Genomic Insights into the Formation of Human Populations in East Asia[

[Ryukendo]

I12976; 1437-1304 BC; Kulala Ula, barrow 2, Ulaankhus sum, Bayan-Ulgiii aimag, Mongolia; Mongolia_LBA_3_MongunTaiga; N1a2-F1360>pre-L666


L666 level: Y3048/FGC10789+ A>G (2G); FGC10768/Y3064+ A>T (4T); Y3067/FGC10811+ A>G (6G); F4309+ A>G (4G); Y3032/FGC10784+ C>T (8T); FGC10801/Y3051+ G>A (1A); L666/F1008- G>A (5G); Y3046/FGC10770/F864- C>T (1C); F1007- C>G (1C); FGC10803/Y3079- C>T (8C); F2017- C>A (6C); F2199- C>G (2C); Y54167- C>A (5C); FT182515- C>T (4C); Y3073/FGC10795- C>G (4C); Y3080/FGC10796- C>G (1C); Y3047/FGC10769- G>A (5G); FGC10767/Y3058- G>A (1G); Y3066/FGC10816- G>A (2G); CTS10075- G>A (4G); Y3077- G>A (1G); F3163- G>A (7G); FGC10823/Y3057- T>A (3T); CTS5442- T>C (2T); F3201/FGC10793/Y3038- T>C (2T); F3290- T>C (2T)


I12976 belongs to an earlier split than Neolithic Baikal samples (DA245, DA248, DA250, DA251, DA362 and DA359).

Hey Pribislav, you might be doing this anyway, but it would be super interesting if you could call the WLR samples later as well. I'm sure ybmpark, Ebizur, and others could chime in on the extent--if any--of uniparental matches between WLR and present-day Koreans and Japanese. My own suspicion is that O2b might be a very Japonic-specific marker with a later uniparental flux responsible for the arrival of Koreanic in the BA-IA, but just that this contribution of continental non-O2b y-haps to modern Koreans could have been quite small.

[ybmpark]

This recent paper on samples from Mongolia was a collaboration between a lab from Seoul University and the Max Planck lab, so it's not like this is an issue of a complete absence of resources and departments dedicated to Ancient DNA study in Korea...

I am sure that the "collaboration" is for Koreans to supply the samples they obtained from Mongolians and for Max Planck to do most of the analysis.
In terms of the general technological aptitude, Koreans are a top notch; they do excell in whatever money is poured into like in fusion research and military technology etc.
But they are also hampered by a sort of political correctness that has plagued them for 70 years.
A top mathematical and scientific talent is socially encouraged to go into engineering because a small and poor country like Korea cannot afford to do pure science.
That might have been a wise choice even 50 years ago but they still have that "small country complex" even after Korea has become rich.

I had communicated with top researchers in Korea and China when this research was at its infancy.
Koreans always complained about not getting enough support and they sometimes lied that their research had some medical implication to get funds.
Chinese who were technologically behind Korea in almost every field at the time (and were much poorer than Korea) had no such problem.
After 20 years we are seeing the result of it.

[altvred]

I13173; 1879-1694 BC; Ulan goviin uzuur barrow 2, Munkhkhairkhan sum, Khovd aimag, Mongolia; Munkhkhairkhan_MBA; N2-Y6503/B482>MF52704>Y147658>(pre?)-Y148364


Y147658 level: Y147658+ A>G (7G); Y147640+ A>T (1T); Y148024+ G>T (2T); Y148885- G>A (1G)

Y148364 level: Y148158+ T>C (1C); Y148670+ C>T (3T); Y148364- A>G (1A)



I13768; 1210-1047 BC; Muunit uul, Ulziit sum, Bayankhongor aimag, Mongolia; Mongolia_LBA; N2-Y6503/B482>MF52704>Y147658>Y148364


Y147658 level: Y147658+ A>G (13G); Y147562+ T>C (4C); Y147640+ A>T (2T); Y148024+ G>T (1T); Y148375+ G>A (1A); Y148832+ G>C (4C); Y148885- G>A (5G)

Y148364 level: Y148158+ T>C (2C); Y148670+ C>T (28T); Y148364+ A>G (3G)


NOTE:

I13173 is the same sample as UAA001 from Jeong et al. 2020, and I13768 is the same sample as I14462/I14463 from Sirak et al. 2020.

The read coverage and quality for the Y148670 call (POS:15,015,469) look solid, DP is referring to the number of reads at a given position.

 

[ybmpark]

In this paper, the authors use West Liao River (WLR) samples to model Koreans and Japanese with good fits, but like you said the uniparentals are a poor match, is your sense that a population high in O2b would be more informative? Where do you think this population would be?

This paper's qpGraph posits a NE Asian farmer group that contributed to WLR and to Taiwan Iron Age Austronesians independently of Yellow River NE Asian farmers further west. Maybe farmers of the Bohai region?

