Ancient mitochondrial DNA sequences of Jomon teeth samples from Sanganji, Tohoku district, Japan
Hideaki KANZAWA-KIRIYAMA et al.
Abstract We investigated mitochondrial DNA haplogroups of four Jomon individuals from the Sanganji shell mound in Fukushima, Tohoku district, Japan. Partial nucleotide sequences of the coding and control region of mitochondrial DNA were determined. The success rate of sequencing increased when we analyzed short DNA sequences. We identified haplogroups from all four samples that were analyzed; haplogroup frequencies were 50% (n = 2) for N9b and 50% (n = 2) for M7a2. Haplogroup N9b has been previously observed in high frequencies in the other Tohoku Jomon, Hokkaido Jomon, Okhotsk, and Ainu peoples, whereas its frequency was reported to be low in the Kanto Jomon and the modern mainland Japanese. Sub-haplogroup M7a2 has previously been reported in the Hokkaido Jomon, Okhotsk, and modern Udegey (southern Siberia) peoples, but not in the Kanto Jomon, Ainu, or Ryukyuan peoples. Principal component analysis and phylogenetic network analysis revealed that, based on haplogroup frequencies, the Tohoku Jomon was genetically closer to the Hokkaido Jomon and Udegey people, than to the Kanto Jomon or mainland modern Japanese. The available evidence suggests genetic differences between the Tohoku and Kanto regions in the Jomon period, and greater genetic similarity between the Tohoku Jomon and the other investigated ancient (Hokkaido Jomon, Okhotsk) and modern (Siberian, Udegey in particular) populations. At the same time, the Tohoku and Hokkaido Jomon seem to differ in sub-haplotype representations, suggesting complexity in Jomon population structure and history.
Link
"the Tohoku and Hokkaido Jomon seem to differ in sub-haplotype representations, suggesting complexity in Jomon population structure and history".
ReplyDeleteI would be far more surprised to learn that the Jomon were a homogeneous population. They must have developed over a considerable period and be the product of numerous migrations into Japan.
I find it strange that the Southern Chinese and Taiwanese Aboriginies are closer to Udegey (northern Asians) than Koreans are.
ReplyDeleteAnother study that shows that mtDNA haplogroups are not a good indicator of genetic distance.
Udegey seem to have very particular mtdna and ydna composition:
ReplyDeletemtdna
N9 30%
M* 29%
C2 17%
CSZ*(S) 15%
F 8.7%
Ydna:
C3c 60 %
M9 25%
Q 10%
R1a 5%
On the East Coast of China men carry quite much ydna C3 and mtdna M9E and F are very common in Taiwanese Aboriginals.
The persistence of non-farmer Jomon genetics in the modern Japanese population through the Neolithic revolution there continues to be remarkable and almost unprecedented to that degree seen in Japan.
ReplyDeleteWhat is the Y-DNA M9? Or is it a misprint for mt-DNA M9? Y-DNA M is New Guinea, with no M9.
ReplyDeleteSorry. Y-DNA M9 is obviously K(xIJ), probably O of some sort.
ReplyDeleteI do not know the original sources of this study's Taiwanese Aboriginals and Udegey data sets, but the data I have on Taiwanese aborigines is nothing like the Udegey data that Kristiina has posted above.
ReplyDeleteTaiwan aborigines (Peng et al. 2011)
100/640 = 15.63% B4a
47/640 = 7.34% B4b1 (This is a sister clade to the American type of Haplogroup B.)
25/640 = 3.91% B4c1b
172/640 = 26.88% B4 total
33/640 = 5.16% B5a
8/640 = 1.25% D4
26/640 = 4.06% D5'6
34/640 = 5.31% D total
77/640 = 12.03% E
14/640 = 2.19% F1a(xF1a1)
21/640 = 3.28% F1a1(xF1a1a)
1/640 = 0.16% F2
54/640 = 8.44% F3
72/640 = 11.25% F4
162/640 = 25.31% F total
1/640 = 0.16% G
2/640 = 0.31% M12
4/640 = 0.63% M10
74/640 = 11.56% M7b(xM7b1)
5/640 = 0.78% M7b1
42/640 = 6.56% M7c1'2
121/640 = 18.91% M7 total
10/640 = 1.56% N9a
9/640 = 1.41% Y
19/640 = 2.97% N9 total
15/640 = 2.34% R9c
The frequency of F is three times greater in Taiwan aborigines than in Udegeys, the frequency of N9 is ten times greater in Udegeys than in Taiwan aborigines, and Taiwan aborigines completely lack the M8 clade, which includes C and Z and seems to be the modal haplogroup among the Udegey.
On the Y-DNA side of things, Taiwan aborigines are heavily O1a-M119, with a few particular subclades of O3-M122 or O2a1-M95 also represented in some tribes. Taiwan aborigines completely lack the typically Siberian C3c and Q lineages that predominate among the Udegey, and they also do not exhibit R1a, which may reflect limited geneflow from the Russian majority population into the indigenous Udegey minority.
Frankly, the mtDNA pool of the Udegey looks quite typically Siberian, with a definite majority of M (and especially M8) subclades accompanied by a small amount of F. N9 is rare in most of Siberia, being generally a more southerly East Asian group, but it is quite common (especially in the form of its Y subclade) in populations around the Sea of Okhotsk in the extreme east of Siberia, who are close to the Udegey geographically and culturally. (This is similar to the distribution of mtDNA haplogroup G, which is also generally rare in Siberia and more typically East/Central Asian, but with a branch extending up the eastern fringe of Siberia into Kamchatka.)
In contrast, the mtDNA pool of Taiwan aborigines is very typical of Southeast Asian populations, with an overwhelming majority of R subclades (especially B and F), and representatives of mtDNA macrohaplogroup M being mostly limited to M7b/c (branches of M7, another branch of which is common in Japan) and E (a branch of M9, another branch of which is common in Tibet).
I cannot explain why the Taiwan aborigines and the Udegey have been placed so close together on this study's PC plot without access to their raw data.
