44) Regional Museum of History, Haskovo, Bulgaria (45) Department of Anthropology, University of Wyoming, 1000 E. University Avenue, Laramie, WY 82071, USA (46) Department of Archaeology, Faculty of Humanities and Social Sciences, University of Zagreb, Ivana Lučića 3, 10000 Zagreb, Croatia (47) Department of Anthropology and Institutes for Energy and the Environment, Pennsylvania State University, University Park, PA 16802 (48) Department of Bioarchaeology, Institute of Archaeology, National Academy of Sciences of Ukraine (49) CHU Sainte-Justine Research Center, Pediatric Department, Université de Montréal, Montreal, PQ, Canada, H3T 1C5 (50) National History Museum of Romania, Calea Victoriei, no. 12, 030026, Bucharest, Romania (51) University of Bucharest, Mihail Kogalniceanu 36-46, 50107, Bucharest, Romania (52) Institute for Pre- and Protohistoric Archaeology and the Archaeology of the Roman Provinces, Ludwig-Maximilians-University, Schellingstr. 12, 80799 Munich, Germany (53) Dipartimento SAGAS - Sezione di Archeologia e Antico Oriente, Università degli Studi di Firenze, 50122 Florence, Italy (54) Museo e Istituto fiorentino di Preistoria, 50122 Florence, Italy (55) School of History, Classics and Archaeology, University of Edinburgh, Edinburgh EH8 9AG, United Kingdom (56) Conservation Department in Šibenik, Ministry of Culture of the Republic of Croatia, Jurja Čulinovića 1, 22000 Šibenik, Croatia (57) Teleorman County Museum, str. 1848, no. 1, 140033 Alexandria, Romania (58) Peter the Great Museum of Anthropology and Ethnography (Kunstkamera) RAS, 199034 St. Petersburg, Russia (59) University of Wisconsin, Madison WI, USA (60) Olga Necrasov Centre for Anthropological Research, Romanian Academy – Iași Branch, Theodor Codrescu St. 2, P.C. 700481, Iași, Romania (61) Dipartimento di Scienze e tecnologie biologiche, chimiche e farmaceutiche, Lab. of Anthropology, Università degli studi di Palermo, Italy (62) Anthropological Center, Croatian Academy of Sciences and Arts, 10000 Zagreb, Croatia (63) Regional Historical Museum Varna, Maria Luiza Blvd. 41, BG-9000 Varna, Bulgaria (64) National Museum in Belgrade, 1a Republic sq., Belgrade, Serbia (65) Department of Human Evolution, Max Planck Institute for Evolutionary Anthropology, 04103 Leipzig, Germany (66) Department of Anthropology, Natural History Museum Vienna, 1010 Vienna, Austria (67) INRAP/UMR 8215 Trajectoires, 21 Allée de l’Université, 92023 Nanterre, France (68) CNRS/UMR 7041 ArScAn MAE, 21 Allée de l’Université, 92023 Nanterre, France (69) Institute of Ethnology and Anthropology, Russian Academy of Sciences, Leninsky Pr., 32a, Moscow, 119991, Russia (70) Archaeological Museum of Macedonia, Skopje (71) Regional museum of history, Shumen, Bulgaria (72) Department of Anthropology, University of Toronto, Toronto, Ontario, M5S 2S2, Canada (73) Institute of Archaeology & Ethnography, Siberian Branch, Russian Academy of Sciences, Lavrentiev Pr. 17, Novosibirsk 630090, Russia (74) Satu Mare County Museum Archaeology Department,V. Lucaciu, nr.21, Satu Mare, Romania (75) Municipal Museum Drniš, Domovinskog rata 54, 22320 Drniš, Croatia (76) anthropol - Anthropologieservice, Schadenweilerstraße 80, 72379 Hechingen, Germany (77) Institute for Prehistory, Early History and Medieval Archaeology, University of Tuebingen, Germany (78) Institute of Latvian History, University of Latvia, Kalpaka Bulvāris 4, Rīga 1050, Latvia (79) Department of Archaeology, Durham University, UK (80) School of Environmental Sciences: Geography, University of Hull, Hull HU6 7RX, UK (81) Department of Biology, Grand Valley State University, Allendale, Michigan, USA (82) Ephorate of Paleoanthropology and Speleology, Athens, Greece (83) The Italian Academy for Advanced Studies in America, Columbia University, 1161 Amsterdam Avenue, New York, NY 10027, USA. 114 Abstract 115 Farming was first introduced to Europe in the mid-7th millennium BCE–associated with 116 migrants from Anatolia who settled in the Southeast before spreading throughout 117 Europe. To understand the dynamics of this process, we analyzed genome-wide ancient 118 DNA data from 225 individuals who lived in southeastern Europe and surrounding 119 regions between 12,000 and 500 BCE. We document a West-East cline of ancestry in 120 indigenous hunter-gatherers and–in far-eastern Europe–early stages in the formation of 121 Bronze Age Steppe ancestry. We show that the first farmers of northern and western 122 Europe passed through southeastern Europe with limited hunter-gatherer admixture, 123 but that some groups that remained mixed extensively, without the male-biased hunter- 124 gatherer admixture that prevailed later in the North and West. Southeastern Europe 125 continued to be a nexus between East and West, with intermittent genetic contact with 126 the Steppe up to 2000 years before the migrations that replaced much of northern 127 Europe’s population. 128 129 Introduction 130 The southeastern quadrant of Europe was the beachhead in the spread of agriculture from its 131 source in the Fertile Crescent of southwestern Asia. After the first appearance of agriculture 132 in the mid-7th millennium BCE,1,2 farming spread westward via a Mediterranean and 133 northwestward via a Danubian route, and was established in both Iberia and Central Europe 134 by 5600 BCE.3,4 Ancient DNA studies have shown that the spread of farming across Europe 135 was accompanied by a massive movement of people5-8 closely related to the farmers of 136 northwestern Anatolia9-11 but nearly all the ancient DNA from Europe’s first farmers is from 137 central and western Europe, with only three individuals reported from the southeast.9 In the 138 millennia following the establishment of agriculture in the Balkan Peninsula, a series of 139 complex societies formed, culminating in sites such as the mid-5th millennium BCE necropolis 140 at Varna, which has some of the earliest evidence of extreme inequality in wealth, with one 141 individual (grave 43) from whom we extracted DNA buried with more gold than is known 142 from any earlier site. By the end of the 6th millennium BCE, agriculture had reached eastern 143 Europe, in the form of the Cucuteni-Trypillian complex in the area of present-day Moldova, 144 Romania and Ukraine, including “mega-sites” that housed hundreds, perhaps thousands, of 145 people.12 After around 4000 BCE, these settlements were largely abandoned, and 146 archaeological evidence documents cultural contacts with peoples of the Eurasian steppe.13 147 However, the population movements that accompanied these events have been unknown due 148 to the lack of ancient DNA. 149 150 Results 151 We generated genome-wide data from 225 ancient humans (216 reported for the first time), 152 from the Balkan Peninsula, the Carpathian Basin, the North Pontic Steppe and neighboring 153 regions, dated to 12,000-500 BCE (Figure 1, Supplementary Information Table 1, 154 Supplementary Information Note 1). We extracted DNA from skeletal remains in dedicated 155 clean rooms, built DNA libraries and enriched for DNA fragments overlapping 1.24 million 156 single nucleotide polymorphisms (SNPs), then sequenced the product and restricted to 157 libraries with evidence of authentic ancient DNA.7,10,14 We filtered out individuals with fewer 158 than 15,000 SNPs covered by at least one sequence, or that had unexpected ancestry for their 159 archaeological context and were not directly dated. We report, but do not analyze, nine 160 individuals that were first-degree relatives of others in the dataset, resulting in an analysis 161 dataset of 216 individuals. We analyzed these data together with 274 previously reported 162 ancient individuals,9-11,15-27 777 present-day individuals genotyped on the Illumina “Human 163 Origins” array,23 and 300 high coverage genomes from the Simons Genome Diversity Project 164 (SGDP).28 We used principal component analysis (PCA; Figure 1B, Extended Data Figure 1), 165 supervised and unsupervised ADMIXTURE (Figure 1D, Extended Data Figures 2&3),29 D- 166 statistics, qpAdm and qpGraph,30 along with archaeological and chronological information 167 (including 137 newly reported AMS 14C dates) to cluster the individuals into populations and 168 investigate the relationships among them. 169 170 We described the individuals in our dataset in terms of their genetic relatedness to a 171 hypothesized set of ancestral populations, which we refer to as their genetic ancestry. It has 172 previously been shown that the great majority of European ancestry derives from three 173 distinct sources.23 First, “hunter-gatherer-related” ancestry that is more closely related to 174 Mesolithic hunter-gatherers from Europe than to any other population, and can be further 175 subdivided into “Eastern” (EHG) and “Western” (WHG) hunter-gatherer-related ancestry.7 176 Second, “NW Anatolian Neolithic-related” ancestry related to the Neolithic farmers of 177 northwest Anatolia and tightly linked to the appearance of agriculture.9,10 The third source, 178 “steppe-related” ancestry, appears in Western Europe during the Late Neolithic to Bronze 179 Age transition and is ultimately derived from a population related to Yamnaya steppe 180 pastoralists.7,15 Steppe-related ancestry itself can be modeled as a mixture of EHG-related 181 ancestry, and ancestry related to Upper Palaeolithic hunter-gatherers of the Caucasus (CHG) 182 and the first farmers of northern Iran.19,21,22 183 184 Hunter-Gatherer substructure and transitions 185 Of the 216 new individuals we report, 106 from Paleolithic, Mesolithic and eastern European 186 Neolithic contexts have almost entirely hunter-gatherer-related ancestry (in eastern Europe, 187 unlike western Europe, “Neolithic” refers to the presence of pottery,31-33 not necessarily to 188 farming). These individuals form a cline from WHG to EHG that is correlated with geography 189 (Figure 1B), although it is neither geographically nor temporally uniform (Figure 2, Extended 190 Data Figure 4), and contains substructure in phenotypically important variants 191 (Supplementary Information Note 2). 192 193 From present-day Ukraine, our study reports new genome-wide data from seven Mesolithic 194 (~9500-6000 BCE) and 30 Neolithic (~6000-3500 BCE) individuals. On the cline from WHG- 195 to EHG-related ancestry, the Mesolithic individuals fall towards the East, intermediate 196 between EHG and Mesolithic hunter-gatherers from Scandinavia (Figure 1B).7 The Neolithic 197 population has a significant difference in ancestry compared to the Mesolithic (Figures 1B, 198 Figure 2), with a shift towards WHG shown by the statistic D(Mbuti, WHG, 199 Ukraine_Mesolithic, Ukraine_Neolithic); Z=8.5 (Supplementary Information Table 2). 200 Unexpectedly, one Neolithic individual from Dereivka (I3719), which we directly date to 201 4949-4799 BCE, has entirely NW Anatolian Neolithic-related ancestry. 202 203 The pastoralist Bronze Age Yamnaya complex originated on the Eurasian steppe and is a 204 plausible source for the dispersal of steppe-related ancestry into central and western Europe 205 around 2500 BCE.13 All previously reported Yamnaya individuals were from Samara7 and 206 Kalmykia15 in southwest Russia, and had entirely steppe-related ancestry. Here, we report 207 three Yamnaya individuals from further West – Ukraine and Bulgaria – and show that while 208 they all have high levels of steppe-related ancestry, one from Ozera in Ukraine and one from 209 Bulgaria (I1917 and Bul4, both dated to ~3000 BCE) have NW Anatolian Neolithic-related 210 admixture, the first evidence of such ancestry in Yamnaya–associated individuals (Figure 211 1B&D, Supplementary Data Table 2). Preceding the Yamnaya culture, four Copper Age 212 individuals (I4110, I5882, I5884 and I6561; Ukraine_Eneolithic) from Dereivka and 213 Alexandria dated to ~3600-3400 BCE have ancestry that is a mixture of hunter-gatherer-, 214 steppe- and NW Anatolian Neolithic-related (Figure 1D, Supplementary Data Table 2). 215 216 At Zvejnieki in Latvia (17 newly reported individuals, and additional data for 5 first reported 217 in Ref. 34) we observe a transition in hunter-gatherer-related ancestry that is opposite to that 218 seen in Ukraine. We find (Supplementary Data Table 3) that Mesolithic and Early Neolithic 219 individuals (Latvia_HG) associated with the Kunda and Narva cultures have ancestry 220 intermediate between WHG (~70%) and EHG (~30%), consistent with previous reports.34-36 221 We also detect a shift in ancestry between the Early Neolithic and individuals associated with 222 the Middle Neolithic Comb Ware Complex (Latvia_MN), who have more EHG-related 223 ancestry (we estimate 65% EHG, but two of four individuals appear to be 100% EHG in 224 PCA). The most recent individual, associated with the Final Neolithic Corded Ware Complex 225 (I4629, Latvia_LN), attests to another ancestry shift, clustering closely with Yamnaya from 226 Samara,7 Kalmykia15 and Ukraine (Figure 2). 227 228 We report new Upper Palaeolithic and Mesolithic data from southern and western Europe.17 229 Sicilian (I2158) and Croatian (I1875) individuals dating to ~12,000 and 6100 BCE cluster with 230 previously reported western hunter-gatherers (Figure 1B&D), including individuals from 231 Loschbour23 (Luxembourg, 6100 BCE), Bichon19 (Switzerland, 11,700 BCE), and Villabruna17 232 (Italy 12,000 BCE). These results demonstrate that WHG populations23 were widely 233 distributed from the Atlantic seaboard of Europe in the West, to Sicily in the South, to the 234 Balkan Peninsula in the Southeast, for at least six thousand years. 235 236 A particularly important hunter-gatherer population that we report is from the Iron Gates 237 region that straddles the border of present-day Romania and Serbia. This population 238 (Iron_Gates_HG) is represented in our study by 40 individuals from five sites. Modeling Iron 239 Gates hunter-gatherers as a mixture of WHG and EHG (Supplementary Table 3) shows that 240 they are intermediate between WHG (~85%) and EHG (~15%). However, this qpAdm model 241 does not fit well (p=0.0003, Supplementary Table 3) and the Iron Gates hunter-gatherers 242 show an affinity towards Anatolian Neolithic, relative to WHG (Supplementary Table 2). In 243 addition, Iron Gates hunter-gatherers carry mitochondrial haplogroup K1 (7/40) as well as 244 other subclades of haplogroups U (32/40) and H (1/40) in contrast to WHG, EHG and 245 Scandinavian hunter-gatherers who almost all carry haplogroups U5 or U2. One interpretation 246 is that the Iron Gates hunter-gatherers have ancestry that is not present in either WHG or 247 EHG. Possible scenarios include genetic contact between the ancestors of the Iron Gates 248 population and a NW Anatolian-Neolithic-related population, or that the Iron Gates 249 population is related to the source population from which the WHG split during a re- 250 expansion into Europe from the Southeast after the Last Glacial Maximum.17,37 251 252 A notable finding from the Iron Gates concerns the four individuals from the site of Lepenski 253 Vir, two of whom (I4665 & I5405, 6200-5600 BCE), have entirely NW Anatolian Neolithic- 254 related ancestry. Strontium and Nitrogen isotope data38 indicate that both these individuals 255 were migrants from outside the Iron Gates, and ate a primarily terrestrial diet (Supplementary 256 Information section 1). A third individual (I4666, 6070 BCE) has a mixture of NW Anatolian 257 Neolithic-related and hunter-gatherer-related ancestry and ate a primarily aquatic diet, while a 258 fourth, probably earlier, individual (I5407) had entirely hunter-gatherer-related ancestry 259 (Figure 1D, Supplementary Information section 1). We also identify one individual from 260 Padina (I5232), dated to 5950 BCE that had a mixture of NW Anatolian Neolithic-related and 261 hunter-gatherer-related ancestry. These results demonstrate that the Iron Gates was a region of 262 interaction between groups distinct in both ancestry and subsistence strategy. 