ArticlePDF Available

Ancient mitochondrial diversity reveals population homogeneity in Neolithic Greece and identifies population dynamics along the Danubian expansion axis

Article

Ancient mitochondrial diversity reveals population homogeneity in Neolithic Greece and identifies population dynamics along the Danubian expansion axis

Abstract and Figures

The aim of the study is to investigate mitochondrial diversity in Neolithic Greece and its relation to hunter-gatherers and farmers who populated the Danubian Neolithic expansion axis. We sequenced 42 mitochondrial palaeogenomes from Greece and analysed them together with European set of 328 mtDNA sequences dating from the Early to the Final Neolithic and 319 modern sequences. To test for population continuity through time in Greece, we use an original structured population continuity test that simulates DNA from different periods by explicitly considering the spatial and temporal dynamics of populations. We explore specific scenarios of the mode and tempo of the European Neolithic expansion along the Danubian axis applying spatially explicit simulations coupled with Approximate Bayesian Computation. We observe a striking genetic homogeneity for the maternal line throughout the Neolithic in Greece whereas population continuity is rejected between the Neolithic and present-day Greeks. Along the Danubian expansion axis, our best-fitting scenario supports a substantial decrease in mobility and an increasing local hunter-gatherer contribution to the gene-pool of farmers following the initial rapid Neolithic expansion. Οur original simulation approach models key demographic parameters rather than inferring them from fragmentary data leading to a better understanding of this important process in European prehistory.
This content is subject to copyright. Terms and conditions apply.
1
Vol.:(0123456789)
Scientic Reports | (2022) 12:13474 | https://doi.org/10.1038/s41598-022-16745-8
www.nature.com/scientificreports
Ancient mitochondrial
diversity reveals population
homogeneity in Neolithic Greece
and identies population dynamics
along the Danubian expansion axis
Nuno M. Silva1,25, Susanne Kreutzer2,24,25, Angelos Souleles3, Sevasti Triantaphyllou4,
Kostas Kotsakis4, Dushka Urem‑Kotsou5, Paul Halstead6, Nikos Efstratiou4, Stavros Kotsos7,
Georgia Karamitrou‑Mentessidi8, Fotini Adaktylou9, Areti Chondroyianni‑Metoki10,
Maria Pappa11, Christina Ziota12, Adamantios Sampson13, Anastasia Papathanasiou14,
Karen Vitelli15, Tracey Cullen16, Nina Kyparissi‑Apostolika17, Andrea Zeeb Lanz18,
Joris Peters19,20, Jérémy Rio1, Daniel Wegmann21,22, Joachim Burger2,24,26,
Mathias Currat1,23,26* & Christina Papageorgopoulou3,26*
The aim of the study is to investigate mitochondrial diversity in Neolithic Greece and its relation to
hunter‑gatherers and farmers who populated the Danubian Neolithic expansion axis. We sequenced
42 mitochondrial palaeogenomes from Greece and analysed them together with European set of 328
mtDNA sequences dating from the Early to the Final Neolithic and 319 modern sequences. To test for
population continuity through time in Greece, we use an original structured population continuity
test that simulates DNA from dierent periods by explicitly considering the spatial and temporal
dynamics of populations. We explore specic scenarios of the mode and tempo of the European
OPEN
1Department of Genetics & Evolution, University of Geneva, Geneva, Switzerland. 2Palaeogenetics Group,
Institute of Organismic and Molecular Evolution (iomE), Johannes Gutenberg University of Mainz, 55099 Mainz,
Germany. 3Laboratory of Physical Anthropology, Department of History & Ethnology, Democritus University
of Thrace, 69100 Komotini, Greece. 4Faculty of Philosophy, School of History and Archaeology, Aristotle
University of Thessaloniki, 54124 Thessaloniki, Greece. 5Department of History & Ethnology, Democritus
University of Thrace, 69100 Komotini, Greece. 6Emeritus, Department of Archaeology, University of
Sheeld, Sheeld S1 3NJ, UK. 7Ephorate of Antiquities of Thessaloniki City, Hellenic Ministry of Culture
and Sports, 54003 Thessaloniki, Greece. 8Ephor Emerita of Antiquities, Hellenic Ministry of Culture & Sports,
10682 Athens, Greece. 9Ephorate of Antiquities of Chalcidice and Mount Athos, Hellenic Ministry of Culture
and Sports, 63100 Poligiros Chalcidice, Greece. 10Ephorate of Antiquities of Kozani, Hellenic Ministry of
Culture and Sports, 50131 Kozani, Greece. 11Ephorate of Antiquities of Thessaloniki Region, Hellenic Ministry
of Culture and Sports, 54646 Thessaloniki, Greece. 12Ephorate of Antiquities of Florina, Hellenic Ministry of
Culture and Sports, 53100 Florina, Greece. 13Department of Mediterranean Studies, University of Aegean,
85132 Rhodes, Greece. 14Ephorate of Paleoanthropology and Speleology, Hellenic Ministry of Culture and
Sports, 11636 Athens, Greece. 15Prof. Emerita, Department of Anthropology, Franchthi Cave Project, Indiana
University Bloomington, Bloomington, USA. 16American School of Classical Studies at Athens, Princeton, NJ,
USA. 17Ephor Emerita of the Ephorate of Paleoanthropology and Speleology, Hellenic Ministry of Culture
and Sports, 11636 Athens, Greece. 18General Direction for Cultural Heritage of Rhineland-Palatinate, Speyer,
Germany. 19Institute of Palaeoanatomy, Domestication Research and the History of Veterinary Medicine, LMU
Munich, Munich, Germany. 20SNSB, State Collection of Palaeoanatomy Munich, Munich, Germany. 21Department
of Biology, University of Fribourg, 1700 Fribourg, Switzerland. 22Swiss Institute of Bioinformatics, 1700 Fribourg,
Switzerland. 23Institute of Genetics and Genomics in Geneva (IGE3), University of Geneva, Geneva,
Switzerland. 24Present address: Functional Genomics Center Zurich/GEML, Department of Biology, ETH
Zurich, Zurich, Switzerland. 25These authors contributed equally: Nuno M. Silva and Susanne Kreutzer. 26These
authors jointly supervised this work: Joachim Burger, Mathias Currat and Christina Papageorgopoulou. *email:
Mathias.Currat@unige.ch; cpapage@he.duth.gr
Content courtesy of Springer Nature, terms of use apply. Rights reserved
2
Vol:.(1234567890)
Scientic Reports | (2022) 12:13474 | https://doi.org/10.1038/s41598-022-16745-8
www.nature.com/scientificreports/
Neolithic expansion along the Danubian axis applying spatially explicit simulations coupled with
Approximate Bayesian Computation. We observe a striking genetic homogeneity for the maternal line
throughout the Neolithic in Greece whereas population continuity is rejected between the Neolithic
and present‑day Greeks. Along the Danubian expansion axis, our best‑tting scenario supports a
substantial decrease in mobility and an increasing local hunter‑gatherer contribution to the gene‑pool
of farmers following the initial rapid Neolithic expansion. Οur original simulation approach models
key demographic parameters rather than inferring them from fragmentary data leading to a better
understanding of this important process in European prehistory.
In the last decade, ancient DNA (aDNA) studies have provided rst insights into the genetic diversity and
population structure of hunter-gatherers and early farmers in Europe and southwestern Asia, leading to a bet-
ter understanding of the process of Neolithisation18. ese palaeogenomic data imply immigration of Early
Neolithic farmers from southwestern Asia to Europe3,9,10 following two major routes: a maritime route along
the Mediterranean coastline and a mainland route along the Danube, connecting central Anatolia, Greece,
the Balkans and central Europe3,11,12. Although this general pattern of the spread of agriculture, along with its
signicant demographic and socioeconomic implications, is well established today, little is known regarding
regional heterogeneities.
Particularly, the region of present-day Greece at the crossroads of southwestern Asia, the Balkans, and the
eastern Mediterranean, played an important role in the Neolithisation of Europe. Farming reached present-day
Greece around c. 6,700 BCΕ. e earliest Neolithic sites are found on the island of Crete13,14, in the Peloponnese
(Franchthi cave Initial Neolithic strata, 7028–6648cal BCE15; Alepotrypa Cave, 6220–6030cal BCE16) central
Greece (Sarakenos Cave Initial Νeolithic, 6976–6685cal BCE17,18) and Macedonia (Mavropigi-Fillotsairi and
Paliambela-Kolindrou date to 6700–6600 BCE1921. Available dates for essaly, from the sites of Argissa, Gendiki
and Sesklo are slightly later at 6500–6400 BCΕ22,23. e newest archaeological ndings and radiocarbon dates
from northern Greece suggest that initial Neolithisation possibly occurred almost simultaneously on both sides
of the Aegean20. From there, the Neolithic dispersal reached the northern Balkans and central Europe following
three major routes: (1) the Struma Basin24, (2) race25 and (3) the Black Sea26.
Overall, the Neolithic in Greece spans a period of nearly 4,000years and comprises four main chronological
phases, i.e., Early (6700/6500–6000/5600 BCE), Middle (6000/5600–5400/5300 BCE), Late (5400/5300–4700/4300
BCE), and Final Neolithic (4700/4300–3300/3100 BCE)19,2729. e preceding Mesolithic period starts with the
onset of the Holocene but is represented by only few human burials with dates spanning from 8600 to 6500
BCΕ. ese ndings derive from caves i.e., Franchthi in the Peloponnese, eopetra in essaly, Cyclops Cave
on Youra island, Sarakenos cave in Boeοtia15,18,3035,37,38 and open-air sites i.e., Maroulas on the island of Kyth-
nos, 8800–8700 BC35. Lithic technology from the Final Paleolithic site of Ouriakos on Lemnos, an island in the
northeastern Aegean (10,500cal BCΕ36) and Maroulas, a Mesolithic settlement on Kythnos in the southwestern
Aegean (8500–6500 BCE37,38) indicate cultural contacts with hunter-gatherers of southwestern Anatolian cave
sites of Öküzini, Direkli, and Girmeler respectively39,40. is supports the hypothesis of a coastal movement of
hunter-gatherers across southeastern Mediterranean18,4143 in parallel to the rise of sedentary communities in
central Anatolia (Așikli, Boncuklu, Pinarbai44,45).
