Between fishing and farming. The Neolithic in the Danube Gorges. Preprint.

[real expert]

​Between fishing and farming: palaeogenomic analyses reveal cross-cultural interactions triggered by the arrival of the Neolithic in the Danube Gorges

Abstract

While early Neolithic populations in Europe were largely descended from early Aegean farmers, there is also evidence of episodic gene flow from local Mesolithic hunter-gatherers into early Neolithic communities. Exactly how and where this occurred is still unknown. Here we report direct evidence for admixture between the two groups at the Danube Gorges in Serbia. Analysis of palaeogenomes recovered from skeletons revealed that second-generation mixed individuals were buried amidst individuals whose ancestry was either exclusively Aegean Neolithic or exclusively local Mesolithic. The mixed ancestry is also reflected in a corresponding mosaic of grave goods. With its deep sequence of occupation and its unique dwellings that suggest at least semi-sedentary occupation since the late Mesolithic, the area of the Danube Gorges has been at the center of the debate about the contribution of Mesolithic societies to the Neolithisation of Europe. As suggested by our data, which were processed exclusively with uncertainty-aware bioinformatic tools, it may have been precisely in such contexts that close interactions between these societies were established, and Mesolithic ancestry and cultural elements were assimilated.


https://www.biorxiv.org/content/10.1101/2022.06.24.497512v1.article-info



https://www.biorxiv.org/content/10.1101/2022.06.24.497512v1.full.pdf+html

 

[real expert]

To ensure comparability and increase the sensitivity of population genomic analyses, we focused on 54 whole genomes and 54 5Mb neutralomes (with seven overlaps, Table 1, S1) that passed rigorous quality assessments mainly evaluating the reproducibility of genetic diversity estimates in face of data bootstrapping (see Methods, Figure S1-S2). Further, we based ou ranalyses on uncertainty-aware inference methods that use genotype likelihoods, which werei nferred from raw sequence data (unless not publicly available) using a bioinformatic pipeline dedicated to ancient DNA (see Methods, Supplementary Data Table 1; (Link et al., 2017)). Finally, we excluded an individual from Dillingen, Germany (Dil15), which we identified as a brother of Dil16, as well as one from Barcın, Turkey (Bar15), which we identified to have aparent-child relationship with Bar8 (Figure S2D).Early-generation admixed individuals at the Danube Gorges Two Danube Gorges individuals, LEPE18 (LV 27d, 6,126 ± 100 cal BC) and LEPE46 (LV 93,6,120 ± 102 cal BC), display substantial ancestry from both clusters. To shed more light on their admixture status, we used a Bayesian approach (Shastry et al., 2021) to infer genome-wide ancestry proportions (q1) jointly with inter-population ancestry proportions (Q12), i.e. the fraction of the genome at which a sample is heterozygous for the different ancestries (Fig 2D). These estimates indicate that both samples were second-generation admixed individuals: both had amixed first-generation parent, while the other parent was unmixed of either Meso European-like(LEPE18) or Neo Aegean-like ancestry (LEPE46). Thus, LEPE46 had one Meso European-like and three Neo Aegean-like grandparents, while LEPE18 had one Neo Aegean-like and three Meso European-like grandparents.We identified three additional individuals with considerable ancestry from both clusters: Bla32,Bla59 and Bla45 from the Blätterhöhle cave in Westfalia, Germany (Fig. 2A) dating to 4th millennium cal BC (3900-3000 cal BC). In contrast to LEPE18 and LEPE46, they appear to be later generation admixed (Fig. 2D), in line with their more recent age (see Table 1) and the previously reported ongoing admixture at this site (Bollongino et al., 2013; Lipson et al., 2017) and in the wider region (Haak et al., 2015).