263 264 Population transformations in the first farmers 265 Neolithic populations from present-day Bulgaria, Croatia, Macedonia, Serbia and Romania 266 cluster closely with the NW Anatolian Neolithic (Figure 1), consistent with archaeological 267 evidence.39 Modeling Balkan Neolithic populations as a mixture of NW Anatolian Neolithic 268 and WHG, we estimate that 98% (95% confidence interval [CI]; 97-100%) of their ancestry is 269 NW Anatolian Neolithic-related. A striking exception is evident in 8 out of 9 individuals from 270 Malak Preslavets in present-day Bulgaria.40 These individuals lived in the mid-6th millennium 271 BCE 272 populations (Figure 1B,D, Extended Data Figures 1-3, Supplementary Tables 2-4); a model of 273 82% (CI: 77-86%) NW Anatolian Neolithic-related, 15% (CI: 12-17%) WHG-related, and 4% 274 (CI: 0-9%) EHG-related ancestry fits the data. This hunter-gatherer-related ancestry with a 275 ~4:1 WHG:EHG ratio plausibly represents a contribution from local Balkan hunter-gatherers 276 genetically similar to those of the Iron Gates. Late Mesolithic hunter-gatherers in the Balkans 277 were likely concentrated along the coast and major rivers such as the Danube,41 which 278 directly connects the Iron Gates with Malak Preslavets. Thus, early farmer groups with the 279 most hunter-gatherer-related ancestry may have been those that lived close to the highest 280 densities of hunter-gatherers. and have significantly more hunter-gatherer-related ancestry than other Balkan Neolithic 281 282 In the Balkans, Copper Age populations (Balkans_Chalcolithic) harbor significantly more 283 hunter-gatherer-related ancestry than Neolithic populations as shown, for example, by the 284 statistic D(Mbuti, WHG, Balkans_Neolithic, Balkans_Chalcolithic); Z=4.3 ( Supplementary 285 Data Table 2). This is roughly contemporary with the “resurgence” of hunter-gatherer 286 ancestry previously reported in central Europe and Iberia7,10,42 and is consistent with changes 287 in funeral rites, specifically the reappearance around 4500 BCE of the Mesolithic tradition of 288 extended supine burial – in contrast to the Early Neolithic tradition of flexed burial.43 Four 289 individuals associated with the Copper Age Trypillian population have ~80% NW Anatolian- 290 related ancestry (Supplementary Table 3), confirming that the ancestry of the first farmers of 291 present-day Ukraine was largely derived from the same source as the farmers of Anatolia and 292 western Europe. Their ~20% hunter-gatherer ancestry is intermediate between WHG and 293 EHG, consistent with deriving from the Neolithic hunter-gatherers of the region. 294 295 We also report the first genetic data associated with the Late Neolithic Globular Amphora 296 Complex. Individuals from two Globular Amphora sites in Poland and Ukraine form a tight 297 cluster, showing high similarity over a large distance (Figure 1B,D). Both groups of Globular 298 Amphora Complex samples had more hunter-gatherer-related ancestry than Middle Neolithic 299 groups from Central Europe7 (we estimate 25% [CI: 22-27%] WHG ancestry, similar to 300 Chalcolithic Iberia, Supplementary Data Table 3). In east-central Europe, the Globular 301 Amphora Complex preceded or abutted the Corded Ware Complex that marks the appearance 302 of steppe-related ancestry,7,15 while in southeastern Europe, the Globular Amphora Complex 303 bordered populations with steppe-influenced material cultures for hundreds of years44 and yet 304 the individuals in our study have no evidence of steppe-related ancestry, supporting the 305 hypothesis that this material cultural frontier was also a barrier to gene flow. 306 307 The movements from the Pontic-Caspian steppe of individuals similar to those associated 308 with the Yamnaya Cultural Complex in the 3rd millennium BCE contributed about 75% of the 309 ancestry of individuals associated with the Corded Ware Complex and about 50% of the 310 ancestry of succeeding material cultures such as the Bell Beaker Complex in central 311 Europe.7,15 In two directly dated individuals from southeastern Europe, one (ANI163) from 312 the Varna I cemetery dated to 4711-4550 BCE and one (I2181) from nearby Smyadovo dated 313 to 4550-4450 BCE, we find far earlier evidence of steppe-related ancestry (Figure 1B,D). 314 These findings push back the first evidence of steppe-related ancestry this far West in Europe 315 by almost 2,000 years, but it was sporadic as other Copper Age (~5000-4000 BCE) individuals 316 from the Balkans have no evidence of it. Bronze Age (~3400-1100 BCE) individuals do have 317 steppe-related ancestry (we estimate 30%; CI: 26-35%), with the highest proportions in the 318 four latest Balkan Bronze Age individuals in our data (later than ~1700 BCE) and the least in 319 earlier Bronze Age individuals (3400-2500 BCE; Figure 1D). 320 321 A new source of ancestry in Neolithic Europe 322 An important question about the initial spread of farming into Europe is whether the first 323 farmers that brought agriculture to northern Europe and to southern Europe were derived from 324 a single population or instead represent distinct migrations. We confirm that Mediterranean 325 populations, represented in our study by individuals associated with the Epicardial Early 326 Neolithic from Iberia7, are closely related to Danubian populations represented by the 327 Linearbandkeramik (LBK) from central Europe7,45 and that both are closely related to the 328 Balkan Neolithic population. These three populations form a clade with the NW Anatolian 329 Neolithic individuals as an outgroup, consistent with a single migration into the Balkan 330 peninsula, which then split into two (Supplementary Information Note 3). 