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.

Full text

Between fishing and farming: palaeogenomic analyses reveal cross-cultural
interactions triggered by the arrival of the Neolithic in the Danube Gorges
Zuzana Hofmanová1,2,3,4,5,28, Carlos S. Reyna-Blanco1,2,28, Camille de Becdelièvre6, Ilektra Schulz1,2,5, Jens Blöcher5,
Jelena Jovanović7, Laura Winkelbach5, Sylwia M. Figarska5, Anna Schulz8, Marko Porčić7,9, Petr Květina10,
Alexandros Tsoupas11, Mathias Currat11,12, Alexandra Buzhilova13,14, Fokke Gerritsen15, Necmi Karul16, George
McGlynn17, Jörg Orschiedt18,19, Rana Özbal20, Joris Peters17,21, Bogdan Ridush22, Thomas Terberger23, Maria
Teschler-Nicola24,25, Gunita Zariņa26, Andrea Zeeb-Lanz27,Sofija Stefanović9, Joachim Burger5,29*, Daniel
Wegmann1,2,29*
Affiliations:
1Department of Biology, University of Fribourg; 1700 Fribourg, Switzerland.
2Swiss Institute of Bioinformatics; 1015 Lausanne, Switzerland.
3Max Planck Institute for Evolutionary Anthropology; 04103 Leipzig, Germany.
4Department of Archaeology and Museology, Faculty of Arts, Masaryk University; 60177 Brno, Czech Republic.
5Palaeogenetics Group, Institute of Organismic and Molecular Evolution (iomE), Johannes Gutenberg University
Mainz; 55099 Mainz, Germany.
6Aix-Marseille University, CNRS, EFS, ADES, Marseille, France.
7BioSense Institute, University of Novi Sad, Bulevar Zorana Đinđića 1; 21000 Novi Sad, Serbia.
8Centre for the Study of Manuscript Cultures, Cluster of Excellence Understanding Written Artefacts, Hamburg
University; 20354 Hamburg, Germany.
9Laboratory for Bioarchaeology, Department of Archaeology, Faculty of Philosophy, University of Belgrade; Čika
Ljubina 18-20, 11000 Belgrade, Serbia.
10Institute of Archaeology of the Czech Academy of Sciences; Prague, Czechia.
11Laboratory of Anthropology, Genetics and Peopling History, Department of Genetics and Evolution - Anthropology
Unit, University of Geneva; Quai Ernest-Ansermet 30, 1205 Geneva, Switzerland
12Institute of Genetics and Genomics in Geneva (IGE3), University of Geneva Medical Centre (CMU); 1 rue
Michel-Servet, 1211 Geneva, Switzerland
13Research Institute and Museum of Anthropology at Moscow State University; Mokhovaya Street 11, 125009
Moscow, Russia.
14Institute of Archaeology and Ethnography; Prospekt Lavrentieva 17, 630090 Novosibirsk, Russia.
15Netherlands Institute in Turkey; İstiklal Caddesi 181, Merkez Han, 34433 Beyoğlu/İstanbul, Turkey.
16Department of Prehistory, İstanbul University; 34434 Istanbul, Turkey
17SNSB, State Collection for Anthropology and Palaeoanatomy; Karolinenplatz 2a, 80333 Munich, Germany.
18Institut für Prähistorische Archäologie, Freie Universität Berlin, Fabeckstr. 23-25; 14195 Berlin, Germany.
19Landesamt für Denkmalpflege und Archäologie Sachsen-Anhalt, Richard-Wagner-Straße 9; 06114 Halle (Saale),
Germany.
20Department of Archaeology and History of Art, Koc¸ University; 34450 Istanbul, Turkey.
21Institute of Palaeoanatomy, Domestication Research and the History of Veterinary Medicine, LMU Munich; 80539
Munich, Germany.
22Department of Physical Geography, Geomorphology and Paleogeography, Yuriy Fedkovych Chernivtsi National
University; Kotsubynskogo 2, 58012 Chernivtsi, Ukraine.
23Seminar for Pre- and Protohistory, University of Göttingen, Nikolausberger Weg 15, 37073 Göttingen, Germany.
24Department of Anthropology, Natural History Museum of Vienna; 1010 Vienna, Austria.
25Department of Evolutionary Anthropology, University of Vienna; 1030 Vienna, Austria.
26Institute of History, University of Latvia, Riga, Latvia.
27Generaldirektion Kulturelles Erbe Rheinland-Pfalz; 67346 Speyer, Germany.
28These authors contributed equally.
29Senior authors.
*Corresponding authors: daniel.wegmann@unifr.ch;jburger@uni-mainz.de ;zuzana_hofmanova@eva.mpg.de
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Summary
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.
Introduction
The Danube Gorges area, or Đerdap in Serbian, is situated in the Central Balkans and an early
contact zone between migrating early European farmers and local foragers (Oross and Bánffy,
2009). With more than 25 prehistoric sites (Fig. 1), numerous Mesolithic and Neolithic human
burial finds and a deep Final Paleolithic to Neolithic sequence of occupation it has contributed
significantly to our understanding of the introduction of agriculture to the European continent
and Mesolithic-Neolithic interactions (Porčić, Blagojević and Stefanović, 2016; de Becdelièvre
et al., 2020).
After a Final Paleolithic occupation in some rock-shelter sites (ca. 13,000-9,500 cal BC), the
human presence at the Danube Gorges is documented on open-air sites in river terraces at
several locations, such as Lepenski Vir and Padina during the Early Mesolithic period (ca.
9,500-7,400/7,300 cal BC; (Boroneanţ, 1999; Borić, 2011)). The population then likely became
more numerous during the Late Mesolithic period (ca. 7400/7300-6300/6200 cal BC), and some
finds indicate that the Mesolithic population lived partly semi-sedentary lives and their mobility
was greatly reduced compared to earlier periods (Dimitrijević, Živaljević and Stefanović, 2016;
de Becdelièvre et al., 2021). This semi-sedentary lifestyle was undoubtedly associated with the
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

consumption of aquatic food: the community relied heavily on fishing and benefited especially
from the large numbers of anadromous species that populated the narrow gorge during their
upward migration (Bonsall et al., 1997; Bartosiewicz et al., 2001; Nehlich et al., 2010; Borić et
al., 2014; Jovanović et al., 2019). The diet was further enriched by game, different plant species,
and possibly dogs (Cristiani et al., 2016; Jovanović et al., 2019, 2021).
At the end of the 7th millennium BC, the first settlements with farming subsistence appeared in
the Central Balkans and the southern part of the Pannonian plain: the Neolithic
Starčevo-Körös-Criş cultural complex (Garašanin, 1982; Tringham, 2000; Krauß, 2011), which
later developed important sites, such as Starčevo and the early layers of Vinča tell, in the
immediate vicinity of the Danube Gorges (Porčić, Blagojević and Stefanović, 2016; Jovanović et
al., 2019; de Becdelièvre et al., 2020; Porčić et al., 2020, 2021). Around this time, trapezoidal
houses were built at the Lepenski Vir site. Such houses, while similar to Late Mesolithic
dwellings at the nearby site Vlasac, have a shape not known outside the Danube Gorges (Boric,
French and Dimitrijević, 2008), but might have some structural analogies, such as plastered
floors, in Neolithic Anatolia (Srejović, 1969; Borić, 2011; Borić, Radović and Stefanović, 2012).
Mesolithic foragers and newcomers with a Neolithic background used the settlement site during
this period simultaneously (Borić and Price, 2013; Hofmanová, 2016; González-Fortes et al.,
2017; Mathieson et al., 2018) and, as indicated by isotopic data in skeletons from the time of
their arrival, newcomers had a more terrestrial diet (Borić and Price, 2013). This period of
interaction between foragers and farmers at the site is referred to as the "Transformation Phase"
and dates ca 6200-5950 cal BC (Borić et al., 2018). Besides the trapezoidal houses, other
elements of the Neolithic way of life, such as typical ornaments, raw materials and ceramics,
now appear (Garašanin and Radovanović, 2001). In addition, so-called ancestor statues with
fish-like features have been found in the corresponding layers of Lepenski Vir. Over time,
Neolithic features increased at the site, leading to the fully developed Neolithic period
(5,950-5,550 cal BC), in which the trapezoidal houses were abandoned (citation), the Neolithic
suite of domesticated animals appeared (Boric and Dimitrijevic, 2005; Borić and Dimitrijević,
2007) and grains of cereals have been evidenced in the dental calculus of some Neolithic
individuals (Jovanović et al., 2021). However, the use of wild resources remained dominant
(Boric and Dimitrijevic, 2005; Borić and Dimitrijević, 2007; Cramp et al., 2019; Jovanović et al.,
2019).
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

This wealth of evidence makes Lepenski Vir one of the best-studied sites of the
Mesolithic-Neolithic transition and provides a unique opportunity to explore foragers and farmers
interactions. However, key aspects of the site’s population history remain unclear, including how
the mixing between the Danubian and Aegean Neolithic members of the society took place, i.e.
when and with what proportions of the two populations it took place. Furthermore, it is not
known how these findings from the Danubian Gorges relate to other forager or farming
populations in SW Asia and Europe.
Using uncertainty-aware bioinformatic tools on high quality genomic data from 51 available and
50 newly produced samples with a geographical and chronological focus on the Meso-Neolithic
transition in the Danube Gorges and key sites from the Central Balkans, Lower Austria,
Southern Germany, the Marmara region and the Baltics, we show that the two contributing
populations were genetically well differentiated but episodically interacted, as evidenced not
least by the presence of first generation mixed individuals.
Results
In order to better understand the reciprocal relationship between the populations originating
from the Neolithic area in the wider Aegean including the Sea of Marmara and the indigenous
foragers in Serbia, we produced (Table 1): 1) six whole genomes at a sequencing depth of 1-5X
and ten neutralomes of 5Mb length (see Methods) from the Danube Gorges, 2) ten Neolithic
neutralomes from the Aegean/Marmara region 3) 37 neutralomes from Neolithic Central Europe,
4) five neutralomes from Mesolithic Central Europe, 6) 12 neutralomes from hunter-gatherers of
North-Eastern Europe, 7) one neutralome from Lesnik Cave presumably dating to the Final
Paleolithic, and 8) 78 mtDNA genomes from the same regions (Supplementary Data Table 1).
We complemented these with 53 chronologically similar whole genomes available from Europe,
NW and Central Anatolia and the Caucasus ((Gamba et al., 2014; Lazaridis et al., 2014; Olalde
et al., 2014, 2015; Skoglund et al., 2014; Jones et al., 2015, 2017; Broushaki et al., 2016;
Hofmanová et al., 2016; Kılınç et al., 2016; González-Fortes et al., 2017; Sikora et al., 2017;
Günther et al., 2018; Marchi et al., 2022), Table S1), including four from the Danube Gorges
((Marchi et al., 2022), Table 1), and 20 modern samples from Africa and West Eurasia retrieved
from the SGDP database ((Mallick et al., 2016), Table S1).
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

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 our
analyses on uncertainty-aware inference methods that use genotype likelihoods, which were
inferred 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 a
parent-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 a
mixed 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
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

the 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 the
idea 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, Figure
S3B, 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 ZVEJ162 (grave ZV162, 4,470 ± 72
cal BC) clustering with Meso European-like samples from Western Europe, while the youngest
sample 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.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Neolithic ancestry in Lepenski Vir resembles that from Early Neolithic Northern-Greece
To characterize the Neolithic population appearing at the Danube Gorges around 6,200 cal BC,
we next focused on all samples with >96% Neo Aegean-like ancestry in the Admixture analysis.
On a projection-free PCA (Figure 2C, Figure S3A), the first axis separates Neo Aegean-like
samples of the Marmara region (Aktopraklık and Barcın sites) from those of Greece and Central
Europe, a distinction usually not seen on projected PCAs (e.g. (Mathieson et al., 2018)). The
three Neo Aegean-like individuals from Lepenski Vir as well as Greek individuals from the same
time (Nea2, Nea3) fall closest to those from Marmara, unlike the Greek individuals from later
periods (Pal7, Klei10). The same chronological signal is also seen among the Lepenski Vir
individuals, among which the two from the Transformation Period, LEPE48 (LV 122, 5,939.5 ±
72.5 cal BC) and LEPE39 (LV 82, 6,075 ± 125 cal BC) appear closer to Marmara samples than
LEPE52 (LV 73, 5,812 ± 119 cal BC), the sample from the later period of the fully developed
Neolithic. This could suggest extended gene flow with other Neolithic sites in the later phases of
Lepenski Vir, which would also explain the observed shift in the composition of the mtDNA gene
pool towards those sites over time (Supplementary Data Table 1). However, as far as the further
Neolithic expansion along the Balkan route from the Aegean to Central Europe is concerned,
the PCA does not show a clear spatiotemporal signal of differentiation. This also applies to the
individuals from Asparn Schletz and Herxheim, which are thought to derive from a massacre or
ritual background potentially involving individuals from a larger area (Boulestin et al., 2009;
Orschiedt and Haidle, 2012; Schulting and Fibiger, 2012; Boulestin, 2015). At the level of
genomic depth studied here, they do not appear particularly diverse, comparable to the
individuals from more regular Neolithic burial sites such as Kleinhadersdorf.
Heterozygosity is higher in farmers than in foragers and decreases along the Neolithic
route of expansion
With the exception of a single individual from the Herxheim site in SW-Germany (grave
281-19-6, 5,078 ± 85), the genome-wide heterozygosity estimated from neutralomes for all Neo
Aegean-like individuals was consistently higher than for post-LGM Meso European-like
individuals, both overall as well as when comparing samples from the Danube Gorges only (Fig
3A). In line with the idea of the Neolithic expansion along a Balkan route, we found generally the
highest diversity both in terms of differentiation on the PCA (Figure 2C) and heterozygosity
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

(Figure 3A,B) in the Marmara region (Aktopraklık and Barcın). While there is a general spatial
trend of decreasing heterozygosity from the Marmara region towards SW Germany, there
appears to be no temporal change in the Marmara region. Rather, the Chalcolithic sample AKT6
(grave H17/50.1; 5,584 ± 49 cal BC), which dates ~1000 years later, shows heterozygosity
comparable to that of the Early Neolithic skeletons from the same region.
Consistent with their admixed status, the genome-wide heterozygosity of LEPE18 and LEPE46
was estimated above that of most Meso European-like and Neo Aegean-like samples,
especially those from the Danube Gorges.
Heterozygosity estimates from whole genomes confirmed these general patterns, albeit the
smaller sample size. The heterozygosity of Neo Aegean-like samples appears to match that of
modern Europeans (Fig 3B), with the exception of Bar31, Klei10 and Pal7 that had slightly lower
levels similar to that of WC1 from Wezmeh Cave, an Early Neolithic individual from Iran with a
very different demographic history (Broushaki et al., 2016; Marchi et al., 2022). For Bon002, we
inferred particularly low levels of heterozygosity, in line with previous reports (Kılınç et al., 2016).
The lower heterozygosity of post-LGM foragers compared to Neolithic individuals was previously
seen at larger geographic scales (Fu et al., 2016; Kılınç et al., 2016; Posth et al., 2016;
Kousathanas et al., 2017; Renaud et al., 2019) and interpreted as a result of their long-term
demography, such as a severe bottleneck during the LGM and their generally low population
size (Gamble et al., 2004; Fernández-López de Pablo et al., 2019; Marchi et al., 2022). In
support of a strong LGM bottleneck, we estimated elevated genome-wide diversity for the Meso
European-like individuals from Sunghir predating the LGM (Sikora et al., 2017). In support of
low post-LGM population size, we found all Meso European-like individuals, including those
from the Danube Gorges, to have longer total length of runs of homozygosity (ROH) than
Neolithic samples (but not LEPE52, which was likely recently inbred) (Fig 3E). Despite their
elevated diversity, however, the three pre-LGM Sunghir individuals had the longest total ROH
and the highest number of very long ROH segments of all Meso European-like individuals, in
line with the interpretation of very small populations embedded in larger mating networks
(Sikora et al., 2017). Interestingly, we estimated elevated diversity comparable to that of the
Sunghir samples also for Lec2, an individual from Lesnik Cave with unclear dating but possible
of pre-LGM origin.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