High genetic diversity in Danube Gorges foragers

To characterize the forager population present at the Danube Gorges during the Mesolithic and the Transformation phase, we first focused on individuals with <4% Neo Aegean-like ancestry in available under aCC-BY-NC-ND 4.0 International license.(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made bioRxiv preprint doi: https://doi.org/10.1101/2022.06.24.497512; this version posted June 28, 2022. The copyright holder for this preprintthe admixture analysis. As attested by a projection-free PCA (Figure 2B), the Meso European-like individuals from the Danube Gorges are most similar to those from Western Europe, albeit slightly shifted towards individuals from North-Eastern Europe, in line with previous reports (Mathieson et al., 2018). In contrast to the Western European individuals, their cluster appears rather diverse (Figure 2B), which likely reflects a locally large population,elevated gene flow from neighboring populations, or both. This interpretation, consistent with theidea of a partially sedentary and prosperous fishing society of the Transformation period, is corroborated by Danube Gorges individuals generally having the highest genome-wide heterozygosity levels and shortest total lengths of runs of homozygosity (ROH) among all post-LGM Meso European-like individuals (Figure 3A, B, E), albeit some individual variation.Interestingly, the three Danube Gorges individuals from the Vlasac site that fall most distantly from the other Western European Meso European-like samples on the PCA (Fig 2A, B, FigureS3B, VLASA10, VLASA32, VLASA41) were among the only four buried with disarticulated skulls. A similarly diverse cluster is observed for the individuals of the site of Zvejnieki, with the two older samples ZVEJ25 (grave 93, 5,738 ± 102 cal BC) and ZVEJ 162 (grave ZV162, 4,470 ± 72cal BC) clustering with Meso European-like samples from Western Europe, while the younge stsample ZVEJ317 (grave ZV317, 3,890 ± 67 cal BC) does not (Figure 2B). The influx of a rather distinct ancestry into Zvejnieki during the Neolithic has been previously reported (Jones et al.,2017). As our data shows, the source of the influx was genetically very close to samples from Minino, which lies around 1,600 km to the east. Since the Baltic Sea region was covered with ice until at least the late 10th millennium cal BC, it is reasonable to assume that the two distinct ancestors discovered at Zvejnieki ultimately came from two different glacial refugia of the late Ice Age, possibly one in southern France and one on the Black Sea coast (Mittnik et al., 2018). Despite this diverse origins, we estimate low levels of heterozygosity for all Zvejnieki and Minino samples but no evidence of recent inbreeding for ZVEJ25 (the only sample with sufficient quality whole genome data).

Noteworthy, we estimated the lowest heterozygosity among all neutralomes for the late hunter-gatherer from the site Criewen (GR2, (Terberger et al., 2018)), a female from Northern-Eastern Germany, with a date around 4,500 cal BC (dated to 4,600 ± 60 cal BC but likely with reservoir effect). It represents the most recent Central European individual with essentially 100% Meso European-like ancestry analyzed here.

Pdf;Page 5.

 

[Angela]

Very interesting. Did they mention why then LBK had so little Euro Mesolithic ancestry? I also remember a paper saying this group did not contribute to the Neolithic settlements. Did the paper address that?

 

[real expert]

Very interesting. Did they mention why then LBK had so little Euro Mesolithic ancestry? I also remember a paper saying this group did not contribute to the Neolithic settlements. Did the paper address that?

The authors addressed this issue in their conclusion.