"I cannot explain why the Taiwan aborigines and the Udegey have been placed so close together on this study's PC plot"
ReplyDeleteThanks for that information. I was more than just a little surprised at the connection too.
Terry, M9 must be just K. Do not know more precisely as I only copied the values.
ReplyDeleteTo me Udegey mtDNA pool is not so typically Siberian or Korean. They lack mtDNA haplogroups D, G and A and yDNA N and O. The lack of D is very unusual, as D is extremely widespread in Northern East-Asia, including Korea. Their CSZ*(S) is specific to Udegey. I do not really say that Udegey and Taiwanese aboriginals have the same mtDNA pool. I just find some interesting lacks (G, A, M10, and Taiwanese Aboriginals have a low frequency of D haplogroups) and similarities (presence of N9, F) which may explain why the Udegey are placed closer to Taiwanese than Koreans. Mtdna haplogroups frequencies of Udegey were taken from Starikovskaya et al study. According to this paper, one of Udegey M* lines has been detected in the Vietnamese and Taiwanese Han, and the other M* line has similarities with a M line detected in China. Taiwanese aboriginals also have a considerable amount of M7 which was found in this ancient Jomon sample.
Kristiina wrote,
ReplyDelete"Terry, M9 must be just K. Do not know more precisely as I only copied the values."
The "M9" (i.e. K-M9(xQ, R1a, ...?)) category is almost certainly not "just K," but rather a hodgepodge of derivatives of N-M231 and O-M175. Many researchers, unfortunately, continue to provide readers with very shallow phylogenetic resolution of their samples. However, in the case of the Udegeys, at least one other study has demonstrated that some of them possess Y-DNA derived from O-M175, such as O2b-SRY465.
Kristiina wrote,
"To me Udegey mtDNA pool is not so typically Siberian or Korean. They lack mtDNA haplogroups D, G and A and yDNA N and O. The lack of D is very unusual, as D is extremely widespread in Northern East-Asia, including Korea. Their CSZ*(S) is specific to Udegey. I do not really say that Udegey and Taiwanese aboriginals have the same mtDNA pool. I just find some interesting lacks (G, A, M10, and Taiwanese Aboriginals have a low frequency of D haplogroups) and similarities (presence of N9, F) which may explain why the Udegey are placed closer to Taiwanese than Koreans. Mtdna haplogroups frequencies of Udegey were taken from Starikovskaya et al study. According to this paper, one of Udegey M* lines has been detected in the Vietnamese and Taiwanese Han, and the other M* line has similarities with a M line detected in China. Taiwanese aboriginals also have a considerable amount of M7 which was found in this ancient Jomon sample."
Would you mind sharing the title of that study by Starikovskaya et al. with us? Are you sure that the "M* 29%" figure for this study's sample of Udeghe explicitly excludes haplogroup D and/or haplogroup G, and not only some subclades (including that allegedly Udeghe-specific type) of M8CZ? It is difficult for me to imagine that the Udeghe population could entirely lack mtDNA haplogroups D and G; G is very common in other populations of the greater circum-Okhotsk region that also share haplogroup N9Y with the Udeghe, and haplogroup D is generally the second most common mtDNA haplogroup in North Asia after haplogroup C. Please keep in mind that the Udeghe are not isolated in any way; they are just one of many populations of Tungusic speakers in the general region, whose language is quite cosmopolitan for a Tungusic language, sharing some features with the dialects of the more northerly Evenks and other features with the dialects of the other Amurians. They do not even live on an island or in mountains, but rather along rivers in the lower watershed of the Amur.
As for the high frequency of M7 among the aborigines of Taiwan, it is completely attributable to the typically South Chinese/Southeast Asian subclades M7b and M7c. The M7 that is often brought up in discussions about the ethnogenesis of populations of the Japanese Archipelago actually refers to the M7a subclade, which has been found almost exclusively in ancient remains from the Japanese Archipelago (e.g. Jōmon) and in modern Japanese, Ryūkyūans, Ainus, and Koreans.
Ebizur, please, go and see the study yourself (Table 2). I am not making up things. http://onlinelibrary.wiley.com/doi/10.1046/j.1529-8817.2003.00127.x/pdf
ReplyDeleteIf you are aware of another study where the Udegey have been sampled, please let me know.
The same applies to this ydna study. Please go and check what kind of M9 it is. I would also like to know!
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC384887/
I agree that their M9 is not at all very clear, but this M9 is anyway absent in Buryat and Yenisey Evenks, so it should not be O line, as Buryats have O2b and O3. They also say that ”The Tat-C haplogroup was absent in the Lower Amur and Sea of Okhotsk region populations that have maintained greater geographic and/or linguistic isolation (e.g., the Udegeys, Nivkhs, and Upriver Negidals) and was only detected in the populations likely to have had recent contact or shared origins with the populations of southern Middle Siberia (e.g., the Okhotsk Evenks, Ulchi/Nanai, and Downriver Negidals)”. M9 could be upstream N3 but I am not aware of N1a or N1b or N1 LLY22g being found in Amur area.
Ebizur, would you kindly like to share the link (or the name) of the study where O2b is found in Udegey.
"the mtDNA pool of Taiwan aborigines is very typical of Southeast Asian populations, with an overwhelming majority of R subclades (especially B and F)"
ReplyDeleteIndeed. So much so that I am certain that research will eventually reveal that mt-DNA R developed from N in Southern Wallacea, specifically in the triangle formed by Sumba, Timor, Alor and Flores. R22's root is claimed to lie in the Lombok/Alor region, although it has been found elsewhere in Malaysia and Indonesia. R14's root is also most likely in the triangle, although it has been found as far away as the Nicobar Islands in the west and New Guinea in the east. R23 straddles the Wallace Line to the west of the triangle. To me it seems obvious that P developed to the east of the island triangle: in Australia. Both R9/F and the haplogroup collection making up the R11'B6, B4'5 and R24 group obviously developed northwest of the Wallace Line, as did all the other R haplogroups.
"Terry, M9 must be just K. Do not know more precisely as I only copied the values".
I don't think so. K is unlikely so far north whereas N and/or O are extremely likely, and not listed as being present. Y-DNA M9 includes both these haplogroups.
