PDA

View Full Version : Sub-Saharan African Admixture in North Africa



Logistics
07-09-2016, 11:38 PM
What are some consistent numbers for SSA admixture in North Africa? Most numbers I see are basically the following:

-Northern Morocco = 10% SSA
-Southern Morocco = 25% SSA
-Algeria = 10-15% SSA
-Tunisia = 15%
-Libya = 15%
-Northern Egypt = 15%

Rather accurate? No. I know for a fact the SSA admixture in North African Arabs is largely from the slave trade and same can be said for the admixture in Gulf Arabs (10-15% SSA) as this admixture is almost exclusively of a maternal (haplogroup L) origin and various reputable sources have confirmed the admixture to be largely a legacy of the slave trade. What about in Berbers? Is SSA admixture in Berbers a legacy of the trans-saharan slave trade? Are black Berbers (Tuaregs, Masmuda's, etc.) slave descendants?

Sikeliot
07-09-2016, 11:45 PM
Most North African Arabs are Arabized Berbers actually.

tippy
07-09-2016, 11:47 PM
We don't have enough berber samples to come to a good conclusion.

Logistics
07-10-2016, 12:11 AM
Most North African Arabs are Arabized Berbers actually.

How many Arabs in NA are actually of Arab ancestry? Not Arabized Berbers -- actual Arabs of Arabian Peninsula stock.

Megalophias
07-10-2016, 12:16 AM
I know for a fact the SSA admixture in North African Arabs is largely from the slave trade and same can be said for the admixture in Gulf Arabs (10-15% SSA) as this admixture is almost exclusively of a maternal (haplogroup L) origin and various reputable sources have confirmed the admixture to be largely a legacy of the slave trade.

Gulf Arabs:
Qatar (n=72): 3% B2, 4% E2, 3% E-M191. Total 10% probably Sub-Saharan Y DNA.
UAE (n=164): 5.5% E-M191. Total 5.5%.
Oman (n=121): 1% A-M13, 2% E2, 2% M191, 5% other M2. Total 9%.

Doesn't look "almost exclusively maternal" to me.

Logistics
07-10-2016, 12:31 AM
Gulf Arabs:
Qatar (n=72): 3% B2, 4% E2, 3% E-M191. Total 10% probably Sub-Saharan Y DNA.
UAE (n=164): 5.5% E-M191. Total 5.5%.
Oman (n=121): 1% A-M13, 2% E2, 2% M191, 5% other M2. Total 9%.

Doesn't look "almost exclusively maternal" to me.

Largely maternal should've been a better word. Not like 50% of Saudia Arabians are E1b1a.

fished
07-10-2016, 12:31 AM
How many Arabs in NA are actually of Arab ancestry? Not Arabized Berbers -- actual Arabs of Arabian Peninsula stock.

Very few. If you mean "pure" Arabian, with 0 Berber admixture, probably none.

All Berbers have ancient, "stabilized" SSA admixture in the same way all Indians have ASI admixture. The "North African" component itself in most admixture calculators is a mixture of West Eurasian and African, although mostly West Eurasian. It will take ancient samples from the Maghreb to tell us exactly when and from where this admixture originated.

As you say, however, there are widely variable degrees of more recent West African admixture in many North Africans, especially in urban areas, associated with the Arab slave trade.

Sikeliot
07-10-2016, 04:47 AM
The northern coasts of North Africa also have less SSA but once you get into southern Morocco, southern Egypt, etc. it gets much higher.

Morges
07-10-2016, 06:26 AM
I have some Tunisian friends and they have wide range of phenotypes, one girl looks almost full black (I guess some sort of around 70% SSA) and she must be recent black ancestors, while others show some SSA traits but not so much visible. One girl of there I know could pass for European. Anyway we need to ancient Berber samples to see how much do they have change and if the SSA admix is recent or not.

ffoucart
07-10-2016, 07:41 AM
could be helpfull:

http://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1001373

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3718088/

E_M81_I3A
07-10-2016, 07:52 AM
There was an interesting Genetic Analysis of North West African Populations (http://www.dnatribes.com/dnatribes-digest-2012-02-01.pdf) by DnaTribes in 2012.