In Greece, farming is considered to have spread rapidly, with most settlements being established between
6600 and 6400 calBC. is is inferred by the simultaneous appearance of the “Neolithic package, i.e., the set of
domesticated animals (sheep, goat, pig and cattle) and crops (wheat, barley, pulses), pottery, ground-stone arte-
facts, schematic gurines and the increasing number of Early Neolithic sites, particularly when compared to the
previous sparsely populated Mesolithic period. Similarities in material culture (clay stamps, schematic gurines,
ear-plugs, hooks) with the Near East have supported the idea of a migration of Near Eastern farmers15, but these
parallels point variously to the Levant or central Anatolia and in some cases remain contextually isolated15. For
example, rectangular or clustered houses, painted oors and walls, which are characteristic features of central
and western Anatolia, are strikingly absent in Early Neolithic Greece15,46. Moreover, lithic industry from Early
Neolithic Knossos shows common features to the Mesolithic of the Aegean islands, and pre-Neolithic ake
industries from Cyprus47.
e ambiguous references to dierent pre-Neolithic and Neolithic landscapes underline the complexity of
the process of Neolithisation in the Aegean and the diculty of identifying a single source region15. Having said
that, it is plausible that two or more waves of farmers, one following an island-coastal dispersal route originat-
ing in the Levant and another mainland route originating in central Anatolia, met in the Aegean. e selective
and at the same time heterogeneous nature of Early Neolithic material culture in the Aegean has variously been
interpreted as a deliberate loss of cultural identity48, a loss of cultural diversity from the core to the periphery49,
or as a consequence of migrations in the Aegean, predating the Neolithic expansion, resulting in considerable
variability and hybridity of cultural forms46.
roughout the Neolithic, material culture changed as seen in ceramic traditions50, burial customs51 and lithic
technology52. From the Middle Neolithic (6000/5600–5300 BCE) onward, a remarkable increase is observed in
the number of settlements, even in less favourable environments28,53. e formation of more and larger commu-
nities and the use of secondary products of animals including traction and milk, promoted more complex social
structures and larger economic networks. Communities were not isolated and networks of communication are
already observed since the beginning of the Neolithic46. During the Early Neolithic, however, the networking
inferred from the distribution of ceramics was indicative of local exchange between neighbouring communities,
whereas by the Late Neolithic stable networks were established over a radius of at least 200 km5457. Intensication
Content courtesy of Springer Nature, terms of use apply. Rights reserved
3
Vol.:(0123456789)
Scientic Reports | (2022) 12:13474 | https://doi.org/10.1038/s41598-022-16745-8
www.nature.com/scientificreports/
of this trend during the Final Neolithic laid one of the key foundations for the development of the next major
cultural transformation, namely that of the Bronze Age (BA).
From a population genetics perspective, this important period in European prehistory has only been mar-
ginally explored. Although hundreds of ancient genomes are available from southeastern Europe for a period
spanning 12,000 to 500 BC7,8,5860, only a single Early (Revenia3) and seven Late/Final Neolithic genomes from
northern (Paliambela, Kleitos3) and southern Greece (Alepotrypa Diros61, Franchthi Caves8) are currently avail-
able. e genome from Early Neolithic Revenia in northern Greece shows strong similarities to human genomes
from contemporaneous sites in northwestern Anatolia (i.e., Barcιn3) and to early farmers in central and western
Europe. e Final Neolithic genomes show an additional signal of gene-ow with a population that has genomic
anities with hunter-gatherers from the Caucasus3,8. Moreover, the mitochondrial haplogroups of the only two
Mesolithic individuals analysed from the Aegean so far, belong to lineages reported in central Anatolian and
Aegean Neolithic populations, but not in central and western European hunter-gatherers3.
As the area of present-day Greece constitutes the rst stepping-stone in the spread of agriculture towards
Europe, human genetic diversity and genetic dierentiation between individuals from dierent Neolithic peri-
ods can be expected to be particularly informative regarding the dynamics of the Neolithisation process. Here,
we sampled 70 individuals from three Mesolithic and 12 Neolithic sites across Greece covering the period
from 7050 to 3300 BCE (Supplementary Information). We were able to acquire mitochondrial genomes from
42 Neolithic individuals and document diachronic genetic diversity in relation to other European Mesolithic,
Neolithic, and modern populations. Together with 18 newly reported and 310 previously published mitochon-
drial genomes from the Balkans and central Europe, we evaluated our dataset in order to enhance insight into
population dynamics along the Danubian route using a spatially explicit computational simulation approach and
Approximate Bayesian Computation (ABC). In particular, we wanted to clarify whether the higher amount of
hunter-gatherer ancestry observed in the later Neolithic stage62,63 resulted from admixture at a constant rate or
whether this rate increased over time, meaning that more and more people with hunter-gatherer ancestry became
integrated into farming communities. In addition, we also wanted to investigate whether the fast migration of
early farmers from the Aegean area towards central Europe3,64 was accompanied by substantial admixture with
hunter-gatherers in the early stage.
Material and methods
Sample preparation and enrichment of the mitochondrial genomes. Ancient DNA (aDNA)
analysis of prehistoric specimens was performed in the dedicated cleanroom facilities of the Palaeogenetics
group at the University of Mainz, Germany. In total we sampled nine individuals from the Mesolithic Period,
17 from the Early Neolithic, nine from the middle Neolithic, 18 from the Late Neolithic and 17 from the Final
Neolithic (TableS1). DNA was extracted from teeth, long and petrous bones via phenol/chloroform extraction
and concentrated by Amicon Ultra-15 centrifugation (Supplementary Information). Illumina sequencing librar-
ies were prepared according to conventional protocols for ancient DNA65 using dierent indexing strategies
for bone/tooth and petrous bone samples (Supplementary Information). Ancient DNA preservation was deter-
mined by quantitative Real-Time PCR and a shallow shotgun sequencing approach3. e mitochondrial genome
was enriched by Agilent’s SureSelect Target enrichment (custom design) with adapted protocols for highly frag-
mented, deaminated and low copy number of endogenous mitochondrial DNA molecules. e enriched librar-
ies were sequenced on Illumina platforms (MiSeq—50bpSE, 150bpSE; HiSeq 2500—100bpPE) targeting 1–2
million reads per sample. Blank controls were processed, sequenced and screened for contaminating molecules
in each step of the protocol. Data analysis was performed as described elsewhere3 and tools to estimate the
authenticity of ancient DNA sequence data were applied to the compiled dataset66. Several independent extrac-
tions of a sample were merged and contamination estimates67 were drawn from the combined dataset per sample
(TableS2).
Genetic diversity in Neolithic Greece. To explore the genetic diversity of the population from Neo-
lithic Greece, we divided the initialdataset into three chronological groups: Early Neolithic (n = 17), Middle/
Late Neolithic (n = 27) and Final Neolithic (n = 17). All sequences were cut to the HVS-I region at position
16.051–16.400bp in the reference. e C-stretch polymorphism 16189C/T is excluded from the analysis. Arle-
quin 3.568 was used to explore the molecular diversity within each population of the dataset, and compute indi-
ces of genetic dierentiation (Fst) between pairs of samples using the Kimura P2 model to describe molecular
distances between sequences. In our dataset we also included ve previously published individuals: three from
the Early Neolithic sites of Nea Nikomedeia (Nea2 and Nea369) and Revenia (Rev53) and two from the Final
Neolithic sites of Paliambela (Pal7) and Kleitos (Klei10)3 (TableS1).
To visualize relations among samples, we conducted a multidimensional scaling analysis (MDS) using the R
function isoMDS from the package MASS on our samples, complemented with a reference panel consisting of
751 ancient (dating from the Upper Palaeolithic to the Final Neolithic) and 1719 present-day individuals from
southeastern and central Europe (TableS3).
Spatially explicit simulation framework. To investigate population continuity in northern Greece and
the relationship between Neolithic farmers along the Danubian expansion route from the northern Aegean to the
Balkans and central Europe, we adapted the spatially explicit simulation framework initially designed by Cur-
rat and Excoer70 and later improved by Silva etal.71 using a modied version of the program SPLATCHE272.
is framework allows the simulation of mitochondrial lineages at dierent points in time and space under
alternative scenarios of population dynamics. Based on the two-layer spatially explicit model, we simulated two
consecutive human expansions in a virtual European map divided into cells of 100 × 100km (Fig.S1). Each
Content courtesy of Springer Nature, terms of use apply. Rights reserved
4
Vol:.(1234567890)
Scientic Reports | (2022) 12:13474 | https://doi.org/10.1038/s41598-022-16745-8
www.nature.com/scientificreports/
cell is made up of two separate demes representing hunter-gatherers and farmers, respectively, resulting in two
superimposed layers of demes over the whole map. e two superimposed populations in one cell compete and
may admix, as described below. is modied version of SPLATCHE2 allows for varying admixture through
time, varying competition through space, and a reduction of the local migration rate using the new parameter
Mdec (see below).
e rst layer, termed HG for hunter-gatherers, represents the expansion of a Palaeolithic population starting
around 1600 generations ago (~ 40,000years considering a generation time of 25years) with 100 individuals
from an arbitrary deme set in the Near East (P in Fig.S1, Table1). e parameters used for the HG layer were
xed based on previous knowledge. ey serve to ll the map with pre-Neolithic populations and no inferences
were made for this layer. Each deme has a carrying capacity (KHG) of 100 eective haploid females, which cor-
responds to 200 individuals (males and females) per cell and a human density of 200*3/10,000 = 0.06 individuals/
km273,74, assuming that the census size was three times the eective population size75. e migration rate (mHG)
and growth rate (rHG) were set to 0.15 and 0.2, respectively, to achieve a colonization of Europe in approximately
500 generations, following Silva etal.71 (Table1). e mHG represents the proportion of individuals in each
deme emigrating to neighboring demes at each generation and the rHG represent the intrinsic rate of population
growth per generation.
e second layer, termed FA for farmers, represents the Neolithic and subsequent periods with a population
expansion starting 400 generations ago (~ 10,000years with an average human generation time of 25years)
from a deme set arbitrarily in eastern Anatolia with 100 individuals (N in Fig.S1). e initial carrying capacity
of these demes (KFA) was estimated using a uniform prior distribution of U[500, 1000] individuals, the upper
limit corresponding to the maximum density estimated for the Linear Pottery culture (LBK) of ~ 0.6 individuals/
km276. Eighty generations before the end of the simulations (~ 2000years ago), the value of KFA is set to 24,000 to
reect an increase in population density during the Roman period (to ~ 14.4 individuals/km276). e migration
rate (mFA) and growth rate (rFA) were set at to 0.4 and 0.53, respectively, in order to t the dates of the advance of
the Neolithic from eastern Anatolia to central Europe via Greece, following Silva etal.71 Table1).
A Lotka-Volterra model of competition77,78 with density dependent coecients of competition is used to
mimic the progressive disappearance of the hunter-gatherer subsistence strategy and its replacement by the
farming strategy70. Under this model, and thanks to a higher carrying capacity, farmers from the Neolithic
deme have a competitive edge over hunter-gatherers from the Palaeolithic deme located in the same geographic
cell. To cope with the long persistence of hunter-gatherers in central and northern Europe1,79, we modied the
original model71 by lengthening the cohabitation period with farmers up to 224 generations (until 4400years
ago) in an area representing central Europe in our simulations (Fig.S1). is was done by setting the coecients
of competition to 0 in this area, then setting KHG to 0 at generation 1424. is prolonged cohabitation period
allows us to sample the hunter-gatherer layer at the time corresponding to the real samples from the analysed
dataset, which otherwise would be impossible since the Palaeolithic deme would be empty at the sampling date.
Gene ow can occur from the Palaeolithic/Mesolithic to the Neolithic layer to represent hunter-gatherers who
adopted farming or the birth of a child in the farming population with a parent from each of the two popula-
tions. e amount of hunter-gatherer gene ow toward the farming population is regulated by the assimilation
rate γ, where γ = 0.0 indicates the absence of gene ow and γ = 1.0 full mixing; γ represents the proportion of
contacts between hunter-gatherers to farmers within a deme that results in gene ow from hunter-gatherers to
farmers at each generation (see70 for details). erefore, this rate, which we estimate from the data, quanties
the relevance of cultural transmission during the Neolithic expansion: a value γ = 0.0 reects a purely demic and
γ = 1.0 a purely cultural diusion of the Neolithic package.
