|
Post by Admin on Jan 18, 2022 20:25:18 GMT
Population Replacement in Early Neolithic Britain Selina Brace, Yoan Diekmann, Thomas J. Booth, Zuzana Faltyskova, Nadin Rohland, Swapan Mallick, Matthew Ferry, Megan Michel, Jonas Oppenheimer, Nasreen Broomandkhoshbacht, Kristin Stewardson, Susan Walsh, Manfred Kayser, Rick Schulting, Oliver E. Craig, Alison Sheridan, Mike Parker Pearson, Chris Stringer, David Reich, Mark G. Thomas, Ian Barnes doi: doi.org/10.1101/267443Abstract The roles of migration, admixture and acculturation in the European transition to farming have been debated for over 100 years. Genome-wide ancient DNA studies indicate predominantly Anatolian ancestry for continental Neolithic farmers, but also variable admixture with local Mesolithic hunter-gatherers1–⇓⇓⇓⇓⇓⇓⇓9. Neolithic cultures first appear in Britain c. 6000 years ago (kBP), a millennium after they appear in adjacent areas of northwestern continental Europe. However, the pattern and process of the British Neolithic transition remains unclear10–⇓⇓⇓⇓15. We assembled genome-wide data from six Mesolithic and 67 Neolithic individuals found in Britain, dating from 10.5-4.5 kBP, a dataset that includes 22 newly reported individuals and the first genomic data from British Mesolithic hunter-gatherers. Our analyses reveals persistent genetic affinities between Mesolithic British and Western European hunter-gatherers over a period spanning Britain’s separation from continental Europe. We find overwhelming support for agriculture being introduced by incoming continental farmers, with small and geographically structured levels of additional hunter-gatherer introgression. We find genetic affinity between British and Iberian Neolithic populations indicating that British Neolithic people derived much of their ancestry from Anatolian farmers who originally followed the Mediterranean route of dispersal and likely entered Britain from northwestern mainland Europe. The transition to farming marks one of the most important shifts in human evolution, impacting on subsistence, social organisation, health and disease vulnerabilities, economy, and material culture. The processes by which this transition occurred have been a matter of intense debate for over a century10–⇓⇓⇓⇓15, although across continental Europe recent ancient DNA studies indicate a predominant role for expanding Neolithic farmer populations of mostly Anatolian ancestry (Anatolian farmers - ANF)1–⇓⇓⇓⇓⇓⇓⇓9. Anatolian farmer-derived populations dispersed throughout Europe via two major routes - one along the Mediterranean and the other through Central and into Northern Europe3–⇓⇓⇓7. Both dispersals involved repeated, but mostly subsequent introgressions with local Mesolithic foragers, producing distinct cultural and genetic trajectories. The nature of the Neolithic transition in Britain remains a puzzle because of the millennium-long delay in its appearance after the establishment of farming in adjacent regions of continental northwestern Europe10–⇓⇓⇓⇓15, and the lack of genome-wide data from British Mesolithic hunter-gatherers. Whilst there is universal agreement amongst archaeologists that there was a dramatic change in material culture in Britain after 6 kBP, there are divergent views regarding the extent to which this change was influenced by cultural or demographic processes1–⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓15. One interpretation of the archaeological evidence is that British Mesolithic hunter-gatherers adopted Neolithic cultural practices abruptly at c.6 kBP without substantial immigration following prolonged contact with their continental neighbours15. This view is inconsistent with the predominantly Anatolian farmer-related ancestry in published data from British Early Neolithic farmers8, but the extent to which local British hunter-gatherer populations contributed to the first British farming populations, as well as the relationship of British hunter-gatherers to continental hunter-gatherer populations remains unresolved. These questions are of interest as Britain is situated between two genetically-distinct contemporaneous groups of Mesolithic hunter-gatherers – Western European and Scandinavian (WHGs & SHGs)16. The British Isles could also have putatively harboured ancestry from hunter-gatherers related to earlier Magdalenian Palaeolithic groups that recolonised Europe from the southwest after the Last Glacial Maximum (~21 to 17 kBP)17. Here, we report the first genome data from six Mesolithic (including the ‘Cheddar Man’ skeleton from Gough’s Cave, Cheddar Gorge, Somerset) and 16 Neolithic British individuals, and combine it with new and already reported data from 51 previously published Neolithic British individuals8 to characterise the Mesolithic and Neolithic populations of Britain (Figure 1). We combined data generated in two different ways. For 35 individuals we generated new whole genome shotgun sequencing data (median coverage 0.09x) including the first full genomes from the British Mesolithic (at 2.3x) and Neolithic (at 10.7x). For all individuals we enriched next generation sequencing libraries for sequences overlapping about 1.24 million single nucleotide polymorphisms (SNPs) (median coverage 0.88x). We merged data obtained from both methods when it was available and chose the most likely base to represent the allele at each SNP (see Material and Methods). We merged the British Mesolithic and Neolithic data with 67 previously reported ancient DNA samples1–⇓3, 5–⇓⇓8, 10, 17–⇓⇓⇓⇓⇓23 (see Supplementary Table S1) and finally also with sequencing data from present-day individuals from diverse populations around the world47.
