October 12, 2013

Ancient Y chromosomes of West Liao River valley

I added the results in my compendium.

BMC Evolutionary Biology 2013, 13:216 doi:10.1186/1471-2148-13-216

Y Chromosome analysis of prehistoric human populations in the West Liao River Valley, Northeast China

Yinqiu Cui et al.

Abstract

Background

The West Liao River valley in Northeast China is an ecologically diverse region, populated in prehistory by human populations with a wide range of cultures and modes of subsistence. To help understand the human evolutionary history of this region, we performed Y chromosome analyses on ancient human remains from archaeological sites ranging in age from 6500 to 2700 BP.

Results

47 of the 70 individuals provided reproducible results. They were assigned into five different Y sub-haplogroups using diagnostic single nucleotide polymorphisms, namely N1 (xN1a, N1c), N1c, C/C3e, O3a (O3a3) and O3a3c. We also used 17 Y short tandem repeat loci in the non-recombining portion of the Y chromosome. There appears to be significant genetic differences between populations of the West Liao River valley and adjacent cultural complexes in the prehistoric period, and these prehistoric populations were shown to carry similar haplotypes as present-day Northeast Asians, but at markedly different frequencies.

Conclusion

Our results suggest that the prehistoric cultural transitions were associated with immigration from the Yellow River valley and the northern steppe into the West Liao River valley. They reveal the temporal continuity of Y chromosome lineages in populations of the West Liao River valley over 5000 years, with a concurrent increase in lineage diversity caused by an influx of immigrants from other populations.

Link

27 comments:

Lathdrinor said...

The gradual subsumption of N1 by immigrants looks valid, but 47 / 70 over 5,000 years is still too small of a sample size to decide whether this is just the effects of geographically associative sample bias. The uniformly C3e results for one of the sites says it all - people of the same patriline traveled in groups. For all we know, the proliferation of N1 was due to the samples all being from the Neolithic equivalent of 'clan villages', though it helps that the samples are from different archaeological sites.

The authors mention the migration of farmers from the Yellow River valley to the West Liao valley and states that these farmers had different haplotypes than the locals. This process is reminiscent of the farmer vs. hunter-gatherer divide in Neolithic Europe, but there is an important difference: the West Liao population were also farmers. I wonder whether this East Eurasian version of the great farming expansion is still capable of being explained in the same way given the diverse profile of early farmers.

terryt said...

"Our results suggest that the prehistoric cultural transitions were associated with immigration from the Yellow River valley and the northern steppe into the West Liao River valley".

I presume 'O3a3c' refers to what is now called 'O3a2c1-M134. But either or both haplogroups arriving 'from the Yellow River valley' strongly suggests a Neolithic expansion. From SE Asia? Unlikely. I would bet that O3's main expansion expansion is Neolithic, as probably is the whole Y-DNA O expansion. O is not some Paleolithic SE Asian haplogroup in spite of what many wish to believe.

Ed said...


The Yellow River Valley had mainly Siberian paternal lineages during pre-historic times. There was clearly a demographic shift that came with the arrival of Rice agriculture from either the Yangtze or Pearl River area. This corresponds with pockets of rice cultivation appearing in the area during the Yangshao culture and becoming the dominant food source for human consumption during the Longshan culture.

http://cdmd.cnki.com.cn/Article/CDMD-10183-1012365432.htm

http://anthropology.usf.edu/news/pdfs/Lanehart%20et%20al.%202011.pdf

http://www.humpopgenfudan.cn/p/E/E3.pdf

terryt said...

"I wonder whether this East Eurasian version of the great farming expansion is still capable of being explained in the same way given the diverse profile of early farmers".

I doubt that either C3e or N1 introduced farming into the region.

Mason said...

N1(xN1a, N1c) was once widespread in northern China around 3000~5500 years ago. There was a Chinese aDNA paper showing that N1(xN1a, N1c) was found among ancient DNA of not only northeast, but also northern and northwest China's neolithic cultures.

Please see: http://www.ranhaer.com/viewthread.php?tid=21547&page=24#pid302164

Lathdrinor said...

@Ed I see no evidence of what you state in the papers you cited, though I didn't read the Chinese language paper due to it being, obviously, in Chinese. Do you have a specific reference / translation?