To be able to reproduce both the interpopulational and intrapopulational patterns of mitochondrial diver-
sity along the Danubian route, we extended the original model of Silva etal.71 with the additional parameter
Mdec, which reects the factor by which the Neolithic migration rate (mFA) is divided in a deme once 90% of
the Neolithic carrying capacity (KFA) is reached in the same deme, and thus represents a reduction of mobility
Table 1. Input parameter for SPLATCHE2 for the various scenarios simulated. SPC corresponds to the
parameter used for Structured Population Continuity test and SN1 to SN4 represent the four alternative
scenarios of the Neolithic spread along the Danubian expansion axis. e parameters of interest were drawn
from prior distributions to make inferences by estimating their posterior distributions using the ABC (see
text). HG stands for the hunter-gatherers population layer and FA for the farmer population layer; rHG and rFA
stand for the growth rate in hunter-gatherers and farmers, respectively; mHG and mFA for the migration rates;
KHG and KFA for the carrying capacities; γ for the assimilation rate and Mdec for the factor of migration rate
decrease in farmers when their carrying capacity reaches 90%.
Analysis Scenario name Paleo. layer Neo. layer rHG mHG KHG rFA mFA KFA γMdec Admixture model
Continuity in Greece SPC Yes 0.53 [0.3–0.5] [100–2000] 1
Neolithic spread-Danu-
bian expansion axis
SN1 Yes Yes 0.2 0.15 100 0.53 0.4 [500–1000] [0.0–0.4] [1–20] Constant in all cells
SN2 Yes Yes 0.2 0.15 100 0.53 0.4 [500–1000] [0.0–0.4] [1–20] Increasing with time in
all cells
SN3 Yes Yes 0.2 0.15 100 0.53 0.4 [500–1000] [0.0–0.4] [1–20] Constant in central
Europe only
SN4 Yes Yes 0.2 0.15 100 0.53 0.4 [500–1000] [0.0–0.4] [1–20] Increasing with time in
central Europe only
Content courtesy of Springer Nature, terms of use apply. Rights reserved
5
Vol.:(0123456789)
Scientic Reports | (2022) 12:13474 | https://doi.org/10.1038/s41598-022-16745-8
www.nature.com/scientificreports/
in farmers aer their initial spread when its value is larger than 1. Mdec was xed at 1.0 (no reduction in mobil-
ity) for the continuity tests (see below) but inferred from the data when investigating the Danubian route of
Neolithisation (Table1).
For each simulated demographic scenario, we simulated mitochondrial DNA sequences of 347bp length
under the coalescent as described in Currat and Excoer70 using a mutation rate of 7.5 × 10–6 mutation/genera-
tion and a transversion rate of 0.98412. e computation of statistics both in simulated and real data was done
using the program Arlequin 3.5 and parameters of interest were inferred with ABC-GLM80 as implemented in
ABCtoolbox281 available at https:// bitbu cket. org/ wegma nnlab/ abcto olbox.
Simulation of the Neolithic spread along the Danube expansion axis. Τhe Neolithic spread along
the Danubian expansion axis was simulated following four dierent scenarios (SN1-4) diering in the mode and
tempo of the admixture between hunter-gatherers and farmers:
SN1: constant admixture along the Danubian expansion axis. is scenario represents an expansion of farmers
from southeastern to central Europe with various proportions of hunter-gatherer contribution regulated by γ.
Admixture between hunter-gatherers and farmers occurs all along the Danubian Neolithic expansion axis from
Greece to central Europe at a rate constant over space and time.
SN2: admixture increasing with time along the Danubian expansion axis. is scenario represents a rapid dis-
persal of Neolithic farmers from Greece to central Europe in an early phase, and a later assimilation of local
hunter-gatherers in a second phase. e admixture between hunter-gatherers and farmers occurs all along the
Danubian Neolithic route from Greece to central Europe, but it increases linearly during the coexistence of both
populations in the same cell till reaching its maximum (γ) at the end of the cohabitation period (i.e., aer some
time hunter-gatherers disappear due to competition with the farmers).
SN3: constant admixture only in central Europe. is scenario represents a fast migration wave of farmers from
Greece to central Europe without admixture along the way. To investigate whether admixture between hunter-
gatherers and farmers could have been limited in the southern part of the Danubian route, we tested a scenario
where the admixture occurs only in central Europe (Fig.S1), regulated by γ.
SN4: increasing admixture only in central Europe. is scenario represents a fast migration of farmers from
Greece to central Europe without admixture in an early phase, then an assimilation of local hunter-gatherers in
a second phase but restricted to central Europe, contrary to SN2 where admixture occurs all along the Danu-
bian route. It is similar to SN3, i.e., the admixture between hunter-gatherers and farmers occurs only in central
Europe (Fig.S1), except that admixture increases linearly with time during the cohabitation period till reaching
its maximum (γ) at the end of the coexistence of the two populations, similarly to SN2.
For each scenario, we performed 160,000 simulations with values for the three parameters of interest in our
study γ, KFA and Mdec drawn from prior distributions: γ [0–0.4]; KFA [500–1000] and Mdec [1–20] (Table1). We
set the maximum value for γ to 0.4, at which the assimilation of hunter-gatherers into farming populations is at
its maximum. e other parameters for the FA layer (rFA and mFA) were xed to t the extremely rapid Neolithic
spread from the Αegean area to central Europe.
Simulation of genetic diversity. Each combination of parameters is used as input to SPLATCHE2 in order to
generate genetic diversity in mtDNA sequences sampled at the precise geographic location and at the chrono-
logical date estimated for the real data available for Greece, the Balkans and central Europe (Fig.1). We compiled
328 ancient sequences for this analysis, which we grouped into seven population samples according to their
subsistence, chronological, and geographical characteristics: (1) hunter-gatherers from central Europe; Early
Neolithic farmers from (2) Greece, (3) Hungary/Croatia and (4) central Europe; and Middle and Late/Final
Neolithic farmers from (5) Greece, (6) Hungary and (7) central Europe (TableS4).
For each simulation, the genetic diversity was summarized in 14 statistics: the mean and standard deviation
of the number of haplotypes (k), heterozygosity (H), mean pairwise dierences (π) across all populations and
of the Fst between each pair of (1) hunter-gatherers and Neolithic central European farmers (Early and Late);
(2) Early Neolithic farmer samples (central Europe, Balkans and Greece); (3) Late Neolithic farmer samples
(central Europe, Balkans and Greece) and (4) Early and Late Neolithic farmer samples from the same area (in
central Europe, Balkans and Greece) (TableS5). All statistics were computed using the program Arlequin 3.5
under the Kimura P2 model82.
Model choice and parameter estimation. e ability of each model to reproduce the observed data was assessed
with ABCtoolbox2 by computing a marginal density P-value ranging from 0 (no t) to 1 (good t). Model choice
was then performed using two approaches. First, we calculated Bayes factors BA in favour of scenario MA over all
other scenarios using ABCtoolsbox2. To validate this procedure, we used the cross-validation procedure avail-
able in ABCtoolbox2 that determines PR, the probability of recovering the correct model among the simulated
data sets. Second, we used the model choice acceptance method of Pritchard etal.83, which assesses the relative
fraction of each model among the best simulations (those with the smallest distances) among the combined set
of simulations (here 640,000 simulations, 160,000 per scenario). We used ABCtoolbox2 to calculate distances
and assessed the robustness of this approach by retaining the best 0.25%, 1% and 2.5% simulations.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
6
Vol:.(1234567890)
Scientic Reports | (2022) 12:13474 | https://doi.org/10.1038/s41598-022-16745-8
www.nature.com/scientificreports/
We estimated the three varying parameters γ, KFA and Mdec under the most likely model using the ABC-
GLM80 method implemented in ABCtoolbox2. is method retains a small proportion δ of all N simulations
based on the minimized Euclidean distance calculated between the simulated and observed statistics. e pos-
terior distribution for each parameter is then obtained by approximating the truncated likelihood function
using a general linear model (GLM). We set δ to 1.0% but also tested other fractions (0.25%, 2.5%) to ensure the
robustness of the estimation.
Histograms representing the distribution of ‘posterior quantile’ and ‘posterior HDI’ for each parameter were
plotted and used to test whether the posterior distribution of the parameters is biased compared to its prior
distribution84. A total of 1,000 sets of statistics was generated under the best model with parameters drawn from
the posterior distributions and considered as if they had been observed in reality (pseudo-observations). e
positions of these true parameters are distributed uniformly in the marginal cumulative posterior distribution,
if it is unbiased85. Deviation from the uniform distribution was detected by the Kolmogorov–Smirnov test. We
also computed three dierent indices to evaluate the precision of our estimate:the relative bias (BIAS), the rela-
tive mean square error (RMSE) and the Factor2.
e Factor2 is dened as the proportion of n estimated values
θi
lying in an interval bounded by 50% and
200% of their “true” value
θi
.
θ
is the average of
θi
over n.
Testing population continuity in northern Greece. We used a Structured Population Continuity test
(SPC) to investigate the relationship between mitochondrial lineages from dierent chronological phases in
northern Greece. e aim was to assess whether populations from the earliest phases of the Neolithic can be
considered directly ancestral to populations from the later Neolithic periods, but also to present-day popula-
tions from the same region. We thus grouped ancient mitochondrial sequences into two chronological phases
corresponding to Early and Middle and to Late/Final Neolithic, thereaer, called serial population samples, and
we compared them. For this we used the dataset of the 42 newly acquired mtDNA sequences excluding two
Late/Final Neolithic samples from southern Greece, one from Franchthi Cave (Fra8) and one from arrounia
(a2), due to their location in southern Greece, distant from the other sequences (Fig.1, TableS4). To the 40
mtDNA sequences from northern Greece we added ve already published mtDNA sequences3,86 as described
above. We also compared those two Neolithic serial population samples with present-day mtDNA sequences
from northern Greece for which we compiled a dataset of 319 mtDNA sequences87 (Fig.1, TableS4).
(1)
BIAS
=
1
n
n
i=1
|θiθi
|
θi
(2)
RMSE
=
1
θ
1
n
n
i
=
1(
θiθi)2
Figure1. Geographical distribution of the mtDNA sequences (individual samples) used in the spatially
explicit simulation framework. e coloured dots represent the geographical location of the mtDNA lineages
fromnorthern Greece (n = 45), central Europe (n = 200) and northern Balkans (n = 83). Hunter-gatherers
(n = 19) are represented in green, Early Neolithic farmers (n = 177) in red, Middle and Late/Final Neolithic
farmers (n = 132) in blue and present-day Greeks (n = 319) in yellow. Name (indicated in numbers) and precise
chronology of the archaeological sites (locations) can be found in TableS4.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
7
Vol.:(0123456789)
Scientic Reports | (2022) 12:13474 | https://doi.org/10.1038/s41598-022-16745-8
www.nature.com/scientificreports/
e principle of this test, inspired by Bramanti etal.2, is to contrast the genetic dierentiation (Fst) observed
between samples from dierent chronological phases to that expected under a model of population continuity.
As Silva etal.88 showed, population structure has a signicant eect on the genetic dierentiation between serial
samples and, to get unbiased results, any such test must account for gene ow between subpopulations in the
studied area. We thus rst obtained ABC posterior samples of Nm values compatible with the observed genetic
diversity among the samples. Specically, we conducted 20,000 simulations under the SPC scenario (Table1)
with Nm values taken from 30 to 1000 using uniform distributions on the migration rate (mFA) and carrying
capacity (KFA) centered on the values used for the simulation of the Danubian route and equal to [0.3–0.5] and
[100–2000], respectively. We used larger priors to ensure a sucient exploration of this important parameter for
the population continuity test. Here, the Palaeolithic layer is only used to create the initial Neolithic population
source. Each simulation was summarized with three statistics quantifying the intrapopulation diversity of the
more ancient population sample among the pair under comparison: heterozygosity (H), pairwise dierences (π)
and number of segregating alleles (k). We then retained the 1,000 simulations closest to the observed statistics
and conrmed their match with the observed data using the marginal density P-value outputted by ABCtoolbox2
(i.e., requesting P > 0.05).
e 1000 Fst values associated with the retained simulations constitute samples from the expected Fst distri-
bution under a model of population continuity and with Nm values drawn from the ABC posterior. We used the
proportion of those Fst values larger than that observed as a one-tail P-value to reject the model of continuity
at the α = 5% threshold. If continuity is rejected, it suggests that the genetic dierentiation observed between
population samples cannot be explained by the stochastic processes of genetic dri, migration and sampling and
hence an additional process should be invoked. Such an event could involve a population replacement (complete
or partial) from a genetically distinct population occurring between the two sampling periods88.