No evidence for strong purifying selection in the Neolithic population
We next evaluated whether these demographic events led to a reduction in the efficacy of
purifying selection by inferring the heterozygosity at exons relative to that at introns for each
whole-genome sample (Fig 3C). As expected, we estimated lower diversity at exons than at
introns for all individuals, but found this reduction was much more pronounced in modern than
ancient individuals, which hints at increased purifying selection in recent times, probably as a
result of population growth (Gazave et al., 2013). However, we inferred similar diversity ratios
and thus a similar degree of purifying selection for Meso European-like and Neo Aegean-like
samples, albeit considerable individual variation. This suggests that the larger diversity seen
among Neo Aegean-like samples is not simply the result of a larger effective population size, but
rather of a particularly diverse source population, maybe as a result of past admixture (Marchi et
al., 2022).
Low X/A diversity of a Mesolithic female
To test for differences in sex-biased gene flow, we next compared the heterozygosity at neutral
regions on the X chromosome and autosomes (X/A diversity) for all female individuals with
whole-genome data (Fig 3D). Low X/A ratios may reflect demographic events such as recent
bottlenecks or relatively low effective sizes for females compared to males (Pool and Nielsen,
2007; Amster and Sella, 2020; Amster et al., 2020). The lowest diversity ratio inferred was for
the Early Mesolithic individual LEPE51, the only Meso European-like female for which sufficient
sequence data is currently available. Among modern samples, equally low ratios were found for
two individuals from forager populations (KhomaniSan-1, Saami-1). For other modern and all
Neo Agean-like individuals we inferred higher ratios, albeit with substantial variation.
Neolithic individuals (in the Danube Gorges) were smaller and tended to have lighter
pigmentation than indigenous foragers.
The genetic differences between local Mesolithic and incoming Neolithic populations at the
Danube Gorges translated into observable phenotypic differences. Here, we focused on the four
pigmentation phenotypes, skin pigmentation,eye pigmentation, hair pigmentation and hair
shade, that we predicted using the HIrisPlex-S system. Focusing on non-admixed Meso
European-like and Neo Aegean-like individuals along the Danubian corridor, none of these
phenotypes were significantly different between these groups. In combination, however, these
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

four pigmentation phenotypes were predictive of ancestry: using a linear discriminant analysis
(LDA), we identified a combination of lighter skin pigmentation, bluer eyes, lighter hair
pigmentation and lighter hair shade that predicted Neo Agegean-like ancestry for 30 (63%) of all
48 Neo Aegean-like individuals and two (40%) of the five admixed individuals with posterior
probability > 0.75, while all 15 Meso European-like had posterior probabilities < 0.5 and nine
(60%) even < 0.25 (Figure 4). The remaining 18 (37%) Neo Aegean-like individuals, however,
appear to have had phenotypic combinations rather similar to Meso European-like individuals
(i.e. darker pigmentation), as did the two second-generation admixed individuals from Lepenski
Vir. Notably, Neo Aegean-like individuals from the Marmara region and Greece south of the
Danube Gorges (the sites of Aktopraklık, Barcın, Nea Nikomedeia, Kleitos and Paliambela)
were more easily distinguished from Meso European-like individuals (13/15 or 87%) than Neo
Aegean-like individuals from the Danube Gorges or further north (17/33 or 52%, χ2=4.04,
p=0.044). Nonetheless, more than half of all Neolithic immigrants could easily be told apart from
local Mesolithic individuals just based on these four phenotypes, and likely even more based on
the entire habitus.
A particular phenotype previously shown to differ between these groups, for instance, was a
larger body size for Mesolithic than Neolithic individuals (Olalde et al., 2014; Ju and Mathieson,
2021; Marchi et al., 2022). Given the low number of Danube Gorges samples with reliable
osteological estimates, evidence for differences in body size mostly stems from comparisons
between periods, with Neolithic samples generally inferred as smaller than those from the
Mesolithic period (Macintosh, Pinhasi and Stock, 2016; Jovanović, 2017; de Becdelièvre et al.,
2020). However, it is interesting to note that the anthropological sexing errors discovered
through our genetic analysis almost exclusively involved relatively tall Meso European-like
females interpreted as males and relatively short Neo Aegean-like males as females
(Supplementary Data Table 1, (Borić and Price, 2013; Budd et al., 2013; Roodenberg, Gerritsen
and Özbal, 2013); personal communication).
Discussion
Genetic diversity in the Danube Gorges in a supra-regional context
On a projection-free PCA, Meso European-like individuals from the Danube Gorges spanning
more than 1,500 years are well differentiated from other Meso European-like individuals,
including those from Central Europe, North-Eastern Europe and the Black Sea region. This
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

differentiation was initially hypothesized to be the result of admixture with more eastern
hunter-gatherers (Mathieson et al., 2018; Feldman et al., 2019). However, it can also be
explained by bidirectional gene-flow between different hunter-gatherer groups (Feldman et al.,
2019), a hypothesis more in line with the deep split recently inferred between multiple
hunter-gatherer individuals, including Meso European-like individuals from the Danube Gorges
(Marchi et al., 2022). According to this scenario, the foraging communities of Europe suffered
from a major population reduction and diverged into several smaller groups during the LGM.
Members of one of those groups settled in the Danube Gorges, where they developed a
semi-sedentary lifestyle certainly as of the Late Mesolithic period (Dimitrijević, Živaljević and
Stefanović, 2016; de Becdelièvre et al., 2021). The level of gene flow with neighboring groups
remains unknown. But the rather high pairwise diversity observed among the Danube Gorges
individuals compared to, for instance, individuals from Central Europe, as well as the elevated
heterozygosity of at least some Danube Gorges individuals is indicative of a relatively large and
well connected population, an interpretation well in line with the richness of archaeological finds
from this period (Borić and Stefanović, 2004; Borić et al., 2014; Borić, 2016, 2021). The
difference in diversity appears particularly stark when compared to the genomic data from
Criewen (GR2), the most recently dated Central European individual with 100%
Meso-European-like ancestry, or the Baltic site of Zvejnieki. In contrast to the site of Zvienjeki,
however, we found no indication of an influx of Mesolithic individuals with very distinct
ancestries at the Danube Gorges.
That changed towards the end of the 7th millennium BC when individuals with Neo Aegean-like
ancestry appeared in the Danube Gorges. The Mesolithic and Neolithic sites in the Danube
Gorges provide a unique opportunity to genetically study forager-farming interactions at high
resolution and interpret them in a European context. Whether Lepenski Vir was a forager
community attracting individuals from the farming frontier, or whether the site was possibly
newly occupied around c. 6,200 BC by arriving migrants from the Aegean region attracting
forager individuals, has been a long-standing archeological debate (Srejović, 1969; Garašanin
and Radovanović, 2001; Borić and Price, 2013). Signals of admixture between individuals with
Meso European-like and Neo Aegean-like ancestry have been reported previously for several
other prehistoric sites, where the rate of forager introgression was found to have been very low
initially but to have increased significantly over time (Lipson et al., 2017). In contrast, evidence
for farming individuals joining previously established forager societies remains rare (Bramanti et
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

al., 2009; Bollongino et al., 2013; Hofmanová, 2016; Hofmanová et al., 2016; Mathieson et al.,
2018), but early farmers, who were adapting their subsistence to new environmental conditions,
might have been attracted by forager communities that provided local knowledge and access to
resources, such as the permanent abundance in fish species in the Gorges at the time of the
8.2 cal BC cooling event. For local foragers, in turn, the resulting mutualistic relationships
allowed for the exchange of goods and an access to a wider mating network (Bocquet-Appel
and Bar-Yosef, 2008).
Admixed individuals were also admixed culturally
This study brings evidence that admixture between individuals with Meso European-like and
Neo Aegean-like ancestries certainly occurred before 6000 cal BC in the Central Balkans: the
second-generation admixed individuals (male LEPE18, LV 27d; female LEPE46, LV 93), as well
as the previously reported admixed case from Lepenski Vir (LEPI_61, LV 61; (Mathieson et al.,
2018) were all dated to the period of Transformation (ca 6200-5950 cal BC). Although they were
buried in extended supine position, following local foragers’ funerary customs, various
contextual and bioarchaeological evidence indicate that admixture may have locally triggered a
complex pattern of cultural mixing. For instance, the admixed male child LV 61 (6225-5915 cal
BC; 4-7 years old) has been buried through the plastered floor of one of the oldest building from
Lepenski Vir (building 40), in a practice with Neolithic Southern Balkans and Anatolian
similarities (Borić and Stefanović, 2004; Stefanović and Borić, 2008). Mostly reserved at
Lepenski Vir to infants and young children, and to a few adults with non-local strontium
signatures (Borić and Price, 2013), this practice may have been associated with their social
construction or status. According to stable isotope values, this child has been fed with large
amounts of aquatic resources, consistent with nutritional socialization (de Becdelièvre, 2020).
Similarly, the second-generation admixed female with ¾ Neo Aegean-like ancestry (LEPE46, LV
93; 6226-6026 cal BC) was also buried into a building (building 72) with various cultural
elements pointing to the Early Neolithic cultural sphere (including numerous limestone beads,
as well as a fragmented stone ring and adze). In contrast, the second-generation admixed male
individual with ¾ Meso European-like ancestry (LEPE18, LV 27d; 6226-6026 cal BC) was
discovered in a primary disturbed burial that contained grave goods associated with both
Mesolithic (deer antler) and Neolithic (pottery fragments) communities. The other individuals,
buried in extended or slightly flexed positions in continuity with local Mesolithic traditions,
included the Meso European-like LEPE53 (LV 27a) and LEPE17 (LV 27b) with a mtDNA
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

haplogroup frequently observed in early Neolithic farmers in Europe (N1a; Supplementary Data
Table 1). All had a typically local diet, rich in aquatic proteins (Supp Info). The funerary practices
associated with these individuals thus reflect the mosaic pattern of the Mesolithic and Neolithic
cultural assimilation at Lepenski Vir.
First-generation immigrants were culturally inter-connected
Some individuals, namely LEPE39 (LV 82), LEPE48 (LV 122) and LEPE52 (LV 73), likely
represent first-generation immigrants that did not admix: they all had >96% Neo Aegean-like
ancestry, had a diet distinct from that of local Meso European-like samples (Supplementary
Data Table 1), and the two samples for which strontium was measured (LEPE52, LEPE48)
showed non-local signatures (Borić and Price, 2013). Of those, the male LEPE52 from the Early
Neolithic Period was buried in a flexed position, which is considered typical for Anatolian and
Balkan Early Neolithic communities. However, the burial practices associated with the two
samples from the earlier Transformation Period reflect elements likely associated with both
Neolithic and Mesolithic funerary rites: The male LEPE39 has been discovered disarticulated
and only the isolated skull (calvaria) of the non-local young (15-20 years) female LEPE48 was
found beneath a building floor (building 47). The practice of disarticulating the skull, while also
found among Early Neolithic Anatolian communities and, albeit more scarcely, in the Southern
Balkans (Talalay, 2004), was common during the Mesolithic at the Danube Gorges: among the
Meso European-like samples from the Danube Gorges studied here, four samples (VLASA4,
VLASA10, VLASA32, VLASA41), all predating the appearance of Neo Aegean-like ancestry,
were buried with signs of disarticulation. Being buried in a dwelling, on the other side, is a
funerary practice frequently found in Early Neolithic Anatolian contexts (Adams and King, 2011;
Brami, 2017). Together, these samples thus again attest to the gradual pattern of social
integration between the groups as well as the cultural transformation triggered by this interaction
at the Danube Gorges.
Interactions between foragers and farmers in the Danube Gorges
While the results shown above attest to a certain degree of cultural syncretism during the
Transformation Period, this does not seem to apply to all sites in the Danube Gorges. All
Danube Gorges individuals for which we estimate at least some Neo Aegean-like ancestry were
buried at Lepenski Vir. A single additional individual with some degree of Neo Aegean-like
ancestry was previously reported from Padina (PADN_4; (Mathieson et al., 2018)), the only
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

other Danube Gorges site at which trapezoidal houses were found. In contrast, at the nearby
site of Vlasac, the genome-wide ancestry of all seven individuals analyzed, as well as mtDNA
haplogroups identified for additional 9 individuals, are all consistent with Meso European-like
ancestry (Supplementary Data Table 1), in line with previous reports for Vlasac and Hajducka
Vodenica (Mathieson et al., 2018). The majority of these samples likely pre-date the arrival of
the Neolithic. However, at least seven individuals are confidently dated to the Transformation
period, making it unlikely to have missed Neo Aegean-like ancestry if it was common during this
period at those sites.
A few hundred years later, in the immediate vicinity of the Danube Gorges, the sites of Vinča-
Belo Brdo and Starčevo were established (Whittle et al., 2002; Tasić et al., 2015; Porčić et al.,
2021). The sites, both associated with the Neolithic Starčevo-Körös-Criş cultural complex and a
subsistence more oriented towards the consumption of C3 plants (such as crops), meat/dairy
products of domesticates and wild game (Filipović and Obradović, 2013; Jovanović et al., 2019;
Stojanovski et al., 2020), paint a highly contrasting picture: For the only individual with
whole-genome data available from these sites (Star1; (Marchi et al., 2022)), we estimate >96%
Neo Agean-like ancestry, a signal confirmed by seven new mtDNA lineages (Supplementary
Data Table 1).
Collectively, this observation suggests the continued co-existence (6200-5950 cal BC) of
foraging and early farming communities if not at the same site, then at least in the same
settlement area. Only in Lepenski Vir and possibly in Padina does the interaction take place at
the same site -perhaps even into the Neolithic period. However, the Vlasac site was possibly no
longer used as a settlement during the Transformation period, but only as an ancestral burial
site by people with Meso European-like ancestry (Borić et al., 2014).
Mesolithic cultural elements disappeared gradually. Several elements of the Neolithic culture,
such as domesticated animals, crop consumption and a typical Neolithic symbolic repertoire,
appear at Lepenski Vir only after the Transformation phase when trapezoidal houses were also
abandoned and the flexed position became the new dominant mortuary canon (Porčić,
Blagojević and Stefanović, 2016; Blagojević et al., 2017; Jovanović et al., 2019; de Becdelièvre
et al., 2020). This cultural change coincides with a significant population increase at Early
Neolithic sites in the Central Balkans and is associated with a general population increase and a
higher percentage of individuals with non-local isotope signals, suggesting a second wave of
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Neolithic immigrantion (Borić and Price, 2013; Porčić, Blagojević and Stefanović, 2016;
Blagojević et al., 2017; de Becdelièvre et al., 2021). In line with this view, the three
first-generation immigrants at Lepenski Vir (LEPE39, LEPE48 and LEPE52) were all found to be
unrelated and most likely date to different generations (Table 1). In addition, we found a lower
fraction of the typical Meso European-like mtDNA haplogroup (U5) among individuals dating to
the Early Neolithic Period than among individuals dating to the Transformation Period (3/15 vs.
12/23). Hence, the Neolithisation of the Danube Gorges should not be understood as a
straightforward process of acculturation or a sudden behavioral shift. Results rather reflect a
mosaic picture of complex behavioral interactions and increased immigrations which triggered
gradual socio-cultural changes within the framework of local economic and ecological continuity.
Conclusion
The analyses presented here consolidate the picture of the Neolithisation of South-Eastern and
Central Europe within the framework of a demic diffusion. Using heterozygosity estimates, we
show the decrease in genetic diversity from sites in NW Anatolia to those in Central Europe
resulting from the demic expansion along the archaeologically attested expansion route.
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”
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv 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.
STAR Methods
Key Resources Table
REAGENT or RESOURCE
SOURCE
IDENTIFIER
Biological samples
Ancient human bone
material. Danube Gorges
whole genome data
This study
VLASA16, VLASA37, LEPE46, LEPE53, LEPE45,
LEPE51
Ancient human bone
material. Danube Gorges
neutralome data
This study
VLASA4, VLASA10, VLASA32, VLASA41, VLASA44,
LEPE39, LEPE52, LEPE53, LEPE18, LEPE46
Ancient human bone
material. Lec2 neutralome
data
This study
Lec2
Ancient human bone
material. North Eastern
Europe neutralome data
This study
Min2, Min3, Min5, Min8, Min10, Min11, ZVEJ317,
ZVEJ39, ZVEJ76. ZVEJ122, ZVEJ162, ZVEJ170
Ancient human bone
material. Central Europe
Meso nuclear capture data
This study
Fre3, Fr1, GrO1, Gr2, Bla20
Ancient human bone
material. North Western
Anatolia neutralome data
This study
AKT6, AKT16, AKT18, AKT20, AKT26, Bar11, Bar15,
Bar16, Bar20, Bar32
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Ancient human bone
material. Lower Austria
neutralome data
This study
Asp1, Asp2, Asp3, Asp4, Asp6, Asp8, Asp10, Klein1,
Klein2, Klein3, Klein4, Klein5, Klein8, Klein10
Ancient human bone
material. Southern
Germany neutralome data
This study
Ess7, Dil15, Dil16, Herx, Herx2, Herx3, Herx4, Herx5,
Herx6,Herx7,Herx8,Herx9
Ancient human bone
material. Blätterhöhle
Mid-Neo neutralome data
This study
Bla1, Bla10, Bla13, Bla15, Bla17, Bla28, Bla29, Bla75,
Bla32, Bla59, Bla45
Ancient human bone
material. MT capture data
This study
78 as listed in Supplementary Data Table 1
Chemicals, peptides, and recombinant proteins
Phenol/chloroform/isoamyla
lcohol (25:24:1)
Roth,
Karlsruhe,
Germany
Cat#A156.1
AmpliTaq Gold ® Buffer II
(10x)
Life
Technologies ™
Cat#4311816
AmpliTaq Gold ® DNA
Polymerase
Life
Technologies ™
Cat#4311816
ATP Solution (100 mM)
Life
Technologies ™
Cat#R0441
Bovine Serum Albumin
(BSA) (20 mg/ml)
Roche
Diagnostics
Cat#10711454001
Bst Polymerase, Large
Fragment (8 U/μl)
New England
Biolabs GmbH
Cat#M0275S
dNTPs (each 10 mM)
QIAGEN,
Hilden,
Germany
Cat#201901
dNTPs (each 25 mM)
Agilent
Technologies
Cat#600677
EDTA (0.5 M), pH 8.0
Ambion/Applied
Biosystems,
Life
Technologies
™, Darmstadt,
Germany
Cat#AM9262
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Ethanol 96% spoilt
Carl Roth
GmbH + Co.
KG
N/A
Herculase II Fusion ® DNA
Polymerase
Agilent
Technologies
Cat#600677
Herculase II Reaction
Buffer
Agilent
Technologies
Cat#600677
MgCl2(25 mM)
Life
Technologies ™
Cat#4311816
Sodium N-lauryl
sarcosinate
Merck Millipore,
Darmstadt,
Germany
Cat#428010
Nuclease-free H2O
Life
Technologies ™
Cat#AM9932
PEG-4000
Thermo
Scientific ™
Cat#EL0011
Proteinase K
Roche
Diagnostics,
Mannheim,
Germany
Cat#3115828001
T4 DNA Ligase (5 U/μl)
Thermo
Scientific ™
Cat#EL0011
T4 DNA Ligase Buffer
(10X)
Thermo
Scientific ™
Cat#EL0011
T4 DNA Polymerase (5
U/μl)
Thermo
Scientific ™
Cat#EP0062
T4 Polynucleotide Kinase
Invitrogen ™
Cat#EK0032
Tango Buffer (10x)
Life
Technologies ™
Cat#BY5
ThermoPol Buffer (10X)
New England
Biolabs GmbH
Cat#M0275S
USERTM enzyme
New England
Biolabs GmbH
Cat#M5505L
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Trichlormethan/Chloroform
Roth,
Karlsruhe,
Germany
Cat#3313.1
Critical commercial assays
Agilent 2100 Expert
Bioanalyzer System and
High Sensitivity DNA
Analysis Kit
Agilent
Technologies
Cat#5067-4626 (kit)
Qubit Fluorometric
quantitation and dsDNA HS
Assay Kit
Invitrogen
Cat#Q32854 (kit)
Cat#Q32856 (tubes)
Deposited data
Sequencing data (this
study)
European
Nucleotide
Archive
https://www.ebi.ac.uk/ena/browser/view/PRJEB47916
Human reference sequence
(hs37d5)
(Mallick et al.,
2016)
ftp://ftp.1000genomes.ebi.ac.uk/vol1/ftp/technical/referenc
e/phase2_reference_assembly_sequence/hs37d5.fa.gz
20 samples from Simons
Genome Diversity Panel
(Mallick et al.,
2016)
https://www.ebi.ac.uk/ena/browser/view/PRJEB9586.
LP6005592-DNA_C01_S_Saami-1,
LP6005442-DNA_D10_S_Finnish-1,
LP6005442-DNA_E10_S_English-1,
LP6005442-DNA_B11_S_Spanish-2,
LP6005441-DNA_B05_S_French-2,
LP6005441-DNA_B06_S_Bergamo-2,
LP6005441-DNA_D11_S_Sardinian-2,
LP6005443-DNA_H05_S_Czech-2,
LP6005592-DNA_E02_S_Polish-1,
LP6005442-DNA_B08_S_Hungarian-1,
LP6005442-DNA_A03_S_Bulgarian-1,
LP6005677-DNA_B01_S_Albanian-1,
LP6005442-DNA_G07_S_Greek-1,
LP6005677-DNA_A03_S_Turkish-2,
LP6005442-DNA_B04_S_Georgian-1,
LP6005441-DNA_B02_S_BantuKenya-2,
LP6005677-DNA_D03_S_Khomani_San-1,
LP6005592-DNA_C05_S_Khomani_San-2,
LP6005441-DNA_F07_S_Mandenka-2,
LP6005441-DNA_B08_S_Mbuti-2
Danube genomes
(Marchi et al.,
2022)
VLASA7, VLASA32, LEPE48, LEPE52
North Western Anatolia and
Aegean genomes
(Hofmanová et
al., 2016;
Marchi et al.,
2022)
Bar25, AKT16, Bar31, Bar8, Rev5, Klei10, Rev5
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

