What are the markers used to donate ancestry?

[doctorisscientia]

General Question:

What are the markers used to donate specific ancestral information?

For example, tribe A is 45% African.

Is it uniparental ancestry?

[sundiata]

Is it uniparental ancestry?

Yes because they change very slowly and can be traced in a linear fashion.

[doctorisscientia]

Thanks Sundiata, I'm asking in regard to the Beja people.

The somewhat recent Tishkoff study stated that the European input into the Beja to be approximately 35%, but various other studies and genetists have replyed negatively towards the results, being that the European input according to uniparental markers into the Beja people being only 5%. The same thing has been discussed in regard to the Dogon results in the same study.

Is it likely that they didn't differentiate ancestral lineages shared between both Africans and non-Africans and those which are largly due to recent non-African admixture as is the case of the northern coastal population?

[sundiata]

^It is likely that the Beja share some genetic characteristics with non-Africans due to chance, common ancestry via out of Africa, or recent admixture. Tishkoff is a good source as she's the one who pointed out that since horn Africans possess a sub-set of overall sub-Saharan African genetic diversity and that Europeans contain of sub-set of Horn African diversity, that genetic affinity is more explained by Horn Africans being the parent populations for the ancestors of Europeans. If Beja are related to Horn Africans, surely this extends to them. Also, interesting that you bring up the discrepancy between the 35% and 5% figures. I just commented in Zarahan's Cavalli-Sforza critique thread on a quote from Keita with data that relates directly to this:

Jun 1, 2010 12:49:09 GMT -5 sundiata said:

Harrison (1984:61), considering the results of
simulations which evaluate interpopulation gene flow
with and without selection, points to
another important issue concerning migration and
population affinities: the causes of inter-population
patterns of gene frequency variation, and the meaning
of affinities based not on ancestry but on genetic
similarity. Clearly with varying levels of selection,
populations which are genetically similar may have
a much more remote common ancestry than populations
which are genetically more different.

This is a beautiful point which to me effectively kills any debate with Eurocentrists. Consider if my brother and I had different fathers but same mother; his father being white and mine black. Then say I pick a random black person on the street. Would I be more related to this fellow, or my half white blood brother? The answer is obvious and the example no where needed to be so extreme to demonstrate this point since obviously NE Africans are not "half white".

[doctorisscientia]

Jun 1, 2010 14:30:46 GMT -5 sundiata said:

^It is likely that the Beja share some genetic characteristics with non-Africans due to chance, common ancestry via out of Africa, or recent admixture. Tishkoff is a good source as she's the one who pointed out that since horn Africans possess a sub-set of overall sub-Saharan African genetic diversity and that Europeans contain of sub-set of Horn African diversity, that genetic affinity is more explained by Horn Africans being the parent populations for the ancestors of Europeans. If Beja are related to Horn Africans, surely this extends to them. Also, interesting that you bring up the discrepancy between the 35% and 5% figures. I just commented in Zarahan's Cavalli-Sforza critique thread on a quote from Keita with data that relates directly to this:

Jun 1, 2010 12:49:09 GMT -5 sundiata said:

This is a beautiful point which to me effectively kills any debate with Eurocentrists. Consider if my brother and I had different fathers but same mother; his father being white and mine black. Then say I pick a random black person on the street. Would I be more related to this fellow, or my half white blood brother? The answer is obvious and the example no where needed to be so extreme to demonstrate this point since obviously NE Africans are not "half white".

Thanks for the input.

When you say:

"It is likely that the Beja share some genetic characteristics with non-Africans due to chance, common ancestry via out of Africa."

Via Out of Africa, in the sense that the Beja along with other Horn of Africans represent the local descendants of the Africans who migrated out of Africa?

And being that they are the local descendants them possessing lineages which are "ancestral" to those found in non-Africans, i.e. M1, N, and even IJK, shouldn't be a surprise in that regard.

On another note:

Have you heard about the possible location of J in Africa, i.e. East Africa/ Southern Yemen/Socotra.

From what I understand the J found among East Africans is not only not of recent or historic admixture, but that they don't even match those found in Arabia or Yemen.

Could it be that these lineages developed among a population staddling East Africa/Southern Yemen/ Socotra, with expansions west and north, with the East African and Arabian versions developing on their own, parallel evolution.

I find it intresting that J, J1 and J* in particular, are found in higher freguencies in areas in more southern locals, with the freguency being highest in Socotra, Southern Yemen, and East Africa then places to the north, including Egypt. Northern Sudan, followed by Upper Egypt, and decreasing into Lower Egypt.

note: Both Southern Yemen, i.e. Hadhramaut, and Socotra have obvious biological and cultural links in East Africa.