"Ebizur, would you kindly like to share the link (or the name) of the study where O2b is found in Udegey".
At this stage I don't require a reference to accept that this is an extremely likely find.
In my first post, I mixed up lines F and Y and wrote that Udegey have 8.7% of F, but it is in fact Y, descendant of N9. So, what Taiwanese aboriginals and Udegey have in common, is not what they share but what they lack.
ReplyDeleteAnyway, in the Starikovskaya et al paper they say that “branching patterns of the Lower Amur and adjacent Okhotsk Sea coast/Kamchatkan populations, relative to those located in Western and Middle Siberia, may reflect relatively recent demic expansions which have brought haplogroups G, S, Y and Z into the Sea of Okhotsk/ Kamchatka region. The observed geographical distribution of these haplogroups (tables 1 and 2) would be expected from anthropological, archaeological and linguistic evidence indicating the expansion of the Altaic-speakers from the Lower Amur/Manchuria region during the Neolithic.”
Then there was another incongruity in my previous mail. I should have used according to the new nomenclature N1c and not N3.
Thank you for the link, Kristiina.
ReplyDeleteFirst, I would like to correct a slight error in your previous tabulation of the Udegey mtDNA data from Starikovskaya et al. 2005. According to this study's Table 2, the authors' sample of Udegey mtDNA contained 14/46 = 30.4% N9(xY) and 4/46 = 8.7% Y for a total of 18/46 = 39.1% N9. Y was found in the following samples besides the Udegeys: 37/56 = 66.1% Nivkhi, 33/87 = 37.9% Ulchi, 7/33 = 21.2% Negidal, 15/155 = 9.7% Koriak, 2/47 = 4.3% Itelmen. This is consistent with other studies' findings of high frequencies of mtDNA haplogroup Y in populations surrounding the Sea of Okhotsk, especially in the vicinity of its southwestern shores (i.e. Tungusic peoples of the lower Amur region, Nivkhs, and Ainus). Haplogroup N9(xY), which tends to be a more typical East/Southeast Asian mtDNA haplogroup, was found with high frequency only in the Udegey sample, but it was also detected in the samples of Tubalars (5/72 = 6.9%), Ulchi (6/87 = 6.9%), and Tuvans (3/95 = 3.2%). They have reported finding mtDNA haplogroup F, a typically Chinese/Vietnamese haplogroup, mainly in the Yeniseyan Kets (9/38 = 23.7% F) and the Turkic Tofalars (4/46 = 8.7% F) and Tuvans (4/95 = 4.2% F), with singletons found in their samples of Evenki (1/71), Tubalars (1/72), Ulchi (1/87), and Mansi (1/98). MtDNA haplogroup F was absent from the study's samples of Nganasans (0/24), Buryats (0/25), Negidals (0/33), Udegey (0/46), Itelmens (0/47), Nivkhi (0/56), Chukchi (0/66), Eskimos (0/79), and Koriaks (0/155).
Anyway, you are correct that the Udegeys do stand out in the Siberian genetic landscape for their high frequencies of N9(xY) and M(xM8CZ, D, G), which are generally much more common in East/Southeast Asia than in North Asia. However, despite this vague "East Asian" tendency, the mtDNA pool of the Udegeys according to this study is still nowhere close to that of the Taiwan aborigines, as the Taiwan aborigines exhibit a typically Southeast Asian mtDNA pool dominated by R (B & F & R9) and complemented by subclades of M7 and M9, whereas this sample of Udegeys is completely devoid of any representative of haplogroup R.
The following are the only data I have regarding Udegey Y-DNA:
ReplyDeleteLell et al. 2002
Udegey
12/20 = 60% C3c-M48
5/20 = 25% K-M9(xO1a-M119, N1c-Tat, P-M45) [It now appears that these should turn out to be O2b* or O3.]
2/20 = 10% P-M45(xR1a1a-M17, Q1a3a-M3)
1/20 = 5% R1a1a-M17
Jin, Han-Jun; Kim, Ki-Cheol; Kim, Wook (2009). "Genetic diversity of two haploid markers in the Udegey population from southeastern Siberia," American Journal of Physical Anthropology. doi:10.1002/ajpa.21232
14/21 = 66.7% C-RPS4Y711
2/21 = 9.5% K-M9(xNO-M214)
2/21 = 9.5% O2b-SRY465(xO2b1-DXYS5Y)
1/21 = 4.8% O3-M122(xLINE1)
1/21 = 4.8% O3-M122(+LINE1)
1/21 = 4.8% Y*(xC, DE, K)
I would like to know more precisely the classification of the 2/20 = 10% P-M45(xR1a1a-M17, Q1a3a-M3) in the Udegey sample of Lell et al. 2002 and the 2/21 = 9.5% K-M9(xNO-M214) in the Udegey sample of Jin et al. 2009. Is it really some sort of haplogroup Q, as Kristiina seems to have assumed, or is it rather some type of R (R1b, R2, etc.)? Both R1b and R2 have been found in some Buryats.
In any case, the Y-DNA pool of the Udegeys seems to share an extremely high frequency of C3c-M48 with Northern Tungusic (Evenk, Even, Negidal) populations of North Asia, but the Udegeys also exhibit O3-M122 and O2b-SRY465, which is similar to Koreans and Southern Tungusic peoples of Manchuria (Manchus, Solon Evenks, Hezhes).
"what Taiwanese aboriginals and Udegey have in common, is not what they share but what they lack".
ReplyDeleteBut that cannot be used to explain any genetic connection. Taiwanese and Europeans have very few haplogroups in common for example. I see no surprises in the data for either Taiwanese aboriginals or Udegey. Unless Udegey do actually have a complete absence of either N or O Y-DNA.
If an analysis shows an unexpected result, I think that it is only normal to try to find an explanation. So, it is a fundamental thing to look into the haplogroup distribution of that particular group, and that is what we have done. I think that picture is now much clearer to everybody.
