Read this article/blog of an attempt at mapping out the Somali/Horner genome using "more accurate" proxies and it came to quite a strange conclusion..

[Nilo-Saharan]

I was browsing around trying to look for some information on Mota-Hunter Gatherers and noticed an interesting and what appeared to be a thoroughly put-together article on the "Land of punt" website modeling the genome of Afro-Asiatic Horners with proxies that were more accurate.
Here's the link: https://landofpunt.wordpress.com/2022/11/

Anyways, I skimmed through it all for a bit and was quite impressed by the depth put into the article and the author seemed to be quite knowledgeable in a lot of areas he was discussing as well as a rich reference to sources. So I jumped ahead to the conclusion to see what the main findings were and came to the realisation that the author also had some very odd observations and statements which didn't sit right with what most of my Genetic research regarding this topic has suggested. But then again I didn't want to be an ignorant bastard and assume I must be right simply based on my own assertions that the research I made was the undeniable truth. So here I am on here to see what anyone else has to say about this, in case I may have either read and interpreted this wrong or the author of the article doesn't have a clue what they're talking about. For some, this might be quite hilariously stupid which is what I thought too but the author of the article came off as quite well-spoken and I'm just here to double-check that my confusion isn't for nothing.

 

[The alchemist]

I don't want to be rude, the brother seems to have put a lot of effort into it, but the models are all wrong. We have no steppe ancestry unless you model minorities with South Asian genetics. Nakht-Ankh, an interesting sample, has low coverage, so we know it can give some errors. When the results say 70% non-African genes for Somalis, and it prefers East Eurasian genetics, one has to wonder if he mixed up unscaled with scaled, target, and sources from the datasheets, or vice versa.

I use the same samples he used for the results (only limited to that), you can already see how the proportions align based on basal preference quite divergent from the guy's result:

No ancient steppe, European or East Eurasian DNA. Observe that the "Sub-Saharan" ancestry is leveled to the norm as well.

I have not properly vetted the sample myself but Kakapel_300BP is not close to Dinka. You can see how terrible the fit is. It's very recent as well so it is likely mixed with various distinct African components, and differentiated from the Kakapel_3900BP which is on the East African hunter-gatherer cline. I think the sample from 300 BP is heavily West African, with around 23% Dinka-like and the rest real Kakapel_3900BP and very minor PN (unsure if it catches non-African stuff or shared SSA in PN). The point is, it is definitely not our model Sub-Sahran African.

 

[Nilo-Saharan]

Thanks for the time taken in your response bro. I had a feeling something was wrong somewhere but just didn't know where to start lol especially with so much proxies being unecessarily introduced into the model.

Also the guy seems to mention "Eurasian ancestry" among Dinkas, is he making a reference to E1b1 presence and writing it off for Admixture or what?

 

[The alchemist]

Thanks for the time taken in your response bro. I had a feeling something was wrong somewhere but just didn't know where to start lol especially with so much proxies being unecessarily introduced into the model.

Also the guy seems to mention "Eurasian ancestry" among Dinkas, is he making a reference to E1b1 presence and writing it off for Admixture or what?

Besides very deep ancestry that is related to Eurasian back migration, and/or paleo-extinct peoples in northeast Africa that carried heterogeneity of heritage, I don't think the average Dinka got much admixture. It can vary slightly from person to person, but the average Dinka is pretty "Sub-Saharan African".

 

[Nilo-Saharan]

Besides very deep ancestry that is related to Eurasian back migration, and/or paleo-extinct peoples in northeast Africa that carried heterogeneity of heritage, I don't think the average Dinka got much admixture. It can vary slightly from person to person, but the average Dinka is pretty "Sub-Saharan African".

I don't think vahaduo even has a way for detecting this deep ancestry anyway no?

 

[The alchemist]

I don't think vahaduo even has a way for detecting this deep ancestry anyway no?

Vahadou gives tremendousinsight. Don't underestimate that tool. You can delineate evidence of deep ancestry through various modeling methodologies you make along the way.