331 332 In contrast, five southern Greek Neolithic individuals (Peloponnese_Neolithic) – three (plus 333 one from Ref. 26) from Diros Cave and one from Franchthi Cave – are not consistent with 334 descending from the same source population as other European farmers. D-statistics 335 (Supplementary Information Table 2) show that in fact, these “Peloponnese Neolithic” 336 individuals dated to ~4000 BCE are shifted away from WHG and towards CHG, relative to 337 Anatolian and Balkan Neolithic individuals. We see the same pattern in a single Neolithic 338 individual from Krepost in present-day Bulgaria (I0679_d, 5718-5626 BCE). An even more 339 dramatic shift towards CHG has been observed in individuals associated with the Bronze Age 340 Minoan and Mycenaean cultures,26 suggesting gene flow into the region from populations 341 with CHG-rich ancestry throughout the Neolithic, Chalcolithic and Bronze Age. Possible 342 sources are related to the Neolithic population from the central Anatolian site of Tepecik 343 Çiftlik,21 or the Aegean site of Kumtepe,11 who are also shifted towards CHG relative to NW 344 Anatolian Neolithic samples, as are later Copper and Bronze Age Anatolians.10,26 345 346 Sex-biased admixture between hunter-gatherers and farmers 347 We provide the first evidence for sex-biased admixture between hunter-gatherers and farmers 348 in Europe, showing that the Middle Neolithic “resurgence” of hunter-gatherer-related 349 ancestry7,42 in central Europe and Iberia was driven more by males than by females (Figure 350 3B&C, Supplementary Data Table 5, Extended Data Figure 5). To document this we used 351 qpAdm to compute ancestry proportions on the autosomes and the X chromosome; since 352 males always inherit a maternal X chromosome, differences imply sex-biased mixture. In the 353 Balkan Neolithic there is no evidence of sex bias (Z=0.27 where a positive Z-score implies 354 male hunter-gatherer bias), nor in the LBK and Iberian_Early Neolithic (Z=-0.22 and 1.09). In 355 the Copper Age there is clear bias: weak in the Balkans (Z=1.66), but stronger in Iberia 356 (Z=3.08) and Central Europe (Z=2.74). Consistent with this, hunter-gatherer mitochondrial 357 haplogroups (haplogroup U)46 are rare and within the intervals of genome-wide ancestry 358 proportions, but hunter-gatherer-associated Y chromosomes (haplogroups I2, R1 and C1)17 359 are more common: 7/9 in the Iberian Neolithic/Copper Age and 9/10 in Middle-Late Neolithic 360 Central Europe (Central_MN and Globular_Amphora) (Figure 3C). 361 362 No evidence that steppe-related ancestry moved through southeast Europe into Anatolia 363 One version of the Steppe Hypothesis of Indo-European language origins suggests that Proto- 364 Indo-European languages developed north of the Black and Caspian seas, and that the earliest 365 known diverging branch – Anatolian – was spread into Asia Minor by movements of steppe 366 peoples through the Balkan peninsula during the Copper Age around 4000 BCE.47 If this were 367 correct, then one way to detect evidence of it would be the appearance of large amounts of 368 steppe-related ancestry first in the Balkan Peninsula, and then in Anatolia. However, our data 369 show no evidence for this scenario. While we find sporadic steppe-related ancestry in Balkan 370 Copper and Bronze Age individuals, this ancestry is rare until the late Bronze Age. Moreover, 371 while Bronze Age Anatolian individuals have CHG-related ancestry,26 they have neither the 372 EHG-related ancestry characteristic of all steppe populations sampled to date,19 nor the 373 WHG-related ancestry that is ubiquitous in Neolithic southeastern Europe (Extended Data 374 Figure 2&3, Supplementary Data Table 2). An alternative hypothesis is that the ultimate 375 homeland of Proto-Indo-European languages was in the Caucasus or in Iran. In this scenario, 376 westward movement contributed to the dispersal of Anatolian languages, and northward 377 movement and mixture with EHG was responsible for the formation of a “Late Proto-Indo 378 European”-speaking population associated with the Yamnaya Complex.13 While this scenario 379 gains plausibility from our results, it remains possible that Indo-European languages were 380 spread through southeastern Europe into Anatolia without large-scale population movement 381 or admixture. 382 Discussion 383 Our study shows that southeastern Europe consistently served as a genetic contact zone. 384 Before the arrival of farming, the region saw interaction between diverged groups of hunter- 385 gatherers, and this interaction continued after farming arrived. While this study has clarified 386 the genomic history of southeastern Europe from the Mesolithic to the Bronze Age, the 387 processes that connected these populations to the ones living today remain largely unknown. 388 An important direction for future research will be to sample populations from the Bronze 389 Age, Iron Age, Roman, and Medieval periods and to compare them to present-day 390 populations to understand how these transitions occurred. 391 Methods 392 393 Ancient DNA Analysis 394 We extracted DNA and prepared next-generation sequencing libraries in four different 395 dedicated ancient DNA laboratories (Adelaide, Boston, Budapest, and Tuebingen). We also 396 prepared samples for extraction in a fifth laboratory (Dublin), from whence it was sent to 397 Boston for DNA extraction and library preparation (Supplementary Table 1). 398 399 Two samples were processed at the Australian Centre for Ancient DNA, Adelaide, Australia, 400 according to previously published methods7 and sent to Boston for subsequent screening, 401 1240k capture and sequencing. 402 403 Seven samples were processed27 at the Institute of Archaeology RCH HAS, Budapest, 404 Hungary, and amplified libraries were sent to Boston for screening, 1240k capture and 405 sequencing. 406 407 Seventeen samples were processed at the Institute for Archaeological Sciences of the 408 University of Tuebingen and at the Max Planck Institute for the Science of Human History in 409 Jena, Germany. Extraction48 and library preparation49,50 followed established protocols. We 410 performed in-solution capture as described below (“1240k capture”) and sequenced on an 411 Illumina HiSeq 4000 or NextSeq 500 for 76bp using either single- or paired-end sequencing. 412 413 The remaining 199 samples were processed at Harvard Medical School, Boston, USA. From 414 about 75mg of sample powder from each sample (extracted in Boston or University College 415 Dublin, Dublin, Ireland), we extracted DNA following established methods48 replacing the 416 column assembly with the column extenders from a Roche kit.51 We prepared double 417 barcoded libraries with truncated adapters from between one ninth and one third of the DNA 418 extract. Most libraries included in the nuclear genome analysis (90%) were subjected to 419 partial (“half”) Uracil-DNA-glycosylase (UDG) treatment before blunt end repair. This 420 treatment reduces by an order of magnitude the characteristic cytosine-to-thymine errors of 421 ancient DNA data52, but works inefficiently at the 5’ ends,50 thereby leaving a signal of 422 characteristic damage at the terminal ends of ancient sequences. Some libraries were not 423 UDG-treated (“minus”). For some samples we increased coverage by preparing additional 424 libraries from the existing DNA extract using the partial UDG library preparation, but 425 replacing the MinElute column cleanups in between enzymatic reactions with magnetic bead 426 cleanups, and the final PCR cleanup with SPRI bead cleanup.53,54 427 We screened all libraries from Adelaide, Boston and Budapest by enriching for the 428 mitochondrial genome plus about 3,000 (50 in an earlier, unpublished, version) nuclear SNPs 429 using a bead-capture55 but with the probes replaced by amplified oligonucleotides synthesized 430 by CustomArray Inc. After the capture, we completed the adapter sites using PCR, attaching 431 dual index combinations56 to each enriched library. We sequenced the products of between 432 100 and 200 libraries together with the non-enriched libraries (shotgun) on an Illumina 433 NextSeq500 using v2 150 cycle kits for 2x76 cycles and 2x7 cycles. 