At the genetic level, the interaction of early Aegean farmers with European hunter-gatherer groups along the expansion route has been demonstrated mostly indirectly: while most studies agree that about 2-6% of the genome of early Neolithic European people derives from admixture with hunter-gatherers during the Early Neolithic period, direct genetic evidence for hunter-gatherers in an early Neolithic context is limited to a single individual reported from the Körös site Tiszaszolos-Domaháza in Hungary (Gamba et al., 2014), an agricultural settlement at the frontier of the Neolithic Expansion that persisted for a few generations only. Considering the large number of individuals studied from Early Neolithic sites so far, the scarcity of individuals with predominantly Meso European-like ancestry and the complete absence of early-generation admixed individuals is remarkable. In this study, for instance, we newly analyzed 21 individuals with genomic data from typical Early Neolithic sites across Europe (Herxheim, Kleinhadersdorf, Dillingen-Steinheim, Asparn-Schlelz), but have not found a single individual that shows substantial Meso European-like ancestry. Cultural practices such as differentiated burial rites may be responsible for this. Another, equally plausible explanation would be that intermarriage was not tolerated at typical Neolithic core sites itself, but perhaps only in the periphery. So is Lepenski Vir a model for an experimental outpost on the Neolithic expansion front? Could intercultural practices have been tried out here that Neolithic societies, with their “colonist ethos”available under aCC-BY-NC-ND 4.0 International license.(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made bioRxiv preprint doi: https://doi.org/10.1101/2022.06.24.497512; this version posted June 28, 2022. The copyright holder for this preprint and entrenched narrow cultural practices, did not tolerate on their own land (Lüning, 2000;Özdoğan, 2011; Shennan, 2018)? Even if agriculture was not possible in the Danube Gorges, a connection to agricultural communities must have existed, as the temporal distribution of nitrogen isotope ratios shows. Whether this connection was also accompanied by gene flow from Lepenski Vir to the Neolithic communities has not been shown yet, but seems likely. Thus,sites like Lepenski Vir could well have been extramural contact zones between hunter-gatherers and early farmers, and thus responsible for the introgression of hunter-gatherer ancestry into Neolithic communities. This would not completely invalidate the alternative "dead-end theory"according to which Lepenski Vir was merely a failed early Neolithic experiment with a modified way of life, for both may be true.

 

[Angela]

I can buy that as a possibility, certainly, although it's also still possible it was a "dead end" in terms of European admixture.

When the Dutch settled the Cape, men far outnumbered women. Some of them took "native" spouses. Through constant intermixing with "pure" Dutch people, the "outsider" ancestry was forgotten or deliberately swept under the rug. Once large numbers of Dutch women arrived, the offspring of whatever alliances still took place went into the "native" or "colored" group. The proof is still there in the genomes, however, with very large numbers of Dutch looking Afrikaners having up to 9% non-white ancestry.

Much the same thing happened in Cuba, where many "white" Cubans were shocked to find they had similar percentages of non-white ancestry.

After 1000 years or more, something happened and there was more admixture. From about 6% to about an average of 25%, and EEF like men can be found carrying I2a. I'm still not completely clear about what happened.

 

[bicicleur 2]

Lepenski Vir is a unique site. I doubt that whatever happened there was representative for the rest of Europe.

 

[MOESAN]

Anthropology found the same results of partial intermixing there, if I recall well.
Maybe few "border-line" farmers at the fringes, with more contacts with "native" HG's of Balkans, and without too much succes later?

 

[I1a3_Young]

I can buy that as a possibility, certainly, although it's also still possible it was a "dead end" in terms of European admixture.

When the Dutch settled the Cape, men far outnumbered women. Some of them took "native" spouses. Through constant intermixing with "pure" Dutch people, the "outsider" ancestry was forgotten or deliberately swept under the rug. Once large numbers of Dutch women arrived, the offspring of whatever alliances still took place went into the "native" or "colored" group. The proof is still there in the genomes, however, with very large numbers of Dutch looking Afrikaners having up to 9% non-white ancestry.

Much the same thing happened in Cuba, where many "white" Cubans were shocked to find they had similar percentages of non-white ancestry.

After 1000 years or more, something happened and there was more admixture. From about 6% to about an average of 25%, and EEF like men can be found carrying I2a. I'm still not completely clear about what happened.

It seems plausible to me that there was a multiple wave scenario where subsequent waves of neo farmers expanded, bringing the meso down. Central Europe seems to have always been in a population flux due to it's location being the crossroads of people migrations. The subsequent changes in percentages could reflect the ebb and flow of the various powers over hundreds or thousands of years. It's rarely as simple as "pop X" immediately took over and then was static for 1,000 years. I know Angela realizes this, I'm reiterating this concept for the general reader.