Ethics approval. We were given permission by the Greek Ministry of Culture and Sports to sample and
extract DNA as well as to radiocarbon date all human remains mentioned in this study according to Greek law
for destructive sampling of archaeological material (Ν.3028/02).
Results
Authenticity of ancient DNA results. e preservation state of the bone and tooth samples from the
Mesolithic and Neolithic individuals was rather low. With few exceptions, the endogenous DNA content was
below 0.5% (TableS2). Only the petrous bone samples showed higher DNA content.
Out of 70 samples screened, 42 mitochondrial genomes from Neolithic Greece were enriched and sequenced
to a depth between 19 × and 300 × (TableS2). With the exception of St3A, Pal1 and Pal6, all results were replicated
by two independent extractions. Samples Mau1, Mau2, and Krk2 were replicated by PCR and Sanger sequencing,
which resulted in the same polymorphisms in the HVS-I region as determined by capture enrichment and NGS.
e rate of post-mortem deamination at the rst 5`position of DNA molecules ranged from 20 to 68% (mean
40%). DNA fragments averaged 86bp (54–148bp). e estimated fragment length obtained from 150bp single
end or 100bp paired end runs shows a strong correlation with the deamination rate at the 3position. Overall,
98.2% (6.6–99.9%) of all reads showed deamination patterns typical for highly degraded ancient DNA. Data
showing signs of contamination (Mau1: 86.6% authentic data, Krk2: 94.4% % authentic data) were cleaned using
PMDtools64 (TableS2).
Descriptive statistics of diversity. e Mesolithic and Neolithic mitochondrial lineages from Greece
correspond to the previously dened family of lineages—called haplogroups—H, T, K, J, N1, U, and HV (Fig.2,
TableS6). In our sample, the lineages belonging to Η and Κ have a frequency of 31% and 28.6%. Haplogroups
T1 and T2 are observed predominantly during the EN, whereas the haplogroup J occurs from the MN onwards.
Lineages belonging to haplogroup U* (U3, U4, U7 and U8) appear only during the LN and FN with a frequency
of 9.5% while haplogroup HV is observed only during the FN. e entire dataset shows no U5 dening muta-
tions (U5: position 3197, U5a: position 14793, U5b: 14182).
Gene diversity is at maximum and there is no dierence between the EN, M/LN and the FN in Greece as all
haplotypes are dierent within each sample (Ĥ = 1.0). Nucleotide diversity is low for all three Neolithic groups
varying from π ~ 0.0114±0.0067 for the Early, 0.0138±0.0079 for the Middle/Late and 0.0089±0.0056 for the
Final Neolithic group.
e genetic distance between the EN and the M/LN (Fst = 0.0, P = 0.501) and FN group in Greece (Fst = 0.0,
P = 0.419), as well as between the M/LN and the FN group (Fst = 0.0, P = 0.845), is not statistically signicant,
supporting an absence of genetic dierentiation between those populations.
Dierences on haplotypic variation between the Neolithic sites can be observed in Fig.2a. e computa-
tion of pairwise Fst between archaeological sites show that most, are genetically undierentiated (TableS7),
in keeping with the very low sample sizes which demands caution in interpretation (2 < n < 8, Franchthi and
arrounia were not included because they contain a single sequence). Among contemporaneous sites, EN Nea
Nikomedeia (n = 5, Nea3: 6379–6091cal BCE; Nea2: 6225–6075cal BCE) shows signicant genetic dierences
with Mavropigi-Fillotsairi (n = 4, Mau 16,333 ± 56cal BCE). EN Mavropigi-Fillotsairi (n = 4) is also dierentiated
from LN Toumba Kremastis Koiladas (n = 7).
e genetic distance between the populations of the wider Aegean region (EN in northwestern Anatolia,
EN, M/LN and FN in northern Greece) and the EN and MN in the Balkans and Carpathian Basin (Starcevo,
Körös, Alföld-Linear Pottery phase I) is low and not statistically signicant, supporting an absence of genetic
dierentiation between those populations. Signicant Fst values > 0.045 (P < 0.05) can be measured in central
Europe between the earlier and the nal Neolithic periods (TableS8).
Content courtesy of Springer Nature, terms of use apply. Rights reserved
8
Vol:.(1234567890)
Scientic Reports | (2022) 12:13474 | https://doi.org/10.1038/s41598-022-16745-8
www.nature.com/scientificreports/
In a multidimensional scaling (MDS) analysis, based on Fst values calculated from the mitochondrial
sequences (TableS8), all three Neolithic groups from present-day Greece fall close to each other (Fig.3). e
EN farmers from Greece plot on an axis connecting EN groups from northwestern Anatolia, Greece, the Balkans,
and central Europe on the right of the graph. A similar axis is observed for the M/LN groups, among which the
FN population from Greece also gures. On the contrary, FN central Europeans fall near the Palaeolithic and
Mesolithic hunter-gatherers. Moreover, all M/LN and FN groups from Greece, Balkans, and central Europe shi
to the le compared to their EN ancestors. Present-day populations from southeastern Europe and Hungary
form a distinct group that falls between Palaeolithic and Mesolithic hunter-gatherers and Neolithic farmers.
Structured population continuity test. e present-day inhabitants of northern Greece are signi-
cantly dierent from the Neolithic population from the same geographical area (Fst = 0.034, P = 0.002 between
EN and present-day Greek populations and Fst = 0.039, P < 0.001 between ML/FN and present-day Greek popu-
lations). Even if the observed genetic dierentiation is signicant, it may be due solely to stochastic processes of
sampling, local migration and genetic dri over time, without necessarily involving population turnover. e
structured population continuity test checks this possibility by taking into account the spatiotemporal variance
among sequences and it clearly rejects population continuity between the Neolithic sample and the present-day
population (P = 0.010 between EN and modern and 0.012 between ΜL/FN and modern). In contrast, population
continuity is not rejected between the earlier and later phases of the Neolithic (P = 0.338). e marginal density
P-values are 0.44 and 0.60 for the simulations of mitochondrial diversity in EN and ML/FN, respectively, show-
ing that the simulation framework is able to reproduce the empirical values and consequently that the structured
population continuity test is valid in the current context.
Figure2. (a) Haplogroup frequency at dierent Neolithic periods in Greece. (b) Haplogroup frequency at
dierent Neolithic sites in Greece (EN: Early Neolithic n = 13, MN: Middle Neolithic, n = 5, LN: Late Neolithic
n = 13, FN: Final Neolithic n = 11).
Content courtesy of Springer Nature, terms of use apply. Rights reserved
9
Vol.:(0123456789)
Scientic Reports | (2022) 12:13474 | https://doi.org/10.1038/s41598-022-16745-8
www.nature.com/scientificreports/
Neolithic spread along the Danube route. Estimation of the best scenario. e values presented in
Table2 show that all tested scenarios can reproduce the observed statistics (marginal density P-values > 0.58).
e cross-validation shows that the probability of recovery is similar for all four scenarios (0.49 < PR < 0.54).
Moreover, all results are robust to variation of the tolerance level δ.
We note that the two best scenarios are those where admixture between hunter-gatherers and farmers is
increasing with time (posterior probabilities of SN2 = 58.8% and SN4 = 38.5% at δ = 1.0), whereas the two sce-
narios with constant admixture can be signicantly rejected (SN1 and SN3 < 2% at any δ). Among the scenarios
with the same admixture model, it was not possible to distinguish whether admixture occurred all along the
Danubian route or only in central Europe (posterior probability of 48.1% for SN1 versus SN3 and 60.4% for SN2
versus SN4 at δ = 1.0).
Parameter estimation. For the best scenario SN2 (admixture increasing with time all along the Danubian
route), we estimated γ at 0.107 (Highest Density Interval 90 = [0.062–0.176], Table3, Fig. 4), the migration
decrease Mdec at 4819 times (HDI 90 = [1.00–14.103]), and KFA at 771 (HDI 90 = [539–974]). ese point esti-
mates reect the mode for γ and Mdec and the mean for KFA, as we found those to be the most accurate on
pseudo-observed data sets (Table3).
Figure3. MDS with the 47 Greek Neolithic samples (42 newly sequenced and 5 published,EL, plain circles)
and a reference panel of 26 ancient and present-day populations (2470 individualsin total), stress = 0.099 (EN:
Early Neolithic, MN: Middle Neolithic, LN: Late Neolithic, FN: Final Neolithic, EUHG: hunter-gatherers,
H_EUHG: Holocene EUHG, UP_EUHG: Upper Palaeolithic EUHG, LP_EUHG: Lower Palaeolithic EUHG,
NWTR: north western Turkey, EL: Greece, DE: Germany, HR: Croatia, HU: Hungary, grey stars = modern
populations abbreviations can be found on TableS3).
Table 2. Model choice results. Marginal densities, posterior probabilities, Bayes factors (Model i against the
others) and probability of recovery for each simulated scenario with a tolerance level of 1% (0.25% and 2.5% in
italic).
Tol. δ Scenario SN1 Scenario SN2 Scenario SN3 Scenario SN4
Marginal density P
0.25 0.893 0.970 0.868 0.798
1.00 0.858 0.936 0.877 0.584
2.50 0.869 0.845 0.873 0.441
Posterior probability
0.25 0.019 0.633 0.020 0.328
1.00 0.013 0.588 0.014 0.385
2.50 0.014 0.575 0.016 0.395
Bayes factor
0.25 0.019 1.727 0.021 0.487
1.00 0.013 1.426 0.015 0.626
2.50 0.014 1.351 0.016 0.653
Probability of recovery
0.25 0.442 0.440 0.442 0.517
1.00 0.540 0.493 0.509 0.522
2.50 0.466 0.468 0.490 0.523
Content courtesy of Springer Nature, terms of use apply. Rights reserved
10
Vol:.(1234567890)
Scientic Reports | (2022) 12:13474 | https://doi.org/10.1038/s41598-022-16745-8
www.nature.com/scientificreports/
To check if posterior estimates are unbiased, we inferred the distribution of quantiles and HDI of the true
values among the posterior distributions on pseudo-observed data sets, which are expected to be uniform84.
Uniformity was indeed rejected for γ (quantiles and HDI P < 0.001, Fig.S2), but not for Mdec (quantiles P = 0.131
and 0.863, Fig.S2). For KFA, uniformity was rejected for the quantiles at the 5% level (P = 0.030), but not for HDI
(P = 0.472, Fig.S2). We thus consider the posterior estimates to be unbiased for KFA and Mdec, but not for γ,
which is slightly overestimated (Fig.S2).
Discussion
Haplotypic variation among ancient sites. Among the 42 ancient individuals newly reported here, we
identied 7 dierent mtDNA haplogroups (K, H, HV, U, T, J and N1), all typical for populations of the Neolithic
and highly related to the Neolithic expansion67.
As far as the distribution of mitochondrial haplogroups is concerned, the prehistoric individuals with hap-
logroup K in Greece are particularly interesting. Haplogroup K is a common lineage in our Neolithic sample.