ReplyDeleteThat M9 in the Lell study should not be O3 as they specifically say that ”Interestingly, none of the Y chromosomes analyzed in the present study contained the M122 marker, which has been shown to represent a large proportion of northern Chinese and East Asian Y chromosomes”. So O2b-SRY465(xO2b1-DXYS5Y) seems to be the most plausible option. In that case, O2b (and O2b1) would be a clear link between Koreans and Japanese, on one hand, and Udegey, Nivkh, Ulchi, on the other, if this M9 is now O2b in all these Okhotsk groups.
It seems that O2b-SRY465(xO2b1-DXYS5Y) is not very widespread in Siberia. Daur have it, and they also have O2a-M95(xO2a1-M88), but Mongols have instead O2*-P31(xO2a-M95, O2b-SRY+465). This O2*-P31(xO2a-M95, O2b-SRY465) is detected in outer and inner Mongolians and Daur.
In a study on Nivkh ydna (Tajima et al.) they have used the formula 28.6% of O-AS1(xO3-M122, O1a-M119).
Moreover, I think that there is surprisingly much O1-M119 in the easternmost populations of the Northeast: Daur 5.1% O1a-M119 (Xue et al. 2006), Yenisey Evenks 3.2% O1a-M119 (Jeffrey et al, 2001), Ulchi/Nanai 3.8% O1a-M119 (Jeffrey et al, 2001) and Nivkh 5.9% O1a-M119 (Jeffrey et al, 2001).
That P-M45(xR1a1a-M17, Q1a3a-M3) could, in fact, be R1b or R1a, if Udegey have some western mtdna haplogroups as it seems on the basis of this summary (http://onlinelibrary.wiley.com/doi/10.1002/ajpa.21232/abstract).
Udegey seem to differ from other Siberians in that they lack ydna N and mtdna D, G and A, and if their P-M45 is only R, they lack also Q. As O3 is detected only in Han-Jun et al. study, it is possible that it is quite recent in Udegey. The Japanese seem to consider that O3a lines are quite recent:
Early Yayoi Intra-Mainland Migration - Circa 800-400 BCE (Northern Kyushu)
O3a4-LINE1 3.10% Sanmiao (Dongye-Huiji) Jar Borial Lower Dolmen Lelang Sea Folk Wo
O2b1-47z 22.00% Wo (Songunni-Ongagawa)
Middle Yayoi Immigration - Circa 400-1 BCE (North Kyushu)
O3-M122* 6.60% Jizi Emigrant Hebei Farmers Higher Dolmen Bronze Dagger
Tumuli Period Circa 369 AD
(Yuezhi-Silla Immigration Ojin aka "Hormusta" Dynasty Yamato-Baekje Pact/Seven Branch Sword)
(Purported Equestrian Conquerers' Dynasty)
O3a3c-M134 10.40% Qi-Qins Emigre
I found also this comment on Internet ”previous O3-LINE1 among Miao and Yao people are proven mostly O3a3b-M7, but in other populations, ex: previous O3-LINE1 among Taiwan Han people are now mostly O3a3-P201 and O3a4-002611. I tend to suppose O3-LINE1 in Koreans and Japanese are O3a3-P201 and O3a4-002611”.
I still think that Udegey are not typical Siberians or extreme North-Easterners.
I got acces to the data in “Genetic Diversity of Two Haploid Markers in the Udegey Population From Southeastern Siberia” (Jin et al. 2009). It seems that they share a strange Y*(xC, DE, K) line at a frequency of 1/21 = 4.8% with at least the following groups: Beijing Han (the highest frequency), Koreans, Buryats, Khalk, Yunnan, Philippines, Thais. They also share with the same groups K-M9 (xO) that should not be R or Q as they say-that “in addition, unlike on the maternal side, very little contribution from European Y chromosome haplogroups was observed in the Udegeys studied here, although non-differentiated Y chromosomes (e.g., haplogroups Y* and K-M9) need further analysis of their classification. It is unusual to find European mtDNA input, and not European Y input in the Udegey population, since male-biased Europeangene flow is much more common.” According to Jin paper, Udegey have some D4c, but this clade seems to be, in fact, typical of East-Asians and not of Northeast-Asians and Siberians who carry other D4 clades, for example D4e and D4j.
ReplyDeleteOn the origin of O2-SRY465 they say that “In the median-joining network, the haplogroup O2-SRY465 chromosomes of the Udegeys are also connected to the East Asian populations (e.g., Koreans and Manchurians). This finding of the occurrence of haplogroup O2-SRY465 chromosomes in the Udegeys may be considered concordant with the historical views about the early connection between ancient Udegeys and northeast Asians (e.g., Qing Dynasty of China and the Ancient Koguryo/Parhae States of Korea).”
To me the original Udegey group consisted of ydna C3c-M48 and mtdna N9. According to Zhong study (Zhong et al. 2010), the age of STR variation of C3c is 10.8kya and the divergence time 9.3kya, and the highest frequencies are in Oroquens, Ewenki and Mongolians of Inner Mongolia, Manchu and Hezhe of Heilongjiang and Kyrgyz of Xinjiang. They also say that “We believe that M48 originated in NEAS populations, which agrees well with the suggested recent migration (for example, the Mongol expansion) of M48-derived individuals into Central Asia and Siberia”. On C3 they say that “The Han Chinese display a high STR diversity, especially those in the eastern coastal region (0.467) as well as other eastern populations (Korean,0.463; Japanese, 0.453), whereas populations in the north and west show low diversities (Altaic, 0.281; Tibetan, 0.366). Therefore, the distribution and gene diversities of the M217-derived lineages support a single eastward migration through the southern route and the subsequent northward migration of Hg C along the coastline of mainland East Asia in prehistory. The evidence from dental morphological
traits pointed to the same direction.”
"This O2*-P31(xO2a-M95, O2b-SRY465) is detected in outer and inner Mongolians and Daur".
ReplyDeleteThat is extrememly interesting. I don't think O2*-P31(xO2a-M95, O2b-SRY465) has been found anywhere in the south so this finding indicates a northern origin for O2 as a whole, as I have long suspected.
"O3a3c-M134 10.40% Qi-Qins Emigre"
O3-M134 is now O3a2c1. This haplogroup is strongly associated with Sino-Tibetan speakers although it is spread from Tibet to Japan.
"previous O3-LINE1 among Miao and Yao people are proven mostly O3a3b-M7, but in other populations, ex: previous O3-LINE1 among Taiwan Han people are now mostly O3a3-P201 and O3a4-002611".
O3-M7 is now O3a2b, predominantly a Hmong-Mien/Mon-Kmer haplogroup (ie. south of the Yangtze). O3a3-P201 is now basal O3a2, widespread through East Asia but especially in the Han. O3a4-002611 is O3a1c, also high in China, especially south of the Yangtze.
Tha above would suggest that, apart from O2 and C3, the Udegey are partly the product of a northward movement from within China. So they may have some K-M9(xNOPS).
Kristiina wrote,
ReplyDelete"That M9 in the Lell study should not be O3 as they specifically say that ”Interestingly, none of the Y chromosomes analyzed in the present study contained the M122 marker, which has been shown to represent a large proportion of northern Chinese and East Asian Y chromosomes”. So O2b-SRY465(xO2b1-DXYS5Y) seems to be the most plausible option. In that case, O2b (and O2b1) would be a clear link between Koreans and Japanese, on one hand, and Udegey, Nivkh, Ulchi, on the other, if this M9 is now O2b in all these Okhotsk groups."
Nice find regarding the lack of M122 in the samples of Lell et al. 2002. That means that their "M9" category is actually K-M9(xN1c1-Tat, O1a-M119, O3-M122, P-M45).
They have reported detecting such Y-DNA in their samples of Udegey (5/20 = 25%), Ulchi/Nanai (8/53 = 15.1%), and Nivkh (2/17 = 11.8%) from the lower Amur region, and Brib[r]i/Cabecar from Costa Rica (1/13 = 7.7%). The Brib[r]i/Cabecar sample from Costa Rica also contains some sort of apparent African admixture (Y*(xDE-YAP, C-RPS4Y, F-M89)), so I would guess that the K-M9(xN1c1-Tat, O1a-M119, O3-M122, P-M45) Y-DNA in their case is probably T1a-M70 from Southern Europe or vicinity.
In the case of the indigenous populations of the lower Amur region, K-M9(xN1c1-Tat, O1a-M119, O3-M122, P-M45) could be any sort of N-M231(xN1c1-Tat), such as N1c2a-M128 (formerly N1 > N1a), N1c2b-P43 (formerly N2 > N1b), or any of the diverse varieties of N*. It could be any sort of O2-P31, such as O2a1-M95, O2b-SRY465, or O2*-P31. There is even an off chance that some or all of it could be L-M20, which has been detected in some nearby populations (e.g. Koreans).
Pending further information, I bet that it should turn out to be some assortment of N-M231(xN1c1-Tat) and O2b-SRY465, with perhaps an occasional instance of O2*-P31.
Consider the Hezhe (PRC Nanai) data of Xue et al. 2006 for comparison:
Hezhe (Xue et al. 2006)
3/45 = 6.7% C-RPS4Y(xC1-M8, C3-M217)
5/45 = 11.1% C3-M217(xC3c-M48)
5/45 = 11.1% C3c-M48
1/45 = 2.2% NO-M214(xN1-LLY22g, O-M175)
8/45 = 17.8% N1c2b-P43
1/45 = 2.2% O2-P31(xO2a1-M95, O2b-SRY465)
2/45 = 4.4% O2b-SRY465(xO2b1a-47z)
7/45 = 15.6% O3-M122(xO3a2a-M159, O3a2b-M7, O3a2c1-M134) [LINE1 -]
4/45 = 8.9% O3-M122(xO3a2a-M159, O3a2b-M7, O3a2c1-M134) [LINE1 +]
2/45 = 4.4% O3a2c1-M134(xO3a2c1a-M117)
7/45 = 15.6% O3a2c1a-M117
12/45 = 26.7% K-M9(xN1c1-Tat, O1a-M119, O3-M122, P-M45) [8 N1c2b-P43 + 2 O2b-SRY465(xO2b1a-47z) + 1 NO-M214(xN1-LLY22g, O-M175) + 1 O2-P31(xO2a1-M95, O2b-SRY465)]
Even with the extremely high frequency of O3-M122 (20/45 = 44.4%) in this sample of Nanais from China, the types equivalent to the K-M9(xN1c1-Tat, O1a-M119, O3-M122, P-M45) category of Lell et al. 2002 are still found in about one quarter of the sample.
Kristiina wrote,
"In a study on Nivkh ydna (Tajima et al.) they have used the formula 28.6% of O-AS1(xO3-M122, O1a-M119)."
The O-AS1 category of Tajima et al. 2004 seems to be equivalent to NO-M214 of the present ISOGG tree. That would explain why "O-AS1" was found in all this study's samples, including 6/61 = 9.8% "Buryat" and 11/49 = 22.4% "Northern Han," whereas K-M9(xAS1, P-P27) was found only in "Philippine (Tagalog language group)" (5/50 = 10.0%) and "ordinary Malay near Kuala Lumpur" (1/12 = 8.3%).
Sorry for the double post.
ReplyDeleteKristiina wrote,
"That P-M45(xR1a1a-M17, Q1a3a-M3) could, in fact, be R1b or R1a, if Udegey have some western mtdna haplogroups as it seems on the basis of this summary (http://onlinelibrary.wiley.com/doi/10.1002/ajpa.21232/abstract)."