434 435 In Boston, we performed two rounds of in-solution enrichment (“1240k capture”) for a 436 targeted set of 1,237,207 SNPs using previously reported protocols.7,14,23 For a total of 34 437 individuals, we increased coverage by building one to eight additional libraries for the same 438 sample. When we built multiple libraries from the same extract, we often pooled them in 439 equimolar ratios before the capture. We performed all sequencing on an Illumina NextSeq500 440 using v2 150 cycle kits for 2x76 cycles and 2x7 cycles. We attempted to sequence each 441 enriched library up to the point where we estimated that it was economically inefficient to 442 sequence further. Specifically, we iteratively sequenced more and more from each individual 443 and only stopped when we estimated that the expected increase in the number of targeted 444 SNPs hit at least once would be less than about one for every 100 new read pairs generated. 445 After sequencing, we trimmed two bases from the end of each read and aligned to the human 446 genome (b37/hg19) using bwa.57 We then removed individuals with evidence of 447 contamination based on mitochondrial DNA polymorphism58 or difference in PCA space 448 between damaged and undamaged reads59, a high rate of heterozygosity on chromosome X 449 despite being male59,60, or an atypical ratio of X-to-Y sequences. We also removed individuals 450 that had low coverage (fewer than 15,000 SNPs hit on the autosomes). We report, but do not 451 analyze, data from nine individuals that were first-degree relatives of others in the dataset 452 (determined by comparing rates of allele sharing between pairs of individuals). 453 454 After removing a small number of sites that failed to capture, we were left with a total of 455 1,233,013 sites of which 32,670 were on chromosome X and 49,704 were on chromosome Y, 456 with a median coverage at targeted SNPs on the 216 newly reported individuals of 0.90 457 (range 0.007-9.2; Supplementary Table 1). We generated “pseudo-haploid” calls by selecting 458 a single read randomly for each individual at each SNP. Thus, there is only a single allele 459 from each individual at each site, but adjacent alleles might come from either of the two 460 haplotypes of the individual. We merged the newly reported data with previously reported 461 data from 274 other ancient individuals9-11,15-27, making pseudo-haploid calls in the same way 462 at the 1240k sites for individuals that were shotgun sequenced rather than captured. 463 464 Using the captured mitochondrial sequence from the screening process, we called 465 mitochondrial haplotypes. Using the captured SNPs on the Y chromosome, we called Y 466 chromosome haplogroups for males by restricting to sequences with mapping quality ≥30 and 467 bases with base quality ≥30. We determined the most derived mutation for each individual, 468 using the nomenclature of the International Society of Genetic Genealogy 469 (http://www.isogg.org) version 11.110 (21 April 2016). 470 471 Population genetic analysis 472 To analyze these ancient individuals in the context of present day genetic diversity, we 473 474 475 merged them with the following two datasets: 476 477 478 479 1. 300 high coverage genomes from a diverse worldwide set of 142 populations sequenced as part of the Simons Genome Diversity Project28 (SGDP merge). 2. 777 West Eurasian individuals genotyped on the Human Origins array23, with 597,573 sites in the merged dataset (HO merge). 480 481 We computed principal components of the present-day individuals in the HO merge and 482 projected the ancient individuals onto the first two components using the “lsqproject: YES” 483 option in smartpca (v15100)61 (https://www.hsph.harvard.edu/alkes-price/software/). 484 485 We ran ADMIXTURE (v1.3.0) in both supervised and unsupervised mode. In supervised mode 486 we used only the ancient individuals, on the full set of SNPs, and the following population 487 labels fixed:     488 489 490 491 492 493 Anatolia_Neolithic WHG EHG Yamnaya For unsupervised mode we used the HO merge, including 777 present-day individuals. We 494 flagged individuals that were genetic outliers based on PCA and ADMIXTURE, relative to 495 other individuals from the same time period and archaeological culture. 496 497 We computed D-statistics using qpDstat (v710). D-statistics of the form D(A,B,X,Y) test the 498 null hypothesis of the unrooted tree topology ((A,B),(X,Y)). A positive value indicates that 499 either A and X, or B and Y, share more drift than expected under the null hypothesis. We 500 quote D-statistics as the Z-score computed using default block jackknife parameters. 501 502 We fitted admixture proportions with qpAdm (v610) using the SGDP merge. Given a set of 503 outgroup (“right”) populations, qpAdm models one of a set of source (“left”) populations (the 504 “test” population) as a mixture of the other sources by fitting admixture proportions to match 505 the observed matrix of f4-statistics as closely as possible. We report a p-value for the null 506 hypothesis that the test population does not have ancestry from another source that is 507 differentially related to the right populations. We computed standard errors for the mixture 508 proportions using a block jackknife. Importantly, qpAdm does not require that the source 509 populations are actually the admixing populations, only that they are a clade with the correct 510 admixing populations, relative to the other sources. Infeasible coefficient estimates (i.e. 511 outside [0,1]) are usually a sign of poor model fit, but in the case where the source with a 512 negative coefficient is itself admixed, could be interpreted as implying that the true source is a 513 population with different admixture proportions. We used the following set of seven 514 populations as outgroups or “right populations”: 515 516 517 518 519 520 521 522         Mbuti.DG Ust_Ishim_HG_published.DG Mota.SG MA1_HG.SG Villabruna Papuan.DG Onge.DG Han.DG 523 524 For some analyses where we required extra resolution (Supplementary Data Table 4) we used 525 an extended set of 14 right (outgroup) populations, including additional Upper Paleolithic 526 European individuals17: 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541               ElMiron Mota.SG Mbuti.DG Ust_Ishim_HG_published.DG MA1_HG.SG AfontovaGora3 GoyetQ116-1_published Villabruna Kostenki14 Vestonice16 Karitiana.DG Papuan.DG Onge.DG Han.DG 542 We also fitted admixture graphs with qpGraph (v6021)30 (https://github.com/DReichLab/ 543 AdmixTools, Supplementary Information Note 3). Like qpAdm, qpGraph also tries to match a 544 matrix of f-statistics, but rather than fitting one population as a mixture of other, specified, 545 populations, it fits the relationship between all tested populations simultaneously, potentially 546 incorporating multiple admixture events. However, qpGraph requires the graph relating 547 populations to be specified in advance. We tested goodness-of-fit by computing the expected 548 D-statistics under the fitted model, finding the largest D-statistic outlier between the fitted and 549 observed model, and computing a Z-score using a block jackknife. 550 551 For 114 individuals with hunter-gatherer-related ancestry we estimated an effective migration 552 surface using the software EEMS (https://github.com/dipetkov/eems)62. We computed 553 pairwise differences between individuals using the bed2diffs2 program provided with EEMS. 554 We set the number of demes to 400 and defined the outer boundary of the region by the 555 polygon (in latitude-longitude co-ordinates) [(66,60), (60,10), (45,-15), (35,-10), (35,60)]. We 556 ran the MCMC ten times with different random seeds, each time with one million burn-in and 557 four million regular iterations, thinned to one in ten thousand. 