^sometimes incorrectly labeled Veddoid in regard to Socotra.

[sundiata]

Have you heard about the possible location of J in Africa, i.e. East Africa/ Southern Yemen/Socotra.

From what I understand the J found among East Africans is not only not of recent or historic admixture, but that they don't even match those found in Arabia or Yemen.

Given this, your scenario seems highly likely when piled with other data showing that east African maternal lineages outside of Africa has its highest frequencies in Yemen. I'd argue that southern Yemenis are relic populations from the horn that received a recent influx from northern Arabians. In addition, that the J haplogroup has its highest concentration in Yemen as well, may point to the fact that J could have emerged outside of Africa, but among African descent populations on the other side of the red sea.
.............................

Another point, as I had an epiphany while commenting in this thread. Recall Bowcock (1991) showing Europeans as being 1/3 African (recently derived). Connect this to Tishkoff showing Europeans to contain a sub-set of Horn African diversity. Think of the claim that Beja share up to 35% of their genome with Europeans.

Now, does it take a huge leap to propose, maybe since Europeans are 1/3 African (33.3%) that the alleles shared between some NE Africans and Europeans are actually African to begin with, and not European? If not nearly the entire 35%, would it not be expected that much or most of the similarity is due to the sharing of common African genes? I really think that this should be looked into further. If you ever speak with Dr. Keita again, can you relay this inquiry? I know he's made similar arguments concerning uniparental markers.

[doctorisscientia]

Jun 1, 2010 20:14:02 GMT -5 sundiata said:

Have you heard about the possible location of J in Africa, i.e. East Africa/ Southern Yemen/Socotra.

From what I understand the J found among East Africans is not only not of recent or historic admixture, but that they don't even match those found in Arabia or Yemen.

Given this, your scenario seems highly likely when piled with other data showing that east African maternal lineages outside of Africa has its highest frequencies in Yemen. I'd argue that southern Yemenis are relic populations from the horn that received a recent influx from northern Arabians. In addition, that the J haplogroup has its highest concentration in Yemen as well, may point to the fact that J could have emerged outside of Africa, but among African descent populations on the other side of the red sea.
.............................

Another point, as I had an epiphany while commenting in this thread. Recall Bowcock (1991) showing Europeans as being 1/3 African (recently derived). Connect this to Tishkoff showing Europeans to contain a sub-set of Horn African diversity. Think of the claim that Beja share up to 35% of their genome with Europeans.

Now, does it take a huge leap to propose, maybe since Europeans are 1/3 African (33.3%) that the alleles shared between some NE Africans and Europeans are actually African to begin with, and not European? If not nearly the entire 35%, would it not be expected that much or most of the similarity is due to the sharing of common African genes? I really think that this should be looked into further. If you ever speak with Dr. Keita again, can you relay this inquiry? I know he's made similar arguments concerning uniparental markers.

Exactly, Southern Arabia, i.e. Yemen, has been historically linked to populations and civilizations in the proximity of the East African Red Sea and beyond (other parts of Africa) then they are to more northerly Arabian sites. I would argue that Haplogroup J would have had originated among a population which straddled both East Africa/ Hadhramaut, having a common origin on the African continent. The fact that the J found in East Africa, i.e. Sudan and Ethiopia, for the large part did not back-migrate into the area or people from either Yemen or Saudi Arabia... suggesting a parallel evolution between the two intertwined regions. I would also have to agree with you in regard to that modern Yemeni populations experiencing relatively recent gene-flow from more northerly areas, i.e. Northern Saudi Arabia, Levant, and even Iran. Their African ancestry is still remarkably evident and obvious, among all populations culturally, but biologically most pronounced among Hadhremaut and Tihama populations.

You know what I never thought about that, but it does make so much sense in regard to your point. And since the Beja are less than 5% "European" in regard to their uniparental ancestry, any shared commonality between East Africans, including the Beja, and Western Eurasians would be significant gene-flow into Western Eurasia from particular African populations.

Do East Africans, being the local descendants of the OOA population, show up intermediate between East Asians or other non-Western Eurasian populations and other Africans? Or is this trend only seen in regard to 1/3 African Western Eurasians?

Western Eurasians have already showed up to be much closer to Africans then other human populations. Therefore suggesting gene-flow from Africans into the Western Eurasian gene-flow. In which would cause Western Eurasians to cluster closer to Africans.

I'll try to contact Keita with this info.