4000-year-old human genome sequenced

February 11, 2010 • 10:08 am

This week’s Nature reports that DNA from an ancient human has been recovered and sequenced.  The DNA is from a hair sample found several decades ago in Greenland, and just now recognized as human. It came from a male of the Saqqac Culture, which was active in Alaska, Canada, and Greenland from about 5,000 to about 2,500 years ago.

Surprisingly, the authors found that DNA could be sequenced from a hair lacking a follicle; apparently some cells were trapped in the hair shaft. About 80% of the genome was sequence-able. What does it tell us about this guy?  A few things:

1.  He had type A blood and was Rh positive. That’s a common blood type in Siberians and Asians.

2.  He had a haplotype consistent with having brown eyes. No surprise there.

3.  Other SNPs (single-nucleotide polymorphisms) associated with physical trait suggested that the man had darker skin, thick, dark hair, dry earwax (!), and a tendency to baldness (but remember, the DNA came from a thick swatch of hair).   A combination of SNPs from different areas of the genome suggested that he had a stocky body.  That’s no surprise, either.  (This, by the way, conforms to “Allen’s Rule,” the biogeographic observation that populations from colder areas have relatively smaller protruding limbs and parts, and stockier frames—all of which conserve heat.) All of these genotypes are consistent with what we know about people who lived in northern Asia and Alaska.

It’s great that they could find out this stuff, but it’s really no surprise.  What made the paper Nature-worthy is the recovery and sequencing of ancient DNA from a human. Oh, and the really interesting result is this: the DNA suggests that the individual had components of genes still present in East Asian and Siberian populations, but not found in modern-day Inuits or people from South and Central America.  This suggests that there were two separate invasions of North America from Asia: the one that gave rise to native Americans, South Americans, and modern Inuit on the one hand,  and that leading to the presence of Saqqaq in Greenland.  Those latter individuals probably came across the Bering Strait, and then, hugging the Arctic, made their way eastward across North America and then to Greenland.

That conclusion is of course tentative because it’s based on only this single genome.  Still, based on the sequence, and the tentative phylogeny showing that this individual’s ancestors split off from the ancestors of their closest living relatives (the Chukchis of eastern Siberia) about 5,000 years ago, anthropologists may have to revise their conclusion that there was one invasion of North America from eastern Asia around 18,000 years ago.


Rasmussen, M. et al. (lots of authors!). 2010. Ancient human genome sequence of an extinct Palaeo-Eskimo.  Nature online, doi:10.1038/nature08835.  Note that this group previously published a sequence of mtDNA from this individual (Gilbert, M. T. P. et al., 2008. Science 320:1787-1789)

12 thoughts on “4000-year-old human genome sequenced

  1. Why does the absence of the specific DNA components suggest that the exodus from Asia happened twice?

    Maybe the DNA components stopped being selected for in American Indians and NA Eskimos, but not in Greenland Eskimos.

  2. I think the previous consensus, based on genetic and linguistic evidence, was that there had been two or three waves of migration to the Americas. The recently sequenced ancient Greenlander would be an additional immigration (i.e. the 3rd or 4th, not the 2nd).

    1. I still don’t see why there had to be more than one migration from Asia. If the genetic differences between the Greenlanders and the Siberians are smaller than the differences between the Americans and the Siberians, it could simply be a result of more drastic adaptations in the Americans.

  3. “Those latter individuals probably came across the Bering Strait, and then, hugging the Arctic, made their way eastward across North America and then to Greenland.”

    They must have really liked cold weather.

  4. I wonder how the hair was dated; you’d need a few strands to attempt 14C dating (maybe only 1 strand if you have an accelerator mass spectrometer).

    Rh+ is also very common (still the predominant type) in caucasians. You can also find asians with type 0. There is hardly anything at all that general blood typing can tell us about geographical origins with any certainty.

    1. Details on pages 6-7 of the Supplement:
      22.8 mg of human hair.
      “The yield after pre-treatment suggests that the material is excellently preserved… Sub-samples of the pre-treated hair or collagen were combusted and mass spectrometrically analysed using a Europa ANCA Roboprep interfaced to a Europa 20/20 MS operating under continuous-flow mode.”

