April 6, 2014

Revised Lazaridis study on ancient ancestry of Europeans

The already famous Lazaridis et al. study on the contribution of various ancient populations to modern European genetics has gone through a revision which does not alter the fundamental conclusions reached in the past but does add some interesting nuances, new graphs and some new data.

Iosif Lazaridis, Nick Patterson, Alissa Mittnik, et al., Ancient human genomes suggest three ancestral populations for present-day Europeans. BioArxiv 2013 (preprint). Freely accessibleLINK (last version) [doi:10.1101/001552]

Most up to date supplementary info → LINK

For background see this previous entry.


Scandinavian hunter-gatherers deviate towards Siberia

Among the new data (or maybe data I skipped in the first read?) is the fact that the ancient Epipaleolithic individuals from Motala (Sweden) deviate towards Mal'ta-1 (Siberia), something that neither Lochsbour (representing Western hunter-gatherers) nor Stuttgart (representing early farmers) do.

This implies that there were already some differences in the Epipaleolithic era among European hunter-gatherers, with those of Magdalenian background lacking the Siberian (ANE) component, which is found however in Scandinavian ones (of Ahrensburgian background?) This may help explaining the extra ANE affinity in Northwest Europe, which is otherwise hard to understand. 

It also suggests that Eastern European hunter-gatherers were already in the Epipaleolithic more akin to Siberian ones than those from the West and South of the subcontinent, as well as those from West Asia (otherwise Stuttgart, which has partial West Asian ancestry would show increased affinity). Of course this can only be confirmed by direct analysis of Eastern European Paleolithic remains but seems quite likely in any case.


Principal component analyses (lots of them!)

Ancient samples projected on the global PCA:

Fig S1-10: projection of ancient samples onto global PCA dimensions 1 & 2

In this global PCA, EEFs overlap well with the reduced modern European sample (Basques and Sardinians only) and the West Asian one (Georgian, Palestinian Bedouins). However projected hunter-gatherers from Europe and Siberia show a clear "other Asian" deviation. Why? For the very same reason that South Asians and Melanesians do, even if they are clearly distinct populations: because the frame gives them no other choice: they are not quite like modern Europeans and they do not have any African tendency either, so the other populations that are somewhat akin to them are other Asians and there they go. 

That's why PCAs must always be taken with a preventive dose of salt: they are very nice visualization tools but they depend too much on the sampling strategy and its intrisical bidimensional limitations. 


Modern West Eurasians projected to a PCA of three ancient samples:

Fig. S10-3 Projection of West Eurasian populations onto the first two principal components
inferred using Loschbour, Stuttgart, and MA1 (full version).

Quite curious: all West Eurasians cluster tightly in comparison to their ancient "ancestors". It is likely that dimension 2 should be scaled down because the second component is always smaller than the first one (often around half). However I could not find a clear datum to proceed so I retained the original equal scale even if it can be a bit misleading. 

While not exactly, Lochsbour and Stuttgart explain the bulk of European (and West Eurasian!) ancestry, at least in comparison to the quite outlying Mal'ta-1 sample, representing ancient Siberians. 

Detail (zoom in) of this PCA:

Fig. S10-4 (I annotated the three ancestral tendencies with arrows for easy of view)

As expected Eastern/SE populations deviate more towards Stuttgart, Western/NW do towards Lochsbour and in general Northern populations deviate slightly towards Mal'ta, also in West Asia (i.e. Iranians, Turks).

Modern European PCA with ancient samples projected on it:

Figure S10.5: Projection of ancient samples onto the “European” PCA (annotations in gray are mine)

Something the authors notice is that their PCA does not approximate a map of Europe, as happens in some cases. They dedicate some time to evaluate this discrepancy, comparing with the PCA of Novembre et al. 2008. The differences are caused because the latter used many more NW and Central European samples and instead had way too little Eastern European ones. 

This highlights that sampling strategy is of utmost importance when analyzing autosomal DNA, not just in PCAs because oversampled populations would tend to cope the axes (or components). This should be obvious but is way too often ignored, what may result in spectacular magic hat tricks but hampers serious science. 

This is one of the reasons why I do not trust too much autosomal DNA analysis: lack of the fundamental rigor. This is of course not a defect in this particular study but it happens often in many others, scholarly and amateur alike. 

The authors believe that the main axis of differentiation in Europe when the subcontinent is considered as a whole may tend to Northeastern Europe rather than SSE/NNW⁸, something that is consistent with their ancient admixture findings elsewhere in the study. 


FineStructure PCA output

A key point in this study is that only Sardinians and Basques can be modeled as simple EEF-WHG admixture, all the rest of Europeans needing of the MA1 component to be explained. This is much easier to visualize in the following graph.
We also processed the ChromoPainter/ChromoCombine output with fineSTRUCTURE1 using 250,000 burnin and 2,500,000 runtime MCMC iterations. Fig. S19.2 shows a Principal Components Analysis by fineSTRUCTURE which strongly resembles that of Fig. 1B.
Fig. S19-2 (annotated in gray by me)
Here we can see that only Sardinians (and quite insistently Canarians), Basques and some populations related to these (North Iberians, South French) are actually close to the Stuttgart-Lochsbour axis. All the rest need a third ancestry for explanation, which is approximated by MA1 (not plotted but whose tendency I annotated).

Notice that in this case the two ancient samples are not projected, as in the previous graphs but actually computed as part of the wider West Eurasian population.

This does not deny that other NE European populations have greater affinity to Lochsbour (fig S19-3) but it seems clear that this affinity must be mediated by another branch of ancient European hunter-gatherers, one that existed in Eastern Europe, rather than in the West, and that it had more ancient Siberian (MA1) affinity. It is also likely that this Eastern European aboriginal population was the one which brought the extra MA1 affinity to the rest of Europe, most likely in the context of Indoeuropean (Kurgan) migrations. Along with it they probably also brought extra WHG-like admixture (but actually from an Eastern European source).

The extra MA1 tendency is also present in West Asia. This may have two alternative or complementary explanations:
  1. There have been also significant Siberian-like intrusions in West Asia after the Neolithic.
  2. This extra ancient Siberian affinity is in fact (largely?) pre-Neolithic but the founder population of West Asian roots which triggered the European Neolithic in Thessaly was particularly removed from this admixture and more akin to Palestinians or peninsular Arabs than to other West Asians.


Sicilians, Maltese and Ashkenazi Jews are different

These three are the only European populations which have a poor fit with the triple admixture model (EFF+WHG+ANE), suggesting that they have fourth party inputs, most likely extra admixture from West Asia. 

This is apparent in the previous graph too (among several).


Tree modeling for the origins of the ancestral populations

From fig. S16-2 (allowing for five admixture edges)

From fig. 16-4 (full-genome coverage, allowing for 5 admixture edges and using Dai instead of Onge)

The basic topology of the tree is consistent (excepted the partial change of the location of Karitiana Native Americans, which depends on the greater affinity of the East Asian sample used and is essentially neutralized by the admixture edge with Onge or Ma1 respectively). The main admixture events are:
  1. The Karitiana (and Amerindians by extension) are clearly a mix of East Asians plus Ancient Siberians of Western affinity (MA1), which is represented differently in both trees.
  2. Early European Farmers (Stuttgart, Iceman) have clear "Basal Eurasian" admixture (which can be interpreted as North or East African input and/or a residual ancient Arabian element, probably both)
  3. La Braña also has "Basal Eurasian" admixture (surely from NW Africa, what implies that the North African component in Western Iberia is pre-Neolithic)
  4. Motala has Ancient Siberian (MA1-like) admixture
  5. Mal'ta 1 probably has some East Asian admixture
  6. Ötzi the Iceman might have some Western Hunter-Gatherer admixture (~3%)

Of the three Western branches, MA1 is the more distant one. That implies that West Asia and Europe were also exchanging genetics in the Upper Paleolithic, while Siberia remained more isolated in comparison. That stands even when East Asian admixture into MA1 is accounted for. 


A Lochsbour's cousin in West Asia

The authors compare and analyze many models of possible admixture leading to the known ancient and modern populations. They seem to favor this one in the end:

Figure S14.20: A model for Near Eastern populations with Ancient North Eurasian admixture.
Stuttgart is a mixture of Near_East and a sister group of Loschbour (UHG: Unknown Hunter-
Gatherers); A Test population (shown here) is a mixture of Near_East and a sister group of MA1.

This scheme suggests that ancient West Asian ("Near East") populations were closer to ancient Europeans (Lochsbour) than to ancient Siberians (MA1). It also suggests an unknown relative population of Lochsbour (UHG) as partial ancestor of early European farmers (Stuttgart). This population is speculated to have lived in the Balcans. 

An issue here is that most modern West Asians and all Caucasian peoples actually have too much MA1 affinity to be a good fit for the (ancient) Near East proto-population concept. About 12-13% among West Asians (Cypriot, Druze) and as much as 29% among Caucasians. They are actually more like "Test" than like "Near East".

The authors conclude that they don't really know if this extra MA1-like ancestry is old or recent. If old, it would imply either two different populations of West Asians or, as they say, the expansion of a West Asian population with extra "Basal Eurasian" ancestry. 

This brings us to a key question: what is actually "Basal Eurasian"?


What is "Basal Eurasian"?

Notice that "Basal Eurasian" is defined as phylogenetically intermediate between the Mbuti and Eurasian-plus populations. Quite misleadingly the node is described as "Non-Africans" but that does not need to be true at all. It is just downstream of one of the most ancient African sub-branches, that of Pygmies, so it can still represent African populations which are or were closer to the out-of-Africa branch. 

There is no formal ascertainment whatsoever of what is "Basal Eurasian", no comparison with other African populations and no even formal consideration of the (very likely) possibility of various isolate ancient populations existing in NW Africa or Arabia. This is clearly a flaw. 

A key piece of information here is that La Braña (ancient NW Iberian hunter-gatherer) consistently shows "Basal Eurasian" admixture. This admixture is much more likely to have arrived from North Africa than anywhere else. NW African genetic markers are still apparent in Western Iberia in fact and there is strong archaeological support for Iberia-NW Africa interaction in Solutrean/Oranian times. 

We can only consider in fact this "Basal Eurasian" idea as a mere indicator of African-like affinity, even if it's not Mbuti but something else. This something else can be in fact several things:
  • NW African Aterian residual in the case of La Braña
  • Arabian OoA residual influence in the case of EEF
  • Direct NE African admixture in the proto-EEF West Asian population, strongly indicated by the NE African E1b-M78 (notably its subclade E1b-V13) in ancient Neolithic and modern European Y-DNA.
I am particularly inclined to suspect an almost direct migration from Palestine to Thessaly at the origins of European Neolithic. After all both non-European lineages found in early European farmers (E1b-V13 and G) are common in that area. But of course an Anatolian intermediate station cannot be excluded.

In any case I'd suggest to change the terms "Basal Eurasian" and "non-Africans" by something more neutral, maybe "Ultra-Mediterranean" and "Proto-Eurasian" respectively, where both concepts are allowed to be African, at least partly so.


Update (Apr 23): see also here for some curious aspects of Lochsbour's IBD ancestry.

64 comments:

  1. Quite curious: all West Eurasians cluster tightly in comparison to their ancient "ancestors". It is likely that dimension 2 should be scaled down because the second component is always smaller than the first one (often around half). However I could not find a clear datum to proceed so I retained the original equal scale even if it can be a bit misleading.

    Maju, quick note on this, I'm pretty sure the ancient samples are far apart compared to the recent here because of the PCA is largely based on their intra individual variability.

    If you do a PCA based on, for example. 1 each of present day South Asians, Europeans and North Africans, you'd get a similar spread when you projected other present day South Asians, Europeans and North Africans onto it, with each of the populations bunching up near the middle and overlapping noisily and tilting in the direction of their "representative".

    This is because most what the PCA found based on the 1 each of the individuals simply won't be related to population level variation.

    (For another comparison, if we did a Maju-Matt-someone else PCA, then our respective populations wouldn't tilt very much towards either of us, because most of our genetic differences are individual to us, so they'd sit more or less at around 0 and overlap a lot, with only slight tendencies towards us (perhaps unless they were actually close family with one of the 3).)

