Totalt antall sidevisninger

lørdag 26. mai 2012

Fennoscandians and Neolitic Gotlanders

EDIT 30/08-2013: I now consider this analysis outdated. Please check the more recent posts with reanalysis of Ajv52, Ajv70, Ire8, Gok4 and Ste7.

In end of April 2012 the ancient Gotlander hunter gatherers genome was published by the University of Uppsala. I will now provide an ADMIXTURE based analysis of these samples using project and public published population panels.

The samples was downloaded from authors website and then subject to most of the quality controls as done by the authors. I extracted the relevant SNP for the Ajv's and Ire8 ending up with 22k SNP using three individuals combined into a composite individual.

The composite sample was then subject to control ADMIXTURE and MDS analysis using "haplotized" (phased in BEAGLE) French, Yoruba and Han's as reference populations. The composite individual clustered entirely with French.




I then merged the composite individual with an "haplotized" (phased in BEAGLE) European panel with populations all within the European continent from different public available population panels and run it through ADMIXTURE at K3. Closely related individuals in the panel was removed. One Russian was removed because of odd clustering, reason unknown.

The result was then averaged over populations to get an overview:



As we can see the individuals clustered into 3 main components, North-East Europe, Caucasus and the Mediterian. The Gotland composite neolitic hunter gatherer tops the list of North-East Europe populations, then the Saami and very close follow up the Finns.

As we can see the old Gotlanders appears to have significant Mediterian influence but almost zero Caucasian influence. This may be due admixture or common ancestry with the Mediterian populations. The Caucasian influence appears to be almost as low as the Basque. This may suggest that the Caucasian influence arrived later than the Mediterian one,

In modern individuals you may see that that Scandinavians, British, Lithuanians, Estonians and Finns appears less affected by Caucasian influence than continental European populations like Chuvash, Vologda Russians, Belorussians, Hungarians,French, Spanish, Romanians, Italians and Sardinians. Among the Saamis we see an elevated Caucasian influence. Its possibly linked somehow to the Chuvash. See posts below.

The Mediterrian influence appears to have been less in the North-East among Chuvash, Saami, Finns, Vologda Russians and Estonians, Note the somewhat higher rate among the old Gotlander. This all suggest that the Meditterian influence arrived second after the initial post glacier migration then later a migration from the Caucasus. The latter from Caucasus didnt reach the Basque. There may be support for this suggestion in the paper by Huyghe 2010:


The Caucasion migration appears to have reached the Saamis in particular from the vicinity of Chuvash and then through the Vologda Russians. See Chromopainter analysis below.

SUMMARY: It appears like Finns and Saamis have the best match of modern variation with these ancient Gotlander hunter gatherers. Note however that in this analysis not enough Saamis to provide proper clustering in ADMIXTURE or MDS. This means Finns and Saamis quite often have been lumped togheter in this kind of analysis.

(Updated 31/5/2012)


17 kommentarer:

  1. This finding of yours also showed up in some of Dienekes' elaborations, I think: the Caucasus component is more recent in most of Europe and must have arrived with Indoeuropeans.

    Otherwise I must say I have to disagree with your description of the NE European component as "Mesolithic", more so when it seems tightly linked to Finnic genetics, which are probably Neolithic (even if hunter-gatherer in some areas). You also seem to imply this way that the Mediterranean cluster would be Neolithic, what is not necessarily the case (i.e. it's higher among Basque than Spaniards, when anything "Neolithic" should be the other way around).

    SvarSlett
  2. Denne kommentaren har blitt fjernet av forfatteren.

    SvarSlett
  3. The Mediterrian influence appears to have been less in the North-East among Chuvash, Saami, Finns, Vologda Russians and Estonians, Note the somewhat higher rate among the old Gotlander. This all suggest that the Meditterian influence arrived second after the initial post glacier migration then later a migration from the Caucasus.

    Less "Mediterranean" component in Saamis, Chuvash, Finns and Vologda Russians than in the Neolithic hunter-gatherer composite should be due to the Mongoloid admixture in the former. The Mediterranean-peaking components are usually the farthest away from Mongoloid genetics among the Caucasoid components, so in the absence of a Mongoloid component usually the Caucasoid componenets that peak somewhere away from the Mediterranean eat the Mongoloid-related genetic portions. In this case it is what you call the "Mesolithic" component and the "Caucasus" component that seem to eat the Mongoloid admixture in populations. So those Mongoloid-admixed populations should actually have the North European-peaking and the Caucasus-peaking components in somewhat lower amounts than they show in this analysis.

    BTW, it would be better if you included the Neolithic farmer sample (Gok4) and Turks in this analysis.

    SvarSlett
  4. Hmmm it was late when I published this so it may need some editing :)

    SvarSlett
  5. The neolithic farmers in Scandinavia are different from the Hunter-Gatherers in more physical respect too. They have another NI (nasal index) and CI (crural index) than the H-G in Gotland (T.Ahlström, Univ. of Lund, 1997)

    Even farming people is investigated in the first thesis written on this subject(H. Malmström, Uppsala Univ. 2009). The Hunter-Gartherers from Gotland have lived in vicinity of cold climate thousands of years already before the farming populations arrived. The farmers is seen to have been living in warmer climate.

    Read also B. Fossum, Umeå univ. 2006/7.

    SvarSlett
  6. Anders, as for the Mongoloid admixture, to test for it you need Mongoloid reference populations (e.g., Han Chinese). To test for the Negroid admixture, you need Negroid reference populations (e.g., Yoruba).

    SvarSlett
  7. The ADMIXTURE graph have been updated with new label.

    SvarSlett
  8. Anders, it is not just a matter of labeling. You need to take into account the non-Caucasoid admixture in several of the populations. Otherwise, the results can be misleading in some respects.

    SvarSlett
  9. Saamis are very clearly VERY EUROPEAN. They are that most compared with all other European populations. Look at the green colour of the last fig.

    Fits very well with the idéa that Saamis in Scandinavia were the first population arrived during an Interglacial period ca 25000 years ago, and some stayed as Komsa Culture during the last glacial period.

    Later individuals arrived in the end of the last part of the glacial period from Basque refuge and through Westeurope and the British Islands. Some moved along the iceborder to east and then moved up north to meet those who came from south and those who stayed during the last phace.

    Then later came agriculture populations to Scandinavia, moved up very north but had to return back south or change the way of living as a period of the climate went bad. People split up in smaller bands and became nomadic hunters and abandoned their settlements. This smaller bands interacted with neighbors in trade. Now we are in Neolithicum and early Bronzeage.

    SvarSlett
    Svar
    1. What you say only makes sense if you consider "European" only the Green component, what is at the very least questionable. These divisions change depending on what peoples and sample numbers you use.

      There's no 1000 Genomes Saami sample (yet) but has Lithuanians who are akin, at least in the European aspect of their ancestry. In this case, the algorithm produces a diffuse "Mediterranean" (brown) component that I can understand why you would consider it to be non-euro (based only on this I'd think so as well). Using different samples, etc., I once arrived to this graph (form this entry) where West Eurasian components are clearly divided in four:

      1. the Selkup/Siberian component (of East Asian affinity and rare in Europe other than among NE peoples) - cyan
      2. the North African component (rare outside its region, except in West Asia) - green
      3. the European component, clearly dominant among all Europeans (weaker in the Balcans and Italy and strongest among Basques and Lithuanians) - purple
      4. the West Asian component, weaker in Anatolia, and penetrating through the Mediterranean and Balcans - red

      If you want to discuss "europeanness", I'd suggest to first find it as I did, with a proper extended sample and not one that obscures some areas and highlights others.

      Furthermore, the Saami are also the represented population (here) that has greater Siberian component, which is consistent with their very high Y-DNA N, a clade that appears to have arrived from Siberia with the Uralic languages themselves (the Selkups for example are still clearly Siberian in everything but the details of geography but early Finno-Ugrians obviously passed through a period of intense europeanization, mostly by mother side - this IMO happened already at the Volga before the Saami and other West Finnic peoples migrated westward as the ice sheet melted away).