Lell et al. 2002
Lower Amur/Okhotsk Coast:
P-M45(xM17, M3)
6/17 = 35.3% Nivkh
2/20 = 10.0% Udegey
2/53 = 3.8% Ulchi/Nanai
0/16 Okhotsk Evenk
0/10 Upriver Negidal
0/7 Downriver Negidal
Kamchatka:
P-M45(xM17, M3)
5/27 = 18.5% Koryak
0/18 Itelmen
Figure 3
3 M45(xM17, M3, M173) [Okhotsk/Amur]
5 M45(xM17, M3, M173) [Okhotsk/Amur]
3 R1-M173(xM17) [2 Kamchatka + 1 Okhotsk/Amur]
15 R1-M173(xM17) [14 North/Central America + 1 Okhotsk/Amur]
Eight of the ten P-M45(xM17, M3) Y-chromosomes from the Amur/Okhotsk samples of Lell et al. must be negative for M173, and therefore Q(xQ1a2a1a1-M3), R(xR1-M173), or (an otherwise unconfirmed as far as I know) P*-M45(xQ, R). However, two of them are definitely R1-M173(xR1a1a-M17), so probably some sort of R1b. The Udegey sample of Lell et al. 2002 and the Nivkh sample of Tajima et al. 2004 contain some R1a1 Y-chromosomes, so I suspect some sort of Western admixture here (either recent Slavic or ancient Indo-Iranian).
In any case, subtracting the two R1-M173(xR1a1a-M17) individuals from Lell's pool of Lower Amur/Okhotsk Coast samples leaves 8/123 = 6.5% P-M45(xR1-M173, Q1a2a1a1-M3) in this region, so some sort of Q or R2 is definitely present in some members of these populations, especially the Nivkhs. However, it is impossible to deduce whether the two Udegeys are P-M45(xR1-M173, Q1a2a1a1-M3) or R1-M173(xR1a1a-M17).
Kristiina wrote,
"I got acces to the data in “Genetic Diversity of Two Haploid Markers in the Udegey Population From Southeastern Siberia” (Jin et al. 2009). It seems that they share a strange Y*(xC, DE, K) line at a frequency of 1/21 = 4.8% with at least the following groups: Beijing Han (the highest frequency), Koreans, Buryats, Khalk, Yunnan, Philippines, Thais."
Where did you obtain that study? Do you have paid access to it?
Anyway, this Y*(xC, DE, K) is probably a catch-all for various F-M89(xK-M9) derivatives, which generally have been found to belong to typical South Asian (e.g. H-M69) or West Asian/European (e.g. J-M304, I-P19, G-M201) haplogroups upon further testing.
Kristiina wrote,
"They also share with the same groups K-M9(xO) that should not be R or Q as they say-that “in addition, unlike on the maternal side, very little contribution from European Y chromosome haplogroups was observed in the Udegeys studied here, although non-differentiated Y chromosomes (e.g., haplogroups Y* and K-M9) need further analysis of their classification."
It could easily be Q or N, since most researchers do not assume that either of these haplogroups represents "contribution from European Y-chromosome haplogroups." Frankly, it could even be some sort of haplogroup R for all we know. It is just not O-M175.
In that other study, Jin etal., they use the following typings (with Udegey frequencies):
ReplyDeletey*(xC/DE/K) 4.8%
C-RPS4Y 66.7%
DE-YAP 0%
K-M9 (xO) 9,5%
N/O-M124 (xO-M175) 0%
O-M175 (xO-M119/P31/M122) 0%
O*-M119 0%
O-P31 (xO-M95/SRY+465) 0%
O*-M95 0%
O-SRY+465 (xO-DXYS5Y) 9.5%
O*-DXYS5Y 0%
O-M122 (xO-LINE1) 4.8%
O*LINE1 4.8%
Am I right that based on the above, Udegey K-M9 in Jin paper cannot be N-M231(xN1c1-Tat) or any other upstream NO line. On the basis of above, it should not be O2-P31 either. So, what is left for this K-M9, seems to be either Q, R or L. According to Malyarchuk et al. study, Buryats, Koryaks and Tungusic Evens and Evenks and Koreans have the following frequencies:
Buryats R2-M124 2.7%, R1b1b2-M269 0.7%
Koryaks Q1a*-MEH2 10.3%
Evenks R1b1b2-M269 2.4%
Evens Q1a3a-M3 3.2%
Koreans 0% all
The typings were the following: Q1a*-MEH2, Q1a2-M25, Q1a3*M346, Q1a3a-M3, R2-M124, R1b1b1-M73, R1b1b2-M269.
According to Zhong et al. 2010 study, Heilongjang Han have 1.5 % of R1b-M269. In China R1b-M73 is found in Gansu and Shanxi Han and Heilongjang Mongolians. However, also Q is found in China: Q1a1-M120, Q1a2-M25, Q1a3-M346. Liaoning and Heilongjang Man seem to have Q1a1-M120, and they are not that far from Amur area
So Udegey K-M9 could then be R2-M124, R1b-M73 or R1b1b2-M269 or Q1a lines. Anyway, the comment according to which ”very little contribution from European Y chromosome haplogroups was observed” could have been clearer and they could have said that K-M9 is actually (in part) haplogroup R, if that is the case
As for this Y*(xC, DE, K), according to this same Zhong et al. paper, Gansu Han, Shandong Han, Heilongjang Han, Ningxia Hui and Inner Mongolians have J2. Henan Han, Liaoning Han, Ningxia Hui and Mongolians have G2. Ningxia Hui have also H. Interestingly, L3 has been detected in Heilongjang Han and Ningxia Hui. I would vote for J, as it has the highest frequencies.
Thank you for additional information from the Jin et al. 2009 paper, Kristiina.
ReplyDeleteAt the risk of being importunate, I repeat my previous question: where did you obtain that study (Jin et al. 2009)? I do not have access to it, so I can only refer to people's comments that I have found online.
Ebizur, we both have a Google account, so is it possible that you send me a personal email. I can then send you the PDF. If it is not working that way, can you accept me in your "circle". (I do not understand very well how this Google profile works and my profile is not complete!)
ReplyDeleteKristiina,
ReplyDeleteIf it would not be too much trouble, may I ask you to send the Jin et al. 2009, Zhong et al. 2010, and the Malyarchuk et al. study to which you have referred to daikun083@gmail.com? I would be very grateful.
