Genetic Makeup of Kurds and Coastal Western Anatolian Turks Revisited

a subsequential short article to "Genetic Impact of Iranic and Turkic Expansions from Central to West Asia"

Please read the previous post first if you have not already done so.

Due to some concerns of a daemon unceasingly whispering to the ears of the skeptic regarding the usage of mixed data types (shotgun and capture) in a qpAdm analysis, we have decided to test the solidity of our previous work by using one type of data only (capture). It should be emphasized that using mixed data types does not make an analysis invalid, but there are convincing arguments for using one type of data being more reliable.

Since being limited with one type of data restricts the number of eligible ancient samples, we will have to compromise some standards that we followed in our previous article, like the exclusive usage of samples that have coverage higher than 1. Instead all the available relevant samples will be included this time as long as they are not labeled as “QUESTIONABLE” in the AADR files due to ancient DNA damage. Also it will not be possible in this analysis to use some important samples (like Yaz II and Karluk samples) since there are no available capture versions of them. While there are eligible alternatives for the mediaeval Turkic migrants that are genetically quite close to the previously used Karluk samples, like those from mediaeval Karakaba, Eastern Kazakhstan, it does not seem to be that we have good alternatives for the Yaz II sample, DA382. Therefore we will try different Iranic sources (Otrar_LIA, Ksirov_Kushan, Sarmatian_IA) for Kurds, knowing that probably none of them resembling the original Western Iranic speaking Central Asian migrants of the Early Iron Age closely enough. Still, these results will provide us some reliable clue to decide whether our previous results are solid or not. Also we will not be able to use Andronovo samples in our distal models, that are, the models that try to reflect an earlier layer of an ancestral breakdown; rather we will be using samples from the Srubnaya Culture, which is generally regarded as the Proto-Iranic culture, that are almost indistinguishable from Andronovo samples.

While all the models below have passing values, the greener boxes indicate better statistical fits. The used 11 reference populations or right pops (can be seen not in the table but in the outputs) are fixed in all the models with three exceptions, all being the ones with ancient targets (line 2, 14 and 15). The list of the IDs of the samples that have been used as well as the qpAdm outputs of each model, which include all the technical details needed to be sure of their validity, can be found in the link provided below the table.

For the qpAdm outputs, please click here.

Although somewhat different, the results are complementary to the conclusions of our previous article.

If you have read this article and are satisfied with it, please consider supporting my work on Patreon.

I am an archaeologist who mainly works on ancient Indo-Iranic cultures. I have published many articles and videos on archaeogenetics and ancient religious traditions in English and Turkish, some of which have been translated into Kurdish and Persian. Eliminating the fallacies surrounding these areas of research (that are generally born out of ideological sentimentality or romanticism regarding one's own ancestors) and providing reliable information in a clear, systematic and analytical way are my main objectives. I would be more than glad if you choose to support my work!

May the xᵛarənah be with you!