Stupid Turk study. Nogais are 60-80% East Asian Mongoloid , not over 70% Caucasoid west Eurasian

[Flashball]

In reality, Nogais are genetically closer to Central European subgroups (such as Hungarians, Czechs, Slovaks, etc.) than any other Turkic subgroups, which on average typically have more Mongoloid admixture than (modern) Nogais do, which are obviously way more "Caucasian" (I'm using this term for BOTH the racial sense and the geographical sense, since they the Nogai tribes already come from the Caucasus, unironically speaking) in West-Eurasian DNA. It also works like this in a similar way ethnic Hungarians were not direct descendants of Turks, not always because every individual of them look so Mongoloid.

Just a reminder, phenotypes and physical traits don't always show enough well to genetics, and linguistically speaking, mother-tongues, though they may still in fact be related with each other.

If I could recall correctly, this would mostly likely be one reason that they evolved through the process of borealization which gives them such dependent traits such as stronger epicanthic folds and/or slanting eyes. Therefore, Nogais would have 90-100% West-Eurasian (Caucasoid) DNA, leaving only 5-10% East-Eurasian ("the so-called Mongoloid") DNA, according to several genetic studies.



Yes, the phenotype corresponds well to the European and non-European genes present in the individual and population. I think it is absurd to say that a Nogais with more "Siberian" traits has no relation to another Nogais with more "European" traits. : both expressed different genes from Europe and Siberia differently.

To summarize, we can say: their traits correspond to the expressed genes (phenotype) of all of their genes (genotype), one is linked to the other, except that phenotype does not express all the genes of the genotype.

On the Global 25 (better than Gedmatch), the European contribution is detectable (barcin + yamnaya), like the “Asian” (Han) and “Siberian” (Nganassan) contribution.

So, it's not surprising to see different traits within this population with their genome and their genotype.

https://image.noelshack.com/fichiers/2021/38/7/1632632184-bp4c4j8z.jpg

[Dick]

I'm confused. When was there ever a smart Turk study?

[Komintasavalta]

Nogais are genetically closer to Central European subgroups (such as Hungarians, Czechs, Slovaks, etc.) than any other Turkic subgroups, which on average typically have more Mongoloid admixture than (modern) Nogais do ...

Therefore, Nogais would have 90-100% West-Eurasian (Caucasoid) DNA, leaving only 5-10% East-Eurasian ("the so-called Mongoloid") DNA, according to several genetic studies.

Gagauzes are probably the least Mongoloid Turkic people. Some older Nogai samples from Karachay-Cherkessia are only about 30% Mongoloid. But in Jeong et al. 2019, there's new Nogai samples from Astrakhan and Stavropol which can be over 50% Mongoloid depending on what metric you use.

I merged samples for populations that have been viewed as possibly Ural-Altaic from 1240K+HO (https://reich.hms.harvard.edu/allen-ancient-dna-resource-aadr-downloadable-genotypes-present-day-and-ancient-dna-data), the Busby dataset (https://data.mendeley.com/datasets/ckz9mtgrjj/3), and Tambets et al. 2018 (https://evolbio.ut.ee/Tambets2018/). The list below shows the percentage of the Mongoloid component in a K=2 ADMIXTURE run of the samples.

For each pair of samples with PI_HAT over .3, I deleted one of the samples. There were relatively few East Asian samples, so the Mongoloid component ended up being heavily influenced by Siberian and Far Northeastern populations. In reality, I think even Salars might have higher Mongoloid ancestry than Nganasans.