      25,000 years ago and 10,000 years ago as well, Scandinavia was as habitable as Antarctica or the colder Greenland that killed the Vikings.

      Slett
  10. Discuss "europeness", I don't know? I see that we have different opinions already about the Y-DNA for Saamis. In my opinion it is I1 that is typical Saami, N1 is from Finland, and a much younger haplogroup for Saami, it came in very late...

    Saamis are often N1 today, they were moved in from Finland by the swedish governments policy, their extended reindeerherding was quite different from the swedish Saamis. Swedish Saamis were even not allowed to registrate a reindeermark, if they had too few reindeers. It was the government who alone decided who were allowed to have reindeer. Intensive reindeerherding and nomadic hunter-fishing was the way of living and also trade - furs and products from the Sea (tran...)

    About the Lithuanians saamis differ a lot from Lithuanians as you can see from this project. But Finlanders are closer to them.

    Astrid Cleve-Euler studied the Icetimes and she showed that it had been an Interglacial. Of course the climate was harsh, but people can live, where flower can grow... During this period also there were summers.

    Vikings was an agricultural population, and also had still traits from "a tropical animal" as the Farmers had that came to Southern Scandinavia (Sweden) in their CI(Crural Index) and NI(Nasal Index) (see above)

    SvarSlett
    Svar
    1. Maybe you have a better source and I must be corrected but according to the classic reference by McDonald, Saamis have like 45% N, what is consistent with the Y-DNA ancestry of other Finno-Ugric peoples and suspected former FU peoples assimilated into Indoeuropean identities (North Russian, Latvian). I is certainly a bit more important than among Finns but I'm not sure what this means, specially since many Finnic populations seem to have gone through quite drastic founder effect bottlenecks - and I would expect that to be the case with the remote Saami nation certainly, whose numbers are very small even today (130,000, my municipality alone has thrice that figure).

      Otherwise I generally feel sympathetic with the Saami struggle for keeping their identity and self-rule but I do not think that really matters much in this discussion.

      "About the Lithuanians saamis differ a lot from Lithuanians as you can see from this project".

      Of course but I meant in the big picture: they share North-Eastern European blood and that tends to cluster together, although admittedly Finnic peoples always look very peculiar in their bottlenecked uniqueness. I just did not have any Finnic people other than Selkups in my little Admixture exercise, and that was why I mentioned Lithuanians as proxy for the European (as oppossed to Siberian) part of the ancestry.

      I totally decline to discuss the anthropometric part of the matter as being supposed adaptions to anything. Neanderthals were more subarctic than our species and they were much more dolicocephalic and had much broader noses, all that is probably nothing but speculative junk.

      Slett
  11. Samequeen, what is your evidence for your claims?

    SvarSlett
  12. The anthropometric I described isn't the old racistic measuring. It's modern measuring on found humans in archaeological diggings, when a human is reconstructed.

    Here I use a famous swedish scientist (who have reconstructed danish bogfinds as an example of his work). Torbjörn Ahlström 1997.

    The description of interglacials come from a scientific paper on Sandarne findings, Gothenburg area, by Astrid Cleve-Euler.

    The botanic reference is to arctic plants that doesn't exist in the Alps, only in the mountains along the Norwegian coast.

    Birgitta Fossum, doctorial thesis on nortern Norways inhabitants "Vilka är våra förfäder?", Umeå universitet 2007.
    I have more works to this, as I've been studying this many years. Also Weinstock,Laufer about the reindeer.
    I'll come back with more later.

    SvarSlett
  13. Samequeen, I want to see genetic evidence (including ancient DNA stuff if available); it would give much more accurate and clear information about origins.

    SvarSlett
  14. Hey Anders,

    There seems to be something wrong with the (modern) Swedish K3 result, it doesn´t add up to 100%. I presume that the Med part should be 34,6% and if so that would be important because the Cau/Med quota would be in the same neighbourhood as the ancient Gotlander and thus implying no further immigrations from Southern or Eastern Europe.

    SvarSlett
  15. Thanks. I found the error and updated the table. The Swedes seems after this to be similar to Norwegians. I also found a similar error in the Adigei. Should be correct now.

    SvarSlett