Sunday, March 23, 2008

"Ethno-specific" markers?

Is it possible to trace a marker to a single specific designated ethnic group, by DNA motifs alone?

Well, it is plausible that certain markers can be relatively more frequent in some highly culturally-"conservative" and relatively socially-isolated ethnic group—that just so happens to have marital customs which generally discourage exogamy with other ethnic groups—but rarer in other groups outside the said socially-isolated ethnic group. In such a scenario, it would mean that the said rare marker had in all likelihood emanated from a mutational event within a lineage that was inherited from TMRCA, but only occurred in the said population after it diverged from the source population to which the said TMRCA belonged. The resulting cluster in such a scenario could be inflated in its distribution in the said socially-isolated ethnic group as an after-effect of random but neutral genetic drift. However, is it possible even in such a scenario, for such an ethnic group to have always been completely isolated from nearby populations, or potential "invading" populations from far-off lands, throughout the lifetime of such a "unique" or one-time mutational event?

Even if one were to accept that possibility, then it would in reality be an issue of frequency and distribution pattern of the marker or cluster in question, rather than an issue of some kind of "ethnic" mark or tattoo of some sort on the cluster. That issue would be one that is centered on the observation of high frequency in one group and rarer distribution elsewhere.

From personal encounter, the present author is reminded of an exchange in a discussion from the past, wherein someone spoke of "maternal Tuareg ancestry" in African Americans. The obvious question here, is what constituted maternal ancestry from Tuareg or Tamasheq kels/groups, for it should be obvious to anyone who is remotely familiar with the Tamasheq, that they aren't exactly an isolated group of people, and have been interacting and living with other west African groups for a great deal of their history. Here is how that exchange went:

Originally posted by MyRedCow:

Al Takruri and Supercar,

This is the Kayser study:
http://www.genome.org/cgi/content/full/13/4/624

There is another .pdf file with the latest data with more Haplogroups. But, I've never once seen an African American man with the Berber E3b Y chromosome on any chart. The main Y chromosomes are always E3a and R1b. I said that the African Americans have maternal Tuareg ancestry. The slavetraders did not alway take the men and women from the same place. There maybe and African American maan out of about 17 - 20 million men who has the Berber E3b Y chromosome. But, I haven't seen one yet.

My response was thus:

That may well be the case, but the point is that Tuareg females in all likelihood share markers that you deem "Tuareg" with their west African neighbours. At least that is what I understand from the study cited in your post in the thread centered on the Fulani. Maybe there is a "Tuareg" specific maternal marker, in which case, it won't hurt to know what that is. — Mystery Solver

...and of course, the poster "MyRedCow" never followed through with that request, because in all likelihood, the present author reckons, he could not dig out the answer from anywhere, what would characterize a "Tuareg"-only maternal marker, and so, only traceable to Tuaregs/Tamasheq of the west African Sahel. As noted, the Tuaregs in all likelihood share markers with their neighbors north, south, west and east to their territorial expanse.

By the way, the study that the present author was talking about in the citation above, is this:

From
http://www.biomedcentral.com/1741-7007/4/34

Among the AFDIL sequences with more than five matches to various African ethnic groups, most language diversity was within the various subfamilies of the Niger-Congo family. These subfamiliesinclude Atlantic Congo (e.g., the ethnic groups Fula, Yoruba, Wolof, Balanta) and Mande (e.g., the ethnic groups Mandingo, Mende, Bambara). However, in some of the sequence matches, different linguistic families were represented altogether, including the Afro-Asiatic (e.g., the Tuareg ethnic group) and Nilo-Saharan (e.g., the Dinka ethnic group) families, along with members of the Niger-Congo family.

What one can take from the above...is that the samples from the said groups fall into said sequence patterns [yet to be identified] found in the intra-West African region, from which I infer that this could be due to geographical proximity and corresponding inter-ethnic miscegenation. It says nothing of ethno-specific maternal markers. However, if I missed it somewhere, I would welcome the prospect of it being brought to my attention. — Mystery Solver

As an illustrative example, let us consider Salas et al.'s examination of L1a mtDNA:

Haplogroup L1a (fig. 4a) is common (∼20%–25%) in East, Central, and southeastern Africa, and is almost absent in North, West, and southern Africa. The main subclade, L1a1, is ∼33,350 (SE 16,600) years old and is quite starlike, with a predominantly East/southeastern African distribution and a root type that is common in East Africa.

...L1a seems likely to have been brought to southeastern Africa by the eastern stream of the Bantu expansion, having been picked up in East Africa...

...The analysis of Soodyall et al....They showed an association between an intergenic COII/tRNALys 9-bp deletion and a subset of L1a types lacking the transitions from the CRS at both 16129 and 16168—that is, within L1a2. This deletion is common in southeastern African Bantu speakers, as well as some East and Central African groups. It was absent not only in all Khoisan groups but also in virtually all southwestern African Bantu speakers (with the exception of three Ambo individuals from Namibia, for whom a southeastern Bantu origin was proposed; see also Soodyall and Jenkins Soodyall and Jenkins, 1993). They propose a Central rather than an East African origin for the deletion; we concur that, although L1a seems most likely to have originated in East Africa, L1a2 may have emerged in Central Africa.

...A predominantly East African origin for L1a types also explains its relative scarcity in America, in comparison with other African types. Most American representatives of L1a, in fact, match types from southeastern Africa, and probably derive directly from that region. — by Salas et al. 2002

As observed from the above, Salas et al. trace the likely American representatives of L1a markers back to a particular region in Africa; in this case, southeastern Africa. To demonstrate this, it came down to frequency and distribution pattern of the sub-types examined. First it was stated that haplogroup L1a is rare in North, West and Southern Africa. Then, it is demonstrated that certain mutational events common in the east, central, and southeast African L1a2-bearing groups were rarer in southwestern African Bantu speaking groups. Finally, as already mentioned, the American representatives of L1a2 are said to match those of southeast African subtypes. The point that should be taken away, is one from the general understanding that a major portion of African American ancestry comes from western Africa, where haplogroup L1a is said to be rare; it is thus no wonder, according to the authors, that L1a representation in America is relatively scarce as well...BUT, the little representation that does appear therein, happen to match the southeast African subtypes, i.e. examples more frequent in southeastern Africa than elsewhere. Notice as well, that no attempt was made to implicate a singular ethnic group, but rather, a somewhat general but specific intra-African region. As the piece above suggests, the subtypes in southeastern Africa transcend a singular ethnic group; so, short of finding and sharing the micro-locus/loci details of "unique" subtypes in an even smaller geographical unit within southeastern Africa, tracing the American L1a examples back to southeastern Africa is as specific as the authors can get!