    IIRC, there's a similar figure in the recent Anzick genome paper that does the same on Anzick vs some other Native American samples to this effect.

    When lots of individuals are included, intra-individual variance gets cancelled out, which is why with more population samples (even 10 individuals from each) we get nice smooth PCA.

    On a similar note, I'm not divided on whether the axes would really need much rescaling here. On the one hand, the null assumption would be there's no reason for more intra-individual variability between any dyad of these individual and so each dimension mostly representing a contrast between one individual and the other two should be about the same size... But if there were, by chance, that would be distorting. It would have been interesting.to know the relative size of each anyway..

    ReplyDelete
    Replies
    1. Interesting meditation, Matt. Honestly I can't say if you're right or wrong but I can ascertain that it is not the same "projecting" than factoring samples together, as seems quite obvious from the comparison between the variousprojections and the FineStructure PCA. It is interesting that at least the Lochsbour distinctiveness is enough to get him to dominate one of the polarities on his own, in spite of being a single individual.

      Delete
  2. Maju,
    I think you hit the nail on the head questioning the assumptions on "Basal Eurasian". A commentor on Davidski's site suggested the so-called "basal" admixture coming from the Natufians who, among others, were some of the earliest people to be absorbed by the earliest NE farmers. As you mentioned, aboriginal peoples of Arabian peninsula are probably worthy of more scrutiny since their skeletal features would suggest significant African ancestry and they certainly mixed with the early famers.

    Early Southern Europeans, on the other hand, may have already had some early African admixture in the late Paleolithic, at least from a Grimaldi-like population. That, or another source, may have contributed to Mesolithic La Brana's heritage, which was significantly more African shifted that modern European populations. (thanks for posting the thumbnail of La Brana, btw) I believe Coon had originally hypothesized that early brutish Western Eurasians were originally gracialized by African admixture, leading to Mediterranean Caucasoids.
    In any case, I don't think we know for certain that ANE, WHG and EEF are three discrete things instead of one thing, mixed with three unknown things. Given that the EEF people must certainly have been a racially mixed population, I think we may be looking at the latter.

    ReplyDelete
    Replies
    1. NW African influence in Europe seems quite restricted to Western Iberia and in any case, other than punctual founder effects like some Welsh village whose name I don't recall, it has remained restricted to the Iberian peninsula or almost so. So it's not something that can be generalized to all Southern Europe.

      "a Grimaldi-like population"...

      Long not heard of Grimaldi! Actually it was demonstrated years ago that Grimaldi is a typical Caucasoid specimen with some misleading post-mortem deformation of the skull, particularly the jaw. It's a bit like the Portuguese boy skull claimed once to be a half-Neanderthal and which is actually post-mortem deformation of the skull by mere gravitational compression under the soil.

      There should not be anything too obvious in the appearance of NW African admixed people because, on one side, NW African themselves are dominantly West Eurasian by origin (and partly Iberian in fact) and, even more importantly, the score of their genetic influence is rather low: less than 10% today, not much more in the past, judging on La Braña and some other scattered data.

      It seems to me that Gravetto-Solutrean South Iberians migrated to North Africa in large numbers at the very genesis of Oranian culture (what has left ~30% Iberian like mtDNA and autosomal legacy in the area) and that, in return, some NW African ancestry reached the Western parts of Iberia (notably Estremadura and Asturias initially, later expanded to all the Western third, which was not as populated).

      Of course I can't exclude that this backflow is of a later date, Neolithic for example, but the La Braña data seems to suggest that it's not the case.

      " I believe Coon had originally hypothesized that early brutish Western Eurasians were originally gracialized by African admixture, leading to Mediterranean Caucasoids".

      I wouldn't bet my money on Coon's Nordocentric theory, really. If he can't discern a Spaniard from a Greek, he's obviously blind to "racial" variation. When I used to read on the subject years ago I rather preferred Lundman, although he also has some of that Nordic bias.

      My own opinion is that a robust individual like Crô-Magnon 1 is simply exceptional, individual variability, and that most other remains are more average. But well, I lost most of my interest on these matters long ago because physical anthropology is a very slippery terrain with low informative value, at the very least when compared with genetics.

      "I don't think we know for certain that ANE, WHG and EEF are three discrete things instead of one thing, mixed with three unknown things".

      Indeed. Particularly ANE seems a bit slippery and is probably rather a proxy for Eastern European Hunter-Gatherers, who must have more ancient Siberian admixture (but otherwise were relatively close to WHG.

      EEFs were for sure a mixed population, however I would not say "racially mixed" because they are the ones who best fit with the modern "Caucasoid race" samples. There's no such thing as a "pure race", only variable mixtures.

      Delete
  3. 'Basal Eurasian' has always been right under your nose Maju, so to speak, it is nothing new really, Cavalli Svorza outlined almost 20 years ago that Europeans had 1/3 African Affinity. I had also, as I am sure you do recall, attempted to hammer in the very simple and basic concept that at K=2, West Asians showed significant Affinity to Africans relative to East Asians. This Basal Eurasian statistical concept is part and parcel of this rudimentary and simple observation that is most often ignored (intentionally?) by geneticists, amateur and professional alike, now the proverbial chicken is coming home to roost.

    Global contour map, K=2

    The World at K=2

    ReplyDelete
    Replies
    1. But Luigi Cavalli-Sforza was not right, exactly the same you are not: K=2 at the global level is always a function of sample size, just try super-oversampling Australian Aborigines or Khoisan peoples (this was done in fact, as I showed you to no avail) or any other very distinctive population and all the rest will cluster as non-them or intermediate between the super-sample and the "non-them". Of course that the level and pattern of distinctiveness also matter but sample size does too and Admixture/Structure will always be a function of these factors: nothing more, nothing less.

      At K=2 global you have such a brutally bad cross-validation value that nobody would think that means anything at all. But there you go beating the dead horse.

      It's plain and clear.

      West Eurasians have some African and surely also quasi-African (OoA isolates') admixture, many Africans also have West Eurasian admixture, even more than the inverse (for example Ethiopians seem to be a 50-50 mix). But you can hardly tell from a global K=2 because it's just meaningless noise.

      You do some interesting stuff, Ethio but in this issue you are absolutely wrong.

      Delete
  4. "We can only consider in fact this "Basal Eurasian" idea as a mere indicator of African-like affinity, even if it's not Mbuti but something else."

    It shows affinities to Mbuti as well, assuming Sardinians as the study seems to show clearly are a good proxy for EEF.


    I also posted this on Davidski's blog, but it might be of some value in discussion here as well. Below is a Chromopainter/FINESTRUCTURE heatmap of world populations from Anders Pålsen's project. The affinity of Sardinians to the African groups, be they Ethiopians, Khoisan, West Africans or Pygmies, is the highest in Europe.

    https://sites.google.com/site/fennobga/CCAggrWorld240413.png

    I fully agree about the study being lacking when it comes to exploring what the hypothesized Basal Eurasian really is.

    ReplyDelete
    Replies
    1. "It shows affinities to Mbuti as well"...

      What's your foundation to claim that? All I can see in the TreeMix algorithm is that it is somewhere between Mbuti and West Eurasians, what can be almost anything. I would say, based on the full genome coverage ones that it is quite downstream of the node of Pygmy divergence, something like "North African" (very loose sense) but the matter merits more investigation in any case, no doubt.

      "Below is a Chromopainter/FINESTRUCTURE heatmap of world populations from Anders Pålsen's project."

      That kind of hyper-high resolution image really clutters my computer, and possibly others. I can't see anything with so many tiny names anyhow, even if it manages to load. It's something I've told Anders but to no avail.

      "The affinity of Sardinians to the African groups, be they Ethiopians, Khoisan, West Africans or Pygmies, is the highest in Europe".

      Maybe but that only means an African tendency of the same uncertain kind that we are discussing when talking about this "Basal Eurasian" mystery. I'm quite sure that they are not more akin to Africa than North Africans or Arabs: it's a tendency.

      My question is: is that African-like tendency or minor component wholly African or partly something else (residual aboriginal Arabian, for example, retained in the area since the OoA migration, but very diluted anyhow)? Whatever part is African, can we identify it with the trail left by yDNA E1b-M78/V13?

      And also: why do modern West Asians appear to lean less in that direction (excepted Palestinians and peninsular Arabs? Is that because the European founding farmers arrived from Palestine, maybe in a coastal migration, or is it that the demographics of West Asia has changed quite a bit since those times?

      Delete
    2. The Chromopainter affinity, and assumption that Sardinians are basically a WHG-shifted EEF-population, is my foundation for the claim. Sardinians show more Mbuti affinity than any other Europeans. It does not need to be direct Mbuti admixture though, any kind of African is enough to increase the affinity to Pygmies and Khoisan given the diversity of their autosomal haplotypes.

      "Maybe but that only means an African tendency of the same uncertain kind that we are discussing when talking about this "Basal Eurasian" mystery. I'm quite sure that they are not more akin to Africa than North Africans or Arabs: it's a tendency."

      It still has to come from somewhere, the problem is that modern North Africans and Arabs can be expected to also have recent Sub-Saharan admixture which complicates investigations. As for modern West Asians showing the shift, depends a lot. Cypriots or Druze show more African affinity than Sardinians in Chromopainter/FINESTRUCTURE and do have more Middle-Eastern like ancestry. One would have to go to Caucasus to find ones with equivalent shift to Sardinians, and I don't think there's any doubt about demographic shifts happening there. Same goes for Anatolia, mesolithic aDNA from there would no doubt be informative about the spread of farmer ancestry.

      "My question is: is that African-like tendency or minor component wholly African or partly something else (residual aboriginal Arabian, for example, retained in the area since the OoA migration, but very diluted anyhow)? "

      I think it's residual African in Middle East. I wouldn't go as far as to say it's been preserved since original OOA, there's more likely been several African-ME contacts with small effects between Neolithic and the OOA event. I'm a bit wary about linking autosomal DNA and haplogroups, but let's say that in this case I don't see other possible links to Y-DNA for the pre-neolithic Africa-ME mixtures than E1-variants. After it got spread into the Middle Eastern population any migration from the area would have carried it.

      Delete
    3. I would need a screen of the side of the wall (and a better computer possibly) to be able to study such a huge image. Or at the very least I would need the graphic's coordinates and key ID of the populations mentioned to find it. Or simpler: I will take your word for it because once you refined your original statement, it does not seem different from what I can appreciate in the data from this study: an African-like (or negative-Eurasian) tendency, which can be anything African or para-African.

      "It still has to come from somewhere, the problem is that modern North Africans and Arabs can be expected to also have recent Sub-Saharan admixture which complicates investigations".

      If an amateur like myself could find it in the first relevant Admixture tries, any researcher with the insight can, using much more powerful tools (or the same ones but refinedly so). Naturally you do not stay at K=2, K=5 or K=8 levels, you need to go beyond K=10. You also need a wise sampling strategy: what is what I want to study, possible comparisons. And you need to evaluate the relative Fst distances between components. There's no point on doing a K=5 or K=8 global Admixture graph each time we analyze something: the resulting graphs are shallow and the populations studied unnecessarily too large, complicating things a lot. That's OK for a preliminary or generic study but not for "microscoping" this kind of stuff.

      That's why in this Lazaridis' paper Admixture takes such a low prominence: it merely says nothing we did not know already. Maybe if they had tried another sampling strategy approach... we would be looking at different images with similar results.

      "As for modern West Asians showing the shift, depends a lot. Cypriots or Druze show more African affinity than Sardinians in Chromopainter/FINESTRUCTURE and do have more Middle-Eastern like ancestry. One would have to go to Caucasus to find ones with equivalent shift to Sardinians"...

      That's not the data-point I gathered: ~10% MA1-shift in Cypriot and Druze (they don't mention others). From the PCAs, particularly figh. S19-2, Turks and Iranians are even more shifted, with ANE affinity not far from that of Caucasians, for which is reported ANE affinity levels of as much as 29% (higher than any European). This is of course just a preliminary assessment because without ancient West Asian genotypes for comparison all becomes quite blurry, but that's at least what seems from comparison with EEF and WHG, for whatever is worth. Naturally Turks and Iranians are the most steppe-influenced populations in the area but they should also be more influenced than Caucasian peoples, which are believed to be refuge populations, so... still confusing.