By the way, in the meanwhile, I have noticed that the mtDNA haplogroup M individuals in Starikovskaya's Udegey sample have been defined more precisely as M7 (9/46 = 19.6%, three different but apparently closely related haplotypes) and M9 (4/46 = 8.7%). Two different M7 haplotypes, which are not shared with the Udegeys, also have been found in the nearby and linguistically related Ulchis (2/87 = 2.3%). Starikovskaya's sample of Tuvans from south-central Siberia also contains two different M7 haplotypes (3/95 = 3.2%), shared with neither the Udegeys nor the Ulchis. From another paper, I recall that the M* found in this sample of Mansis from western Siberia also belongs to M7 (1/98 = 1.0%). M7 is generally a coastal East Asian haplogroup, prevalent in populations from Japan, eastern China (and Taiwan, including the aborigines of that island), and Vietnam. I suppose that the Udegeys and Ulchis represent the northernmost extension of this haplogroup along the eastern coast of Asia; maybe its presence in these two populations is somehow connected with the spread of O-M175, which also seems to find its northern terminus in the same populations (excepting singletons that have been reported from some samples of Koryaks and Yukagirs). M7 in the Tuvans is probably indicative of a secondary East Asian influence on this population, related to the East Asian influence(s) on their Southern Altayan, Kazakh, and Mongol neighbors (also reflected by e.g. Y-DNA haplogroups O3 and D3), and is probably separate from the East Asian influence on populations of the Amur River basin/Greater Manchuria.
M9 was found only in this study's sample of Udegeys. It is most common in Tibet, but also found with low frequency elsewhere in continental East Asia and Central Asia, and it is related to haplogroup E, which is common in Austronesians from Taiwan, the Malay Archipelago, and the Mariana Islands.
I didn't know where to post this.
ReplyDeleteRecently (a few months ago) the Jomon have been shown to have carried not only Neanderthal DNA, but also Denisovan DNA. This evidences their ties with Melanesians/Papuans.
http://www.saitou-naruya-laboratory.org/assets/files/Seminar-Saitou-03-10-14.pdf
The study by Jinam et al. (2012) was shared by Dienekes on this blog when it was published, but it was behind a paywall. I have found it for free here:
http://www.saitou-naruya-laboratory.org/assets/files/pdf/J_Hum_Genet_57-787_2012.pdf
The study shows that Ainu are related to Sahulians (Papuans/Australian Aborigines). The level of Ainu admixture in ethnic Japanese highlighted by this study is consistent with higher Papuan+Melanesian admixture detected by others among Japanese than among other East Asians (http://oi57.tinypic.com/34ga841.jpg)
Thanks Vincent. Very interesting. It has taken me a while to check it all out. I see my first comment at this post was, 'I would be far more surprised to learn that the Jomon were a homogeneous population'. Seems I have no need to be surprised. From your link:
ReplyDelete"the Ainu population seems to have experienced gene flow
with two different populations, the Mainland Japanese and the yet
unknown population in the north".
What northern population is an interesting question. The authors propose Okhotsk or Nivkh. Quite possibly correct. On the other hand their figure 2 argues against the Ryukyu islanders and the Ainu being particularly close. It seems the similarity is as much the lack of other connection for the two regions as it is common origins. For example as far as I'm aware Y-DNA D2 is common to both Ainu and Ryukyu but C1a1, although common in the Ryukyu Islands, is unknown in the Ainu.
Dienekes has some language connections between Okhotsk and Ainu here (in case you haven't seen it):
http://dienekes.blogspot.co.nz/2013/04/okhotsk-and-ainu-linguistic-connection.html
"The study shows that Ainu are related to Sahulians (Papuans/Australian Aborigines)".
This particular study doesn't show that. It refers to a 1995 study. They quote:
"It should be noted that Omoto53 conducted a pioneering study on
the phylogenetic relationship of the Ainu population considering
various degrees of admixture. When a 60% admixture with the
Mainland Japanese was assumed for the modern Ainu population, the
ancestral Ainu population was clustered with Sahulian (Papuan and Australian)".
Mind you I have seen a study on Malay Orang Asli that claims a connection between some elements of that population with both Australian Aborigines and Ainu. And until the recent past many assumed such a relationship existed, along with a connection of both with West Eurasian populations.
"What northern population is an interesting question. The authors propose Okhotsk or Nivkh. Quite possibly correct. On the other hand their figure 2 argues against the Ryukyu islanders and the Ainu being particularly close. It seems the similarity is as much the lack of other connection for the two regions as it is common origins. For example as far as I'm aware Y-DNA D2 is common to both Ainu and Ryukyu but C1a1, although common in the Ryukyu Islands, is unknown in the Ainu."
ReplyDeleteI think that the unknown population was the one from which both the Nivkhs and a part of the Ainu descended, that is, the Okhotsk population.
Regarding the Ryukyuan component, I also think it is correlated to C1a1-M8, as I don't see any Ryukyuan-specific lineage, either paternal or maternal. I am inclined to think that both C1a1 and the Rykyuan admixture were more frequent in Kyushu and Honshu. Japanese archaeologists recognize two distinct Upper Paleolithic cultures in the Ryukyuan archipelago. The remains of one group were found only in the northern part of the archipelago. This Northern group is thought to have come from Siberia through the Korean peninsula. After its arrival, the Northern group spread in the whole Japanese archipelago (the remains of this group are found also in Kyushu and Shikoku). The remains of another group, the Southern, were only found in the southern and central Ryukyus. This group is thought to have come from Southeast Asia. Both groups were contemporary (30-40 kya), and both were composed of people with a sundadont dental morphology. I think that the Southern group (C1a1) expanded into Kyushu and Honshu and mixed with the Northern group (D1b, ex D2), except in Hokkaido, where the Northern group survived unadmixed. This would explain why the Kanto Jomon are very different from the Tohoku and Hokkaido Jomon (these two groups instead are more related; see the paper linked by Dienekes), and also why C1a1 is not present among the Ainu and the presence of the Ryukyuan component among all Japanese (except the Ainu). To get a better picture of what I mean, please read this http://heritageofjapan.wordpress.com/pacing-the-paleolithic-path/origins-of-the-paleolithic-people/
The Hokkaido Jomon were not distinct from the Kanto Jomon only on a maternal, paternal and autosomal basis, but also in cranial morphology. If you compare Hokkaido Jomon crania to Kanto Jomon crania and to modern groups from across the world, you will notice that the Kanto Jomon crania (and in general non-Hokkaido Jomon crania) tend markedly towards some kind of Mongoloid, while the Hokkaido Jomon crania are markedly similar to the Australian Aboriginal ones (not only to historical crania, but also to prehistoric specimens like Keilor, Kow Swamp etcetera) and very different from the Kanto group.
