Wednesday, November 3, 2010 Westward Across the Asian Steppe; Southward Through the Himalayas Ever see patterns just above the noise floor of data?
As I was staring across Dienekes' admixture results, I noticed a minor "signal" drifting across various populations. At low levels, this signal seems to be everywhere east of Romania. After analyzing it, I realized it was like a seesaw, with a dominant Northeast Asian component in West Asia and a dominant East Asian component in Southern India.
Here, I plot populations that show evidence of this Northeast-East Asian seesaw. In the plot, I also keep the North European component in the data, as it seems somehow to follow along with the Northeast and East Asian components. However, because there seems to be evidence of multiple admixture events between Northeast Asian-East Asian-North European peoples and North European peoples, I normalize the data only on the Northeast Asian + East Asian components. I then organize the populations in a Northeast Asian dominant to East Asian dominant ordering:
On the x axis, the numbers represent the following populations:1=Chuvash 2=Lezgins 3=Sindhi 4=Adygei 5=Pathan 6=Turks 7=Syrians 8=Romanians 9=Jordanians 10=Ashkenazis 11=Uygurs 12=Burusho 13=NorthKannadi 14=GujaratiJust above the xaxis of the graph, you see the Northeast Asian component in turquoise and the East Asian component in lime. Above that is yellow North European component, which I didn't include in the normalization.What does this say about Eurasian gene flow? First, while the North European component is present almost everywhere, it is very non-uniformly distributed. In a few cases, such as the North Kannadi and the Uygurs, it is virtually absent. This would indicate that East Asian populations were in contact with South Indian populations, in the absence of any North European contact.
The likely path for that contact is either across the Himalayas or through Southeast Asia.On the other end of the spectrum, we see Northeast Asian contact and admixture with Eastern European, West Asian and Northern Indian groups. Their low level Northeast Asian-East Asian signature is tilted toward Northeast Asia and the people of the Central Asian Steppe. While this signature is at low level, it is a clear indication of an ancient flow of peoples from the Central Asian Steppe into West and South Asian.The clear similarity of this signature between Lezgins, Sindhi, Adygei, and Pathans indicates gene flow from the Central Asian Steppe into the Eastern Caucasus and into Afghanistan, Pakistan and Northwest India. Carried along with this low level Asian Steppe signature is a contribution to the gene pool from North Europeans among the Lezgins, Adygei, Sindhis, and Pathans. The North European contribution occurs at 24.5, 15.7, 7.4, and 13.2 percent, respectively. All four of these populations also carry a contribution from West Asians at greater than 35%.
We can't know for certain the ordering of these admixture events. However, it does seem probably that a Steppe people joined a West Asian/North European people and then proceeded southward into Afghanistan, Pakistan and India.At very low levels, the Northeast Asia/East Signature is present in people of the Levant. Looking at Jordanians, for example, the North European component is quite suppressed. This would indicate that a separate migration or diffusion from the Steppe occurred for the Levant, absent the North European component. To better understand this, it would be necessary to examine populations from neighboring Iraq and Iran.The Chuvash, on the far left of the graph, show indictation of absorbing people from Northeast Asia, West Asia and Northern Europe, but with proportionally little contribution from East Asia. The similarity of proportion of their Northeast Asian-East Asian signature with the Adygei and Lezgins is suggestive. http://linearpopulationmodel.blogspot.com/2010/11/westward-across-asian-steppe-southward.html