This study, which used data from Henn et al. 2012 (http://journals.plos.org/plosgenetics/article?id=10.1371%2Fjournal.pgen.1002397), is interesting because, contrary to other studies (like Henn 2012) it excludes local North African admixture which allows to show West and East African contributions in North West African populations. As per the study, probably that most of SSA is rather old in North Africa : "This suggests consistent patterns of contact between North Africa and Sub-Saharan regions, possibly dating to early periods such as the Capsian culture that was in contact with East Africa during the “Green Sahara” period. "

"Results identified a primary Middle Eastern component and a secondary European and Sub-Saharan African components"

"Regional admixture components in several North African populations (excluding North African admixture) were quantified using autosomal SNP data. Results are summarized in Table 2 and illustrated in Figure 3."

http://img15.hostingpics.net/pics/383441DnaTribesNAF22012.jpg (http://www.hostingpics.net/viewer.php?id=383441DnaTribesNAF22012.jpg)

http://img15.hostingpics.net/pics/840946DnaTribesNAF2012.jpg (http://www.hostingpics.net/viewer.php?id=840946DnaTribesNAF2012.jpg)

Discussion: Results in Table 2 indicate that excluding North African admixture, the largest admixture component for sampled North African populations was from the Arabian region (average 60.3%). This region includes populations of the Arabian Peninsula, Egypt, and the Levant. The pattern of Arabian admixture is consistent with periodic North African links with the Near East via the Southern Mediterranean coast (such as Phoenician migrations) and via the Sahara Desert (such as Arab migrations). Results also indicated an Atlantic European component for all studied populations. This was largest in North Morocco (27.2%) and Tunisia (23.4%). Although based on separate SNP analysis, this Atlantic European component is similar to the more specific Portuguese genetic link identified based on STR data (see previous section). This Atlantic European admixture might express ancient contacts between North Africa and the West Mediterranean dating to the prehistoric Iberomaurusian and Capsian cultures, as well as Phoenician and Moorish links with Southwest Europe in later periods. In addition, results identified Horn of Africa (average 8.5%) and West African (average 13.3%) components. These were most pronounced for the South Moroccan population, but were found for all studied North African populations. This suggests consistent patterns of contact between North Africa and Sub-Saharan regions, possibly dating to early periods such as the Capsian culture that was in contact with East Africa during the “Green Sahara” period.


Same conclusions for Kamla-Raj 2015 "Variation of Rhesus Haplotype Frequencies in North Africans and in Worldwide Population Analyses" :

"Historic data are in favor of the oldness of this sub-Saharan component in North Africa. In fact since about 5000 years before present (BP) the immense Sahara desert already had the current severe climate that represents a considerable barrier to human migration, but before it had better climate more accessible to people movements (Pons et al. 1991). Besides the first stage of Neolithic Age in North Africa designated “Neolithic with Sudan origins” is the only period characterized by some sub-Saharan ethnic contribution from Sudan. This cultural period was started about 9000 years BP in the extreme south of North Africa, where the current Saharan regions of Algeria are (Gragueb and Mtimet 1989). Hence the possible presence of slight sub-Saharan component in current North African populations could be traced back to the first stage of Neolithic Age. Namely it would be an original component already present in the gene pool of Berber ancestors’ natives of North Africa."


Regarding mtdna L lineages in North Africa, 2/3 seems to be very old (around 8000 BP) and only 1/3 due to slave trade and contemporaneous events. Indeed as per recent study by Hernández et al. 2015 (http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0139784) : "The most intense input of L lineages into North Africa (Fig 4A) seems to have occurred at two time points, one of which was very recent (peak at 0) and the other around the period of the Holocene climatic optimum, HCO (6.6 ky—f1 criterion—; and 11.8ky—f2 criterion). When trying to impose two migration periods, the first at 0.5 ky (f1) (representing the slave trade from sub-Saharan western Africa to the Maghreb and other contemporaneous events) and the other at 8.0 ky (f2) (around the middle of the HCO), it is possible to ascertain (see Fig K in S1 File) that the older period was responsible for the migration of approximately 2/3 of current sub-Saharan L lineages observed presently in North Africa; the slave trade and contemporaneous events there would have introduced 1/3 of this pool."


