Europe's First Farmers

new

Admin
Administrator

Posts: 73,016

Europe's First Farmers Mar 1, 2024 22:34:29 GMT

Quote

Post by Admin on Mar 1, 2024 22:34:29 GMT

Fig. 4: Genetic legacy of ancient Danish individuals.

PCA of 2,000 modern Danish genomes from the iPSYCH study62 in the context of ancient western Eurasian individuals. Coloured symbols indicate sample age for ancient Danish individuals, whereas grey symbols indicate 1,145 ancient imputed individuals from across Western Eurasia3. Modern Danish individuals are indicated by black filled circles and are shown on the right. Inset, a magnified view of the cluster with modern Danes. The colour scale in the inset represents the age range of the ancient samples within the magnified region only.

Later Neolithic and Bronze Age
Europe was transformed by large-scale migrations from the Pontic–Caspian Steppe around 5,000–4,800 cal. BP. This introduced steppe-related ancestry to most parts of the continent within a 1,000-year span and gave rise to the Corded Ware culture (CWC) complex1,2. In Denmark, this coincided with the transition from the FBC to the SGC, the regional manifestation of the CWC complex. The transition to single graves in round tumuli has been characterized archaeologically by two expansion phases: a primary and rapid occupation of central, western and northern Jutland (west Denmark) starting around 4,800 cal. BP and a later and slower expansion across the Eastern Danish Islands starting around 4,600 cal. BP53,54. In the eastern parts of the country, SGC traits are less visible, whereas FBC traditions such as burial in megalithic grave chambers persisted55. This cultural shift represents another classical archaeological enigma, with explanations favouring immigration versus cultural acculturation competing for generations19,56.

Insights from a few low-coverage genomes1,57 have indeed shown a link to the Steppe expansions, but by mapping out ancestry components in the 100 ancient genomes we now uncover the full impact of this event and demonstrate a second near-complete population turnover in Denmark within just 1,000 years. This genetic shift was evident from PCA and ADMIXTURE analyses, in which Danish individuals dating to the SGC and Late Neolithic and Bronze Age (LNBA) cluster with other European LNBA individuals and show large proportions of ancestry components associated with Yamnaya groups from the Steppe (Figs. 1 and 3 and Extended Data Fig. 1). We estimate around 60–85% of ancestry related to Steppe groups (Steppe_5000BP_4300BP), with the remainder contributed from individuals with farmer-related ancestry associated with Eastern European GAC (Poland_5000BP_4700BP; 10–23%) and to a lesser extent from local Neolithic Scandinavian farmers (Scandinavia_5600BP_4600BP; 3–18%) (Extended Data Fig. 6a,b). Although the emergence of SGC introduced a major new ancestry component in the Danish gene pool, it was not accompanied by apparent shifts in dietary isotopic ratios, or Sr isotope ratios (Fig. 3). Our complex trait predictions, however, indicate an increase in height (Fig. 3 and Supplementary Note 2), which is consistent with ancient Steppe individuals being predicted taller than average European Neolithic individuals before the steppe migrations32,49,58.

Because of poor preservation conditions in most of western Denmark, we do not have skeletons from the earliest phase of the SGC (around 4,800 cal. BP) so we cannot unequivocally demonstrate that these people carried steppe-related ancestry. SGC burial customs were implemented in different ways in the southern and the GAC-related northern parts of the peninsula, respectively18 and considering recent genetic results in other regions59, it is plausible that differing demographic processes unfolded within Denmark. However, we know that steppe ancestry was present 200 years later in SGC-associated skeletons from the Gjerrild grave57. The age of the Gjerrild skeletons (from around 4,600 cal. BP) matches the earliest example of steppe-related ancestry in our current study, identified in a skeleton from a megalithic tomb at Næs (NEO792). We estimated around 85% of Steppe-related ancestry in this individual, the highest amount among all Danish LNBA individuals (Extended Data Fig. 6a). Notably, NEO792 is also contemporaneous with the two most recent individuals in our dataset showing Anatolian farmer-related ancestry without any steppe-related ancestry (NEO580, Klokkehøj and NEO943, Stenderup Hage) testifying to a short period of ancestry co-existence before the FBC disappeared—similar to the disappearance of the Mesolithic Ertebølle people of hunter-gatherer ancestry a thousand years earlier. Using Bayesian modelling we estimate the duration between the first appearance of Anatolian farmer-related ancestry to the first appearance of Steppe-related ancestry in Denmark to be between 876 and 1,100 years (95% prob. interval, Supplementary Note 3) implying that the former type of ancestry was dominant for less than 50 generations.

