Thursday, March 19, 2015

The Population Genetics Of The British Isles

A major new study of the whole genomes of the people of the British Isles has been published in the journal Nature.

A few of the "forest" level conclusions:

* The sample size is 2,039 people in the British Isles compared to 6,209 European individuals.  This is pretty much as big a sample as you get in historical population genetics.  The significance of these sample size is magnified by the fact that these are whole genomes, and not just Y-DNA or mtDNA haplogroup data.  Even very small samples of whole genomes can be highly informative, and these samples aren't small.

* All of the peoples of the British Isles are very homogeneous genetically, and the people of Central and Southern Britain (as well as Cornwall), are extremely homogeneous genetically.

This is particularly notable, given that Britain has more than four cultural/linguistic units that are recognized politically as different enough to require autonomy to some extent (Britain, Scotland, Wales, and Northern Ireland, plus some minor dependencies on nearby islands).

There is far more dialect variation in the British Isles than there is in the United States to the extent that one could make a very well informed estimate about someone's origins that is even finer grained than the genetic clusters within the British Isles identified by this study based upon that person's dialect and accent in the English language.

The British Isles are perhaps one of the clearest examples of strong cultural and linguistic substructure in a population that is genetically very homogeneous.  In most areas of the world, the kind of cultural and linguistic substructure seen in the British Isles corresponds to fairly dramatic population genetic differences between the various groups.  In the British Isles, in contrast, populations that are genetically almost identical and not all that far from each other geographically have very distinct cultural identities.

* The fine grained differences between British subpopulations correspond to influences from different parts of the Continental Europe and Scandinavia that largely correspond to population genetic events in the historic era and archaeologically well documented late prehistoric eras, more or less as one would expect with only minor surprises (like Cornwall that one might have expected to be more like  the Welsh people).

For example, as expected from our historical knowledge, the population genetics of Northern Ireland overlap heavily with the population genetics of the area around the western Scottish-English borderlands.

* The Danish Vikings who imposed Danelaw on Britain in the 1st millennium, while they had significant cultural and linguistic impact, had almost no genetic impact outside of the Orkney Islands.  This is analogous to the situation in Hungary, where the people who are the source of Hungary's current language have left almost no genetic trace in the country.

* The Anglo-Saxon contribution to the British gene pool in the 1st millennium is about 10% to 40% of the total (the spread is disappointingly large for such an impressive data set).  This is a significant minority contribution, but not population replacement.  This is reflected mostly in the Central-Southern British cluster that out of 17 clusters makes up about half of the total sample.  Many of the other clusters are in areas that have some level of political autonomy.  Wales, for example, appears to have five different small regional genetic clusters.

* The pre-Anglo-Saxon Celtic substrate in Britain was not uniform; it varied by region.

* There was substantial migration to Britain from Continental Europe in the Neolithic and later eras prior to the arrival of the Romans (and therefore also prior to the arrival of the Danes, the Anglo-Saxons, the Normans, and recent immigration from the 19th century onwards).

* I suspect, but don't know for a fact, that the sample was limited to people who are "ancestrally" British and hence excludes individuals with known recent immigrant ancestry.  Thus, Britain today is probably much more genetically diverse than this sample which was probably picked to be as informative as possible about the ancient and prehistoric genetic history of Britain would indicate.  For example, Britain has a significant South Asian minority population that is largest in greater London, but is found throughout the British Isles, that is not reflected in this data.  These huge genetic chasms, however, are not reflected in Britian's system of regional autonomy and instead is blended into preexisting communities across the British Isles.

No comments: