August 21, 2014

Ancient Y-DNA from China

From the paper:
Dividing the samples further using social status shows that the six aristocrats had haplogroups Q1a1, O3a, and N, the 14 commoners had haplogroups Q1a1, O3a, and O*, and the seven slaves had haplogroups O3a, O2a, and O* (Fig. 2).
Am. J. Hum. Biol. DOI: 10.1002/ajhb.22604

Ancient DNA evidence reveals that the Y chromosome haplogroup Q1a1 admixed into the Han Chinese 3,000 years ago

Yong-Bin Zhao et al.

Objectives

Y chromosome haplogroup Q1a1 is found almost only in Han Chinese populations. However, it has not been found in ancient Han Chinese samples until now. Thus, the origin of haplogroup Q1a1 in Han Chinese is still obscure. This study attempts to provide answer to this question, and to uncover the origin and paternal genetic structure of the ancestors of the Han Chinese.

Methods

Eighty-nine ancient human remains that were excavated from the presumed geographic source of the Han Chinese and dated to approximately 3,000 years ago were treated by the amelogenin gene polymerase chain reaction test, to determine their sex. Then, Y chromosome single nucleotide polymorphisms were subsequently analyzed from the samples detected as male.

Results 

Samples from 27 individuals were successfully amplified. Their haplotypes could be attributed to haplogroups N, O*, O2a, O3a, and Q1a1. Analyses showed that the assigned haplogroup of each sample is correlated to the suspected social status and observed burial custom associated with the sample.

Conclusions

The origins of the observed haplotypes and their distribution in present day Han Chinese and in the samples suggest that haplogroup Q1a1 was probably introduced into the Han Chinese population approximately 3,000 years ago. Am. J. Hum. Biol., 2014. © 2014 Wiley Periodicals, Inc.

Link

20 comments:

andrew said...

"Dividing the samples further using social status shows that the six aristocrats had haplogroups Q1a1, O3a, and N, the 14 commoners had haplogroups Q1a1, O3a, and O*, and the seven slaves had haplogroups O3a, O2a, and O* (Fig. 2)."

Ergo:

1. Q1a1 and N - superstrate population (probably ANE autosomally and Northeast Asian in origins).

2. O3a - predominant pre-conquest population found in all three strata.

3. O2a and O* - substrate population.

terryt said...

Eurogenes has a bit more:

http://eurogenes.blogspot.co.nz/2014/08/complex-paternal-origins-of-han-chinese.html

"Their haplotypes could be attributed to haplogroups N, O*, O2a, O3a, and Q1a1".

I have no doubt there will still be those who refuse to see a northern origin for any of these haplogroups apart, perhaps, for Q1a1.

TheXanian said...

I believe that this is not a new paper. I remember that I saw these results on a Chinese internet site back in 2012.

The ancient Chinese human remains came from the tombs of the Peng Kingdom, a small kingdom located in present-day Shanxi province in northern China. Peng aristocracy belonged to Q1a1.

bicicleur said...

from the presumed geographic source of the Han Chinese

can anybody tell where these human remains were found?

apostateimpressions said...

D, http://dienekes.110mb.com/ is down?

Model-Based Clustering of World Craniometric Variation
Anthropological Evidence and the Fallmerayer Thesis
Greek Y Chromosomes
Greek mtDNA
Greek autosomal DNA

tew said...

Those Q individuals might have been originally an intrusive Central Asian elite, a pattern that has been repeated throughout Chinese history. NB that in some studies on ancient Y-DNA from NW China nomads almost all individuals were Q of some sort.

Hector said...

Like many jingoistic Chinese papers this will probably turn out to be a farce. How did they define "Han Chinese" of 3000 years ago?

Lathdrinor said...

The Hengbei site in this paper is well studied archaeologically and is associated with a fief called Peng, whose rulers were recorded to have been migrants from the northern steppes - a region called Guifang during earlier times. Insofar as we accept that the 'rich graves' in the paper corresponding to the graves of the Peng rulers, the distribution of haplogroups matches with the records: Q1a was found among nomads in northwestern China around the same time, while N was found in Neolithic sites in Inner Mongolia.