My impression is that O1b2(formerly O2b) predates even Japonic in the Korean peninsula.
O1b2 decreases in frequency as you go north and west to such an extent PyongAn province(north west Korea) probably has more C2 than O1b2.
But 2000 years ago O1b2 might have been a marker for Japonic speakers, especially O1b2-47z if Vovin's theory is right.
But Koreans may have brought some of them too especially L682.
Whatever we speculate now may look stupid once the ancient remains in Korea and Manchuria are analyzed.

This paper also casts doubt on Vovin's "Out of Southern China" theory.
We just don't see much connection between Japanese and Thai or Southern Chinese at all.

This paper has something like 92 percent Koreans(~WLR) and 8 percent Jomon for Japanese even though in both paternal and maternal HG the indigenous Japanese portion is much higher.
Maybe the same thing happened in Korea. The indigenous population with O1b2 endured the massive influx of both Japonic and Koreanic speakers and remained a very significant force until today.
It also follows that the "Koreans" who peopled Japan were not the direct WLR population but were Japonic speakers with some indigenous Korean(but not Koreanic speaking) uni-parental influence.

We just need ancient Korean samples analyzed. All these can be wild speculations.

P.S. I have a model for this. At least the persistence of paternal lineage. This can explain some of the European puzzles where indigenous WHG y haplogroups pop up later.

[talljimmy0]

Is there some particular reason as to why we don't have any aDNA from the Korean peninsula? Obviously, North Korea isn't really an option, but you'd figure that we would have at least a couple of samples from South Korea at this point.

Hate to assume the worst, but I always thought it was some politically-driven nonsense that forbids collaboration with labs in Japan and China...

Their notion of Koreaness is based on “blood purity”. Clearly incompatible with concepts in modern genomic research. In contrast, Chinese and Japanese have more awareness of their mixed origins. I’ve seen a few Koreans have a mini identity crisis when they get unexpected results from commercial dna tests. Imagine the reaction if it was a society-wide revelation.

[Pribislav]

I1202; 996-830 BC; Pospelovo 1, Primorsky Krai, Russia; Yankovsky_IA ; N1a2-F1360* (xL666)


F1360 level: F1360+ T>C (8C); FGC10797/Y3071+ A>T (4T); Y3049/FGC10810+ T>C (1C); CTS7235+ T>C (4C-1T)

L666 level: L666/F1008- G>A (2G); Y5298/FGC10766- G>A (1G); Y3048/FGC10789- A>G (2A); FGC10768/Y3064- A>T (2A); Y3067/FGC10811- A>G (1A); Y3046/FGC10770/F864- C>T (2C); F1007- C>G (1C); Y3032/FGC10784- C>T (2C); F2017- C>A (1C); Y54167- C>A (3C); Y3047/FGC10769- G>A (4G); FGC10801/Y3051- G>A (1G); FGC10767/Y3058- G>A (4G); FGC10804/Y3065- G>C (1G); CTS10075- G>A (1G); F3163- G>A (6G); FGC10823/Y3057- T>A (1T); Y3078/FGC10775- T>G (2T)


This one seems to be a genuine F1360*. He will be submitted to YFull, so maybe he'll form new subclade with Chinese F1360* that is already in the YFull tree.

[Ryukendo]

My impression is that O1b2(formerly O2b) predates even Japonic in the Korean peninsula.
O1b2 decreases in frequency as you go north and west to such an extent PyongAn province(north west Korea) probably has more C2 than O1b2.
But 2000 years ago O1b2 might have been a marker for Japonic speakers, especially O1b2-47z if Vovin's theory is right.
But Koreans may have brought some of them too especially L682.
Whatever we speculate now may look stupid once the ancient remains in Korea and Manchuria are analyzed.

This paper also casts doubt on Vovin's "Out of Southern China" theory.
We just don't see much connection between Japanese and Thai or Southern Chinese at all.

This paper has something like 92 percent Koreans(~WLR) and 8 percent Jomon for Japanese even though in both paternal and maternal HG the indigenous Japanese portion is much higher.
Maybe the same thing happened in Korea. The indigenous population with O1b2 endured the massive influx of both Japonic and Koreanic speakers and remained a very significant force until today.
It also follows that the "Koreans" who peopled Japan were not the direct WLR population but were Japonic speakers with some indigenous Korean(but not Koreanic speaking) uni-parental influence.

We just need ancient Korean samples analyzed. All these can be wild speculations.

P.S. I have a model for this. At least the persistence of paternal lineage. This can explain some of the European puzzles where indigenous WHG y haplogroups pop up later.

Didn't Korea get agriculture fairly late? Cheulmun culture similar to Jomon in Japan and all. Don't you think O2b could only have come with the rice farmers?

[Ryukendo]

I am sure that the "collaboration" is for Koreans to supply the samples they obtained from Mongolians and for Max Planck to do most of the analysis.
In terms of the general technological aptitude, Koreans are a top notch; they do excell in whatever money is poured into like in fusion research and military technology etc.
But they are also hampered by a sort of political correctness that has plagued them for 70 years.
A top mathematical and scientific talent is socially encouraged to go into engineering because a small and poor country like Korea cannot afford to do pure science.
That might have been a wise choice even 50 years ago but they still have that "small country complex" even after Korea has become rich.

