Classify these Turkic girls

[travv]

https://i.imgur.com/5yfbR4k.jpg

[Jaromir]

beauties, I am ready to move to Ankara.... wait - not those Turkic girls they are

[Hektor12]

beauties, I am ready to move to Ankara.... wait - not those Turkic girls they are

Unfortunately bro, we lost them.. Now Turkic girls from Ankara look like this=

https://www.cumhuriyet.com.tr/Archive/2020/12/14/1798490/kapak_101657.jpg


https://youtu.be/FdiRGSlsHXg

:picard2:

[travv]

beauties, I am ready to move to Ankara.... wait - not those Turkic girls they are

Yes not a single drop of Moorish blood about them. They are Chuvash girls from Tatarstan.

Girl on the left.


https://i.imgur.com/4kNo0pX.jpg

[Asten]

The numero 9 is good looking . 9/10

[Altaylı]

beauties, I am ready to move to Ankara.... wait - not those Turkic girls they are

Fun fact i am living ankara Turkey at the moment and there are a lot of tatar and nogai in ankara Turkey and they are genetically more mongoloid than chuvashes

Chuvash people are 20% mongoloid
Turkish tatars nogais are 30% mongoloid :cool:

Here is a turkish nogai result from my hometown:
https://scontent.fesb2-1.fna.fbcdn.net/v/t1.6435-9/109340910_125195889238577_5183497296515056670_n.pn g?_nc_cat=110&ccb=1-3&_nc_sid=8024bb&efg=eyJpIjoidCJ9&_nc_ohc=q_jnxXrgY5oAX8ocLfr&_nc_oc=AQlN2jsTwoZ_D6P2o7IEFk5BM24WWnjNIYf9oqbYdoz PRZZRV1n2p2HN82U4Fi1pjqw&_nc_ht=scontent.fesb2-1.fna&oh=1bfd4f5d5cd008e2ab31ed64a80f6a37&oe=60B53B4D
30%mongoloid :cool:

User Edgü is a Tatar too and he is from Ankara, Turkey

[Komintasavalta]

Fun fact i am living ankara Turkey at the moment and there are a lot of tatar and nogai in ankara Turkey and they are genetically more mongoloid than chuvashes

Chuvash people are 20% mongoloid
Turkish tatars nogais are 30% mongoloid :cool:

I think Chuvashes are more Mongoloid than Tatars or at least equally Mongoloid. But if you use a model with only an East Asian or a Southern Siberian Mongoloid source, it puts Chuvashes at a disadvantage, because they have more Uralic-like central-north Siberian ancestry than Tatars. (By putting Chuvashes at a disadvantage, I mean that it makes them look less Mongoloid than they actually are.)

For example the ternary diagram below displays a K=3 ADMIXTURE run of Turkic and Uralic samples in the 1240K+HO dataset. Chuvashes are clearly closer to Nganasans than Tatars, even though Tatars have more of the Salar component. Also the Chuvash samples in 1240K+HO are less Mongoloid on average than the Chuvash samples in G25.

https://i.ibb.co/By5x1ph/turk-ural-ternary-admixture.png

Also here's a K=5 ADMIXTURE run. The clustering is based on an FST matrix generated with SmartPCA. Now Kazan Tatars have about 7% of the Salar component and 10% of the Nganasan component, but Chuvashes have about 14% of the Nganasan component and 4% of the Salar component. So it might actually be a close tie, because Salars are more Mongoloid than Nganasans. However Mishars make up about 45% of Volga Tatars according to Wikipedia, so if you calculate the Mongoloid percentage of Tatars by taking the average of Mishars and Kazan Tatars, then Tatars clearly lose to Chuvashes. The blue Finnic and orange Turkish components are of course also part Mongoloid, and I could've increased the Mongoloid percentage of Turko-Uralics by including more IEs in this run.

https://i.ibb.co/HqjQvMH/turk-uralic-admixture-dendro.png

[Kaspias]

.


He talks about Steppe Crimean Tatars(20-30% EE) and Nogais(35-45% EE). The East Asian source of Chuvash is not mainstream Turkic despite having a great deal of East Eurasian, therefore he might be right, too. Anyway, you should not be using Salar for East Asian proxy. They are a mix of Oghuz(~25) and Han(~75) therefore should not be representative of any population. However, that's a good way of investigation. You can try to use Levanluhta or Chalmny Varre and BA Khovsgol and we will have a clear separation of Uralic and Turkic. What would you say?

[Komintasavalta]

He talks about Steppe Crimean Tatars(20-30% EE) and Nogais(35-45% EE). The East Asian source of Chuvash is not mainstream Turkic despite having a great deal of East Eurasian, therefore he might be right, too. Anyway, you should not be using Salar for East Asian proxy. They are a mix of Oghuz(~25) and Han(~75) therefore should not be representative of any population. However, that's a good way of investigation. You can try to use Levanluhta or Chalmny Varre and BA Khovsgol and we will have a clear separation of Uralic and Turkic. What would you say?

But I thought Edgü was a Volga Tatar / larping as Volga Tatar...

It was unsupervised ADMIXTURE so I didn't manually pick any population as a Mongoloid proxy. There were only 8 Salar samples in the run, so the components that were maximal in Salars probably didn't even depend that much on Salars.

Anyway, here's another unsupervised run where I removed Salars and I added Khövsgöl (it sounds 100x more Türkö-Ürälö-Möngölic with the umlauts).

Now even though the top component is maximal in Khövsgöl, a lot of weight in the top component seems to be given to Tubalars, Shors, and Chelkans. They were also at one extreme of a cline on PC3 and PC4 in a SmartPCA run using the same samples. Even many Bashkir samples have more than 50% of the top component. So the top component isn't really a Khövsgöl component even though it's maximal in Khövsgöl.

https://i.ibb.co/8cQ3L3W/a.jpg

You can see the Tubalar-Chelkan-Shor cline in the second image below:

https://i.ibb.co/Fhtrcqb/turkural4-i-1.jpg
https://i.ibb.co/VDdh4g8/turkural4-i-3.jpg

[Altaylı]

But I thought Edgü was a Volga Tatar / larping as Volga Tatar...


no Edgü is a real Ankara Tatar probably he is Çöl Tatarı(Crimean_Tatar_Steppe)
Yes there are Tatars Nogais in Turkey and their population isn`t small

Distance to: Crimean_Tatar_Steppe
8.11380305 Lipka_Tatar
15.08643099 Uzbek
15.34141780 Turkmen_Afghanistan
16.06060709 Bashkir_South
16.26522671 Bashkir_Central
16.60028012 Uzbeki
17.88100389 Turkmen_Uzbekistan
18.03205479 Tatar_Siberia
18.48025974 Nogai_Stavropol
19.38624254 Bashkir_North
19.73826233 Nogai_Astrakhan
19.85630378 Karakalpak
20.03978792 Tatar_Kazan
21.51643325 Turk_Trakya
21.64653552 Nogai
21.83349262 Turk_Deliorman
22.04075770 Turkmen_Iran
23.41294087 Crimean_Tatar_Mountain
23.89151104 Tajik_Mountain
23.94644232 Uyghur
24.30244844 Besermyan_Udmurtia
24.52976763 Turk_Northwest
24.69033212 Tatar_Mishar
25.78689590 Chuvash
25.81228777 Turk_Makedonya

[Komintasavalta]

Here's a K=5 ADMIXTURE model where the Khövsgöl component is differentiated from the Tubalar-Chelkan-Shor component. Now Kazan Tatars have about about 8% of the Khövsgöl component and 8% of the Nganasan component. Chuvashes have 6% of the Khövsgöl component and 12% of the Nganasan component. So this time it looks like Chuvashes win.

Nogai_Stavropol and Nogai_Astrakhan both have over 50% of the last two components. They're the same samples from Jeong et al. 2019 as in G25. But the Nogais in Turkey are probably more mixed? (And yeah I know that Noğay (Steppe Crimean Tatars) are different from Nogai.)

https://i.ibb.co/KN06N4L/a.jpg

[Kaspias]

.

Edgü is Noğay, a separate tribe under Steppe Tatars. Tatars are settled in Turkey into masses and the people who preserved their Tatar ethnicity so far are supposedly still pure Tatar. Referencing Stavropol and Astrakhan Nogais, Steppe Tatars would get something like 25-30 in total.

The 3rd admixture that peaks in Chelkans, what is it stands for? Did you take Khövsgöl as a late-comer(from further East) to the region and the Chelkans as Turkic or both stand for Turkic, Chelkans just being an alternative?

Inspiring from you, I tried to plot the differentiation of East Asian admixtures across Eurasia.

#1 I run by using these references:


West_Eurasian:Baltic_LVA_HG
West_Eurasian:WHG
West_Eurasian:LUX_Loschbour
West_Eurasian:TUR_Barcin_N
West_Eurasian:Levant_Natufian
West_Eurasian:GEO_CHG
West_Eurasian:TKM_Gonur1
West_Eurasian:HUN_Avar_Szolad
West_Eurasian:DEU_MA
MNG_Khovsgol_BA
CHN_Amur_River_Xianbei_IA
CHN_Upper_Yellow_River_IA
FIN_Levanluhta_IA


https://i.ibb.co/TmtvxHN/xx.png

#2 Ignored West Eurasian and plotted East Eurasian differentiation.

https://i.ibb.co/v3BPWKj/pca2.png

#3 Equalized all scores to 100% and run hierarchical clustering in Chord's method to ignore score differences, but focus on variability.

https://i.ibb.co/3h8mMXf/clustering.png

And percentages:

https://i.ibb.co/MGyMt3J/percentage.png

[Komintasavalta]

The 3rd admixture that peaks in Chelkans, what is it stands for? Did you take Khövsgöl as a late-comer(from further East) to the region and the Chelkans as Turkic or both stand for Turkic, Chelkans just being an alternative?

For pairs of samples with PI_HAT over .35, I excluded the sample that sorted alphabetically first. At first I tried a threshold of .25, but it removed all Nganasan samples. But now some of the samples with the highest PI_HAT included Tubalars and Chelkans, so maybe that's why a component for Altaians was differentiated early on.

Even some of the samples with the very highest PI_HAT included Shors and Tubalars:

$ x=turkural4;plink --bfile $x --genome --out $x
[...]
$ awk '{print$10,$2,$4}' $x.genome|sort -rn|head -n1000|awk 'NR==FNR{a[$1]=$3;next}{print$1,a[$2]":"$2,a[$3]":"$3}' v44.3_HO_public.ind -|head -n10
0.6255 Tofalar:Vgut8 Tofalar:Vgut12
0.6143 Shor_Khakassia:KHS-035 Shor_Khakassia:KHS-036
0.6091 Tubalar:Tuba23 Tubalar:Tuba24
0.5874 Khakass_Kachin:Khs-493 Khakass_Kachin:Khs-513
0.5763 Tubalar:ALT-116 Tubalar:Tuba2
0.4722 Tofalar:Vgut11 Tofalar:Vgut13
0.4600 Nganasan:Tuebingen23 Nganasan:Tuebingen124
0.4600 Nganasan:Tuebingen19 Nganasan:Tuebingen116
0.4421 Nganasan:Tuebingen23 Nganasan:Tuebingen121
0.4363 Tofalar:Vgut1 Tofalar:Vgut4

... But maybe my theory is not correct, because even after I did a new PCA using a PI_HAT threshold of .25, there was still an Altaian-Shor cline on PC3 and PC4. And because all Nganasans had been removed by the PI_HAT filtering, there was no Nganasan component in ADMIXTURE, and an Altaian-Shor component already became differentiated at K=4.

[Komintasavalta]

#2 Ignored West Eurasian and plotted East Eurasian differentiation.

https://i.ibb.co/v3BPWKj/pca2.png

I hadn't thought of doing a biplot before where you show the loadings of the source populations. Here's my version that includes population labels and convex hulls:

https://i.ibb.co/bPnxnGB/1.jpg


library(tidyverse)
library(ggrepel)
library(ggforce)

t=read.table(text="Target CHN_Amur_River_Xianbei_lA CHN_Upper_Yellow_River_IA FIN_Levanluhta_lA MNG_Khovsgol_BA West_Eurasian
Gagauz 0 0 0 0.4 99.6
Bulgarian 0 0 1 0 99
Hungarian 0 0 2.4 0 97.9
Azeri_Dagestan 0 2.5 0.4 1 96.1
Azeri_Turkey 2.1 1.8 0 0 96.1
Azeri 1.2 3.8 0.1 0 94.9
Turkish_Deliorman 3.5 1.2 1 0 94.3
Balkar 0 0.2 0 8 91.8
Karachay 0 0.3 0 8.5 91.2
Turkish_Central 3.7 2.9 3.3 0 90.1
Kumyk 2.6 0 5.6 1.8 90
Turkish_North 7.7 1.9 0.9 1.1 88.4
Turkish_Rumeli 0 0.9 5.4 5.9 87.8
Turkish_Northwest 6.5 2.7 1.6 3.5 85.7
Turkish_South 8.6 2.4 3.9 0 85.1
Turkish_Southwest 7.5 2.7 6.5 0 83.3
Finnish 0 0 22.6 0 77.4
Tatar_Lipka 14.6 6.7 4.9 4 69.8
Karelian 0 0 31.4 0 68.6
Tatar_Mishar 6.4 2.6 22.5 0 68.5
Finnish_East 0 0 32.4 0 67.6
Turkmen 11.2 7.3 7.2 9 65.3
Tatar_Crimean_steppe 16.5 9.6 7.4 2.7 63.8
Turkmen_Uzbekistan 13.9 7.8 10.3 5.8 62.2
Uzbek 5.6 18.2 5.5 16.3 54.4
Tatar_Kazan 4.6 2.7 40.6 1.5 50.6
Hazara 11.9 21.6 1 17.5 48
Hazara_Afghanistan 2 20.4 3.9 27.2 46.5
Uygur 5 26.4 3.8 18.7 46.1
Karakalpak 28.2 12.9 4.7 13.6 40.6
Kazakh 29.9 15.3 4.1 16.6 34.1
Saami_Kola 0.7 0 66.5 0 32.8
Kirghiz 34.2 18 0.8 17.9 29.1
Bashkir 4.5 2.5 43.6 25.1 24.3
Tubalar 0 0 4.2 71.8 24
Tatar_Siberian 6.8 2.5 42.5 30.4 17.8
Mongolian 50.3 16.8 0 15.4 17.5
Buryat 64.8 3.9 0 14.3 7
Altaian 32.7 5 3.9 42 16.4
Shor_Mountain 0 0 12.5 71.6 15.9
Chuvash 0.6 0 84.8 0 14.6
Udmurt 0 0 84.1 1.4 14.5
Shor_Khakassia 0 0 14.6 71.4 14
Althai_Kizhi 30.3 4 6.1 46.7 12.9
Saami 0.5 0 88.5 2.4 8.9
Yukagir_Forest 29.3 0 64 0 6.7
Tibetan_Gangcha 12.2 73.3 0 7.9 6.6
Mari 5.4 0 89.5 0 5.1
Tibetan_Shannan 3 80.6 0 11.6 4.8
Khamnegan 66.7 8.5 7 13.7 4.1
Tatar_Siberian_Zabolotniye 0 0 53.5 43 3.5
Tuvinian 44.4 0 8.9 43.5 3.2
Yugur 0 91.2 2.1 3.5 3.2
Yakut 73.1 0 9.8 15 2.4
Mongola 19.7 74.3 2.3 1.8 1.9
Tibetan_Yunnan 1.4 90.8 0 6.2 1.6
Han_Shanxi 0 98.6 0.4 0 1
Nivkh 69.6 26.3 0 3.9 0.2
Selkup 0 0 29.5 70.5 0
Nenets 17.2 0 32.9 49.9 0
Ket 0 0 51.8 48.2 0
Todzin 46.9 0 7.1 46 0
Mansi 0 0 58 42 0
Yukagir_Tundra 80.2 0 9.5 10.3 0
Ulchi 76.8 18.2 0 5 0
Nganassan 86.7 0 13.3 0 0
Dolgan 76 0 24 0 0
Orogen 66.9 30.7 2.4 0 0
Evenk 92.9 0 7.1 0 0
Even 95.2 0 4.8 0 0
Han_Hubei 0 100 0 0 0",header=T,row.names=1,check.names=F,sep=" ")
t=t/100

p=prcomp(t)
pct=paste0(colnames(p$x)," (",sprintf("%.1f",p$sdev/sum(p$sdev)*100),"%)")
p2=as.data.frame(p$x)
p2$k=as.factor(cutree(hclust(dist(t)),k=16))

load=p$rotation

i=1
x=sym(j("PC",i))
y=sym(j("PC",i+1))
mult=min(max(p2[,i])/max(load[,i]),max(p2[,i+1])/max(load[,i+1]))

ggplot(p2,aes(!!x,!!y))+
geom_segment(data=load,aes(x=0,y=0,xend=mult*!!x,y end=mult*!!y),arrow=arrow(length=unit(.3,"lines")),color="gray60",size=.4)+
annotate("text",x=(mult*load[,i]),y=(mult*load[,i+1]),label=row.names(load),size=2.3,vjust=ifelse(load[,i+1]>0,-.5,1.4))+
ggforce::geom_mark_hull(aes(color=as.factor(k),fil l=as.factor(k)),concavity=100,radius=unit(.2,"cm"),expand=unit(.2,"cm"),alpha=.2,size=.13)+
# geom_polygon(data=p2%>%group_by(k)%>%slice(chull(!!x,!!y)),alpha=.2,aes(color=k,fill=k ),size=.3)+
geom_point(aes(color=k),size=.6)+
# geom_text(aes(label=row.names(t),color=k),size=2.3 ,vjust=-.6)+
ggrepel::geom_text_repel(aes(label=row.names(p2),c olor=as.factor(k)),size=2.2,max.overlaps=Inf,box.p adding=.05,segment.size=.3,force=5,min.segment.len gth=.1)+
labs(x=pct[i],y=pct[i+1])+
coord_fixed()+
scale_x_continuous(breaks=seq(-1,1,.2),expand=expansion(mult=.08))+
scale_y_continuous(breaks=seq(-1,1,.2),expand=expansion(mult=.08))+
scale_color_manual(values=hcl(head(seq(15,375,leng th=length(unique(p2$k))+1),-1),120,50))+
theme(
aspect.ratio=1,
axis.text=element_text(color="black",size=6),
axis.ticks=element_line(size=.3,color="gray60"),
axis.ticks.length=unit(-.1,"cm"),
axis.text.x=element_text(margin=margin(.2,0,0,0,"cm")),
axis.text.y=element_text(angle=90,vjust=1,hjust=.5 ,margin=margin(0,.2,0,0,"cm")),
axis.title=element_text(color="black",size=8),
legend.position="none",
panel.background=element_rect(fill="white"),
panel.border=element_rect(color="gray60",fill=NA,size=.4),
panel.grid=element_blank()
)

ggsave(paste0(i,".png"),width=6,height=6)

[Komintasavalta]

Edgü is Noğay, a separate tribe under Steppe Tatars.

Russian Wikipedia says that all Steppe Crimean Tatars are called Nogai or Noğay (https://ru.wikipedia.org/wiki/Степные_крымские_татары):


Steppe Crimean Tatars (Nogai; self-name - Crimean Tatar noğaylar, Nogailar [1] [2] [3]) is one of the three sub-ethnic groups of Crimean Tatars [4] [5] [6] [7], which formed and lived in the steppe parts of Crimea Crimean Cat. Çöllü Qırım) until 1944, when all [4] Crimean Tatars were deported from Crimea. In the formation of the steppe Crimean Tatars, the Western Kypchaks (in Russian historiography - the Polovtsy) played an important role. Similar in origin and culture to the North Caucasian and Stavropol Nogais [8] [4] [9] [10].

Wikipedia says that one variety of the Crimean Tatar language is this: "Northern (Steppe Crimean Tatar/Nogay Steppe) (should not be confused with Nogai people of the Northern Caucasus and the Lower Volga)" (https://en.wikipedia.org/wiki/List_of_Turkic_languages).

Turks are so confusing...

I still can't tell if this girl is suposed to be Steppe Crimean Tatar or the other type of Nogai: https://gramho.com/explore-hashtag/nogaikhan.

[travv]

Unfortunately bro, we lost them.. Now Turkic girls from Ankara look like this=

https://www.cumhuriyet.com.tr/Archive/2020/12/14/1798490/kapak_101657.jpg

:picard2:

Looks Latina/Hispanic. This is what VURians girls would look like if Chad Spaniards spread their blood in VUR.:scared:

[placebo]

both can pass as turkish (i'm not talking about nogay or tatar)

[lei.talk]

is karina toktamysheva's name typically turkic (https://en.wikipedia.org/wiki/Turkic_peoples) or chuvash (https://en.wikipedia.org/wiki/Chuvash_people)? :icon1:

is her appearance typical for either? :confused:

or of "tartarstan (https://en.wikipedia.org/wiki/Almetyevsky_District)"? :confused3:

was she selected based on her cooking and singing? :fponder:


:dunno:

[Jana]

Woggy, pass as honorary Cypriots.

[Jaromir]

^^
oh no! it hurts travv

[Jana]

^^
oh no! it hurts travv

The painful truth :p

[travv]

is karina toktamysheva's name typically turkic (https://en.wikipedia.org/wiki/Turkic_peoples) or chuvash (https://en.wikipedia.org/wiki/Chuvash_people)? :icon1:

is her appearance typical for either? :confused:

or of "tartarstan (https://en.wikipedia.org/wiki/Almetyevsky_District)"? :confused3:

was she selected based on her cooking and singing? :fponder:


:dunno:

Tuktamysheva is Tatar surname. Many Chuvash and Mari people living in Tatarstan have Tatar surnames. I think Karina Tuktamysheva not enough Mongoloid to be typical Chuvash.

[travv]

^^
oh no! it hurts travv

Not really. I think Stearsolina means they don’t look mainstream Europeans as well as Cypriots.
I can see it only as compliment.