O2a is an aberration this far north, given that O2a and its sub-branches are only found in about 2% northern Chinese today. Here it is only found among sacrificial slaves. I wonder whether they are war captives.

An additional issue arises in what the O* - the other 'commoner/slave' haplogroup - is. The paper mentions that the subjects under O were only typed for O, O1a, O2a, O3, O3a, and O3a3. That leaves other branches of O1 and O2 (xO2a) untested. My thinking is that it is O2*-M268, given the present distribution of haplogroups in northern China.

O3a/O3a3 is represented in all three social classes, and I imagine that makes them the best candidate for the locals, whose existing social structure merged with that of the migrants'. Intermarriage surely occurred between the two, as was recorded in Chinese records, which state that the rulers of Peng intermarried with the nobility of the Zhou.

Unfortunately, the authors admit to not testing for consanguinity within the samples. Thus, we don't know whether this is a fair sampling of the families of Hengbei at the time vs. a case of one Q1a patrilineal clan showing up repeatedly.

terryt said...

"1. Q1a1 and N - superstrate population (probably ANE autosomally and Northeast Asian in origins).
2. O3a - predominant pre-conquest population found in all three strata.
3. O2a and O* - substrate population".

Agreed.

"O2a is an aberration this far north, given that O2a and its sub-branches are only found in about 2% northern Chinese today".

The extent to which O2a is a 'southern' haplogroup is debatable. O2a and O2b share a common ancestor and O2b is undoubtedly northern. And O2a was present at the Middle Yangtze 3500 years ago. O2a's mere 2% presence in northern Chinese today may be the product of replacement and a push southward through later expansions.

Ebizur said...

terryt wrote,

"The extent to which O2a is a 'southern' haplogroup is debatable. O2a and O2b share a common ancestor and O2b is undoubtedly northern."

O2a is found with highest frequency today among populations that have been long isolated either physically (Nicobarese in the Nicobar Islands) or socially (the more tribal/primitive among the Mundaic adivasis of India, e.g. Juang). If it has come from northern China, it must have done so in the distant past (e.g. very early Neolithic), especially since it seems to be correlated with the "Austroasiatic" autosomal profile (and language phylum, for that matter), which seems to form a substrate under Austronesian-speaking populations of the Malay Peninsula and western Indonesia; assuming the above to be true, the O2a-carrying Austroasiatic substrate of Southeast Asia must have been in place at least 5000 years ago. Furthermore, the modern Austroasiatic languages are so diverse that the relationships among them are barely discernible, and no branching order is readily apparent; their ancestors must have spread very rapidly over a wide territory because of little resistance. This suggests that the Austroasiatic peoples were either the first settlers of Southeast Asia or the first newcomers who had a distinct advantage (e.g. Neolithic culture) over the previous inhabitants.

As for O2b, the relationship between that clade and O2a is only about as close as the relationship between J1 and J2, Q and R, or I1 and I2 (cf. Wang et al. 2014). Estimations of the TMRCA based on the molecular clock hypothesis are controversial, of course, but I think the most widely accepted estimate for each of these splits is somewhere around 30,000 YBP, plus or minus 5000 years or so. O2a has expanded early and left branches of about the same time depth as early branches of T or O3; the branches that are most divergent from the main core of this haplogroup have been found with low frequency over a wide area from Ho Chi Minh City and Sipsongpanna to Tokyo and Beijing, in peoples like the Naxi, Dai, Kinh, Han, and Japanese, by the way. O2b, on the other hand, seems to have been bottlenecked or failed to expand until about the same era as N, O1, O3a1c, or O3a2c1. O2b most definitely does not exhibit a trace of rapid expansion after its MRCA with O2a. On the basis of the available evidence, there is almost zero probability of O2a and O2b reflecting the same, singular Neolithic expansion; they already existed in separate Palaeolithic populations. Actually, even O2a itself most likely already existed in several interrelated but separate Palaeolithic populations prior to the Neolithic revolution in eastern Asia (the MRCA of all extant O2a is significantly older than the MRCA of all extant R1, for example), although it is possible that O2b might have persisted only in a small, monolithic population as late as the advent of the Neolithic. The long isolation and relatively recent expansion of O2b is also suggested by its present geographic distribution, which is almost entirely limited to areas surrounding the Sea of Japan; it has not even been found in human remains associated with cultures of Northeast China from the Neolithic through the Late Bronze
Age, so it most likely existed only in the Korean Peninsula or the Japanese Archipelago until very recently.