North Greece genomes
(Marchi et al.,
2022)
Nea2, Nea3
Serbia genomes
(Marchi et al.,
2022)
STAR1, VC3-2
Austria genomes
(Marchi et al.,
2022)
Asp6, Klein7
Germany genomes
(Lazaridis et al.,
2014; Marchi et
al., 2022)
Dil16, Ess7, Herx, Stuttgart
Sunghir genomes
(Sikora et al.,
2017)
SunghirI, SunghirII, SunghirIII, SunghirIV
Abdul Hosein, Wezmeh
Cave and Boncuklu
genomes
(Broushaki et
al., 2016; Kılınç
et al., 2016)
AH1, AH2, AH4, WC1, Bon002
Kotias, Satsurbia, Balkan
and Scandinavia genomes
(Lazaridis et al.,
2014; Jones et
al., 2015, 2017;
Günther et al.,
2018)
KK1, Satsurbia, Latvia_HG2, Latvia_HG3, Latvia_MN2,
Latvia_HG1, Hum1, Hum2, SF9, Steigen, Motala12
Central Europe Meso
genomes
(Lazaridis et al.,
2014; Olalde et
al., 2014; Jones
et al., 2015;
González-Forte
set al., 2017)
Bichon, Loschbour, LaBrana, Canes1-Meso, Chan-Meso,
OC1-Meso, SC2-Meso, GB1-Eneo
Hungarian genomes
(Gamba et al.,
2014)
KO1, NE1
Cova Bonica and Ajvice
genomes
(Skoglund et
al., 2014;
Olalde et al.,
2015)
CB13, Ajv58
1240K capture sites
(Mathieson et
al., 2015)
Allen Ancient DNA
Resource v42.4 and v37.2
Reich lab public
data release
https://reichdata.hms.harvard.edu/pub/datasets/amh_repo
/curated_releases/index_v42.4.html ;
https://reich.hms.harvard.edu/allen-ancient-dna-resource-
aadr-downloadable-genotypes-present-day-and-ancient-d
na-data
genomic masks
UCSC genome
browser
http://hgdownload.cse.ucsc.edu/goldenpath/hg19
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

strict callability mask from
1000G
(1000
Genomes
Project
Consortium et
al., 2015)
http://ftp.1000genomes.ebi.ac.uk/vol1/ftp/release/201305
02/supporting/accessible_genome_masks/
37,574 1kb long genomic
loci
(Gronau et al.,
2011)
http://compgen.cshl.edu/GPhoCS/data.php
Known InDel positions
Consortium and
The 1000
Genomes
Project
Consortium,
2015
ftp://gsapubftp-anonymous@ftp.broadinstitute.org/bundle/
b37/1000G_phase1.indels.b37.vcf.gz
Known InDel positions
Consortium and
The 1000
Genomes
Project
Consortium,
2015
ftp://gsapubftp-anonymous@ftp.broadinstitute.org/bundle/
b37/Mills_and_1000G_gold_standard.indels.b37.vcf.gz
GERP scores
(Cooper et al.,
2005),
ENSEMBL
http://ftp.ensembl.org/pub/release-96/compara/conservati
on_scores/88_mammals.gerp_conservation_score/gerp_
conservation_scores.homo_sapiens.GRCh38.bw
Oligonucleotides
MYBait kit
Arbor
biosciences;
https://arborbio
sci.com/genomi
cs/targeted-seq
uencing/mybait
s/mybaits-custo
m/
N/A
SureSelectTM XT in-solution
target enrichment kit
Agilent
Technologies
(custom
design); (Gnirke
et al., 2009)
N/A
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

P5 and P7
(Meyer and
Kircher, 2010)
IDT, Leuven,
Belgium
N/A
IS4, IS5, IS6 and IS7
(Meyer and
Kircher, 2010)
IDT, Leuven,
Belgium
N/A
Software and algorithms
fastqc
www.bioinforma
tics.babraham.a
c.uk/projects/fa
stqc/
www.bioinformatics.babraham.ac.uk/projects/fastqc/
Trim Galore!
https://www.bioi
nformatics.babr
aham.ac.uk/proj
ects/trim_galore
/
https://www.bioinformatics.babraham.ac.uk/projects/trim_
galore/
bwa mem -
Burrows-Wheeler
Alignment Tool
Li 2013
https://github.com/lh3/bwa
SAMtools
(Li et al., 2009)
https://github.com/samtools/samtools
Picard-tools
http://broadinstit
ute.github.io/pic
ard/
http://broadinstitute.github.io/picard/
seqtk
https://github.co
m/lh3/seqtk
https://github.com/lh3/seqtk
GATK
(DePristo et al.,
2011)
https://github.com/broadinstitute/gatk/releases
Snakemake
(Koster and
Rahmann,
2012)
https://snakemake.readthedocs.io/en/stable/getting_starte
d/installation.html
ATLAS
(Link et al.,
2017)
https://bitbucket.org/wegmannlab/atlas/wiki/Installing%20
and%20Running%20ATLAS
ContamMix
(Fu et al., 2014)
N/A
MIA
https://github.co
m/mpieva/mapp
ing-iterative-ass
embler
https://github.com/mpieva/mapping-iterative-assembler
mafft
(Katoh et al.,
2002)
https://mafft.cbrc.jp/alignment/software/
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

ANGSD
(Korneliussen,
Albrechtsen
and Nielsen,
2014)
https://github.com/ANGSD/angsd
Haplogrep
(Weissensteine
ret al., 2016)
https://haplogrep.i-med.ac.at/category/haplogrep2/
entropy2
(Shastry et al.,
2021)
https://bitbucket.org/buerklelab/mixedploidy-entropy/src/m
aster/
NGSadmix
(Skotte,
Korneliussen
and
Albrechtsen,
2013)
http://www.popgen.dk/software/index.php/NgsAdmix
pcAngsd
(Meisner and
Albrechtsen,
2018)
https://github.com/Rosemeis/pcangsd
bedtools
(Quinlan and
Hall, 2010)
https://bedtools.readthedocs.io/en/latest/content/installatio
n.html
hgLiftOver
https://genome.
ucsc.edu/cgi-bi
n/hgLiftOver
https://genome.ucsc.edu/cgi-bin/hgLiftOver
NRE
(Arbiza, Zhong
and Keinan,
2012)
http://nre.cb.bscb.cornell.edu
plink2
(Chang et al.,
2015)
https://www.cog-genomics.org/plink/2.0/
convertf
https://reich.hm
s.harvard.edu/s
oftware/InputFil
eFormats
https://reich.hms.harvard.edu/software/InputFileFormats
R
R Core Team
(2019). R: A
language and
environment for
statistical
computing. R
Foundation for
Statistical
Computing,
Vienna, Austria
https://www.R-project.org/
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

hapROH
(Ringbauer,
Novembre and
Steinrücken,
2021)
https://pypi.org/project/hapROH/
HIrisPlex-S webtool
(Chaitanya et
al., 2018)
https://hirisplex.erasmusmc.nl/
Other
Amicon Ultra-15 Centrifugal
Filter Units, 30kDa and
50kDa
Merck Millipore,
Darmstadt,
Germany
Cat#UFC803096 and Cat#UFC905096
MinElute ® PCR
Purification Kit
QIAGEN,
Hilden,
Germany
Cat#28006
MSB ® Spin PCRapace
Invitek, Stratec
Molecular,
Berlin,
Germany
Cat#1020220400
QIAquick PCR Purification
Kit
Qiagen
Cat#28106
Resource Availability
Lead Contact
Further information and requests for resources and reagents should be directed to and will be
fulfilled by the Lead Contacts, Joachim Burger (jburger@uni-mainz.de) and Daniel Wegmann
(daniel.wegmann@unifr.ch).
Materials Availability
Genomic data are available at the European Nucleotide Archive under the accession number
PRJEB47916 in BAM and FASTQ format. Mitochondrial capture data are available at GenBank.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Data and Code Availability
All genomic data used in this study is publicly available at the European Nucleotide Archive
under the accession number PRJEB47916 or at the sources listed in the Key Resources Table.
The pipeline used to process the raw data is available on
https://bitbucket.org/wegmannlab/atlas-pipeline/src/master/. The code used to plot relatedness
is available on Bitbucket at https://bitbucket.org/wegmannlab/atlas/downloads/Relatedness.R.
Experimental Model and Subject Details
Archaeological context
The samples analyzed in this study have been obtained from archaeological collections with
permission of the responsible curators or excavators.
Barcın
The Barcın Höyük was occupied without interruptions between 6,600-6,000 cal BC (Gerritsen,
Özbal and Thissen, 2013) and shows continuity from pre-Fikirtepe to Fikirtepe horizons of the
Neolithic period ((Hofmanová et al., 2016); Suppl. S4-S5). While the site shares common
elements with other Fikirtepe sites in the Marmara region, there were differences in architecture
and dietary habits noted between flat sites (e.g., Aktopraklık) and tell sites (e.g., Barcın) (Karul
and Avcı, 2013). This site is the oldest known Neolithic occupation in NW Anatolia (Gerritsen,
Özbal and Thissen, 2013) and from the start, the food economy was fully agrarian with absence
of an earlier transitional phase from foraging to farming (Arbuckle et al., 2014). The dead were
buried at several locations within the settlement: neonates and infants were buried within or
close to the houses, generally next to the walls, whereas juveniles and adults were buried in
primary single burials in the central courtyard (Alpaslan Roodenberg, Gerritsen and Özbal,
2013).
Aktopraklık
Aktopraklık is a flat inland Fikirtepe site about 25 km from the city of Bursa (Marmara region)
and its excavation showed uninterrupted occupation from the middle of the 7th millennium BC to
the middle of the 6th millennium BC (Karul and Avcı, 2013). Site Aktopraklık C served as a
settlement (with associated human remains) during the earlier phases and as a cemetery during
the later phases (the settlement moved to Aktopraklık B during the Chalcolithic) (Karul and Avcı,
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

2011). In the Neolithic phase, the individuals were buried close to buildings or beneath them
(Songül Alpaslan-Roodenberg and Roodenberg, 2020).
Asparn-Schletz
Asparn-Schletz is a large settlement enclosed by two ditch systems (Windl, 2009) with signs of
occupation (Windl, 1999). There were 20 burials in the ditch and 130 individuals interred without
a classical burial, often with perimortem trauma and likely buried some time after death
(Teschler-Nicola, 2012). The burials predate the presumably violent event that has taken place
in the end of the LBK, ca. 7150-6900 BP (Wild et al., 2004) and possibly the site was
abandoned after this (Windl, 2009).
Kleinhadersdorf
The classical LBK cemetery, the largest such site in Austria, was located close to a LBK
settlement and is dated to the second half of the 6th millennium BC (Neugebauer-Maresch and
Lenneis, 2015). Features of the lithic technology and red ochre have been interpreted in the
past as possible signals of continuation of Mesolithic traditions (Mateiciucová, 2015). Most
graves were single primary inhumations in a contracted position on the left side typical of LBK
with some exceptions (burials on the back, possible cenotaphs) (Neugebauer-Maresch and
Lenneis, 2015; Tiefenböck and Teschler-Nicola, 2015).
Essenbach-Ammerbreite
Essenbach-Ammerbreite was an LBK cemetery close to an LBK settlement (Brink-Kloke, 1990).
The same individual as in (Marchi et al., 2022) has been analysed in this study, namely Ess7, a
child from grave 7.
Dillingen-Steinheim
Dillingen-Steinheim is a LBK cemetery of 27 classical burials (Nieszery, 1995), some in a ditch
(Marchi et al., 2022). Two such individuals, analyzed in this study were Dil15 and Dil16 (graves
23 and 24, respectively) and they have been dated to 5,116 ± 118 cal. BC (Pechtl, 2015). They
have been found to be brothers (see Results).
Herxheim
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Herxheim is a well-known site, especially in relation to violence during the Early Neolithic period.
The site is enclosed by two ditch segments that contained remains of more than 500 individuals
(Haack, 2016; Zeeb-Lanz, 2016, 2019) with signs of perimortem and postmortem violence on
the skeletons (Boulestin and Coupey, 2015). Interestingly, individuals formally buried showed a
local range of Sr values in contrast to the scattered remains that showed more often a non-local
signal (Turck, 2019).
Lesnik cave
The site of Lesnik cave is located near Yalta, Ukraine. The analyzed sample (Lec2) was not
directly dated, but based on the contextual finds, Lec2 was placed in the early Mesolithic or Late
Palaeolithic of the region. The other human sample (Lec1) from this cave, was dated to
11260+/-45 BP (OxA-19112), cal Oxcal 4.4 11292-11144 BC (Ridush, 2009; Schulz, 2016).
Zvejnieki
Zvejnieki in northern Latvia is a cemetery where individuals were buried from the middle
Mesolithic to Neolithic period, from the 8th to 4th millenia BC (Eriksson, Lõugas and Zagorska,
2003; Stutz, Larsson and Zagorska, 2013). Individuals at the site were partially mobile and the
inhumations did not respect one another, suggesting the absence of knowledge of previous
burial placement in some later periods (Larsson et al., 2017).
Minino
While the site of Minino, Russia, is dated from the Palaeolithic to Neolithic, the main part is
assigned to the Mesolithic period when burial complexes are very rare. The ages of the samples
range from 5650-4600 cal BC to 8671±48 cal BC (Wood et al., 2013).
Criewen
Criewen, discovered in northern Germany, provided this study with samples associated with a
non-agricultural context that were dated to 4770±40 cal BC (Gr1) and 4600±60 cal BC (Gr2)
(Geisler and Wetzel, 1999). One of the samples was buried with ca. 3,000 perforated shells
(Street et al., 2001).
Groß Fredenwalde
At Groß Fredenwalde, at least eight burials have been found with 12 individuals. The site was
dated from 6,400 cal BC to 4,900 cal BC with one individual overlapping with the presence of
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

farmers in the region in approx. 5,200 cal BC (Terberger et al., 2015; Kotula, Piezonka and
Terberger, 2020).
Große Ofnet Höhle
33 skulls separated from their postcranial skeletons have been found at the site. They date to
5770 cal BC (Walde et al., 1986; Orschiedt, 1998).
Blätterhöhle
Blätterhöhle shows signs of human occupation from the Mesolithic to Late Neolithic. During the
Mesolithic, the site has been used as a sporadic settlement. Human remains have been
discovered inside the cave in a disturbed context, showing presence of hunter-gatherers and
farmers through time but also in parallel (human remains from individuals with different
subsistence were differentiated by isotopic analysis) (Orschiedt et al., 2012; Bollongino et al.,
2013).
Lepenski Vir
This large and famous settlement in the Danube Gorges is a type-site of the Lepenski Vir culture
(sometimes called Lepenski Vir-Schela Cladovei culture). The site with its abundance of burials
has been under discussion ever since its excavation (started 1965 by Dragoslav Srejenović)
and the chronology of the area has been revised with the use of radiocarbon dating corrected
for freshwater reservoir effect (Borić, 2002). The sampling strategy and further details about the
site are provided in other parts of this study.
Vlasac
The Vlasac site is geographically very close to Lepenski Vir (cca 3 km downstream) and was
first excavated in the same period as Lepenski Vir (Boroneanţ, 2011). The site was assigned to
the Lepenski Vir culture and is mostly dated to Late Mesolithic, while there are dates as old as
9,800 cal BC known from the site (Bonsall et al., 2000; Borić and Stefanović, 2004; Borić,
French and Dimitrijević, 2008). Additionally, new excavations (during seasons 2006–2009)
showed that there was also an occupation parallel to the Transformation phase of Lepenski Vir
with appearance of features influenced by the Neolithic (Early Starčevo ceramics, Spondylus
shells and discoid beads) (Borić et al., 2014). Most of the settlement was abandoned
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

