May 9, 2011

More ancient mtDNA maps of Europe: the Neolithic and Chalcolithic

With these two maps of the Neolithic/Chalcolithic period, I will consider my mapping project more or less complete. It can still be improved with a map from the Bronze Age, I guess, specially as this is the time when mtDNA H seems to advance to Northern Europe, but by the moment I consider the collection quite complete.

As with my previous maps of the Paleolithic period, I have not just taken for granted the haplogroups reported by the authors (sometimes quite surprisingly) but I have checked myself which haplogroup they should belong to with the help of PhyloTree

The only exception are the peripheral Basque samples from Izagirre and De la Rúa, which I could not manage to find out which markers were detected, so I have mentioned them in a separate inset - because they are under the strict identification standards I have followed for all the rest.

The Neolithic and Chalcolithic ancient mtDNA maps are as follow:

The time-frame descriptive tags correspond to pan-European chronological frames, regardless of local development. 

Previously I posted (in November) three maps for the Paleolithic and Epipaleolithic (this last one including a contemporary sample from Syria/Kurdistan, which is already Neolithic). They were:

All the data I used came via the Ancient DNA pages of Jean Manco, though in some cases I had to browse the relevant papers as well. 

For the record, the first clear appearances of the following large haplogroups are as follow:
  • U2: 30 Ka BP: Kostenki, Russia
  • H: 25 Ka BP: Sunghir, Russia (H17'27). 
    • If you don't like this one, then Epipaleolithic Portugal (H1b)
  • U5 and JT (JT*): 18 Ka BP: Nerja, Andalusia, Spain 
  • U4: 12 Ka BP: Taforalt, Morocco
  • HV0 (probably V): Neolithic Portugal and East Germany
  • K: 11 Ka BP: Tell Halula (Neolithic Syria/Kurdistan)
  • N1, W, J and T: Neolithic Central Europe 
  • X: Neolithic West France 
I also spotted an "Oriental" lineage in Chalcolithic Castelló: either D1 or G1a1 (one of the five L3(xR) in the map).
    Disclaimer: I carefully checked all reported sequences against the most recent builds of PhyloTree, trying to assume nothing. Surely most sequences have been correctly identified as far as the data and scientific knowledge allow, however if you spot any mistake, please let me know. I hate mistakes.

    PS- Spreadsheet (ods format) that I used in this work is uploaded here (for one year).


    1. Despite the increasing number of data points, I still find it hard to make much sense of it. If anything, Derenburg reminds us that you need many samples from one site (or area) to get useful statistics and a useful breadth of haplotypes.

      There is perhaps some inflow from the East discernible just around and after the Younger Dryas (U5; Ahrensburg and Hamburg Cultures) along the northern edge of the ice sheets, but other than that, I don't see much yet to base any solid theory on.

    2. We can't take the data at face value for the Paleolithic: excepting Taforalt and Epipaleolithic Portugal, all the samples are minuscle.

      Said that and with the earliest U5 located in Nerja, Spain, I do not understand what you say about Eastern influences in Ahrensburgian.

      U2 does not make U*, much less U5 and U4. While there is U5 located in the East, this shows up much later than in Andalusia and it is roughly simultaneous to other samples from several European places.

      I think that U5 spread with Aurgignacian (or Gravettian) and so did H (harder to identify but surely there in many of the R* cases, as well as in the few ones that are clearly identified). More difficult for me is to me to pinpoint the source of U4 but probably spread together with U5, even if it has almost vanished (because of drift?) in the West or expanded beyond original numbers (because of founder effects in the East).

      I'd say that the original identifiable haplogroups of pre-Neolithic Europeans were H, U5 and U4. Possibly others too but not identified (except L3b2, identified in Epipaleolithic Portugal, which surely arrived from North Africa in the LGM). Also add some rarer lineages like U*-CRS, HV*-CRS, etc. which have been found on occasion both among fossil and living Europeans.

      It'd be interesting to find out the "true" origins of N1a, J, T, X, W, even V and K... the "Neolithic" clades, only located since the Neolithic expansion. K is the only one that has been found first in a West Asian sample (Tell Halula) - but there are no more West Asian samples so far, so it's like a little bit frustrating.

    3. Maju : "I'd say that the original identifiable haplogroups of pre-Neolithic Europeans were H, U5 and U4"

      I think there were probably (many?) others.
      The haplogroup generally reported as mtDNA N of the paglicci cave could easily have something to do with N1a (IIRC it's been shown a few months ago that some of the European N1a is probably not from a neolithic flow, am I wrong?).
      I think we could also maybe put a few HV0 (or at least HV0a) and V in the lot, maybe several others too.