The following Late Neolithic ‘Dagger’ epoch (around 4,300–3,700 cal. BP) in Denmark has been described as a time of integration of culturally and genetically distinct groups54. Bronze became dominant in the local production of weapons while elegantly surface-flaked daggers in flint were still the dominant male burial gift. Unlike the SGC epoch, this period is richly represented by human skeletal material. Although broad population genomic signatures suggest genetic stability in the LNBA (Figs. 1 and 3), patterns of pairwise IBD-sharing and Y chromosome haplogroup distributions in a temporal transect of 38 LNBA Danish and southern Swedish individuals indicate at least three distinct ancestry phases during this approximately 1,000-year time span (Extended Data Figs. 4c and 8).

LNBA phase I: an early stage between around 4,600 and 4,300 cal. BP, in which Scandinavians cluster with early CWC individuals from Eastern Europe, rich in Steppe-related ancestry and males with an R1a Y chromosomal haplotype (Extended Data Fig. 8a,b). Archaeologically, these individuals are associated with the later stages of the Danish SGC and the Swedish Battle Axe Culture.

LNBA phase II: an intermediate stage largely coinciding with the Dagger epoch (around 4,300–3,700 cal. BP), in which Danish individuals cluster with central and western European LNBA individuals dominated by males with distinct sub-lineages of R1b-L513 (Extended Data Fig. 8c,d). Among them are individuals from Borreby (NEO735, 737) and Madesø (NEO752).

LNBA phase III: a final stage from around 4,000 cal. BP onwards, in which a distinct cluster of Scandinavian individuals dominated by males with I1 Y-haplogroups appears (Extended Data Fig. 8e). Y chromosome haplogroup I1 is one of the dominant haplogroups in present-day Scandinavians, and we here document its earliest occurrence in an approximately 4,000-year-old individual from Falköping in southern Sweden (NEO220). The rapid increase in frequency of this haplogroup and associated genome-wide ancestry coincides with increase in human mobility seen in Swedish Sr isotope data, suggesting an influx of people from eastern or northeastern regions of Scandinavia, and the emergence of stone cist burials in Southern Sweden60, which were also introduced in eastern Denmark during that period54,61.

Using genomes from LNBA phase III (Scandinavia_4000BP_3000BP) in supervised ancestry modelling, we find that they form the predominant ancestry source for later Iron and Viking Age Scandinavians (Extended Data Fig. 6d) and other ancient European groups with a documented Scandinavian or Germanic association (for example, Anglo-Saxons and Goths; Extended Data Fig. 6e). When projecting 2,000 modern Danish genomes62 on a PCA of ancient Eurasians, the modern individuals occupy an intermediate space on a cline between the LNBA and Viking Age individuals (Fig. 4). This result shows that the foundation for the present-day gene pool was already in place in LNBA groups 3,000 years ago, but the genetic structure of the Danish population was continually reshaped during succeeding millenia.

Admin
Administrator

Posts: 73,016

Europe's First Farmers Mar 4, 2024 20:12:52 GMT

Quote

Post by Admin on Mar 4, 2024 20:12:52 GMT

Environmental impact
The two documented major population turnovers were accompanied by substantial changes in land use, as apparent from the high-resolution pollen diagram from Lake Højby in Northwest Zealand (Fig. 3) reconstructed using the landscape-reconstruction algorithm (LRA; Supplementary Note 6). We uncovered a direct synchronic link between shifts in a populations’ ancestry profile and land use. During the Mesolithic, the landscape was dominated by primary forest trees (Tilia, Ulmus, Quercus, Fraxinus, Alnus and so on). At the onset of the Neolithic, the primary forest diminished, cleared by FBC farmers. A new type of forest with more secondary and early successional trees (Betula and then Corylus) appeared, whereas the proportion between forest and open land remained almost unaltered. From about 5,650 cal. BP deforestation intensified, resulting in an open grassland-dominated landscape. This open phase was short-lived, and the secondary forest expanded again from around 5,500 to 5,000 cal. BP, until another episode of forest clearance occurred during the last part of the FBC epoch. We conclude that the agricultural practice during the FBC was characterized by repeated clearing of the forest followed by regrowth. After about 4,600 cal BP, this strategy changed with the emergence of the SGC and the arrival of Steppe-related ancestry in Denmark. In Western Denmark (Jutland), the arrival of the SGC was characterized by permanent large-scale opening of the landscape to create pastureland63,64 and we observe here a similar increase in grassland and cropland at Højby Sø in Eastern Denmark around 4,600 cal. BP (Fig. 3). Notably, this was accompanied by an increase in primary forest cover, especially Tilia and Ulmus, probably reflecting a development of a more permanent division of the landscape into open grazing areas and primary forests.

Drivers of change
We have demonstrated examples of both cultural and demic diffusion during the Mesolithic and Neolithic periods in Denmark. Shifts in the Mesolithic material culture appeared without any detectable levels of changes in ancestry, whereas the two cultural shifts in the Neolithic period were clearly driven by new people coming in. Accordingly, groupings of artefacts and monuments into archaeological cultures do not always represent genetically distinct populations and the underlying mechanisms responsible for prehistoric cultural shifts must be examined on a case-by-case basis.

It remains a mystery why the Neolithic farming expansion came to a 1,000-year standstill before entering Southern Scandinavia. It may be that it was complicated by a high Mesolithic hunter-gatherer population density owing to a very productive marine and coastal environment20,65. Further, the Danish Ertebølle population may have been acquainted with armed conflict11,66 enabling territorial defence against intruders. Alternatively, it has been argued that changing climatic conditions around 6,000 cal. BP became a driver since it enhanced the potential for farming further north67, but other studies have not confirmed this68. The second population turnover in the late Neolithic resulted in a short period of three competing cultural complexes in Denmark, namely the FBC, the PWC and the SGC. The latter introduced the steppe-related ancestry which has prevailed to this day. There is archaeological evidence that this was a violent time, both in Denmark69 and elsewhere70,71. Additionally, ancient DNA evidence has demonstrated that plague was widespread during this period72,73. In tandem with other indicators of population declines74, and widespread reforestation after 5,000 cal. BP75, it suggests that the local populations of Central and Northern Europe may have been severely impacted prior to the arrival of newcomers with Steppe-related ancestry. This could explain the rapid population turnover and limited admixture with locals we observe.

While the two major shifts in Danish Mesolithic and Neolithic material culture may have had different drivers and causes, the outcomes were ultimately the same: new people arrived and rapidly took over the territory. With this arrival, the local landscape was modified to fit the lifestyle and culture of the immigrants. This is the hallmark of the Anthropocene, observed here in high resolution in prehistoric Denmark.

Admin
Administrator

Posts: 73,016

Europe's First Farmers Mar 5, 2024 20:03:31 GMT

Quote

Post by Admin on Mar 5, 2024 20:03:31 GMT

Ancient genomic analyses
The 100 ancient Danish genomes analysed here contribute to the 317 shotgun-sequenced genomes in Allentoft et al.3. All details concerning sampling, DNA extraction, library preparation, sequencing, basic bioinformatics, authentication and dataset construction are found in ref. 3 together with all site descriptions and sample metadata. A condensed list of metainformation on the 100 Danish individuals is released here (Supplementary Data 1) together with a text summarizing the study sites and skeletons (Supplementary Note 1). In brief, laboratory work was carried out in dedicated ancient DNA cleanlab facilities (University of Copenhagen) using optimized ancient DNA methods1,77. Double-stranded blunt-end libraries were sequenced (80 bp and 100 bp single-end reads) on Illumina HiSeq 2500 and 4000 platforms. Initial shallow shotgun screening was used to identify samples with sufficient DNA preservation for deeper genomic sequencing. Of the 100 Danish samples that qualified for this, 65 were from tooth cementum, 29 were petrous bones, and 6 were obtained from other bones (Supplementary Data 1). Sequence reads were bioinformatically mapped to the human reference genome (build 37), filtered and merged to sample level followed by estimates of genomic overage, post-mortem DNA damage, contamination, and genetic sex ID (see3). For these 100 samples we observed C-to-T deamination fractions ranging from 12.2% to 66.7%, with an average of 34.9% across all samples (Supplementary Data 1), consistent with highly degraded ancient DNA. We genetically identified 67 males, 32 females and one undetermined in our dataset (Supplementary Data 1).

We utilized a new computational method optimized for low-coverage data21, to impute genotypes based on genotype likelihoods of ancient individuals with the samtools/bcftools pipeline, and using the 1000 Genomes phased data78 as a reference panel. To generate the main dataset in3 this was jointly applied to 1,664 shotgun-sequenced ancient genomes, including our 100 ancient Danish genomes, and resulted in a dataset of 8.5 million common SNPs (>1% minor allele frequency and imputation info score > 0.5) for imputed diploid ancient genomes. After removing genomes with low coverage (<0.1X), low imputation quality (average genotype probability <0.98), contamination estimates >5%, or close relatives (first or second degree, lowest coverage relative removed), 67 of the 100 Danish genomes were retained as imputed in downstream analyses. The remaining 33 genomes were analysed as pseudo-haploid genotypes.

For population genetic analyses, we combined ancient samples with two different modern reference panels:

‘1000 G’ dataset: whole-genome sequencing data of 2,504 individuals from 26 world-wide populations from the 1000 Genomes project, with genotypes at 7,321,965 autosomal SNPs.