It's obvious that this current paper considers the O3 patriline to be well established in the Yellow River valley in prehistoric times; otherwise its argument of a Central Plains migration for this haplotype in the mid Neolithic Liao River valley makes no sense.

Ed said...


@Lathdrinor Right, I was never trying to claim otherwise. Sorry for the misunderstanding.

The question I was trying to raise is “when did O3 become dominant in the Yellow River Valley during the prehistoric period”. The prehistoric Longshan culture (3000BC – 1900 BC) in the Yellow River Valley was primarily Y-DNA O3 The prehistoric Yanghsao culture(5000BC – 3000BC) in the Yellow River valley was primarily Y-DNA N. This was stated in this paper and those I previously linked.

The Longshan period was also a period of technological advancement/transition, including bronze weapons and tools. It was also a very violent time in the pre-history of Northern China where mass grave sites were commonplace.
http://archaeology.about.com/od/lterms/qt/longshan.htm

It was also a period when rice replaced millet as the primary food source for human consumption
“The Liangchengzhen results indicate that by the Longshan period, people in southeastern
Shandong no longer relied as heavily on millet, while it was probably a primary source of fodder
for domesticated animals such as pigs and dogs. Figure 5 reveals that Liangchengzhen apatite
values (δ
13C = -9.8‰) were similar to apatite values at Jiahu (δ
13C = -10.4‰), a site known for
rice cultivation. This suggests that other agricultural crops, especially rice, had increased in
importance at Liangchengzhen”

“Rice may have been cultivated at large centers such as Liangchengzhen as a preferred
food and used for public ceremonies or as gifts to influence and acquire allies as during the
Shang period (see Underhill and Fang, 2004). A higher social status of some human consumers
at Liangchengzhen, in comparison to residents of smaller settlements, could partially explain the
pattern of greater availability of rice for ritual and other social contexts. Increased social
stratification may have limited access to this preferred food at smaller sites (Underhill et al.,
2008).

Lathdrinor said...

@Ed correct me in case I'm wrong, but isn't the N bearing Yangshao site they're referring to in *south-central Inner Mongolia?* I know the Yellow River curves upwards into Inner Mongolia in west to east course, but the areas they're referring to aren't exactly what I regard to be the Yellow River Valley.

Yangshao was a rather widespread culture, and I'm not even sure I believe that the site they're talking about is typical of Yangshao sites in the Yellow River valley. While it's conceivable that N was the primary haplotype in Yangshao cultural sites - remembering that Yangshao was concentrated in northwest China - it is rather difficult to conceive that it is the primary haplotype of contemporary sites to the east of Yangshao.

Indeed, we find from the authors' results that Hongshan culture, which was contemporary with Yangshao, had 1/6 sample turn out to be O3. Provided this isn't a result of sample contamination, it tells us O3 was already in the Liao River region in ~3000-5000 BC. Unfortunately, no samples of contemporary sites south of the Liao - ie in the lower and middle Yellow River valley - were provided with this article, but I have a difficult time believing that O3 wasn't already one of the primary haplotypes in those regions then, especially given the observation that the O3 found in southern China during the late Neolithic were of a different subclade than the ones found in northern China during the late Neolithic.

terryt said...

"The question I was trying to raise is 'when did O3 become dominant in the Yellow River Valley during the prehistoric period'. The prehistoric Longshan culture (3000BC – 1900 BC) in the Yellow River Valley was primarily Y-DNA O3 The prehistoric Yanghsao culture(5000BC – 3000BC) in the Yellow River valley was primarily Y-DNA N".

Seems O3 has 'always' been present in the Yellow River region. The paper claims:

"The Miaozigou site, about 500 km west of the West Liao River valley in the central/southern region of Inner Mongolia, was settled by people of the northern branch of the Yangshao culture, an important Neolithic farming culture along the Yellow River. Our analysis of three ancient Miaozigou individuals revealed that they all belong to haplogroup N1(xN1a, N1c), while the main lineage of the Yellow River valley culture is O3-M122"

N1 was an immigrant from 'the central/southern region of Inner Mongolia'. That places O3 in the Yellow River before N1's arrival there. In fact it seems likely O, and perhaps N, coalesced from NO in that region. The paper further states:

"O3a continued to enter the West Liao River valley during the expansion of the Yellow River valley culture, displacing or replacing the original lineages".

That further supports the idea that O3 originated in the Yellow River region, not further south. N looks to have originated slightly further north, which makes sense as NO appears to have basically split simultaneoulsy into four subgroups: O1, O2, O3 and N. That makes it likely that neither O1 nor O2 are 'southern' haplogroups either. For example:

"haplotype O3d, found at high frequency in the ancient Daxi site in the middle reaches of the Yangtze River, is very rare in living people, except in the Hmong-Mien population of southern China and Southeast Asia. This suggests the ancient Daxi migrated south and became the ancestors of the present Hmong-Mien"

And:

"The Upper Xiajiadian individuals of the late Bronze Age had different subtypes of O3a-M324, O3a3c-M117. O3a-M324 is found today in most East Asian populations, and its subtype O3a3c-M117 occurs at the highest frequency in modern Sino-Tibetan populations"

We can be fairly sure that Sino-Tibetan-speaking populations have moved south from the region of the Upper Yangtze and Yellow Rivers.

Ed said...

@Lathdrinor
Right, O3 clearly was present ~3000-5000BC. This coincides with small pockets of rice cultivation appearing in the north. I don’t think it was the dominant haplotype even in the Northeast sites. You can take a look for yourself and many of the “North East” Sites Y-DNA O is indeed present ~3000BC, but not the dominant group quite yet. In the later sites O3 has in many cases become dominant. http://eurogenes.blogspot.com/2013/01/lots-of-ancient-y-dna-from-china.html
I just find it hard to believe that the technological, cultural, agricultural and demographic changes in Northern China during this timeframe aren’t related somehow.

@terryt
I stumbled upon the ‘northern route vs southern route’ debate between you and Maju a year back or so. We don’t need to reiterate those points here. However, at this time, I do find Maju and the research supporting a southern route of O much more convincing. I know better than to engage you further on this point.
http://forwhattheywereweare.blogspot.com/2011/04/haplogroup-o3-downstream-structure.html#comment-form

Lathdrinor said...

@Ed I'm not contesting the North East sites, but the Yellow River valley sites, of which we have no samples before ~3000 BC. While the idea of O3 = rice expansion is not altogether ridiculous, one has to keep in mind that shifting between different types of crops does not ultimately require a revolutionary change in life-style and technology. Examples of new crops - ie wheat, corn, etc. - being introduced into an agricultural zone without attendant demographic change are plentiful. Further, crops ie wheat and millet remained important in northern China well into the historic, even though the samples after 3,000 BC reveal an overwhelmingly O3 population.

At the same time, the idea that O3 expanded with Longshan from the south is contradicted by the fact that the Longshan archaeological complexes, despite being mainly rice farming, does not match contemporary traditions found in southern China, but developed out of earlier archaeological complexes in northern China, such as Dawenkou and Yangshao. There IS a southern contribution from Hemudu-Liangzhu, but tests at Liangzhu sites reveal an O1 population. Contemporaneously, the Daxi sites in southwest China were occupied by O3d populations, which are rare in modern Chinese.

Given the data, I have to think that the primary O3e and O3* subclades found in Chinese today were already well established in northern China, even in the case that they were not necessarily the main haplotype in the region ~5000 BC. The idea of a massive mid Neolithic expansion tied to intensive agriculture is not illogical, and indeed the new article Dienekes posted supports it. But I don't think this expansion started during the Longshan and I don't think it came out of a southern archaeological complex.

terryt said...

"Yangshao was a rather widespread culture, and I'm not even sure I believe that the site they're talking about is typical of Yangshao sites in the Yellow River valley".

The authors specifically say:

"The Miaozigou site, about 500 km west of the West Liao River valley in the central/southern region of Inner Mongolia, was settled by people of the northern branch of the Yangshao culture, an important Neolithic farming culture along the Yellow River. ... The existence of N1(xN1a, N1c) in the Miaozigou site could be evidence for the expansion of the Hongshan culture during its heyday, a view supported by archaeological evidence of Hongshan influences at the Miaozigou site".

So, yes, a culture at the margin of the Yanghsao, likely to have become mixed with neighbouring populations.

"I don’t think it [O3] was the dominant haplotype even in the Northeast sites".

I am sure it, along with O1 and O2, was the dominant haplogroup that began the Chinese Neolithic though. In fact the Neolithic's expansion was largely led by Y-DNA O.

"at this time, I do find Maju and the research supporting a southern route of O much more convincing".

I don't deny that O's deeper origins are via a southern route. O is descended from MNOPS after all. Anyway, check this out:

http://dienekes.blogspot.co.nz/2013/10/neolithic-super-grandfathers-of-chinese.html

Quote:

"three strong star-like Neolithic expansions at ~6 kya (thousand years ago) (assuming a constant substitution rate of 1e-9/bp/year) indicates that ~40% of modern Chinese are patrilineal descendants of only three super-grandfathers at that time. This observation suggests that the main patrilineal expansion in China occurred in the Neolithic Era and might be related to the development of agriculture".

Still going to claim a major Paleolithic O presence in SE Asia?

"the idea that O3 expanded with Longshan from the south is contradicted by the fact that the Longshan archaeological complexes, despite being mainly rice farming, does not match contemporary traditions found in southern China, but developed out of earlier archaeological complexes in northern China, such as Dawenkou and Yangshao".

As I said, there is no way O is ancient SE Asia or even South China.

"Given the data, I have to think that the primary O3e and O3* subclades found in Chinese today were already well established in northern China"

And probably originated there. That is the conclusion I came to years ago.

Ed said...

@ Lathdrinor responses to your 3 paragraphs

-You’re right, the switch to rice agriculture was probably secondary to the development of bronze tools and weapons actually.

-I never meant to say they recently came from southern China but initially. The Longshan culture was clearly not a continuation of the Yangshao, they were completely different. The Daxi are actually not a bad candidate. There were other O3 besides O3d present and the sample size was small. Perhaps the small sample size is to blame? There are also cultural similarities (wall building and rice farming for example). The Dawenkou is probably the best candidate. I’ve heard this mentioned more than a few times on Chinese History Forum and similar sites.

-After reading the new paper I’d say that 6kya isn’t too out of line with the Longshan culture and definitely not out of line with the Dawenkou (assuming they transitioned into the Longshan).

Va_Highlander said...

Ed:

"The prehistoric Yanghsao culture(5000BC – 3000BC) in the Yellow River valley was primarily Y-DNA N. This was stated in this paper and those I previously linked."

Could you be more specific? Perhaps I am just missing something, but I am not seeing this stated in anything you cite.

The paper under discussion states that the Yellow River valley culture was primarily O3-M122, citing Li, et al, to which you linked, above. Miaozigou is located somewhat east of the northeast corner of the Ordos Loop. While the Miaozigou culture is considered a northern expression of Late Yangshao, I do not see that this necessarily supports your claim that Yangshao was primarily Y-DNA N.

Ed said...

@terryt
Also the paper continues:
“However, the small sample size of our current ancient genetic material and the lack of data for earlier time periods means an alternate explanation [16], in which N1(xN1a, N1c) existed across the region prior to the Neolithic, is still a possibility.”

“three strong star-like Neolithic expansions at ~6 kya (thousand years ago) (assuming a constant substitution rate of 1e-9/bp/year) indicates that ~40% of modern Chinese are patrilineal descendants of only three super-grandfathers at that time. This observation suggests that the main patrilineal expansion in China occurred in the Neolithic Era and might be related to the development of agriculture”
A relatively recent expansion from N.China doesn’t disprove that O3 initially departed from S.China. Differences between architecture and culture of southern and northern Neolithic sites don’t disprove an initial southern origin of the northern sites. There was actually a lot of contact, trade and shared culture between the pre-Huaxia cultures and southern sites.


@VA_highlander
The archeological evidence matches the genetic evidence of Longshan replacing Yangshao, not being descended from Yangshao. Whether or not it is a ‘good enough’ sample is arguable, but it’s the best we have (that I’m aware of).

The Yellow River valley culture referred to is apparently the Longshan culture.

terryt said...

"The Yellow River valley culture referred to is apparently the Longshan culture".

Both the Chinese Neolithic and the Longshan developed in much the same place: almost certainly in the low hill country between, and north of, where the Yangtze and Yellow rivers empty onto the North China Plain.

