Early contact between late farming and pastoralist societies in southeastern Europe

Abstract

Archaeogenetic studies have described two main genetic turnover events in prehistoric western Eurasia: one associated with the spread of farming and a sedentary lifestyle starting around 7000-6000 BC (refs. ^1-3) and a second with the expansion of pastoralist groups from the Eurasian steppes starting around 3300 BC (refs. ^4,5). The period between these events saw new economies emerging on the basis of key innovations, including metallurgy, wheel and wagon and horse domestication^6-9. However, what happened between the demise of the Copper Age settlements around 4250 BC and the expansion of pastoralists remains poorly understood. To address this question, we analysed genome-wide data from 135 ancient individuals from the contact zone between southeastern Europe and the northwestern Black Sea region spanning this critical time period. While we observe genetic continuity between Neolithic and Copper Age groups from major sites in the same region, from around 4500 BC on, groups from the northwestern Black Sea region carried varying amounts of mixed ancestries derived from Copper Age groups and those from the forest/steppe zones, indicating genetic and cultural contact over a period of around 1,000 years earlier than anticipated. We propose that the transfer of critical innovations between farmers and transitional foragers/herders from different ecogeographic zones during this early contact was integral to the formation, rise and expansion of pastoralist groups around 3300 BC.

Fig. 1. Geographical locations, genetic analyses and chronology of newly reported ancient Copper Age, Eneolithic and Early Bronze Age individuals.

a, Map of sites and relevant archaeological cultures discussed in the text. Maps were made with Natural Earth (https://naturalearthdata.com). b, PCA of newly reported individuals (coloured symbols with black outline) and relevant published groups (coloured symbols, no outline) projected onto the West Eurasian genetic variation of 1,253 individuals from 77 populations. c, Mean radiocarbon dates of relevant published and newly reported individuals from southeastern Europe plotted according to the regional chronology. The suffixes in the group labels present archaeological time periods and geographical regions: N, Neolithic; EN, MN, LN, Early, Middle, Late Neolithic, respectively; CA, Copper Age; EBA, Early Bronze Age; SEE, southeastern Europe; WHG, EHG, WSHG, CHG, oHG, Western, Eastern, West Siberian, Caucasus, outlier Hunter–Gatherers, respectively.

Fig. 2. Outgroup f₃ statistics for SEE CA, Ukraine Eneolithic and Early Bronze Age groups.

Outgroup f₃ statistics of the form f₃(test1, test2; Mbuti) plotted onto maps of central/SEE highlighting the shifting genetic affinities over time, separated temporally. Test1 includes groups and individuals newly reported in this study (headers) and their respective location is given by a grey circle. Test2 includes all relevant ancient populations from the respective time period (Supplementary Table C) and their locations are indicated as filled circles with black outlines. Higher f₃ statistics (red colours) indicate more shared drift with the respective group in Test1. All f₃ statistics, including outlier individuals and published ancient populations further west, can be found in Supplementary Table C. The maps were made in R and the tile sets are copyright of Stamen Design, under a Creative Commons Attribution (CC BY 3.0) licence.

Fig. 3. Distal and proximal qpAdm results for the Copper Age, Ukraine Eneolithic and Early Bronze Age groups.

a–c, Distal models with Turkey_N, WHG, EHG and CHG as sources for the three sampled time periods: SEE CA (a); Ukraine Eneolithic (b); and Early Bronze Age (c). d–f, Geographically and temporally proximal models of the three sampled time periods: SEE CA (d); Ukraine Eneolithic (e); and Early Bronze Age (f) (Supplementary Tables G, H, P, Q, V and W). All results shown here were run with the parameter ‘allSNPs: NO’ (Supplementary Information 6). * Indicates non-supported/rejected/failed models when applying a P value cut-off of less than 0.05 (shown in italics). Ancestry proportions are shown with one standard error. Standard errors were computed with the default block jackknife approach.