∼6,300/6,200 cal BC, but the site might have served as an “ancestral” place and a cemetery for
burial rites of Mesolithic tradition during the last centuries of the 7th millennium BC (Borić et al.,
2014).
Grivac
This well-stratified Neolithic site in central Serbia contains both proto-Starčevo and Vinča layers
(Bogdanović, 2008; Porčić, Blagojević and Stefanović, 2016). The individual included in this
study (Gri1) was buried in grave 1.
Padina
Padina is a Danube Gorges site with a superposition of Mesolithic and Early Neolithic
structures. It is associated with the Lepenski Vir culture, with additional later layers of Late
Neolithic cultures present at the site (Borić and Miracle, 2004). The only ancient individual
successfully analysed from this site (Pad 11, burial 30) is assigned to the Early/Middle
Mesolithic (∼9,500-7,400 cal BC) (Borić and Price, 2013).
Ostrovul Corbului
Ostrovul Corbului is located on the Romanian side of the Danube Gorges, downstream from
Lepenski Vir. It was originally assigned to the Schela Cladovei culture, which was later
connected to the Lepenski Vir culture (Boroneanţ, 2011). There are Mesolithic and Neolithic
layers at the site (Roksandić, 1999).
Vinča-Belo Brdo
Vinča-Belo Brdo is a typical site of the eponymous Vinča culture with key Neolithic
developments such as the formation of large settlements and tells, the intensification of farming
subsistence and the expansion of material networks (Tasić et al., 2015). The culture occupied a
large region of Serbia and several bordering countries between the late 6th millennium BC and
middle 5th millennium BC (Borić, 2015). The only burial at the site is a collective burial at the
lowermost level (Dimitrijević, 2014) and it is still under discussion whether this Starčevo level
represents continuous occupation to later phases or an earlier abandoned settlement (Borić,
2009). The collective burial can be dated to 5,476-5,304 cal BC (Borić, 2009) and the samples
analyzed here are coming from this context.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Sultana Malu Roşu
This Eneolithic site is located on the bank of the Mostiştea River, about 15 km from the Danube
(approximately 500 km from the Danube Gorges) (Lazăr et al., 2008).
Method Details
Sample preparation
Sample preparation took place in a dedicated ancient DNA facility of the Paleogenetics Group at
the Institute of Organismic and Molecular Evolution (iomE) at the Johannes
Gutenberg-University in Mainz, following and further improving guidelines for good practices in
ancient DNA analysis (Bramanti et al., 2009; Bollongino et al., 2013; Scheu et al., 2015) . We
processed blank controls alongside milling, extraction and library built to control for the
decontamination procedure of the devices used. Sample treatment and library preparation were
performed as described in (Kircher, Sawyer and Meyer, 2012) with the adaptations described in
(Scheu et al., 2015) and (Hofmanová et al., 2016). Sample specific modifications/adaptations
are noted in Supplementary Data Table 1. 12 samples underwent a pre-lysis step during DNA
extraction as described in (Scheu et al., 2015) in order to increase the percentage of
endogenous yields. USERTM (NEB) treatment of the DNA extract was performed for 181 libraries
prior to Library preparation. To increase library complexity, we amplified each library in three or
six PCR-parallels.
Quality assessment
With the first library of each sample, we estimated library complexity by quantitative real time
PCR (qPCR) as described in (Meyer and Kircher, 2010; Hofmanová et al., 2016). Additionally,
we performed a shallow screening on an Illumina Miseq (50 bp, single end sequencing) that
was analyzed with the pipeline described in (Hofmanová et al., 2016).
Library preparation and Target enrichment
Based on the endogenous DNA content (fraction of reads aligning to the reference genome)
and the qPCR results (Figure S1A) we chose three strategies on a per sample basis, for which
additional extractions and libraries were created: 1) We chose samples of highest quality for
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

whole genome sequencing, aiming to obtain genomes of differing ages from the Danube
Gorges. Selected samples were sequenced on an Illumina Hiseq2500 (100bp paired-end or
single-end) and an Illumina NextSeq (75bp paired-end). 2) We also selected a total of 75 high
quality samples to perform a nuclear target enrichment (nuclear capture). The capture array of
5Mb sites (i.e. neutralomes) followed the protocol as described in (Veeramah et al., 2018) with
some alterations. Sequencing was performed on an Illumina Hiseq 2500 with 100bp single-end
or paired-end runs and on an Illumina NextSeq with 75bp paired-end sequencing at the
University of Mainz. 3) We performed mitochondrial target enrichment for samples with relatively
low DNA quality. Target enrichment was performed two times (double capture) following the
methods described in (Hofmanová et al., 2016), with the exception, that PCR purification was
performed with QIAquick PCR Purification Kit columns (QIAGEN®) according to the companies
protocol, eluting in 33µl preheated (65°C) elution buffer for 5 minutes. For 9 samples
(Supplementary Data Table 1), the supernatants from both nuclear target enrichment steps
containing non-target DNA, were used to perform an additional mitochondrial capture
experiment (supernatant capture). The supernatants were purified with QIAquick PCR
Purification Kit columns prior to the first mitochondrial target enrichment. Samples were pooled
equimolarly and sequenced at GENterprise GENOMICS (Mainz) on an Illumina MiSeq
sequencing system with 50 cycles single-end or on an Illumina HiSeq sequencing system with
100 cycles paired-end. Mitochondrial data was additionally gained as by-products of the
previous two strategies.
Bioinformatic processing
The mitochondrial captures were processed as described in (Hofmanová, 2016; Hofmanová et
al., 2016). Read statistics can be found in Supplementary Data Table 1. For all 134 ancient
genomes and neutralomes, alignment, local realignment, PMD and recal estimation were
processed with commit 9ec713b of the ATLAS-Pipeline
(bitbucket.org/wegmannlab/atlas-pipeline/wiki/Home) with some minor changes depending on
how the data were obtained as indicated below and in Supplementary Data Table 1. Reads
were trimmed with length filter ≥ 30 (TrimGalore, v0.6.4,
https://github.com/FelixKrueger/TrimGalore), aligned with bwa-mem (v0.7.17, (Li, 2013)) to the
hs37d5 reference (Mallick et al., 2016), filtered for mapping quality < 30, sorted and indexed
(SAMtools, v.1.9, (Li et al., 2009)). Read groups were added with picard-tools (v2.21.1,
http://broadinstitute.github.io/picard/) to keep track of libraries. Unmapped reads, orphans and
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

secondary alignments were removed with SAMtools, and duplicates were marked with
picard-tools before and after merging the library BAMfiles per sample. For neutralomes, we
merged all BAMfiles into one master BAMfile for further processing. A custom script was run to
download data, mark duplicates and filter out reads as above for 20 SGDP BAMfiles (Mallick et
al., 2016). All samples underwent Local Realignment (GATK, v.3.7; (DePristo et al., 2011)) using
a union interval set of 30 samples plus the target sample (RealignerTargetCreator), and a
guidance set of 12 samples for realigning along the target sample with IndelRealigner
(Supplementary Data Table 2). We used ATLAS (v0.9; (Link et al., 2017)) with commit 7c1e6a4,
unless indicated otherwise, and filterSoftClips option to split/merge single-end/paired-end reads
(task=splitMerge) which generated our final processed BAMfiles. See Table 1 and S1 for more
information.
Quantification and Statistical Analysis
Library and Sample Statistics
We used ATLAS (task=BAMDiagnostics and task=depthPerSiteDist) and SAMtools flagstat to
determine read counts, sequencing depth, endogenous DNA-content and further statistics that
are listed in Supplementary Data Table 1 for each library parallel and merged samples.
Molecular Sex Determination
Using the script by (Skoglund et al., 2013) obtained from
https://github.com/pontussk/ry_compute), we determined the molecular sex of individuals by
calculating the ratio of reads aligned to the Y chromosome over the total number of reads
aligned to X and Y. Ratios of 0.075 or higher indicate males and ratios below 0.016 indicate
females (Figure S1C). It was run with default settings as suggested by Skoglund's
documentation.
Ancient DNA Authenticity
The blank controls from milling, extraction, library and capture experiments were measured by
Qubit® Fluorometric quantitation (dsDNA HS assay, Invitrogen, Carlsbad, California, United
States) and on an Agilent 2100 Bioanalyzer (HS DNA, Agilent Technologies, Waldbronn,
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Germany). Mitochondrial capture controls were sequenced alongside the capture sequencing.
The concentration of potential contaminants was never higher than 2.1 ng/ul and the screening
results showed a maximum of 39 aligning reads per blank control (out of a potential share of
80.000 reads). PMD-patterns were checked for all samples to show the characteristic
exponential pattern of ancient DNA (Figure S1B). We estimated contamination on mitochondrial
regions with ContamMix (v1.0; (Fu et al., 2014)) and for male individuals on the X-chromosomal
regions with ANGSD (v0.917-108-g2f9cc4b; (Rasmussen et al., 2011; Korneliussen,
Albrechtsen and Nielsen, 2014)). As the nuclear capture regions do not span the mitochondrial
DNA and only contain few regions on the X-chromosome, we require a minimum of 10X depth
over the mitochondrial genome (to assure correct consensus-calls) combined with a mt/nuc ratio
below 200 (Furtwängler et al., 2018), as well as a minimum of 100 SNPs for ANGSD estimation
to rely on the results of contamination estimation (Nägele et al., 2020). The detailed results can
be found in Supplementary Data Table 1.
mt-DNA analysis
mt-Capture of nuclear capture supernatants
An additional mitochondrial target enrichment experiment on the supernatant of a nuclear target
enrichment experiment - meaning the DNA not hybridized on the nuclear target - yielded a
significantly higher percentage of endogenous reads aligning to the mitochondrial genome than
the conservative mt-capture (t-test 4.0, p=2.3e-04). This could be explained by a higher
percentage of mitochondrial fragments in the hybridisation process as well as a potential lack of
steric hindrance in the hybridisation process as much longer nuclear molecules have been
removed. Yet, the fold-coverage is significantly lower for supernatant captures (t-test -6.5,
p=5.75e-08). This is expected, as several additional purification steps are performed on
supernatant captures, accompanied by a severe loss of molecules. As a supernatant capture
will only be performed after a nuclear capture, and therefore on high-quality samples, it can be
recommended for further experiments to reduce the amount of sample material used.
mt-DNA Haplogroups
In order to determine the mitochondrial haplotypes from nuclear genomic data, majority allele
calls on the MT genome were created with ATLAS (task=call method=majorityBase). The output
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

VCF files were merged and uploaded to the HaploGrep 2.0 Website (Weissensteiner et al.,
2016). The mitochondrial haplotypes for mitochondrial capture experiments were obtained as
described in (Hofmanová, 2016; Hofmanová et al., 2016).
Genotype likelihoods estimations
Post mortem damage patterns
We used ATLAS (task=PMD) to infer position-specific PMD patterns as described in
(Kousathanas et al., 2017) from the tabulated mismatches between the raw reads and the
hs37d5 reference genome. These patterns were mainly inferred independently for all
whole-genome individuals and read groups. However, for neutralomes we used the master
BAMfile to pool all read groups that came from the same sample and had similar PMD pattern
(option poolReadGroups).
Base quality score recalibration
We used ATLAS (task=recal) to recalibrate base quality scores with the method described in
(Kousathanas et al., 2017) using the model qualFuncPosSpecificContext. This reference-free
approach exploits a set of known homozygous sites and is extended to additional covariates
beyond the original quality score, in our case the specific position within the sequencing read
and the nucleotide context. As known homozygous sites, we used 10 million sites highly
conserved among mammals as reflected by high RS-Scores (also called GERP scores; (Cooper
et al., 2005)) calculated across the multiple sequence alignments of 88 mammals and provided
by Ensembl
(http://ftp.ensembl.org/pub/release-96/compara/conservation_scores/88_mammals.gerp_conser
vation_score/gerp_conservation_scores.homo_sapiens.GRCh38.bw). For single-end
sequencing data, the read groups were split into two new read groups during the raw data
processing, so we provided the names of these split read groups to be merged for recalibration
(option poolReadGroups). For neutralomes, we used the master BAMfile to pool all read groups
that were generated with the same sequencing run and lane. This increased the power when
estimating recalibration parameters for all neutralome samples.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Genotype likelihoods
We used ATLAS (task=GLF) to infer genotype likelihoods for all individuals at autosomal neutral
capture 5Mb sites. We considered the PMD and the recalibration parameters previously
estimated. This created GLF files for all our samples at the selected sites.
Downstream quality filtering
Downsampling experiments
For further analysis, we tested the impact of recalibration and post-mortem damage in 12
low-coverage and medium-coverage whole-genome samples. The tests consisted in
downsampling the BAMfiles by probabilities ranging from 1 to 0.05, and estimating their
heterozygosity in windows (ATLAS, task=thetaQC), either ignoring or considering PMD and
recalibration parameters, respectively. For more details, see Figure S2B.
Depth and Heterozygosity Filtering
Based on the downsampling experiments, we estimated heterozygosity once with full data and
50 times using a downsampling probability of 0.5 (ATLAS, task=thetaQC) for every sample. The
PMD and recalibration parameters inferred from full data were used in all heterozygosity
estimations so that we can assess how well the error-rates are recalibrated between the full
data and downsampled versions. We calculated the log ratio between the median sampled
estimate with full data and the median of the 50 downsampling median sampled estimates. For
the neutralome data, we bootstrapped 100 times the genome wide heterozygosity on the
capture neutral regions and took the medians for the full data and the 50 downsampling median
estimates and proceeded to calculate the log ratio. A ratio <0.239 and >-0.239 and a depth of
coverage >1.5x for whole-genome and >4x for neutralome were used as quality filters (Figure
S2A) to ensure comparability among samples and increase the sensitivity of our population
genomic analyses. A total of 54 out of 79 whole-genomes and 56 out of 75 neutralomes passed
these filters (7 overlaps between them) with two neutralome samples being further removed as
mentioned below (Kinship analysis).
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Population genetic analysis
Kinship analysis
We used ATLAS (task = geneticDist) to estimate the euclidean distances between all pairs of
ancient samples that passed our quality control filters, using the GLFs created during the
estimation of genotype likelihoods. We then used a custom R script
(https://bitbucket.org/wegmannlab/atlas/downloads/Relatedness.R) to apply the method of
(Waples, Albrechtsen and Moltke, 2019) to transform these distances into estimates of genetic
relatedness. No relatedness was detected among Danube Gorges samples, but for Dil15 and
Bar15 which were filtered out for downstream analysis. See details in Figure S1D.
Major/minor
We used ATLAS (task = majorMinor) to estimate the major and minor alleles (default
parameters) from our sample-specific GLFs and output the genotype likelihoods for those in a
vcf file, which then was converted to a Beagle format (task=VCFToBeagle) for the population
genetic analysis. This was run for different sets of populations.
PCA
We used PCAngsd (v0.986; (Meisner and Albrechtsen, 2018)) with default parameters
(MAF>=0.05) on different sets of populations to estimate the covariance matrix and perform
Principal Component Analysis (PCA).
Admixture analysis
We used NGSAdmix (Skotte, Korneliussen and Albrechtsen, 2013) with -minMaf 0.05 to infer
admixture proportions for different numbers of clusters (K=2 to K=7). Each K was run 10 times
with a different seed and the best K was estimated by applying the Evanno method (Evanno,
Regnaut and Goudet, 2005). In addition, Entropy (v2.0; (Shastry et al., 2021)) and PCAngsd
were run as well for different Ks to estimate admixture proportions. The deviance information
criterion (DIC) was used for entropy models with K=2 to K=7, lower values of DIC correspond to
better model fit. For PCAngsd, a different eigenvector value was provided each time since the
best K was empirically set to 2 based on PC loadings. The admixture proportions were then
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

compared among the three (Figure S3B). The results were qualitatively the same, in all cases
K=2 is the best fit, and only the proportions from NGSadmix were used for population structure.
Inter population ancestry
We used the second model of Entropy (Shastry et al., 2021) that takes into account the
combination of ancestry states across all loci in diploid individuals. The benefit of this model is
that it allows distinguishing among early generations of admixed individuals (i.e., F1, BC1) when
combined with their global ancestry. We followed the user manual as described in the bitbucket
repository (https://bitbucket.org/buerklelab/mixedploidy-entropy/src/master/vignette_entropy.pdf)
to format the input genotype likelihood data and run Entropy with the right options. The model
was run with five chains simultaneously, where posterior distributions were estimated with
100,000 iterations, sampled every 10th iteration and with 10,000 burn-in.
Intra-inter genetic diversity
Genomic regions of interest
We extracted exons and introns from the human genome annotation file
(http://ftp.ensembl.org/pub/grch37/release-104/gff3/homo_sapiens/Homo_sapiens.GRCh37.87.
gff3.gz) by using a custom R script. We used bedtools (v2.27.1; (Quinlan and Hall, 2010)) to
merge intervals and to subtract the exons regions that were also in the introns. We combined
tracks from build GRCH37 of the UCSC Genome Browser (Kent et al., 2002) to exclude
simple-repeats, segmental duplications, self Chains, regions with high CpG content and the
strict callability mask from 1000G (1000 Genomes Project Consortium et al., 2015). We ended
up with ~50Mb in both regions (introns were downsampled to the same amount of exons sites).
Additionally, we identified 17,737 neutral 1kb autosomal loci based on 37,574 autosomal neutral
regions (Gronau et al., 2011) that were lifted over from hg18 to GRCH37
(https://genome.ucsc.edu/cgi-bin/hgLiftOver). To update and provide an extra layer of
stringency while accounting for the liftover, we used NRE (Arbiza, Zhong and Keinan, 2012)
to mask all sites that do not fall in the 37,574 lifted loci and removed mammalian conserved
noncoding elements plus 100bp each side (PhastConsElements46WayPlacental GRCH37
track; (Pollard et al., 2010)). Regions with recombination rate <0.1 and >10 cM/Mb, nearest
gene distance <0.01 cM, simple repeats, BG selection coefficient <0.85 and not separated at
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

