44 replies to this topic

[mongobanjum]

I don't believe that Mongolians are genetically closer to Han Chinese than to Koreans. Modern Y-haplogroup markers suggest that Koreans and Mongolians are more related than the relation between both Han Chinese and Mongolians and Japanese and Mongolians. About 15% of Y-haplogroup diversity of Koreans are haplogroup C3, while for Japanese and Han Chinese, the figures are much lower.

Also this study only used 200K SNP markers, while the human genome has more than 3 million. Typically, about 1.5 million SNPs are unique to an ethnicity. So this study is just a mere glimpse.

[LongMa]

mongobanjum:

You don't really need to see a resemblence in Mongolians and North Han. There can be gene flow in such a way that it is not fast enough to change the phenotype.

Here is an example:

A Korean man marries a Chinese Han woman that looks more like what he thinks is a "beautiful woman"...she is not necessarily attractive among the Han...but she had Han genes.

This guys kids, are raised as Mongols and they marry Mongols.

If most of the gene flow happened in this way...phenotype would not radically changed, there is "selection bias". All traits are not equally selected for and never have been.

[mongobanjum]

mongobanjum:

You don't really need to see a resemblence in Mongolians and North Han. There can be gene flow in such a way that it is not fast enough to change the phenotype.

Here is an example:

A Korean man marries a Chinese Han woman that looks more like what he thinks is a "beautiful woman"...she is not necessarily attractive among the Han...but she had Han genes.

This guys kids, are raised as Mongols and they marry Mongols.

If most of the gene flow happened in this way...phenotype would not radically changed, there is "selection bias". All traits are not equally selected for and never have been.

Well according to classical genetic markers (autosomal DNA), Koreans and Mongolians are very close ethnic groups. It's also true that about 15% of South Koreans have haplogroup C3, whereas Mongolians typically have it around 30-50%. Han Chinese and Japanese have it at much lower frequencies, yet Mongolians are closer to Han Chinese than to Koreans, according to this single study.

First of all, I think the sample size is too small for determining genetic distances. Also, it would have been interesting if more genetic markers were tested, since 200K is very limited in making any conclusions about the whole genome.

Edited by mongobanjum, 08 December 2008 - 08:10 AM.

[LongMa]

Well according to classical genetic markers (autosomal DNA), Koreans and Mongolians are very close ethnic groups. It's also true that about 15% of South Koreans have haplogroup C3, whereas Mongolians typically have it around 30-50%. Han Chinese and Japanese have it at much lower frequencies, yet Mongolians are closer to Han Chinese than to Koreans, according to this single study.

First of all, I think the sample size is too small for determining genetic distances. Also, it would have been interesting if more genetic markers were tested, since 200K is very limited in making any conclusions about the whole genome.

I'm sorry but you are confusing some issues. Allow me to explain...
1st problem:


C3 is OLD. Quite old. The estimates I have seen is that C3, although a rather recent Haplogroup is about 15-20,000 years old. If we are talking about 25 years per generation, we are talking about 800 generations! There are people in Eastern Europe who have C3 who are completely European in appearance and mostly European in DNA, but they had a distant ancestor from Asia (likely around the time of the Mongol/Turkic expansions).

E3b is even older and you can find holders of that Haplogroup who look like black Africans, Spaniards, or Amerindian. Origin does not necessarily have anything to do with current genetic affinity. E3b is thought to have originated in the MIddle East or even East Africa. There are obviously people with that haplogroup today who live in Mexico (for example) when given an autsomal DNA test, are more related to local INdian populations. C3 means you had one male ancestor who had that, it says nothing about who they married.
Older than Korean people...older than Mongol people.

So lower levels of a haplogroup have little to do with present day relationships.

Problem 2:

You are forgetting about the input of women. We are talking about autosomal testing...only 50% of DNA comes from men. Women are key in this, IMHO. You seem to be (correct me if I'm wrong) assuming that MtDNA and Y Chromsone haplogroups always track together, that is not so. History often shows men moving independent of women and marrying women outside their ethnicity, but often minimizing or killing off local male populations who are weaker. This creates gender bias and is the reason, everywhere in the world, Y Chromosone Haplogroups for a given population are less diverse than MtDNA...most of the genetic change to an ethnicity, usually comes from men aquiring women.

