New Dating for Ancient European admixture

[real expert]

Target Region WHG EHG AHG SE_WHG SE_EHG SE_AHG p-value

IronGates Central Europe 0.638 0.224 0.138 0.109 0.043 0.072 0.926236

Maybe if it's repeated often enough, this will get through, along with an understanding of how this might confound results.

"Previous analysis has suggested that early Anatolian farmers can be modeled as a mixture of Anatolian hunter-gatherers (AHG) and Iran Neolithic. The AHG were in turn were a mixture of WHG and Levant Neolithic groups (23). We confirmed this model using previously suggested sources and outgroups using Model-B (Supplementary Table S5.3). The early Anatolian farmers mixed with Levant Neolithic groups to form the genetic ancestry of Anatolian farmers, that migrated to the west to Europe and in the east to form the Chalcolithic groups of Seh Gabi andHajji Firuz. Using Model-D and Anatolian farmers and Iran Neolithic groups as sources, weobtained a good fit for Iran Seh Gabi individuals. We find that Hajji Firuz individuals were better modeled as a mix of Iran Seh Gabi and Anatolian farmer-related ancestry using Model-C (24). Wereport details of all models and estimated admixture proportions in Supplementary Table S5.3.".........

Weren't the Natufians/ Levant Neo Anatolian/Pinarbasi like themselves just plus ANA?

 

[Angela]

Angela: No I recognized what you were hinting at. The Models for the 3 Italian populations in Table S4.1 are "uninformative" as you stated because they were rejected as specifications to model the Target populations (North Italian, Tuscan, and West Sicilian) in Adler, which I think is what Reich uses a lot at the Harvard Lab?. Thanks for posting that link " admixturemap". As I noted, I don't have this paper, thus never read it. Posting the admixture llink allows a very easy comparison of all the models Chintalapti et al 2022 estimated in Table S4.1 to the ones Hellenthal et al 2014 ran. It is clear the Hellenthal et al 2014 models for admixture were not replicated by the Chintalapati, Patterson and Moorjani 2022 paper (See Table S4.1, pp 47-48). On the other hand, the Results from Table S5.4 do show the models fit quite well for these ancient populations, even though there may be some differences as in admixture proportions in this study compared to others, as you noted.

The inelegance wasn't yours. I was referring to the post calling the Hellenthal series of papers "crap". Inelegant and even vulgar, but I think my posts on those three threads discussing those papers made my feelings plain; I most definitely wasn't a fan.

 

[Palermo Trapani]

The inelegance wasn't yours. I was referring to the post calling the Hellenthal series of papers "crap". Inelegant and even vulgar, but I think my posts on those three threads discussing those papers made my feelings plain; I most definitely wasn't a fan.

Ok, got it thanks. I wasn't sure if my post was off base. As I said I wasn't around here for the discussion on the Hellenthal paper back in 2015. I do appreciate you linking it as I went through the thread and there were some concerns about it that you clearly articulated, as did others.

 

[Angela]

Will take a look at this more closely, because this sounds like a pretty big deal. One reason why Y-DNA analysis has been so popular is because of dates. If you can date autosomal mixtures, then it'll change the entire ballgame.

Well, it would be if it worked better with situations where, for instance, there has been continuous gene flow from the same general area, or founder effect etc. It's not that it can't be used. It's just always going to favor the last pulse of admixture with the older date being at the midpoint of the continuous pulses. It's still going to take very careful analysis, and always doing qpAdm for every step.

It does seem to work very well for the much more ancient admixtures where there was one major admixture event and it involved very discrete groups.

 

[Jovialis]

Whatever happened to the Lazaridis 2018 pre-print? How does this factor into everything? Also, so the case is closed on the ethnogenesis on Anatolia_N? I thought it was still up for debate... Considering how much Dzudzuana is supposed to be Anatolian-like.

Frankly, I think Anatolian_N is a continuation of Dzudzuana, and early Levantines are a mix of Ancestrial North African plus Dzudzuana, while CHG is ANE-enriched Dzudzuna.

What ever happened to this study????

 

[torzio]

Whatever happened to the Lazaridis 2018 pre-print? How does this factor into everything? Also, so the case is closed on the ethnogenesis on Anatolia_N? I thought it was still up for debate... Considering how much Dzudzuana is supposed to be Anatolian-like.

Frankly, I think Anatolian_N is a continuation of Dzudzuana, and early Levantines are a mix of Ancestrial North African plus Dzudzuana, while CHG is ANE-enriched Dzudzuna.

What ever happened to this study????

maybe Anatolian_N represented the thracian tribe , Bithynia

 

[Angela]

Weren't the Natufians/ Levant Neo Anatolian/Pinarbasi like themselves just plus ANA?

That's certainly what some of the papers indicated.

Patterson doesn't address that and just talks about the gene flow from Levant Neo into AHG and then again into Anatolian farmers before they go to Europe.

That directly contradicts, it seems to me, the paper which said there was no gene flow accompanying the adoption of agriculture by the AHG group. Unless, like the flow of Iran Neo and WHG, the first pulse happened before the adoption of agriculture and the last one after it. Strange when you think of it that unless that paper was wrong, the pulses from Levant Neo took place before and after the adoption of agriculture, but not with it.

 

[real expert]

That's certainly what some of the papers indicated.

Patterson doesn't address that and just talks about the gene flow from Levant Neo into AHG and then again into Anatolian farmers before they go to Europe.

That directly contradicts, it seems to me, the paper which said there was no gene flow accompanying the adoption of agriculture by the AHG group. Unless, like the flow of Iran Neo and WHG, the first pulse happened before the adoption of agriculture and the last one after it. Strange when you think of it that unless that paper was wrong, the pulses from Levant Neo took place before and after the adoption of agriculture, but not with it.

Interesting stuff. By the way, I read on AG & Eurogenes that Levant_N has ~10% ANA on top of what Barcin_N has, which is probably ~6%. There are some posters on AG that seem to be versed in using qpAdm. According to them, all the Anatolian farmers (except Boncuklu) show some level of ANA, and this becomes clear by using ZlatyKun, who is pretty much equally related to all Eurasians except those with ANA ancestry.

 

[Angela]

Interesting stuff. By the way, I read on AG & Eurogenes that Levant_N has ~10% ANA on top of what Barcin_N has, which is probably ~6%. There are some posters on AG that seem to be versed in using qpAdm. According to them, all the Anatolian farmers (except Boncuklu) show some level of ANA, and this becomes clear by using ZlatyKun, who is pretty much equally related to all Eurasians except those with ANA ancestry.

Well, given how abysmally wrong they've been on many occasions with percentages of this, that or the other thing, I'm not going to take those numbers to the bank. The first example which comes to mind is the incredibly high percentages they supposedly saw of "steppe" in Pakistanis and even Northern Indians.

It's not enough to say you're using qpAdm; you have to know how to use it, and most amateurs don't.

I also think that a lot of times the conclusions to which the scientists at someplace like the Reich Lab come to are based on the results of dozens and dozens of samples which haven't yet made their way into papers.