You also should take note of the fact that the split between O2a and O2b is older than the TMRCA of Japanese D1b according to Lippold et al. 2014. In other words, it is quite possible (though not necessarily true) that O2b might represent a substrate underlying D1b in the Japanese Archipelago and Korean Peninsula (in any case, it is not at all clear which of these haplogroups has arrived earlier to the vicinity of the Sea of Japan).

Lathdrinor said...

"The extent to which O2a is a 'southern' haplogroup is debatable. O2a and O2b share a common ancestor and O2b is undoubtedly northern. And O2a was present at the Middle Yangtze 3500 years ago. O2a's mere 2% presence in northern Chinese today may be the product of replacement and a push southward through later expansions."

As far as I know, O2a has not been found in any Neolithic site in northern China. This sample of O2a from the Bronze Age is the furthest north I've seen it in ancient East Asian samples. You are correct about it being found around the middle Yangtze River but that is hardly northern. It is also rarely found alongside O2b. A replacement scenario is not very parsimonious. It is cleaner to posit a bifurcated distribution resulting from separate evolution of the ancestral haplogroup O2*.

Kristiina said...

Ebizur, I just sent you my very delayed reply to your Tohoku post!

terryt said...

"You are correct about it being found around the middle Yangtze River but that is hardly northern".

The Yangtze has been used very often to define the boundary between 'northern' and 'southern' China.

"If it has come from northern China, it must have done so in the distant past (e.g. very early Neolithic)"

That is what I have long assumed: the very early Neolithic.

"especially since it seems to be correlated with the 'Austroasiatic' autosomal profile (and language phylum, for that matter), which seems to form a substrate under Austronesian-speaking populations of the Malay Peninsula and western Indonesia"

But that Austro-Asiatic substrate may not be particularly ancient, if this link at Wikipedia is correct:

http://en.wikipedia.org/wiki/Proto-Mon%E2%80%93Khmer_language

Quote:

"Much work has been done on the reconstruction of Proto-Mon–Khmer in Harry L. Shorto's Mon–Khmer Comparative Dictionary. Sidwell (2007, 2009, 2011)[full citation needed] suggests that the likely homeland of Austroasiatic is the middle Mekong, in the area of the Bahnaric and Katuic languages (approximately where modern Laos, Thailand, and Cambodia come together), and that the family is not as old as frequently assumed, dating to perhaps 2000 BCE".

I agree its disjunct distribution indicates it has been replaced through much of its earlier range, but the above places its expansion (although not its presence in SE Asia) as pre-dating Austronesian by a very short period.

"the modern Austroasiatic languages are so diverse that the relationships among them are barely discernible, and no branching order is readily apparent; their ancestors must have spread very rapidly over a wide territory because of little resistance. This suggests that the Austroasiatic peoples were either the first settlers of Southeast Asia or the first newcomers who had a distinct advantage (e.g. Neolithic culture) over the previous inhabitants".

That seems quite likely to me too. In fact to me it is almost certain that southern China was very sparsely inhabited until the Neolithic. The survival of megafauna such as panda, orang-utan and tapir support this idea.

"Estimations of the TMRCA based on the molecular clock hypothesis are controversial, of course, but I think the most widely accepted estimate for each of these splits is somewhere around 30,000 YBP, plus or minus 5000 years or so".

Extremely likely dates. In fact the diversity of the whole NO clade seems to date from around then. In other words NO broke rapidly into N, O1, O2a, O2b and O3. The O3a1, O3sa2, O3a3 and O3a4 split is probably somewhat later.

"the branches that are most divergent from the main core of this haplogroup have been found with low frequency over a wide area from Ho Chi Minh City and Sipsongpanna to Tokyo and Beijing"

That doesn't help much in discerning its origin though. In fact it looks very much like a relatively recent wide expansion of the haplogroup.

"O2b most definitely does not exhibit a trace of rapid expansion after its MRCA with O2a".

It would be normal for a haplogroup that manages to move from the region of common origin of the wider clade to expand rapidly while anything that remained behind would continue to be limited in its ability to expand and diversify.

"On the basis of the available evidence, there is almost zero probability of O2a and O2b reflecting the same, singular Neolithic expansion; they already existed in separate Palaeolithic populations".

But those populations are likelyto have been reasonably close geographically, at least to start with.

Hector said...

There has been a lot of unpublished work done on O2 clade by amateur scientists.

Every single O2* subjected to further SNP analysis so far has been found to harbor mutations that are shared with O2a but not O2b so they will be in the O2a clade as soon as those SNPs are verified as genuine. In fact they probably form a single clade of shallow time depth so that the current O2a will become O2a1 and they, O2a2 etc.

Unfortunately for Jake(or Ebizur, whatever he wishes to call himself), the SNP analysis of O2b indicates that its history in Japan cannot predate that in Korea. Rare types that are more upstream are found only in Korea so far. Koreans have their own "star cluster" comparable to O2b-47z which accounts for nearly 90 percent of Japanese O2b's. In fact they are "brother clades" not too distant from each other.
But Koreans also have other types more upstream.

TheXanian said...

@ terryt

So far there's not enough evidence to prove that haplogroup O and O2a originated from Northeast Asia or Siberia (I'm assuming that by "northern" you mean Northeast Asia, cause Yangtse River is hardly northern).

Let's think for a while. If NO and O were really originated from the north or migrated to the north very early on, then we would expect them to have a much wider distribution, in particular we would expect to find them in Europe and the Americas just like their cousins Q and R. However, the fact that O is completely absent from both Europe and the Americas and N is only present on the periphery of Europe make me think that N and O were not the first settlers of the northern Eurasian steppes but rather they were the late comers.

And we do have evidences about Q and R being the early northern settlers. The 24,000 year old Mal'ta child found in Siberia belongs to haplogroup R, and another 17,000 year old individual from Siberia belongs to Q. So far haplogroups N and O had not been found in Siberia at such early dates. Of course you could always expect that some Paleolithic bones of a haplogroup NO hunter-gatherer might turn up in Siberia some day in the future, but when taking into consideration of the current distribution of haplogroups, I feel that it's very unlikely that N and O were in the same place as Q and R at the same time, cause otherwise N and O should have a much wider distribution in Europe and the Americas.

So far all the evidences point to a southern origin of NO.

Ebizur said...

A summary of heretofore published ancient Y-DNA from China in (roughly) chronological order:

Daxi site (Daxi Town, Wushan County, Chongqing)/Daxi culture/approx. 6400 YBP ~ 5300 YBP/Neolithic (Li et al. 2007)
9/20 Undetermined (?= Y*(xO1a-M119, O2a1-M95, O3-M122))
5/20 O3a2b-M7
4/20 Missing (= "No amplifiable product could be obtained")
1/20 O2a1-M95
1/20 O3-M122(xO3a2b-M7, O3a2c1-M134)

Niuheliang site (the border of Lingyuan County and Jianping County, Liaoning Province, PRC)/Hongshan culture/approx. 6500 YBP ~ 5000 YBP/Neolithic (Cui et al. 2013)
2/6 N-M231(xN1c2a-M128, N1c1-Tat)
2/6 N-M231(xN1c1-Tat)
1/6 C-M216 (subclade undetermined)
1/6 O3-M122 (subclade undetermined)

Miaozigou site (Chahar Right Front Banner, Ulanqab, Inner Mongolian Autonomous Region, PRC)/Miaozigou phase of the Yangshao culture (Yangshao with Hongshan influence?)/approx. 6000 YBP ~ 5000 YBP/Neolithic (Cui et al. 2013)
3/3 ?N1c2a-M128/?N-M231(xN1c2a-M128, N1c1-Tat) (There is a discrepancy between Table 1 of the main article, in which these three Y-DNA samples are described as N1(xN1a, N1c), and Table S1 of the data supplement, in which the Y-SNP genotyping data indicate an assignment of N1c2a-M128.