  5. As Lewis Thomas once wrote, this paper deserves hiring a plane and painting the sky with exclamation marks.

    Hyperbole aside, for me as an archaeologist (with much interest but precisely zero expertise in Arctic archaeology, I hasten to add) this is extremely exciting, though perhaps least of all for the extremely hypothetical conclusions regarding the putative third wave of immigration from Siberia. This kind of inference from just one individual is unwarranted as yet, and the authors are extremely careful in their wording. The media, needless to say, are jumping on this most speculative part of the study.

    A few points I find remarkable after a ***first*** reading, apart from all the spectacular work on the DNA (and the general importance of the work as described by Jerry Coyne):
    1. a number of very clever statistical techniques, especially Admixture;
    2. the functional SNP assessment and its consequences;
    3. Fig. 3b: the PCA plot illustrating the close relation between the Saqqaq Inuk and Koryak, Yukaghir and Chukchi samples;
    4. Fig. 3c: the Admixture simulation of ancestry proportions, similarly highlighting the close genetic relation;
    5. Fig. S15 (in the Supplement) with further runs of the Admixture program;
    6. Fig. 1f: The excellent agreement between the calibrated radiocarbon date corrected for marine reservoir effect from the Saqqaq Inuk skeleton (OxA 20656) and the archaeologically associated reindeer cranium from the same layer (OxA 18749).

    Points 3-6 are really spectacular: Koryaks and Chukchi are closely related, their languages belonging to the Chukotko-Kamchatkan family. Joseph Greenberg, the late, controversial linguist, postulated (and others before him: Pedersen, Collinder, Ruhlen) a kinship between this family and the Yukaghir-Uralic languages. Now we’re talking about a geographic spread from Kamchatka to Finland! I am careful to mention just linguistic groups. In the case of Koryaks, Yukaghirs and Chukchi it is probably safe to speak of ethnic/ethonogenetic groups also. (This seems overly cautious, but much confusion arises from using linguistic, ethnic, and genetic denominations as if they were interchangeable.)
    One point deserves particular emphasis: all these groups relied heavily on reindeer, in all aspects of their lives; their coastal subgroups were also maritime hunters. The Rasmussen/Willerslev study shows through isotope analyses of the Saqqaq hair (carbon and nitrogen) that the individual relied on high trophic level marine food resources (hence the need for marine reservoir correction in the radiocarbon calibration). Yet the human remains were found next to a reindeer cranium. Amazing!

    A few puzzling points:
    i. The PCA plot of genomes lists ‘Na-Dene’ as one of the ethnic — and presumably genetic — groups. Yet ‘Na-Dene’ is primarily a linguistic term, coined by Edward Sapir nearly a century ago, including languages from Alaska to New Mexico. What gives?
    ii. The Admixture plots are grouped by conventional geographic denominations, on a linear nominal scale. It would be more informative to see a proper spatial projection, combining genetic/ancestry and spatial distances. It would be even more interesting to include chronological layers.
    iii. The map projection in Fig. 3a is absurdly distorted, presumably for ‘better readability’. A proper circumpolar map would show how relatively close the circumpolar world is, once we step out of our Mercator flatland.

  6. Wait a second, didn’t the reviewers ask the obvious question?
    Surely this is simply a case of an Eskimo wearing a wig made from Siberian hair!

  7. [Scratching head]

    So, the comments clear up the post confusion on the number of invasions.

    But what about the claim that “surprisingly” DNA could be sequenced from follicle less hair? Wasn’t that the big news from the ancients, oh say two years ago? IIRC someone found such cells in hairs of a badly degraded mammoth that couldn’t be sequenced successfully elsewhere due to mold.

    Apparently hair is hard to degrade for bacteria and fungus, making them as perfect little undiluted test tube archives for posterity as we can wish for. (I.e. still not very good, I suspect.) I thought it was a standard technique by now. Am I mistaken?

  8. I thought it was pretty well established there were at least two migrations to the Americas from Asia, the last being approximately 1000 years ago that brought the contemporary Inuit, and the earlier one that brought the rest, including the older Inuit-esque group that the Inuit replaced.

    In the other direction, I seem to remember that some cultural anthropologists claim to show commonality for groups across Siberia (Chukti, Nanai, etc.) all the way to the Sami and then down to the Ainu and across to the Inuit, but I can’t remember any references now, alas.

Leave a Reply