      It's not impossible that Northern Ancient West Asians from the Zagros-Taurus arch were more ANE-like than those from the Levant. They were almost certainly two distinct populations and it's even possible that Eastern Europe had influenced the former at some point. If this is correct, then the EEFs had a founder effect from the Levant rather than from the Northern Highlands. But much of this can only be speculative until ancient West Asian genomes are studied in greater depth than just mtDNA sequences.

      ....

      Delete
    4. ...

      "I think it's residual African in Middle East. I wouldn't go as far as to say it's been preserved since original OOA"...

      I tend to that idea because of E1b-M78. However, I am also reasonably certain that Saudi Arabians and some Low Egyptians (but not others) retain a component that appears equally distant (at "continental" distance scores) from Africans and West and East Eurasian main components alike.

      The same happens in NW Africa (particularly Southern Morocco) but it is a different component (ref. links in main entry). So I believe that isolate OoA populations existed in these two regions but were eventually overrun by mostly West Eurasian (UP) and possibly some secondary African expansions, remaining however in form of a very diluted component and some more or less notorious mtDNA lineages (whose L(xM,N) adscription make them appear as "African" but are often distinctive).

      Sadly, it will take time for academic research to confirm this because those areas are not so well studied as Europe is (neither in the genetic nor in other aspects like archaeology) but I am personally very persuaded that it is the case: there is an "Aterian" and a distinct "early OoA" residuals in these two regions and they can be found if you pay attention, if you study the genome with sufficient depth and the proper sampling strategies (just overloading your samples is of no help: better to select just specific controls and a good sample of the studied regional population).

      "there's more likely been several African-ME contacts with small effects between Neolithic and the OOA even"...

      That I do not know, except for the Eurasian back-migration to parts of Africa, which seems confirmed by just looking at haploid DNA (and confirmed by autosomal DNA) as follows:
      1. West Asian flow into NE Africa, probably in the early UP, leading to the LSA.
      2. Iberian flow into NW Africa, surely at the Oranian genesis, of Solutrean influence.

      But in what regards to secondary flows from Africa to those areas, I can only discern the E1b (and related mtDNA) current, which seems rather Mesolithic than Paleolithic (and some backflow in West Iberia related to point #2).

      "I'm a bit wary about linking autosomal DNA and haplogroups".

      Me too but, on one side, mtDNA seems to show often a good correlation with autosomal DNA (probably because mothers carried both on in spite of fathers' changes) and, on the other, yDNA usually shows relationship with minor autosomal elements in populations which have gone through such extensive admixture that their patrilineal origins are hard to recognize beyond the yDNA (examples: Native Americans, Uralic peoples, Palestinian-European Neolithic vector of African origin). So it is not so thoughtless when properly pondered: there may be and often are relationships, just that we have to look at them from all angles before reaching to conclusions.

      " I don't see other possible links to Y-DNA for the pre-neolithic Africa-ME mixtures than E1-variants. After it got spread into the Middle Eastern population any migration from the area would have carried it".

      Yes. But E1b variants in West Asia are only important in the Levant and Arabia Peninsula, precisely the populations which show the least ANE-like drift. Turkey for example only has c. 10% E1b, much less than Greece or Albania. So this, like the generic "Basal Eurasian" issue, points to a Palestinian affinity of EEFs, which may be interpreted as either direct origin or as radical population changes further North after the Early Neolithic.

      Delete
  5. "I would need a screen of the side of the wall (and a better computer possibly) to be able to study such a huge image. Or at the very least I would need the graphic's coordinates and key ID of the populations mentioned to find it."

    Odd, I don't have any problems doing it with my 4 years old laptop using the scroll feature.

    "That's not the data-point I gathered: ~10% MA1-shift in Cypriot and Druze (they don't mention others). From the PCAs, particularly figh. S19-2, Turks and Iranians are even more shifted, with ANE affinity not far from that of Caucasians, for which is reported ANE affinity levels of as much as 29% (higher than any European)."

    I'm specifically talking about African shift of Cypriots and Druze being > that of Sardinians. African shifts don't show on West Eurasian PCA's with no Africans though. They have more ANE than Sardinians, but are much more Middle Eastern and less WHG, so they are more African shifted than Sardinians in global PCA's. To make the African shift of a West Asian or Middle Eastern population lower than that of Sardinians, lots of ANE is needed. This can be found in Caucasus.

    You can see the African shift of Georgians (higher than Basques but lower than Sardinians) in Dimension 1 of Fig S1-10.
    Any high-resolution global PCA will show Middle Eastern and Caucasus populations as African-shifted compared to non-Southern Europeans. ANE-shift of West Asians (and WHG shift of Europeans) is more visible on the Sardinians vs East Eurasians axis (usually dimension 2 of any global PCA that has Sardinians).

    Personally I think some small admixture events with Africans between OOA and Neolithic are the most plausible explanation for the creation of "ancient Middle Eastern" or "Basal Eurasian" which is the African-shifted non-WHG component in Stuttgart. If there was no African beyond the initial event into the Middle Eastern population, an OOA isolate would look as diverged from Africa as Mal'ta boy who, I assume, had no African influences beyond OOA from his ancestors and places accordingly compared to Africans. The WHG component the farmers could have acquired in Europe. Only ancient autosomal DNA from Middle East and North Africa can settle this though. To prove existence of OOA isolates in the Middle East into the neolithic or Mesolithic, much more ancient DNA is also needed to find out what an OOA looks like in the first place. The Ust-Ishim genome might not be old enough.

    ReplyDelete
    Replies
    1. "I'm specifically talking about African shift of Cypriots and Druze being > that of Sardinians."

      Alright. It's true that the African-like shift is not captured on those graphs but it does seem to be in general opposite to the Siberian (ANE-like) shift. So Sardinians who have a very low ANE score (c. 4%) correspondingly show a relatively high African-like score. On the other hand WHG (La Braña excepted) do not show either deviation, their distinctiveness is unique of their own, so they probably had no meaningful Siberian nor African input yet (at the very least they are a reference baseline).

      "They have more ANE than Sardinians, but are much more Middle Eastern and less WHG"...

      Logical, isn't it? The West Asian ancient populations surely were only distantly related to WHG, barely more than MA1, judging on reconstructive models. The non-European element in EEFs (and hence in Sardinians) seems to be made of Ancient West Asian plus some "Basal Eurasian" (i.e. African-like). To that we should add the UHG which would be related to Lochsbour but possibly Balcanic by origin.

      "they are more African shifted than Sardinians in global PCA's"

      I don't perceive that, sincerely. In the main PCA (see previous entry), Bedouins are clearly African-shifted but Cypriots or Druze are about the same as Norther-NE Europeans in this regard. Difficult to judge because there are no African-specific controls but that's the main stretch axis of West Asians: between the African-shift o Bedouins and the Siberian-shift of Caucasian peoples and other "highlanders".

      "You can see the African shift of Georgians (higher than Basques but lower than Sardinians) in Dimension 1 of Fig S1-10".

      You must mean S10-1 (there's no S1-10), the global plot. That is probably misleading in fact, exactly the same of what I said about South Asians, Melanesians and ancient hunter-gatherers looking as intermediate towards East Asia. Other studies using West Eurasian and North African populations do not show such tendency in fact. Sampling strategy is crucial, global analyses are relatively shallow and potentially misleading. In this case they are probably pulled in the African direction by Mozabite and other African "West Asianness" rather than by any genuine "Africanness".

      "Any high-resolution global PCA will show Middle Eastern and Caucasus populations as African-shifted compared to non-Southern Europeans".

      Maybe but it's only an illusion caused by the unspecific globality of such sampling strategy. As I just said: it's almost certain that what pulls them towards Africa is the "West Asianness" of some Africans rather than the opposite. Global comparisons alone are not too informative. Please don't be misled by them without due contrast.

      ...

      Delete
    2. ...

      "If there was no African beyond the initial event into the Middle Eastern population, an OOA isolate would look as diverged from Africa as Mal'ta boy"...

      No! Because Mal'ta, as WHG, etc. belongs to the Asian clade which migrated to South Asia (and possibly even reached SE Asia in some instances, only to turn back at some point). That is the "Eurasian" clade. Everything else will look as partly African-like, just because it was neither this nor that: neither "Eurasian" nor "African". Actually the category "African" is extremely diverse, even in spite of the semi-homogeneization caused by Bantu expansion, and includes all non-Eurasian basal populations of Humankind. These aboriginal Arabs (OoA remnant) would look on first sight "African" or partly so just because they diverged before the Eurasian expansion proper, which began only when their descendants reached South Asia. The same happens with NW African aboriginal population isolates of Aterian roots, which may have remained on their own for some 100,000 years. So they look "African" because "African" means everything before the Eurasian-specific expansion, which began in South Asia c. 100 Ka ago (judging on the latest archaeological data).

      Comparatively there is such a strong founder effect in the Eurasian-plus macro-population that all the rest looks like a single amorphous clade by (shallow) comparison and we identify it with Africa, logically. But part of that pre-Eurasian may well have been in Arabia since the OoA.

      "To prove existence of OOA isolates in the Middle East into the neolithic or Mesolithic, much more ancient DNA is also needed to find out what an OOA looks like in the first place".

      Not necessarily so. As I said I believe I found it in modern autosomal DNA: a minor component that has continental-size Fst distances to both Eurasians and Africans alike. Same in NW Africa. I also think it is apparent in mtDNA and even in some Y-DNA like A00 and A0, which must have arrived to NW Africa in very ancient times, judging only on their frequencies and rarity.

      → http://forwhattheywereweare.blogspot.com/2011/12/north-african-genetics-through-prism-of.html
      → http://forwhattheywereweare.blogspot.com/2012/01/egyptian-genetics-in-regional-context.html

      Is it enough evidence? Maybe not yet but it is a quite strong principle of evidence.

      Anyhow there's no reason to think that the ancient populations of Arabia or NW Africa left zero legacy. They may have been overrun by others, even repeatedly so, but there was no inter-species barrier of any sort as happened with Neanderthals, which hindered admixture. So in principle some of their legacy must remain. We just have to look for it.

      Delete
  6. " In the main PCA (see previous entry), Bedouins are clearly African-shifted but Cypriots or Druze are about the same as Norther-NE Europeans in this regard. "

    That PCA is not informative regarding sub-saharan African affinity because they are not present on the plot. For that matter, to truly explore relationship of Eurasians and sub-saharans, we need sub-saharans in the comparison. IBS and Finestructure, and PCA's all require this.

    "No!..."

    My point is that there is additional African output in the Middle East and North Africa including pre-neolithic Middle East compared to the original OOA event which would also have brought with it everything that is in the "Eurasian expansion" and MA-1. If we leave the special case of Oceanians out, everything Eurasian is ultimately African + drift (and some Neanderthal but that is equal for MENA and wider Eurasia) and any hypothetical OOA isolate would have drifted as far from Africa as something like Mal'ta did unless it had additional African input.

    "As I said I believe I found it in modern autosomal DNA: a minor component that has continental-size Fst distances to both Eurasians and Africans alike. Same in NW Africa. I also think it is apparent in mtDNA and even in some Y-DNA like A00 and A0, which must have arrived to NW Africa in very ancient times, judging only on their frequencies and rarity. "

    The problem is that we'd need ancient DNA to compare such results with, as well as more distant African groups and East Asians. That analysis is within Mediterranean context without wide comparisons.

    In any case, my views about what so-called Basal Eurasian is are clear. I doubt further analysis of Stuttgart will change them.

    ReplyDelete
  7. "That PCA is not informative regarding sub-saharan African affinity because they are not present on the plot".

    We know from other tests in this study that Bedouins have African affinity or admixture but otherwise you are correct, I mean: in purely abstract terms.

    But there are other studies, for example check this old entry: the Mediterranean divide is pretty neat and more or less equal for all populations, even Asturians who have greater E1b-M81 than the rest.

    Check also this one, which includes Sardinians in the pop-res analyses. Sardinians are not significantly more Yoruba-like than CEU at all.

    "My point is that there is additional African output in the Middle East and North Africa including pre-neolithic Middle East compared to the original OOA event"...