Xindili site (Tongxiang City, Zhejiang)/Liangzhu culture/approx. 5300 YBP ~ 4100 YBP/Neolithic (Li et al. 2007)
5/9 O1a-M119
3/9 Undetermined
1/9 Missing

Maqiao site (Maqiao Town, Minhang District, Shanghai)/Liangzhu culture/approx. 5300 YBP ~ 4100 YBP/Neolithic (Li et al. 2007)
4/6 O1a-M119
2/6 Undetermined (?= Y*(xO1a-M119, O2a1-M95, O3-M122))

Halahaigou site (Yuanbaoshan County, Chifeng City, Inner Mongolian Autonomous Region, PRC)/Xiaoheyan culture/approx. 5000 YBP ~ 4200 YBP/Neolithic (Cui et al. 2013)
8/12 N-M231(xN1c2a-M128, N1c1-Tat)
4/12 N-M231(xN1c1-Tat)

Taosi site (Xiangfen County, Linfen City, Shanxi)/Longshan culture/approx. 4500 YBP ~ 3900 YBP/Late Neolithic (Li et al. 2007)
3/5 O3-M122(xO3a2b-M7, O3a2c1-M134)
1/5 O3a2c1-M134
1/5 Missing

Dadianzi site (Aohan Banner, Chifeng City, Inner Mongolian Autonomous Region, PRC)/Lower Xiajiadian culture/approx. 4200 YBP ~ 3600 YBP/Early Bronze Age (Cui et al. 2013)
3/5 N-M231(xN1c1-Tat)
2/5 O3-M122 (subclade undetermined)

Wucheng site (Wucheng Town, Zhangshu City, Yichun Municipality, Jiangxi)/Wucheng culture/approx. 3500 YBP ~ 3100 YBP/Early Bronze Age (Li et al. 2007)
2/4 O2a1-M95
1/4 O3-M122(xO3a2b-M7, O3a2c1-M134)
1/4 Missing

Hengbei site (Hengbei Village, Hengshui Town, Jiang County, Yuncheng Municipality, Shanxi)/Western Zhou vassal state of Peng/approx. 3000 YBP/Late Bronze Age (Zhao et al. 2014)
11/27 Q1a1a1-M120
5/27 O3a2-P201
4/27 O3a-M324 (could not be amplified at locus of P201)
4/27 O-M175(x?) (Unfortunately, the authors have not indicated clearly which SNPs under M175 have been amplified successfully and found to be negative/ancestral in each of these O-M175 individuals.)
2/27 O2a1-M95
1/27 N-M231(xN1c2a-M128, N1c1-Tat) (no amplification at the locus of P43)

Dashanqian site (Harqin Banner, Chifeng City, Inner Mongolian Autonomous Region, PRC)/Upper Xiajiadian culture/approx. 3000 YBP ~ 2700 YBP/Late Bronze Age (Cui et al. 2013)
3/9 N1c1-Tat
2/9 O3a-M324(xO3a2-P201)
2/9 O3a2c1a-M117
1/9 N-M231(xN1c2a-M128, N1c1-Tat)
1/9 C2c-P53.1

Jinggouzi site (Linxi County, Chifeng City, Inner Mongolian Autonomous Region, PRC)/Northern nomad culture/approx. 3000 YBP ~ 2500 YBP/Late Bronze Age (Cui et al. 2013)
12/12 C2c-P53.1

Zhongzhuang and Wangdahu sites (Pengyang County, Guyuan Municipality, Ningxia)/Northern nomad culture/approx. 2800 ~ 2300 YBP/Late Bronze Age (Zhao et al. 2010)
4/4 Q-M242

Taojiazhai site (Hehuang area, Xining, Qinghai)/Han culture (Eastern a.k.a. Later Han Dynasty colony in the Western Territories)/approx. 1900 ~ 1700 YBP/Imperial China (Zhao et al. 2011)
12/12 O3-M122

Ebizur said...

In the Neolithic sites tested so far, Y-DNA haplogroup N-M231(xN1c1-Tat) predominates in Inner Mongolia and Liaoning, O1a-M119 predominates in the area around the mouth of the Yangtze River, and O3-M122 predominates in the Central Plain (though with only four samples, all from the Taosi site of the terminal Neolithic, tested so far). O3-M122 is also common in the samples from the most ancient site tested so far (Daxi in the Three Gorges area around the middle reaches of the Yangtze River), but 5/6 of the O3-M122 individuals from that site belonged to the O3a2b-M7 subclade; furthermore, one of the individuals has been determined to belong to O2a1-M95, and the Y-DNA haplogroups of most individuals sampled from the Daxi site remain "undetermined" ("For some of the samples, not all of the five SNPs could be amplified, thus haplogroups could not be determined if no mutated alleles were found (shown as undetermined in Table 1)"). This might suggest that the population of the Daxi site (proto-Hmong-Mien?) retained an indigenous element despite receiving some influence from (presumably) the Central Plain.

Lathdrinor wrote,

"As far as I know, O2a has not been found in any Neolithic site in northern China. This sample of O2a from the Bronze Age is the furthest north I've seen it in ancient East Asian samples. You are correct about it being found around the middle Yangtze River but that is hardly northern. It is also rarely found alongside O2b. A replacement scenario is not very parsimonious. It is cleaner to posit a bifurcated distribution resulting from separate evolution of the ancestral haplogroup O2*."

I agree. O2a1-M95 has been found in 1/7 successfully genotyped Y-DNA samples from the Neolithic Daxi site along the middle Yangtze and in 2/3 successfully genotyped Y-DNA samples from the Bronze Age Wucheng site, much later and further downriver (on a right tributary of the lower Yangtze). Neither of those sites may be described as "northern" (Daxi is in Chongqing, and Wucheng is in Jiangxi), and the frequency of O2a1-M95 in the older Daxi sample is not particularly high. O2a1-M95 also has been found in two immolated slaves out of a total sample of 27 individuals from the Hengbei site in southern Shanxi, which is probably only a few hundred years more recent than the Wucheng site. It seems very likely that these slaves (actually, sacrificial victims is a more apt description) were forcibly brought (perhaps as prisoners of war) from the southern neighbors of the Zhou. Actually, the Hengbei site dates to an era when the early Zhou Dynasty, not long after conquering the preceding Shang, was expanding into semi-barbarian areas to the south, and enfeoffing their vassals as rulers of those new territories. The domain of Chu was founded in this way in the latter half of the 11th century BCE.

Kristiina wrote,

"Ebizur, I just sent you my very delayed reply to your Tohoku post!"

Thank you for the notice, Kristiina.

Ebizur said...

I apologize for my back-to-back comments. Continuing...

terryt wrote,

"The Yangtze has been used very often to define the boundary between 'northern' and 'southern' China."

All right. Both the Daxi site and the Wucheng site are located to the south of the Yangtze River, so both fall within "Southern China" under that definition (although the Daxi site is practically on its right bank).

terryt wrote,

"I agree its disjunct distribution indicates it has been replaced through much of its earlier range, but the above places its expansion (although not its presence in SE Asia) as pre-dating Austronesian by a very short period."

I am not sure how the authors of that study have reached that conclusion. Personally, I have compared the lexica of many Austronesian languages and many supposedly Austroasiatic languages, and whereas (most) members of the former group are similar and obviously interrelated, the similarity among members of the latter group is significantly less salient. If one wants to claim that the expansion and diversification of the proto-Austroasiatic language has occurred not long before the expansion and diversification of the proto-Austronesian language, I think one should provide an alternative explanation for the much higher degree of similarity among the descendants of the latter protolanguage than among descendants of the former protolanguage (e.g. greater influences from diverse substrate languages on Austroasiatic languages, although that would be special pleading IMHO, and difficult to falsify) because the standard explanation in this situation would be that the proto-Austroasiatic language had expanded and diversified significantly earlier.

terryt wrote,

"That doesn't help much in discerning its origin though. In fact it looks very much like a relatively recent wide expansion of the haplogroup."

The very early branches of O2a to which I have referred in my previous comment to you are actually O2a(xM95): O2a-PK4(xM95), e.g. O-CTS11792, and O2a-K18(xPK4), e.g. O-CTS10887. Members of these haplogroups are, by definition, not part of O-M95, which is the Y-DNA haplogroup that has been found to be particularly common among present-day speakers of Austroasiatic languages. The TMRCA of O-CTS11792 and O-M95 appears to be comparable to that of all extant J2a or I2, and just slightly less than the TMRCA of all extant O3a1 or O2b (although these intraclade estimates may change pending the discovery of more divergent members). The spread of I2 in Europe is certainly of at least Mesolithic time depth; O-M95 proper may also be roughly so old.

