GENETICS There Was No Natufian Back Migration

[Xabeshi]

 

[Arabsiyawi]

mhhhh nice distance

 

[Fez]

Model on different regions

 

[The alchemist]

Natufian is a RELATIVELY good proxy. No one here with even an ounce of competence would say Natufian are the cause of our ancestry. It is the other way around. We share ancestry because Natufian is substantially northeast African HG:

Basically Natufian has extra 5% Southwest Asian HG-like (the Iran Neolithic has exessive Basal ancestry that is African-like, and that is what defines much of that affinity, not real prehistoric Iranian ancestry) ancestry which is irrelevant to us. We only capture the ~40-50% we need which Natufian has more than.

This is all proxy. I'd rather use Natufian and Sudanese7 instead of simulated that cut genetics, all to narrow distance (still, those shortening of gaps gives us insight). I accept adequate simuations where we have constraints and discrete/isolated values (ANA, spesifically); also use these methodologies to inspect and measure proportional sub-structure (that is what these things are best for, honestly). Natufian is already substantially similar to our non-AEA side so it is sufficient. I am not trying to narrow to 0% distance because often times, when we use Natufian it is to shape basal layers, not mix in with recent ones. The simulated contrived coordinate sure gives good fit, still it is not delineated enough where we can't tell if potential "Basal Eurasian" is not taken from the AEA side. Natufian functions very, in fact, it might be better because it does not blur the lines. It provides worse fit but we have better delineation. And it is consistent across the board. The first coordinate posted had substantial Dinka-like enrichment.

I would not even use mixed populations that are closer in proximity. If the Skhirat-Rouazi had closer distance, I would still prefer Natufian, no question. That is why I am reluctant to use the Old-Kingdom samples; they have too much Levantine food producing layers. I might as well use Arabians at that point. It gets the same job done.

As I said, I rarely use simulated samples that are chopped up like that exept one for the sake of modelling alone, but here is one dubbed "Basa Eurasian" which consists of Zlaty Kun, removing Bacho-Kiro and "Western Eurasian." In reality this is just closer to Africican drift rahter than "Basal Eurasian" but such hypothetical might capture something that we see fall in Iran Neolithic:

As you see again, higher basal drift for this simulated coordinate with the difference being Natufian having ~5% higher southwest Asian.

I'm not sure why this individual used Mota. Mota is not a proxy for our ancestry actually. It only has part AEA ancestry with the rest being divergent, especially when it comes to the non-AEA ancestry. Very strange choice. According to Lipson et al 2022, Mota is a basal lineage that has existed for at least since the OOA ancestry and recieved a Ghost population admixture of 30% post 50 kya. I don't know why people include this in models unless they are checking for actual Ethiopian HG ancestry, or modelling HG diversity toward southern African clines.

The "Ethio-Somali" component mentioned in the paper from 2014 was talking about the backmigration of southwest Asians some 23kya that was terminal Levantine Aurignicians who were basically related to Anatolian HG and Dzudzuana (these populations are separate related lineages). It does not describe the entire Cushitic ancestry because we're talking about a deeper population period.

Please don't post Reddit threads about these topics here as if has something we have not exhausted to a greater degree. Every time I see something written on that forum it is always incorrect. Much of what is written there is directly copped from this forum, yet they still did not take notes, since they deliver their own clusterfuckery.