I would imagine most of the gene flow between CHinese and Mongols was Mongol men takign Chinese wives and not the other way around. Also you must keep in mind that Mongols are a much smaller population. It does not take a lot of Han input to influence Mongols. In this case, I would be looking at MtDNA not Y Chromosones. What I've seen in East Asia, is the father's family often determines the ethnicity of the children. I seriously doubt many Han men and many Mongol or Turkic men (in mass) move to each others nations (but under force of arms) and impregnated local women and then acculterated into each others culture. That did happen, but I doubt it happens at such a large amount to change the overall gene pool.

What is likely is Han men who lived along the border took Mongol and Turkic wives and Mongol men (during their expansion) took Han wives...from the North.

Over time you have situations, where a 1/4 Han/3/4 Mongol man is marrying a 1/4 Mongol 3/4 Han woman...etc. At this point the people are fairly related, but not perfectly so...this type of thing happens along border regions all the time.

We know during the Mongol invasion, many Han women were taken into slavery and sold...(not just Han either, but also Eastern European women, which explains the MtDNA diveristy of Mongols that I posted in another thread).




I'm not attempting to argue that Korean and Mongonls have a closer (in historical terms) common ancestry than Chinese or Koreans originated near Mongolia.

That is not the issue, that is quite a seperate argument.

THese old autsomal tests used how many markers???

Do you have the studies? I would like to see them. I believe what you say but I want to analyze it for more detail.

Edited by LongMa, 08 December 2008 - 08:58 AM.

[mongobanjum]

I'm sorry but you are confusing some issues.
1st problem:


C3 is OLD. Quite old. The estimates I have seen is that C3, although a rather recent Haplogroup is about 15-20,000 years old. If we are talking about 25 years per generation, we are talking about 800 generations! There are people in Eastern Europe who have C3 who are completely European in appearance and mostly European in DNA, but they had a distant ancestor from Asia (likely around the time of the Mongol/Turkic expansions).

E3b is even older and you can find holders of that Haplogroup who look like black Africans, Spaniards, or Amerindian. Origin does not necessarily have anything to do with current genetic affinity. E3b is thought to have originated in the MIddle East or even East Africa. There are obviously people with that haplogroup today who live in Mexico (for example) when given an autsomal DNA test, are more related to local INdian populations. C3 means you had one male ancestor who had that, it says nothing about who they married.
Older than Korean people...older than Mongol people.

So lower levels of a haplogroup have little to do with present day relationships.

Problem 2:

You are forgetting about the input of women. We are talking about autosomal testing...only 50% of DNA comes from men. Women are key in this, IMHO. You seem to be (correct me if I'm wrong) assuming that MtDNA and Y Chromsone haplogroups always track together, that is not so. History often shows men moving independent of women and marrying women outside their ethnicity, but often minimizing or killing off local male populations who are weaker. This creates gender bias and is the reason, everywhere in the world, Y Chromosone Haplogroups for a given population are less diverse than MtDNA...most of the genetic change to an ethnicity, usually comes from men aquiring women.

I would imagine most of the gene flow between CHinese and Mongols was Mongol men takign Chinese wives and not the other way around. Also you must keep in mind that Mongols are a much smaller population. It does not take a lot of Han input to influence Mongols. In this case, I would be looking at MtDNA not Y Chromosones. What I've seen in East Asia, is the father's family often determines the ethnicity of the children. I seriously doubt many Han men and many Mongol or Turkic men (in mass) move to each others nations (but under force of arms) and impregnated local women and then acculterated into each others culture. That did happen, but I doubt it happens at such a large amount to change the overall gene pool.

What is likely is Han men who lived along the border took Mongol and Turkic wives and Mongol men (during their expansion) took Han wives...from the North.