least 50kb from each other were excluded too following (Gronau et al., 2011; Veeramah et al.,
2018). Since the list of 37,574 loci does not include the X chromosome, we applied the filters
from above and (Gronau et al., 2011) on the X chromosome with slight modifications to be less
stringent. We ended up with 5,505 neutral >1kb X loci (~11Mb).
Genome-wide Heterozygosity (θ)
We used ATLAS (task=theta thetaGenomeWide minDepth=2 bootstraps=100) to infer genome
wide heterozygosity with 100 bootstraps on autosomal neutral sites. The nuclear capture data in
fact contain 5Mb of autosomal neutral sites at high depth; however, for ancient whole-genomes,
more neutral sites were necessary for increasing the sensitivity of the estimates. Hence, we
provided the autosomal neutral ~18Mb sites (section “Genomic regions of interest”) for the
whole-genome data and the autosomal neutral 5Mb sites for the nuclear capture data to
estimate single genome-wide estimates for θ.
Heterozygosity ratio (θ1/θ2)
We first tested the statistical power of the heterozygosity ratio model in ATLAS by simulating
genomic data with default parameters (task=simulate) as shown in Figure S4. It was concluded
that a depth >1.5x and a window size >10Mb for both regions of interest provide good
estimates. We then used ATLAS (task=thetaRatio) to estimate genetic diversity between exons
and introns with a prior=0 for all whole-genome samples, and also between X and autosome
neutral sites with a prior=log(¾) for all female whole-genome samples. These genomic regions
are above 10Mb and they were extracted as explained in section “Genomic regions of interest”.
ROHs
We used ATLAS (task=call method=majorityBase) to produce haploid calls on known alleles
(1240K sites; (Mathieson et al., 2015)) while taking PMD and recalibration parameters into
account for the whole-genome samples. We merged our calls with the 1240K reference panel
using a custom script and plink (v1.9; (Chang et al., 2015)) and converted them into
EIGENSTRAT format (convertf -p, https://reich.hms.harvard.edu/software/InputFileFormats). We
then used hapROH (v0.3a4; (Ringbauer, Novembre and Steinrücken, 2021)) with default
parameters to identify runs of homozygosity (ROHs) in whole-genome modern and ancient
samples. The hapROH output files were merged and the ROHs were binned on genetic length
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

ranges (2-5cM, 5-10cM and >10cM) to calculate the total length and number of ROHs for each
bin.
Phenotyping prediction
HIrisPlex analysis
We used ATLAS (taks=call method=MLE) to produce diploid calls on a set of 41 SNPs
(Chaitanya et al., 2018) while taking PMD and recalibration parameters into account for our
nuclear genomic data. We then used the HIrisPlex-S webtool (Walsh et al., 2014, 2017;
Chaitanya et al., 2018) to predict possible pigmentation phenotypes. We therefore created a csv
file by converting genotypes to allele counts for all alleles of interest needed for the prediction
and uploaded it at https://hirisplex.erasmusmc.nl/. We obtained posterior probabilities for four
phenotypes, each of them consisting of different categories that sum up to one: skin
pigmentation (dark, intermediate, pale and very pale), hair pigmentation (black, brown, blond
and red), eye pigmentation (brown, intermediate and blue) and hair shade (dark and light)
(Figure S5). In order to reduce noise, we turned phenotypes with more than two categories into
a binary form by summing up the corresponding probabilities falling in the darker (dark,
intermediate, black and brown) or lighter (pale, very pale, blonde, red and blue) spectrum. We
then proceeded to train a LDA model based on the summarized probabilities as input and using
a set of Meso European-like and Neo Aegean-like samples as a grouping factor, excluding
admixed samples and those from a different geographical context (see Figure 4).
Supplemental Information
Document S1: Supporting Figures S1–S5 and Table S1.
Extended data tables:
Supplementary Data Table 1: Sample processing including detailed information on library
preparation, read- and sample statistics, isotope- haplogroup and contamination
information for sequenced libraries and ancient reference samples.
Supplementary Data Table 2: List of samples used in the interval-set and guidance set for local
in-del realignment.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Acknowledgements
This work was supported by Swiss National Science Foundation (31003A_173062 and
310030_200420 to DW, 31003A_182577 to MC), Marie Skłodowska-Curie actions ITN ‘‘BEAN’’.
ZH was further supported by the European Research Council (856453 ERC-2019-SyG), Czech
Grant Agency (GACR 21-17092X) and EMBO Long-Term Fellowship (ALTF 445-2017). CB was
supported by the Fyssen Foundation (Fondation Fyssen, post-doctoral research grant).
We also thank the IBU cluster and sequencing facility of the University of Bern.
GENterprise, SAPM Munich,
Declaration of Interests
The authors declare no competing interests.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Figures and Tables
Figure 1 - Spatial and temporal distribution of the ancient nuclear genomic samples sequenced in this study.
(A) Location of archeological sites of Danube Gorges, North Aegean and Central, North and East Europe.
(B) Observed summed probability distribution (SPD, solid black line) of calibrated radiocarbon dates from Danube Gorges along with
a simulation envelope of the fitted model (shaded gray area). Adapted after (de Becdelièvre et al., 2021).
(C) Chronological distribution of whole-genome and nuclear capture data sequenced in this study (see details in Table 1). Both
symbols and labels are shown for each directly-14C or approximately dated sample, except for the ones who were filtered out for
downstream analysis (labels shown in gray) or whose dates were not available.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Figure 2.PCA and Admixture of Meso European-like and Neo Aegean-like individuals.
(A) PCA that includes Neo Aegean-like individuals in red-like colors and Meso European-like individuals in blue-like colors. Admixed
individuals are shown in purple.
(B) PCA using only Meso European-like populations.
(C) PCA using only Neo Aegean-like populations.
(D) Entropy analysis with the same dataset as in (A) but excluding Lec2, Motala12 and North-Eastern European hunter-gatherers.
The y axis denotes the proportion of the genome in which one gene copy comes from Neo Aegean-like ancestry and the other from
Meso European-like ancestry; the position of our Lepenski Vir individuals are associated with backcross lineages. The x axis
denotes the proportion for Neo Aegean-like genome ancestry estimated by Entropy. Inner triangles represent the samples of the
posterior distribution and the curved dotted line is based on Hardy-Weinberg equilibrium (HWE).
(E) NGSadmix results for K=2 using the set from Entropy (D); blue represents the Meso European-like ancestry and red the Neo
Aegean-like ancestry.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Figure 3 -Intra and inter genetic diversity of ancient and modern samples.
(A) Heterozygosity (θ) inferred genome-wide along the 5Mb neutral capture sites for neutralome samples. The dashed line indicates
the expected value, the points represents the MLE estimate and the error bar the min/max of 100 bootstrap values.
(B) Heterozygosity (θ) inferred genome-wide along the ~18Mb customized neutral genome sites for whole-genome samples.
(C) Exon/Intron heterozygosity (θ) ratio inferred for whole-genome samples. The dashed lines indicate the median of each group
ratio value.
(D) Heterozygosity (θ) ratio estimated between X neutral sites (~11Mb) and autosomal neutral genome sites (~18Mb) for
whole-genome female samples. The black dashed line indicates the expected X/A ratio value (3/4), the other colored dashed lines
represent the median of each group ratio value.
(E) Runs of homozygosity (ROHs) for all whole-genome samples; the fill color represents the broad cultural periods, the genetic
lengths were binned in 2-5 cM, 5-10 cM and >10 cM, and the observed counts are written inside each bar category.
Figure 4 - Posterior probabilities of HIrisPlex data based on a Linear Discriminant model (LDA).
The model was trained on a set of samples (indicated by lines below x-axis) representing the Meso European-like and Neo
Aegean-like populations, where the first discriminant function is a linear combination of four different phenotypes: skin pigmentation,
hair pigmentation, eye pigmentation and hair shade. To reduce noise, we turned each phenotype into a binary configuration (see
Methods). The trained model was then used to obtain posterior assignment probabilities for all samples, including those not used for
training (the admixed samples and those from a different geographical context shown on the right).
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Table 1 -Archaeological and genetic information of all sequencing data either produced in this study or
available for the Danube Gorges.
Sample ID
Site
Phasea
Archeological/
Burial ID
Age
(cal. BC)
Mean
Depth
Genetic
sex
Ry
MT-Haplogroup
Danube Gorges whole genomes
VLASA16*
Vlasac
LM
VL 53
6,650 ± 174
0.99
XX
0.0012
U5a1c
VLASA37
Vlasac
LM
VL 24
6,614 ± 153
5.06
XY
0.0829
K1f
LEPE45
Lepenski Vir
TEN
LV 91
6,492 ± 96
4.03
XY
0.0754
U5a2d
LEPE46*,#
Lepenski Vir
TEN
LV 93
6,120 ± 102
0.74
XX
0.0014
H
LEPE51
Lepenski Vir
EM
LV 68
7,756 ± 184
3.74
XX
0.0004
U4a2
LEPE53*,#
Lepenski Vir
TEN
LV 27/a
6,111.5 ± 104.5
0.91
XX
0.0013
U5b2c1
Danube Gorges neutralomes
VLASA4
Vlasac
LM
VL 18/a
~7,400-6,200
85.82
XY
0.0815
U5a1c
VLASA10
Vlasac
LM
VL 41
~7,400-6,200
41.74
XY
0.0978
K1f
VLASA32#
Vlasac
LM
VL 16
7,604.5 ± 136.5
43.08
XY
0.08
U5a2a
VLASA41
Vlasac
LM
VL 30
~7,400-6,200
55.97
XX
0.0005
U5b2b
VLASA44
Vlasac
LM
VL 47
~7,400-6,200
64.96
XY
0.0819
U5b2b
LEPE18
Lepenski Vir
TEN
LV 27/d
6,126 ± 100
78.71
XY
0.0822
U5a2
LEPE39
Lepenski Vir
TEN
LV 82
6,075 ± 125
23.26
XY
0.0798
T2e
LEPE46#
Lepenski Vir
TEN
LV 93
6,120 ± 102
52.09
XX
0.0005
H
LEPE52#
Lepenski Vir
E-MN
LV 73
5,812 ± 119
61.44
XY
0.082
H2a
LEPE53#
Lepenski Vir
TEN
LV 27/a
6,111.5 ± 104.5
36.39
XX
0.0004
U5b2c1
Danube Gorges whole genomes - published data (Marchi et al., 2022)
VLASA7
Vlasac
LM
VL 31
6,552 ± 212
15.21
XY
0.0754
U5a2a
VLASA32#
Vlasac
LM
VL 16
7,604.5 ± 136.5
12.66
XY
0.0753
U5a2a
LEPE48
Lepenski Vir
TEN
LV 122
5,939.5 ± 72.5
10.93
XY
0.0755
K1a1
LEPE52#
Lepenski Vir
E-MN
LV 73
5,812 ± 119
12.38
XY
0.0751
H3
non-Danube Gorges nuclear capture genomes
Lec2
Lesnika Cave,
Crimea Ai-Petri
EM
Lsa 031
-
12.906
XY
0.0952
U5a1
Min2
Minino
M
Minino I, 11
-
16.212
XX
0.0006
U4
Min3
Minino
M
Minino II, III
7,472 ± 52
58.044
XY
0.089
U4a1
Min5*
Minino
M
Minino I, 3
8,580 ± 160
3.365
XY
0.077
U4
Min8*
Minino
M
Minino I, 5
6,125 ± 325
0.812
XY
0.0826
U4a1
Min10*
Minino
M
Minino I, 20
5,125 ± 525
3.917
XX
0.0006
U4d
Min11
Minino
M
Minino II, V
8,671 ± 48
8,092 ± 94
39.340
XY
0.1181
U4a1
ZVEJ317
Zvejnieki
M
ZV317
3,890 ± 67
28.716
XY
0.1415
U4a1
ZVEJ39*
Zvejnieki
M
ZV39
5,681 ± 36
1.707
XY
0.0788
U5a2c
ZVEJ76*
Zvejnieki
M
ZV76
5,802 ± 73
1.164
XX
0.0006
U5a2
ZVEJ122*
Zvejnieki
M
ZV122
5,383 ± 68
0.971
XY
0.0889
U5a2d
ZVEJ162
Zvejnieki
M
ZV162
4,470 ± 72
17.264
XY
0.1132
U4a1
ZVEJ170*
Zvejnieki
M
ZV170
7,182 ± 107
0.251
XY
0.0793
U5a2d
Bla20
Blätterhöhle
M
BH 04/174
8,652 ± 58
6.558
XX
0.0005
U5a2c
Fre3*
Groß Fredenwalde
M
Ind 1
5,800 ± 400
0.315
XY
0.0762
H2a2a
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Gr1*
Criewen
M
Criewen
1961:82 Grab1
4,770 ± 40
5.104
XY
0.1072
U5b2a2
Gr2
Criewen
M
1961: 82/7
4,600 ± 60
21.134
XX
0.0005
U4b1b1
GrO1*
Große Ofnethöhle
LM
Ind 12, Ktg nr
SV 001/12
(2485)
-
11.351
XY
0.0795
U5b1d1
AKT6
Aktopraklık
EN
H17/50.1
5,584 ± 49
32.182
XY
0.0815
T2
AKT16#
Aktopraklık
EN
D89/14.1
6,547 ± 87
39.813
XX
0.0019
K1a3
AKT18*
Aktopraklık
EN
D89/17.1
6,431 ± 38
3.76
XX
0.0006
H2a2a
AKT20
Aktopraklık
EN
E89/9.3
6,456 ± 37
20.96
XY
0.0775
J2b1
AKT26
Aktopraklık
EN
D90/4.4
~6,500 - 6,000
18.77
XX
0.0004
J2a
Bar11
Barcın
EN
M11/93
~6,600 - 6,000
24.359
XY
0.0801
X
Bar15*,1
Barcın
EN
M10/115
6,131 ± 82
25.558
XX
0.0004
K1a2
Bar16
Barcın
EN
L10/187
6,159 ± 74
22.579
XX
0.0004
K1a1
Bar20
Barcın
EN
M11S/401
6,348 ± 90
13.561
XX
0.0006
W5
Bar32
Barcın
EN
L11/604
6,329 ± 90
13.688
XX
0.002
K1a2
Asp1
Asparn-Schletz
EN (LBK)
93/7//93/8, Nr.
4451/Schnitt
22
5,250 ± 250
21.639
XY
0.0945
K1a
Asp2
Asparn-Schletz
EN (LBK)
Ind 10, S6/
282
5,250 ± 250
14.771
XY
0.085
N1a1a1a2
Asp3
Asparn-Schletz
EN (LBK)
Nr.
4455/Schnitt
22 1993/19
5,250 ± 250
33.465
XX
0.0018
J1c17
Asp4
Asparn-Schletz
EN (LBK)
2490, Ind 7,
7//restl.
Schädel 5/02
5,250 ± 250
29.061
XX
0.0021
T2b13b
Asp6#
Asparn-Schletz
EN (LBK)
Ind 44 (646
Part 152,
Schnitt 10
LM70
5,525 ± 50
23.815
XY
0.0928
U5a1c1
Asp8
Asparn-Schletz
EN (LBK)
Ind 24, 374
S7/LM72 Grab
1
5,250 ± 250
76.578
XX
0.0019
X2b
Asp10
Asparn-Schletz
EN (LBK)
Ind 6
5,250 ± 250
48.450
XY
0.1047
K1a
Klein1
Kleinhadersdorf
EN (LBK)
25,939
~5,950 - 5,150
85.4511
XX
0.0018
T2b
Klein2
Kleinhadersdorf
EN (LBK)
25,941
~5,950 - 5,150
44.729
XX
0.0019
H
Klein3
Kleinhadersdorf
EN (LBK)
25,945
~5,950 - 5,150
30.345
XX
0.002
J1c2
Klein4
Kleinhadersdorf
EN (LBK)
25,926
~5,950 - 5,150
49.749
XX
0.0019
T2b
Klein5
Kleinhadersdorf
EN (LBK)
25,925
~5,950 - 5,150
44.777
XY
0.0957
N1a1a1a3
Klein8
Kleinhadersdorf
EN (LBK)
25,923
~5,950 - 5,150
41.197
XX
0.0006
U5b
Klein10
Kleinhadersdorf
EN (LBK)
25,936
~5,950 - 5,150
49.393
XX
0.0021
N1a1a1
Dil15*,1
Dillingen-Steinheim
EN (LBK)
Grab 23,
Befund 24
5,116 ± 118
28.668
XY
0.0918
J1c
Dil16#,1
Dillingen-Steinheim
EN (LBK)
Grave 24,
Befund 24
5,116 ± 118
25.186
XY
0.0952
J1c
Ess7#
EssenbachAmmerbreit
EN (LBK)
Grave 2
4,975 ± 75
12.2172
XY
0.0793
U5b2c1
Herx#
Herxheim
EN (LBK)
281-19-6
5,078 ± 85
9.23
XX
0.0004
K1a4a1i
Herx2
Herxheim
EN (LBK)
282-126-7
~5,000
55.87
XY
0.1266
J2b1
Herx3
Herxheim
EN (LBK)
282-13-7
~5,000
16.14
XX
0.0006
W1+119
Herx4
Herxheim
EN (LBK)
282-23-1
~5,000
115.12
XY
0.1039
K1a
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Herx5
Herxheim
EN (LBK)
282-94-11
~5,000
30.32
XY
0.087
HV+16311
Herx6
Herxheim
EN (LBK)
7034-12
~5,000
36.42
XX
0.0005
U3
Herx7
Herxheim
EN (LBK)
282-104-4
∼5,000
34.39
XY
0.0787
K1a1a
Herx8
Herxheim
EN (LBK)
282-126-16
∼5,000
28.77
XX
0.0009
T2e
Herx9
Herxheim
EN (LBK)
282-88-2
∼5,000
57.05
XX
0.0006
K1a+150
Bla1*
Blätterhöhle
MN
BH 04/011
3,508 ± 102
0.689
XY
0.0981
H1bm
Bla10*
Blätterhöhle
MN
BH 04/012 B
3,418 ± 63
3.528
XY
0.0781
H2a2a
Bla13*
Blätterhöhle
MN
BH 04/034
3,513 ± 102
4.549
XX
0.0006
H2a2a
Bla15*
Blätterhöhle
MN
BH 04/041
3,571 ± 47
0.509
XY
0.0845
H2a2a1
Bla17*
Blätterhöhle
MN
BH 04/032
3,681 ± 19
1.663
XY
0.1036
H1ba
Bla28*
Blätterhöhle
MN
BV 06G5d/Po.
49.2
3,196 ± 103
0.4235
XY
0.0813
J1c1
Bla29*
Blätterhöhle
MN
BH 07 I4/Po 1
3,020 ± 61
1.282
XX
0.0009
H2a2a1
Bla75*
Blätterhöhle
MN
BH14
Qu7c/Bla75
-
0.303
~XY
0.3047
-
Bla32
Blätterhöhle
MN
BH 04/072
-
47.839
XX
0.0005
H5
Bla45
Blätterhöhle
MN
BH14
Qu7c/Bla45
3,616 ± 56
3,922 ± 60
32.4236
XY
0.095
U5b2b
Bla59
Blätterhöhle
MN
BH14
Qu7c/Bla59
3,869 ± 59
10.879
XY
0.0943
U5b2a2
aM, Mesolithic; EM, Early Mesolithic; LM, Late Mesolithic; TEN, Transformation/Early Neolithic; EN, Early Neolithic; E-MN, Early-Middle
Neolithic; MN, Middle Neolithic.
* Samples were discarded for downstream analysis.
#Data for both whole genome and capture nuclear genome.
1Related samples, Bar8-Bar15; DIl15-Dil16.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