|
|
|
Post by Admin on Jan 18, 2022 22:17:52 GMT
Figure 1: Map of sample locations. Geographical locations of British samples analysed here. Numbers indicate the number of samples from a given location. We used principal components analysis to visualise some of the affinities of British Mesolithic and Neolithic genomes alongside those from ancient and modern West-Eurasian populations (Figure 2). The British Mesolithic individuals all cluster with Western and Scandinavian hunter-gatherers. By contrast, all directly-dated individuals who post-date 6 kBP, and undated individuals associated with Neolithic monuments, cluster tightly near Iberian and Central European Middle Neolithic individuals. Figure 2: PCA of modern and ancient West-Eurasians. British and additional ancient samples are projected onto the reference space computed on modern West-Eurasian populations. See Material and Methods for computational details and Supplementary Table 1 for information on the samples. Abbreviations: European (Eur.), Pleistocene (Plei.), hunter-gatherer (H.-G.), British Isle (Brit.-I.), Middle Neolithic (M.-Neo.) By examining the degree of allele sharing of British Mesolithic individuals to various European hunter-gatherer individuals or groups (SHG, EHG and El Miron, see Supplementary Figures S1-S4), we were able to attribute them confidently to the WHG. Comparison of British Mesolithic samples to different Mesolithic WHGs (Loschbour - Luxembourg, La Brana - Spain, KO1 - Hungary; Supplementary Figures S5-S6) indicates that they all resemble Loschbour most closely (i.e. the geographically most proximate Mesolithic genome available). When we compared the remaining British Mesolithic genomes to Loschbour and Cheddar Man (our highest-coverage British Mesolithic sample; ~2.3x), we found no significant excess of shared drift for either individual, indicating that Loschbour and the British Mesolithic samples do not form separate clusters (Supplementary Figure S7). To investigate the proportions of Anatolian farmer-related ancestry in the British samples we modeled them as mixtures of ANFs and European WHGs (Figure 3, Supplementary Figure S8). All British Mesolithic samples could be explained entirely by WHG ancestry within the error bounds of the test. The majority (~75%) of ancestry in all British Neolithic individuals could be attributed to ANFs, indicating a substantial demographic shift with the transition to farming. These proportions of British Neolithic ANF/WHG ancestry are similar to Early Neolithic Iberian and Middle Neolithic Central European samples. We inferred some geographic structure in WHG admixture proportions among the British Early Neolithic samples; individuals from Wales retain the lowest levels of WHG admixture, followed by those from South-West and Central England. South East England and Scotland show the highest WHG admixture proportions. These proportions remain stable for over a millennium, from the Early into the Middle/Late Neolithic.