Over time you have situations, where a 1/4 Han/3/4 Mongol man is marrying a 1/4 Mongol 3/4 Han woman...etc. At this point the people are fairly related, but not perfectly so...this type of thing happens along border regions all the time.

We know during the Mongol invasion, many Han women were taken into slavery and sold...(not just Han either, but also Eastern European women, which explains the MtDNA diveristy of Mongols that I posted in another thread).




I'm not attempting to argue that Korean and Mongonls have a closer (in historical terms) common ancestry than Chinese or Koreans originated near Mongolia.

That is not the issue, that is quite a seperate argument.

THese old autsomal tests used how many markers???

Do you have the studies? I would like to see them. I believe what you say but I want to analyze it for more detail.

Mtdna wise, Koreans are probably the most distant from Chinese, even more than Japanese. I've read many articles about this. The reason why Koreans are less related to Mongols in that study is probably because some members of the CHB group had heavy Mongol/Tungusic influence, while other CHB had SE Asian influence, like the Vietnamese, Cambodians or Yi.

If you look at one Hezhe (Nanai) on the graph, it is very close to the Korean group, more so than to Japanese or CHB. Also, I would bet money that if the sample size was about 5000+, then Koreans and Tungusic/Mongolic people would overlap. You can just see, the Tungusic groups (specifically the Nanai and Xibe samples) are closer to Korean population than to Japanese/CHB.

Here is an autosomal DNA study

Origin of the Koreans: a population genetic study.

Saha N, Tay JS.

Department of Paediatrics, National University of Singapore.

A population genetic study was undertaken to investigate the origin of Koreans. Thirteen polymorphic and 7 monomorphic blood genetic markers (serum proteins and red cell enzymes) were studied in a group of 437 Koreans. Genetic distance analyses by both cluster and principal components models were performed between Koreans and eight other populations (Koreans in China, Japanese, Han Chinese, Mongolians, Zhuangs, Malays, Javanese, and Soviet Asians) on the basis of 47 alleles controlled by 15 polymorphic loci. A more detailed analysis using 65 alleles at 19 polymorphic loci was performed on six populations. Both analyses demonstrated genetic evidence of the origin of Koreans from the central Asian Mongolians. Further, the Koreans are more closely related to the Japanese and quite distant from the Chinese. The above evidence of the origin of Koreans fits well with the ethnohistoric account of the origin of Koreans and the Korean language. The minority Koreans in China also maintained their genetic identity.

Unfortunately you have to subscribe to get the actual genetic distance maps. But all the information should give you an idea.

Edited by mongobanjum, 08 December 2008 - 09:09 AM.

[mongobanjum]

I found a relatively recent study by Korean geneticists on bone genes (rather than blood genes which was studied in the study the OP posted).

http://www.pubmedcen...bmedid=18036257

http://www.pubmedcen...e...table&id=T3

The genetic distance table shows that:

Fst distance between (lower is closer) KOR= Koreans, JPT= Japanese, CHB= Chinese

1. KOR and CHB is 0.05259
2. KOR and JPT is 0.05775
3. JPT and CHB is 0.04840

The study also showed that both JPT and CHB genomes correspond highly to KOR genome. But overall genetic distance, JPT and CHB are closest.

For those who do not know what Fst means, here's a definition

FST is the proportion of the total genetic variance contained in a subpopulation (the S subscript) relative to the total genetic variance (the T subscript). Values can range from 0 to 1. High FST implies a considerable degree of differentiation among populations.

source: http://www.uwyo.edu/...opGenGloss.html

Korean samples were from Gyeonggi province, near Seoul.

The study didn't show a map, so you will have to read.

Results
We identified 942 variants, including 888 SNPs, 43 insertion/deletion polymorphisms, and 11 microsatellite markers. Of the SNPs, 557 (63%) had been previously identified and 331 (37%) were newly discovered in the Korean population. When compared SNPs in the Korean population with those in HapMap database, 1% (or less) of SNPs in the Japanese and Chinese subpopulations and 20% of those in Caucasian and African subpopulations were significantly differentiated from the Hardy-Weinberg expectations. In addition, an analysis of the genetic diversity showed that there were no significant differences among Korean, Han Chinese and Japanese populations, but African and Caucasian populations were significantly differentiated in selected genes. Nevertheless, in the detailed analysis of genetic properties, the LD and Haplotype block patterns among the five sub-populations were substantially different from one another.

MAYBE, this is why Koreans and Chinese don't look similar- because of high (FST >0.5) differentiation of genetic markers in the bone.

Plus, the study LongMa posted uses data from different sources, and hence the study only used supplementary data, so the results are not as reliable as it should be. Plus, as someone pointed out on: http://dienekes.blog...asia-using.html

http://www.plosone.o...ne.0003862.t001

I have the impression something is wrong with the Fst table. I don't see JPT's distance to any population other than CHB; and the authors seem to have computed the distance between a population and itself -- resulting in somewhat elevated distances. FIL's Fst distance to CHB is 0.0140, but its distance to itself is 0.0182. :S

by Blogger Rafinha, at Sunday, December 07, 2008 7:58:00 PM

the table is very flawed. How can Koreans to Koreans FST distance be 0.0028? I think that the authors might have made an error in displaying the data.

Edited by mongobanjum, 08 December 2008 - 10:41 AM.

[LongMa]

Mtdna wise, Koreans are probably the most distant from Chinese, even more than Japanese. I've read many articles about this. The reason why Koreans are less related to Mongols in that study is probably because some members of the CHB group had heavy Mongol/Tungusic influence, while other CHB had SE Asian influence, like the Vietnamese, Cambodians or Yi.

If you look at one Hezhe (Nanai) on the graph, it is very close to the Korean group, more so than to Japanese or CHB. Also, I would bet money that if the sample size was about 5000+, then Koreans and Tungusic/Mongolic people would overlap. You can just see, the Tungusic groups (specifically the Nanai and Xibe samples) are closer to Korean population than to Japanese/CHB.

Here is an autosomal DNA study


Unfortunately you have to subscribe to get the actual genetic distance maps. But all the information should give you an idea.

The Korean study did:

47 alleles controlled by 15 polymorphic loci. A more detailed analysis using 65 alleles at 19 polymorphic loci was performed on six populations.

This one has 200, not saying the old one was c**p, but higher resolution is always better.


As far as "Mongol/Tungusic influence,"

Yes, I agree, but those are Northern Han too. There is not "standard Han Chinese". The further South you go in China the more people trend to Southeast Asia, the further North the more they trend to Northeast Asia...history tells us why this is.

I commented about this on my blog awhile ago, it is interesting that Han Y Chromosone DNA is fairly uniform...most of this "variance" in Han is due to who Han men married. Still, we do know in history entire populations of "barbarian" (according to the Han) that were absorbed into the Han, this happened several times, especially in the North.

http://pmsol3.wordpr...mosome-testing/

Still, I admit you have a good point when you said this:

Also, I would bet money that if the sample size was about 5000+, then Koreans and Tungusic/Mongolic people would overlap.

I wonder would this depend where you sample from in Korea? I wonder is their a cline in Korea from North to South? I doubt it as Korea is quite small (even when you include the North, but you never know).


Going back to Han MtDNA and Mongol MtDNA...

Hum Genet.

Genetic imprint of the Mongol: signal from phylogeographic analysis of mitochondrial DNA

Cheng B, Tang W, He L, Dong Y, Lu J, Lei Y, Yu H, Zhang J, Xiao C.