But, of course we need Ancient DNA to know more about North African Genetics.

Odyss
07-10-2016, 08:14 AM
A good chunk of the African ancestry (12%) in Berbers is East African and is crucial in the ethnogensis of Berber languages. People confuse the Middle East with North Africa too much. Berbers are their own ethnie group with a culture that has deep African strain. In North Africa it is the Eurasian ancestry that is intrusive and not the other way around. Berbers have several types of Eurasian ancestry just like they have several types of African ancestry as well.

Morges
07-10-2016, 08:30 AM
This study have analyzed mtDNA.

http://bmcevolbiol.biomedcentral.com/articles/10.1186/1471-2148-10-138

E_M81_I3A
07-10-2016, 08:35 AM
This study have analyzed mtDNA.

http://bmcevolbiol.biomedcentral.com/articles/10.1186/1471-2148-10-138

Yes, Hernández et al. 2015 mentions this study from 2009 but Hernández et al. 2015 is much more recent and analysed a lot of samples from both Iberia and North Africa.

ffoucart
07-10-2016, 08:52 AM
Although, it must be said that some of Hernández et al. 2015 conclusions can no more be followed, as they considered U6 as a North African marker, but U6 has been found recently in paleolithic Europe:
http://www.nature.com/articles/srep25501

MfA
07-10-2016, 09:10 AM
Very few. If you mean "pure" Arabian, with 0 Berber admixture, probably none.

All Berbers have ancient, "stabilized" SSA admixture in the same way all Indians have ASI admixture. The "North African" component itself in most admixture calculators is a mixture of West Eurasian and African, although mostly West Eurasian. It will take ancient samples from the Maghreb to tell us exactly when and from where this admixture originated.

As you say, however, there are widely variable degrees of more recent West African admixture in many North Africans, especially in urban areas, associated with the Arab slave trade.

This maybe the truth but can't say for certainty as there is no aDNA evidence. Looking at the very young age(TMRCA 2100kya) of modern M81 descendents, North Africa suffered from huge bottleneck quite recently. It might be much less SSA affinities for North Africans not that long ago.

E_M81_I3A
07-10-2016, 09:34 AM
Although, it must be said that some of Hernández et al. 2015 conclusions can no more be followed, as they considered U6 as a North African marker, but U6 has been found recently in paleolithic Europe:
http://www.nature.com/articles/srep25501

Not exactly, it is known for a very long time that basal U6* is of Eurasian origin and its presence if North Africa is due to a back migration. Hernández is of course aware of this (he refers to the studies that demonstrated this like Maca-Meyer 2003, Secher 2014 etc).

In his study, Hernández is analysing the numerous derived U6 haplotypes (not the basal U6* which is very old, 50 ky, and not very informative in that case) which are much more younger and formed, for most of them, into North Africa and then were introduced from North Africa to Iberia.

ffoucart
07-10-2016, 10:04 AM
In his study, Hernández et al. 2015 is analysing the numerous derived U6 haplotypes (not the basal U6* which is very old, 50 ky, and not very informative in that case) which are much more younger and formed, for most of them, into North Africa and then were introduced from North Africa to Iberia.
Except, that they themselves discuss of the origin of U6, as they couldn't prove an North African origin of U6b U6c... and considered only U6a as a proof of North African migration into Europe in part because U6 was not proven to be an European subclade.

So, with only modern samples available for North Africa, you can't rule anymore an European origin of U6a.