558 559 To analyze potential sex bias in admixture, we used qpAdm to estimate admixture proportions 560 on the autosomes (default option) and on the X chromosome (option “chrom: 23”). We 561 562 computed Z-scores for the difference between the autosomes and the X chromosome as 𝑍 = 563 the X chromosome, and σA and σX are the corresponding jackknife standard deviations. Thus, 564 a positive Z-score means that there is more hunter-gatherer admixture on the autosomes than 565 on the X chromosome, indicating that the hunter-gatherer admixture was male-biased. 566 Because X chromosome standard errors are high and qpAdm results can be sensitive to which 567 population is first in the list of outgroup populations, we checked that the patterns we observe 568 were robust to cyclic permutation of the outgroups. To compare frequencies of hunter- 569 gatherer uniparental markers, we counted the individuals with mitochondrial haplogroup U 570 and Y chromosome haplogroups C1, I2 and R1, which are all common in Mesolithic hunter- 571 gatherers but rare or absent in Anatolian Neolithic individuals. The Iron Gates hunter- 572 gatherers also carry H and K1 mitochondrial haplogroups so the proportion of haplogroup U 573 represents the minimum maternal hunter-gatherer contribution. We computed binomial 574 confidence intervals for the proportion of haplogroups associated with each ancestry type 575 using the Agresti-Coull method63,64 implemented in the binom package in R. 𝑝𝐴 −𝑝𝑋 2 +𝜎2 √𝜎𝐴 𝑋 where pA and pX are the hunter-gatherer admixture proportions on the autosomes and 576 577 Given autosomal and X chromosome admixture proportions, we estimated the proportion of 578 male and female hunter-gatherer ancestors by assuming a single-pulse model of admixture. If 579 the proportions of male and female ancestors that are hunter-gatherer-related are given by m 580 and f, respectively, then the proportions of hunter-gatherer-related ancestry on the autosomes 𝑚+𝑓 2 and 𝑚+2𝑓 . 3 581 and the X chromosome are given by We approximated the sampling error in 582 the observed admixture proportions by the estimated jackknife error and computed the 583 likelihood surface for (m,f) over a grid ranging from (0,0) to (1,1). 584 585 Direct AMS 14C Bone Dates 586 We report 137 new direct AMS 14C bone dates for 136 individuals from multiple AMS 587 radiocarbon laboratories. In general, bone samples were manually cleaned and demineralized 588 in weak HCl and, in most cases (PSU, UCIAMS, OxA), soaked in an alkali bath (NaOH) at 589 room temperature to remove contaminating soil humates. Samples were then rinsed to 590 neutrality in Nanopure H2O and gelatinized in HCL.65 The resulting gelatin was lyophilized 591 and weighed to determine percent yield as a measure of collagen preservation (% crude 592 gelatin yield). Collagen was then directly AMS 14C dated (Beta, AA) or further purified using 593 ultrafiltration (PSU, UCIAMS, OxA, Poz, MAMS).66 It is standard in some laboratories 594 (PSU/UCIAMS, OxA) to use stable carbon and nitrogen isotopes as an additional quality 595 control measure. For these samples, the %C, %N and C:N ratios were evaluated before AMS 596 14 597 collagen preservation.68 For 119 of the new dates, we also report δ13C and δ15N values 598 (Supplementary Table 6). C dating.67 C:N ratios for well-preserved samples fall between 2.9 and 3.6, indicating good 599 600 All 14C ages were δ13C-corrected for mass dependent fractionation with measured 13C/12C 601 values69 and calibrated with OxCal version 4.2.370 using the IntCal13 northern hemisphere 602 calibration curve.70 For hunter-gatherers from the Iron Gates, the direct 14C dates tend to be 603 overestimates because of the freshwater reservoir effect (FRE), which arises because of a diet 604 including fish that consumed ancient carbon, and for these individuals we performed a 605 correction (Supplementary Information Note 1),71 assuming that 100% FRE = 545±70 yr, and 606 δ15N values of 8.3% and 17.0% for 100% terrestrial and aquatic diets, respectively. 607 608 Data Availability 609 The aligned sequences are available through the European Nucleotide Archive under 610 accession number PRJEB22652. The pseudo-haploid genotype dataset used in analysis and 611 consensus mitochondrial genomes are available at https://reich.hms.harvard.edu/datasets. 612 Code Availability 613 Software used to analyze the data is available from the following sources: 614 smartpca, qpAdm, qpDstat, qpGraph: https://github.com/DReichLab/AdmixTools/ 615 ADMIXTURE: https://www.genetics.ucla.edu/software/admixture/ 616 EEMS: https://github.com/dipetkov/eems/ 617 bwa: http://bio-bwa.sourceforge.net 618 OxCal: https://c14.arch.ox.ac.uk/oxcal.html 619 620 Acknowledgments 621 We thank David Anthony, Iosif Lazaridis, and Mark Lipson for comments on the manuscript, 622 Bastien Llamas, Alan Cooper and Anja Furtwängler for contributions to laboratory work, 623 Richard Evershed for contributing 14C dates and Friederike Novotny for assistance with 624 samples. Support for this project was provided by the Human Frontier Science Program 625 fellowship LT001095/2014-L to I.M.; by DFG grant AL 287 / 14-1 to K.W.A.; by Irish 626 Research Council grant GOIPG/2013/36 to D.F.; by the NSF Archaeometry program BCS- 627 1460369 to DJK (for AMS 14C work at Penn State); by MEN-UEFISCDI grant, Partnerships 628 in Priority Areas Program – PN II (PN-II-PT-PCCA-2013-4-2302) to C.L.; by Croatian 629 Science Foundation grant IP-2016-06-1450 to M.N.; by European Research Council grant 630 ERC StG 283503 and Deutsche Forschungsgemeinschaft DFG FOR2237 to K.H.; by ERC 631 starting grant ADNABIOARC (263441) to R.P.; and by US National Science Foundation 632 HOMINID grant BCS-1032255, US National Institutes of Health grant GM100233, the 633 Howard Hughes Medical Institute, and an Allen Discovery Center grant from the Paul Allen 634 Foundation to D.R. 635 636 Author Contributions 637 SAR, AS-N, SVai, SA, KWA, RA, DA, AA, NA, KB, MBG, HB, MB, ABo, YB, ABu, JB, 638 SC, NC, RC, MC, CC, DD, NE, MFr, BGal, GG, BGe, THa, VH, KH, THi, SI, IJ, IKa, DKa, 639 AK, DLa, MLa, CL, MLe, KL, DLV, DLo, IL, MMa, FM, KM, HM, MMe, PM, VM, VP, 640 TDP, ASi, LS, MŠ, VS, PS, ASt, TS, MT-N, CT, IV, FVa, SVas, FVe, SV, EV, BV, CV, JZ, 641 SZ, PWS, GC, RK, DC, GZ, BGay, MLi, AGN, IP, AP, DB, CB, JK, RP & DR assembled 642 and interpreted archaeological material. CP, AS-N, NR, NB, FC, OC, DF, MFe, BGam, GGF, 643 WH, EH, EJ, DKe, BK-K, IKu, MMi, AM, KN, MN, JO, SP, KSi, KSt & SVai performed 644 laboratory work. IM, CP, AS-N, SM, IO, NP & DR analyzed data. DJK, ST, DB, CB 645 interpreted 14C dates. JK, RP & DR supervised analysis or laboratory work. IM & DR wrote 646 the paper, with input from all co-authors. 647 648 Author Information 649 Reprints and permissions information is available at www.nature.com/reprints. The authors 650 declare that they have no competing financial interests. Correspondence and requests for 651 materials should be addressed to DR (reich@genetics.med.harvard.edu) or RP 652 (ron.pinhasi@ucd.ie) or IM (mathi@pennmedicine.upenn.edu) 653 Figure captions 654 655 Figure 1: Geographic and genetic structure of 216 newly reported individuals. A: 656 Locations of newly reported individuals. B: Ancient individuals projected onto principal 657 components defined by 777 present-day West Eurasians (shown in Extended Data Figure 1). 