I had communicated with top researchers in Korea and China when this research was at its infancy.
Koreans always complained about not getting enough support and they sometimes lied that their research had some medical implication to get funds.
Chinese who were technologically behind Korea in almost every field at the time (and were much poorer than Korea) had no such problem.
After 20 years we are seeing the result of it.

This is very much the case in Singapore as well. Narasimhan (yes, that Narasimhan) is a classic example of a brilliant Singaporean researcher who didn't want to join the national research pipelines, which are super commercially oriented and "strategically placed for Dominance in Biotechnology!" and for that reason incredibly boring.

[ybmpark]

Didn't Korea get agriculture fairly late? Cheulmun culture similar to Jomon in Japan and all. Don't you think O2b could only have come with the rice farmers?

But from where? Martine Robbeet(linguist who proposed TransEurasian language family this paper also discussed - this paper cast doubt) speculated that Japonic speakers arose from "Thai husbands and Japanese wives" - from "genetic evidence".
That was quite comical. She probably had , in mind, O1b1 and O1b2 having a common origin and also Vovin's "Out of Southern China". O1b1 and O1b2 diverged more than 30000 years ago while OoSC is an event less than 5000 years ago. Totally incompatible.
Christopher Beckwith performed a similar comedy when he claimed there is even a genetic evidence supporting Koguryo's Japonic language. Linguists are pretty bad geneticists.

However there is a rumor level report that 7000 years old remains on Kadeok Island yielded HG D for men and Jomon affinity for autosomes.
So maybe southern Korea was Jomon-like but O1b2 has to have been somewhere. Maybe central and northern Korea?
As I said previously, Yfull is quite misleading; too many Japanese and Chinese. Given enough samples, O1b2 seems solidly centered within Korea whatever the language the people spoke. O1b2-F940 that mainly appears on the continet was probably mediated by Manchu instead of being the remnants of O1b2 before it entered Korea.

There is a youtube made by Park Jonghwa who was a coauthor of several recent papers on Korean genetics. He candidly admitted that he was wrong when he speculated Koreans are a mix of Devil's Gate and Taiwanese aborigine-like people in the famous 2017 paper.
instead he pointed out some ancient northern coastal Chinese who are quite a bit closer to Koreans than even to modern Han Chinese.
He did not seem to be aware of WLR. If he was, he could have been rather shocked at the closeness.

[Ryukendo]

But from where? Martine Robbeet(linguist who proposed TransEurasian language family this paper also discussed - this paper cast doubt) speculated that Japonic speakers arose from "Thai husbands and Japanese wives" - from "genetic evidence".
That was quite comical. She probably had , in mind, O1b1 and O1b2 having a common origin and also Vovin's "Out of Southern China". O1b1 and O1b2 diverged more than 30000 years ago while OoSC is an event less than 5000 years ago. Totally incompatible.
Christopher Beckwith performed a similar comedy when he claimed there is even a genetic evidence supporting Koguryo's Japonic language. Linguists are pretty bad geneticists.

However there is a rumor level report that 7000 years old remains on Kadeok Island yielded HG D for men and Jomon affinity for autosomes.
So maybe southern Korea was Jomon-like but O1b2 has to have been somewhere. Maybe central and northern Korea?
As I said previously, Yfull is quite misleading; too many Japanese and Chinese. Given enough samples, O1b2 seems solidly centered within Korea whatever the language the people spoke. O1b2-F940 that mainly appears on the continet was probably mediated by Manchu instead of being the remnants of O1b2 before it entered Korea.

There is a youtube made by Park Jonghwa who was a coauthor of several recent papers on Korean genetics. He candidly admitted that he was wrong when he speculated Koreans are a mix of Devil's Gate and Taiwanese aborigine-like people in the famous 2017 paper.
instead he pointed out some ancient northern coastal Chinese who are quite a bit closer to Koreans than even to modern Han Chinese.
He did not seem to be aware of WLR. If he was, he could have been rather shocked at the closeness.

Where is this rumour from, did the Kadeok island remains get sent to Max Planck? (seems like the proper romanization is Gadeokdo for those who wanna look this up).
Can you send the youtube link for the talk? He's probably referring to Boshan.
About the recent Wang et al stuff, he eventually included Ning et al genomes but not the northern genomes of Fu's paper (Yumin, Boshan, Bianbian etc.). Have you run across any statistics comparing the closeness of Boshan vs WLR to Koreans and Japanese? Won't be surprised if e.g. Boshan was closer.
Lol @ "Thai husbands and Japanese wives" concept, but a "Northeastern agricultural" contribution to WLR and Taiwan_IA, say from the Shandong-Bohai area, is a good population from which Japonic at least could have come, no?