(((Davidski)))

[FinalFlash]

I don't have a problem with your conclusions. Others will argue things that are more false, but argue them better. You're arguing all this because of your "Fuck Israel" crusade. In that case, in your cluster world, where do all these Jewish groups belong? Which group is your Levantine reference? Are Armenians a tightly packed cluster?

You were right about the Egyptian Jews btw. They are legit Levantines and likely the best representatives of the original Jews. However, I won't give you the Iraqi and Syrian Jews because the former has excess Meso-Iranic input while the latter seems to have elevated steppe and Med affinities. That might be because they are largely derived from Spharidic Jews, no?

[vbnetkhio]

Right, it was Mingle who told us he had contacted Davidski and that the man didn't want to upload any more samples 'for technical reasons'. That would imply he wanted to say the other ones aren't of decent enough quality. If he really did select them at random, it makes his excuse even more invalid and insulting.

he definitely did that.

these are the official labels of Montenegrin samples from the study. Yes, they are labeled a bit randomly, there is no Montenegrin2 or 3.
1,4,5,6,7 are in g25.

Montenegro1 Montenegro1 0 0 2 -9
Montenegro4 Montenegro4 0 0 2 -9
Montenegro5 Montenegro5 0 0 2 -9
Montenegro6 Montenegro6 0 0 1 -9
Montenegro7 Montenegro7 0 0 2 -9
Montenegro9 Montenegro9 0 0 1 -9
Montenegro10 Montenegro10 0 0 1 -9
Montenegro11 Montenegro11 0 0 1 -9
Montenegro14 Montenegro14 0 0 2 -9
Montenegro15 Montenegro15 0 0 1 -9
Montenegro17 Montenegro17 0 0 1 -9
Montenegro19 Montenegro19 0 0 1 -9
Montenegro20 Montenegro20 0 0 1 -9
Montenegro21 Montenegro21 0 0 2 -9

[Pine]

You were right about the Egyptian Jews btw. They are legit Levantines and likely the best representatives of the original Jews. However, I won't give you the Iraqi and Syrian Jews because the former has excess Meso-Iranic input while the latter seems to have elevated steppe and Med affinities. That might be because they are largely derived from Spharidic Jews, no?

I wasn't right about Egyptian Jews (assuming you mean Egyptian Karaites) in the way you think I was right. They do plot in the Levant with most models, as that's what you're solely focused on - so I provided you the counter example. However, their plotting so close is likely due to offsetting admixture, which is why I brought up the example of a European admixed Yemeni Arab who'd plot in the Levant, despite having 0 Levantine ancestry. This is why plotting doesn't tell as much of the story as you think it does. Samaritans likely have a similar issue. Samaritans are an outlier among Levantines. They're the only Levantine group which is closer to MBA Canaanites than to later samples. For whatever reason this ended up the case, they're not a good proxy for ancient Jews either. Other Jews separated from them over 700 years before exile. As for Syrian Jews, they're not a monolith. Some plot exactly in the Levant and it's not utterly clear why yet. Same goes for Iraqi Jews - some plot in the Levant. Iraqi Jews barely have any detectable Sephardic admixture. Syrian Jews have more Sephardic admixture, but have a lot of substructure. There are likely at least 3 subgroups of Syrian Jews. If Roman Era Judean samples changed in the direction if IA III Beirut, then Romaniotes may be the closest to pre-exilic Judeans. Libyan Jews and the "native" subset of Tunisian Jews are also a contender. The big mystery are Crimean Karaites, who score very high WANA on 23andMe, but otherwise, there is very little data on them.

