Population genomics of postglacial western eurasia

Abstract

The transitions from foraging to farming and later to pastoralism in Stone Age Eurasia (c. 11-3 thousand years before present, BP) represent some of the most dramatic lifestyle changes in human evolution. We sequenced 317 genomes of primarily Mesolithic and Neolithic individuals from across Eurasia combined with radiocarbon dates, stable isotope data, and pollen records. Genome imputation and co-analysis with previously published shotgun sequencing data resulted in >1600 complete ancient genome sequences offering fine-grained resolution into the Stone Age populations. We observe that: 1) Hunter-gatherer groups were more genetically diverse than previously known, and deeply divergent between western and eastern Eurasia. 2) We identify hitherto genetically undescribed hunter-gatherers from the Middle Don region that contributed ancestry to the later Yamnaya steppe pastoralists; 3) The genetic impact of the Neolithic transition was highly distinct, east and west of a boundary zone extending from the Black Sea to the Baltic. Large-scale shifts in genetic ancestry occurred to the west of this "Great Divide", including an almost complete replacement of hunter-gatherers in Denmark, while no substantial ancestry shifts took place during the same period to the east. This difference is also reflected in genetic relatedness within the populations, decreasing substantially in the west but not in the east where it remained high until c. 4,000 BP; 4) The second major genetic transformation around 5,000 BP happened at a much faster pace with Steppe-related ancestry reaching most parts of Europe within 1,000-years. Local Neolithic farmers admixed with incoming pastoralists in eastern, western, and southern Europe whereas Scandinavia experienced another near-complete population replacement. Similar dramatic turnover-patterns are evident in western Siberia; 5) Extensive regional differences in the ancestry components involved in these early events remain visible to this day, even within countries. Neolithic farmer ancestry is highest in southern and eastern England while Steppe-related ancestry is highest in the Celtic populations of Scotland, Wales, and Cornwall (this research has been conducted using the UK Biobank resource); 6) Shifts in diet, lifestyle and environment introduced new selection pressures involving at least 21 genomic regions. Most such variants were not universally selected across populations but were only advantageous in particular ancestral backgrounds. Contrary to previous claims, we find that selection on the FADS regions, associated with fatty acid metabolism, began before the Neolithisation of Europe. Similarly, the lactase persistence allele started increasing in frequency before the expansion of Steppe-related groups into Europe and has continued to increase up to the present. Along the genetic cline separating Mesolithic hunter-gatherers from Neolithic farmers, we find significant correlations with trait associations related to skin disorders, diet and lifestyle and mental health status, suggesting marked phenotypic differences between these groups with very different lifestyles. This work provides new insights into major transformations in recent human evolution, elucidating the complex interplay between selection and admixture that shaped patterns of genetic variation in modern populations.

Imputation of ancient genomes

“…Finally, we assessed imputation performance and compared the downstream analyses results obtained with high-coverage and imputed genotypes. Three out of the 43 ancient genomes in this study constitute a trio (mother, father and son) that was recently re-sequenced and is not yet fully public 23,30 , in contrast to the remaining 40 genomes. This dataset of 43 ancient genomes is a diverse dataset in regard to their sequencing/study, as well as epoch and continent the ancient individuals lived in, with about half of the individuals being from Europe and the other half from Africa, America and Asia (Figure 1B).…”

“…The availability of an ancient trio (mother, father, son) allowed us to use an orthogonal approach based on Mendel's rules of inheritance to measure imputation and phasing quality. This trio was sampled in a Late Neolithic mass burial at Koszyce 23,30 and was re-sequenced in the context of the 9 study of Allentoft et al 23 resulting in genome coverages of 27.5x (mother, RISE1159), 18.9x (father, RISE1168), 5.4x (son, RISE1160). In this analysis, imputation errors corresponded to sites where parental and offspring genotypes disagreed with Mendel transmission rules.…”

“…CC-BY-NC-ND 4.0 International license made available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It 23 ) to impute low-coverage ancient genomes, using present-day haplotypes, assuming matching ancestry. However, aDNA introduces extra challenges, including damage and potential contamination 24 , and it is not clear whether ancient individuals' ancestries are well captured by reference panels of presentday individuals.…”

“…Present-day genotypes have been imputed with increasing accuracy due to improved imputation methods on the one hand, and increased reference panel size and diversity on the other hand, such as the Haplotype Reference Consortium (HRC) 15 , the 1000 Genomes Project 16 and TOPMed 17 . These advances have also been exploited by some (e.g., Martiniano et 23 ) to impute low-coverage ancient genomes, using present-day haplotypes, assuming matching ancestry. However, aDNA introduces extra challenges, including damage and potential contamination 24 , and it is not clear whether ancient individuals' ancestries are well captured by reference panels of presentday individuals.…”

Timesweeper: Accurately Identifying Selective Sweeps Using Population Genomic Time Series

“…Genomic time-series data has been used to great effect to find positively selected loci in experimentally controlled studies (reviewed in [45]). There is growing interest in performing similar analyses in natural populations, as the acquisition of time-series genomic data has become feasible both in species with short generation times such as Drosophila [35,36,68,69], and in those for which ancient DNA is available, such as humans [70].…”