Admixture

Comment's on RM's "Foreign Admixture in Europeans"

RM has put up a page which supposedly attempts to quantify foreign admixture throughout Europe by adding up non-European mtDNA and Y-chromosome haplogroups.

Here are a few issues that need to be pointed out:

• Mongoloid and Negroid admixture are not equivalent

  It is misleading to lump "Mongoloid and Negroid Ancestry" together. This was done on some mtDNA studies where levels of non-Caucasoid admixture are relatively low. But once "Mongoloid" Y-chromosome haplogroups are added, combining "Mongoloid and Negroid" can only confuse the issue.

  Negroids are much more genetically distinct from Caucasoids than are Mongoloids. Northern Mongoloids are the major race to which Caucasoids are closest. According to Cavalli-Sforza's data, the distance between English and Chuckchi is 114, compared to an English-Bantu distance of 462, and an English-Sardinian distance of 65.

  A given amount of Mongoloid admixture will alter a European population less than the same amount of Negroid ancestry.

• "Mongoloid" Y-chromosome marker frequencies in NE Europe do not accurately reflect Mongoloid admixture

  As has been explained to RM repeatedly, "Mongoloid" Y-chromosome haplogroup HG16 is present in Finland, for example, at levels that far exceed any actual Mongoloid contribution in the gene pool. There are at most a total of 10% Mongoloid genes in Finland from all sources, and HG16 in Sweden contributes at most perhaps 1% Mongoloid genes in Sweden. And that's if Tat C is even Mongoloid. The most recent research indicates Tat C probably arose in a Caucasoid population, since Tat C lineages show more diversity among NE Europeans than among Mongoloids. Tat C was probably brought to Siberia by Caucasoids, and the high levels in some Northern Mongoloid populations may be accounted for by genetic drift or the founder effect. Or, as some researchers have speculated, Tat C may may have been selected for through offering a survival advantage in northern climates.

• Frequencies of "Mongoloid" mtDNA markers may have been amplified by selection in northern climates

  Looking at levels of "Mongoloid" mtDNA markers in Northern Europe, one would probably overestimate any actual Mongoloid maternal contribution. Haplogroups A, C, and D "improve the body's resistance to cold" and are positively selected for by northern climates. Northern Europeans already have lower levels of non-Caucasoid DNA than Southern Europeans. But, when we take this environmental selection of mtDNA into account, the difference may be even greater.

• RM confirms that NW Europeans have the least non-Caucasoid admixture

  Inhabitants of the British isles have comparatively tiny amounts of non-Caucasoid admixture, as even RM is forced to admit. Once Tat C lineages are removed from consideration, Scandinavians too have extremely low levels of non-Caucasoid ancestry. RM has deceptively included Tat C lineages in an attempt to obscure actual trends in admixture, confuse his readers, and draw attention away from the fact that Italians have much more non-Caucasoid admixture than the British. RM's data -- after one subtracts Tat C lineages -- simply confirm that admixture is highest in the east and south, and lowest in the northwest.

As mentioned above, with autosomal data, we can quickly dispose of idiocies like RM's claim that Finns are 30% Mongoloid. Y-chromosomes and mtDNA are not perfect, but until we see more work on autosomal DNA, they will have to do.

Maternal admixture

The trend is clear. Just as common sense would indicate, admixture is highest in the south and east, and lowest in the northwest. Contrary to the hopes of some southern Europeans, even Mongoloid admixture is higher in southern Europe than in northern Europe. And, as discussed above, haplogruops A, C, and D, are positively selected for by northern climates, so the real difference may be even larger than indicated here.

Also, keep in mind that examining mtDNA haplogroups is by no means a perfect method for determining admixture. mtDNA is more sensitive to genetic drift than nuclear genes (the effective population size is reduced by a factor of four). Over time, mtDNA lineages will die out. Some women will have sons but no daughters. This means that foreign mtDNA markers can be lost while foreign autosomal genes remain. This is true even given a stable population size, but it is especially true in periods of population decline, of which southern Europe has experienced several.

At the time of the Roman Empire [the population of Italy] was estimated to be around 6-7 million, about 10% of the total population of the empire. In the Middle Ages, and probably as early as the third century, socioeconomic conditions began to deteriorate. The wars between Byzantines and Goths, and later with Langobards, caused a population decrease, down to a minimum in the seventh century of about 50% of the maximum at the time of the Roman Empire. Numbers started to grow again (McEvedy and Jones 1978) and were 10 million by A.D. 1300. After the serious losses caused by the epidemic of plague known as the Black Death in A.D. 1340, they were down to 7 million and returned to 10 million in A.D. 1500. C. Cipolla (pers. comm.) gave values about 3 million higher with the same trend. [Cavalli-Sforza et al., HGHG, 280]

Similarly, the Peloponnese was depopulated on some occasions during Byzantine rule, and again during the Greek War of Independence. So, mtDNA most likely does not reflect the full extent of admixture in southern Europe.