Previously, it has been used as a marker for the Neolithic expansion because it is virtually absent in Mesolithic
hunter-gatherers from central and northern Europe89. However, lineages belonging to K1a have been recorded
in two Mesolithic hunter-gatherers (eo1, eo5, ca. 7000 BCE) from the eopetra Cave in Greece3 making
them currently the only hunter-gatherers in Europe with a dierent haplogroup than U5. is lineage has been
also found in pre-pottery Neolithic individuals from Israel60 and in central Anatolian Early Neolithic farmers
(Boncuklu and Tepecik-Çilik4) signifying its presence in Anatolia and the Aegean even before the Neolithic
expansion.
Table 3. Parameter estimation results. Limits and characteristics of the prior and posterior distributions of
the parameters estimated for the Neolithic expansion along the Danubian route under Scenario SN2 with a
tolerance level δ of 1% (0.25% and 2.5% in italic). γ = assimilation rate between hunter-gatherer and farmer
layer; Mdec = factor of migration decreases aer reaching carrying capacity in farmers; KFA = carrying capacity
of the farmer demes. HDI = Highest Density Interval. e precision of the mode and the mean of the posterior
distributions are also given with three statistics (BIAS, RMSE and Factor2, see text for details).
Parameters Prior
distribution Tol. δ
Posterior distribution characteristics Estimation precision
Mode Mean HDI 50 HDI 90 BIAS mode/
mean RMSE mode/
mean Factor 2
mode/mean
γ 0.0–0.4
0.25 0.103 0.120 0.089–0.139 0.063–0.174 0.01/0.11 0.33/0.32 0.98/0.98
1.00 0.107 0.120 0.0870.135 0.0620.176 0.01/0.10 0.30/0.30 0.99/0.99
2.50 0.109 0.121 0.086–0.138 0.059–0.182 0.02/0.13 0.33/0.33 0.98/0.97
Mdec 1–20
0.25 6.729 7.620 2.289–7.576 1.00–14.005 0.26/0.51 0.57/0.48 0.68/0.77
1.00 4.819 7.501 2.2697.636 1.0014.103 0.23/0.50 0.58/0.49 0.69/0.77
2.50 4.915 7.662 1.907–7.558 1.00–14.222 0.32/0.51 0.55/0.47 0.71/0.77
KFA 500–1000
0.25 970 780 785–994 577–999 0.12/0.03 0.24/0.18 1.00/1.00
1.00 822 771 661884 539974 0.11/0.02 0.23/0.18 1.00/1.00
2.50 907 774 770–981 572–996 0.12/0.03 0.24/0.18 1.00/1.00
0
3
6
9
0.0 0.10.2 0.30.4
Assimilation rate (γ)
Density
Neolithic Farmers Carrying Capacity (
K
)
FA Neolithic Farmers migration decrease (
Mdec
)
0.0008
0.0012
0.0016
0.0020
0.0024
500600 700 800 900 1000
0.000
0.025
0.050
0.075
0.100
5101
52
0
Figure4. Prior (red line) and posterior (black line) distributions of the parameters estimated for the Neolithic
expansion along the Danubian route under Scenario SN2. Τhe assimilation rate (γ) corresponding to the
maximum gene ow from hunter-gatherer to the Neolithic farmer population, the carrying capacity of Neolithic
farmers (KFA), the ratio of decrease of migration rate in Neolithic farmers aer the colonization phase (Mdec).
Content courtesy of Springer Nature, terms of use apply. Rights reserved
11
Vol.:(0123456789)
Scientic Reports | (2022) 12:13474 | https://doi.org/10.1038/s41598-022-16745-8
www.nature.com/scientificreports/
e entire dataset shows no U5 dening mutations, which is the predominant lineage of western European
hunter-gatherers and is observed at high frequency in later phases of the central European Neolithic9,62 as well
as in Mesolithic and Neolithic Italy90 and the Mesolithic Iron Gates59. is is remarkable, as it suggests that the
local hunter-gatherer population in the Aegean and northern Greece may have been genetically distinct from
those in the northern Balkans and other parts of Europe—as supported by the mitochondrial DNA evidence of
eopetra cave3.
Haplogroup N1a1a is one of the two lineages -the other being K1a- observed in Early Neolithic farmers of
central Anatolia4. In Greece it is recorded only during the Middle and Late Neolithic Period.
Haplogroups N1, H, HV, J1, T1 and T2, have all been identied in Early Neolithic populations along the
Danubian Neolithisation route i.e., in present-day Serbia, Bulgaria, Hungary and Germany79,63,91,92.
Genetic anities between Neolithic sites in Greece, Anatolia and the Danube Neolithic expan‑
sion axis. Overall, we observe small dierences in the frequency of haplogroups between Early, Middle, Late
and Final Neolithic Greece. Consequently, genetic distances between these groups were low and not signicant,
supporting the scenario of a continuous, relatively homogeneous maternal population over the entire course of
the Neolithic in Greece. In addition, most archaeological sites were not genetically dierentiated, except for the
site of Nea Nikomedeia that displays signicant dierences to EN Mavropigi-Fillotsairi. Although an intriguing
nding especially in respect to the heterogenous and selective pattern observed in the material culture during
the earliest phases of the Neolithic in Greece54, we refrain from making further inferences given the small sample
size and our spatially explicit modelling results that indicate population continuity throughout the Neolithic.
Interestingly, the remarkable increase in the number of settlements from the Middle Neolithic onwards28,53,
the formation of more and larger communities that promoted more complex social structures and larger eco-
nomic networks, were not triggered by a strong population turnover, at least not in the maternal line. Instead,
endogenous population growth, possibly combined with increased local mobility, is likely to be the underlying
phenomenon.
is is also supported by the low dierences in gene and nucleotide diversity observed in the three Neolithic
datasets from Greece especially when compared to the values of other Neolithic populations from the Balkans
and central Europe93. Early Neolithic farmers from central Europe and Late Neolithic farmers from western
Anatolia show similarly low gene diversity4,8. An overall increase of haplotype diversity from the Aegean to
central Europe is only observed in early farmers along the Danubian route. is is in accordance with our results
from the spatially explicit modelling that support an increasing gradient of hunter-gatherer contribution to the
gene pool of farmers along the axis of the Danubian expansion, thus increasing genetic diversity by merging
two dierentiated genetic pools.
Population continuity between Neolithic populations but discontinuity with present‑day
Greeks. e continuity between the Early and the Later Periods of the Neolithic in Greece has been veried
by the application of an original test based on spatially explicit modelling, which considers ongoing local migra-
tion. is means that no fundamental population changes have taken place on the female side, but does not
exclude minor and/or male migration. is changes during the very last phase of the Neolithic period.Previous
work on two Final Neolithic (4500–4000 BCE) whole genomes from Greece, indicated gene-ow from popula-
tions with evidence of Caucasus hunter-gatherer-like ancestry3,8 that becomes stronger during the Early Bronze
Age94. A similar trend is visible on the MDS plot (Fig.3) as the Middle/Late and Final Neolithic groups shi away
from the Early Neolithic groups but the dierence is not big enough to result from population discontinuity
according to the results of the structured population continuity test. Although we could not identify a signicant
external gene-ow for the maternal line, these signals could be related to the developments that took place in
the Aegean at the end of the Neolithic. During the later phases of the Neolithic, maritime contacts and trade
exchange intensied. e coastal zones and the islands were now populated and, on the mainland, certain settle-
ments seem to have acquired considerable economic importance95, indicating that the more advanced economy
of the following BA was unfolding.
On the contrary, present-day Greeks are descended not only from Neolithic Aegeans. Our results support pop-
ulation discontinuity between the Neolithic era and present-day Greece due to a genetically distinct population(s)
immigrating to this region between those two periods. is observation is in line with the conclusion drawn
from the analysis of three Neolithic genomes from this area for which DNA was retrieved at the genomic scale3.
us, the results obtained for the maternal line are in accordance with those from biparental molecular markers.
In this line of evidence, the recent analysis of six Early (3300–2000 BCE) and Middle (~ 2000 BCE) Bronze Age
genomes from Greece showed that present-day northern-Greeks are genetically similar to 2000 BCE Aegeans
from the same region94. Although they derive part of their ancestry from Neolithic farmers, a Neolithic Caucasus-
like and BA Pontic-Caspian Steppe-like gene ow shaped the Aegean aer the Neolithic period and may explain
the population discontinuity we observe in our analyses.
Neolithic spread along the Danube route. To trace the further spread of the Neolithic population
from Greece to the Balkans and central Europe, we applied spatially explicit simulations and the ABC approach.
We tested four competing scenarios, that consider dierent conditions of admixture between Neolithic farm-
ers and hunter-gatherers: whether this happened everywhere in the simulation area or only in central Europe;
and whether this occurred constantly or increased in intensity during the cohabitation period. e dierent
scenarios were explored by varying three important parameters: (1) gene ow from hunter-gatherers to farmers,
(2) the maximum farmer’s density and (3) the migration rate for farmers aer their initial settlement. isinitial
period corresponds to the time it takes for a deme to reach 90% of its carrying capacity and it varies approxi-
Content courtesy of Springer Nature, terms of use apply. Rights reserved
12
Vol:.(1234567890)
Scientic Reports | (2022) 12:13474 | https://doi.org/10.1038/s41598-022-16745-8
www.nature.com/scientificreports/
mately between 200 and 400years (8–16 generations) depending on the combination of parameters considered
and the deme’s location. We found that the scenarios SN2 and SN4, under which admixture increases with time
along the Danubian route, tted the observed data much better than scenarios postulating constant admixture.
In contrast, the data was inconclusive regarding the geographic area where admixture took place, i.e., whether
hunter-gatherer assimilation occurred only in central Europe or also along the way from Greece. is diculty
in dierentiating between scenarios with and without admixture in the rst section of the Danubian route may
be attributed, at least partly, to the fact that hunter-gatherer data is available only for central Europe. e fact that
we modeled an extended cohabitation time in central Europe may also play a role because the resulting admix-
ture is always higher in central Europe than elsewhere, independently of the scenario simulated.
Our results propose a cultural process by which hunter-gatherer ancestry has increased in European popula-
tions since the Middle Neolithic62,63, which did not occur: by accumulation through constant assimilation of
hunter-gatherers into the farming community, but as the result of an increasing rate of assimilation, which did not
take place early in the process, but only once farming was fully established. In our model, the rate of assimilation
increases progressively throughout the duration of cohabitation between hunter-gatherers and farmers in each
deme, which varies according to the combination of parameters and the scenario considered. It is generally of
the order of 10 generations, but can go up to 100 generations in central Europe, i.e., roughly between 250 and
2500years. Such dynamics correspond roughly to the ‘leapfrog’ colonization model96,97 with a rapid colonization
of suitable niches for establishing agriculture followed by acculturation and genetic exchange with external non-
farmers98. Under the most probable scenario, the assimilation rate γ was estimated at 11%. However, the HDI and
quantile plots suggest that the point estimate for γ tends to be overestimated by about 20–25% (Table3), and we
thus conclude that along the Danubian route the proportion of contacts between hunter-gatherers and farmers
that resulted in hunter-gatherer assimilation was probably around or slightly less than 10%. is is in line with the
results of Silva etal.71, who estimated γ in central Europe to be ~ 2% from mtDNA (HDI90 = [0%–6%]) and ~ 9%
from autosomal data (HDI90 = [5–14%]). e admixture rates estimated in the two studies cannot be directly
compared, however, because they dier in the quantity they measure due to the dierent underlying admixture
models (constant over time in Silva etal.71, but increasing over time in the current study). Our study therefore
estimates a maximum admixture rate at the end of the cohabitation period (i.e., of a common land use), whereas
Silva etal.71, estimated an average value during the same period, which is logically lower. In addition, our study
includes a larger dataset covering the whole Danubian route, including the Balkans and Greece, whereas Silva
etal.71, was geographically limited to central Europe.
Another important, but unexpected insight from the simulation study is that the high migration rate required
to achieve a fast Neolithic spread between Greece and central Europe is only compatible with mitochondrial
diversity if the migration rate decreases substantially aer the original Neolithic settlement. In our simulation
framework, it happens a few centuries aer the colonization of a deme by farmers (i.e. aer approximately
200–400years), i.e., once 90% of the demes carrying capacity has been reached. We estimate that the dispersal
rate of farmers must decrease by at least two-fold aer the initial phase of a Neolithic settlement, and we estimate
a ve-fold reduction of the migration rate. is decrease of mobility aer the phase of population growth that fol-
lowed the initial Neolithic settlement (a few centuries in our simulation framework) corroborates anthropological
observations99 as well as the estimation made from paleogenomic data with a dierent statistical approach100 and
may be related to an increased sedentism89. is result is consistent with a gradual, almost plasmodic expansion
of the Neolithic lifeway and the observation that each new settlement step was followed not only by a regional
increase in population size, but also by pauses in expansion of up to several hundred years12.
Finally, the value estimated for the farmer’s carrying capacity points to the mean value of the range explored
(modemean = 771, prior = [500, 1000]), which may be translated to a density of ~ 0.46 individuals/km2, lower
than the maximum density of 0.6 individuals/km2 estimated for the Linear Pottery culture (LBK)76, but as the
condence interval is large, one should treat this estimate with some caution.
Conclusion
We performed spatially explicit modelling with mitochondrial data whose resolution is lower than genomic mul-
tilocus data but which are abundant in an area from which whole genome data are sparse. Our results prove for
the maternal lineage a homogeneous population inhabiting northern Greece throughout the Neolithic, without
signicant external gene ow until the end of this period. Our best tting scenario suggests that Neolithic farmers
expanded from the Aegean area to central Europe during a rapid migration event, with initially little admixture
with local hunter-gatherers and increased gene ow over time. According to our results, the very high migra-
tion rate during the rst phase of the Neolithisation process necessary in order to t the temporal framework of
the Neolithic spread must be followed by a substantial decrease in mobility to be compatible with the observed
mitochondrial diversity. is phase of prolonged cohabitation may have been accompanied by an increasing
local assimilation of hunter-gatherers and may explain the resurgence of hunter-gatherer ancestry from the
Middle Neolithic onwards. In conclusion, the simulation approach presented here provides a solid framework
for investigating the mechanisms of past population dynamics, which may be used for investigating ancestral
genetic patterns in various spatio-temporal contexts of human migration and evolution.
Data availability
e accession number for the bam les of the ancient mtDNA genomes reported in study is European Nucleotide
Archive: PRJEB52148 (https:// www. ebi. ac. uk/ ena/ brows er/ view/ PRJEB 52148). e setting les and executable
used for the simulation part of the paper can be accessed in the Zenodo public repository under: https:// doi.
org/ 10. 5281/ zenodo. 63856 10. Supplementary information to the present article, in addition to the supplemental
Content courtesy of Springer Nature, terms of use apply. Rights reserved
13
Vol.:(0123456789)
Scientic Reports | (2022) 12:13474 | https://doi.org/10.1038/s41598-022-16745-8
www.nature.com/scientificreports/
tables and gures, are available in Document S1. e sources of the published sequences used for comparison
with our newly produced sequences are provided in TablesS3 and S4.
Received: 25 February 2022; Accepted: 14 July 2022
References
1. Bollongino, R. et al. 2000 years of parallel societies in stone age central Europe. Science 342, 479–481 (2013).
2. Bramanti, B. et al. Genetic discontinuity between local hunter-gatherers and central Europe’s rst farmers. Science 326, 137–140
(2009).
3. Hofmanová, Z. et al. Early farmers from across Europe directly descended from Neolithic Aegeans. Proc. Natl. Acad. Sci. 113,
6886–6891 (2016).
4. Kilinç, G. M. et al. e demographic development of the rst farmers in Anatolia. Curr. Biol. 26, 2659–2666 (2016).
5. Kilinç, G. M. et al. Archaeogenomic analysis of the rst steps of Neolithization in Anatolia and the Aegean. Proc. R. Soc. B Biol.
Sci. 284, 20172064 (2017).
6. Marcus, J. H. et al. Genetic history from the Middle Neolithic to present on the Mediterranean island of Sardinia. Nat. Commun.
11, 939 (2020).
7. Mathieson, I. et al. Genome-wide patterns of selection in 230 ancient Eurasians. Nature 528, 499–503 (2015).
8. Mathieson, I. et al. e genomic history of southeastern Europe. Nature 555, 197–203 (2018).
9. L azaridis, I. et al. Ancient human genomes suggest three ancestral populations for present-day Europeans. Nature 513, 409–413
(2014).
10. Omrak, A. et al. Genomic evidence establishes Anatolia as the source of the European Neolithic gene pool. Curr. Biol. 26, 270–275
(2016).
11. Horejs, B. et al. e Aegean in the early 7th millennium BC: maritime networks and colonization. J. World Prehistory 28, 289–330
(2015).
12. Shennan, S. J. e First Farmers of Europe: An Evolutionary Perspective (Cambridge University Press, 2018).
13. Douka, K., Efstratiou, N., Hald, M. M., Henriksen, P. S. & Karetsou, A. Dating Knossos and the arrival of the earliest Neolithic
in the southern Aegean. Antiquity 91, 304–321 (2017).
14. Efstratiou, N., Karetsou, A. & Ntinou, M. e Neolithic Settlement of Knossos in Crete New Evidence for the Early Occupation of
Crete and the Aegean Islands (INSTAP Academic Press, 2013).
15. Perlès, C., Quiles, A. & Valladas, H. Early seventh-millennium AMS dates from domestic seeds in the Initial Neolithic at Franch-
thi Cave (Argolid, Greece). Antiquity 87, 1001–1015 (2013).
16. Papathanasiou, A. Alepotrypa cave: e site description and its cultural and chronological range. In Neolithic Alepotrypa Cave
in the Mani (eds Papathanassiou, A. et al.) 10–24 (Oxbow Books, 2018).
17. Sampson, A. Palaeolithic and Mesolithic sailors in the Aegean and the Near East (Cambridge Scholars Publishing, 2019).
18. Sampson, A. e excavation at Sarakenos Cave in Boeotia, Greece (Cambridge Scholars Publishing, Newcastle 2022).
19. Maniatis, Y. Radiocarbon dating of the major cultural phases in prehistoric Macedonia: Recent developments. In A Century of
Research in Prehistoric Macedonia 1912–2012 (eds Stefani, E. et al.) 205–222 (Springer, 2014).
20. Kotsakis, K. Domesticating the periphery. Pharos 20, 41–73 (2014).
21. Karamitrou-Mentessidi, G., Efstratiou, N., Kaczanowska, M. & Kozłowski, J. Early Neolithic settlement of mavropigi in Western
Greek Macedonia. Eurasian Prehist. 12, 47–116 (2015).
22 . Reingruber, A. e Argissa Magoula and the beginning of the Neolithic in essaly. in How did Farming Reach Europe? Anatolian-
European relations; Anatolian-European Relations from the Second Half of the 7th rough the First Half of the 6th Millennium
cal BC, 155–172 (2008).
23. Reingruber, A. et al. Neolithic essaly: Radiocarbon dated periods and phases. Doc. Praehist. 44, 34–53 (2017).
24. Krauß, R., Marinova, E., De Brue, H. & Weninger, B. e rapid spread of early farming from the Aegean into the Balkans via
the Sub-Mediterranean-Aegean Vegetation Zone. Quat. Int. 496, 24–41 (2018).
25. Lichardus-Itten, M., Demoule, J.-P., Perničeva, L., Grębska-Kulova, M. & Kulov, I. Kovačevo, an Early Neolithic site in South-
West Bulgaria and its importance for European Neolithization. in Aegean-MarmaraBlack Sea: e Present State of Research on
the Early Neolithic (eds. Gatsov, I. & Schwarzberg, H.) 83–94 (Beier & Beran, 2006).
26. Özdoğan, M. Neolithic cultures at the contact zone between Anatolia and the Balkans-diversity and homogeneity at the Neolithic
frontier. in Aegean--Marmara--Black Sea: e Present State of Research on the Early Neolithic (eds. Gatsov, I. & Schwarzberg, H.)
21–28 (Beier and Beran, 2006).
27. Reingruber, A. & issen, L. Depending on 14 C data: Chronological frameworks in the Neolithic and Chalcolithic of South-
eastern Europe. Radiocarbon 51, 751–770 (2009).
28. Andreou, S., Fotiadis, M. & Kotsakis, K. Review of Aegean prehistory V: e Neolithic and bronze age of Northern Greece. Am.
J. Archaeol. 100, 537 (1996).
29. Sampson, A. Late Neolithic remains at arrounia Euboea: A model for the seasonal use of settlements and caves. BSA 87,
61–101 (1992).
30. Sampson, A. e Neolithic and Bronze Age Occupation of the Sarakenos Cave in Boeotia. Cave Settlement Patterns and Population
Movements in Central and Southern Greece. (University of the Aegean and Polish Academy, 2008).
31. Kyparissi-Apostolika, N. e essalian Mesolithic: Evidence from eopetra Cave. J. Greek Archaeol. 6, 25–43 (2021).
32. Sampson, A. et al. Sarakenos Cave in Boeotia, from the palaeolithic to the early bronze age. Eurasian Prehistory 6, 1–33 (2009).
33. Cullen, T. Mesolithic mortuary ritual at Franchthi Cave, Greece. Antiquity 69, 270–289 (1995).
34. Facorellis, Y. Radiocarbon dates from archaeological sites in caves and rockshelters in Greece. in Stable Places and Changing
Perceptions: Cave Archaeology in Greece (eds. Mavridis, F. & Jensen, J. T.) 19–72 (BAR International Series 2528, 2013).
35 . Sampson, A., Kaczanowska, M. & Kozłowski, J. K. Τhe Prehistory of the Island of Kythnos and the Mesolithic Settlement at Maroulas
(Polish Academy of Sciences and Arts and University of the Aegean, 2010).
36. Efstratiou, N., Biagi, P. & Elisabetta, S. e epipalaeolithic site of ouriakos on the Island of Lemnos and its place in the late
pleistocene peopling of the east Mediterranean region. Adalya XVII (2014).
37. Sampson, A. e Cave of the Cyclops: Mesolithic and Neolithic Networks in the Northern Aegean, Greece. Intra-Site Analysis, Local
Industries, and Regional Site Distribution. (INSTAP Academic Press, 2008).
38. Sampson, A. e Cave of the Cyclops: Mesolithic and Neolithic Networks in the Northern Aegean, Greece II: Bone Tool Industries,
Dietary Resources and the Paleoenvironment, and Archaeometrical Studies (INSTAP Academic Press, 2011).
39. Kaczanowska, M., Kozłowski, J. K. & Sampson, A. Le Mésolithique du bassinégéen. Études Balk. 15, 85–99 (2008).
40. Takaoğlu, T., Korkut, T., Erdoğu, B. & Işın, G. Archaeological evidence for 9th and 8th millennia BC at Girmeler Cave near Tlos
in SW Turkey. Doc. Praehist. 41, 111–118 (2014).