    According to my own research, see links above, Palestinians have 1.2% East African ancestry and Turks, Saudis, etc. 0% or such a tiny residual that is effectively zero. North Africans like Egyptians or Moroccans have a greater amount: 8-12%. Instead the "Arab2" component, which fits with an OoA origin, is 12% among Saudis and 11% in one but not the other Egyptian sample.

    Also the South Moroccan component (Aterian IMO) is almost 15% strong in Southern Morocco, where pseudo-Khoisanid phenotypes are found, but only ~1% in other parts of North Africa. Southern Moroccans are also outstanding in West African (Mandenka) affinity: c. 15%, while other North Africans have variable figures around just 5%. In The North African study Saudis showed 0.4% of Mandenka affinity plus a 0.2% of Fulani affinity and a 3.4% of Ethiopian affinity (however this last component is clearly a mix of West Asian and East African and only shows up intermittently).

    The direct Tropical African influence is not that important in West Asia or even in much of North Africa. And certainly is not more important in many cases than the aboriginal regional components from OoA times (estimated on Fst distances).

    "... any hypothetical OOA isolate would have drifted as far from Africa as something like Mal'ta did unless it had additional African input".

    Not at all: the direction of Ma1 and other Eurasian hunter-gatherers in the PCA (projected, remember!) is caused by their intrinsic Eurasianness, i.e. founder effect at the South Asian origin of the expansion: they have something to attach to: Eurasian populations. However the OoA residuals have nothing to attach to, so, if they existed as such pure individuals, they would split their allegiance (nil) between Eurasia and Africa, hanging between them both, just like Ethiopians.

    "... we'd need ancient DNA"...

    Sure: the more, the merrier.

    ReplyDelete
  8. Would this "basal Eurasian" admixture explain why Davidski's work was showing La Brana as having SE Asian / Melanesian admixture? http://eurogenes.blogspot.ca/2014/03/4-ancestors-oracle-results-for-anzick-1.html

    That's the deepest split in the Out-of-Africa tree, isn't it? Would other OoA isolates cluster with groups like Melanesians and Adamanese simply by virtue of not being on the same branch as the "derived" Eurasians.

    La Brana was Y haplogroup C-V20 though. C-V20's closest relative is C-M8, which is found in Japan and believed to be associated with the Jomon culture. I'm not sure how that can be reconciled into this picture.

    With regards to admixture from an OOA isolate or directly from tropical Africa, couldn't both be correct to some degree? The Sahara wasn't exactly a huge barrier during the timeframe we're talking about.

    ReplyDelete
    Replies
    1. It could be the case: I have seen (and commented here) various apparent minor component affinities of La Braña in the past, mostly towards Africa (varied subgroups) but sometimes also towards Asia. This fluctuation may well be because none of them really represents at all the original population behind that component, which could well be Aterian, so it sticks to some oddball rather randomly instead.

      "That's the deepest split in the Out-of-Africa tree, isn't it?"

      I'd say it's not a perfect tree but if we want to simplify the Eurasian population (excluding ancient Arabian left-behind remnants) it is probably best to set the first split between East and West at approx. the Indian-Burmese border. Then the Eastern branch splits between continental and Australasian and the Western one between South Asian and West Eurasian, probably some time later.

      But it's a bit more complex than that. Personally I think that the division between West (originally South Asia) and East was blurry and included some migrations forth and back between parts of India and SE Asia. It only consolidated later on, about at the same time when the Western Eurasian colonization began (early UP) and the first outreach to NE Asia did as well. All three processes were probably caused by demographic growth, which limited migrations within the (sub-)Tropical Asian core areas and pushed some groups to explore the ultimate frontiers.

      " C-V20's closest relative is C-M8, which is found in Japan".

      I was not aware of this phylogenetic affiliation until now. It is indeed interesting.

      "I'm not sure how that can be reconciled into this picture."

      Me neither honestly. Do you know which is the recent study which redefined the C phylogeny; it seems a very interesting study to read and discuss.

      "With regards to admixture from an OOA isolate or directly from tropical Africa, couldn't both be correct to some degree?"

      It's possible indeed but I would not think that the two or three components potentially hidden behind the "Basal Eurasian" label are the same at all: they just share the quality of not belonging to the main Eurasian macro-population but otherwise they are quite different from each other as far as I can tell.

      Personally I'd expect a NE African element in EEFs via Palestine, associated to yDNA E1b and also maybe an Aboriginal Arabian one. The NW African Aboriginal element would be the third potential candidate and it should only be found in La Braña and modern West Iberians - unless there are confounding factors at play, for example if Iberians scored closer to EEF because of La Brana-like influence, what would be a total artifact.

      The issue of what exactly is "Basal Eurasian" in each case really merits further study.

      Delete
    2. This seems to be the study: http://www.nature.com/jhg/journal/v58/n4/full/jhg2012159a.html

      Sadly it is pay per view. I would appreciate if someone with access could send me a copy, really.

      Delete
    3. @Ryan
      "La Brana was Y haplogroup C-V20 though. C-V20's closest relative is C-M8, which is found in Japan and believed to be associated with the Jomon culture. I'm not sure how that can be reconciled into this picture."

      I can imagine two possibilities

      1) A population starting at point A (India?) expands both west to point B (La Brana) and east to point C (Jomon). Later two distinct populations develop and expand in between point A and B (Egypt-Arabia) and in between point A and point C (China) thus splitting the populations at A, B and C into three separate populations.

      2) As before but the population expanding west to point B (La Brana) reached the mammoth steppe sooner as it extended furthest south at the western end and those mammoth hunters spread east to point C (Jomon) directly across the mammoth steppe.

      mammoth steppe: http://img571.imageshack.us/img571/6525/7dt9.png

      Delete
    4. I must mention here that, for what was explained to me recently in a private exchange, the reorganization of the C tree detaches first of all former C3 (now C2), and then another subclade including all other C sublineages (there is doubt about the exact affiliation of Australian former C4). So C still seems to expand from the Far East, even more solidly than before.

      It is true than within this new C1 subclade, European hyper-rare C-V20 and Japanese (also very rare) C-M8 are most closely related but my most parsimonious tentative conclusion is that C-V20 migrated westwards, rather than C-M8 doing the opposite.

      Nothing is certain anyhow but I must say that neither La Braña nor Ainu are known for hunting mammoths (rather deer and fish respectively). I wouldn't build much on mammoth hyping because, other than is some specific areas, these were not the main prey of Paleolithic hunter-gatherers.

      Sadly there's no published study of any sort on the new classification (seems all FTDNA findings), so I decided to leave discussion on this reorganization of Y-DNA hap. C for the future.

      Delete
    5. "Nothing is certain anyhow but I must say that neither La Braña nor Ainu are known for hunting mammoths (rather deer and fish respectively)."

      Yeah I am more inclined to the first explanation - both expanded from the same start location (somewhere south of both La Brana and Jomon) and then separated by later expansions.

      Delete
  9. In terms of Siberian affinities, the Eurasian steppe runs straight from the Ukraine across to Lake Baikal, so I don't see why it should be all that surprising the populations living adjacent to it would have some greater affinities to people living there.

    ReplyDelete
    Replies
    1. First of all the Altai "Aurignacian" is contemporaneous with the earliest European "Aurignacoid" cultures, both surely stemming from West Asia or somewhere around Pakistan or Uzbekistan. They are "sibling" cultures but quite apparently each has its own distinct founder effects and such.

      Between the Asian and the European steppe it is one of the largest Eurasian rivers, the Volga, which is a natural barrier, and also some quite desertic areas. It's not exactly a "highway", although it's not impossible either that there were some occasional contacts.

      There is no clear evidence of interaction between Europe and Central Asia in the Paleolithic, excepted maybe the Gravettian period (Mal'ta boy belongs to that era), which could also have been centered in West Asia.

      Personally I would indeed expect some European flow to Altai and backflow from there within Gravettian particularly but I would also expect both populations to have kept much of their original distinctiveness (for example modern Altaians still retain lots of yDNA Q1, while in Europe this is extremely rare). I may be wrong in this of course.

      Delete
    2. Yah, I definitely agree with that assessment. Only thing I'd caution is re: assuming today's river system is representative of what it was like earlier. During the LGM the river system would have been very different. There's a theory out there that Lake Baikal may have had to drain west first to the Aral, then the Caspian, and then to the Black Sea and beyond. The Volga's drainage basin would have been much smaller, and not yet included the Kama. Many areas that now drain south would have drained north initially, only shifting direction as isostatic rebound takes its course after the glaciers retreated. There would have been much less rainfall in the area too. The Volga may not have been a big barrier at the time, but there could have been other less obvious barriers. Perhaps other opportunities as well for people to fish along these river systems as well as along the massive lakes that would be formed by the retreating ice.



      Delete
    3. See: → http://donsmaps.com/icemaps.html

      Scroll down for the North Eurasian Ice Age maps, which are extremely good AFAIK. However the river systems appear to be the same, just that there were huge glacial lakes and inner seas that are not any more (or have been reduced, like the Caspian Sea). The Kama was anyhow a Volga affluent, at least in those high quality maps.

      But maybe you know something Don doesn't...

      A clear barrier at the LGM was the huge glacial lake fed by the Obi and to lesser extent the Yenisei, at some times feeding to Aral Sea (and this one to the Caspian). But otherwise the mighty Volga was there in any case, as well as a much enlarged Caspian Sea.

      An even more mighty barrier was probably the ice corridor going between Tibet and NE Siberia via Lake Baikal, according to another map in that series. That would have kept West and East Asia clearly segregated at the LGM and may have accelerated the proto-Amerindian "orientalization" process, as they were already East of it by all accounts.

      Delete
    4. Bah, I was getting different glaciations confused. The Kama was captured in the previous interglacial. My mistake.

      All I know is from reading a few articles on Google Scholar. Apparently the exact details of the Volga during the Last Glacial Maximum are a bit unknown as the Soviets obscured much of the information when they built hydro electric dams in the area.

      The map does correspond to what I read (though the exact extent of Siberian glaciation seems to be a matter of debate still). The climate was much drier though. The Volga may not have been as wide or as fast.

      Amerindians were likely cut-off from the Siberian population, with the Siberians more connected to central Asia and Europe, at least according to that map. Thanks for the link - the maps I saw were of lower quality.

      Delete
    5. "A clear barrier at the LGM ... An even more mighty barrier"

      Did these barriers separate mammoth into distinct east and west populations?

      If not I doubt they separated the mammoth hunters.

      Delete
  10. "This implies that there were already some differences in the Epipaleolithic era among European hunter-gatherers, with those of Magdalenian background lacking the Siberian (ANE) component, which is found however in Scandinavian ones (of Ahrensburgian background?) This may help explaining the extra ANE affinity in Northwest Europe, which is otherwise hard to understand."

    mammoth steppe

    http://img571.imageshack.us/img571/6525/7dt9.png

    If you assume ANE -> WHG ->(EEF) -> "Basal Eurasian" go from north to south then it makes perfect sense.

    If ANE was mammoth steppe (northern) and not Siberian (eastern) - which is what makes the most sense - then it is a latitudinal distinction not a longitudinal one and the extra ANE in NW Europe is easy to understand. It was already there from the end of the LGM.

    The map also explains why you would expect a higher proportion of ANE east of the Black Sea than west of the Black sea. There are physical geographical barriers in the east: Black Sea, Caucasus, Caspian but not in the west. If ANE and WHG were latitudinal bands then as the ice retreated after the LGM then west of the Black Sea you would expect both ANE and WHG to move north because there are no physical barriers but from the Black Sea eastwards there are geographical barriers stopping the more southerly HG populations moving north hence mostly ANE.

    So WHG/ANE in north western europe and mostly ANE in north eastern Europe pre-Kurgan with additional ANE coming west with the Kurgans.

    #

    "The authors believe that the main axis of differentiation in Europe when the subcontinent is considered as a whole may tend to Northeastern Europe rather than SSE/NNW⁸, something that is consistent with their ancient admixture findings elsewhere in the study."

    Alternatively there were two (or more): an earlier SSE/NNW cline and a later NE cline with the later one(s) partially over-writing the earlier ones.

    #

    "This brings us to a key question: what is actually "Basal Eurasian"?"

    Afro-Asiatic (Egypt/Arabia) imo.

    north to south:
    - ANE (mammoth steppe) (northern HG across all of Eurasia)
    - WHG (western *branch* of southern HG) (southern Europe, north Africa, near east)
    - EEF (mix of WHG and "Basal") (Levant)
    - "Basal" (Afro-Asiatic) (Egypt-Arabia)

    The labeling of the populations has tilted the axis of the debate; it's north to south not east to west.

    ReplyDelete
    Replies
    1. I would expect ANE and EHG (Eastern-European Hunter Gatherers) to be somewhat different. I.e. EHG should look as a "mix" of WHG and ANE, a bit like Motala but more towards ANE maybe. That's because the core origin of ANE should be from West Asia, while the core origin of EHG should be from Central Europe, much as WHG - of course this ultimately comes from West Asia too but there is an intermediate downstread node in Central Europe which is not shared by ANE.

      So when we see that Lithuanians, for example, are very high in the WHG component, we must ponder how much of that appearance is actually directly from Lochsbour-like ancestry or a more distant relationship via EHGs. The ANE component (absent in WHG) gives us an important clue but even more important is probably the IBD ancestry analysis, which I forgot to discuss but I plan to, maybe today.

      In the last supp. material article, we can see that, while Sardinians share both the greatest number of segments and length of them with the EEF population (represented by Stuttgart), Baltic Europeans only share the greatest amount of segments with WHG, the ones who share the lengthiest segments with Lochsbour are the French, what implies a difference in the patterns of ancestry: French are less % Lochsbour but more recent descendants of the population he represents, Finns or Balts are more % of Lochsbour but more distant descendants, surely because their ancestry is mediated by EHGs, who diverged from Lochsbour before the LGM.

      Probably a comparison with La Braña or Motala would generate different results, because Lochsbour can't represent single-handedly the complex genetics of European Paleolithic. Ideally an Eastern European hunter-gatherer would illuminate this matter even better. We need more research on this matter.

      Just because we only have three clades with known samples, it does not mean that there aren't more. In fact there must be greater complexity in the European ancestry, from Europe and, of course, from outside it (still unexplored in any direct way other than ANE).

      Delete
    2. "Afro-Asiatic (Egypt/Arabia) imo."

      Not in the case of La Braña at least. In the case of EEFs "Basal Eurasian" can and probably is partly that, but I cannot exclude "aboriginal Arabian" (OoA residual) from it. In the case of La Braña, it should be an aboriginal NW African element (Aterian residual, today only detectable at significant amounts in Southern Morocco).

      "If you assume ANE -> WHG ->(EEF) -> "Basal Eurasian" go from north to south then it makes perfect sense".

      I can't assume that (all archaeological and genetic evidence is against it). Your map (which exaggerates the extent of the steppe-tundra, a better name than "mammoth steppe") is meaningless unless you can identify the ecology with archaeological evidence, which is at least quite hard.

      Actually:

      West Eurasian → ANE
      West Eurasian → Paleolithic European (WHG but also UHG, EHG, etc.)
      West Eurasian → West Asian HGs (WAHG),

      In some cases WAHG+OoA residue = "Basal Eurasian" (at least one possible such element), in other cases WAHG+NE African ="Basal Eurasian", finally it is also possible to see "Basal Eurasian" as WAHG+OoA residue+NE African.

      In the case of La Braña instead it's nothing of that: just Aterian residue via NW Africa.

      Delete
    3. Fair points. I over stress the mammoth steppe point because it seems to me the labeling of the Laziridis components leads people to forget the mammoth zone extended from France to Siberia, not just Siberia.

      Delete
  11. Update: new entry on Lochsbour's IBD relatedness, which I forgot to mention here but I find quite interesting: → http://forwhattheywereweare.blogspot.com/2014/04/lochsbours-ibd-in-modern-europeans-is.html

    ReplyDelete
    Replies
    1. Those Somali samples are not actually representative of Somalis, they seem to be some sort of strange intermediate between Somalis and "Southern"/ Borana Oromos.

      I explain this here:

      http://www.forumbiodiversity.com/showthread.php/43408-Lazaridis-quot-Kenyan-Somalis-quot-not-representative-A-mix-between-Boranas-and-Somalis

      Delete
    2. Is there any sample of Somalis in the Lazaridis study? Why should it even matter if they are a bit more this of that when it is a study on European origins? Sorry, but it seems you mixed this entry with some other.

      Delete
    3. Yeah, in the supplemental... In that very plot you share in this blog post. My point is that about 8 to 6 of those 13 samples are seemingly to possibly mixed with Borana Oromos who have "Omotic" admixture and are seemingly less Middle Eastern admixed than your average Somali so my point was simply that I figured you'd be interested in their mixed-ness.

      It would also be liking sharing Amhara samples that are seemingly half Somali and acting like they fully represent Amharas.

      Delete
    4. Look again. No Somalis in this entry, Awale (and nothing to do with the thread discussed, which is European origins based on ancient DNA). Please search for the entry you're interested in.

      Delete
  12. This comment has been removed by a blog administrator.

    ReplyDelete
    Replies
    1. ^^ There are highly irrelevant commercial links (spam) in your comment: hence I have to delete it. Try reposting without them, please.

      Delete
  13. "We can only consider in fact this "Basal Eurasian" idea as a mere indicator of African-like affinity, even if it's not Mbuti but something else. This something else can be in fact several things:

    NW African Aterian residual in the case of La Braña

    Arabian OoA residual influence in the case of EEF

    Direct NE African admixture in the proto-EEF West Asian population, strongly indicated by the NE African E1b-M78 (notably its subclade E1b-V13) in ancient Neolithic and modern European Y-DNA..."

    I'm now starting to see how this might be the case for A-M13 in Sardinia. I got my NGS results from Yfull about a month ago, on their Ytree they've placed the Saudis, Sardinians, South-Sudanese (Dinka) and Ugandan nilote (me:) as having a TMRCA at 10,500ybp ("https://www.yfull.com/arch-4.08/tree/A-M13*/"). I didn't make the EEF, BE connection until a member on a forum mentioned it.

    I did a little digging and subsequently replied with what I thought might be supporting papers demonstrating the EEFs of the Aegean region and the peopling of Sardinia. ("http://www.anthrogenica.com/showthread.php?7187-A-M13-in-an-Englishman&p=245845#post245845")



    I think this might support the idea of the "OoA residual influence" in SW Asia:

    When running admix calc's on GEDmatch, I noticed a specific correlation between an Mbuti-like affinity and "Natufian" on GedrosiaDNA (NEN-K13, AE-K6). It's a detection at somewhat minor levels <9% but clearly non-negligible enough to demand an explanation. The correlation of kits with the least amount of Mbuti-like affinity and those with the least "Natufian" is the most apparent. I noticed this with over a dozen nilotes from Uganda and neighboring South-Sudan. *It's worth noting, none of us have non-negligible Eurasian ancestry on modern calculators (FTDNA,23andMe,Ancestry). When there is, it's on a GEDmatch calc with what I assume to be ancient references - we usually get SW-Asian/Middle-East, ASI and SE-Asian (in that order - collectively under ~10%).

    Depending which GedrosiaDNA calc is used, I noticed the Natufian affinity gets split in half with Ancestral South Eurasian (ASI). Perhaps this favors the idea of this being some kind of "OoA residue" as you've nicely coined it. This hunter gather affinity has become quite the enigma, it's the most inconsistent in admix calculators -- it varies from <1% to ~15% depending which of the big 3 DNA companies let alone the GEDmatch calc's.



    "...In any case I'd suggest to change the terms "Basal Eurasian" and "non-Africans" by something more neutral, maybe "Ultra-Mediterranean" and "Proto-Eurasian" respectively, where both concepts are allowed to be African, at least partly so."

    I strongly agree with your suggestion to change the term "Basal Eurasian" to something more neutral.




    Out of curiosity, what's your take on Natufians having minor, if any SSA affinity?

    I questioned Lazardi's conclusion on a forum, using the afmd mbuti/natufian correlation as a counter-argument, let's just say it wasn't well received. In hindsight, to what you stated earlier, I'm also starting to see how "Basal Eurasian" defined as intermediate between Mbuti and Eurasian-plus populations could be problematic -- at this point how does one demarcate what's SSA, or "Tropical African"(*I like this term, I think I'll use it more often - it's growing on me)

    ReplyDelete
    Replies
    1. Hi, Benjamin. What you say about A-M13 makes good sense to me, even if I'm quite reluctant to take molecular-clock guesstimates at face value (they could perfectly be doubly older or whatever).

      As for your Wikipedia quote (Brigaglia) on Anthrogenica, if you go to the paper (in Italian) it doesn't say that: it just says that there are two pre-Neolithic sites but no clue as where those sea-daring peoples arrived from (I would imagine that Liguria, because that was the shortest leap to Corsica-Sardinia, then a single island).

      "It's a detection at somewhat minor levels <9% but clearly non-negligible enough to demand an explanation."

      Do you mean less than 1/10 people? Well, that's intriguing but can also be a semi-random product, maybe convergence of "true BA" and African inputs in some area like Palestine. The whole matter definitely demands more research in depth in any case.

      "Depending which GedrosiaDNA calc is used, I noticed the Natufian affinity gets split in half with Ancestral South Eurasian (ASI)."

      The ASI relation could be and surely is an effect of what else you're comparing against, sometimes pseudo-affinities are produced that only mean "not the other" or, if split, "neither this nor that". It's like if you ask the algorithm if a lion is an elephant or a mouse: the answer will be half-half (roughly) but what it actually means is "neither one nor the other".

      We see in Lazaridis 2016 that Natufian and Levant Neolithic are clearly distinct from Zagros or Iran Neolithic. And we also see that Natufian and Levant Neolithic were strongly dominated by E1b sublineages, strongly suggesting African roots.

      This should be different from "true BA" found in CHG. I say this may be "true BA" because nothing indicates influence from Africa. Of course there could also be some "true BA" remaining in Natufian and Levant Neolithic but at least a good share should be East/NE African influence of Afroasiatic and/or Eastern Sudanic language/macro-ethnicity (not sure if you know but I do suspect Basque, and hence the Vasconic language family of ancient European farmers, is partly "Nubian").

      "This hunter gather affinity has become quite the enigma, it's the most inconsistent in admix calculators -- it varies from <1% to ~15% depending which of the big 3 DNA companies let alone the GEDmatch calc's".

      Such contradictory results strongly suggest that it's largely a "ghost" of some sort, i.e. not something too reliable. But, of course, more research is needed.

      ...

      Delete
    2. ...

      "I questioned Lazardi's conclusion on a forum, using the afmd mbuti/natufian correlation as a counter-argument, let's just say it wasn't well received".

      Sadly most anthro/genetic forums are dominated by "militant" people who lack the necessary open-mindedness to address these issues scientifically (many may be rather fascistoid in fact, because this is a field that, sadly, attracts that kind of people). That's one of the reasons I became a blogger: truly tired of arguing with fanatics of their own prejudices.

      ""Basal Eurasian" defined as intermediate between Mbuti and Eurasian-plus"...

      In some cases it is rather between Yoruba and Greater (Eur-)Asian. But that's enough to leave a significant blank in NE Africa (by that I mean Sudan and The Horn primarily), where "the L3 population" (lax label) originated and is still largely dominant (even after removing M/N backflows from Eurasia). Yorubans and other West African/Bantus would be rather "the L2 population", while West Pygmies are rather "the L1 pop." and Khoisan are "the L0 pop". (all labels are lax and only refer to the core or most important mtDNA macro-lineage).

      Since maybe the early UP/LSA c. 50 Ka BP there must have been some backflows from West Asia (and also later from Europe into North Africa) and (less clearly) maybe vice-versa. This Africa-to-WestAsia flow is however very clear at the Mesolithic, and originates surely in Nubia (i.e. North Sudan and southernmost Egypt). This makes "BA" analysis very complicated: first there is the L3-M/N relation at the very macro-Asian founder effect, then the almost certain backflows from West Asia in the Paleolithic (for example mtDNA X1 in Egypt, necessarily very old, or the very diverse NE African Y-DNA J1, almost as diverse as West Asian one), and finally the African (Afrasian/Nubian, E1b-M78) migration to Palestine in the Mesolithic with the very rudiments of agriculture (they were already gathering cereals, although seemingly not yet farming them). How can you filter all that? Probably the only safe way is to go to Sudan, Egypt, etc. and get ancient DNA (which is complicated because of climate but not impossible) and compare with that once achieved.

      "*I like this term, I think I'll use it more often"...

      Cool. This issue of "SSA" grew on my nerves early on: the term was adopted by UN bodies upon the political transition in South Africa, earlier the term "Black Africa" was used instead (but South Africa got excluded). I think "Black Africa" is fine, although maybe some people would consider it "maybe racist" or with a colonial aftertaste (but it can also be read positively as in Black Pride and Black Power), but "Sub-Saharan" is definitely all kinds of wrong: it's geographically wrong (the south is not "under", under the Sahara there may be water, oil, uranium or whatever but not Tropical Africa) and it suggests inferiority (sub-human, sub-normal, sub-standard, etc.), so I began using "Tropical (and Southern) Africa" instead.

      Delete
    3. On the "Sub-Saharan" terminology issue, see: http://leherensuge.blogspot.com/2008/05/supersaharan-africa.html

      Since then I've thinking on making t-shirts with the concept: maybe a map of Africa with south on top, optionally a dotted line or color change across the Sahara, and the Superman logo in "African" colors (something like red, yellow, green and/or black, typical of many flags over there) and the legend around "Super-Saharan Africa". Maybe you can borrow the idea now that Africa is experiencing a good economic moment, I won't ask for designer royalties, don't worry, that concept should belong to Africans and Afrodescendants.

      Delete
  14. Hi Maju, pardon the initial omission of salutation – I guess I was too excited to jump into convo:
    “As for your Wikipedia quote (Brigaglia) on Anthrogenica, if you go to the paper (in Italian) it doesn't say that: it just says that there are two pre-Neolithic sites but no clue as where those sea-daring peoples arrived from (I would imagine that Liguria, because that was the shortest leap to Corsica-Sardinia, then a single island). “
    Thank you for clarifying.
    So any of those folks from Liguria were overrun by WHGs?


    It's a detection at somewhat minor levels <9% but clearly non-negligible enough to demand an explanation.
    “Do you mean less than 1/10 people? Well, that's intriguing but can also be a semi-random product, maybe convergence of "true BA" and African inputs in some area like Palestine. The whole matter definitely demands more research in depth in any case.”
    I meant to say the Natufian affinity of Central and Eastern Sudanics (*small sample set of immediate family members) is less than 9% on GEDmatch’s GedrosiaDNA calc . Here’s a comparison chart I made in an attempt to illustrate each kit’s correlation between the Mbuti-like affinity on other calcs (puntDNAL, Ethihelix) and GedrosiaDNA. To push the point further, I displayed each kit’s “Omotic, NiloSaharan and Western-Semitic” clusters (puntDNAL calc) to demonstrate the correlation lies solely with the Mbuti-like affinity and nothing that might be considered Eurasian shifted:

    [img]http://i.imgur.com/w6r6Fea.jpg[/img]

    Depending which GedrosiaDNA calc is used, I noticed the Natufian affinity gets split in half with Ancestral South Eurasian (ASI).

    “The ASI relation could be and surely is an effect of what else you're comparing against, sometimes pseudo-affinities are produced that only mean "not the other" or, if split, "neither this nor that". It's like if you ask the algorithm if a lion is an elephant or a mouse: the answer will be half-half (roughly) but what it actually means is "neither one nor the other".

    I see.
    A moot point here but I thought I’d mention, on second examination, the lesser “Natufian” affinity on GedrosiaDNA ‘s NEN K13 vs. AE K6 might be comparatively lower because the Kusunda , a tribe of former hunter-gatherers of the forests of western Nepal were used instead of the Onge/Andamanese. It’s said this was done because “it produced Iran Neolithic levels in South and West Asians, somewhat more consistent with those in the recent papers”.

    ReplyDelete
    Replies
    1. Greetings are optional, worry not.

      "So any of those folks from Liguria were overrun by WHGs?"

      They should be WHGs themselves. I mentioned because the Wikipedia article said something (with that cite) about anywhere from Iberia to the Aegean as the origin. Well, probably something much more specific: Liguria or somewhere nearby, because of the distance of navigation required. We know from archaeology that the Neolithic settlers of Sardinia were probably from Central Italy instead, but those people were the most advanced sailors of West Eurasia back in the day, so no wonder (the Balearic Islands were probably settled directly from Provence in another sailor feat, also in the Neolithic).

      "I meant to say the Natufian affinity of Central and Eastern Sudanics (*small sample set of immediate family members) is less than 9% on GEDmatch’s GedrosiaDNA calc."

      OK, that's significant to my eyes and, I'd dare say, expected. However it can be read both ways (i.e. Sudanics to Natufians or Natufians to Sudanics), any reason why it cannot be the latter?

      "Depending which GedrosiaDNA calc is used, I noticed the Natufian affinity gets split in half with Ancestral South Eurasian (ASI)."

      This might be explained better if the flow is (primarily) from Asia to Africa and quite old, i.e. before West Eurasian distinctiveness emerged, for instance in the early Upper Paleolithic/LSA. At that time ASI and West Eurasian gene pools should not be easy to discern yet, IMO.

      "... the lesser “Natufian” affinity on GedrosiaDNA ‘s NEN K13 vs. AE K6 might be comparatively lower because the Kusunda , a tribe of former hunter-gatherers of the forests of western Nepal were used instead of the Onge/Andamanese. It’s said this was done because “it produced Iran Neolithic levels in South and West Asians, somewhat more consistent with those in the recent papers”."

      That's quite interesting but very complicated to discern. Do the Kusunda get like 100% ASI in intra-Indian comparisons? If so it may be correct. Onge are not a South Asian population but a distinct population on their own right, with only very weak connection to South Asia, so I have argued in the past against considering them a valid referece for Indian Paleolithic populations. In principle having a local reference like the Kusunda should be better but we can't know for sure if the Kusunda are also admixed or not until ancient DNA is produced in the subcontinent.

      Complicated issue, really. I still think that the best reference for African influence on Natufian is Y-DNA E1b-M78 and archaeological evidence, autosomal DNA is quite tricky.

      Delete
  15. “This should be different from "true BA" found in CHG. I say this may be "true BA" because nothing indicates influence from Africa. Of course there could also be some "true BA" remaining in Natufian and Levant Neolithic but at least a good share should be East/NE African influence of Afroasiatic and/or Eastern Sudanic language/macro-ethnicity (not sure if you know but I do suspect Basque, and hence the Vasconic language family of ancient European farmers, is partly "Nubian").”

    This intrigues me – an anthrogenica forum member, represented w/ Italian and Belgian flags, seems to think the same – he replied today rather emphatically to the post made yesterday.
    Have you come across this blog before: https://nigercongoindoeuropean.net/category/language/
    Apart from the lengthy linguistic comparisons, the “A haplogroups” and R1b R-V88 in Chad are mentioned as links from Africa to Basques and Sardinians.
    On a thread titled, “Who settled Ireland before the Celts?” I asked whether the few detected A-M13s from Ireland/Scotland could be remnants of first arrivals: http://www.anthrogenica.com/showthread.php?7545-Who-settled-Ireland-before-the-Celts/page2
    - I’m interested to hear your take on it.

    This is my own subjective rambling and musings:
    I was asked once if I was from the Balkans judging by my surname once “Yebuga”, I found that odd.
    I’m not familiar with the Basque language but to my subjective ears, Spanish sounds very African compared to most European languages, Italian would be second-up. I unfortunately don’t speak any of my parent’s native languages but I feel as though listening to them talk amongst themselves gave me an edge over anglo students when learning Spanish – especially with the running “R”s and comparatively fluid vowel structure (*comparatively speaking when compared to other neighboring IE languages). My mindset at the time, in my mid-teens, was that if I couldn’t learn an African language, I’d learn one that sounded closest to one – Spanish was easier than French at the time but ultimately revived my interest to pursue French; I can converse and write at a low-intermediate level in both (*grammar has taken a big hit since uni days though).
    I found what I thought to be coincidences between Castellano and my mother’s Eastern-Sudanic language, only to find the similarities are probably derived from different dialects of Arabic. An example, the word grandmother in my mother’s tongue is “Abuba” vs. the Spanish “Abuela” – I was certain the connection was through Arabic (the Islamic period in Iberia?) but none of the Arabic Speakers I asked (Egyptian, Palestinian, Syrian, Lebanese: from what I recall) had anything remotely similar. It wasn’t until I asked a North-Sudanese colleague (an Egyptian Copt with a Sudanese grandmother); they have the same term but pronounce with an “H” in the front “Hahbuba” – she thinks it’s a play-on-word from the Arabic “habibi”
    My father’s Nilotic tribe “Aringa” means herring-fish in Italian. The larger encompassing ethnic group of my mother’s Kakwa tribe, “Bari”, is spelt the same as the southern-Italian town. These are not at all arguments for relatedness with Mediterraneans but I find it interesting that these fluid vowel combinations are more common in the Mediterranean than say in stereotypical IE languages in northern Europe.

    ReplyDelete
    Replies
    1. "Apart from the lengthy linguistic comparisons"...

      Yes, better leave them appart: there's no method, only an opportunistic pseudocientific list, with many Basque words being wrong or Latin-derived. Eek!

      "... the “A haplogroups” and R1b R-V88 in Chad are mentioned as links from Africa to Basques and Sardinians".

      Basques have neither, only Sardinians (and to lesser extent Italians) do. R1b-V88 is indeed an Afro-Mediterranean lineage which may originate in West Asia, the Italo-Balcanic area or the Nile Basin (ultimately always from West Asia as R1b-root). I lean for West Asia but awaiting more detail from African V88.

      See:
      · http://leherensuge.blogspot.com/2010/01/r1b1-origin-italy-or-west-asia.html
      · http://leherensuge.blogspot.com/2010/01/african-r1b-is-distinct-single.html

      "I asked whether the few detected A-M13s from Ireland/Scotland could be remnants of first arrivals"...

      Looks like. Remind me of a Frisian guy who asked me for help for understanding his E1b lineage (quite rare over there) and eventually persuaded me and himself that it was a Neolithic legacy (originally I thought it'd be Sicilian). Can't recall the details though but he was very solid in his conclusions, very persuasive. So these A in Ireland or also maybe the B in Iran and Afghanistan might well be from Neolithic flows.

      In the Irish case the Sardinian connection seems quite relevant and the Saudi one should be considered as quasi-proof of Natufian/Levant Neolithic connection (Arabia Peninsula was essentially settled from the Levant in the Neolithic).

      BTW, a side note: don't take at face value whatever Jean Manco says, she doesn't just have an agenda of her own but she pretends to know things she does not know and even denies hard facts when they contradict her hypothesis, having no respect for anything or anyone who stands in her way.

      ...

      Delete
    2. ...

      As for Spanish, abuelo/-a comes from Latin avus (grandfather), the /v/>/b/ shift is very normal in Iberian languages and a Basque-like trait. I don't know enough about African languages to judge but you could be right on the issue of preferred phonetics, such as strong R (thrill) or plenty of vowels (all these are Basque traits as well and therefore likely Neolithic traits, better preserved in Southern Europe apparently), however, for what I know, most African languages are tonal, while no European language is (not hard evidence of anything, as there was surely founder effects in Palestine first and the Aegean later, which may have trimmed the trait easily).

      When I found the striking Vasco-Nubian appearance of relatedness with my mass lexical comparison, I though about contacting G. Starostin, son of the "pope" of Soviet linguistics Sergei Starostin, because I realized I was working on his materials and that he has been actively researching Nubian languages recently. But I didn't (lack of self-confidence even if my linguist friends, all them retired, have not dared to criticize my humble work). In any case I think this is a matter for academic linguists, because most amateurish work is just junk: coincidences happen all the time, you need a very high frequency of coincidences, above 10% at least in a neutral/random list such as Swadesh lists, in order to have some indication, not yet strong evidence.

      "My father’s Nilotic tribe “Aringa” means herring-fish in Italian."

