Kazakh DNA

[vbnetkhio]

I tried plotting the data from Ilumäe et al. 2016 (in the second link), but I don't know if I did it right, because the clustering doesn't make much sense. Anyway, the heatmap became too long because there's too many different Y-STR haplotypes.

i'm still trying to fully understand how STRs work. I got the idea to divide them with their mutation rates. But I didn't find any study where something like this was done.

https://yhrd.org/pages/resources/mutation_rates
https://isogg.org/wiki/Mutation_rates#Resources

[Kaspias]

Hi, @Kaspias abi!

Could you post the Kirgiz DA86 sample? I wondered it so much

Hi Sora! Sure,

Distance to:	KGZ_Turk:DA86
0.03072051	Tubalar
0.04585718	Shor
0.04754714	Shor_Mountain
0.05315575	Shor_Khakassia
0.05786717	Khakass
0.06058749	Kazakh
0.06378500	Karakalpak
0.07969243	Nogai
0.08235853	Tatar_Siberian
0.08254243	Kirghiz
0.08355578	Kirghiz_China
0.09041583	Tatar_Siberian_Zabolotniye
0.09367360	Khakass_Kachins
0.09537103	Hazara
0.09637459	Uygur
0.09841909	Altaian
0.09996264	Kazakh_China
0.10210658	Althai_Kizhi
0.10279122	Hazara_Afghanistan
0.11684931	Bashkir
0.12253006	Yukagir_Forest
0.12882304	Uzbek
0.12985034	Mansi
0.13069383	Khanty
0.13366074	Ket

Distance to:	KGZ_Turk:DA86
0.03550427	KAZ_Pazyryk_IA
0.03780590	KAZ_Nomad_MA
0.04362356	MNG_KHI001
0.05903482	MNG_Altai_MLBA
0.06005560	MNG_Munkhkhairkhan_MBA_2
0.06198999	KAZ_Karakhanid
0.06766592	MNG_Early_Med_Turk
0.08264622	MNG_Xiongnu_Central_Asian
0.08970675	KAZ_Karluk
0.09132537	CHN_Shirenzigou_IA
0.09217692	MNG_UUS001
0.09404971	KAZ_Kimak
0.09651550	MNG_EIA_5
0.09728359	KAZ_Kipchak
0.09808974	Saka_Kazakh_steppe
0.09827559	RUS_Marinskaya_IA
0.09916759	RUS_Khaptsagai_IA
0.10196789	MNG_Sagly_EIA_4
0.10325966	RUS_Altai_IA
0.10341865	Scythian_Aldy_Bel_IA
0.10389521	RUS_Yakutia_LUP
0.11692137	MNG_Mongun_Taiga_LBA_1
0.11710770	Scythian_Zevakino_Chilikta_IA
0.11825634	MNG_Early_Med_Uigur
0.11932985	MNG_Mongol

Presence of the same profile in more than one tribe is interesting. Note the shift of these two through Huns, while the rest shifts towards Sogds. DA23 and DA86, in this sense, might be representing "Eastern Turkicness" and might have joined the migration later than the rest. Pazyryk comes as a possible spawn point for their grandfathers.

Or, depends on the date which I can't recall now, it could be simply a mix of mainstream Turkic and Mongol. DA23 was C if I recall.

[Kaspias]

i'm still trying to fully understand how STRs work. I got the idea to divide them with their mutation rates. But I didn't find any study where something like this was done.

https://yhrd.org/pages/resources/mutation_rates
https://isogg.org/wiki/Mutation_rates#Resources

This was on my mind lately. You need to increase the significance of the STRs that have a lesser mutation rate. You can do it by simply duplicating them. Take this spreadsheet as an example: https://drive.google.com/file/d/16LV...lCFutXqD2/edit

DYS388 has 0.007270 mutation rate, DYS391 has 0.003220, and DYS390 has 0.006220. Presume that we are using only 3 STRs. Add these three together and it will make 0.01671. Then divide it into three in order to find the base parameter. It makes 0.00557. Divide all the STR rates with this number. For example, 0.007270/0.00557=1.30xxxxx. Then, multiple it with 10(even could be better if multiple with 100, but requires more work if we proceed with manual way). You will find 13. And for the other two, you will have 6, and 11.

Now create a spreadsheet and repeat STRs according to their value. This will create a differentiation between the individuals as the diversification between fast mutated will have a lesser effect(4-5 rows) while slower ones will pull harder with like 10 rows.



There might be better solutions, but if not, an automated version of this would work smoothly.

[Leto]

Yes, in terms of autosomal.

50-50 for Altaians, 60% Mongol for Tuvans. However, I'm not really sure about the Medieval Turkic form of them.

You think the Turkic Altaians are half Mongolian?

[Kaspias]

You think the Turkic Altaians are half Mongolian?

Not sure honestly. This is what models indicate when using mainstream Turkic samples in the models -giving decent fit, btw- but I have reasons to think that the Altay region did not undergo the same mixes as Turks who migrated to the West experienced. I wouldn't even surprise if I see some strong Yeniseian absorb in the region rather than Mongolian.

[Leto]

Not sure honestly. This is what models indicate when using mainstream Turkic samples in the models -giving decent fit, btw- but I have reasons to think that the Altay region did not undergo the same mixes as Turks who migrated to the West experienced. I wouldn't even surprise if I see some strong Yeniseian absorb in the region rather than Mongolian.

Well, they are both culturally and linguistically isolated from the wider Turkic-speaking world. I thought they should be more Siberian than Central Asians given their location. Most of them are heavily R1a (30-50%), strangely enough.

[Komintasavalta]

In an ADMIXTURE run of modern Turkic samples in the 1240K+HO dataset, I got a pole for Altaians already at K=3. Maybe it's partially because there are only about 2,000 Tubalars and 1,000 Chelkans, so the Tubalar and Chelkan samples included some related individuals with high PI_HAT, even though I excluded one sample from each pair of samples with PI_HAT over .35.



At K=5, I got a new ghost component that the highest proportion in Karakalpak (41%), followed by Khakass (27%), Altaian (26%), Altaian_Chelkan (24%), and Bashkir (24%). Now there's also a new component that is maximal in Chuvashes. Tatar_Siberian and Tatar_Siberian_Zabolotniye both have 59% of the Chuvash component, so I don't know if Siberian Tatars are that close to the original Turks, because they seem to have additional Uralic-like ancestry. (Even though there's huge diversity among the different subgroups of Siberian Tatars, and this run didn't include include the populations Tatar_Tomsk.SG or Tatar_Irtysh_Barabinsk.SG.) Now the Yakut and Salar components that broke off at K=4 have merged again, so it causes Chinese Kazakhs and Kyrgyzes to plot closer to other Kazakhs and Kyrgyzes.