0 (0-0) Tambets_Gagauzes (n=1)
0 (0-2) Hungarian (n=20)
1 (0-2) Gagauz (n=9)
1 (0-5) Estonian (n=10)
1 (0-5) Tambets_Estonians (n=36)
4 (4-4) Busby_Finnish (n=2)
5 (1-7) Tambets_Ingrian (n=6)
5 (3-7) Tambets_Karels (n=2)
5 (1-10) Tambets_Finns (n=19)
5 (3-8) Finnish (n=8)
7 (2-9) Mordovian (n=22)
7 (5-8) Busby_Mordvin (n=15)
7 (6-8) Tambets_Vepsas (n=2)
7 (3-10) Karelian (n=15)
7 (1-9) Kumyk (n=8)
7 (0-17) Turkish (n=50)
8 (5-11) Busby_Turkish (n=19)
8 (3-12) Busby_Turkish_Kayseri (n=11)
8 (1-15) Busby_Turkish_Istanbul (n=10)
8 (6-11) Azeri (n=17)
9 (7-10) Veps (n=10)
9 (0-18) Busby_Turkish_Aydin (n=13)
9 (7-11) Balkar (n=10)
10 (7-12) Karachai (n=11)
10 (5-14) Busby_Kumyk (n=14)
15 (13-17) Turkish_Balikesir (n=5)
16 (14-19) Tatar_Mishar (n=10)
18 (11-23) Tambets_Saami_Kola (n=8)
19 (12-25) Tambets_Tatars (n=2)
20 (13-24) Chuvash (n=6)
20 (18-23) Tatar_Kazan (n=13)
21 (16-28) Busby_Nogai (n=8)
21 (21-21) Finland_Saami_Modern.SG (n=1)
23 (4-27) Busby_Chuvash (n=17)
23 (17-34) Busby_Turkmen_Turkmenistan (n=10)
24 (23-26) Besermyan (n=6)
26 (20-33) Nogai_Karachay_Cherkessia (n=8)
27 (22-30) Tambets_Saami_SWE (n=13)
27 (24-31) Udmurt (n=10)
28 (28-28) Tambets_Maris (n=1)
29 (29-29) Mari.SG (n=1)
33 (2-60) Tambets_Mansi (n=23)
34 (18-45) Bashkir (n=53)
35 (30-46) Turkmen (n=6)
39 (18-55) Uzbek (n=27)
47 (37-55) Tatar_Siberian (n=19)
47 (42-52) Yukagir_Forest (n=5)
48 (35-58) Nogai_Stavropol (n=10)
52 (50-54) Tatar_Siberian_Zabolotniye (n=5)
54 (51-57) Nogai_Astrakhan (n=6)
54 (46-58) Mansi (n=8)
55 (48-62) Karakalpak (n=14)
57 (44-64) Altaian_Chelkan (n=8)
58 (56-64) Tambets_Khanty (n=11)
62 (56-67) Tubalar (n=23)
63 (37-79) Busby_Selkup (n=10)
63 (58-68) Kazakh (n=18)
64 (62-65) Shor_Mountain (n=7)
64 (61-66) Shor_Khakassia (n=5)
65 (61-68) Kyrgyz_Tajikistan (n=4)
68 (51-78) Khakass (n=15)
68 (59-74) Kyrgyz_China (n=13)
68 (55-73) Busby_Kyrgyz (n=16)
70 (32-99) Even (n=9)
70 (60-91) Selkup (n=18)
72 (66-81) Kazakh_China (n=8)
73 (62-77) Busby_Altaian (n=11)
75 (72-79) Altaian (n=24)
75 (71-77) Khakass_Kachin (n=7)
75 (48-98) Busby_Dolgan (n=5)
80 (73-85) Enets (n=3)
81 (80-85) Salar (n=8)
81 (66-100) Busby_Yakut (n=5)
82 (79-86) Kalmyk (n=10)
82 (51-100) Busby_Nganasan (n=7)
84 (64-100) Evenk_FarEast (n=5)
85 (81-87) Tuvinian (n=11)
85 (83-89) Busby_Tuvan (n=13)
86 (37-94) Tofalar (n=9)
86 (78-90) Mongol (n=34)
88 (79-92) Buryat (n=37)
89 (86-91) Busby_Buryat (n=17)
89 (81-96) Dolgan (n=4)
90 (88-92) Todzin (n=3)
95 (92-97) Xibo (n=7)
96 (94-98) Daur (n=9)
97 (93-100) Yukagir_Tundra (n=13)
97 (93-100) Evenk_Transbaikal (n=8)
99 (97-100) Japanese (n=29)
99 (97-100) Korean (n=6)
100 (98-100) Nganasan (n=23)
100 (100-100) Nanai (n=10)
100 (100-100) Negidal (n=3)

The Nogai samples from Busby included 9 samples that were duplicates of the Nogai_Karachay_Cherkessia samples from 1240K+HO, so I removed them. The rest also look similar:

16 Busby_Nogai:nogay6
17 Busby_Nogai:nogay7
19 Busby_Nogai:nogay9
20 Nogai_Karachay_Cherkessia:nogay42
21 Busby_Nogai:nogay12
21 Busby_Nogai:nogay8
22 Nogai_Karachay_Cherkessia:nogay45
22 Nogai_Karachay_Cherkessia:nogay44
23 Busby_Nogai:nogay5
25 Nogai_Karachay_Cherkessia:nogay46
25 Nogai_Karachay_Cherkessia:nogay33
25 Busby_Nogai:nogay11
28 Busby_Nogai:nogay13
29 Nogai_Karachay_Cherkessia:nogay35
29 Nogai_Karachay_Cherkessia:nogay24
33 Nogai_Karachay_Cherkessia:nogay25
35 Nogai_Stavropol:NOG-125
38 Nogai_Stavropol:NOG-129
44 Nogai_Stavropol:NOG-220
45 Nogai_Stavropol:NOG-133
50 Nogai_Stavropol:NOG-217
50 Nogai_Stavropol:NOG-224
51 Nogai_Stavropol:NOG-357
51 Nogai_Astrakhan:NOG-18
51 Nogai_Stavropol:NOG-127
52 Nogai_Astrakhan:NOG-42
53 Nogai_Astrakhan:NOG-17
54 Nogai_Astrakhan:NOG-15
54 Nogai_Astrakhan:NOG-19
55 Nogai_Stavropol:NOG-263
57 Nogai_Astrakhan:NOG-21
58 Nogai_Stavropol:NOG-130