References
1000 Genomes Project Consortium et al. (2015) ‘A global reference for human genetic
variation’, Nature, 526(7571), pp. 68–74.
Adams, R.L. and King, S.M. (2011) Residential Burial: A Multiregional Exploration. Wiley.
Alpaslan Roodenberg, S., Gerritsen, F.A. and Özbal, R. (2013) ‘Neolithic burials from Barcin
Höyük: The 2007-2012 excavation seasons’, Anatolica, 39, pp. 93–111.
Amster, G. et al. (2020) ‘Changes in life history and population size can explain the relative
neutral diversity levels on X and autosomes in extant human populations’, Proceedings of the
National Academy of Sciences of the United States of America , 117(33), pp. 20063–20069.
Amster, G. and Sella, G. (2020) ‘Life History Effects on Neutral Diversity Levels of Autosomes
and Sex Chromosomes’, Genetics, 215(4), pp. 1133–1142.
Arbiza, L., Zhong, E. and Keinan, A. (2012) ‘NRE: a tool for exploring neutral loci in the human
genome’, BMC bioinformatics, 13, p. 301.
Arbuckle, B.S. et al. (2014) ‘Data sharing reveals complexity in the westward spread of domestic
animals across Neolithic Turkey’, PloS one, 9(6), p. e99845.
Bartosiewicz, L. et al. (2001) ‘New data on the prehistoric fauna of the Iron Gates: a case study
from Schela Cladovei, Romania’, From the Mesolithic to the Neolithic, pp. 15–22.
de Becdelièvre, C. et al. (2020) ‘Direct insight into dietary adaptations and the individual
experience of Neolithisation: comparing subsistence, provenance and ancestry of Early
Neolithic humans from the Danube Gorges c. 6200--5500 cal BC’, Farmers at the Frontier: A
Pan European Perspective on Neolithisation, p. 45.
de Becdelièvre, C. (2020) Eco-ethology of prehistoric populations living in the Danube Gorges c.
10.000-5500 BC. Dietary strategies, feeding practices and nutritional choices over the
Mesolithic and Neolithic Transformations. Ph.D. University of Belgrade.
de Becdelièvre, C. et al. (2021) ‘Palaeodemography of the Foraging to Farming Transition:
insights from the Danube Gorges Mesolithic-Neolithic transformations’, in Degioanni, A., Estelle,
H., and Naji, S. (eds) Journey of a committed paleodemographer. Farewell to Jean-Pierre
Bocquet-Appel. coll. Préhistoire de la méditerranée, Aix-en-Provence: Presses Universitaires de
Provence, pp. 7–14.
Blagojević, T. et al. (2017) ‘Early Neolithic population dynamics in the eastern Balkans and the
Great Hungarian Plain’, Documenta Praehistorica, 44, pp. 18–33.
Bocquet-Appel, J.-P. and Bar-Yosef, O. (eds) (2008) The Neolithic Demographic Transition and
its Consequences. Springer, Dordrecht.
Bogdanović, M. (2008) Grivac: Settlements of Proto-Starčevo and Vinča Culture. Center for
Scientific Research of Serbian Academy of Sciences and Arts.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Bollongino, R. et al. (2013) ‘2000 years of parallel societies in Stone Age Central Europe’,
Science, 342(6157), pp. 479–481.
Bonsall, C. et al. (1997) ‘Mesolithic and early Neolithic in the Iron Gates: a palaeodietary
perspective’, Journal of European Archaeology, 5(1), pp. 50–92.
Bonsall, C. et al. (2000) ‘Stable isotopes, radiocarbon and the Mesolithic-Neolithic transition in
the Iron Gates’, Documenta Praehistorica, 27(7), p. 119:132.
Borić, D. (2002) ‘The Lepenski Vir conundrum: reinterpretation of the Mesolithic and Neolithic
sequences in the Danube Gorges’, Antiquity, 76(294), pp. 1026–1039.
Borić, D. (2009) ‘Absolute Dating of Metallurgical Innovations in the Vinča Culture of the
Balkans’, in Kienlin, T.L. and Ben W, R. (eds) Metals and Societies. Studies in honour of
Barbara S. Ottaway. Rudolf Habelt GMBH, Bohn, pp. 191–245.
Borić, D. (2011) ‘Adaptations and transformations of the Danube Gorges foragers (c.
13,000–5500 cal. BC): an overview’, in Beginnings – New Research in the Appearance of the
Neolithic between Northwest Anatolia and the Carpathian Basin. Papers of the International
Workshop 8th - 9th April 2009, Istanbul. Organized by Dan Ciobotaru, Barbara Horejs und
Raiko Krauß. Beginnings - new research in the appearance of the Neolithic between Northwest
Anatolia and the Carpathian Basin : papers of the international workshop, Rahden/Westf.:
Verlag Marie Leidorf GmbH (Menschen – Kulturen – Traditionen: Studien aus den
Forschungsclustern des Deutschen Archäologischen Instituts; ForschungsCluster 1), pp.
157–203.
Borić, D. et al. (2014) ‘Late Mesolithic lifeways and deathways at Vlasac (Serbia)’, Journal of
Field Archaeology, 39(1), pp. 4–31.
Borić, D. (2015) ‘The End of the Vinča World: Modelling the Neolithic to Copper Age Transition
and the Notion of Archaeological Culture’, in Hansen, S. et al. (eds) Neolithic and Copper Age
Between the Carpathians and the Aegean Sea: Chronologies and Technologies from the 6th to
the 4th Millennium BCE.International Workshop Budapest, Verlag Marie Leidorf (Archäologie in
Eurasien ), pp. 157–217.
Borić, D. (2016) Deathways at Lepenski Vir: Patterns in Mortuary Practice. Serbian
Archaeological Society.
Borić, D. (2021) ‘The Danube Gorges Mesolithic: The first fifty years’, in Borić, D., Antonović, D.,
and Mihailović, B. (eds) Foraging Assemblages. Serbian Archaeological Society, Belgrade,
Republic of Serbia. The Italian Academy for Advanced Studies in America, Columbia University,
New York, USA.
Borić, D., French, C. and Dimitrijević, V. (2008) ‘Vlasac revisited: formation processes,
stratigraphy and dating’, Documenta Praehistorica, 35, pp. 261–287.
Borić, D. and Miracle, P. (2004) ‘Mesolithic and neolithic (dis)continuities in the Danube gorges:
New ams dates from Padina and hajducka vodenica (Serbia)’, Oxford Journal of Archaelogy,
23(4), pp. 341–371.
Borić, D. and Price, T.D. (2013) ‘Strontium isotopes document greater human mobility at the
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

start of the Balkan Neolithic’, Proceedings of the National Academy of Sciences of the United
States of America, 110(9), pp. 3298–3303.
Borić, D. and Stefanović, S. (2004) ‘Birth and death: infant burials from Vlasac and Lepenski
Vir’, Antiquity, 78(301), pp. 526–546.
Boroneanţ, V. (1999) ‘The Mesolithic Habitation Complexes in the Balkans and the Danube
Basin’, Living Past, 1. Available at:
http://www.cimec.ro/Arheologie/livingpast/nr1/boroneant/art.htm.
Boroneanţ, V. (2011) ‘A Suggested Chronology for the Iron Gates Mesolithic’, Buletinul Muzeului
Judeţean Teleorman, (3), pp. 21–39.
Boulestin, B. et al. (2009) ‘Mass cannibalism in the Linear Pottery Culture at Herxheim
(Palatinate, Germany)’, Antiquity, 83(322), pp. 968–982.
Boulestin, B. (2015) ‘Cannibalism in the linear pottery culture: The human remains from
Herxheim’, Cannibalism in the Linear Pottery Culture, pp. 1–152.
Boulestin, B. and Coupey, A.-S. (2015) Cannibalism in the linear pottery culture: The human
remains from herxheim. Oxford, England: Archaeopress Archaeology.
Bramanti, B. et al. (2009) ‘Genetic discontinuity between local hunter-gatherers and central
Europe’s first farmers’, Science, 326(5949), pp. 137–140.
Brami, M.N. (2017) The Diffusion of Neolithic Practices from Anatolia to Europe: A Contextual
Study of Residential Construction, 8,500-5,500 BC Cal. BAR Publishing.
Brink-Kloke, H. (1990) ‘Das linienbandkeramische Gräberfeld von Essenbach-Ammerbreite,
Ldkr. Landshut, Niederbayern’, Germania, 68(2), pp. 427–481.
Broushaki, F. et al. (2016) ‘Early Neolithic genomes from the eastern Fertile Crescent’, Science,
353(6298), pp. 499–503.
Budd, C. et al. (2013) ‘Stable isotope analysis of Neolithic and Chalcolithic populations from
Aktopraklık, northern Anatolia’, Journal of Archaeological Science, pp. 860–867.
doi:10.1016/j.jas.2012.09.011.
Chaitanya, L. et al. (2018) ‘The HIrisPlex-S system for eye, hair and skin colour prediction from
DNA: Introduction and forensic developmental validation’, Forensic science international.
Genetics, 35, pp. 123–135.
Chang, C.C. et al. (2015) ‘Second-generation PLINK: rising to the challenge of larger and richer
datasets’, GigaScience, 4, p. 7.
Cooper, G.M. et al. (2005) ‘Distribution and intensity of constraint in mammalian genomic
sequence’, Genome research, 15(7), pp. 901–913.
Cristiani, E. et al. (2016) ‘Dental calculus reveals Mesolithic foragers in the Balkans consumed
domesticated plant foods’, Proceedings of the National Academy of Sciences of the United
States of America, 113(37), pp. 10298–10303.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

DePristo, M.A. et al. (2011) ‘A framework for variation discovery and genotyping using
next-generation DNA sequencing data’, Nature genetics, 43(5), pp. 491–498.
Dimitrijević, V. (2014) ‘The provenance and use of fossil scaphopod shells at the Late
Neolithic/Eneolithic site Vinča – Belo Brdo, Serbia’, in Szabó, K. et al. (eds) Archaeomalacology:
Shells in the Archaeological Record. Archaeopress Publishers of British Archaeological Reports
(BAR International Series 2666).
Dimitrijević, V., Živaljević, I. and Stefanović, S. (2016) ‘Becoming sedentary? The seasonality of
food resource exploitation in the Mesolithic-Neolithic Danube Gorges’, Documenta
Praehistorica, pp. 103–122. doi:10.4312/dp.43.4.
Eriksson, G., Lõugas, L. and Zagorska, I. (2003) ‘Stone Age hunter–fisher–gatherers at
Zvejnieki, northern Latvia: Radiocarbon, stable isotope and archaeozoology data’, Before
Farming: The Archaeology and Anthropology of Hunter-Gatherers, pp. 1–25.
Evanno, G., Regnaut, S. and Goudet, J. (2005) ‘Detecting the number of clusters of individuals
using the software STRUCTURE: a simulation study’, Molecular ecology, 14(8), pp. 2611–2620.
Feldman, M. et al. (2019) ‘Late Pleistocene human genome suggests a local origin for the first
farmers of central Anatolia’, Nature communications, 10(1), p. 1218.
Fernández-López de Pablo, J. et al. (2019) ‘Palaeodemographic modelling supports a
population bottleneck during the Pleistocene-Holocene transition in Iberia’, Nature
communications, 10(1), p. 1872.
Filipović, D. and Obradović, Đ. (2013) ‘Archaeobotany at Neolithic sites in Serbia: a critical
overview of the methods and results’, Bioarheologija na Balkanu: bilans i perspektive, pp.
25–55.
Fu, Q. et al. (2014) ‘Genome sequence of a 45,000-year-old modern human from western
Siberia’, Nature, 514(7523), pp. 445–449.
Fu, Q. et al. (2016) ‘The genetic history of Ice Age Europe’, Nature, 534(7606), pp. 200–205.
Furtwängler, A. et al. (2018) ‘Ratio of mitochondrial to nuclear DNA affects contamination
estimates in ancient DNA analysis’, Scientific reports, 8(1), p. 14075.
Gamba, C. et al. (2014) ‘Genome flux and stasis in a five millennium transect of European
prehistory’, Nature communications, 5, p. 5257.
Gamble, C. et al. (2004) ‘Climate change and evolving human diversity in Europe during the last
glacial’, Philosophical transactions of the Royal Society of London. Series B, Biological
sciences, 359(1442), pp. 243–53; discussion 253–4.
Garašanin, M. and Radovanović, I. (2001) ‘A pot in house 54 at Lepenski Vir I’, Antiquity, pp.
118–125. doi:10.1017/s0003598x00052819.
Gazave, E. et al. (2013) ‘Population growth inflates the per-individual number of deleterious
mutations and reduces their mean effect’, Genetics, 195(3), pp. 969–978.
Geisler, H. and Wetzel, G. (1999) ‘Mittelsteinzeitliche und mittelalterliche Bestattungen vom
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

“Rollmannsberg” bei Criewen, Lkr : Uckermark’, in Cziesla, E. and Kersting, T. (eds) Den Bogen
spannen ... Festschrift für Bernhard Gramsch. Beier & Beran (Beiträge zur Ur- und
Frühgeschichte Mitteleuropas), pp. 259–280.
Gerritsen, F.A., Özbal, R. and Thissen, L.C. (2013) ‘The earliest neolithic levels at Barcin Höyük,
Northwestern Turkey’, Anatolica, 39, pp. 53–92.
Gnirke, A. et al. (2009) ‘Solution hybrid selection with ultra-long oligonucleotides for massively
parallel targeted sequencing’, Nature biotechnology, 27(2), pp. 182–189.
González-Fortes, G. et al. (2017) ‘Paleogenomic Evidence for Multi-generational Mixing
between Neolithic Farmers and Mesolithic Hunter-Gatherers in the Lower Danube Basin’,
Current biology: CB, 27(12), pp. 1801–1810.e10.
Gronau, I. et al. (2011) ‘Bayesian inference of ancient human demography from individual
genome sequences’, Nature genetics, 43(10), pp. 1031–1034.
Günther, T. et al. (2018) ‘Population genomics of Mesolithic Scandinavia: Investigating early
postglacial migration routes and high-latitude adaptation’, PLoS biology, 16(1), p. e2003703.
Haack, F. (2016) ‘The early Neolithic ditched enclosure of Herxheim - architecture, fill formation
processes and service life’, in Zeeb-Lanz, A. (ed.) Ritualised Destruction in the Early Neolithic -
The Exceptional Site of Herxheim (Palatinate, Germany). Speyer: Generaldirektion Kulturelles
Erbe Rheinland-Pfalz (Forschungen zur Pfälzischen Archäologie 8.1), pp. 19–118.
Haak, W. et al. (2015) ‘Massive migration from the steppe is a source for Indo-European
languages in Europe’. doi:10.1101/013433.
Hofmanová, Z. et al. (2016) ‘Early farmers from across Europe directly descended from
Neolithic Aegeans’, Proceedings of the National Academy of Sciences of the United States of
America, 113(25), pp. 6886–6891.
Hofmanová, Z. (2016) Palaeogenomic and biostatistical analysis of ancient DNA data from
Mesolithic and Neolithic skeletal remains. Edited by J. Burger. PhD. Johannes Gutenberg
University Mainz.
Jones, E.R. et al. (2015) ‘Upper Palaeolithic genomes reveal deep roots of modern Eurasians’,
Nature communications, 6, p. 8912.
Jones, E.R. et al. (2017) ‘The Neolithic Transition in the Baltic Was Not Driven by Admixture
with Early European Farmers’, Current biology: CB, 27(4), pp. 576–582.
Jovanović, J. (2017) ‘The diet and health status of the early neolitihic communities of the Central
Balkans (6200-5200 BC)’, Универзитет у Београду [Preprint]. Available at:
https://nardus.mpn.gov.rs/handle/123456789/8238.
Jovanović, J. et al. (2019) ‘Last hunters–first farmers: new insight into subsistence strategies in
the Central Balkans through multi-isotopic analysis’, Archaeological and anthropological
sciences, 11(7), pp. 3279–3298.
Jovanović, J. et al. (2021) ‘Microbotanical evidence for the spread of cereal use during the
Mesolithic-Neolithic transition in the Southeastern Europe (Danube Gorges): Data from dental
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

