Substack cometh, and lo it is good. (Pricing)

Admixture in South African Afrikaners

I teased this yesterday, and I don’t like to do that, so I’m going to put up a quick post. Something more thorough will go up on the Family Tree DNA blog at some point soon. Basically I have heard through the grapevine that something on the genomewide patterns of Afrikaners would be published “soon” in the fall of 2012, but that hasn’t happened. So a few weeks ago I went looking in the Family Tree DNA database, and extracted individuals who stated that both parents were born in South Africa. Twelve of those had more than 90 percent European ancestry. So it is likely that these individuals are either Afrikaner or non-Afrikaner South African whites. I’ll hold off on the admixture results until the Family Tree post, but visual inspection made it clear that most of them they had non-European ancestry. American whites and Europeans generally have no non-West Eurasian/North African ancestry, with the exception of some Spaniards, who have small Sub-Saharan segments, probably mediated through the Moors. Many of these individuals had affinities at low levels to Khoisan, South Asians, Southeast Asians, and West Africans. These are as it happens the non-European groups who contributed ancestry to the Cape Coloureds.

So as a preview, below are MDS plots of various populations. I filtered the SNPs down to ~150,000.

SA1

Compared to the British and Polish Northern European samples all but one of the South Africans is shifted toward Africans. One of the individuals is probably Xhosa, and a few Cape Coloured. But the others are over 90 percent European, in particular Northern European.

SA2

Here you see both South and Southeast Asian admixture in these individuals, by and large. A few individuals now overlap with the Northern European British/Polish cluster. They had lower levels of Asian ancestry.

SA3

Now you see that the South African whites are clearly like the British, not the Poles or Spaniards, but skewed in the direction of non-European ancestry. This makes sense since the parent populations of Afrikaners, French Hugenonts, Dutch, and Germans, are more like the British. One of the individuals fits nicely into the British cluster. I presume that this is a person of non-Afrikaner British background.

Around 60 percent of whites are Afrikaner in South Africa (there is some intermarriage obviously as well). This sample seems skewed toward that subset, so one has to question the ascertainment here. Perhaps people who got their DNA tested had suspicions or questions about their ancestry? This is plausible. But on the other hand the fraction of non-European ancestry is pretty much on average what the genealogical estimates of ~5 percent are. And, it’s relatively evenly spread through the white sample here that has admixture, indicating it probably dates to the founding generations of the Afrikaner ethnic group, which until recently has been endogamous.

More later.

Note: Click images for larger version.

Posted in Uncategorized

Comments are closed.