Here's the K=3 run visualized as a ternary diagram and the K=4 run as a square version of a ternary diagram. In the K=4 run, you can see the combined proportion of the Siberian and East Asian components from how right a sample plots. Out of the four Nogai populations, Nogai_Astrakhan plots the furthest right and Busby_Nogai plots the furthest left. (Nogai can also refer to Steppe Crimean Tatars, but they are not included here.)

https://i.ibb.co/HFY9tyR/uralaltaicmerge3.png
https://i.ibb.co/SQk9yRq/uralaltaicmerge4.png

[Leto]

I merged samples for populations that have been viewed as possibly Ural-Altaic from 1240K+HO (https://reich.hms.harvard.edu/allen-ancient-dna-resource-aadr-downloadable-genotypes-present-day-and-ancient-dna-data), the Busby dataset (https://data.mendeley.com/datasets/ckz9mtgrjj/3), and Tambets et al. 2018 (https://evolbio.ut.ee/Tambets2018/). The list below shows the percentage of the Mongoloid component in a K=2 ADMIXTURE run of the samples.

Can you merge multiple samples into one for Gedmatch/Admixture Studio? I was thinking adding Khanty and Mansi averages to Dodecad K12b would be a great idea.

[Komintasavalta]

Can you merge multiple samples into one for Gedmatch/Admixture Studio? I was thinking adding Khanty and Mansi averages to Dodecad K12b would be a great idea.

I don't know how to use either of them. What format does the data need to be in? Or can you just use a PLINK dataset with Admixture Studio? There's Khanty samples in Tambets_et_al_2018_610k and 1M here: https://evolbio.ut.ee/Tambets2018/.

I can't use Admixture Studio, because I don't have a Windows VM and I'm not gonna install one. I looked at Gedmatch, but it seems to allow just uploading one file at a time, and you have to enter a bunch of information like the name of the sample. Is there some way you can use it to quickly run a calculator for a batch of samples?

Or is there some way you can run admixture calculators like Dodecad K12b from the command line? I don't want to use any GUI or web UI.

[Leto]

I don't know how to use either of them. What format does the data need to be in? Or can you just use a PLINK dataset with Admixture Studio? There's Khanty samples in Tambets_et_al_2018_1M here: https://evolbio.ut.ee/Tambets2018/.

I can't use Admixture Studio, because I don't have a Windows VM and I'm not gonna install one. I looked at Gedmatch, but it seems to allow just uploading one file at a time, and you have to enter a bunch of information like the name of the sample. Is there some way you can use it to quickly run a calculator for a batch of samples?
Um, I only know of Gedmatch itself and the Studio. Some people know how to merge multiple samples into one. Vbnet, Lemmin do. Adding Khanty from Tambets would be very nice.

[Komintasavalta]

Um, I only know of Gedmatch itself and the Studio. Some people know how to merge multiple samples into one. Vbnet, Lemmin do. Adding Khanty from Tambets would be very nice.

Apparently you can also use DIYDodecad, but I think it only has Linux and Windows binaries and no source code.

http://dodecad.blogspot.com/2011/08/do-it-yourself-dodecad-v-20.html
https://anthrogenica.com/showthread.php?4879-Tutorial-%96-How-to-use-DIY-calculator-files-and-oracle-data-sheets
https://isogg.org/wiki/Admixture_analyses#DIYDodecad

Where can I download the calculator files? The Google Drive links are not public but you need to request access.

[Leto]

Apparently you can also use DIYDodecad, but I think it only has Linux and Windows binaries and no source code.

http://dodecad.blogspot.com/2011/08/do-it-yourself-dodecad-v-20.html
https://anthrogenica.com/showthread.php?4879-Tutorial-%96-How-to-use-DIY-calculator-files-and-oracle-data-sheets
https://isogg.org/wiki/Admixture_analyses#DIYDodecad

Where can I download the calculator files? The Google Drive links are not public but you need to request access.
I'm afraid I cannot answer that question :confused: