G25 coordinates of 42 Inner Mongolians from China

[Tsakhur]

Sorry that I can't spend too much time on this because I've got a ton of other things to do before I release my Valentine's Day update on the 14th, but I'll just say the three ostensible subgroups mentioned in the paper look to blend into each other when I do runs incorporating various East Asian sources (Devil's Gate and Hanben; Devil's Gate and Miaozigou MN; Devil's Gate and Han_North_China). I don't really see any way to divide them easily; it's a big cline with no hard breaks. These three outliers have the most Han-like and the least Devil's Gate ancestry: 52661809219373, 32351809296156, and 67441809296932. This last sample is the one that could just be Han if he wanted to; he plots closest to Han North China and Manchu.

Most everyone else clusters closest to Mongolics already in G25. A few to Xibo.

Point noted. Here's the second datasheet btw just sent to me by Davidski, you can help annotate if you have time.

https://drive.google.com/file/d/1Riq...Be_lF9OSy/view

[altvred]

Mongol_IMAR=Mongol_Inner_Mongolia

 

https://drive.google.com/file/d/1Riq...IzeeDBe_lF9OSy

[Ebizur]

Regarding Mongols in the Northeastern region of Inner Mongolia and Northeastern China, would there also be some who are genetically identical to Proto-Mongols (who seem very Tungusic-like based on the MNG_East_N, Ulaanzukh, Slab Grave samples closest modern day distance) or Tungusics e.g. Oroqen, Nanai, Ulchi with negligible Western Eurasian affinity?

What do you mean by "Proto-Mongols"? The Borjigid circa their conquest of the Jin (金) Dynasty of Manchuria & North China?

These two individuals seem to be ethnic Mongols from Inner Mongolia. One of them and is closest to Oroqen, Daur in HarappaWorld but in the oracle can be modelled as a mix of Mongola+Buryat/Hezhen+Tuvinian/Xibo+Yakut/Daur+Yukaghir, etc. whereas the other one received Buryat as the closest match but possessed much less Western affinity than average and can be modelled as combination of mostly Oroqen+Yukaghir/Ket/Selkup or Tuvinian+Oroqen/Hezhen/Daur, etc in mixed mode sharing. These two results are most likely Northeastern region of Inner Mongolia and Northeastern China right? Could both individuals be a mix of Buryat+ Daur or some other Tungusic tribe?

Would you anticipate that the Uzemchin, Barga and Dariganga minorities in Dornod, Sukhbaatar in Far Eastern part of Outer Mongolia be Daur/Hezhen or Oroqen/Nanai-like or even autosomally resemble the Manchu/Northern Han with very little Western affinity compared to Khalkhas who have much higher Western Eurasian?

I would refrain from making any claims about minority tribes like the Uzemchin and Dariganga until we have actual samples from them because it is clear from currently available data that present-day speakers of Mongolic languages are quite diverse genetically.

The ADMIXTURE bar chart in Supplementary Figure 6 of Bai et al. (2018) suggests that the Buryat should be either the "purest" or the most drifted Mongol(ic)s among the following ethnic (sub)groups:

*Oirat from Bortala Mongol Autonomous Prefecture, Xinjiang, China
*Buryat from Evenk Autonomous Banner, Inner Mongolia, China
*Khalkha from Ulaanbaatar, Mongolia
*Sonid from Sonid Banner, Xilingol League, Inner Mongolia, China
*Abaga from Abaga Banner, Xilingol League, Inner Mongolia, China
*Horchin from Hinggan League, Inner Mongolia, China

However, four of seven Sonid also appear to have roughly the same amount of "Mongolic drift" as these Buryats from Evenk Autonomous Banner have on average.

At the other end of the spectrum, these Horchin from Hinggan League have been assigned rather little of the "Mongolic" component. The Horchin also stand out for having very little of the "Finnish" and "Indigenous American" components compared to the other subgroups of Mongols tested in this study. However, the Horchin have been assigned about the same amount of the "Italian/Spanish" (TSI/IBS) and basal (African) components as the other subgroups of the Mongols. The most important component of the Horchin in this analysis is the "Japanese" (JPT) one, but they also exhibit significant amounts of the "Dai/Vietnamese" (CDX/KHV) component, so I suppose that their "Mongolic" affinity probably has decreased through assimilation of some combination of Han Chinese, Manchus, and Koreans. (The Chinese in this analysis are assigned variable amounts of the "Dai/Vietnamese," "Japanese," and "Mongolic" components, with the "Japanese" and "Mongolic" components positively correlating with each other among the CHB from Beijing but not among the CHS from Hunan and Fujian.)