|
|
|
Post by Admin on Jan 18, 2022 23:32:06 GMT
Figure 3: WHG and ANF ancestry components of British and Central European Neolithic populations. The relative WHG and ANF ancestry in Early and Middle Neolithic British and continental European populations quantified by f4 admixture proportions. Percentages indicate error estimates. See Material and Methods for computational details and Supplementary Table 1 for the lists of samples grouped into the different Neolithic populations. Abbreviations: Neolithic (Neo.), Early-, Middle-, and Late Neolithic (EN, MN, LN), South East (SouthE), South West (SouthW) To investigate the proximate source of ANF ancestry in British Neolithic samples, we examined affinities with available Early Neolithic individuals from Iberia and Central Europe. We chose to compare Early over Middle Neolithic Iberians as the latter are contemporary with the British Early Neolithic, making them an unlikely direct source. For all of our British Neolithic individuals we inferred more shared drift with Early Neolithic Iberians; for the majority of comparisons this was significant (Figure 4A, Supplementary Figure S9). To infer levels of WHG introgression occurring between Iberian Early Neolithic populations – the closest currently available attributable source of farmer ancestry in British Early Neolithic genomes – and early British farmers, we estimated f4 admixture proportions. We detected little excess (~10%) WHG ancestry beyond what was already present in Iberian Early Neolithic populations, suggesting small proportions of admixture, particularly in Wales where we detected no excess WHG ancestry (Figure 4B). The small amounts of WHG introgression inferred here could have occurred on mainland Europe, and there is no need to invoke any genetic contribution from British Mesolithic hunter-gatherers to explain these results, although the significant bias towards British WHGs in some British Neolithic farmers, suggests that at least some of this introgression probably did occur in Britain (Supplementary Figure S7): Figure 4: (A) Affinities of British and continental Neolithic populations. We compute admixture f4 statistics for different British EN, MN, and LN and continental MN populations comparing shared drift with Central European EN and Iberian EN populations. A positive Z-score above 2 corresponds to a significant affinity to the Iberian EN over Central European EN population. (B) Quantifying excess WHG ancestry in British EN compared to Iberian EN and MN populations. We compute f4 admixture proportions of WHG and different ANF populations in EN samples from Wales, England, and Scotland.
|
|
|
Post by Admin on Jan 19, 2022 3:27:19 GMT
The Mesolithic Cheddar Man and the Neolithic sample from Carsington Pasture Cave, Derbyshire (‘Sven’) had sufficient coverage to estimate heterozygosity. Consistent with recent ancestry from larger or more admixed populations, Sven showed slightly higher levels of heterozygosity than Cheddar Man (Supplementary Figure S10). None of the Mesolithic and Neolithic British individuals analysed here had a derived lactase persistence allele (see Supplementary Table S2). We predicted pigmentation characteristics for Cheddar Man and Sven using Hirisplex25 and a recently-developed method for predicting skin pigmentation26. Previously, predictions on the level of skin pigmentation were mostly derived using two SNPs in SLC45A2 and SLC24A5 that indicate lack of hypo-pigmentation when in the ancestral state18. However, here we integrate 36 rather than 2 SNPs allowing more precise prediction26. Cheddar Man is predicted to have had dark or dark to black skin, blue/green eyes and dark brown possibly black hair, whereas Sven most likely had intermediate to dark skin pigmentation, brown eyes and black possibly dark brown hair (see Pigmentation section in the Supplementary Materials for a detailed discussion of the results). This is in line with the current hypothesis that alleles commonly associated with lighter skin were introduced in Western Europe by ANFs19. We also analysed two previously-published WHGs, and find potential temporal and/or geographical variation in pigmentation characteristics. Loschbour22 from Luxembourg is ~2000 years younger than Cheddar Man, and is predicted to have had intermediate skin pigmentation. Furthermore, the Loschbour individual most likely had blue/green eyes. In contrast, La Braña18 from northern Spain who is slightly later than Loschbour is predicted to have had dark to dark to black skin and hazel/green eye colour. Both La Braña and Loschbour were predicted to have had black, possibly dark brown hair. These results imply that quite different skin pigmentation levels coexisted in WHGs at least by around 6000 BC. Discussion The genomes of the six British Mesolithic individuals examined here appear to be typical of WHGs, indicating that this population spread to the furthest northwestern point of early Holocene Europe after moving from southeastern Europe, or further east, from approximately 14,000 years ago17. It is notable that this genetic similarity among British Mesolithic and Loschbour individuals spans a period in Britain (c.10.5-6 kBP) that includes the cultural transition to the Late Mesolithic, the potentially catastrophic Storegga landslide tsunami, and the separation of Britain from continental Europe as sea levels rose after the last ice age. This finding is inconsistent with the hypothesis of pre-c.6 kBP gene flow into Britain from neighbouring farmers in continental Europe15. Our analyses indicate that the appearance of Neolithic practices and domesticates in Britain c.6 kBP was mediated overwhelmingly by immigration of