The Sardinian meter

By Razib Khan | August 21, 2012 11:10 pm

I cropped the image above from the paper Inference of Population Structure using Dense Haplotype Data. The main reason was emphasize the distinctiveness of the Sardinian cluster, on the bottom right. As you can see this population exhibits a lot of coancestry across individuals. This isn’t too surprising, Sardinia is an island, and islands are often genetically distinctive. Random genetic drift prevents populations from diverging through gene flow, but water is a major impediment to gradual isolation by distance dynamics. The original Sardinians are naturally going to diverge from mainlanders over time, and begin to share the same set of common ancestors in the recent past, because their space of reasonable mating possibilities is constrained. The other population which is similar in the heat map above are the residents of the Orkneys, off the north coast of Scotland (the Orkneys has a much smaller population than Sardinia, but, it is also much closer to the mainland).

This is on my mind because Dienekes has a long post where he explores the D-statistic results of various European populations, using Sardinians as one of the references. You don’t need to know the details, just that Northern European populations seem to exhibit an affinity to East Asians. Our favorite human genomicsts at Broad highlighted this tendency, and Dienekes went looking (if David Reich et al. were inclined toward mischief they should just posit some crazy scenario, and see if Dienekes assembles the requisite data set!). Which is all fine, and we’ll see more of this in the near future. It isn’t as if others aren’t using Sardinians.

If you play around with model-based clustering packages (e.g., ADMIXTURE, STRUCTURE, etc.) or PCA the distinctiveness of the Sardinians does jump out at you. In short they are the least Asian of the European populations. If there is a cline, they generally define one of the antipodes of the cline (often, along with the Basques). They’re a natural reference. But we need to be cautious. On occasion a reader has casually asserted that understanding the relationships between populations is easy enough, just find appropriate reference groups. Easier said that done.

A few years ago there was a great deal of debate whether the Sardinians were Neolithic “newcomers” or the descendants of Paleolithic populations. The latter is actually somewhat conjectural in regards to archaeology. There is some evidence of Paleolithic habitation, but it is fragmentary, implying perhaps very low population densities, and likely local extinctions over time. The reason the argument roiled for so long is that if Sardinians are the descendants of Paleolithic Europeans you have the perfect reference. But what if they’re not? I think that question is up in the air, especially in light of the fact that Ötzi the Iceman has affinities to the Sardinians. Ötzi was a mainlander, and an agriculturalist. That does not preclude descent from a common Paleolithic Western European population which adopted agriculture, but I think it tilts us more toward some sort of demographic expansion (likely from the Eastern Mediterranean, but perhaps a secondary expansion from somewhere in the Western Mediterranean).

Ten years ago we were looking at a two-factor model. Europeans the Ice Age hunters, or Europeans the Middle Eastern farmers. Today the past is much more crisp, and yet that crispness has unveiled a befuddling complexity. If I had to put money down I would bet that in another ten years we’ll have to add into our understanding both demographic pulses during the Ice Age, and multiple ones which post-date the Neolithic. It may be that the signals of a relationship between Northern Europeans and the First Americans is very old, and dates back to the Gravettian culture, which spanned Western Europe eastward toward the fringes of the Black Sea. With ancient DNA I believe that we’ll get a sense of who the “First Europeans” were, and it may be that their genetic contribution to modern Europeans is in the same range as the Neandertals whom they assimilated and exterminated!

All something to keep in mind in the coming years, when European paleogenetics may be measured by the Sardinian….

Image credit: Wikimedia

  • http://forwhattheywereweare.blogspot.com/ Maju

    As you bring the matter around, using your same image (fig. 5 of said study) but inconveniently uncropped, we can appreciate that Sardinians but also the “Italian” and the “Tuscan, French” groups show intense (blue, dark purple) transmediterranean affinities (West Asia, Africa), which is suggestive, IMO of Neolithic and post-Neolithic flows.

    These transmediterranean genetic connections are totally missing (low values: yellow, orange, red) among Orcadians and Basques, what makes these Atlantic populations once again apparent legitimate reference for the pre-Neolithic European reality.

    Russians also but these have Siberian/East Asian affinities. The main “French” population are midling.

    I’d suggest using Basques and Orcadians and probably Lithuanians as well, as proxy for the mostly unmixed pre-Neolithic European layer. Sardinians are probably not the thing.

  • http://dienekes.pontikos@gmail.com Dienekes

    Our favorite human genomicsts at Broad highlighted this tendency, and Dienekes went looking (if David Reich et al. were inclined toward mischief they should just posit some crazy scenario, and see if Dienekes assembles the requisite data set!)

    Actually, I highlighted it first:

    http://dienekes.blogspot.com/2011/04/sub-saharan-admixture-in-west-eurasian.html

    “However, this is clearly a case of seeing the glass half full. The authors prefer the hypothesis that some Caucasoid groups have African ancestry, although the hypothesis that other Caucasoid groups have East Asian ancestry can equally well explain the observed pattern. Indeed, both hypotheses may explain the phenomenon they observe.”

    http://dienekes.blogspot.com/2011/05/east-asian-and-african-shift-of-west.html

    “With respect to the Asian- and African- shift of West Eurasian populations, I note that northern Europeans (and Basques) are less African-shifted than southern Europeans, and, at the same time they are more Asian-shifted: the 16 least Asian-shifted populations have a coastline in the Mediterranean (excluding the Portuguese), while the 16 least African-shifted populations do not (excluding the French).”

    So, it’s good to know that people at the Broad institute assembled the requisite data to test my crazy scenario.

  • Torbjörn Larsson, OM

    With ancient DNA I believe that we’ll get a sense of who the “First Europeans” were, and it may be that their genetic contribution to modern Europeans is in the same range as the Neandertals whom they assimilated and exterminated!

    Can the fact that Ötzi was robustly ‘twice as neanderthal’ (whatever that means) tell us something about the early population structure of Europe or the ancestry of Sardinians? (I see you linked to Hawks’ article before, but I don’t think you have discussed this topic – if it is one.)

  • http://forwhattheywereweare.blogspot.com/ Maju

    I’d say it’s simpler than that: tropical Africans or San act as proxy for North/NE African influence. IF a North African sample would be at hand that would be obvious but for some reason North Africans are generally ignored.

    I’m not sure about the validity or usefulness of chromo-painter (it seems to identify well clusters of endogamy but everything looks the same beyond that) but it’s certainly not the same as PCA or other even more useful tools. It’s a novelty but not all novelties are improvements.

    Whatever the case, IMO there is clear minor African influence in Europe in relation with essentially Y-DNA haplogroup E1b, which has an African origin. This influence extends at subtle levels through all Europe but it’s more obvious and intense in some areas of the South. Similarly there is East Asian (or more exactly Siberian) influence in Northern Europe, correlated with Y-DNA N more obviously but also with R1a to lesser extent.

    But what interests me the most is the West Asian influence, which in the relevant uncropped graph is very different: high for Italy, Sardinia, etc. but effectively zero for the Atlantic.

  • pconroy

    For those of you wondering what an Irish person with 17.58% Ameasian admixture might look like, I offer my sister at 6 yo, who had/has a Eurasian look about her:

    http://i48.tinypic.com/ofdn9z.jpg

  • aks

    Razib could you explain your comment about the relations between the Gravettian culture and the First Americans. Are you suggesting this in support for the Solutrean hypothesis?

    Thanks!!

  • https://plus.google.com/109962494182694679780/posts Razib Khan

    #6, treemix seems to indicate a genetic affinity between northern europeans and non-northern native americans. the most recent paper by the reich lab indicates that native americans came in three waves, the ‘first americans’, later the na-dene, and finally the eskimo-aleut. the two later don’t have an affinity with europeans, so it isn’t a function of conventional gene flow. the most parsimonious explanation for this is probably that there were ancient connections between some ancestors of europeans and the siberian ancestors of the first americans. probably we should peg this >15,000 years ago, since that’s around when first americans showed up in the new world (perhaps a little later). the gravettian culture extended toward the eastern borders of european russia until ~20,000 years ago. so no, i’m not talking the souletrean hypothesis.

  • Karl Zimmerman

    Is this Asian element unique to European Caucasoids?

    Although some portion of this was probably historically recent (Mongol, Turkic, Hunnic, etc) judging by the Basque having somewhat reduced levels of Asian admixture compared to surrounding IE-speaking peoples, the proto-Indo-European population probably had a somewhat elevated Asian admixture compared to “first-wavers” (although perhaps somewhat lower than Finnic peoples).

    Thus we’d expect to see a minor echo of Asian admixture not only within Europe, but also within other areas that Indo-European peoples have touched – Iran and South Asia. Obviously some Indo-Aryan populations have had recent historic contact with East Asian peoples, but a more ubiquitous echo would be telling.

  • https://plus.google.com/109962494182694679780/posts Razib Khan

    #8, excellent point. both dienekes & zack have the data sets to test this proposition (as do the ppl in the reich lab). dienekes runs do indicate that non-turk west asians don’t have it. armenians ARE indo-european, and they didn’t.

  • pconroy

    @8, @9,

    As I mentioned on Dienekes’s Blog:

    In terms of my sister and one of my father’s sisters, it’s not just partial epicanthic folds I’m talking about, it’s wide face, low nasal bridge, straight jet black hair, slender body-type.

    People always mistake her for some sort of Eurasian hybrid.

    BTW, Polako/Devidski in his old analysis found some Selkup in my Father and some Chukchi in my Mother, on other analysis my Mother seems to have some Athabascan ancestry. DNA Tribes found a minor Arctic component in my Father.

    On HarappaWorld calculator, my Father shown a minor component of South American Native ancestry, which included Karaitana, and all other samples from South America – so something common to them all or in other words probably “old”.

    So it’s a consistent genetic component, and shows up in the phenotype too.

  • Onur

    BTW, Polako/Devidski in his old analysis found some Selkup in my Father and some Chukchi in my Mother, on other analysis my Mother seems to have some Athabascan ancestry. DNA Tribes found a minor Arctic component in my Father.
    On HarappaWorld calculator, my Father shown a minor component of South American Native ancestry, which included Karaitana, and all other samples from South America – so something common to them all or in other words probably “old”.

    Conroy, you and your parents’ genetic results are no different from the typical Irish/British. Whatever overall genetic connection to Mongoloids you and your parents have is present in equal amount in the average Irish/British.

  • Onur

    continuing from my above post…

    I have seen all the Dodecad, Harappa, Eurogenes and DNA Tribes genetic analysis results of Conroy and his parents, and I know what he is referring to in his last post. In all of those analyses, his and his parents’ results were no different from the average Irish/British results in those analyses, and this includes the inferred exotic connections (however very small and/or questionable). So, based on all those analyses, Conroy’s and his family’s genetic connection to Mongoloid populations is no different from that of the average Irish/British.

  • Onur

    #8, excellent point. both dienekes & zack have the data sets to test this proposition (as do the ppl in the reich lab). dienekes runs do indicate that non-turk west asians don’t have it. armenians ARE indo-european, and they didn’t.

    According to Dienekes’ analysis, Armenians apparently have that East Eurasian element in a bigger proportion than Sardinians but smaller proportion than Greeks:

    http://1.bp.blogspot.com/-qCEiV81NI0c/UDJP0Y-C_LI/AAAAAAAAFjs/WJLXbhylaCM/s1600/f4_Sardinian_X_Han_San.png

    Here is a zoomed in and Euro-focused version of the same analysis (the numbers on the left show the estimated % of the total East Eurasian element, in Vologda Russians approaching 14%):

    http://1.bp.blogspot.com/-T76SQupkgGw/UDJVEtN-qNI/AAAAAAAAFkI/p0371G6gZNk/s1600/f4_Sardinian_X_Han_San_scaled.png

  • ackbark

    A pretty girl, but –what’s up with her elbows, is that a Sardinian trait?

  • pconroy

    @11,12, Onur,

    I’m only familiar with my family’s result in detail, that’s why I mention them. I expect them to be fairly typical of Irish and somewhat typical of British.

    I do know that some samples of Dienekes are not wholly Irish, but label them as such, so their results are different than my family.

    I also know one sample is from a guy who claims to be Native Irish from Donegal, NW Ireland yet it shows 12% SW Asian – a component similar to Bedouin?!

  • pconroy

    @14, Ackbark,

    I have elbows like that, they look normal to me??

  • Sandgroper

    I hadn’t noticed the elbows – I was too busy mentally calculating the waist/hip ratio.

    But now that you mention it, I notice quite wide variation in elbows, including range of movement – some people can straighten their arms so far that they bend back the other way. I’m sure there’s an appropriate anatomical terms for that, but have no idea what it is.

    To me she looks like someone straight out of a Minoan fresco.

  • ackbark

    @16, Great heavens! Do they allow you to move through crowds more easily?

    @17, yes she does look like a Minoan fresco.

  • pconroy

    @18,
    Let’s just say as a soccer player I made a very effective defender ;)

NEW ON DISCOVER
OPEN
CITIZEN SCIENCE
ADVERTISEMENT

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Gene Expression

This blog is about evolution, genetics, genomics and their interstices. Please beware that comments are aggressively moderated. Uncivil or churlish comments will likely get you banned immediately, so make any contribution count!

About Razib Khan

I have degrees in biology and biochemistry, a passion for genetics, history, and philosophy, and shrimp is my favorite food. In relation to nationality I'm a American Northwesterner, in politics I'm a reactionary, and as for religion I have none (I'm an atheist). If you want to know more, see the links at http://www.razib.com

ADVERTISEMENT

See More

ADVERTISEMENT

RSS Razib’s Pinboard

Edifying books

Collapse bottom bar
+

Login to your Account

X
E-mail address:
Password:
Remember me
Forgot your password?
No problem. Click here to have it e-mailed to you.

Not Registered Yet?

Register now for FREE. Registration only takes a few minutes to complete. Register now »