calculus analysis’, Journal of archaeological science, 125, p. 105288.
Ju, D. and Mathieson, I. (2021) ‘The evolution of skin pigmentation-associated variation in West
Eurasia’, Proceedings of the National Academy of Sciences of the United States of America,
118(1). doi:10.1073/pnas.2009227118.
Karul, N. and Avcı, M.B. (2011) ‘Neolithic Communities in the Eastern Marmara Region:
Aktopraklık C’, Anatolica, 37, pp. 1–15.
Karul, N. and Avcı, M.B. (2013) ‘Aktopraklık’, in Özdoğan, M., Başgelen, N., and Kuniholm, P.
(eds) The Neolithic in Turkey: New Excavations and New Research. Northwestern Turkey and
Istanbul. İstanbul: Arkeoloji ve Sanat Yayınları, pp. 45–68.
Katoh, K. et al. (2002) ‘MAFFT: a novel method for rapid multiple sequence alignment based on
fast Fourier transform’, Nucleic acids research, 30(14), pp. 3059–3066.
Kent, W.J. et al. (2002) ‘The Human Genome Browser at UCSC’, Genome Research, pp.
996–1006. doi:10.1101/gr.229102.
Kircher, M., Sawyer, S. and Meyer, M. (2012) ‘Double indexing overcomes inaccuracies in
multiplex sequencing on the Illumina platform’, Nucleic acids research, 40(1), p. e3.
Kılınç, G.M. et al. (2016) ‘The Demographic Development of the First Farmers in Anatolia’,
Current biology: CB, 26(19), pp. 2659–2666.
Korneliussen, T.S., Albrechtsen, A. and Nielsen, R. (2014) ‘ANGSD: Analysis of Next
Generation Sequencing Data’, BMC bioinformatics, 15, p. 356.
Koster, J. and Rahmann, S. (2012) ‘Snakemake--a scalable bioinformatics workflow engine’,
Bioinformatics, pp. 2520–2522. doi:10.1093/bioinformatics/bts480.
Kotula, A., Piezonka, H. and Terberger, T. (2020) ‘The Mesolithic Cemetery of Groß
Fredenwalde (NE Germany) and its Cultural Affiliations’, Lietuvos Archeologija, 46, pp. 65–84.
Kousathanas, A. et al. (2017) ‘Inferring Heterozygosity from Ancient and Low Coverage
Genomes’, Genetics, 205(1), pp. 317–332.
Larsson, L. et al. (2017) ‘New Aspects of the Mesolithic-Neolithic Cemeteries and Settlement at
Zvejnieki, Northern Latvia’, Acta Archaeologica, 88(1), pp. 57–93.
Lazăr, C. et al. (2008) ‘The Eneolithic Cemetery from Sultana-Malu Roşu (Călăraşi county,
Romania)’, Studii de Preistorie, 5, pp. 131–152.
Lazaridis, I. et al. (2014) ‘Ancient human genomes suggest three ancestral populations for
present-day Europeans’, Nature, 513(7518), pp. 409–413.
Li, H. et al. (2009) ‘The Sequence Alignment/Map format and SAMtools’, Bioinformatics , 25(16),
pp. 2078–2079.
Li, H. (2013) ‘Aligning sequence reads, clone sequences and assembly contigs with
BWA-MEM’, arXiv:1303.3997v2. Broad Institute of Harvard and MIT. Available at:
https://arxiv.org/abs/1303.3997.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Link, V. et al. (2017) ‘ATLAS: Analysis Tools for Low-depth and Ancient Samples’, bioRxiv.
doi:10.1101/105346.
Lipson, M. et al. (2017) ‘Parallel palaeogenomic transects reveal complex genetic history of
early European farmers’, Nature, 551(7680), pp. 368–372.
Lüning, J. (2000) Steinzeitliche Bauern in Deutschland: die Landwirtschaft im Neolithikum. R.
Habelt.
Macintosh, A.A., Pinhasi, R. and Stock, J.T. (2016) ‘Early Life Conditions and Physiological
Stress following the Transition to Farming in Central/Southeast Europe: Skeletal Growth
Impairment and 6000 Years of Gradual Recovery’, PloS one, 11(2), p. e0148468.
Mallick, S. et al. (2016) ‘The Simons Genome Diversity Project: 300 genomes from 142 diverse
populations’, Nature, 538(7624), pp. 201–206.
Marchi, N. et al. (2022) ‘The genomic origins of the world’s first farmers’, Cell [Preprint].
doi:10.1016/j.cell.2022.04.008.
Mateiciucová, I. (2015) ‘Silices’, in Neugebauer-Maresch, C. and Lenneis, E. (eds) Das
Linearbandkeramische Gräberfeld von Kleinhadersdorf. Verlag der Österreichischen Akademie
der Wissenschaften (Mitteilungen der Prähistorischen Kommission), pp. 111–122.
Mathieson, I. et al. (2015) ‘Genome-wide patterns of selection in 230 ancient Eurasians’,
Nature, 528(7583), pp. 499–503.
Mathieson, I. et al. (2018) ‘The genomic history of southeastern Europe’, Nature, 555(7695), pp.
197–203.
Meisner, J. and Albrechtsen, A. (2018) ‘Inferring Population Structure and Admixture
Proportions in Low-Depth NGS Data’, Genetics, 210(2), pp. 719–731.
Meyer, M. and Kircher, M. (2010) ‘Illumina sequencing library preparation for highly multiplexed
target capture and sequencing’, Cold Spring Harbor protocols, 2010(6), p. db.prot5448.
Mittnik, A. et al. (2018) ‘The genetic prehistory of the Baltic Sea region’, Nature communications,
9(1), p. 442.
Nägele, K. et al. (2020) ‘Genomic insights into the early peopling of the Caribbean’, Science,
369(6502), pp. 456–460.
Nehlich, O. et al. (2010) ‘Sulphur isotope evidence for freshwater fish consumption: a case
study from the Danube Gorges, SE Europe’, Journal of archaeological science, 37(5), pp.
1131–1139.
Neugebauer-Maresch, C. and Lenneis, E. (2015) Das Linearbandkeramische Gräberfeld von
Kleinhadersdorf. Mitteilungen der Prähistorischen Kommission 82. Wien: Verlag der
Österreichischen Akademie der Wissenschaften.
Nieszery, N. (1995) Linearbandkeramische gräberfelder in Bayern. Rahden, Germany: VML Vlg
Marie Leidorf (Internationale Archäologie).
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Olalde, I. et al. (2014) ‘Derived immune and ancestral pigmentation alleles in a 7,000-year-old
Mesolithic European’, Nature, 507(7491), pp. 225–228.
Olalde, I. et al. (2015) ‘A Common Genetic Origin for Early Farmers from Mediterranean Cardial
and Central European LBK Cultures’, Molecular biology and evolution, 32(12), pp. 3132–3142.
Oross, K. and Bánffy, E. (2009) ‘Three successive waves of Neolithisation: LBK development in
Transdanubia’, Documenta Praehistorica, 36(0), pp. 175–189.
Orschiedt, J. (1998) ‘Ergebnisse einer neuen Untersuchung der spätmesolithischen
Kopfbestattungen aus Süddeutschland’, in Conard, N.J. and Kind, C.-J. (eds) Aktuelle
Forschungen zum Mesolithikum – Current Mesolithic Research. Mo Vince Verlag
(Urgeschichtliche Materialhefte), pp. 147–160.
Orschiedt, J. et al. (2012) ‘The Neolithic and Mesolithic cave site “Blätterhöhle” in Westphalia
(D)’, Notae Praehistoricae, 32, pp. 73–88.
Orschiedt, J. and Haidle, M.N. (2012) ‘Violence against the living, violence against the dead on
the human remains from Herxheim, Germany. Evidence of a crisis and mass cannibalism?’,
Sticks, Stones, and Broken Bones, pp. 121–138.
doi:10.1093/acprof:osobl/9780199573066.003.0007.
Özdoğan, M. (2011) ‘Archaeological Evidence on the Westward Expansion of Farming
Communities from Eastern Anatolia to the Aegean and the Balkans’, Current anthropology,
52(S4), pp. S415–S430.
Pechtl, J. (2015) ‘Zeit und Umwelt. Ein Forschungsprojekt zur ersten bäuerlichen Kultur in
Bayerisch-Schwaben’, Denkmalpflege Informationen, 161, pp. 14–17.
Pollard, K.S. et al. (2010) ‘Detection of nonneutral substitution rates on mammalian
phylogenies’, Genome research, 20(1), pp. 110–121.
Pool, J.E. and Nielsen, R. (2007) ‘Population size changes reshape genomic patterns of
diversity’, Evolution; international journal of organic evolution, 61(12), pp. 3001–3006.
Porčić, M. et al. (2021) ‘The Neolithic Demographic Transition in the Central Balkans: population
dynamics reconstruction based on new radiocarbon evidence’, Philosophical transactions of the
Royal Society of London. Series B, Biological sciences, 376(1816), p. 20190712.
Porčić, M., Blagojević, T. and Stefanović, S. (2016) ‘Demography of the Early Neolithic
Population in Central Balkans: Population Dynamics Reconstruction Using Summed
Radiocarbon Probability Distributions’, PloS one, 11(8), p. e0160832.
Posth, C. et al. (2016) ‘Pleistocene Mitochondrial Genomes Suggest a Single Major Dispersal of
Non-Africans and a Late Glacial Population Turnover in Europe’, Current biology: CB, 26(6), pp.
827–833.
Quinlan, A.R. and Hall, I.M. (2010) ‘BEDTools: a flexible suite of utilities for comparing genomic
features’, Bioinformatics , 26(6), pp. 841–842.
Rasmussen, M. et al. (2011) ‘An Aboriginal Australian genome reveals separate human
dispersals into Asia’, Science, 334(6052), pp. 94–98.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Renaud, G. et al. (2019) ‘Joint Estimates of Heterozygosity and Runs of Homozygosity for
Modern and Ancient Samples’, Genetics, 212(3), pp. 587–614.
Ridush, B. (2009) ‘Bear caves’, Ukraine. Slovenský Kras Acta Carsologica Slovaca, 47, pp.
67–84.
Ringbauer, H., Novembre, J. and Steinrücken, M. (2021) ‘Parental relatedness through time
revealed by runs of homozygosity in ancient DNA’, Nature communications, 12(1), p. 5425.
Roksandić, M. (1999) Transition from Mesolithic to Neolithic in the Iron Gates gorge: Physical
anthropology perspective. PhD. Simon Fraser University. Available at:
https://www.academia.edu/download/4546963/RoksandicPhD.pdf.
Roodenberg, A., Gerritsen, F.A. and Özbal, R. (2013) ‘Neolithic Burials from Barcın Höyük: The
2007-2012 Excavation Seasons’, Anatolica, 39, pp. 93–111.
Scheu, A. et al. (2015) ‘The genetic prehistory of domesticated cattle from their origin to the
spread across Europe’, BMC genetics, 16, p. 54.
Schulting, R.J. and Fibiger, L. (2012) Sticks, Stones, and Broken Bones: Neolithic Violence in a
European Perspective. OUP Oxford.
Schulz, A. (2016) Paläo- und Populationsgenetik von Jäger-Sammler-Fischern und Ackerbauern
in Nordzentraleuropa und angrenzenden Regionen. PhD. Johannes Gutenberg-Universität
Mainz. doi:10.25358/openscience-2832.
Shastry, V. et al. (2021) ‘Model‐based genotype and ancestry estimation for potential hybrids
with mixed‐ploidy’, Molecular Ecology Resources, pp. 1434–1451.
doi:10.1111/1755-0998.13330.
Shennan, S. (2018) The First Farmers of Europe: An Evolutionary Perspective. Cambridge
University Press.
Sikora, M. et al. (2017) ‘Ancient genomes show social and reproductive behavior of early Upper
Paleolithic foragers’, Science, 358(6363), pp. 659–662.
Skoglund, P. et al. (2013) ‘Accurate sex identification of ancient human remains using DNA
shotgun sequencing’, Journal of Archaeological Science, pp. 4477–4482.
doi:10.1016/j.jas.2013.07.004.
Skoglund, P. et al. (2014) ‘Genomic diversity and admixture differs for Stone-Age Scandinavian
foragers and farmers’, Science, 344(6185), pp. 747–750.
Skotte, L., Korneliussen, T.S. and Albrechtsen, A. (2013) ‘Estimating individual admixture
proportions from next generation sequencing data’, Genetics, 195(3), pp. 693–702.
Songül Alpaslan-Roodenberg, M. and Roodenberg, J. (2020) ‘In the light of new data: The
population of the first farming communities in the eastern Marmara region’, Praehistorische
Zeitschrift, 95(1), pp. 48–77.
Srejović, D. (1969) Lepenski Vir: Nova praistorijska kultura u Podunavlju. SKZ Beograd.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Stefanović, S. and Borić, D. (2008) ‘New-born infant burials underneath house floors at
Lepenski Vir: in pursuit of contextual meanings’, The Iron Gates in Prehistory: New
Perspectives. Oxford: Archaeopress, pp. 131–169.
Stojanovski, D. et al. (2020) ‘Living off the land: Terrestrial-based diet and dairying in the farming
communities of the Neolithic Balkans’, PloS one, 15(8), p. e0237608.
Street, M. et al. (2001) ‘Final Paleolithic and Mesolithic Research in Reunified Germany’,
Journal of World Prehistory, 15(4), pp. 365–453.
Stutz, L.N., Larsson, L. and Zagorska, I. (2013) ‘The persistent presence of the dead: recent
excavations at the hunter-gatherer cemetery at Zvejnieki (Latvia)’, Antiquity, 87(338), pp.
1016–1029.
Talalay, L.E. (2004) ‘Heady business: skulls, heads and decapitation in Neolithic Anatolia and
Greece’, Journal of Mediterranean Archaeology, 17(2), pp. 139–163.
Tasić, N. et al. (2015) ‘The end of the affair: formal chronological modelling for the top of the
Neolithic tell of Vinča-Belo Brdo’, Antiquity, 89(347), pp. 1064–1082.
Terberger, T. et al. (2015) ‘Standing upright to all eternity–The Mesolithic burial site at Groß
Fredenwalde, Brandenburg (NE Germany): Aufrecht in die Ewigkeit – Der mesolithische
Bestattungsplatz von Groß Fredenwalde, Brandenburg (Nordostdeutschland)’, Quartär, 62 , pp.
133–153.
Terberger, T. et al. (2018) ‘Step by step – The neolithisation of Northern Central Europe in the
light of stable isotope analyses’, Journal of Archaeological Science, pp. 66–86.
doi:10.1016/j.jas.2018.08.004.
Teschler-Nicola, M. (2012) ‘The Early Neolithic site Asparn/Schletz (Lower Austria)’, in
Schulting, R.J. and Fibiger, L. (eds) Sticks, Stones, and Broken Bones: Neolithic Violence in a
European Perspective. Oxford: Oxford University Press, pp. 101–120.
Tiefenböck, B. and Teschler-Nicola, M. (2015) ‘Teil II: Anthropologie’, in Neugebauer-Maresch,
C. and Lenneis, E. (eds) Das Linearbandkeramische Gräberfeld von Kleinhadersdorf.
Mitteilungen der Prähistorischen Kommission 82. Wien: Verlag der Österreichischen Akademie
der Wissenschaften, pp. 297–392.
Turck, R. (2019) ‘Where did the dead from Herxheim originate? Isotope analyses of human
individuals from the find concentrations in the ditches’, in Zeeb-Lanz, A. (ed.) Ritualised
Destruction in the Early Neolithic – the Exceptional Site of Herxheim (Palatinate, Germany) Vol.
2. Speyer: Generaldirektion Kulturelles Erbe, Direktion Landesarchäologie (Forschungen zur
Pfälzischen Archäologie ), pp. 313–421.
Veeramah, K.R. et al. (2018) ‘Population genomic analysis of elongated skulls reveals extensive
female-biased immigration in Early Medieval Bavaria’, Proceedings of the National Academy of
Sciences of the United States of America, 115(13), pp. 3494–3499.
Walde, T.S. im et al. (1986) ‘Köln Radiocarbon Dates III’, Radiocarbon, 28(1), pp. 134–140.
Walsh, S. et al. (2014) ‘Developmental validation of the HIrisPlex system: DNA-based eye and
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

hair colour prediction for forensic and anthropological usage’, Forensic science international.
Genetics, 9, pp. 150–161.
Walsh, S. et al. (2017) ‘Global skin colour prediction from DNA’, Human genetics, 136(7), pp.
847–863.
Waples, R.K., Albrechtsen, A. and Moltke, I. (2019) ‘Allele frequency-free inference of close
familial relationships from genotypes or low-depth sequencing data’, Molecular ecology, 28(1),
pp. 35–48.
Weissensteiner, H. et al. (2016) ‘HaploGrep 2: mitochondrial haplogroup classification in the era
of high-throughput sequencing’, Nucleic acids research, 44(W1), pp. W58–63.
Whittle, A. et al. (2002) ‘In the beginning : new radiocarbon dates for the Early Neolithic in
northern Serbia and south-east Hungary’, Antaeus., 25, pp. 63–118.
Wild, E.M. et al. (2004) ‘Neolithic Massacres: Local Skirmishes or General Warfare in Europe?’,
Radiocarbon, 46(1), pp. 377–385.
Windl, H.J. (1999) ‘Makabres Ende einer Kultur?’, Archäologie in Deutschland, 1, pp. 54–57.
Windl, H.J. (2009) ‘Zur Stratigraphie der bandkeramischen Grabenwerke von Asparn an der
Zaya-Schletz’, in Zeeb-Lanz, A. (ed.) Krisen – Kulturwandel – Kontinuitäten. Zum Ende der
Bandkeramik in Mitteleuropa. Beiträge der internationalen Tagung in Herxheim bei Landau
(Pfalz) vom 14.-17.06.2007. Rahden: Verlag Marie Leidorf, pp. 191–196.
Wood, R.E. et al. (2013) ‘Freshwater Radiocarbon Reservoir Effects at the Burial Ground of
Minino, Northwest Russia’, Radiocarbon, 55(1), pp. 163–177.
Zeeb-Lanz, A. (ed.) (2016) Ritualised Destruction in the Early Neolithic – the Exceptional Site of
Herxheim (Palatinate, Germany) Vol. 1. Speyer: Generaldirektion Kulturelles Erbe, Direktion
Landesarchäologie (Forschungen zur Pfälzischen Archäologie).
Zeeb-Lanz, A. (ed.) (2019) Ritualised Destruction in the Early Neolithic – the Exceptional Site of
Herxheim (Palatinate, Germany) Vol. 2. Speyer: Generaldirektion Kulturelles Erbe, Direktion
Landesarchäologie (Forschungen zur Pfälzischen Archäologie).
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Document S1
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Figure S1.Quality assesment, sex and relatedness of the genomic samples
(A) Results from shallow MiSeq-screen of Libraries. The number of molecules in the fill-in product, measured with quantitative
real-time PCR (qFI) and the percentage of reads aligning to the human reference (% endogenous). Colors and panels according to
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