Abstract

Mitochondrial deoxyribonucleic acid (DNA) from 201 unrelated Mongolian individuals in the three different regions was analyzed. The Mongolians took the dominant East Asian-specific haplogroups, and some European-prevalent haplogroups were detected. The East Asians-specific haplogroups distributed from east to west in decreasing frequencies, and the European-specific haplogroups distributed conversely. These genetic data suggest that the Mongolian empire played an important role in the maternal genetic admixture across Mongolians and even Central Asian populations, whereas the Silk Road might have contributed little in the admixture between the East Asians and the Europeans.

http://www.ncbi.nlm....pubmed/18769869

Yao YG, Kong QP, Bandelt HJ, Kivisild T, Zhang YP.
Laboratory of Molecular Evolution and Genome Diversity, Kunming Institute of Zoology, Chinese Academy of Sciences, Kunming, Yunnan, China.

To characterize the mitochondrial DNA (mtDNA) variation in Han Chinese from several provinces of China, we have sequenced the two hypervariable segments of the control region and the segment spanning nucleotide positions 10171-10659 of the coding region, and we have identified a number of specific coding-region mutations by direct sequencing or restriction-fragment-length-polymorphism tests. This allows us to define new haplogroups (clades of the mtDNA phylogeny) and to dissect the Han mtDNA pool on a phylogenetic basis, which is a prerequisite for any fine-grained phylogeographic analysis, the interpretation of ancient mtDNA, or future complete mtDNA sequencing efforts. Some of the haplogroups under study differ considerably in frequencies across different provinces. The southernmost provinces show more pronounced contrasts in their regional Han mtDNA pools than the central and northern provinces. These and other features of the geographical distribution of the mtDNA haplogroups observed in the Han Chinese make an initial Paleolithic colonization from south to north plausible but would suggest subsequent migration events in China that mainly proceeded from north to south and east to west. Lumping together all regional Han mtDNA pools into one fictive general mtDNA pool or choosing one or two regional Han populations to represent all Han Chinese is inappropriate for prehistoric considerations as well as for forensic purposes or medical disease studies.

http://www.ncbi.nlm....pubmed/11836649

So some of this depends on where you sample from in Mongolia and China.

I would like to see how close Inner Mongolian samples are to Northern CHina (MtDNA).

I'm not sure of any such study that looks at Mongol and Han DNA in clines and gives specific % of each haplogroup per each area.

My point is, we both could be right.

It depends on where you are sampling...China is just so d**** big..

[LongMa]

mongobanjum:

The bone study was a good find...that is very interesting. I wonder what caused the difference in selection or was it just random...?

[mongobanjum]

mongobanjum:

The bone study was a good find...that is very interesting. I wonder what caused the difference in selection or was it just random...?

It would be random. The study was a HapMap study. There would have been no point in a FST distance table if selection wasn't random.

I wonder would this depend where you sample from in Korea? I wonder is their a cline in Korea from North to South? I doubt it as Korea is quite small (even when you include the North, but you never know).

You can never know. Most genetic studies in Korea have been sampled in Daejeon, where most biotechnology research takes place. Unlike Chinese and Japanese (where there is a fairly large difference between north, central and south), most genetic studies use Koreans from one area. And don't say Japan has little genetic differentiation. If you look at my post on page 1, Kanto-Koshinetsu Japanese lean toward Han Chinese (some overlapping with Han Chinese) and Kyushu Japanese lean toward the Okinawan cluster.

Edited by mongobanjum, 08 December 2008 - 10:50 AM.

[LongMa]

Mongobanjum:

Kanto-Koshinetsu folks (some at least) overlap with Han Chinese or lean toward them? This takes us back to the other topic. Which Han?

There is a theory, based on lingustics more than anything, that Japanese (the Yayoi) came from Goguryeo or to a related group further West into Liaoning, which is part of China.

Xianbei, Goguryeo, Khitan and Jurchen (Manchu) ruled Liaoning...and most people there today consider themselves Han, and there has been some Han Chinese who have lived there since the Han Dynasty (maybe before)...

So to say that people in Central Japan would not be shocking. This does not mean Han colonized these areas or even immigrated here.

What it could easily mean is that a common ancestor to many present day Chinese Han and Japanese came to Japan.


Those two things are quite different.

[Polynesia]

Are you suggesting that Asia as a whole comes from seperate origins or that Asians
over time were isolated, because Southeast Asians not only are Chinese but are East Asians
as well, all of East Asia is the early migration North from Southeast Asia- the forming of China's
Culture itself cut these groups or migration north off from the Southeast of Asia, this is why you find Archaeolical
and Genetic markers from the South in the North , after the Formation of stronger Chinese Cultures this migration from
Southeast Asia along the Pacific Coast slowed to almost nothing after even Stronger Political borders were drawn.
All of Asia itself has it's orgins in the Southeast Asian Pacific along the Oceans border.

[Moonstone]

Ok, I made a mistake in that post.

1) I should have said Tohoku Japanese, not Hokkaido Japanese.

That would make a bit more sense, considering the study's graphs of the principal components. However, I would argue that the study's data do not support a hypothesis that the Tōhoku Japanese have been isolated; they rather suggest that the genetic constitution of the Japanese in Tōhoku reflects this region's geographical location within the island of Honshū, and, more generally, within Japan as a whole. Tōhoku is located at the northeastern end of Honshū and at the northeastern end of the territory that has long been inhabited by ethnic Japanese, so it is expected according to geography that the population of Tōhoku should have drifted to one extreme of the genetic cline within the Japanese population (Hokkaidō must be ignored for this sort of comparison, because the modern population of Hokkaidō is the result of intermixture of people from all different parts of Japan).

2) Kantō-Kōshin'etsu and Kinki Japanese represent your everyday run of the mill Japanese people. Central-southern Honshu is where majority of Japanese live in. I would understand why they would hold an intermediate position between Han Chinese and the central cluster, but do not understand why some would actually fit inside the Han Chinese cluster?

As I see it, there are only perhaps three individuals, all from the Kantō-Kōshin'etsu sample, who could be said to "fit inside the Han Chinese cluster." Kantō-Kōshin'etsu includes Tōkyō and Yokohama, which are the most populous cities of Japan and also relatively cosmopolitan. Yokohama is known for its expansive Chinatown.

Actually, mongobanjum's comment about these outliers has made me wonder about this study's sample sizes and sampling procedure. How many Japanese were sampled overall, and from each of the different regions? How were the individuals in each sample selected; was current place of residence the only criterion for inclusion in each of the regional samples? Did they take precautions to exclude non-ethnic Japanese? According to the year 2007 estimates, approximately 0.48% (606,889/127,433,494) of the population of Japan is Chinese, and approximately 0.47% (593,489/127,433,494) of the population of Japan is Korean. Note that these numbers do not include several hundred thousand "naturalized" Chinese and Koreans, i.e. those who have obtained Japanese citizenship, so the real number of "ethnic Chinese" and "ethnic Koreans" in Japan is even higher than the census estimates would suggest. If approximately 1% of the current population of Japan is ethnically Chinese or Korean, then we should expect about 3 individuals in a random sample of 300 individuals from Japan to be ethnically (and, presumably, genetically) Chinese or Korean. If it were a random sample of 600 individuals from Japan, then we should expect at least three ethnic Chinese and three ethnic Koreans to be included.

In any case, it is important to remember that these individuals are "outliers" precisely because they do not represent the majority of the Japanese population. It might be fun to speculate about the reasons for these outliers' deviation from the Japanese genetic cluster, but in the end, they are only a few individuals who might not even be ethnic Japanese.

3) OK, I think I made a mistake in that point. You can completely disregard what I said there. But I would have thought Kyushu Japanese would actually be closer to Han Chinese than the rest?

I think anyone who has actually lived in Japan would not expect Kyūshū Japanese to be closer to any continental Asian population. In my opinion, the most "Asian-looking" Japanese can be found in the western half of Honshū, especially in the Chūgoku region of westernmost Honshū. People in Kyūshū tend to be relatively "non-Asian-looking," or at least not Korean/Mongol/Northern Asian-looking; Kyūshū people are usually relatively dark-skinned, short, and have huge eyes. That's the impression I get from seeing them, anyway.