658 Includes selected published individuals (faded circles, labeled) and newly reported individuals 659 (other symbols, outliers enclosed in black circles). Colored polygons cover individuals that 660 had cluster memberships fixed at 100% for supervised admixture analysis. C: Date (direct or 661 contextual) for each sample and approximate chronology of southeastern Europe. D: 662 Supervised ADMIXTURE analysis, modeling each ancient individual (one per row), as a 663 mixture of population clusters constrained to contain Anatolian Neolithic (grey), Yamnaya 664 from Samara (yellow), EHG (pink) and WHG (green) populations. Dates in parentheses 665 indicate approximate range of individuals in each population. See Extended Data Figure 2 for 666 individual sample IDs. Map data in A from the R package maps. 667 668 Figure 2: Structure and change in hunter-gatherer-related populations. Inferred ancestry 669 proportions for populations modeled as a mixture of WHG, EHG and CHG (Supplementary 670 Table S3.1.3). Dashed lines show populations from the same geographic region. Percentages 671 indicate proportion of WHG+EHG ancestry. Standard errors range from 1.5-8.3% 672 (Supplementary Table S3.1.3). 673 674 Figure 3: Structure and change in NW Anatolian Neolithic-related populations. A: 675 Populations modeled as a mixture of NW Anatolia Neolithic, WHG, and EHG. Dashed lines 676 show temporal relationships between populations from the same geographic region. 677 Percentages indicate proportion of WHG+EHG ancestry. Standard errors range from 0.7- 678 6.0% (Supplementary Table S3.2.2). B: Z-scores for the difference in hunter-gatherer-related 679 ancestry on the autosomes compared to the X chromosome when populations are modeled as 680 a mixture of NW Anatolia Neolithic and WHG (N=126 individuals, group sizes in 681 parentheses). Positive values indicate more hunter-gatherer-related ancestry on the autosomes 682 and thus male-biased hunter-gatherer ancestry. “Combined” populations merge all individuals 683 from different times from a geographic area. C: Hunter-gatherer-related ancestry proportions 684 on the autosomes, X chromosome, mitochondrial DNA (i.e. mt haplogroup U), and the Y 685 chromosome (i.e. Y chromosome haplogroups I2, R1 and C1). Points show qpAdm 686 (autosomes and X chromosome) or maximum likelihood (MT and Y chromosome) estimates 687 and bars show approximate 95% confidence intervals (N=109 individuals, group sizes in 688 parentheses). 689 690 Extended Data Figure Captions 691 Extended Data Figure 1: PCA of 486 ancient individuals, projected onto principal 692 components defined by 777 present-day West Eurasian individuals (grey points). This differs 693 from Figure 1B in that the plot is not cropped and the present-day individuals are shown. 694 695 Extended Data Figure 2: Supervised ADMIXTURE analysis modeling each ancient 696 individual (one per row), as a mixture of populations represented by clusters that are 697 constrained to contain Anatolian Neolithic (grey), Yamnaya from Samara (yellow), EHG 698 (pink) and WHG (green) populations. Dates in parentheses indicate approximate range of 699 individuals in each population. This differs from Figure 1D in that it contains some previously 700 published samples, and includes sample IDs. 701 702 Extended Data Figure 3: Unsupervised ADMIXTURE plot from k=4 to 12, on a dataset 703 consisting of 1099 present-day individuals and 476 ancient individuals. We show newly 704 reported ancient individuals and some previously published individuals for comparison. 705 706 Extended Data Figure 4: Spatial structure in hunter-gatherers. Estimated effective migration 707 surface (EEMS).62 This fits a model of genetic relatedness where individuals move (in a 708 random direction) from generation to generation on an underlying grid so that genetic 709 relatedness is determined by distance. The migration parameter m defines the local rate of 710 migration, varies on the grid and is inferred. This plot shows log10(m), scaled relative to the 711 average migration rate (which is arbitrary). Thus log10(m)=2, for example, implies that the 712 rate of migration at this point on the grid is 100 times higher than average. To restrict as much 713 as possible to hunter-gatherer structure, the migration surface is inferred using data from 116 714 individuals that date to earlier than ~5000 BCE and have no NW Anatolian-related ancestry. 715 Though the migration surface is sensitive to sampling, and fine-scale features may not be 716 interpretable, the migration “barrier” (region of low migration) running north-south and 717 separating populations with primarily WHG from primarily EHG ancestry seems to be robust, 718 and consistent with inferred admixture proportions. This analysis suggests that Mesolithic 719 hunter-gatherer population structure was clustered and not smoothly clinal, in the sense that 720 genetic differentiation did not vary consistently with distance. Superimposed on this 721 background, pies show the WHG, EHG and CHG ancestry proportions inferred for 722 populations used to construct the migration surface (another way of visualizing the data in 723 Figure 2, Supplementary Table 3.1.3 – we use two population models if they fit with p>0.01, 724 and three population models otherwise). Pies with only a single color are those that were 725 fixed to be the source populations. 726 727 Extended Data Figure 5: log-likelihood surfaces for the proportion of female (x-axis) and 728 male (y-axis) ancestors that are hunter-gatherer-related for the combined populations 729 analyzed in Figure 3C, and the two populations with the strongest evidence for sex-bias. 730 Numbers in parentheses give the number of individuals in each group. Log-likelihood scale 731 ranges from 0 to -10, where 0 is the feasible point with the highest likelihood. 732 Main Text References 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Tringham, R. E. in The Transition to Agriculture in Prehistoric Europe (ed D. Price) 19-56 (Cambridge University Press, 2000). Bellwood, P. First Farmers: The Origins of Agricultural Societies. 2nd edn, (WileyBlackwell, 2004). Golitko, M. in Ancient Europe, 8000 B.C. to A.D. 1000: An Encyclopedia of the Barbarian World (eds P. Bogucki & P.J. Crabtree) 259-266 (Charles Scribners & Sons, 2003). Vander Linden, M. in Investigating Archaeological Cultures: Material Culture, Variability, and Transmission (eds B.W. Roberts & M. Vander Linden) 289-319 (Springer, 2012). Bramanti, B. et al. Genetic discontinuity between local hunter-gatherers and central Europe's first farmers. Science 326, 137-140 (2009). Skoglund, P. et al. Origins and genetic legacy of Neolithic farmers and huntergatherers in Europe. Science 336, 466-469 (2012). Haak, W. et al. Massive migration from the steppe was a source for Indo-European languages in Europe. Nature 522, 207-211 (2015). Cassidy, L. M. et al. Neolithic and Bronze Age migration to Ireland and establishment of the insular Atlantic genome. Proc. Natl. Acad. Sci. U. S. A. 113, 368-373 (2016). Hofmanova, Z. et al. Early farmers from across Europe directly descended from Neolithic Aegeans. Proc. Natl. Acad. Sci. U. S. A. 113, 6886-6891 (2016). Mathieson, I. et al. Genome-wide patterns of selection in 230 ancient Eurasians. Nature 528, 499-503 (2015). Omrak, A. et al. Genomic Evidence Establishes Anatolia as the Source of the European Neolithic Gene Pool. Curr. Biol. 26, 270-275 (2016). Müller, J., Rassmann, K. & Videiko, M. Trypillia Mega-Sites and European Prehistory: 4100–3400 BCE., (Routledge, 2016). Anthony, D. W. The horse the wheel and language. (Princeton University Press, 2007). Fu, Q. et al. An early modern human from Romania with a recent Neanderthal ancestor. Nature (2015). Allentoft, M. E. et al. Population genomics of Bronze Age Eurasia. Nature 522, 167172 (2015). Fu, Q. et al. Genome sequence of a 45,000-year-old modern human from western Siberia. Nature 514, 445-449 (2014). Fu, Q. et al. The genetic history of Ice Age Europe. Nature 534, 200-205 (2016). Gallego Llorente, M. et al. Ancient Ethiopian genome reveals extensive Eurasian admixture in Eastern Africa. Science 350, 820-822 (2015). Jones, E. R. et al. Upper Palaeolithic genomes reveal deep roots of modern Eurasians. Nature communications 6, 8912 (2015). Keller, A. et al. New insights into the Tyrolean Iceman's origin and phenotype as inferred by whole-genome sequencing. Nature communications 3, 698 (2012). Kilinc, G. M. et al. The Demographic Development of the First Farmers in Anatolia. Curr. Biol. (2016). Lazaridis, I. et al. Genomic insights into the origin of farming in the ancient Near East. Nature 536, 419-424 (2016). Lazaridis, I. et al. Ancient human genomes suggest three ancestral populations for present-day Europeans. Nature 513, 409-413 (2014). Olalde, I. et al. Derived immune and ancestral pigmentation alleles in a 7,000-yearold Mesolithic European. Nature 507, 225-228 (2014). Raghavan, M. et al. Upper Palaeolithic Siberian genome reveals dual ancestry of Native Americans. Nature 505, 87-91 (2014). 