41. Erek, C. M. A new epi-paleolithic site in Northeast Mediterranean Region: Direkli Cave (Kahramanmaraş, Turkey). Adalya XIII
1–17 (2010).
Content courtesy of Springer Nature, terms of use apply. Rights reserved
14
Vol:.(1234567890)
Scientic Reports | (2022) 12:13474 | https://doi.org/10.1038/s41598-022-16745-8
www.nature.com/scientificreports/
42. Kaczanowska, M. & Kozowski, J. K. Mesolithic Obsidian networks in the Aegean. in Unconformist Archaeology. Papers in honour
of Paolo Biagi (ed. Starnini, E.) 17–26 (BAR International Series 2528, 2013).
43. Sampson, A. e Mesolithic hunter-gatherers in the Southeastern Mediterranean and their contribution in the Neolithisation
of the Aegean. Archaeol. Cult. 1, 11 (2018).
44. Baird, D. Pinarbasi, from Epipalaeolithic campsite to sedentarising village in central Anatolia. in Neolithic in Turkey new excava-
tions, new discoveries. (eds. Ozdogan, M. & Basgelen, N.) 1–36 (Arkeoloji ve Sanat Yayinlari, 2012).
45. Özbaşaran, M. Re-starting at Aşıklı. Anatolia Antiq. 19, 27–37 (2011).
46. Kotsakis, A. et al. Epidemiological characteristics, clinical outcomes and management patterns of metastatic breast cancer
patients in routine clinical care settings of Greece: Results from the EMERGE multicenter retrospective chart review study.
BMC Cancer 19, 88 (2019).
47. Kaczanowska, M. & Kozlowski, J. K. Lithic industry from the aceramic levels at Knossos (Crete, Greece): An alternative approach.
Eurasian Prehist. 8, 67–87 (2011).
48. Vitelli, K. D. Franchthi Neolithic Pottery 1: Classication and Ceramic Phases 1 and 2, Fascicle 8 (Excavations at Franchthi Cave)
(Indiana University Press, 1993).
49. Özdoğan, M. On arrows and sling missiles: what happened to the arrows? in Mauer Schau: Festschri für Manfred Korfmann
(ed. Aslan, I. R.) 437–444 (Greiner, 2002).
50. Urem-Kotsou, D. & Kotsakis, K. Pottery, cuisine and community in the Neolithic of North Greece. in Cooking up the Past: Food
and Culinary Practices in the Neolithic and Bronze Aegean (eds. Renard, J. & Mee, C.) 225–246 (Oxbow Books, 2007).
51. Triantaphyllou, S. Living with the Dead: a Re-consideration of Mortuary Practices in the Greek Neolithic. in Escaping the
Labyrinth: e Cretan Neolithic in Context (eds. Isaakidou, V. & Tomkins, P.) 139–157 (Oxbow Monographs, 2008).
52. Kakavakis, O. Chipped Stone Aspects of the Interaction among Neolithic Communities of Northern Greece. in Communities,
Landscapes, and Interaction in Neolithic Greece 434–445 (Berghahn Books, 2018). https:// doi. org/ 10. 2307/j. ctvw0 49k3. 35.
53. Demoule, J.-P. & Perlès, C. e Greek Neolithic: A new review. J. World Prehist. 7, 355–416 (1993).
54. Perlès, C. e Early Neolithic in Greece (Cambridge University Press, 2003).
55. Pentedeka, A. Pottery Exchange Networks During Middle and Late Neolithic in essaly (Aristotle University of essaloniki,
2008).
56. Urem-Kotsou, D. et al. Patterns in Contemporaneous Ceramic Traditions. in Communities, Landscapes, and Interaction in
Neolithic Greece 324–338 (Berghahn Books, 2017). https:// doi. org/ 10. 2307/j. ctvw0 49k3. 28.
57. Urem-Kotsou, D., Papaioannou, A., Papadakou, T., Saridaki, N. & Intze, Z. Pottery and stylistic boundaries. Early and middle
neolithic pottery in Macedonia. in A century of research in prehistoric Macedonia 1912–2012 (eds. Stefani, E., Merousis, N. &
Dimoula, A.) 505–517 (2014).
58. Allento, M. E. et al. Population genomics of Bronze Age Eurasia. Nature 522, 167–172 (2015).
59. González-Fortes, G. et al. Paleogenomic evidence for multi-generational mixing between neolithic farmers and mesolithic
hunter-gatherers in the Lower Danube Basin. Curr. Biol. 27, 1801–1810 (2017).
60. Lazaridis, I. et al. Genomic insights into the origin of farming in the ancient Near East. Nature 536, 419–424 (2016).
61. Lazaridis, I. et al. Genetic origins of the Minoans and Mycenaeans. Nature 548, 214–218 (2017).
62. Brandt, G. et al. Ancient DNA reveals key stages in the formation of Central European mitochondrial genetic diversity. Science
342, 257–261 (2013).
63. Lipson, M. et al. Parallel palaeogenomic transects reveal complex genetic history of early European farmers. Nature 551, 368–372
(2017).
64. Posth, C. et al. Reconstructing the deep population history of Central and South America. Cell 175, 1185–1197 (2018).
65. Kircher, M., Sawyer, S. & Meyer, M. Double indexing overcomes inaccuracies in multiplex sequencing on the Illumina platform.
Nucleic Acids Res. 40, e3–e3 (2012).
66. Ginolhac, A., Rasmussen, M., Gilbert, M. T. P., Willerslev, E. & Orlando, L. mapDamage: Testing for damage patterns in ancient
DNA sequences. Bioinformatics 27, 2153–2155 (2011).
67. Fu, Q. et al. A revised timescale for human evolution based on ancient mitochondrial genomes. Curr. Biol. 23, 553–559 (2013).
68. Excoer, L. & Lischer, H. E. L. Arlequin suite ver 3.5: A new series of programs to perform population genetics analyses under
Linux and Windows. Mol. Ecol. Resour. 10, 564–567 (2010).
69. Marchi, N. et al. e genomic origins of the world’s rst farmers. Cell 185, 1842-1859.e18 (2022).
70. Currat, M. & Excoer, L. e eect of the Neolithic expansion on European molecular diversity. Proc. R. Soc. B Biol. Sci. 272,
679–688 (2005).
71. Silva, N. M., Rio, J., Kreutzer, S., Papageorgopoulou, C. & Currat, M. Bayesian estimation of partial population continuity using
ancient DNA and spatially explicit simulations. Evol. Appl. 11, 1642–1655 (2018).
72. Ray, N., Currat, M., Foll, M. & Excoer, L. SPLATCHE2: A spatially explicit simulation framework for complex demography,
genetic admixture and recombination. Bioinformatics 26, 2993–2994 (2010).
73. Alroy, J. A multispecies overkill simulation of the end-pleistocene megafaunal mass extinction. Science 292, 1893–1896 (2001).
74. Steele, J., Adams, J. & Sluckin, T. Modelling Paleoindian dispersals. World Archaeol. 30, 286–305 (1998).
75. Hill, W. G. Eective size of populations with overlapping generations. eor. Popul. Biol. 3, 278–289 (1972).
76. Zimmermann, A., Hilpert, J. & Wendt, K. P. Estimations of population density for selected periods between the Neolithic and
AD 1800. Hum. Biol. 81, 357–380 (2009).
77. Volterra, V. Fluctuations in the abundance of a species considered mathematically. Nature 118, 558–560 (1926).
78. Lotka, A. e growth of mixed populations: Two species competing for acommon food supply. J. Washingt. Acad. Sci. 22, 461–469
(1932).
79. Skoglund, P. et al. Separating endogenous ancient DNA from modern day contamination in a Siberian Neandertal. Proc. Natl.
Acad. Sci. 111, 2229–2234 (2014).
80. Leuenberger, C. & Wegmann, D. Bayesian computation and model selection without likelihoods. Genetics 184, 243–252 (2010).
81. Wegmann, D., Leuenberger, C., Neuenschwander, S. & Excoer, L. ABCtoolbox: A versatile toolkit for approximate Bayesian
computations. BMC Bioinform. 11, 116 (2010).
82. Kimura, M. A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide
sequences. J. Mol. Evol. 16, 111–120 (1980).
83. Pritchard, J. K., Seielstad, M. T., Perez-Lezaun, A. & Feldman, M. W. Population growth of human Y chromosomes: A study of
Y chromosome microsatellites. Mol. Biol. Evol. 16, 1791–1798 (1999).
84. Wegmann, D., Leuenberger, C. & Excoer, L. Ecient approximate bayesian computation coupled with markov chain Monte
Carlo without likelihood. Genetics 182, 1207–1218 (2009).
85. Cook, S. R., Gelman, A. & Rubin, D. B. Validation of soware for Bayesian models using posterior quantiles. J. Comput. Graph.
Stat. 15, 675–692 (2006).
86. Irwin, J. et al. Mitochondrial control region sequences from northern Greece and Greek Cypriots. Int. J. Legal Med. 122, 87–89
(2008).
87. Silva, N. M., Rio, J. & Currat, M. Investigating population continuity with ancient DNA under a spatially explicit simulation
framework. BMC Genet. 18, 114 (2017).
Content courtesy of Springer Nature, terms of use apply. Rights reserved
15
Vol.:(0123456789)
Scientic Reports | (2022) 12:13474 | https://doi.org/10.1038/s41598-022-16745-8
www.nature.com/scientificreports/
88. Isern, N., Fort, J. & de Rioja, V. L. e ancient cline of haplogroup K implies that the Neolithic transition in Europe was mainly
demic. Sci. Rep. 7, 11229 (2017).
89. Antonio, M. L. et al. Ancient Rome: A genetic crossroads of Europe and the Mediterranean. Science 366, 708–714 (2019).
90. Gamba, C. et al. Genome ux and stasis in a ve millennium transect of European prehistory. Nat. Commun. 5, 5257 (2014).
91. Brunel, S. et al. Ancient genomes from present-day France unveil 7,000 years of its demographic history. Proc. Natl. Acad. Sci.
117, 12791–12798 (2020).
92. Kreutzer, S. Populations Genetische Analyse Prähistorischer Individuen aus Griechenland (Johannes Gutenberg University of
Mainz, 2017).
93 Clemente, F. et al. e genomic history of the Aegean palatial civilizations. Cell 184, 2565-2586.e21 (2021).
94. Broodbank, C. e Early Bronze Age in the Cyclades. in e Cambridge Companion to the Aegean Bronze Age 47–76 (Cambridge
University Press, 2008). https:// doi. org/ 10. 1017/ CCOL9 78052 18144 47. 003.
95. Zvelebil, M. e agricultural transition and the origins of Neolithic society in Europe. Doc. Praehist. 28, 1–26 (2001).
96. Zvelebil, M. e social context of the agricultural transition in Europe. in Archaeogenetics: DNA and the population prehistory
of Europe (eds. Renfrew, C. & Boyle, K.) 57–79 (2000).
97. Bentley, D. J. et al. DNA ligase I null mouse cells show normal DNA repair activity but altered DNA replication and reduced
genome stability. J. Cell Sci. 115, 1551–1561 (2002).
98 . Ru, C. B. et al. Gradual decline in mobility with the adoption of food production in Europe. Proc. Natl. Acad. Sci. 112, 7147–7152
(2015).
99. Loog, L. et al. Estimating mobility using sparse data: Application to human genetic variation. Proc. Natl. Acad. Sci. 114, 12213–
12218 (2017).