If you compare with the PCA graph in Supplementary Figure 4, you may see how the Buryat may have been assigned the greatest amount of a "Mongolic" component in the aforementioned ADMIXTURE analysis by virtue of greater genetic drift rather than by virtue of greater "Mongolic" purity. The Oirat and the Khalkha appear to have approximately equal amounts of "Mongolic" drift, but the Oirat are marked by greater levels of Western Eurasian (and especially Italian/Spanish-like as opposed to Finnish-like) admixture, dragging the Oirat downward on PC2. Three of ten Mongola_HGDP are plotted between the centroid of the Buryat and the centroid of the Khalkha; they appear to have relatively little Western admixture. The centroid of the Sonid appears to be close to the "southern" edge of the Khalkha cluster. The centroid of the Abaga appears to be close to the centroid of the Sonid, but with possibly a greater degree of variance introduced by recent East Asian admixture; however, this is difficult to judge because of the small sample size of the Sonid. After the Sonid and Abaga, the next closest Mongolic group to the main East Asian cluster on the PCA graph in Supplementary Figure 4 of Bai et al. (2018) are the Daur, who appear to be those of the HGDP. They form a very tight cluster that may exhibit some degree of "Tungusic" (Oroqen-Hezhen) drift as opposed to the "Mongolic" drift (along a vector pointing toward the Yakut cluster) exhibited by the other Mongolic-speaking populations. (Note that the Yakut were not included in the ADMIXTURE run in which the "Mongolic" drift was defined; it is possible and actually quite likely that the Yakut would have been assigned even more of this so-called "Mongolic" drift than the Buryat if a Yakut sample had been included in the ADMIXTURE run.) Finally, the Horchin from Hinggan League, Inner Mongolia form a loose cluster along with seven of ten Mongola_HGDP, eight of nine Xibo, about three of forty Oirat from Bortala Mongol Autonomous Prefecture, Xinjiang, about one of twenty-nine Abaga from Abaga Banner, Xilingol League, Inner Mongolia, and one of ten Hezhen. The Tu (Monguor) appear slightly to the "WSW" of the centroid of the Horchin cluster, probably reflecting their Tibetan/Qiangic admixture.

Bai 2018 Supplementary Figure 4.png
Bai 2018 Supplementary Figure 6 partial.png

[East-Asia]

The Turk, the Mongol and the Tungus are by far much later populations than the y chromosome hg N population.

The hg N-Z4762 population started its separations in China and the ancestors of hg N1c started to interact with mtDNA B5b4 which is also tied to Shandong in "Maternal genetic structure in ancient Shandong between 9500 and 1800 years ago".

The distribution of mtDNA B5b4 to Europe occured in parallel with the distribution of mtDNA B5b2a2 from the Beiqian settlement of the Dawenkou culture in the western direction through the territory of China after 5400 years ago (y chromosome hg N substituted for mtDNA B5b2 its initial male spouses).
Before this, the mtDNA B5b4 population started to interact with the mtDNA D5a3a1 population reported from the Hongshan culture (for Hongshan culture, Zhaobaogou culture where N1c related to Europeans was found had been a substratum), hence the appearance of mtDNA D5a3a1 in Urallic Mansi, Finns and ancient specimens from Hungary. The ancient specimens from Hungary contain both D5a3a1 from Hongshan culture and mtDNA B5b4.

Those branches of N1c that reached westernmost China in parallel with the B5b2a2 population from the Dawenkou culture interacted with Indo-Iranians and Tocharians in the bronze age less than 5400 years ago, but not 6000-7000 years ago. The populations in China who harbour hg N together with Indo-European uniparental markers R1a and R1b, but whose members speak non-Indo-European languages influenced by universal characteristics common for languages where there is a lot of N haplogroup speakers in China, mean that in these cases it is hg R1a and R1b who were assimilated by hg N speakers in China.

Only much later, the coming of Turk Mongol and Tungus caused the assimilation of the above mentioned speakers in Turk, Mongol and Tungus nationalities. Hg N did not belong to these Altaic languages in the first place.

There was no coming of mtDNA B5b4 into Yakutia_LN, Ymyyakhtakh and even Shamanke_EN, so there won't be modern N-L1026 haplogroups in those populations.





mtDNA D5a3a was reported from the Hongshan cuture, but not from the Yakutia_LN and not from Chukotko-Kamchatkans.

HONGSHAN culture WEST LIAO RIVER_MN BLSM27S F 0.021 n/a 0.02 (0.01-0.03) D5a3a1 n/a
HONGSHAN culture WEST LIAO RIVER_MN BLSM45 F 0.145 n/a 0.01 (0.01-0.02) D5a3a1 n/a

[Tsakhur]

What do you mean by "Proto-Mongols"? The Borjigid circa their conquest of the Jin (金) Dynasty of Manchuria & North China?

I would refrain from making any claims about minority tribes like the Uzemchin and Dariganga until we have actual samples from them because it is clear from currently available data that present-day speakers of Mongolic languages are quite diverse genetically.

The ADMIXTURE bar chart in Supplementary Figure 6 of Bai et al. (2018) suggests that the Buryat should be either the "purest" or the most drifted Mongol(ic)s among the following ethnic (sub)groups:

*Oirat from Bortala Mongol Autonomous Prefecture, Xinjiang, China
*Buryat from Evenk Autonomous Banner, Inner Mongolia, China
*Khalkha from Ulaanbaatar, Mongolia
*Sonid from Sonid Banner, Xilingol League, Inner Mongolia, China
*Abaga from Abaga Banner, Xilingol League, Inner Mongolia, China
*Horchin from Hinggan League, Inner Mongolia, China

However, four of seven Sonid also appear to have roughly the same amount of "Mongolic drift" as these Buryats from Evenk Autonomous Banner have on average.

At the other end of the spectrum, these Horchin from Hinggan League have been assigned rather little of the "Mongolic" component. The Horchin also stand out for having very little of the "Finnish" and "Indigenous American" components compared to the other subgroups of Mongols tested in this study. However, the Horchin have been assigned about the same amount of the "Italian/Spanish" (TSI/IBS) and basal (African) components as the other subgroups of the Mongols. The most important component of the Horchin in this analysis is the "Japanese" (JPT) one, but they also exhibit significant amounts of the "Dai/Vietnamese" (CDX/KHV) component, so I suppose that their "Mongolic" affinity probably has decreased through assimilation of some combination of Han Chinese, Manchus, and Koreans. (The Chinese in this analysis are assigned variable amounts of the "Dai/Vietnamese," "Japanese," and "Mongolic" components, with the "Japanese" and "Mongolic" components positively correlating with each other among the CHB from Beijing but not among the CHS from Hunan and Fujian.)

If you compare with the PCA graph in Supplementary Figure 4, you may see how the Buryat may have been assigned the greatest amount of a "Mongolic" component in the aforementioned ADMIXTURE analysis by virtue of greater genetic drift rather than by virtue of greater "Mongolic" purity. The Oirat and the Khalkha appear to have approximately equal amounts of "Mongolic" drift, but the Oirat are marked by greater levels of Western Eurasian (and especially Italian/Spanish-like as opposed to Finnish-like) admixture, dragging the Oirat downward on PC2. Three of ten Mongola_HGDP are plotted between the centroid of the Buryat and the centroid of the Khalkha; they appear to have relatively little Western admixture. The centroid of the Sonid appears to be close to the "southern" edge of the Khalkha cluster. The centroid of the Abaga appears to be close to the centroid of the Sonid, but with possibly a greater degree of variance introduced by recent East Asian admixture; however, this is difficult to judge because of the small sample size of the Sonid. After the Sonid and Abaga, the next closest Mongolic group to the main East Asian cluster on the PCA graph in Supplementary Figure 4 of Bai et al. (2018) are the Daur, who appear to be those of the HGDP. They form a very tight cluster that may exhibit some degree of "Tungusic" (Oroqen-Hezhen) drift as opposed to the "Mongolic" drift (along a vector pointing toward the Yakut cluster) exhibited by the other Mongolic-speaking populations. (Note that the Yakut were not included in the ADMIXTURE run in which the "Mongolic" drift was defined; it is possible and actually quite likely that the Yakut would have been assigned even more of this so-called "Mongolic" drift than the Buryat if a Yakut sample had been included in the ADMIXTURE run.) Finally, the Horchin from Hinggan League, Inner Mongolia form a loose cluster along with seven of ten Mongola_HGDP, eight of nine Xibo, about three of forty Oirat from Bortala Mongol Autonomous Prefecture, Xinjiang, about one of twenty-nine Abaga from Abaga Banner, Xilingol League, Inner Mongolia, and one of ten Hezhen. The Tu (Monguor) appear slightly to the "WSW" of the centroid of the Horchin cluster, probably reflecting their Tibetan/Qiangic admixture.

Bai 2018 Supplementary Figure 4.png
Bai 2018 Supplementary Figure 6 partial.png

Proto-Mongols as in Shiwei tribal confederation who later became Xianbei, Rouran, Khitan etc.

Can the Abaga, Sonid, Khorchin and other Mongolic samples from Bai et al 2018 be uploaded to David's G25?

Also are there any autosomal studies on the Uzemchin, Dariganga, Barga that can be added to G25 dataset?

[Huck Finn]

There was no coming of mtDNA B5b4 into Yakutia_LN, Ymyyakhtakh and even Shamanke_EN, so there won't be modern N-L1026 haplogroups in those populations.

N-L1026 > N-Y6058 based lineages in Koryaks and Chukchi might be based on some (later?) reindeer herding groups being a part of the Ymyakhtakh Proper stage, maybe after a language switch. That being said, I'd guess that the lineages leading into European N-L1026 were not involved in the Ymyakhtakh Proper, at least if we assume that they spoke Uralic already then, for Ymyakhtakh apparently did not speak Uralic. They however most probably were a part in the Pre Ymyakhtakh stages, apparently somewhere in the Trans Baikal area.

Most of paternal N's found in the Viking Era burials in Sweden, including the ship burials by the Estonian coast, are N-Y6058 too, just like Avars. However those Vikings are already then just slightly Siberian, whereas the Avars are still Siberians, if not just East Asians.

[Ryukendo]

What do you mean by "Proto-Mongols"? The Borjigid circa their conquest of the Jin (金) Dynasty of Manchuria & North China?

I would refrain from making any claims about minority tribes like the Uzemchin and Dariganga until we have actual samples from them because it is clear from currently available data that present-day speakers of Mongolic languages are quite diverse genetically.

The ADMIXTURE bar chart in Supplementary Figure 6 of Bai et al. (2018) suggests that the Buryat should be either the "purest" or the most drifted Mongol(ic)s among the following ethnic (sub)groups:

*Oirat from Bortala Mongol Autonomous Prefecture, Xinjiang, China
*Buryat from Evenk Autonomous Banner, Inner Mongolia, China
*Khalkha from Ulaanbaatar, Mongolia
*Sonid from Sonid Banner, Xilingol League, Inner Mongolia, China
*Abaga from Abaga Banner, Xilingol League, Inner Mongolia, China
*Horchin from Hinggan League, Inner Mongolia, China

However, four of seven Sonid also appear to have roughly the same amount of "Mongolic drift" as these Buryats from Evenk Autonomous Banner have on average.

At the other end of the spectrum, these Horchin from Hinggan League have been assigned rather little of the "Mongolic" component. The Horchin also stand out for having very little of the "Finnish" and "Indigenous American" components compared to the other subgroups of Mongols tested in this study. However, the Horchin have been assigned about the same amount of the "Italian/Spanish" (TSI/IBS) and basal (African) components as the other subgroups of the Mongols. The most important component of the Horchin in this analysis is the "Japanese" (JPT) one, but they also exhibit significant amounts of the "Dai/Vietnamese" (CDX/KHV) component, so I suppose that their "Mongolic" affinity probably has decreased through assimilation of some combination of Han Chinese, Manchus, and Koreans. (The Chinese in this analysis are assigned variable amounts of the "Dai/Vietnamese," "Japanese," and "Mongolic" components, with the "Japanese" and "Mongolic" components positively correlating with each other among the CHB from Beijing but not among the CHS from Hunan and Fujian.)

If you compare with the PCA graph in Supplementary Figure 4, you may see how the Buryat may have been assigned the greatest amount of a "Mongolic" component in the aforementioned ADMIXTURE analysis by virtue of greater genetic drift rather than by virtue of greater "Mongolic" purity. The Oirat and the Khalkha appear to have approximately equal amounts of "Mongolic" drift, but the Oirat are marked by greater levels of Western Eurasian (and especially Italian/Spanish-like as opposed to Finnish-like) admixture, dragging the Oirat downward on PC2. Three of ten Mongola_HGDP are plotted between the centroid of the Buryat and the centroid of the Khalkha; they appear to have relatively little Western admixture. The centroid of the Sonid appears to be close to the "southern" edge of the Khalkha cluster. The centroid of the Abaga appears to be close to the centroid of the Sonid, but with possibly a greater degree of variance introduced by recent East Asian admixture; however, this is difficult to judge because of the small sample size of the Sonid. After the Sonid and Abaga, the next closest Mongolic group to the main East Asian cluster on the PCA graph in Supplementary Figure 4 of Bai et al. (2018) are the Daur, who appear to be those of the HGDP. They form a very tight cluster that may exhibit some degree of "Tungusic" (Oroqen-Hezhen) drift as opposed to the "Mongolic" drift (along a vector pointing toward the Yakut cluster) exhibited by the other Mongolic-speaking populations. (Note that the Yakut were not included in the ADMIXTURE run in which the "Mongolic" drift was defined; it is possible and actually quite likely that the Yakut would have been assigned even more of this so-called "Mongolic" drift than the Buryat if a Yakut sample had been included in the ADMIXTURE run.) Finally, the Horchin from Hinggan League, Inner Mongolia form a loose cluster along with seven of ten Mongola_HGDP, eight of nine Xibo, about three of forty Oirat from Bortala Mongol Autonomous Prefecture, Xinjiang, about one of twenty-nine Abaga from Abaga Banner, Xilingol League, Inner Mongolia, and one of ten Hezhen. The Tu (Monguor) appear slightly to the "WSW" of the centroid of the Horchin cluster, probably reflecting their Tibetan/Qiangic admixture.

Bai 2018 Supplementary Figure 4.png
Bai 2018 Supplementary Figure 6 partial.png

Its truly quite remarkable that they found a component that peaks in both Buryat Mongols and Sonid Mongols because Sonids are in Inner Mongolia on the Khingan range north of Beijing while Buryats are found around Lake Baikal in Russia, so its not possible that this component is caused by a single isolated population with extremely high internal drift as these two populations probably separated quite a whiles back... You can also see differences in their admixture with other components in both PCAs and ADMIXTURE, with the "non-Mongolic" components in Sonid shifted towards East Asians and Buryats towards Siberians; this admixture with other ancestries is usually enough to get rid of drift-dominated components.

I think they might just have found a component that quite accurately maps onto the "proto-Mongolic" component... remarkable stuff!

[East-Asia]

N-L1026 > N-Y6058 based lineages in Koryaks and Chukchi might be based on some (later?) reindeer herding groups being a part of the Ymyakhtakh Proper stage, maybe after a language switch. That being said, I'd guess that the lineages leading into European N-L1026 were not involved in the Ymyakhtakh Proper, at least if we assume that they spoke Uralic already then, for Ymyakhtakh apparently did not speak Uralic. They however most probably were a part in the Pre Ymyakhtakh stages, apparently somewhere in the Trans Baikal area.

Most of paternal N's found in the Viking Era burials in Sweden, including the ship burials by the Estonian coast, are N-Y6058 too, just like Avars. However those Vikings are already then just slightly Siberian, whereas the Avars are still Siberians, if not just East Asians.

N-B202 got to Koryak from Northeast China, it hardly had had anything to do with Ymyakhtakh or reindeer herding groups. It was already a very mixed population and it did not have a certain local component found in Northeast China in the Paleolithic, but it already had at least some component maximal in Han Chinese.

[East-Asia]

Its truly quite remarkable that they found a component that peaks in both Buryat Mongols and Sonid Mongols because Sonids are in Inner Mongolia on the Khingan range north of Beijing while Buryats are found around Lake Baikal in Russia, so its not possible that this component is caused by a single isolated population with extremely high internal drift as these two populations probably separated quite a whiles back... You can also see differences in their admixture with other components in both PCAs and ADMIXTURE, with the "non-Mongolic" components in Sonid shifted towards East Asians and Buryats towards Siberians; this admixture with other ancestries is usually enough to get rid of drift-dominated components.

I think they might just have found a component that quite accurately maps onto the "proto-Mongolic" component... remarkable stuff!

It is possible that Buryat population contained a Turkic-speaking tribe before it shifted to Mongolic languages. It complicates the simplistic approach to the component that they have found.

EDIT: There is no Turkic populations in the Admixture above, only Mongolic and Japanese.
It is a very strange choice for a Mongolic institute in China, unless it is connected to the Japanese people.
By the way, the last time I saw the 100% Tokyo Japanese autosomal component in the article about ancient Uelgi population, 100% Tokyo Japanese autosomal component was scattered throughout the whole of Siberia and Central Asia, while in China, it was always mixed with Chinese Dai from Xishuangbanna component.

[Tsakhur]

It is possible that Buryat population contained a Turkic-speaking tribe before it shifted to Mongolic languages. It complicates the simplistic approach to the component that they have found.

Wouldn't Khalkha and Oirats also possessed this Turkic speaking component due to the notable Western admixture they have similar to Buryats?