‘HO’ dataset: SNP array data of 2,180 modern individuals from 213 world-wide populations, with genotypes at 535,880 autosomal SNPs.

Analyses were based on the 1000 G dataset unless otherwise noted. Individuals not passing imputation quality control cutoffs mentioned above were included in PCA and ADMIXTURE analyses as pseudo-haploid genotypes. Four Danish individuals showed possible signs of DNA contamination (Fig. 3 and Supplementary Data 1) and were excluded from most analyses. To take full advantage of the extensive multiproxy data they were, however, included in Fig. 3. Individual metadata for all genetic analyses related to the ancient Danish individuals as well as selected subset of relevant West Eurasian individuals, are reported in Supplementary Data 3.

For PCA combining ancient and modern Western Eurasians (Fig. 1b), we used the data and framework from3 to capture West Eurasian genetic diversity based on n = 983 modern genomes and n = 1,105 ancient genomes (HO dataset). Data from a total of 179 ancient Danish genomes are shown in Fig. 1b of which 83 are previously published1,47,57,76 (Supplementary Data 3)—the latter being primarily from the Bronze Age and Viking periods. To perform PCA projection for low-coverage individuals, we used smartpca with options ‘lsqproject: YES’ and ‘autoshrink: YES’.

The ADMIXTURE results presented in this study represent subsets of individuals from the full ADMIXTURE runs in3 where 1,593 ancient individuals were analysed (n = 1,492 imputed, n = 101 pseudo-haploid, n = 71 excluded as close relatives or with a contamination estimate >5%; HO dataset). Figure 1c represents 176 ancient Danish genomes after excluding three close relatives (Supplementary Data 1 and 4).

D-statistics were obtained using pseudo-haploid genotypes at transversion SNPs in the 1000 G dataset, grouping the non-Danish individuals into populations using their membership in the genetic clusters inferred from IBD sharing (Supplementary Data IV). We computed D-statistics from genotypes in PLINK format using the qpdstat function implemented in the ADMIXTOOLS 2 R package79.

Analysis of IBD sharing and mixture models were carried out as described3, using the same set of inferred genetic clusters (see Supplementary Data 4). In brief, we used IBDseq80 to detect IBD segments, a carried out genetic clustering of the individuals using hierarchical community detection on a network of pairwise IBD-sharing similarities. IBD-based PCA was carried out in R using the eigen function on a covariance matrix of pairwise IBD sharing between the respective ancient individuals. We estimated ancestry proportion in supervised modelling of target individuals as mixtures of different sets of putative source groups via non-negative least squares on relative IBD-sharing rate vectors.

Admixture time inference for FBC-associated individuals was carried out using the linkage-disequilibrium-based method DATES44 (HO dataset). We estimated admixture time separately for each target individual from Denmark and Sweden, using hunter-gatherer individuals (n = 58) and early farmer individuals (n = 49) as the two source groups.

For the PCAs presented in Fig. 4 including modern Danish samples we projected 2,000 imputed samples81 of individuals born in 1981–2005 from the iPSYCH2012 case-cohort study62 onto the PCA space spanned by the 1,145 non-low coverage or related european and western Asian ancient imputed samples3. Otherwise, the analysis is identical to the one described above. The modern individuals were selected from a subset of the random population subcohort component of iPSYCH2012 having all four grandparents born in Denmark, and being of Danish or European ancestry as determined in a separate already existing PCA of main modern-day ancestry groups81. This was done using Eigensoft 7.2.1 on the intersect of imputed SNPs from the ancient and modern samples, filtered by minor allele frequency 0.05, pruned using PLINK v1.90b6.2182 based on source samples (parameters: –indep-pairwise 1000 50 0.25) leaving 146,895 variants.

The genetic predictions of eye and hair colour were done based on the HIrisPlex system83. We used imputed effect allele dosages of 18 out of 24 main effect HIrisPlex variants, available for the ancient samples, to derive probabilities for brown, blue and grey/intermediate eye colour and blond, brown, black and red hair colour, following HIrisPlex formulas (see further details in Supplementary Note 2). We predicted relative ‘genetic height’ using allelic effect estimates from 310 common autosomal SNPs with robustly genome-wide significant allelic effects (P < 10−15) in a recent GWAS of height in the UK Biobank84. Per-sample height polygenic score (PGS) was calculated for ancient individuals as well as 3,467 Danish ancestry male conscripts from the random population subcohort of the iPSYCH2012 case-cohort study62 by summing allelic effect multiplied with the effect allele imputed dosage81 across the 310 loci. For further details see Supplementary Note 2. Only a fraction of the 100 Danish skeletons were suitable for stature estimation by actual measurement, which is why these values are not reported here.