      A coincidence at least in principle. Aringa (Spanish "arenque") comes from proto-Germanic *hēringaz (or something like that), as the herring fish is only found in cold waters of the North Atlantic and North Pacific oceans. However it is possible that the word is ultimately derived from a relative of Basque "arrain(-a)" = "fish", just as "sardine" is also Vasconic almost certainly: (Basque) "sar-tu" = to enter, "sare" = net, "sarda" = fish school (what enters in the net, what is entered), "sarda-gina" (> "sardina") = what makes the fish school. The word and similar words are also used in inland contexts: "sarda" = pitchfork (the pointy thing that enters into the hemp), Sardanya in Catalonia probably referes to an entrance into the mountains, the Sardana dance (also in Catalonia) is a dance in which everybody can participate (enter, maybe "sar dana" = "get in everybody", in Basque) and makes circles of people inside each other. The very name of Sardinia probably comes from the fish, i.e. "island of sardines", as known to ancient fishermen. Anyway, Germanic has a lot of cryptic Vasconic words according to Theo Vennemann such as "kill" (Basque "hil"), "gore" (plausibly related to Basque "gorri", now "red" but also "bloody" as in curse word, and maybe meaning something related to raw flesh or meat originally according to my linguist friend Roz Frank).

      "These are not at all arguments for relatedness with Mediterraneans but I find it interesting that these fluid vowel combinations are more common in the Mediterranean than say in stereotypical IE languages in northern Europe".

      I tend to agree: it's not hard evidence but it suggests a cultural tendency, which in Europe must be seen as pre-Indoeuropean, i.e. Vasconic for the greatest part. A typical example is Latin "librus" (book), which has the -BR- double consonant typical of IE languages, Basque naturally transformed it in "liburu", because that -BR- group was cacophonic or hard to pronounce.

      I remember years ago that there was an athlete from Sudan named Aldama, which incidentally is also a Basque surname (a precursor of my own surname: Aldamiz). On its own it's just a weird coincidence but when "coincidences" pile up, then we must begin to consider there can well be something more.

      Delete
    3. “Basques have neither, only Sardinians (and to lesser extent Italians) do. R1b-V88 is indeed an Afro-Mediterranean lineage which may originate in West Asia, the Italo-Balcanic area or the Nile Basin (ultimately always from West Asia as R1b-root). I lean for West Asia but awaiting more detail from African V88.”

      Thank you for the links, the topic interests me.

      After FTDNA came out with their “AncientORIGINS” admix calculator, I noticed members of my Central-Sudanic side of the family all had “Metal Age Invader” affinities, albethey at low levels <4%:

      Noted here: http://www.anthrogenica.com/showthread.php?9060-FTDNA-ancientOrigins&p=213329&highlight=nilosaharan#post213329

      Again, although this is a small data-set, I found the contrast quite stark in comparison to my Eastern-Sudanic side of the family, along with myself and siblings who share no “Metal Age Invader” affinities. I hypothesized in the post that this might be related to R1b carriers in the Central-Sudanic (Chad) region where my paternal ethnic group is believed to have migrated. I base this from the Central Sudanic linguistic branch which stretches in an arch from Lake Chad to Lake Albert -- a genetic continuity, seems to be confirmed with the most accurate African admix calc’, IMO, puntDNAL k8, which has the Bulala_Chad (another Central-Sudanic speaking ethnicity) as the closest Oracle match for most family members. I think this along with my paternal-grandmother’s L2e haplogroup, which is concentrated around Lake Chad, seems to confirm a distant connection to the region.

      I make the case for L2e here:
      http://www.anthrogenica.com/showthread.php?241-L2d2-in-French-Guiana-and-beyond

      “..So these A in Ireland or also maybe the B in Iran and Afghanistan might well be from Neolithic flows”

      I noticed quite a number of B in the middle-east through FTDNA’s B Haplogroup Project. I’d have to get him SNP tested but judging from the STRs alone, the genetic steps/distance on mutations from others seems to suggest these MENA Bs are beyond anything in the last 2-3 thousand years:

      https://www.familytreedna.com/public/BYDNA/default.aspx?section=yresults
      *my maternal grandfather (Eastern-Sudanic) is at the bottom:

      "In the Irish case the Sardinian connection seems quite relevant and the Saudi one should be considered as quasi-proof of Natufian/Levant Neolithic connection (Arabia Peninsula was essentially settled from the Levant in the Neolithic)."

      I’m starting to believe this is the case, it seems to be the most reasonable explanation.

      I contacted several of the Boyd matches through BigY, two responded – one believes a connection might be through an NPE event from the Iberian peninsula but through some of the first “Moors” in Scottish courts (James IV) supposedly from a Portuguese ship in the early 16th century:
      https://en.wikipedia.org/wiki/Andrew_Barton_(privateer)

      I’ve read another theory on a 23andme forum, relating to a crashed sailor from the Spanish Armada.
      I’m not sure if any are aware of the Neolithic connection.

      “BTW, a side note: don't take at face value whatever Jean Manco says, she doesn't just have an agenda of her own but she pretends to know things she does not know and even denies hard facts when they contradict her hypothesis, having no respect for anything or anyone who stands in her way. “

      Wow, the two of you are acquainted? The genetic/anthro community must be smaller than I thought. It’s ironic that her counterargument relating “people who had been living in the south European refuges, mainly the Franco-Iberian one” actually makes the case for a connection with the Sardinian A-M13s - wouldn’t you say?

      Delete
    4. I'd encourage you to contact G. Starostin, what's the worst that could happen? As you've mentioned, if multiple people are arriving at similar conclusions, he'd be one of the best candidates to academically confirm a summary of concrete linguistic evidence.

      I know I already mentioned this but the post from the presumably Italian/Belgian member on the forum is rather shocking -- is the Nubian-Vasconic relation pretty well known among fellow Vasconic speakers?

      http://www.anthrogenica.com/showthread.php?7187-A-M13-in-an-Englishman&p=248244#post248244

      Delete
    5. Your family's Central Sudanic data is very interesting. At face value it says "Metal Age Invader" (i.e. Indoeuropeans from the steppe) but those categories (also "Farmer", i.e. Vasconics from the Aegean, "Hunter-Gatherer") only make sense for Europe (that's what the test is designed for) and not for Africa, at least not directly.

      Indirectly however we can tentatively infer some stuff: European/West Anatolian early farmers were a mix of Levant and Zagros neolithic (plus some WHG), while early Indoeuropeans were a near 50-50 mix of Zagros Neolithic and EHG. So "Farmer" (without "Metal Age Invader") in your African case should mean ancestry from Neolithic Levant (or its African partial precursors) and "Invader" should mean Zagros Neolithic affinity (and not Indoeuropean, as they never went so far South before the Persian conquest of Egypt).

      The fractions are small and it might be just noise but the results are consistent within the ethnic groups, so I'm tempted to consider them relevant, at least provisionally. Hence my tentative conclusion is that R1b-V88 originated in or near the Zagros area (this might also apply to R1b in general, as happens with its "cousin" R1a, R1 as such coud be from Pakistan though but is much much older). This area was surely also at the origin of J2, which seems to have expanded southwards maybe since the late Neolithic/Chalcolithic or in the earliest Bronze Age at the latest, and that in Africa is a clear marker of Semitic inflow (usually very shallow anyhow).

      So my guess is that pastoralist from the Zagros area, after migrating to Palestine, went on into Africa, eventually reaching Central Africa as Chadic speakers (influencing Central Sudanics by contact probably) and that at some point in between they had a founder effect leaving them with dominant R1b-V88 and no J2, nor J1, etc. This founder effect could reasonably be related to the adoption of Chadic as their language, which should be of African origin anyhow. For what I read the most common theory about the origin of Chadic languages/peoples imply a late Green Sahara (c. 7000 BP) pastoralist migration northwards first and southwards later but it's debated.

      Does it make any sense to you?

      Delete
    6. Re. Irish Y-DNA A, you say: "I’m not sure if any are aware of the Neolithic connection." And the obvious answer is they are not. While E1b-M78 has been discussed a bit more, A is almost never heard of in relation to Neolithic. In any case, what they say also shows their general ignorance because there's no known Y-DNA A in Iberia (you will more easily find O, from a Japanese embassy that fertilized a whole Andalusian town). But, well, I'm getting too old for fighting the natural human tendency to ignorance and shallowness, really.

      "Wow, the two of you are acquainted?"

      Sure. We collaborated and emailed with some frequency in the past... until she arbitrarily and against all evidence began denying every single instance of Paleolithic mtDNA H, which was inconvenient for her theories (and her economic interest in a book she had just published with all the wrong answers).

      "The genetic/anthro community must be smaller than I thought".

      Not sure how big or small it is but some people like myself, Manco, Davidski/Polaco, Razib, Dienekes or whatshisname-the-Italian-surnamed-guy-behind-FTDNA-and-Eupedia, among others, have been around for long enough, we're talking more than a decade...

      IMO it's probably time for a generational change.

      "It’s ironic that her counterargument relating “people who had been living in the south European refuges, mainly the Franco-Iberian one” actually makes the case for a connection with the Sardinian A-M13s - wouldn’t you say?"

      It's part of her nonsensical approach, I guess. I just don't make any sense of it in any case.

      Delete
    7. "I'd encourage you to contact G. Starostin, what's the worst that could happen?"

      Most probably I'd get silence as answer, no big deal. Academics in general (not all) are that way and amateur linguistics and even more-or-less academic mass lexical comparison doesn't have too good press. What I did instead was to ask my two linguist acquaintances (both recently retired and both experts in Basque) and, well, they were mildly cold but did not reject the hypothesis nor the method.

      In any case it was a true surprise for me to find such level of apparent relatedness between Basque and Nubian, so much that I didn't even write on it until many months after I did the research. Anyway, I don't do much linguistics, because it's very hard to get to clear conclusions: archaeology and genetics are more solid stuff.

      "is the Nubian-Vasconic relation pretty well known among fellow Vasconic speakers?"

      Nope. Only readers of this blog (which may be in the thousands but most are from other areas of the planet) and a few other acquaintances of myself. However I know there have been a few other (much more amateurish) attempts to link Basque with African languages (as with almost any other language on Earth, even Quechua!) but they lack seriousness, a proper method. The mass lexical comparison method may not be the ultimate scientific corroboration (that would be the very slow and painful comparative method) but it does serve as preliminary approach in any case, particularly among languages which do not share a contact zone, i.e. are not in "sprachbund". So I think my method is solid enough but, as always, more research is needed.

      Delete
  16. “Sadly most anthro/genetic forums are dominated by "militant" people who lack the necessary open-mindedness to address these issues scientifically (many may be rather fascistoid in fact, because this is a field that, sadly, attracts that kind of people). That's one of the reasons I became a blogger: truly tired of arguing with fanatics of their own prejudices.”

    That’s really unfortunate, I’ve learned a lot on the forum -- I’ve also noticed somewhat of a “gang” mentality at times, I admit though, the administrators do a fairly good job of keeping a lid on things. Personally speaking, I don’t feel 100% comfortable asking thought provoking questions – often times members are more eager to “knock” someone’s question down rather than provide objective arguments to assist one’s learning.

    “Since then I've thinking on making t-shirts with the concept: maybe a map of Africa with south on top, optionally a dotted line or color change across the Sahara, and the Superman logo in "African" colors (something like red, yellow, green and/or black, typical of many flags over there) and the legend around "Super-Saharan Africa". Maybe you can borrow the idea now that Africa is experiencing a good economic moment, I won't ask for designer royalties, don't worry, that concept should belong to Africans and Afrodescendants.”

    Lol, sounds fair regarding the royalties
    One might brush this off as mere semantics but there’s strong meaning behind these identifiers – I’m seriously considering refraining from the “SSA” usage.

    ReplyDelete
    Replies
    1. "I’ve also noticed somewhat of a “gang” mentality at times"...

      It permeates all the net in fact. I've more than once used the term "indoeuropean pack" for a large group of people pushing for some untenable ideas I dislike.

      "... often times members are more eager to “knock” someone’s question down rather than provide objective arguments to assist one’s learning."

      Sure, I know some people like that, for example certain Australian blogger I won't name. IMO best is to ignore their emotional strike and stick to facts and arguments, but of course it can take a toll: it gets really annoying when all you get is "shut up" and "because I say so", or worse. I'm sorta lucky because I'm a respected blogger nowadays and my style is usually calm (although I can get angry too if pushed) but it's a pain when you painfully bother demonstrating what we know and what we don't and yet they remain unmoved, and usually it's then when they entrench in nonsensical wording ("you will see how we are right") or even personal attacks.