100. Bocquet-Appel, J.-P., Naji, S., Vander Linden, M. & Kozlowski, J. K. Detection of diusion and contact zones of early farming
in Europe from the space-time distribution of 14C dates. J. Archaeol. Sci. 36, 807–820 (2009).
Acknowledgements
We would like to thank the Greek Ministry of Culture and Sports and the Ephorate of Antiquities of Pal-
aeoanthropology and Speleology, essaloniki City, essaloniki Region, Kozani, Pieria, Imathia, Rhodopi
for providing the necessary sampling permissions. We thank the Bavarian State collection for Anthropology
and Palaeoanatomy, namely Michaela Harbeck, for providing skeletal samples from Otzing, Essenbach, and
Dillingen and the General Direction for Cultural Heritage of Rhineland-Palatinate, Speyer, Germany for pro-
viding skeletal samples from Herxheim. We also thank Elissavet Ganiatsou for help with editing the references
and the supplementary information. NS, MC, JB and CP were nanced by the Marie Skłodowska-Curie actions
ITN “BEAN”. NS, JR and MC were nancially supported by Grants 31003A_156853 and 31003A_182577 (to
MC) from the Swiss National Science Foundation. CP, JB were nanced by the Humboldt foundation. CP, SK
and JB were nanced by the German Science Foundation BU 1403/6-1. CP, JB, AS, were co-nanced by the
Greek-German bilateral cooperation program 2017 (General Secreteriat for Research and Innovation, Ministry
of Development and Investments, Greece, and Federal Ministry of Education and Research-BMBF, Germany)
project BIOMUSE-0195 funded by the Operational Programme “Competitiveness, Entrepreneurship and Inno-
vation” (NSRF 2014-2020) and co-nanced by Greece and the European Union (EU Social Fund and European
Regional Development Fund).
Author contributions
J.B., M.C. and C.P. conceived the project idea and designed the study. S.K. and C.P. performed laboratory work.
N.S., S.K., J.R., A.S., D.W., M.C., performed population genetic analyses. S.K. compiled reference datasets. C.P.,
S.T., K.K., D.U.-K., P.H., N.E., S.K., G.K.-M., F.A., A.Ch.-M., M.P., C.Z., A.SA., A.P., K.V., T.C., A.K.-A., A.Z.-L.,
J.P. contributed samples and background archaeological information. M.C. and C.P. wrote the main manuscript
text with critical input of S.K., N.S., D.W., J.B. and all co-authors. All authors reviewed the manuscript.
Competing interests
e authors declare no competing interests.
Additional information
Supplementary Information e online version contains supplementary material available at https:// doi. org/
10. 1038/ s41598- 022- 16745-8.
Correspondence and requests for materials should be addressed to M.C.orC.P.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
Open Access is article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the
Creative Commons licence, and indicate if changes were made. e images or other third party material in this
article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.
© e Author(s) 2022
Content courtesy of Springer Nature, terms of use apply. Rights reserved
1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com
... In fact, it has been found that the Bronze Age genomes from Greece have shown that present-day northern-Greeks are genetically similar to ~2000 BCE Aegeans from the same region. Although they derive part of their ancestry from Neolithic farmers, a population homogeneity in Neolithic Greece, and a Neolithic Caucasus-like and Bronze Age Yamnaya Steppe-like gene flow that shaped the Aegean after the Neolithic period and may explain the population discontinuity observed in recent analyses (Broodbank 2008;Silva et al., 2022)). Yet the Mycenaeans little difference from Minoans derives from additional ancestry related to the hunter-gatherers of eastern Europe and Siberia, introduced via a proximal source related to the inhabitants of the Eurasian steppe and not from Armenia. ...
Article
Full-text available
Alerted by a trio of papers that appear in the journal SCIENCE today (26 th August 2022) I present briefly the essential findings with some extension on the hot subject of ancient DNA studies over all the World but mainly the SE Mediterranean and Euro-Asian neighborhood. These three papers summarize the results of a remarkable Ancient DNA project and involving 202 co-authors. These studies draw on published archaeogenetic records and new DNA analysis of 727 individuals who lived during the Copper and Bronze Ages. These people populated the "Southern Arc," a swath of land connecting Europe to West Asia through Anatolia including coastal Levant. The findings are critically discussed in the wider frame of aDNA results, the migration and mobilities for the past about 8000 years ago and linked with the non-linear evolution processes of ancient cultures. Particular emphasis here is given to the 2016 expedition at the Late Helladic site of Kastrouli near Delphi in Greece, that contribute to these studies.
Book
Full-text available
The book details the importance of the many years of excavation in Sarakenos Cave, Greece, carried out from 1994 until today. The book discusses the cultural sequence from the time when Neanderthals settled in the cave around 40000 years BP until the end of the Bronze Age. The site has been proved to be the largest inhabited cave of the prehistoric period in Greece with 12 phases of habitation. The most important is the Neolithic period from the beginning of the 7th millennium until the end of 4th millennium BC. A special significance has the beginning of the Initial Neolithic with a full Neolithic package, which is the oldest so far in the Aegean area (7980+50 BP, 6976-6685 cal. BC).
Article
Full-text available
The precise genetic origins of the first Neolithic farming populations in Europe and Southwest Asia, as well as the processes and the timing of their differentiation, remain largely unknown. Demogenomic modeling of high-quality ancient genomes reveals that the early farmers of Anatolia and Europe emerged from a multiphase mixing of a Southwest Asian population with a strongly bottlenecked western hunter-gatherer population after the last glacial maximum. Moreover, the ancestors of the first farmers of Europe and Anatolia went through a period of extreme genetic drift during their westward range expansion, contributing highly to their genetic distinctiveness. This modeling elucidates the demographic processes at the root of the Neolithic transition and leads to a spatial interpretation of the population history of Southwest Asia and Europe during the late Pleistocene and early Holocene.
Article
Full-text available
The Cycladic, the Minoan, and the Helladic (Mycenaean) cultures define the Bronze Age (BA) of Greece. Urbanism, complex social structures, craft and agricultural specialization, and the earliest forms of writing characterize this iconic period. We sequenced six Early to Middle BA whole genomes, along with 11 mitochondrial genomes, sampled from the three BA cultures of the Aegean Sea. The Early BA (EBA) genomes are homogeneous and derive most of their ancestry from Neolithic Aegeans, contrary to earlier hypotheses that the Neolithic-EBA cultural transition was due to massive population turnover. EBA Aegeans were shaped by relatively small-scale migration from East of the Aegean, as evidenced by the Caucasus-related ancestry also detected in Anatolians. In contrast, Middle BA (MBA) individuals of northern Greece differ from EBA populations in showing ∼50% Pontic-Caspian Steppe-related ancestry, dated at ca. 2,600-2,000 BCE. Such gene flow events during the MBA contributed toward shaping present-day Greek genomes.
Article
Full-text available
Sarakenos Cave, occupied since Late Palaeolithic, is of crucial importance for the study of the Mesolithic/Neolithic interface. There is a striking contrast between the isolated Mesolithic, particularly Late Mesolithic occupations with flake technology adapted to the local bad quality raw materials, and with subsistence economy based on fowling and plant gathering, also adapted to the local environmental conditions, and the Early Neolithic groups arriving some 120-140 radiocarbon years later, with imported raw materials, mocroblade technology and animal breeding. The Middle and Late Neolithic occupations are represented by numerous symbolic objects such as figurines indicating the important role of the cave in ritual activities.
Article
Full-text available
The island of Sardinia has been of particular interest to geneticists for decades. The current model for Sardinia’s genetic history describes the island as harboring a founder population that was established largely from the Neolithic peoples of southern Europe and remained isolated from later Bronze Age expansions on the mainland. To evaluate this model, we generate genome-wide ancient DNA data for 70 individuals from 21 Sardinian archaeological sites spanning the Middle Neolithic through the Medieval period. The earliest individuals show a strong affinity to western Mediterranean Neolithic populations, followed by an extended period of genetic continuity on the island through the Nuragic period (second millennium BCE). Beginning with individuals from Phoenician/Punic sites (first millennium BCE), we observe spatially-varying signals of admixture with sources principally from the eastern and northern Mediterranean. Overall, our analysis sheds light on the genetic history of Sardinia, revealing how relationships to mainland populations shifted over time.
Article
Full-text available
Ancient Rome was the capital of an empire of ~70 million inhabitants, but little is known about the genetics of ancient Romans. Here we present 127 genomes from 29 archaeological sites in and around Rome, spanning the past 12,000 years. We observe two major prehistoric ancestry transitions: one with the introduction of farming and another prior to the Iron Age. By the founding of Rome, the genetic composition of the region approximated that of modern Mediterranean populations. During the Imperial period, Rome’s population received net immigration from the Near East, followed by an increase in genetic contributions from Europe. These ancestry shifts mirrored the geopolitical affiliations of Rome and were accompanied by marked interindividual diversity, reflecting gene flow from across the Mediterranean, Europe, and North Africa
Article
The Mesolithic from Theopetra Cave was initially presented at the 1st Prehistoric Conference held in Kastoria in 1998,2 and later in 2000 in Documenta Praehistorica and the volume edited by Galanidou and Perlès (2003). Since then, new evidence was produced during the technical works carried out for the development and promotion of the cave (2005–2008), which provided the opportunity to carry out further excavation. Almost two decades later, this paper sets out to, if not provide definite answers regarding the period in Thessaly and Greece generally, at least to contribute to the discussion by presenting new evidence from Theopetra Cave. During these years, apart from our additional excavations, more Mesolithic sites were investigated, mainly on the Aegean islands, including Cyclope Cave on Youra, Maroulas on Kynthos, Kerame on Ikaria, as well as those on the island of Astypalaia8 more recently. Additional sites from mainland Greece include Boila Rockshelter in Epirus, those in the Kopais Basin, and others in the Peloponnese, including Klissoura Cave I and in the Kandia area of the Argolid. The presence of Mesolithic sites on the islands refutes models of the period that had been presented in the literature during previous decades. At this point, it is worth reiterating the following comment made by Demetrios Theocharis in 1967. There are not enough reasons to explain the disappearance of Palaeolithic populations at the end of the Pleistocene and it seems irrational to assume that adaptation here was more difficult than anywhere else. He concludes by saying that if there is any hope of finding the period that connects the Palaeolithic with the Neolithic, we need to undertake more research in caves.
Article
Genomic studies conducted on ancient individuals across Europe have revealed how migrations have contributed to its present genetic landscape, but the territory of present-day France has yet to be connected to the broader European picture. We generated a large dataset comprising the complete mitochondrial genomes, Y-chromosome markers, and genotypes of a number of nuclear loci of interest of 243 individuals sampled across present-day France over a period spanning 7,000 y, complemented with a partially overlapping dataset of 58 low-coverage genomes. This panel provides a high-resolution transect of the dynamics of maternal and paternal lineages in France as well as of autosomal genotypes. Parental lineages and genomic data both revealed demographic patterns in France for the Neolithic and Bronze Age transitions consistent with neighboring regions, first with a migration wave of Anatolian farmers followed by varying degrees of admixture with autochthonous hunter-gatherers, and then substantial gene flow from individuals deriving part of their ancestry from the Pontic steppe at the onset of the Bronze Age. Our data have also highlighted the persistence of Magdalenian-associated ancestry in hunter-gatherer populations outside of Spain and thus provide arguments for an expansion of these populations at the end of the Paleolithic Period more northerly than what has been described so far. Finally, no major demographic changes were detected during the transition between the Bronze and Iron Ages.