Maternal Near Eastern ancestry. Counting up occurences of haplogroup J, for example, is unlikely to tell us much. As I've already discussed, frequencies of haplogroup J may have been influenced by selection. It is true that there are relatively "high" levels of haplogroup J in northern Europe (~14%). Note, the frequency of J is about the same in Iceland as in the "Celtic" areas of Britain. Moreover, the frequency of J mtDNA in northern Europe is dwarfed by the frequency of "Hamito-Semitic" male ancestry in southern Europe.

What mtDNA can tell us is that there has been a great deal of recent gene flow into eastern Mediterranean Europe.

The results are shown in table 5. Strikingly, although the level of recent gene flow surviving under this criterion is similar for most populations, at 5%-9%, the eastern-Mediterranean region (samples from Thessaloniki, Sarakatsani, and Albanians) has a very high value, 20%. This may reflect the heavy historical gene flow known between Greece and other populations of the eastern Mediterranean. (Richards et al. 2000)

Paternal admixture

Male sub-Saharan ancestry has been detected in Portugal (1 lineage in a sample of 93), Sardinia (1/10), France (1/40), Corsica (11/328), and the Greek island Mitilini (1/27).

Male Near Eastern (HG9) and North African (HG21) ancestry. Note: "Neolithic" markers (HG9 and HG21) in Central and Northern Europe are probably genuinely Neolithic. The same can't necessarily be said in Greece, Portugal, southern Spain, or southern Italy, where more recent gene flow may be indicated (cf. Richards 2003).

Summary/Conclusion

The methods I've used on this site to estimate admixture are imperfect, but they do reliably demonstrate trends, which are in accordance with history, geography, and common sense.

In general, northern Europeans have less non-Caucasoid admixture and much less non-European admixture than southern Europeans, and western Europeans have less than eastern Europeans. Europeans have less non-Caucasoid admixture than Middle Easterners, who in turn generally have less admixture than North Africans.

Levels of non-Caucasoid admixture are relatively low throughout Europe, and may be fairly inconsequential; on the other hand, relatively recent non-European Caucasoid influence probably noticeably affected phenotypes in some areas of southern Europe.

We don't really know what effect non-Caucasoid admixture has had on Europe. Roger Pearson seems to think it may have been significant even in Britain:

. . . the West European portion of the Caucasoid gene pool has changed considerably over the past century (as graveyards in London show). This is especially due to an increasing admixture with non-Caucasoid elements

I remain skeptical. If Pearson is correct in claiming the British gene pool has deteriorated, I suspect internal migration and differential reproduction within Britain were the main culprits, the small amounts of foreign admixture being negligible by comparison (of course, with mass third world immigration, this may be changing). But, to whatever extent admixture affected Britain, the effects in southern Europe were many times greater.

Sources

Belledi et al. Maternal and paternal lineages in Albania and the genetic structure of Indo-European populations. Eur J Hum Genet 2000 Jul;8(7):480-6.

Gonzalez et al. Mitochondrial DNA affinities at the Atlantic fringe of Europe. Am J Phys Anthropol 2003 Apr;120(4):391-404.

Helgason et al. mtDNA and the Islands of the North Atlantic: Estimating the Proportions of Norse and Gaelic Ancestry. Table 3. Am J Hum Genet 2001 68:723-737.

Lucotte et al. North African Genes in Iberia Studied by Y-Chromosome DNA Haplotype 5. Human Biology 73.5 (2001) 763-769.

Lucotte et al. Y-chromosome haplotypes in Corsica. C R Biol 2002 Mar;325(3):191-6.

Richards et al. Tracing European founder lineages in the Near Eastern mtDNA pool. Am J Hum Genet 2000 Nov;67(5):1251-76. Supplementary data.

Richards et al. In Search of Geographical Patterns in European Mitochondrial DNA. Am J Hum Genet 2002 Nov;71(5):1168-74.

Richards, M. The Neolithic Invasion of Europe. Annual Review of Anthropology 2003 32:135-62.

Romano et al. Autosomal microsatellite and mtDNA genetic analysis in Sicily (Italy). Ann Hum Genet 2003 Jan;67(Pt 1):42-53.

Rosser et al. Y-Chromosomal Diversity in Europe Is Clinal and Influenced Primarily by Geography, Rather than by Language. Table 1. Am J Hum Genet 2000 67:1526-1543.

Wells et al. The Eurasian Heartland: A continental perspective on Y-chromosome diversity. Proc. Natl. Acad. Sci. USA, Vol. 98, Issue 18, 10244-10249, August 28, 2001.