Regarding the Melanesian affinities of the Ainu and their autosomal component. The study I linked reported Denisovan admixture among Jomon adding to the Neanderthal basis common to all Eurasians. This is enough to differentiate them from extant Asians and to tie them with Sahulians. Of course, this is valid if we consider the Ainu as the purest extant descendants of the Jomon in Japonesia, but I don't think that this is questioned. If you look at the purest representatives of the Ainu, you'll notice that there's more than only a phenotypic similarity: http://cryptomundo.com/wp-content/uploads/ainu-japan.jpg
Thanks for all that Vincent. While I have no doubt about a connection with Sahul for the Ainu I see a problem regarding C1a1 being part of a movement from Southern China or Sahul. I realise you have severe doubts concerning haplogroups but C1a1's closest relation is C1a2, a European haplogroup as far as we know. Of course C1a may have been very widespread at some time but it seems to be a 'northern' version of C1 with C1b being South Asian, C1c being Wallacean and C1d being Australian. The problem becomes did C1 move east to Japan from Central Asia or did it move west to Europe from Japan? My guess is that C1a's origin is Central Asian and it moved both east and west from there, making up a small proportion of the P Upper Paleolithic expansion.
ReplyDeleteI know very well that the closest relation of C-M8 is C-V20. Mine is only a hypothesis to explain the difference between the two UP cultures in the Ryukyus through genetics. The problem is the age of C1a and its descendants. The majority of experts say around 10 kya for C-M8, only one pointed at 42 ky. If the latter date is correct it is not required that the Southern group came from Southeast Asia, C1a1 may have arrived from Central Asia through Eastern China.
ReplyDeleteI take this opportunity to notify you all about interesting "news".
ReplyDeleteI think I am the only one who knows that Y-DNA DE (sub-haplogroup unknown) has been found among Australian Aborigines. Hammer et al. (1994) is the first study about YAP (Y chromosome Alu Polymorphism), a synonym of Y-DNA DE. The authors found 1 out of 7 Aborigines to be YAP+. The authors were unsure about the genuine nature of the YAP individual, but I have found another, recent study that points at an indigenous origin of YAP in Australian Aborigines.
This "study" (better, abstract from a conference) from 2013 investigated the Y-DNA of other 7 Australian Aborigines. Six of them had the usual C and K haplogroups, but one had Y-DNA M. The fact that neither Hudjashov et al. (2007) nor Keyser et al. (2002) found traces of DE and M among their little sample (~200 Australian Aborigines) means that future investigations will report other cases of both haplogroups. Here's the link http://www.ashg.org/2013meeting/abstracts/fulltext/f130121558.htm
Dienekes, sorry for this series of comments, but I also found another abstract that seems to support the theory of Eurasian origin of mtDNA L3 (I know you support this theory), and a back-migration of it to Africa (with an obvious DE/E male counterpart, even if not stated in the abstract). Here's the abstract:http://www.ashg.org/2013meeting/abstracts/fulltext/f130122833.htm
ReplyDelete"I think I am the only one who knows that Y-DNA DE (sub-haplogroup unknown) has been found among Australian Aborigines ... I have found another, recent study that points at an indigenous origin of YAP in Australian Aborigines".
ReplyDeleteI didn't see anything in the link about YAP, but recently a 'D4' has been identified in the Philippines and so it is not altogether impossible. But regarding the 1994 paper, 'The authors were unsure about the genuine nature of the YAP individual'.
"the Y-DNA of other 7 Australian Aborigines. Six of them had the usual C and K haplogroups, but one had Y-DNA M".
That's not really surprising. As the abstract says:
"This last haplogroup has a restricted geographical distribution centred on Papua New Guinea, and illustrates a genetic link between Australia and that region".
It is actually mainly centred on Melanesia and so must have been capable of substantial open water voyaging.
"C1a1 may have arrived from Central Asia through Eastern China".
That seems the most likely situation to me.
"The problem is the age of C1a and its descendants. The majority of experts say around 10 kya for C-M8, only one pointed at 42 ky."
ReplyDeleteThe later age is presumably that of C1a1-M8 whereas the really relevant date is the split between C1a1-M8 and C1a2-V20. C1a1 may have remained a 'private' lineage until 10 kya. and then been able to expand for some reason or other.
"I didn't see anything in the link about YAP..."
ReplyDeletePage 3, Table 1. Page 7: "A single Australian was found to have the Alu insertion. As it is not known if this is due to recent admixture, a more extensive survey is needed to determine if the YAP element is polymorphic in native Australian populations". There is no doubt about the individual being YAP+, it's only that the authors (righteously) would have liked a confirmation of the fact that DE was indigenous.
"That's not really surprising. As the abstract says..."
The aim of my sharing this abstract was certainly not to "surprise" anyone in fact. The point I wanted to make clear is that haplogroup DE in Australia is very likely to be indigenous, as the M individual. Both DE and M were not reported in studies previous to this one (except DE in Hammer et al. 1994), but now they are known to exist and very probably they are indigenous.
Regarding D4 (now D2), the Filipino men who tested positive for it reported to have exact matches with an individual from Guam (this island has been shown by a study to possess D-M174*). For this reason I suspect that the Andamanese and Southeast Asian D* is very likely D2, while real underived D* is present only among Tibetans and nearby populations.
"Regarding D4 (now D2)"
ReplyDeleteThanks for drawing my attention to that Y-DNA D adjustment.
By the way Vincent. Have you seen this:
ReplyDeletehttp://heritageofjapan.wordpress.com/2014/08/16/bold-new-dna-2014-study-proposes-when-and-where-the-three-major-populations-of-africa-europe-and-east-asians-diverged-and-suggests-why-the-ainu-look-like-europeans/
I think it probably can be taken with more than just a grain of salt, but interesting nonetheless.