farmers from continental Europe2,13–14, and strongly reject the hypothesised adoption of farming by indigenous hunter-gatherers as the main process15. British farmers were substantially descended from Iberian Neolithic-related populations whose ancestors had expanded along a Mediterranean route2,7,14, although our analyses cannot rule out the possibility that they also inherited a minority portion of their ancestry from the Danubian route expansion through Central Europe. Indeed, a recent study that investigated continental Neolithic farmer-related ancestry components in Neolithic Britain estimated ⅔ Mediterranean and ⅓ Danubian route8, which may be consistent with the association between Britain’s more easterly-distributed Carinated Bowl tradition14 and the Nord-Pas-de-Calais region of France, as Neolithic people in these regions of mainland Europe are thought to have interacted with populations of Central European Neolithic ancestry14, 27–⇓29. The varied Neolithic cultures of northern France, Belgium and the Netherlands, the most likely continental sources for the British Neolithic14, may represent a genetically heterogeneous population who shared large but variable proportions of their ancestry with Neolithic groups in Iberia via Atlantic and southern France. We caution that our results should not be interpreted as showing the Iberian Neolithic-related ancestry in British Neolithic people derives from migrants whose ancestors lived in Iberia, as we do not have ancient DNA data from yet unidentified source populations — possibly located in southern France — that were ancestral to both Iberian and British farmers. Available Middle Neolithic Iberian individuals are too late to represent the source population for early British farmers, and there is no archaeological evidence for direct immigration from Iberia14. The lack of genome-wide data from Neolithic northern France, Belgium and the Netherlands means that it is not currently possible to identify proximal continental source populations. The limited regional variation in WHG ancestry we see in the British Neolithic samples could reflect subtle but differing degrees of regional admixture between farmers and foragers, and/or multiple continental source populations carrying varying levels of WHG ancestry colonising different regions of Britain. What is clear is that across Britain all of our estimates for admixture between hunter-gatherers and farmers are very small, and that we find no evidence of WHG ancestry increasing as the British Neolithic progressed over time (Figure 3). In contrast, the resurgence of WHG ancestry in all available continental European Middle Neolithic samples, prior to the British Neolithic transition and including a population sample from southern France, means that the level of WHG ancestry we see in most British Neolithic farmers could be accounted for entirely in continental source populations3, 8–⇓10. Evidence for only low levels of WHG introgression among British Neolithic people is striking given the extensive and complex admixture processes inferred for continental Neolithic populations3’ 8–⇓10, 30–⇓32. Low levels of admixture between these two groups on the wavefront of farming advance in continental Europe have been attributed to the groups maintaining cultural and genetic boundaries for several centuries after initial contact30–⇓32. Similarly, isotopic and genetic data from the west coast of Scotland are consistent with the coexistence of genetically distinct hunter-fisher-gatherers and farmers, albeit for a maximum of few centuries 33–34. The resurgence in WHG ancestry after the initial phases of the Neolithic transition in continental Europe indicates that the two populations eventually mixed more extensively3, 8–⇓10. However there is no evidence for a WHG resurgence in the British Neolithic up to the Chalcolithic population movements associated with the Beaker phenomenon (c.4.5 kBP)8. This is consistent with the lack of evidence for Mesolithic cultural artefacts in Britain much beyond 6 kBP14 and with a major dietary shift at this time from marine to terrestrial resources; particularly apparent along the British Atlantic coast33–⇓35. In summary, our results indicate that the progression of the Neolithic in Britain was unusual when compared to other previously studied European regions. Rather than reflecting the slow admixture processes that occurred between ANFs and local hunter-gatherer groups in areas of continental Europe, we infer a British Neolithic proceeding with little introgression from resident foragers – either during initial colonization phase, or throughout the Neolithic. This may reflect the fact that farming arrived in Britain a couple of thousand years later than it did in Europe. The farming population who arrived in Britain may have mastered more of the technologies needed to thrive in northern and western Europe than the farmers who had first expanded into these areas. A large-scale seaborne movement of established Neolithic groups leading to the rapid establishment of the first agrarian and pastoral economies across Britain, provides a plausible scenario for the scale of genetic and cultural change in Britain. Supplementary Material www.biorxiv.org/content/10.1101/267443v1.supplementary-material
|
|
|
Post by Admin on Jan 19, 2022 19:03:48 GMT
Population Replacement in Early Neolithic Britain— SUPPLEMENTARY MATERIAL
Pigmentation Susan Walsh, Manfred Kayser
See Supplementary Table S2 for sequencing data and allelic states that entered the analyses below. Predictions are based on methods and tools published in Walsh et al., Forensic Sci Int Genet, 2013, and Walsh et al., Hum Genet, 2017.