"A relatively recent expansion from N.China doesn’t disprove that O3 initially departed from S.China".

Perhaps not, but you'd be clutching at straws to sustain a southern Chinese origin for the haplogroup, or even O as a whole. A synthesis of several papers I am aware of is revealing. It seems from the Dienekes' blog I linked to above that the 'O3*' that makes up two thirds of the Longshan Neolithic at Taosi, along with one third of the Wucheng and a smaller proportion of the Daxi, is almost certainly ISOGG's O3a1c1-P103. We can presume the very minor presence of its close relations outside the region of the early Chinese Neolithic can be explained as their having been carried along during that haplogroup's expansion.

"The Daxi are actually not a bad candidate".

The Daxi's one third O3d is ISOGG's O3a2b, Longshan O2a2c's 'brother'.
The two haplogroups most probably originated very near each other.

The remaining third of the Taosi sample (O3e in the study), and in that study found only there, is O3a2c1-M134. Two of the three expansions revealed in the above link are within that haplogroup. Both are obviously northern. O3a2c's relation O3a2b also appears in the Neolithic study, in the Daxi of the Three Gorges region of the Yangtze. Its origin is unlikely to have been further south than that river. That leaves no convincing examples of an O3 origin south of the Yangtze. O2a is also present in the Neolithic Daxi but makes up two thirds of the Wucheng sample further down the Yangtze. And let's not ignore O2b on the northern margin of the Chinese Neolithic. O1 is present in the Lower Yangtze Neolithic and it's expansion into the SE Asian mainland and islands is certainly Neolithic. What are we left with as evidence for a southern origin of Y-DNA O?

"Differences between architecture and culture of southern and northern Neolithic sites don’t disprove an initial southern origin of the northern sites".

The southern Neolithic is basically Hoabinhian, a southern invention. Its members were more like Australian Aboriginals or Papuans than like the population in the region today. The change towards a more Mongoloid phenotype began with the diffusion of a more northern-type Neolithic. That phenotype must have had some vector. The only possibility is Y-DNA O.

Ed said...

The Longshan falls within the Neolithic. The Dawenkou developed in the same area. I believe the safest bet is that the Longshan culture is an expansion of the Dawenkou culture.

I wouldn’t call it “clutching at straws”. In fact, every recent research paper I’ve read has supported the southern route. Even in the Shi Yan et al. paper you claimed supported a northern origin of O, the researchers conclude that O entered East Asia near Southern China.
http://arxiv.org/ftp/arxiv/papers/1310/1310.3897.pdf Page 24

-I actually agree that O3’s origin could be along the Yangtze.
-O definitely did not originate in ISEA, the recent Shi Yan et al. paper shows it originating in Southwest China

-I actually think the East Eurasian(“Mongoloid”) phenotype goes further back, possibly NO.
-Yes there are Papuan-like populations living in Southeast Asia even still today, most of them were/are hunter gatherers. There were also Y-DNA O, East Eurasian farming cultures present in Neolithic SEA and Southern China as well.


terryt said...

"Yes there are Papuan-like populations living in Southeast Asia even still today, most of them were/are hunter gatherers. There were also Y-DNA O, East Eurasian farming cultures present in Neolithic SEA and Southern China as well".

There is no evidence at all that the Hoabinhian people carried haplogroup O. Yes, the 'true' Neolithic SE Asian cultures almost certainly carried O, but they entered SE Asia from the north.

"I actually think the East Eurasian('Mongoloid') phenotype goes further back, possibly NO".

I doubt the Mongoloid phenotype goes back to NO. NO is part of MNOPS, including K, and is almost certainly 'Papuan-like' at its origin.

"the researchers conclude that O entered East Asia near Southern China".

I would say it is far more likely that it was NO that 'entered East Asia near Southern China'. It diversified into its four components somewhere near where the Chinese Neolithic developed. From where all four eventually expanded.

Ed said...

-That’s actually true regarding SEA. Vietnam probably didn’t have a farming culture related to East Asians until 4kya. Taiwan wasn’t colonized by East Asians until ~5kya as well. I actually see your point regarding the paleolithic presence of Y-DNA O in SEA.

-There are predominantly ‘mongoloid’ remains from ~60,000 years ago (Liujiang remains). NO seems to have split from K relatively early. I doubt very much that they developed side by side for very long.

-Yes, I’d say that is accurate.

terryt said...

"Vietnam probably didn’t have a farming culture related to East Asians until 4kya. Taiwan wasn’t colonized by East Asians until ~5kya as well. I actually see your point regarding the paleolithic presence of Y-DNA O in SEA".

Thanks. It's great to finally come across someone who can actually see that.

Va_Highlander said...

Ed:

"The archeological evidence matches the genetic evidence of Longshan replacing Yangshao, not being descended from Yangshao."

Do you have a citation for this assertion?

Also, I don't think you answered my question. Where, specifically, is it stated that Yangshao in the Yellow River Valley was primarily Y-DNA N?

Ed said...

Va_highlander:

-Evidence from their writing systems:
http://en.wikipedia.org/wiki/Neolithic_signs_in_China#Dawenkou
-Also if you look at where the Longshan 'center' was, it overlaps with Dawenkou not Yangshao.
-Yangshao culture was egalitarian while both Dawenkou and Longshan were warlike
-Dawenkou also cultivated rice like the Longshan


Like I said, it's the best sample for Yangshao we have. I would not be surprised if they were primarily N carriers.

Ed said...

@Va_Highlander

There is no shortage of evidence

- the 'center' of Longshan overlaps with Dawenkou, not Yangshao

writing system:
http://en.wikipedia.org/wiki/Neolithic_signs_in_China#Dawenkou

-both Dawenkou and Longshan were rice farmers

-both were warlike while Yangshao was egalitarian

-the best sample we have for Yangshao is Y-DNA N, Longshan is O3

I could go on.

Va_Highlander said...

Ed:

"There is no shortage of evidence."

I think not and you are hanging your assertion on a very slender reed. In order to prove your claim, among other things you would have to show cultural discontinuity on actual Yangshao sites and you are not doing that.

Yangshao society became increasingly differentiated through time, so claims that it was egalitarian and Longshan was not are meaningless.

And you have yet to demonstrate that Yangshao was Y-DNA N.

Lathdrinor said...

@Ed at the same time, there is also continuity between Yangshao and subsequent regional cultures. For example, the cultivation of millet continued without abate in north China during Longshan. So did silk worm cultivation, tripod pottery, and various other practices. I do agree that the Longshan horizon replaced Yangshao to a great extent. But in Yangshao centers, mixed Yangshao-Longshan - aka Longshanoid - cultures persisted. None of this obviates your scenario, but the biggest issue with explaining this in terms of the replacement of N by O3 is that the Yangshao regions are actually closer to the area of greatest O3 diversity - southwestern China - than Dawenkou. That, and the complete absence of genetic data from central and western China Yangshao sites.

Whatever the case, I think the idea that Yangshao was N primary is less obviously problematic than the idea that O3 came from the south during Longshan. You yourself stated it best in observing that the Longshan horizon overlaps with Dawenkou, which was in the north, not the south.

Ed said...

@va_highlander
-There is plenty of “N” in the north that isn’t as frequent there anymore
“Miaozigou, Central-South Inner Mongolia, Yangshao Culture, 5500 YBP, all N
http://img826.imageshack.us/img826/1599/ancientchineseydna.png
Not a perfect sample, but the best we have for now.
-Also if you look at those “All O” or “predominantly O” sites, they occur much later.

-I’ve already cited plenty of archeological evidence showing the cultural transition. Even the script is a direct descendant.
“The Dawenkou culture (4300-2500 BC) based in present-day Shandong province overlapped in time with Yangshao culture, and can be considered one of the precursors of the Longshan”
http://depts.washington.edu/chinaciv/archae/2dwkmain.htm



Ed said...


-Thank you, I hear that Longshan descends from Dawenkou so often that I assumed it was common knowledge. I didn’t realize there was still a debate. It’s entirely possible that it is a bad sample. It is on the outskirts of Yangshao territory after all.

-I don’t believe that O3 only entered North China during the Longshan. I think that the ancestors of the Longshan were initially rice farmers from somewhere in the South. The spread of rice was recently found to have a single origin in Southern China, they were from there at some point.