NEW YORK – A genetic map of the Scottish populations broadly reflects the boundaries of Dark Age kingdoms there, a new analysis has found.
Previous population genetic studies of the British Isles have focused largely on southern Britain and Ireland, finding, for instance, the influence of German-related ancestry within southern England stemming from the Anglo-Saxon invasions that occurred between 400 and 650 AD and of English and Scottish admixture into northeastern Ireland dating to the Ulster Plantations during the 17th century.
Researchers led by the University of Edinburgh's James Wilson have extended such population genetic analyses to encompass additional individuals from Scotland and nearby isles including the Hebrides, the Shetland Islands, and the Isle of Man. As they reported this week in the Proceedings of the National Academy of Sciences, the researchers uncovered a prevalent geographic population structure that broadly aligns with the borders of the ancient kingdoms of Scotland.
"It is remarkable how long the shadows of Scotland's Dark Age kingdoms are, given the massive increase in movement from the industrial revolution to the modern era," Wilson, the senior author of the paper, said in a statement. "We believe this is largely due to the majority of people marrying locally and preserving their genetic identity."
The researchers pulled together genotyping data on 2,544 individuals with English, Welsh, Scottish, Manx, or Irish regional ancestry from five different cohorts. After quality control, they analyzed their samples at 341,924 common markers.
Using a suite of population genetics analysis tools — including the haplotype-based fineStructure clustering tool, principal components analysis, dimension reducing t-distributed stochastic neighbor embedding (t-SNE), and more — the researchers began to tease out the population substructure of Scotland.
In particular, they found that their samples largely clustered by geography into six groups: the northeast, the southwest, the Borders, the Hebrides, Orkney, and Shetland. The northeast cluster appears to reflect, for instance, the boundaries of an ancient Pictish kingdom, while the Borders cluster resides within the Brythonic kingdoms of Gododdin and Rheged.
The northern islands, meanwhile, exhibited effects of isolation, the researchers reported. Individuals from Hebrides, for example, clustered independently of mainland Scottish populations in principal components analysis and formed their own island within t-SNE dimensional analysis. The population additionally exhibited long runs of homozygosity and low gene flow rates, all reflecting a small and isolated ancestral population.
Meanwhile, by folding in genotyping data from 2,225 Scandinavian individuals, the researchers teased out varying degrees of Norwegian-related ancestry — representing possible Norse Viking ancestry — among these British Isles populations. They estimated the level of Norwegian-like ancestry of Orkney and Shetland island populations to be about 20 to 25 percent, but about 7 percent among Irish populations, lower than previous estimates.
The researchers additionally noted genetic affinities between ancient Gaelic Icelanders and populations from northwestern Britain and Ireland, suggesting that these areas — which were sites of Viking activity — might have served as a source population for Gaelic Icelanders.
Shedding light on local genetic population structure such as this, the researchers added, could also help in better appreciating rare variants present among these populations with further implications for understanding complex disease. "This work is important not only from the historical perspective, but also for helping understand the role of genetic variation in human disease," first author Edmund Gilbert from the Royal College of Surgeons in Ireland added. "Understanding the fine-scale genetic structure of a population helps researchers better separate disease-causing genetic variation from that which occurs naturally in the British and Irish populations, but has little or no impact on disease risk."