La Braña (Spain, Mesolithic)
Eye colour —All loci are present and have good coverage.
Blue eye 0.459 Int. eye 0.155 Brown eye 0.387
Final prediction: Intermediate (hazel/green) eye colour
Explanation: All probabilities are less than 0.5 so it is a combination of all categories, as brown is relatively high, it is more than likely a light hazel eye colour individual, but perceived green (blue/yellow) cannot be ruled out.
Hair colour—There is 1 locus (TYRP1 rs683) with low coverage (1x), hence a heterozygote is possible. Prediction is a range that includes what the 1x coverage found (derived G allele) and the possibility of an A ancestral allele being present. TYRP1 rs683 TYRP1 rs683 (homozygote GG) (heterozygote GA) Blond 0.018 0.014 Brown 0.612 0.595 Red 0 0 Black 0.37 0.391 Light 0.033 0.025 Dark 0.967 0.975
Prediction range: Brown 0.612 – 0.595 Black 0.37 – 0.391
Final Prediction: Black/Dark Brown hair colour
Explanation: The probability value of black is >0.25 so it has a significant impact on prediction, and will darken the high brown probability. This individual would be perceived to have black hair. Dark Brown however cannot be ruled out.
Skin pigmentation—Only 1 locus (BNC2 rs10756819) is missing, however the profile does contain 3 loci with low coverage (n=1x), hence a heterozygote is possible. When factoring in possible genotype combinations, a prediction range has been generated. The range consists of assuming the 3 loci with low coverage are correct as homozyogote for their sequenced allele (ASIP rs1667394 A allele (derived), OCA2 rs1545397 A allele (ancestral), TYRP1 rs683 A allele (ancestral)) and omitting BNC rs10756819 in the prediction model as it has no coverage, to including this marker with a homozygote ancestral G allele and also derived A allele. The following range for skin colour prediction is possible for this individual with these parameters:
Omitting rs10756819 G ancestral allele A derived allele Very Pale 0 0 0 Pale 0 0 0 Intermediate 0.174 0.042 0.205 Dark 0.463 0.209 0.435 Dark-Black 0.363 0.749 0.360
Prediction range: Very Pale 0 Pale 0 Intermediate 0.042 - 0.205 Dark 0.209 - 0.435 Dark-Black 0.749 - 0.36
Final prediction: Dark/Dark-to-Black skin
Explanation: The combined effect of probabilities in the dark and dark-to-black colour categories provide an indication that the individual has darkly pigmented skin, it is unlikely that this individual has the darkest possible skin pigmentation, however, it cannot be ruled out as the missing marker does influence that detail, but certainly skin colour is dark in complexion.
Cheddar Man (UK, Mesolithic)
Eye colour —There is 1 locus (LOC105374875 (formally known as SLC24A4) rs12896399) with low coverage (1x) hence a heterozygote is possible. Prediction includes a range that includes what the 1x coverage found (ancestral G allele) and the possibility of an A derived allele being present.
LOC105374875 rs12896399 LOC105374875 rs12896399 (homozygote GG) (heterozygote GA) Blue eye 0.564 0.711 Int. eye 0.189 0.143 Brown eye 0.247 0.145
Prediction range: Blue eye 0.564 - 0.711 Int. eye 0.189 - 0.143 Brown eye 0.247 - 0.145
Final prediction: Intermediate (blue/green) eye colour
Explanation: This individual has light or blue/green eye colour, it is not light blue, there are elements of brown/yellow in the eye to give a proposed perceived green colour. Better coverage at the low sequenced marker would clarify this but blue/hazel cannot be ruled out. It is certainly not a brown eyed or clear blue-eyed individual.
Hair colour—There is 1 locus PIGU rs2378249 with low coverage (1x) hence a heterozygote is possible. Prediction is a range that includes what the 1x coverage found, ancestral A allele, but also includes the possibility of a heterozygote being present. PIGU rs2378249 PIGU rs2378249 (homozygote AA ) (heterozygote CA) Blond 0.014 0.013 Brown 0.719 0.759 Red 0.011 0.023 Black 0.257 0.205 Light 1 0.999 Dark 0 0.001
Prediction range: Brown 0.719 – 0.759 Black 0.257 – 0.205
Final Prediction: Dark Brown/Black hair colour
Explanation: The probability value of black is >0.2 so it has an impact on prediction, and will darken the high brown probability. However there are light pigment alleles indicating a lighter shade phenotype. Better coverage at the low sequenced marker would help clarify this. This individual would be perceived as having dark brown hair. Black however cannot be ruled out.
Skin pigmentation—There are 3 loci (BNC2 rs10756819, TYR rs1126809, MC1R rs3212355) missing, and the profile does contain 2 loci (LOC105374875 rs12896399 and PIGU rs2378249) with low coverage (n=1x) hence a heterozygote is possible at those sites. When factoring in possible genotype combinations, a prediction range may be generated. The range consists of assuming the two loci with low coverage are correct as homozygote for their sequenced allele (LOC105374875 rs12896399 G allele and PIGU rs2378249 A allele) and omitting the 3 missing loci from the prediction model as they have no coverage, to including these markers with their ancestral (BNC2 rs10756819-GG, TYR rs1126809-GG, MC1R rs3212355-CC) and also their derived allele counterparts. The following range for skin colour prediction is possible for this individual with these parameters:
ancestral alleles used derived alleles used Very Pale 0 0 Pale 0 0 Intermediate 0.394 0.125 Dark 0 0 Dark-Black 0.606 0.875
Prediction range: Very Pale 0 Pale 0 Intermediate 0.394 - 0.125 Dark 0 - 0 Dark-Black 0.606 - 0.875
If we omit the three missing alleles, our tool produces 0.891 and 0.109 probabilities for the intermediate and dark-black category respectively, changing the prediction ranges to 0.891-0.125 and 0.109-0.875. However, note that this completely removes the locus from the prediction model, hence the prediction will not perform optimally (how the prediction model was made), therefore it is best to have some allele present to infer the most probable range for Cheddar Man and we therefore derive the ranges above from the extreme allele constellations only.
Final prediction: Dark/Dark-to-black skin
Explanation: The missing loci certainly impact on this prediction; however utilizing the input of all ancestral alleles is the preferred option over the use of the derived alleles at these loci – hence 0.394 for intermediate and 0.606 for Dark-black would be the most probable profile. That being said a broad range is present in both the intermediate and dark-black categories due to the missing loci. Also this effect, of skipping a skin colour prediction category with regards probability values, tends to be observed more often in admixed individuals. What is important to note is the input of the dark-black prediction is significant on the intermediate category and therefore it is acceptable to propose a dark complexion individual over an intermediate/light prediction even though the intermediate range is large. It is unlikely that this individual has the darkest possible pigmentation, however it cannot be ruled out. Better sequencing coverage would clarify to what degree this individual has a dark complexion.
|
|