 

[Nilo-Saharan]

Vahadou gives tremendousinsight. Don't underestimate that tool. You can delineate evidence of deep ancestry through various modeling methodologies you make along the way.

Would there be any way of finding this deep ancestry in Dinka samples because not one Dinka sample I've observed shows even a decimal of Eurasian admixture?

 

[The alchemist]

Would there be any way of finding this deep ancestry in Dinka samples because not one Dinka sample I've observed shows even a decimal of Eurasian admixture?

You might find individuals with relatively recent admixture, but that is generally outlier stuff and not important. Dinka types are one of the non-admixed Nilo-Saharans when it comes to modern mixing of "non-SSA" ancestry.

The deep ancestry is something different, and what constitutes Eurasian is also a discussion in itself. Nilotes, similar to other peoples got different deep components. The issue is to give them a fitting model, after adequately describing, and relatively legitimizing the models. This is a prehistoric/stone age time window. The same models might not be a 50kya back migration, for example, but a northeast African component. How to conceptually talk about this is tricky, challenging and therefore fun. In modern terms, it is African ancestry. In ancient terms, it is Africa with different "divergent" histories that might have or might have not been related to migration in southwest Asia. But overall, that DNA is African, another type of African. I'm talking in the hypothetical, there is African ancestry in Dinkas that is from inner Africa too. Nilotics and other peoples of the region got some deep northeast Arican-specific, now extinct, stuff baked into their DNA.

 

[Som]

I've come across this site, it's very focused on minimizing the African ancestry which is predominant in Somalis (60%). If we were only 30% subsaharan we would look like northafricans.

 

[Nilo-Saharan]

You might find individuals with relatively recent admixture, but that is generally outlier stuff and not important. Dinka types are one of the non-admixed Nilo-Saharans when it comes to modern mixing of "non-SSA" ancestry.

The deep ancestry is something different, and what constitutes Eurasian is also a discussion in itself. Nilotes, similar to other peoples got different deep components. The issue is to give them a fitting model, after adequately describing, and relatively legitimizing the models. This is a prehistoric/stone age time window. The same models might not be a 50kya back migration, for example, but a northeast African component. How to conceptually talk about this is tricky, challenging and therefore fun. In modern terms, it is African ancestry. In ancient terms, it is Africa with different "divergent" histories that might have or might have not been related to migration in southwest Asia. But overall, that DNA is African, another type of African. I'm talking in the hypothetical, there is African ancestry in Dinkas that is from inner Africa too. Nilotics and other peoples of the region got some deep northeast Arican-specific, now extinct, stuff baked into their DNA.

Very interesting stuff, but if we really are speaking in Modern Terms this deep ancestry isn't really even acknowledged as Eurasian. It's only in technicality that we can draw connections to possibly Eurasian ancestry based on the fact that going back far enough in time we get different traces of ancestry to what we now consider African and Not African. What would be the use of identifying this deep ancestry if it ages so long ago.

 

[Nilo-Saharan]

I've come across this site, it's very focused on minimizing the African ancestry which is predominant in Somalis (60%). If we were only 30% subsaharan we would look like northafricans.

Yeah, I've come across some wild justifications and irrational conclusions on that website. This is exactly what I was thinking when reading the Ancestry model result.

 

[Som]

Yeah, I've come across some wild justifications and irrational conclusions on that website. This is exactly what I was thinking when reading the Ancestry model result.

And the truth is that out of the major horner groups (Oromo, Amhara, Tigray, afar and Somali) we somalis are the most African. I don't think any studies will ever show that we are anything less than 55% SSA (and in reality we are slightly over 60%).

 

[The alchemist]

Very interesting stuff, but if we really are speaking in Modern Terms this deep ancestry isn't really even acknowledged as Eurasian. It's only in technicality that we can draw connections to possibly Eurasian ancestry based on the fact that going back far enough in time we get different traces of ancestry to what we now consider African and Not African. What would be the use of identifying this deep ancestry if it ages so long ago.

We were talking about deep ancestry. Modern terms would say a person is approximately 100% Somali. Deep ancestry changes that conversation to something else because the conceptual framing speaks with a different vocabulary when we trace it to Stone Age.