Indo-European genetic signatures in an Orcadian and a Lithuanian

In the Bronze Age Indo-European invasion of Europe, I argued that:

A West_Asian genetic component is missing from ancient DNA sampled from Europe as recently as ~5,000 years ago.
This component exists at appreciable frequencies in modern Europeans, reaching minima in non-Indo-European Basques and Finns.
It is likely that the West_Asian component manifested itself in Europe post-5kya, during the Copper and Bronze Ages, and was associated with the bearers of Indo-European languages.

Of course, the West_Asian component is an abstraction created from a study of modern populations. To better understand the identity of this component, I undertook a simple experiment using the (not yet released) K5c and K8c calculators of the Dodecad Project. The are equivalent to the weac2 and K7b calculators released so far by the Project, but make use of the new Pagani et al. (2012) data on East Africa, hence the appearance of an "East_African" component at K=7 pushing back the appearance of the West_Asian component to K=8.

K5c captures the "West Eurasian cline" between the Near East and Europe, and includes the following components:


We can say that the Atlantic_Baltic corresponds to northern Caucasoids, while the Near_East one to southern Caucasoids.

K8c includes the following components:


This differs from the previous one in the appearance of the Siberian and East_African components, as well as the aforementioned West_Asian one within West Eurasia.

Naturally the question arises: what can we say about the origin of West_Asian ancestry in modern Europeans? How is it related to other populations? How is it related to the northern and southern Caucasoids?

To answer these questions, I isolated an Orcadian individual (HGDP00794) and used the DIYDodecad byseg mode to construct a local ancestry map of this individual with sliding windows 500 SNP long, advanced by 50 SNPs.

In essence, this procedure assigns ancestry to local regions of the genome, and can hence be used to identify, e.g., regions with an excess of West_Asian or any other component.

I decided to use an Orcadian for this purpose, since Orcadians are Indo-European speakers from northwestern-most Europe who can be reasonably thought to have minimum admixture in historical times from groups other than (i) the pre-Indo-European substratum, (ii) Celto-Germanic superstratum which dominates western Europe.

It is clear that the K=8 West_Asian segments of the Orcadian individual correspond to the K=5 South_Asian (+0.36) and Near_East (+0.29) segments.

The negative correlation with the K=5 Atlantic_Baltic (-0.4) component further indicates that in genomic regions where the Orcadian has West_Asian segments, there is a deficiency of Atlantic_Baltic ancestry; we can be fairly sure that these segments are not the result of common partial descent of Orcadians and West Asians from a northern Caucasoid population.

I have also repeated the above experiment, but with a Lithuanian individual (GSM536635). Lithuanians are surrounded by Indo-European and Finno-Ugric speakers, and are also relatively unlikely to have experienced other gene flow in more recent times:

The same general pattern emerges, with the K=8 West_Asian segments corresponding to K=5 Near Eastern and South Asian segments.


Chunks of DNA in an Orcadian and a Lithuanian that are labelled West_Asian tend to be related to populations of the Near East and South Asia.

It seems fairly clear that a post-5kya link between the North Atlantic and the Indian subcontinent admits to a single parsimonious explanation: the expansion of the Indo-Europeans out of their West Asian homeland during the Copper and Bronze Ages. July 03, 2012