The TMRCA of O-CTS10887 and O-PK4 is even older by a great margin; that split appears to date back to around the MRCA of all extant O3-M122, or significantly earlier than the MRCA of all extant N-M231 or the MRCA of R1a and R1b (this last is an interclade estimate, so unlikely to change much).

The fact that some representatives of the divergent O-CTS10887 clade have been found among Han and Japanese as well as among Southeast Asians (but nowhere else as far as I know) only points to the likelihood that the collateral relatives of O-M95 have been in East and/or Southeast Asia for a very long time; it does not suggest that they (or O-M95 itself) all have originated in a Neolithic culture of North China and moved ever southward from that origin. Of course, the available evidence does not disprove your hypothesis, either; until either your "North China" hypothesis or the "Southeast Asia" hypothesis of the origin of O-M95 can be falsified, it must remain an open question.

terryt said...

@ TheXanian:

"So far there's not enough evidence to prove that haplogroup O and O2a originated from Northeast Asia or Siberia (I'm assuming that by 'northern' you mean Northeast Asia, cause Yangtse River is hardly northern)".

Not really. By 'northern' I mean northern China, probably some where near the Tsinling Mountains.

"If NO and O were really originated from the north or migrated to the north very early on, then we would expect them to have a much wider distribution"

Obviously they were not as far north as Mongolia. In fact O is barely that far north today.

@ Ebizur:

"the likelihood that the collateral relatives of O-M95 have been in East and/or Southeast Asia for a very long time"

They've certainly been in East Asia for a very long time. In fact must have originated there. The problem lies in SE Asia. It seems very likely that the presence of a 'Mongoloid' phenotype there is no older than the early Neolithic, and it must have been introduced by some population expansion. Populations containing various O haplogroups are the only option I see. I am open to suggestions if anyone can suggest an alternative.

"Members of these haplogroups are, by definition, not part of O-M95, which is the Y-DNA haplogroup that has been found to be particularly common among present-day speakers of Austroasiatic languages".

Something I find extremely interesting is that Munda O is O2a1-M95, specifically O2a1a-M88. That suggests it may be surprisingly recent in South Asia.

"O1a-M119 predominates in the area around the mouth of the Yangtze River, and O3-M122 predominates in the Central Plain (though with only four samples, all from the Taosi site of the terminal Neolithic, tested so far). O3-M122 is also common in the samples from the most ancient site tested so far (Daxi in the Three Gorges area around the middle reaches of the Yangtze River), but 5/6 of the O3-M122 individuals from that site belonged to the O3a2b-M7 subclade; furthermore, one of the individuals has been determined to belong to O2a1-M95"

All those haplogroups are present in Neolithic sites. The Chinese Neolithic originated somewhere between the middle reaches of the Yangtze and Yellow Rivers which implies quite strongly that all those haplogroups originated there too.

Hector said...

Currently O2a is defined by PK4. K17(so far equivalent to K18 K19 K20) is an upstream marker and page 59 is downstream to K17 and appears to encompass all known O2*'s tested so far.(in short page 59 clade and PK4 clade are "brother clades")

The distribution of Page 59 clade(currently O2*) is somewhat higher in the northern East Asia than southern.

Koreans 2 percent
Japanese 1
Koreans in China 3.8
Northern Han 4.5
Hui 3.7
Manchu 8
Inner Mongolians 2.2
Harbin Han 8.6
Uighurs 9.4
Kiang(Tibet) 2.9
Yi 2.3
Taiwan Han 7.1
Toga 4.1
Outer Mongolians 1.3
Nanai(Tungus) 2.2
Dagur 2.6
Orchen(Tungus) 4.5
Evenki(Tungus) 4.9

source: daum cafe - molecular anthropology

I had to trouble myself for this because someone did not lose his habit of talking out of his ass.
NO!! O2* is not restricted to Chinese and Japanese. In particular Japanese are generally irrelevant to these topics.