Extended Data Fig. 1. Tell settlements and burial sites in Southeastern Europe.

a, Aerial view of Tell Măgura Gorgana near Pietrele, Romania (© Konstantin Scheele, German Archaeological Institute, Eurasia Department). b, Detailed view of the 11m stratigraphy at Pietrele (© Svend Hansen, German Archaeological Institute, Eurasia Department). c, Aerial view of Tell Yunatsite, Bulgaria (© Kamen Boyadzhiev). d, Map of the site Orlovka-Kartal, Ukraine. The base map was sourced from Google Earth https://www.google.com/earth/index.html. e, Characteristic finds from the Eneolithic type-site Usatove. f, Characteristic finds attributed to the Cernavoda I phase. g, Burial in flexed position from grave 10 at Kartal (© Igor Bruyako). h, Infant urn-burial from the Early Bronze Age layer south of tell Yunatsite (© Kamen Boyadzhiev).

Extended Data Fig. 2. F₄-statistics for CA groups to determine Neolithic proxies.

F₄-statistics show different attraction of the CA to Neolithic groups conditioned on HG groups. Z-scores outside the threshold of (|Z|≥1) are highlighted in orange, f₄-values are shown with one standard error. Test populations are given on the y-axis. Standard errors (SE) were computed with the default block jackknife approach.

Extended Data Fig. 3. DATES and Y-chromosomal haplogroup diversity.

a, DATES decay curve for the CA outlier individual PIE060 with SEE N and Iron Gates HGs as source populations. b, Changes of Y-chromosomal haplogroup diversity (colour fills) over time (mean ¹⁴C dates cal. BC; x-axis) with respect to changes in autosomal ancestry as reflected in PC2 (y-axis), based on the relative density of female (open circles) and male (colour filled squares) of all newly reported individuals in this study.

Extended Data Fig. 4. Identity-by-descent within and between sites.

a, Results of identity-by-descent (IBD) analysis per pair of individuals. Plotting the sum versus the number of the shared chunks of IBD in window sizes of >12cM resolves degrees of biological relatedness up to the 4-6^th degree. Within (left) and between site (right) relationships are highlighted separately. b, Stacked bar plot showing the number (left) and the cumulative distribution of the sum (right) of IBD blocks that are shared between all individuals within and between sites.

Extended Data Fig. 5. Cumulative distribution of the runs of homozygosity tracts of all newly reported individuals.

Runs of homozygosity were estimated with hapROH. Individuals are grouped in relative chronological order from right to left. Expected parental relationship and simulated effective populations sizes are given.

Extended Data Fig. 6. F₄ statistics for Ukraine Eneolithic groups to determine attraction to ‘cornerstone’ populations.

F₄ statistics show different attractions of Ukraine Eneolithic groups to ‘cornerstone’ ancestry groups conditioned on a, Maykop-associated groups, b, Steppe Eneolithic, and c, SEE CA. Significant Z-scores (|Z|≥3) are highlighted in orange, f₄ values are shown with three standard errors. Test populations are given on the y-axis. Standard errors (SE) were computed with the default block jackknife approach.

Extended Data Fig. 7. F₄ statistics for EBA groups to test for excess HG attraction.

F₄ statistics show different attractions of EBA groups to HG groups conditioned on their respective preceding or contemporaneous group. Significant Z-scores (|Z|≥3) are highlighted in orange, f₄ values are shown with three standard errors. Test populations are given on the y-axis. Standard errors (SE) were computed with the default block jackknife approach.

Extended Data Fig. 8. Testing for Anatolian farmer-related affinity in EBA individuals.

F₄ statistics show different attractions of the EBA groups to VAR_CA conditioned on Yamnaya-associated groups. Significant Z-scores (|Z|≥3) are highlighted in orange, f₄ values are shown with three standard errors. Test populations are given on the y-axis. Standard errors (SE) were computed with the default block jackknife approach.

Extended Data Fig. 9. F₄ statistics to explore excess affinity of EBA ‘steppe ancestry’ groups conditioned on preceding Eneolithic groups from the steppe and the Caucasus.

F₄ statistics show different attractions of the EBA groups to ‘cornerstone’ populations conditioned on pre-Yamnaya groups, Steppe Eneolithic and Caucasus Eneolithic/Maykop. Significant Z-scores (|Z|≥3) are highlighted in orange, f₄ values are shown with three standard errors. EBA groups are given on the y-axis and test populations are given on the x-axis. Standard errors (SE) were computed with the default block jackknife approach.