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 Lazaridis, I. et al. Genetic origins of the Minoans and Mycenaeans. Nature 548, 214218 (2017). Lipson, M. et al. Parallel ancient genomic transects reveal complex population history of early European farmers. Nature 551, 368-372 (2017). Mallick, S. et al. The Simons Genome Diversity Project: 300 genomes from 142 diverse populations. Nature 538, 201-206 (2016). Alexander, D. H., Novembre, J. & Lange, K. Fast model-based estimation of ancestry in unrelated individuals. Genome Res. 19, 1655-1664 (2009). Patterson, N. et al. Ancient admixture in human history. Genetics 192, 1065-1093 (2012). Gronenborn, D. & Dolukhanov, P. in The Oxford Handbook of Neolithic Europe (eds C. Fowler, J. Harding, & D. Hofmann) 195-214 (Oxford University Press, 2015). Telegin, D. Ya. Neolithic cultures of Ukraine and their chronology. Journal of World Prehistory I, 307-331 (1987). Telegin, D. Ya. & Potekhina, I. D. Neolithic cemeteries and populations in the Dnieper Basin. (British Archaeological Reports, 1987). Jones, E. R. et al. The Neolithic Transition in the Baltic Was Not Driven by Admixture with Early European Farmers. Curr. Biol., 2185–2193 (2017). Mittnik, A. et al. The Genetic History of Northern Europe. bioRxiv, https://doi.org/10.1101/113241 (2017). Saag, L. et al. Extensive farming in Estonia started through a sex-biased migration from the Steppe. Curr. Biol. 27, 2185-2193 (2017). Maier, A. The Central European Magdalenian: Regional Diversity and Internal Variability., (Springer, 2015). Borić, D. & Price, T. D. Strontium isotopes document greater human mobility at the start of the Balkan Neolithic. Proc. Natl. Acad. Sci. U. S. A. 110, 3298-3303 (2013). Krauß, R., Marinova, E., De Brue, H. & Weninger, B. The rapid spread of early farming from the Aegean into the Balkans via the Sub-Mediterranean-Aegean Vegetation Zone. Quaternary International XXX, 1-18 (2017). Bacvarov, K. in Moments in time: Papers Presented to Pál Raczky on His 60th Birthday (eds A. Anders & G Kulcsár) 29-34 (L’Harmattan, 2013). Gurova, M. & Bonsall, C. ‘Pre-Neolithic’ in Southeast Europe: a Bulgarian perspective. Documenta Praehistorica XLI, 95-109 (2014). Brandt, G. et al. Ancient DNA reveals key stages in the formation of central European mitochondrial genetic diversity. Science 342, 257-261 (2013). Borić, D. in The Oxford Handbook of Neolithic Europe (eds C. Fowler, J. Harding, & D. Hofmann) 927–957 (Oxford University Press, 2015). Szmyt, M. in Transition to the Bronze Age (Archaeolingua 30) (eds V. Heyd, G. Kulcsár, & V. Szeverényi) 93-111 (Archaeolingua, 2013). Olalde, I. et al. A Common Genetic Origin for Early Farmers from Mediterranean Cardial and Central European LBK Cultures. Mol. Biol. Evol. 32, 3132-3142 (2015). Posth, C. et al. Pleistocene Mitochondrial Genomes Suggest a Single Major Dispersal of Non-Africans and a Late Glacial Population Turnover in Europe. Curr. Biol. 26, 827-833 (2016). Anthony, D. W. & Ringe, D. The Indo-European Homeland from Linguistic and Archaeological Perspectives. Annual Review of Linguistics 1, 199-219 (2015). Additional Methods References 48 Dabney, J. et al. Complete mitochondrial genome sequence of a Middle Pleistocene cave bear reconstructed from ultrashort DNA fragments. Proc. Natl. Acad. Sci. U. S. A. 110, 15758-15763 (2013). 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 Meyer, M. & Kircher, M. Illumina sequencing library preparation for highly multiplexed target capture and sequencing. Cold Spring Harb Protoc 2010, pdb prot5448 (2010). Rohland, N., Harney, E., Mallick, S., Nordenfelt, S. & Reich, D. Partial uracil-DNAglycosylase treatment for screening of ancient DNA. Philos. Trans. R. Soc. Lond. B Biol. Sci. 370, 20130624 (2015). Korlevic, P. et al. Reducing microbial and human contamination in DNA extractions from ancient bones and teeth. BioTechniques 59, 87-93 (2015). Briggs, A. W. et al. Removal of deaminated cytosines and detection of in vivo methylation in ancient DNA. Nucleic Acids Res. 38, e87 (2010). DeAngelis, M. M., Wang, D. G. & Hawkins, T. L. Solid-phase reversible immobilization for the isolation of PCR products. Nucleic Acids Res. 23, 4742-4743 (1995). Rohland, N. & Reich, D. Cost-effective, high-throughput DNA sequencing libraries for multiplexed target capture. Genome Res. 22, 939-946 (2012). Maricic, T., Whitten, M. & Pääbo, S. Multiplexed DNA sequence capture of mitochondrial genomes using PCR products. PLoS One 5, e14004 (2010). Kircher, M., Sawyer, S. & Meyer, M. Double indexing overcomes inaccuracies in multiplex sequencing on the Illumina platform. Nucleic Acids Res. 40, e3 (2012). Li, H. & Durbin, R. Fast and accurate long-read alignment with Burrows-Wheeler transform. Bioinformatics 26, 589-595 (2010). Fu, Q. et al. A revised timescale for human evolution based on ancient mitochondrial genomes. Curr. Biol. 23, 553-559 (2013). Skoglund, P. et al. Separating endogenous ancient DNA from modern day contamination in a Siberian Neandertal. Proc. Natl. Acad. Sci. U. S. A. 111, 22292234 (2014). Korneliussen, T. S., Albrechtsen, A. & Nielsen, R. ANGSD: Analysis of Next Generation Sequencing Data. BMC Bioinformatics 15, 356 (2014). Price, A. L. et al. Principal components analysis corrects for stratification in genomewide association studies. Nat. Genet. 38, 904-909 (2006). Petkova, D., Novembre, J. & Stephens, M. Visualizing spatial population structure with estimated effective migration surfaces. Nat. Genet. 48, 94-100 (2016). Brown, L., Cai, T. & DasGupta, A. Interval Estimation for a Binomial Proportion. Statistical Science 16, 101-133 (2001). Agresti, A. & Coull, B. Approximate is better than "exact" for interval estimation of binomial proportions. The American Statistician 52, 119-126 (1998). Longin, R. New method of collagen extraction for radiocarbon analysis. Nature 230, 241-242 (1971). Brown, T. A., Nelson, D. E., Vogel, J. S. & Southon, J. R. Improved collagen extraction by modified longin method. Radiocarbon 30, 171-177 (1988). Kennett, D. J. et al. Archaeogenomic evidence reveals prehistoric matrilineal dynasty. Nature communications 8, 14115 (2017). van Klinken, G. J. Bone collagen quality indicators for paleodietary and radiocarbon measurements. JAS 26, 687-695 (1999). Stuiver, M. & Polach, H. A. Discussion: Reporting of 14C data. Radiocarbon 19, 355363 (1977). Bronk Ramsey, C. OxCal 4.23 Online Manual, https://c14.arch.ox.ac.uk/oxcalhelp/hlp_contents.html (2013). Cook, G. T. et al. A freshwater diet-derived 14C reservoir effect at the Stone Age sites in the Iron Gates gorge. Radiocarbon 43, 453-460 (2001).