the strategy chosen. SupCap stands for mitochondrial capture on the supernatant of nuclear capture experiment, containing
unhybridized DNA.
(B) Post-mortem damage estimates for the first and last 10 bp of the reads. All samples show expected damage patterns
(C) Fraction of Y- to (X+Y)-chromosomal reads. Up from a ratio above 0.075 (orange), an individual is considered to have a XY
genotype, while a ratio below 0.016 (green) is considered to arise from a XX genotype. (Skoglund 2013).
(D) Relatedness between the samples that passed quality filters. PO: parent-offspring, FS: full-siblings, HS:
half-siblings/avuncular/grandparent-grandchild, C1: first cousin, UR: unrelated, DR: distantly related. All samples are not more than
distantly related except for Bar8 and Bar15 that are PO, and Dil15 and Dil16 that are FS. Bar15 and Dil15 were additionally filtered
out.
Related to STAR Methods, Library preparation and Target enrichment, Ancient DNA Authenticity, Molecular Sex
Determination and Kinship analysis.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Figure S2.Downsampling analysis and quality filtering for comparability among samples
(A) log ratio between the median θ (heterozygosity) sampled estimates using all data and the median θ sampled estimates with data
downsampled 50%. Additional quality filters were applied on all samples (left for capture, right for whole-genomes) based on the log
ratio and sequencing depth (>1.5x and >4x for whole-genome and capture, respectively); both denoted as horizontal and vertical
dotted lines. Samples coloured in red were discarded for downstream analysis.
(B) θ downsampling experiments in low (1-4x, left) and medium-coverage (~10x, right) samples from 100% to 5% data. Four
different tests were done. Raw, the θ window estimates were calculated without taking into account post-mortem damage (PMD)
and recalibration of base quality scores; noPMD, just recalibration parameters were considered; noRecal, only PMD parameters
were used; PMDRecal,; both PMD and recalibration estimates were taken into account, this is the final way all our data were
processed.
Related to STAR Methods, Depth and Heterozygosity Filtering, and Downsampling experiments.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Figure S3. Genetic structure in the ancient samples
(A) PCAs for Neo Aegean-like and Meso European-like with their respective sample labels.
(B) Admixture results from K=2 to K=4 using different softwares. The Evanno method and the deviance information criterion (DIC)
were used for NGsAdmix and Entropy, respectively, to estimate the best K (PCAngsd estimates the best K on the fly). In all cases
K=2 is the best. The set of samples used here excludes Lec2, Motala12 and North-Eastern Europeans compared to PCA.
Related to Figure 2 and STAR Methods, PCA and Admixture analysis.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Figure S4. Theta ratio power analysis.
(A) 100 BAMfiles were simulated with a sequencing depth of 10x and two chromosomes of same or different length (1Mb, 10Mb).
The first chromosome was simulated with θ = 0.75*0.001 and the second with θ = 0.001 to mimic the expected ratio between the X
chromosome and an autosome in humans. All BAMfiles were downsampled from 10x to 0.5x. The distribution of posterior medians
obtained with ATLAS task=thetaRatio; at 0.5x most medians were out the plotted range.
(B) Root mean squared error (RSME) between the posterior median and true value log(¾); when less than 10Mb are available per
region below 2x, the RMSE becomes much larger than the scale of the plot.
(C) The difference between the median of each 100 posterior medians per sequencing depth and window size and the true value
log(¾); below 2x and with less than 10Mb per region the Bias becomes more negative than the scale of the plot.
(D) Three ancient samples with medium coverage depth were chosen and downsampled from 10x to 0.5x to estimate the theta ratio
between neutral sites in X chromosome and autosomes. Expected values are in dashed lines.
Related to Figure 3 and STAR Methods, Heterozygosity ratio.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Figure S5. HIrisPlex analysis.
Probabilities for eye, hair and skin pigmentation phenotypes estimated using the HIrisPlex-S webtool for newly sequenced capture
and whole genomes and previously published individuals. In case an individual has capture and whole-genome sequences
available, the analysis with capture data was just shown. Some SNPs associated with hair colour genotypes could not be called
properly for some individuals.
Related to Figure 4 and STAR Methods, HIrisPlex analysis
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Table S1. Reference whole-genome samples.
Sample ID
Site
Phase
Archeological/
BurialID
Age (cal. BC)
Mean
Depth
Sex
Publication
SunghirII
Sunghir
MUP
2, SII
32,234 ± 1049
3.86
XY
(Sikora et al. 2017)
SunghirIII
Sunghir
MUP
2, SIII
32,092 ± 1061
10.16
XY
(Sikora et al. 2017)
SunghirIV
Sunghir
MUP
2, SIV
31,992 ± 493
3.68
XY
(Sikora et al. 2017)
SunghirI*
Sunghir
MUP
1, SI
30,822 ± 1052
1.05
XY
(Sikora et al. 2017)
Kotias
Kotias Klde
LM
Layer A2
7,712 ± 183
11.93
XY
(Jones et al. 2015)
ZVEJ25
Zvejnieki
LM (Narva)
93
5,738 ± 102.5
3.16
XY
(Jones et al. 2017)
ZVEJ32*
Zvejnieki
MM (Kunda)
313
6,358 ± 109
0.98
XX
(Jones et al. 2017)
Hum1*
Hummervikholmen
M Scand.
-
7,363.5 ± 88.5
0.7
XX
(Günther et al. 2018)
Hum2*
Hummervikholmen
M Scand.
-
7,363.5 ± 88.5
3.64
XY
(Günther et al. 2018)
SF9*
Stora Förvar
M Scand.
G/Layer 9
7,144 ± 156
1.03
XX
(Günther et al. 2018)
Stg001*
Steigen
M Scand.
-
3,857 ± 93
1.11
XY
(Günther et al. 2018)
Motala12
Östergötland
M Scand.
-
5,938.5 ±422.5
1.92
XX
(Lazaridis et al. 2014)
Satsurbia*
Satsurbia
UP
Area B, Y5
11,256 ± 124
1.17
XY
(Jones et al. 2015)
Bichon
Grotte du Bichon
UP
Bichon
11,665 ± 105
7.47
XY
(Jones et al. 2015)
Loschbour
Heffingen
LM
LBK380
6,105 ± 115
12.05
XY
(Lazaridis et al. 2014)
LaBrana*
La Braña-Arintero
M
La Braña 1
5,815 ± 125
2.71
XY
(Olalde et al. 2014)
Canes1-Meso*
Canes
M
I-A
5,115 ± 130
0.49
XX
(González-Fortes et al.
2017)
Chan-Meso*
Chan do Lindeiro
M
Elba
7,131 ± 124
0.95
XX
(González-Fortes et al.
2017)
OC1-Meso*
Ostrovul Corbului
M
24
6,704 ± 269
1.2
XY
(González-Fortes et al.
2017)
SC2-Meso*
Schela Cladovei
M
M96/3
-
2.341
XY
(González-Fortes et al.
2017)
VLASA7
Vlasac
LM
VL 31
6,552 ± 212
15.21
XY
(Marchi et al. 2022)
VLASA32#
Vlasac
LM
VL 16
7,604.5 ± 136.5
12.65
XY
(Marchi et al. 2022)
WC1
Wezmeh Cave
EN
n-10
7,268.5 ± 186.5
12.72
XY
(Broushaki et al. 2016)
AH1*
Tepe Abdul Hosein
EN
13030
-
1.1
XX
(Broushaki et al. 2016)
AH2*
Tepe Abdul Hosein
EN
19001
7,980.5 ± 224.5
0.6
XY
(Broushaki et al. 2016)
AH4*
Tepe Abdul Hosein
EN
10035
7,979.5 ± 224.5
0.82
XX
(Broushaki et al. 2016)
Bon002
Boncuklu
N
ZHB; Grave 9
8,128 ± 151
5.67
XX
(Kılınç et al. 2016)
AKT16#
Aktopraklik
EN
89 D 14.1
6,547.5 ± 87.5
10.85
XX
(Marchi et al. 2022)
Bar31
Barcin
EN
L11W-546
6,328 ± 91
3.56
XY
(Hofmanová et al. 2016)
Bar81
Barcin
EN
M10/106
6,121 ± 91
7.38
XX
(Hofmanová et al. 2016)
Bar25
Barcin
EN
BH 43347,
M10, 455 (lot.
1856)
6,294.5 ± 89.5
12.66
XY
(Marchi et al. 2022)
Nea3
Nea Nikomedeia
EN
T XII
6,183.5 ± 143.5
11.57
XX
(Marchi et al. 2022)
Nea2
Nea Nikomedeia
EN
#7
6,098 ± 75
12.51
XX
(Marchi et al. 2022)
Rev5*
Revenia
EN
Rev5, burial 2
6,351 ± 87
1.14
XX
(Hofmanová et al. 2016)
Klei10
Kleitos
FN
grave 9
4,116 ± 118
2.49
XY
(Hofmanová et al. 2016)
Pal7
Paliambela
LN
Pal7
4,401 ± 51
1.56
XX
(Hofmanová et al. 2016)
LEPE52#
Lepenski-Vir
E-MN
LV 73
5,812 ± 119
12.38
XY
(Marchi et al. 2022)
LEPE48
Lepenski-Vir
TEN
LV 122
5,939.5 ± 72.5
10.93
XY
(Marchi et al. 2022)
STAR1
Grad-Starčevo
EN
(Starčevo)
grave 1
5,532.5 ± 56.5
10.56
XX
(Marchi et al. 2022)
VC3-2
Vinča-Belo Brdo
EN
(Starčevo)
grave V - ND
5,495.5 ± 69.5
11.23
XY
(Marchi et al. 2022)
KO1*
TiszaszőlősDomaháza
EN (Körös)
4
5,715 ± 65
0.73
XY
(Gamba et al. 2014)
NE1
Polgár-Ferenci-hát
MN (ALP)
325 V
5,190 ± 120
17.66
XX
(Gamba et al. 2014)
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Klein7
Kleinhadersdorf
EN (LBK)
Grave 56
(25.937)
4,977 ± 181
11.3
XX
(Marchi et al. 2022)
Asp6#
Asparn-Schletz
EN (LBK)
Ind 44, 646 Part
152, Schnitt 10
LM70
5,524.5 ± 50.5
12.12
XY
(Marchi et al. 2022)
Dil16#,1
Dillingen-Steinheim
EN (LBK)
Grave 24,
Befund 24
5,116.5 ± 118.5
10.61
XY
(Marchi et al. 2022)
Ess7#
Essenbach Ammerbreit
EN (LBK)
Grave 2
4,975 ± 75
12.34
XY
(Marchi et al. 2022)
Herx#
Herxheim
EN (LBK)
281-19-6
5,078 ± 85
11.47
XX
(Marchi et al. 2022)
Stuttgart
Stuttgart Mühlhausen
EN (LBK)
grave (I-78,
area-1)
4,900 ± 150
13.59
XX
(Lazaridis et al. 2014)
CB13*
Cova Bonica
EN (Cardial)
CB13-HH34-
IV 2 -2407
5,415 ± 55
0.84
XX
(Olalde et al. 2015)
ZVEJ27*
Zvejnieki
M/EN
121
5,077 ± 225
0.7
XY
(Jones et al. 2017)
ZVEJ31*
Zvejnieki
MN (Ware)
221
4,014 ± 214.5
1.01
XX
(Jones et al. 2017)
GB1-Eneo*
Gura Baciului
Eneolithic
M1
3,377 ± 77
2.73
XX
(González-Fortes et al.
2017)
Ajv58*
Ajvide
N (PWC)
-
2,750 ± 150
0.96
XY
(Skoglund et al. 2014)
KhSan-1
South Africa
Modern
-
-
38.45
XX
(Mallick et al. 2016)
KhSan-2
South Africa
Modern
-
-
42.42
XX
(Mallick et al. 2016)
Mbuti-2
Congo
Modern
-
-
30.81
XX
(Mallick et al. 2016)
Mandenka-2
Senegal
Modern
-
-
32.44
XX
(Mallick et al. 2016)
BaKenya-2
Kenya
Modern
-
-
31.34
XX
(Mallick et al. 2016)
Turkish-2
Turkey
Modern
-
-
31.05
XX
(Mallick et al. 2016)
Georgian-1
Georgia
Modern
-
-
35.32
XY
(Mallick et al. 2016)
Greek-1
Greece
Modern
-
-
25.45
XY
(Mallick et al. 2016)
Bulgarian-1
Bulgaria
Modern
-
-
34.83
XY
(Mallick et al. 2016)
Albanian-1
Albania
Modern
-
-
24.09
XX
(Mallick et al. 2016)
Hungarian-1
Hungary
Modern
-
-
27.7
XX
(Mallick et al. 2016)
Czech-2
Czech Republic
Modern
-
-
40.77
XY
(Mallick et al. 2016)
Polish-1
Poland
Modern
-
-
38.84
XY
(Mallick et al. 2016)
Saami-1
Finland
Modern
-
-
35.71
XX
(Mallick et al. 2016)
Finnish-1
Finland
Modern
-
-
32.18
XX
(Mallick et al. 2016)
English-1
England
Modern
-
-
34.21
XY
(Mallick et al. 2016)
French-2
France
Modern
-
-
26.68
XX
(Mallick et al. 2016)
Spanish-2
Spain
Modern
-
-
35.72
XX
(Mallick et al. 2016)
Bergamo-2
Italy
Modern
-
-
65.99
XX
(Mallick et al. 2016)
Sardinian-2
Italy
Modern
-
-
28.96
XX
(Mallick et al. 2016)
aUP, Upper Palaeolithic; MUP, Mid Upper Palaeolithic; M, Mesolithic; EM, Early Mesolithic; MM, Middle Mesolithic; LM, Late Mesolithic; TEN,
Transformational/Early Neolithic; N, Neolithic; EN, Early Neolithic; E-MN, Early-Middle Neolithic; MN, Middle Neolithic; LN, Late Neolithic; FN,
Final Neolithic
* Samples were discarded for downstream analysis.
#Data for both whole genome and capture nuclear genome.
1Related samples, Bar8-Bar15; DIl15-Dil16.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint

Supplemental References
Broushaki, Farnaz, Mark G. Thomas, Vivian Link, Saioa López, Lucy van Dorp, Karola Kirsanow, Zuzana Hofmanová, et al. 2016.
“Early Neolithic Genomes from the Eastern Fertile Crescent.” Science 353 (6298): 499–503.
Gamba, Cristina, Eppie R. Jones, Matthew D. Teasdale, Russell L. McLaughlin, Gloria Gonzalez-Fortes, Valeria Mattiangeli, László
Domboróczki, et al. 2014. “Genome Flux and Stasis in a Five Millennium Transect of European Prehistory.” Nature
Communications 5 (October): 5257.
González-Fortes, Gloria, Eppie R. Jones, Emma Lightfoot, Clive Bonsall, Catalin Lazar, Aurora Grandal-d’Anglade, María Dolores
Garralda, et al. 2017. “Paleogenomic Evidence for Multi-Generational Mixing between Neolithic Farmers and Mesolithic
Hunter-Gatherers in the Lower Danube Basin.” Current Biology: CB 27 (12): 1801–10.e10.
Günther, Torsten, Helena Malmström, Emma M. Svensson, Ayça Omrak, Federico Sánchez-Quinto, Gülşah M. Kılınç, Maja
Krzewińska, et al. 2018. “Population Genomics of Mesolithic Scandinavia: Investigating Early Postglacial Migration Routes and
High-Latitude Adaptation.” PLoS Biology 16 (1): e2003703.
Hofmanová, Zuzana, Susanne Kreutzer, Garrett Hellenthal, Christian Sell, Yoan Diekmann, David Díez-Del-Molino, Lucy van Dorp,
et al. 2016. “Early Farmers from across Europe Directly Descended from Neolithic Aegeans.” Proceedings of the National
Academy of Sciences of the United States of America 113 (25): 6886–91.
Jones, Eppie R., Gloria Gonzalez-Fortes, Sarah Connell, Veronika Siska, Anders Eriksson, Rui Martiniano, Russell L. McLaughlin,
et al. 2015. “Upper Palaeolithic Genomes Reveal Deep Roots of Modern Eurasians.” Nature Communications 6 (November):
8912.
Jones, Eppie R., Gunita Zarina, Vyacheslav Moiseyev, Emma Lightfoot, Philip R. Nigst, Andrea Manica, Ron Pinhasi, and Daniel G.
Bradley. 2017. “The Neolithic Transition in the Baltic Was Not Driven by Admixture with Early European Farmers.” Current
Biology: CB 27 (4): 576–82.
Kılınç, Gülşah Merve, Ayça Omrak, Füsun Özer, Torsten Günther, Ali Metin Büyükkarakaya, Erhan Bıçakçı, Douglas Baird, et al.
2016. “The Demographic Development of the First Farmers in Anatolia.” Current Biology: CB 26 (19): 2659–66.
Lazaridis, Iosif, Nick Patterson, Alissa Mittnik, Gabriel Renaud, Swapan Mallick, Karola Kirsanow, Peter H. Sudmant, et al. 2014.
“Ancient Human Genomes Suggest Three Ancestral Populations for Present-Day Europeans.” Nature 513 (7518): 409–13.
Mallick, Swapan, Heng Li, Mark Lipson, Iain Mathieson, Melissa Gymrek, Fernando Racimo, Mengyao Zhao, et al. 2016. “The
Simons Genome Diversity Project: 300 Genomes from 142 Diverse Populations.” Nature 538 (7624): 201–6.
Marchi, Nina, Laura Winkelbach, Ilektra Schulz, Maxime Brami, Zuzana Hofmanová, Jens Blöcher, Carlos S. Reyna-Blanco, et al.
2022. “The Genomic Origins of the World’s First Farmers.” Cell, May. https://doi.org/10.1016/j.cell.2022.04.008.
Olalde, Iñigo, Morten E. Allentoft, Federico Sánchez-Quinto, Gabriel Santpere, Charleston W. K. Chiang, Michael DeGiorgio, Javier
Prado-Martinez, et al. 2014. “Derived Immune and Ancestral Pigmentation Alleles in a 7,000-Year-Old Mesolithic European.”
Nature 507 (7491): 225–28.
Olalde, Iñigo, Hannes Schroeder, Marcela Sandoval-Velasco, Lasse Vinner, Irene Lobón, Oscar Ramirez, Sergi Civit, et al. 2015. “A
Common Genetic Origin for Early Farmers from Mediterranean Cardial and Central European LBK Cultures.” Molecular Biology
and Evolution 32 (12): 3132–42.
Sikora, Martin, Andaine Seguin-Orlando, Vitor C. Sousa, Anders Albrechtsen, Thorfinn Korneliussen, Amy Ko, Simon Rasmussen,
et al. 2017. “Ancient Genomes Show Social and Reproductive Behavior of Early Upper Paleolithic Foragers.” Science 358
(6363): 659–62.
Skoglund, Pontus, Helena Malmström, Ayça Omrak, Maanasa Raghavan, Cristina Valdiosera, Torsten Günther, Per Hall, et al. 2014.
“Genomic Diversity and Admixture Differs for Stone-Age Scandinavian Foragers and Farmers.” Science 344 (6185): 747–50.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(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
The copyright holder for this preprintthis version posted June 28, 2022. ; https://doi.org/10.1101/2022.06.24.497512doi: bioRxiv preprint