[FinalFlash]

I wasn't right about Egyptian Jews (assuming you mean Egyptian Karaites) in the way you think I was right. They do plot in the Levant with most models, as that's what you're solely focused on - so I provided you the counter example. However, their plotting so close is likely due to offsetting admixture, which is why I brought up the example of a European admixed Yemeni Arab who'd plot in the Levant, despite having 0 Levantine ancestry. This is why plotting doesn't tell as much of the story as you think it does. Samaritans likely have a similar issue. Samaritans are an outlier among Levantines. They're the only Levantine group which is closer to MBA Canaanites than to later samples. For whatever reason this ended up the case, they're not a good proxy for ancient Jews either. Other Jews separated from them over 700 years before exile. As for Syrian Jews, they're not a monolith. Some plot exactly in the Levant and it's not utterly clear why yet. Same goes for Iraqi Jews - some plot in the Levant. Iraqi Jews barely have any detectable Sephardic admixture. Syrian Jews have more Sephardic admixture, but have a lot of substructure. There are likely at least 3 subgroups of Syrian Jews. If Roman Era Judean samples changed in the direction if IA III Beirut, then Romaniotes may be the closest to pre-exilic Judeans. Libyan Jews and the "native" subset of Tunisian Jews are also a contender. The big mystery are Crimean Karaites, who score very high WANA on 23andMe, but otherwise, there is very little data on them.

Interesting. So which group is responsible for the offsetting admixture in Karaites that has them plotting in the Levant? I was surprised to see how close they were to Levantines relatives to other Jews. Syrian Jews did seem mixed, yes. Probably a mix of Spharidic and Iraqi Jews. As for Iraqi Jews, I believe they do have excess non-Levantine input when compared to other Levantines. As for Samaritans, if they are not the best proxy for the ancient Jews, then who are in all honesty? Which modern-day group would likely be the best candidate for being the most Jewish genetically?

[Pine]

Interesting. So which group is responsible for the offsetting admixture in Karaites that has them plotting in the Levant? I was surprised to see how close they were to Levantines relatives to other Jews. Syrian Jews did seem mixed, yes. Probably a mix of Spharidic and Iraqi Jews. As for Iraqi Jews, I believe they do have excess non-Levantine input when compared to other Levantines. As for Samaritans, if they are not the best proxy for the ancient Jews, then who are in all honesty? Which modern-day group would likely be the best candidate for being the most Jewish genetically?

Egyptian Karaites may have a combination of Levantine (primarily), Aegean, North African, and Mesopotamian, where the Mesopotamian and North African "cancel out" the shift from the Aegean. Syrian Jews aren't a mix of Sephardim and and Iraqi Jews. You're forgetting their main component, which is from pre-Sephardic and from pre-Iraqi Jews. The closest modern group would be either Lebanese Christians, Palestinian Christians, or some subsets of of Syrian and Iraqi Jews. There are individual Iraqi Jews which plot in the Levant and it's unclear why. Much less is known about Mizrachi Jews,as they're not studied nearly as much. If, however, Roman Era Judeans were similar to IA III Beirut and shift further in the same direction, then possibly Romaniotes or Cypriots are the best modern proxy. This question isn't going to be solved till we get appropriate Roman Era Jewish samples. One also has to consider that there was likely substructure among Judeans.

[Deusex99]

The Serbian academic study had like 15 samples, but the fool only opted for five of them he percieved to be of high SNP quality. The trouble is the average of those five is unrepresentative of the whole study and when Mingle and vbnetkhio told him about it he still didn't want to make any changes to the average.

Indeed: "technical issue" is a good excuse used by (((Davidski))) to replace academic samples with non academic ones: like all Portuguese and half of Spaniards on G25 are from non academic sources as well.

He also eliminated the Central Greeks from Behar et al 2013 for no logical reason.

[JerryS.]

so why would GEDmatch Dodecad V3, GEDmatch MDLP K23b, and GEDmatch Eurogenes K13 provide difference oracle populations for the same data?

[digital_noise]

so why would GEDmatch Dodecad V3, GEDmatch MDLP K23b, and GEDmatch Eurogenes K13 provide difference oracle populations for the same data?

They use different sample populations for each category.

[Chris596]

There might be too much hype around G25, but I can really model my ancestry with it, pretty accurately.

Should you not trust anyone on this basis? I don't think so.

Davidski looked into my DNA file and he told me practically all the important information about my ancestry and has created a model which works very well almost everywhere. My profile description is basically all of my ancestry. He deserves the money for his work, but I think no one was obliged to pay him and use his services.

[JerryS.]

They use different sample populations for each category.

yes, but why? what is wrong with one model's choice of samples that another model chooses to use a different set?