[mongobanjum]

As I see it, there are only perhaps three individuals, all from the Kantō-Kōshin'etsu sample, who could be said to "fit inside the Han Chinese cluster." Kantō-Kōshin'etsu includes Tōkyō and Yokohama, which are the most populous cities of Japan and also relatively cosmopolitan. Yokohama is known for its expansive Chinatown.

Actually, mongobanjum's comment about these outliers has made me wonder about this study's sample sizes and sampling procedure. How many Japanese were sampled overall, and from each of the different regions? How were the individuals in each sample selected; was current place of residence the only criterion for inclusion in each of the regional samples? Did they take precautions to exclude non-ethnic Japanese? According to the year 2007 estimates, approximately 0.48% (606,889/127,433,494) of the population of Japan is Chinese, and approximately 0.47% (593,489/127,433,494) of the population of Japan is Korean. Note that these numbers do not include several hundred thousand "naturalized" Chinese and Koreans, i.e. those who have obtained Japanese citizenship, so the real number of "ethnic Chinese" and "ethnic Koreans" in Japan is even higher than the census estimates would suggest. If approximately 1% of the current population of Japan is ethnically Chinese or Korean, then we should expect about 3 individuals in a random sample of 300 individuals from Japan to be ethnically (and, presumably, genetically) Chinese or Korean. If it were a random sample of 600 individuals from Japan, then we should expect at least three ethnic Chinese and three ethnic Koreans to be included.

In any case, it is important to remember that these individuals are "outliers" precisely because they do not represent the majority of the Japanese population. It might be fun to speculate about the reasons for these outliers' deviation from the Japanese genetic cluster, but in the end, they are only a few individuals who might not even be ethnic Japanese.

I think only ethnically Japanese people were sampled in the data. There were 7003 Japanese sampled, and the study says "Japanese individuals". So unless they were very careless, then I doubt that the data was affected by non-ethnic Japanese.

[Moonstone]

I think only ethnically Japanese people were sampled in the data. There were 7003 Japanese sampled, and the study says "Japanese individuals". So unless they were very careless, then I doubt that the data was affected by non-ethnic Japanese.

So, the total number of individuals sampled from Japan is 7003. Thanks, mongobanjum.

However, I must disagree with you in regard to the meaning of the phrase "Japanese individuals"; in English, this phrase is ambiguous between a less inclusive sense of "ethnic Japanese individuals" and a more inclusive sense of "individuals who are inhabitants of the State of Japan." Do the authors of the study provide any description of their sampling procedure that would allow us to determine whether they checked the ethnic background of each of the sampled individuals? Considering the huge number of sampled individuals (7003), it is possible that they have not gone to the trouble of performing background checks.

Just as a reminder, if one has collected a random sample of 7003 inhabitants of Japan, one should expect the sample to contain (at least) approximately 35 individuals of Chinese origin and approximately 35 individuals of Korean origin.

[mongobanjum]

So, the total number of individuals sampled from Japan is 7003. Thanks, mongobanjum.

However, I must disagree with you in regard to the meaning of the phrase "Japanese individuals"; in English, this phrase is ambiguous between a less inclusive sense of "ethnic Japanese individuals" and a more inclusive sense of "individuals who are inhabitants of the State of Japan." Do the authors of the study provide any description of their sampling procedure that would allow us to determine whether they checked the ethnic background of each of the sampled individuals? Considering the huge number of sampled individuals (7003), it is possible that they have not gone to the trouble of performing background checks.

Just as a reminder, if one has collected a random sample of 7003 inhabitants of Japan, one should expect the sample to contain (at least) approximately 35 individuals of Chinese origin and approximately 35 individuals of Korean origin.

If there should be 35 individuals of Chinese origin, then shouldn't there be at least 35 Japanese dots on the Han Chinese cluster? I think the results are just a random variation of Japanese population rather than the few outliers being actual Han Chinese citizens in Japan. With a huge sample size as 7003, one would expect some outliers and even some who overlap other clusters, even if the people are ethnically Japanese.

Edited by mongobanjum, 08 December 2008 - 07:47 PM.