    4. maju : "I also spotted an "Oriental" lineage in Chalcolithic Castelló: either D1 or G1a1"

      If I understood it correctly they found a M8 in Spanish samples.
      I had assumed it might have arrived with the Alans maybe as we can suppose they had some east Asian haplogroups among them, but if such Asian hgs are found so early in Spain, maybe we need a different explantion (this reminds me of this N9a in south east Hungary in the remains of the Körös culture).

    5. Obviously, there could be other haplogroups - but not many we can identify with any certainty. Blame the researchers for using only HVS-I in most cases. :(

      I'm just saying that these three seem clearly old and established.

      N* does not mean N1 or anything of the like. It just means N*. N1 is not "more N" than any other N subclade like X, W (N2), etc. Anyhow, all these Western N(xR) lineages seem to have higher diversity in West Asia and hence probably originated there.

      N* may have meant something else, now extinct or found in such small amounts that we don't even count it often. I must say that N1-derived I has been argued on occasion to be old in Europe (Gravettian?) but it is not known in the archaeological record before the Iron Age. Also the N1a variants found among Danubian Neolithic peoples have also been argued to be European-specific but this may be a fallacious argument (because the modern survivors with European-specific N1a might well be their descendants). N1 in any case keeps highest diversity in West Asia.

      I feel uncertain about J and T. Both but J specially have been often argued to be of West Asian Neolithic spread and the fossil record would support at least the "Neolithic" aspect of that claim, but JT* has been located in Andalusian Early UP and nowadays seems to be a most rare but real West Mediterranean clade. It is possibly not the ancestor of J and T but a third "sister" anyhow.

      "I think we could also maybe put a few HV0 (or at least HV0a) and V in the lot, maybe several others too".

      I have not located any such haplotype. HV0 should be easy to identify with HVS-I (transition at 16298), so it seems to me that not a single ancient sample before Portuguese Neolithic had it. It could well have been there but it has not been sampled.

      Of course the many R*, specially the R*-CRS, can well be HV (often H probably) but we cannot say that for sure. Similarly we would not be able to discern an "ancestor" of HV0 if it showed up as R*-CRS or R0*-CRS (Paglicci). But we would be able to discern HV0 as such and it does not show up before Neolithic (in Portugal and East Germany first of all).

    6. Not sure if this caused confusion but I wrongly said that HV0 (V?) was first found in Epipaleolithic Portugal. This is wrong (and has been corrected): it was first found in Neolithic Portugal and East Germany (in this case clearly V because coding region was analyzed).

    7. As for the "Oriental" lineages, I am not that much surprised of finding M8 or nearly any other kind of erratic in modern Iberian samples. After all there's a long history of globalization going on since c. 1500.

      But it is really intriguing that one D1 or G1 would have made it so far West so early.

      The reported sequence is 16223T, 16325C, 16362C and was "correctly" identified by the author as D, though more exactly would be D1 (or the alternative of G1a1, which has the same HVS-1 transitions).

      AFAIK G is almost never found in Europe, so D is maybe a safer bet. I understand that D and CZ (and maybe other M8 lineages) arrived to Europe with the Uralic peoples at some unclear point in the Epipaleolithic or early Neolithic. But still explaining its arrival to Castellnovo (Castelló Province) of all places is pretty hard. Maybe it explains (somehow) the "quasi-Mongoloid" looks of the Rafa Nadal type (an Iberian-specific look, IMO).

    8. Will you be covering this new bombshell study about neanderthals disappearance presumably being much older than believed?

    9. I don't think so. I was already persuaded that Mezmaiskaya Neanderthal presence ended c. 40 Ka., so I'm kinda puzzled but what I have read around among the conservative anthropology bloggers, who seem to live in another totally different planet. :/

    10. Staying in the off-topic (sorry), Argiedude: if you can read Spanish, there's an interesting debate on this paper at Mundo Neandertal. Essentially it seems it does not change much but is presented as if it would.

    11. [much thanks, and sorry for off topic]

    12. Where are the Balkans, Caucasians and Anatolians? Those are three key areas for all my Paleo-Meso-Neo questions. :)

    13. Nobody has researched them so early.

      There are a handful of Balcanic samples (Romania, Greece) from the Bronze Age but little more.

      See older maps (without revision: only reported haplogroups, which is often a total mess) here and Jean's page as well.

      There are no Anatolian or Caucasian aDNA samples of any age (Jean's site is pretty much comprehensive of all aDNA research).

    14. Thanks Maju. Jean's site is a treasure! I thought I had read that U3 is quite old in the Southern Caucasus (thought to be UP but no evidence, based on diversity I guess). Didn't see much proof for that in her data so I guess it came from modern data.
      I have enjoyed lurking on your blog and reading your posts on Deinekes. Cheers

    15. U3 should be from around the Black Sea based on modern distribution. Just that it has not been detected in the fossil DNA (as far as I recall).


    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).