E_M81_I3A
07-10-2016, 10:08 AM
Except, that they themselves discuss of the origin of U6, as they couldn't prove an North African origin of U6b U6c... and considered only U6a as a proof of North African migration into Europe in part because U6 was not proven to be an European subclade.

So, with only modern samples available for North Africa, you can't rule anymore an European origin of U6a.

Maybe... So in any case, it just shows that migrations occurred between North Africa and Southern Europe, in both directions, for thousands of years. As the author wrote "in particular, our work supports the existence of an ancient, frequently denied, bridge connecting the Maghreb and Andalusia."

Táltos
07-10-2016, 02:33 PM
Just a heads up. Thread moved from the General forum to the Africa forum. 1 week redirect link left up in the General section.

Ignis90
07-10-2016, 04:37 PM
Without any aDNA, it's all speculation of course.


The way I see it right now is how homogeneous the main population is despite the very big distances, from the Atlantic ocean to Libya. It's pretty much the lenght of West and Central Europe. When excluding the various outliers, most NW Africans have similar types of Eurasian and African ancestries and at similar amounts, plus or minus at best a few percents. This, to me, supports some degree of continuity from at least before the advent of Islam. I could be wrong, and if it was the case, I would be very surprised and it would mean the population size got very small after the fall of the Roman Empire in order to reshape the entire region homogeneously.

As to the African ancestry, part of it is East African-like and part of it is more West-Central African-like. The first one is found in all Afroasiatic speakers (except some Chadics?) while the latter might be a specificity of NW Africans if it is found in old enough aDNA. In any case, African/non OoA diversity is still poorly understood at this point so it might very well be more complex and more exciting than this two signals of African ancestry in NW Africans.
As Odyss said, while Berbers are usually lumped with West Asians (and other West Eurasians) because there is a lot of commonalities (genetically or not), the African ancestry (and maybe the Eurasian one, the way Sardinians are "unique") makes the population very distinct from everybody in a global context (with Egyptians being somewhat close unsurprisingly) and in a West Eurasian context.

As for the presentation at the SMBE event a few days ago (Günther Torsten), it seems they found two individuals from a cave in Morocco being modeled as a mixture of European Mediterraneans and South Africans. Considering it's dated to the Middle Ages, it doesn't tell anything about the genesis of NW Africans as it is very likely the result of Arab era slavery. The only surprise is that they could enslave people from as far as Southern Africa, which is much more astonishing the Icelander slaves.

E_M81_I3A
07-10-2016, 06:11 PM
...it is very likely the result of Arab era slavery. The only surprise is that they could enslave people from as far as Southern Africa, which is much more astonishing the Icelander slaves.

... not linked to the subject, but regarding Icelanders enslaved in the 17th Century in North Africa, they were in fact (like most of the Europeans captured in the 16-17th century by Muslims) captured not by Arabs but by European corsair captains (like the Dutch renegade Jan Janszoon alias Murat Reis) who "work" for North Africans. More than half of the Barbary corsair captains were indeed Europeans (Dutch, English, Italians, Iberians etc converted to Islam or not), then Turks but no Arab...

Catterick
07-11-2016, 06:04 AM
Some SSA in Northwest Africa must surely date back to the Green Sahara period. What's needed is whole genome aDNA from Iberomaurusan and Capsian times. Though most gene flow or migration was southward because the people of the Gobero graves were all closer to Stone Age people around the Mediterranean basin.

Logistics
07-11-2016, 08:45 PM
I have some questions about Northeast African DNA (mostly regarding Horn Africans and the Sudanese) overall. Here the my questions:

1) What is the origin of the SSA admixture in Horn Africans and Sudanese? Is of Bantu origin?
2) How did Sudanese get their Caucasoid admixture? I know Horn Africans got theirs as a result of back-to-Africa migrations from West Asia during neolithic times. What about Sudanese? Same result?
3) When did these West Asians first start settling in the Horn during these back-to-Africa migrations? The specific dates? What is it just like incessant during the neolithic times and there wasn't like one big massive back-to-Africa migration at one time?

Thanks.

B)