June 2, 2013

Ancient Jomon mtDNA from Japan

Udege family
There is some debate about the connection between the Jomon period (Japan's ceramic but pre-agricultural period, extending between c. 16,000 to 2300 years ago) and the Ainu, as well as Ryukyuans and other peoples, including mainstream Japanese. A new study provides some extra bits of information to fuel the debate:

Hideaki Kanzawa-Kiriyama et al., Ancient mitochondrial DNA sequences of Jomon teeth samples from Sanganji, Tohoku district, Japan. Anthropological Science 2013 (advance publication). Open accessLINK [doi:10.1537/ase.121113]

The researchers sequenced ancient mtDNA from Jomon remains from a shell mound of Sanganji (Fukushima), which produced two M7a2, one N9b2 and one (incomplete) N9b* sequences.

Referring to previous similar studies as well, they produced the following tables:

From this data it would appear that the ancient Jomon people would be most closely related to modern Udegey (or Udege) from the Amur region of Eastern Siberia (with the possible exception of the Kanto Jomon, who may be closest to Ryukyuans instead).

The Sanganji sample is included pooled into Tohoku Jomon

The Ainu, it must be said, are next in line after the Udege, and I wonder if recent admixture may be distorting their relation. Another issue is that in such an extensive period of almost all the Holocene and even some millennia into the Pleistocene, there may have been flows and variability also within the Jomons (the Sanganji shell mound is dated to c. 4000-2500 BP, for example).

Whatever the case, it seems clear that N9b was an important matrilineage among ancient Jomon peoples, while M7a (now most common among Ryukyuans) was present but less common, with the Sanganji sample being rather exceptional in this.


  1. there is much confusion over Jomon mtDNA. For some reason the fundamentals of out of Africa migration do not seem understood. It is clear that Hg M and R(B,F) have SEA origins and the N9b for example has a Siberian origin -originally from SW Asia. N9, X; Y and A probably came to Japan by 25 kya as South Siberia was almost totally depopulated during the coldest part of the ice age 28 to 24 kya

    1. Hi Alan. I'd say that ultimately all come from Southeast Asia (SEA). Even if N9b may have spread from Manchuria or SE Siberia, it's not what I'd call a Siberian lineage in the wider sense but rather NE Asian. It's not known (or very rarely so at most) in inner Siberia.

  2. Hi Maju,
    Thanks for your quick response. If you look at the work of Zhong 2010 and 2011 on Y hgs, Tateno 2014 Y and mtDNA, it is clear, but new, that that there are 2 types of East Asians. 1 just from E Asia and the other Eurasians, from which East Asians and Europeans split about 41 kya.

    Sukernik 2012 "Mitochondrial Genome Diversity in the Tubalar, Even, and Ulchi: Contribution to Prehistory of Native Siberians and Their Affinities to Native Americans -"A total of 16 distinct haplotypes attributable to haplocluster N9 (N9a-N9b-Y) were identified in the Tubalar, Even and Ulchi (Table 1)" Quote regarding N9B, (they have a new phylogeny of N9 too).
    The updated genealogy of N9 enriched by 10 new sequences, of which the Tuvan and Udegey mtDNA samples are from our old collection, is shown in Figure 2. The ages of principal offshoots of the N9 lineage are 18.7/16.5 kya for N9a, 16.4/19.2 kya for N9b, and 25.0/30.7 kya for Y, thus falling nicely within the Ice Age, whereas the antiquity of the subhaplogroup Y1a confined to the Even, Koryak, Ulchi, and Nivkhi is much younger, only 6.0/1.8 kya. Spatial patterns andcoalescent dates of the N9a, N9b, and Y haplogroups suggest that the N9 root emerged in southwest Asia 38.7/36.0 kya, but only much later, presumably during the LGM, its particular derivatives might have spread through the corridor provided by southern refugia that stretch from mountainous Altai to the Russian Far East" N9b is also very prominent in the Jomon of Japan and especially in Hokkaido
    Regards Alan

    1. Can you provide a link, doi or title for your first suggested read? Zhong seems a way too common name to be of help in my search.

      By "Europeans" do you mean West Eurasians in general? It is well known that West Asians and Europeans are almost indistinct, with West Asia being for the greatest part the craddle of this macro-population.

      I wouldn't think that the ancestor population to both major Eurasian branches exists anywhere but (as different sister branches, plus whatever admixture) in Southern Asia and Australasia.

      Siberians and Native Americans mostly seem an ancient mix of the West and East Eurasian branches some time after they split apart. Native Americans for example clearly seem to be, on one side (mostly patrilineal Q1 but also mtDNA X2), West Eurasians (South/West Asia originated via Central Asia) and, on the other, (most matrilineages but also Y-DNA C2, former C3) from East Asia. This process can also be tracked in the expansion of the so-called Upper Paleolithic or "mode 4" via Mongolia and North China from Altai, where it is similar to what is found in West Eurasia at similar dates (soon after 50 Ka BP). Recent aDNA studies in Central Siberia add quite conclusive evidence in support of the Central Siberian branch leading (after admixture with East Asians) to Native Americans had strong Western affinities and can today be best represented by Kets (although is quite influential in NE Europe, Afghanistan, etc.)

      The data you present from Sukernik 2012 (sadly pay per view) does not seem to contradict this scenario and would rather suggest that the Tubalar (the westernmost population of all mentioned ones) would have a rather eastern origin at least in part. I can't judge further without looking at the detail of the data but in general it seems that N9b would be centered towards NE Asia and has at most a thin relation with the corridors of Central Siberia. Otherwise we'd see some of that in Russia, Finland, etc. and we do not.

      More common Siberian lineages of Eastern origin are C and D, to lesser extent A and seldom some others like G or F. C has been spotted in Europe since the Epipaleolithic and probably indicates that the (proto-)Uralic migration carrying N1 to the West was already ongoing. D is probably of a later dispersal, very possibly related to the Seima-Turbino phenomenon (Siberian Bronze Age, possibly related to Tungusic peoples). Other Oriental lineages are rare in West-Central Siberia, as well as in NE Europe and most likely arrived only since the Turkic expansion began.

    2. Also a warning against taking age estimates too seriously. They have lots of problems and do not have any meaningful weight as scientific evidence. They are just marginal "good hunch" results that are given too much attention and not at all enough caution, sadly enough.

  3. Check here for some of the recent materials on Siberian genetics and prehistory, discussed by me:

    In general modern Siberians seem intermediate (at variable ranks) between West-South Eurasia on one pole and East Asia on the other and Native Americans on the third vortex. However ancient Central Siberians cluster between modern West-South Eurasians and Native Americans with zero East Asian affinity.

    On the haploid side of the issue Native Americans seem pretty much, as I outlined above, a mix of West-South Eurasians and East Asians, even if an ancient one. This ancient time depth cannot be seen in autosomal genetics, at least not easily. The proto-Amerinds surely migrated eastwards from a "Western" population in early UP Altai, incorporating then massive Eastern genetics, mostly from the maternal side. Inversely, after the LGM, the proto-Uralics did almost exactly the opposite via the Taiga, bringing East Asian genetics to Europe but also admixing heavily with the locals, particularly in the maternal side (Y-DNA N1 is still very important among Uralics but East Asian mtDNA not so much or even almost zero in some populations).


Please, be reasonably respectful when making comments. I do not tolerate in particular sexism, racism nor homophobia. Personal attacks, manipulation and trolling are also very much unwelcome here.The author reserves the right to delete any abusive comment.

Preliminary comment moderation is... ON (your comment may take some time, maybe days or weeks to appear).