      It's a true pity but that's how the Internet is sometimes...

      Delete
  17. “So "Farmer" (without "Metal Age Invader") in your African case should mean ancestry from Neolithic Levant (or its African partial precursors) and "Invader" should mean Zagros Neolithic affinity (and not Indo-European, as they never went so far South before the Persian conquest of Egypt).”

    Thanks for summary, I’m learning more about a region I’ve known very little about.
    What do you mean exactly by "Indo-European" - is this an ancestral component? If you could provide me a link/reference for my own learning, that’d be great.



    “… my tentative conclusion is that R1b-V88 originated in or near the Zagros area (this might also apply to R1b in general, as happens with its "cousin" R1a, R1 as such coud be from Pakistan though but is much much older). This area was surely also at the origin of J2, which seems to have expanded southwards maybe since the late Neolithic/Chalcolithic or in the earliest Bronze Age at the latest, and that in Africa is a clear marker of Semitic inflow (usually very shallow anyhow).”
    This is a tangent – feel free not to reply: Are you familiar with a “Ted Kandell” by any chance? I’m curious to know your stance on the spread of Semitic languages in the Levant. He seems rather confident with his conclusions, I posted his claims from yfull’s official facebook group to gather opinions from more learned members on anthrogenica: http://www.anthrogenica.com/showthread.php?10771-Ancient-Egyptian-mummy-genomes/page8&highlight=nilosaharan



    “For what I read the most common theory about the origin of Chadic languages/peoples imply a late Green Sahara (c. 7000 BP) pastoralist migration northwards first and southwards later but it's debated.”

    I know you’re not a big fan of “molecular-clock guesstimates” :) but these yfull dates corroborate with your guess to a “T”. R-V88 dates within branch R-Y8447; formed 7700 ybp, TMRCA 7000 ybp. The Nigerian caught my eye, since the location is stated as Kano, within the Sahelian geographic region, I wouldn’t doubt there’s ties to neighbouring Chadic speakers. Within this subclade, the Nigerian shares a 7000ybp MRCA with a Pole, Slovak, Saudi, Kuwaiti and Sardinian among others: https://www.yfull.com/tree/R1b/

    ReplyDelete
    Replies
    1. The study this entry is about (Lazaridis 2014, pre-pub in 2013) is seminal precisely in our learning about this Indoeuropean(IE, also "Kurgan" for their typical princely tombs) invasion of much of Europe in the Copper Age (and later to the rest of Europe). The "ANE" component, while not strictly describing the IE input in Europe, is clearly related to them. Other more recent studies have developed the analysis by sampling various early IE cultures (Samara, Yamna, Corded Ware, etc.) and they are now clearly perceived to be a mix of Zagros (or "Iran") Neolithic and Eastern European aboriginals (EHG), who are the ones providing the "ANE" (or Paleosiberian element).

      I'm not sure how much you know or ignore about Indoeuropean languages and their expansion but it's a fundamental force in the shaping of our world, with more than half modern humans speaking natively Indoeuropean languages (Romance, Germanic, Slavic, Indo-Iranian, etc.)

      I dedicated a page to IE early expansion: https://forwhattheywereweare.blogspot.com/p/blog-page.html

      ... some details may need polishing but it's quite good for a basic understanding, much better than the typical Wikipedia map you see in many presentations on the issue. However it does not deal with the latter (and maybe better known) phases of IE expansion into Western Europe (Celts initially), Italy (Italics, incl. Latins), India (Indo-Aryans) and Iran (Iranics, already in the Iron Age). In the Iron Age, Indoeuropeans were gradually replaced by Turkic peoples in the steppes, who formed important historical states/hordes such as the Gökturk and the Hunnic empires (the Mongol Empire can also be considered in this group). Horse-based mobility was very important in their swift expansion, as was for Turkics later on.

      Some entries in this blog that may be of interest regarding this matter:

      · https://forwhattheywereweare.blogspot.com/2016/06/ancient-genomes-from-neolithic-west-asia.html (not really about IEs but about West Asian Neolithic from a genetic viewpoint but relevant for our discussion anyhow)
      · https://forwhattheywereweare.blogspot.com/2015/09/detailed-analysis-of-ancient-atapuerca.html (in the ADMIXTURE analysis you can see how different are French, Spanish and Basques in the Caucasus/Central Asia component, which is directly associated to IE inflow).
      · https://forwhattheywereweare.blogspot.com/2015/06/alentoft-2015-more-ancient-dna-from.html (again the ADMIXTURE graph shows the various Central Eurasian "teal" components and probably IE influence in various European populations ancient and modern).
      · https://forwhattheywereweare.blogspot.com/2015/02/kurgan-ancient-dna-suggests-major.html (another key work on IE migration into non-Eastern Europe)

      Delete
    2. Not familiar with that Ted Kandell guy. But I'd tentatively subscribe to Agamemnon criticism in the thread you linked to. To my knowledge, Semites (people speaking Semitic languages) expanded from the so-called Circum-Arabian Pastoralist Complex (CAPC), which in turn may derive from the more semidesert specialized PPN (and hence Natufian ultimately) branches.

      J1 MUST be much older than Hurrians or Semites in any case, macro-haplogroup IJ surely split in the early Upper Paleolithic, first between I (European in principle but also found at low frequencies in Turkey and Sudan and even in the Guanche mummies of Canary Islands) and J and this one soon after between J2 and J1. J2 looks northerner or "highlander" (i.e. related to the Zagros archaeological area probably), while J1 seems southerner or "lowlander" (i.e. related to the Palestine archaeological area plausibly, but also to NE Africa, where it is very diverse, and, oddly enough, to the Eastern Caucasus). There is a genetic duality in West Asia, which plausibly correlates to the Zagros/Palestine duality in the archaeological record, and this duality is not just in the Y-DNA but also exists in the autosomal DNA. See: http://forwhattheywereweare.blogspot.com/2013/03/west-asian-autosomal-genetics-two.html

      Delete
    3. Pardon my ignorance -- to be quite frank, my initial interest coming into genealogy has been focused primarily on private ethnic genealogy centered within Central-East, Central-North and North-East Africa. As an amateur, I felt it important to first fully grasp an understanding of my own anthro-genealogical history before delving into others.
      I've only fairly recently gained interest in "ANE" and W-Asian components after realizing there'd been an ancient impact on my related ethnic groups, and apparently "us" to them in the case of A-M13's minor prevalence in EEF populations (*perhaps also reflective in some of the "Basal-Eurasian" component)

      That being said, I greatly appreciate your blog and your holistic interest in genealogy. This was (and still is) the only blog on the internet where I found anything related to my maternal haplogroup (L3x2b). There's also been quite a few other esoteric subjects related to Tropical-African populations that I've enjoyed and couldn't find anywhere else.

      Thank you for the links related to the IE expansion -- I've since read through them a few times and found them to be a great reference!

      Delete
  18. “The fractions are small and it might be just noise but the results are consistent within the ethnic groups”

    That’s what I find so intriguing – but then again, my maternal ES ethnic group only arrived in the West-Nile region of Uganda in the last ~400 years. Despite episodic introgression with the local Central-Sudanics, I guess there’s still been enough time to observe slight differences in admixture. Hopefully there’ll be some future studies to confirm this, although I’m skeptical there’ll be any in the region anytime soon.

    *This topic is only indirectly related to Eurasian migrations, if you feel the need not to approve the post, do not fret – we can engage in private correspondence.

    I’d love to get your thoughts on this, it’s another Eurasian element I’ve noticed above non-negligible levels, this time Ancient South Eurasian and East-Asian surprisingly enough.

    Here’s the same Gedrosia K6 results, this time with all components: http://i.imgur.com/SUSwN47.png

    *the “Metal Age Invader” component from FTDNA’s ancient origins is still apparent with these results via the Ancient North Eurasian affinity (albeit at a lower affinity).

    I’m guessing the Ancient South Eurasian and East Asian affinities are the result of back migrations from Asia to Africa – would these statements make light of these affinities? :

    “The highest frequencies worldwide of macrohaplogroup M are observed in Asia, specifically in Bangladesh, China, India, Japan, Nepal, and Tibet, where frequencies range from 60%-80%.”

    “The timing of the proposed migration of M1 and U6-carrying peoples from West Asia to Africa (between 40,000 to 45,000 ybp) is also supported by the fact that it coincides with changes in climatic conditions that reduced the desert areas of North Africa, thereby rendering the region more accessible to entry from the levant. This climatic change also temporally overlaps with the peopling of Europe by populations bearing haplogroup U5, the European sister clade of haplogroup U6.”

    https://en.wikipedia.org/wiki/Haplogroup_M_(mtDNA)#Asian_origin_hypothesis

    More of my musings on the topic with surprising Tamil similarities in Chadic speakers found in Northern Cameroon:

    http://www.anthrogenica.com/showthread.php?7854-Coincidental-Linguistic-Similarities-Found-Outside-of-Africa/page2

    Similar to the Nubian/Vasconic similarities, I’d like to see a linguistic academic disprove or validate these claims.

    ReplyDelete
    Replies
    1. Not sure what "Ancient South Eurasian" is, really: is that ASI (Ancient South Indian, surely pre-Neolithic aboriginal South Asian) or is it West Asian Neolithic stuff?

      What I do find interesting in your link is the ANE (paleo-Siberian) component being clearly stronger in Central Sudanic, much like the "Metal Age Invader" one of the previous analysis, and probably pointing in the same direction.

      Re. mitochondrial DNA expansion in Asia, I have worked hard on it on my own (it's a relatively easy phylogeny to work with) and I strongly stand on M being clearly older than N, after all M is only 3 "control region" mutations downstream of L3, while N is 5. Also N (and its offspring R) are strongly related to the secondary expansion of Y-DNA K2 (incl. P (incl. Q and R) and NO) from SE Asia, surely after the Toba catastrophe, which must have left many niches open.

      See: https://forwhattheywereweare.blogspot.com/p/continuing-with-joint-series-in-spanish.html

      Delete
  19. If interested, here's post I made with other members on the forum (predominantly of African decent). When we compared our ASE and EA affinities, we found the detection more or less wide-spread and variable at 0-minor detection throughout North and East Africa. Nothing was really conclusive despite the consistently non-existent/near-negligible detection in African samples/populations that weren't found in North or East Africa. If anything can truly be deduced from this amateur ancient-admix experiment, I'd guess it'd be a fair assumption that the hypothesized Asia back to Africa migration didn't penetrate into populations found in the deeper hinterland regions of Africa.

    http://www.anthrogenica.com/showthread.php?9831-Ancient-Eurasia-K6-Deducing-Traces-of-Back-to-Africa-Migrations

    ReplyDelete
    Replies
    1. I don't understand everything (ASE??) nor I really have time to go through all the details. But IMO there must have been at least one Paleolithic migration from West Asia to NE Africa (Egypt, Sudan and The Horn, judging on modern DNA), linked to Y-DNA J1 (plausibly also T) and mtDNA X1, M1, etc. In addition to that there may have been some Neolithic inflows and then of course the Semitic expansions (which seem associated to the presence, always weak, of Y-DNA J2).

      Less important but worth considering is the migration from SW Europe into NW Africa, which I associate to Iberomaurusian culture (aka Oranian), this may have expanded eastward into the Nile, what may explain the presence of European autosomal component in Egypt and Y-DNA I in Sudan and the Guanche mummies (in these also R1b, which may be West European but yet to be clarified).

      Delete
    2. PD- In any case it is somewhat interesting the high frequency of ANE in North Africa. This may be an artifact caused by incomplete list of "source" populations. Let's not forget that "ANE" is two Siberian samples from c. 20 Ka ago, and therefore half-way in the process of differentiation of West Eurasian populations. In other words: it can stand for any West Eurasian affinity not considered in the "source" or comparison populations.

      It also offers an alternative possibility for the high ANE in Central Sudanic: it's nothing but a marker of North African admixture.

      Delete

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 (it may take some time before your comment is published, thank that to Trumptards).