Saturday, June 18, 2016

Natufians and Neolithic Levantines lack African admixture?

Whilst Neolithic Levantines and Natufians seem to have a good amount of what looks to be Y-DNA E-M35 (specifically 2 E-Z830 samples among the Natufians) and seemingly no Y-DNA J so far- :


Natufians (Epipaleolithic Levant):

I0861: E1b1b1b2(x E1b1b1b2a, E1b1b1b2b)
I1069: E1b1(xE1b1a1, E1b1b1b1)
I1072: E1b1b1b2(xE1b1b1b2a, E1b1b1b2b)
I1685: CT
I1690: CT

Levant_Neolithic:

I0867: H2 (PPNB)
I1414: E(xE2, E1a, E1b1a1a1c2c3b1, E1b1b1b1a1, E1b1b1b2b) (PPNB)
I1415: E1b1b1 (PPNB)
I1416: CT (PPNB)
I1707: T(xT1a1, T1a2a) (PPNB)
I1710: E1b1b1(x E1b1b1b1a1, E1b1b1a1b1, E1b1b1a1b2, E1b1b1b2a1c) (PPNB)
I1727: CT(xE, G, J, LT, R, Q1a, Q1b) (PPNB)
I1700: CT (PPNC)


-they don't seem to show any affinities for African populations (ones who lack West Eurasian admixture or have negligible amounts of such admixture) on an autosomal level according to this Pre-print:

"However, no affinity of Natufians to sub-Saharan Africans is evident in our genome-wide analysis, as present-day sub-Saharan Africans do not share more alleles with Natufians than with other ancient Eurasians (Extended Data Table 1). (We could not test for a link to present-day North Africans, who owe most of their ancestry to back-migration from Eurasia)."

This is quite interesting, truth be told. It implies that, despite seeming clearly old, the African admixture in the Levant is perhaps, at least, Post-Neolithic? Some of it (particularly non-East African cluster-related admixture), as I've noted in the past, is definitely owed to the Arab Slave Trade but some of it does also seem rather ancient.  




As for PCA (Principal Component Analysis) positions; Natufians in particular look to cluster essentially to the direct east of Anatolian Neolithic and Early European Farmers (and somewhat southwards of some of them) whilst Levantine Neolithic Farmers seem less Basal Eurasian and more northern shifted than their Epipaleolithic predecessors. I'll be more interested in seeing David over at Eurogenes throw these samples into his own Pan-West Eurasia PCA and a global PCA once the genomes are made publicly available when this paper is published in a peer-reviewed journal (a month or less, not sure). The PCAs in these studies can be a bit wonky due to issues like projection-bias.

I'll be interested in seeing if David and other 3rd parties will either confirm or deny that Natufians lack African admixture (whatever Basal Eurasian turns out to be aside).



One somewhat off-topic matter to note is that it seems as though we've more or less discovered a close equivalent to David's "ENF"/Near Eastern cluster from his old Fateful Triangle~K=8 model, a cluster also noted by Lazaridis and company prior to the release of this pre-print. 

The Natufians and Neolithic Levantines are essentially dominated by the blue ADMIXTURE cluster that seems to make up most of the ancestry in the Neolithic Anatolians and Early European Farmers. This seems, to me, like we're seeing Southwest Asian/ENF/Near Eastern in the flesh. The way it's differentiated from Neolithic Iranians is intriguing though given that Natufians (Epipaleolithic Levantines) and Neolithic Iranians are comparable in terms of Basal Eurasian ancestry; it must be the non-Basal Eurasian ancestry that's causing the differentiation or there's just really significant genetic drift at play here or both.


In this model above, where Natufians and the Hotu cave Iranian were not included, the Caucasus Hunter-Gatherer (CHG) samples seem to carry notable EHG/ANE-related ancestry as prior analyses like David's K=8 ADMIXTURE run might've implied. Initially, I figured what was differentiating the Neolithic Iranians from Natufians was perhaps ANE-related admixture in the former but that might not really be the case.

As I said in a prior post; things will become more clear when third-parties like David over at Eurogenes get their hands on these genomes once the paper's published at a peer-reviewed journal and also once we have even more ancient genomes someday.

Reference List:


Note:

1. I also must add that we should be cautious given that these are just 6 Natufian samples from one area. There might be some variation with future samples from other areas in regards to Y-DNA although I doubt there will be very much variation in terms of autosomal DNA given their similarity to Levantines from later periods.

15 comments:

  1. Did the study imply how comparable the modern levantines and Arabians are to the ancient levant samples?

    ReplyDelete
    Replies
    1. Yes, they did, if I recall correctly. Go through the supplementary materials for details. The modern groups seem "closer" to Neolithic and Bronze Age Levantines though which makes perfect sense but the Fst between them and the Natufians is probably exaggerated by the Natufians being foragers and thus having rather low population sizes which would be conducive to drift. You see this with European Hunter-Gatherers as well.

      Delete
  2. Dohaneus Kabylicus Atlas-icusJune 19, 2016 at 2:20 AM

    So Natufians are basically Basal Eurasian + UHG (once related to WHG) ???

    ReplyDelete
    Replies
    1. In my opinion, anyway. We'll see as things progress, ya Dohan. ;-)

      Delete
  3. Dohaneus Kabylicus Atlas-icusJune 19, 2016 at 2:21 AM

    Great coverage of the study my Broski !

    ReplyDelete
  4. I think they do have "African ancestry". It's just not ancestry traditionally assigned to Africans. We can step back and use common sense and make a clear observation that North East African Y-DNA phylogeny thus far has not had a clear associating autosomal signature. I am of the opinion that we are kind of looking right at it.

    And just as ancient Levantines define the ancestry in East Africans better than the farmers that contributed to Europe, I am willing to bet that SOME ancient NE African genomes would be even better in describing that ancestry (of course I am excluding the affinity of later migration into the horn.)

    ReplyDelete
  5. @astenb

    'I think they do have "African ancestry"'

    Absolutely.

    The answer to the question of where exactly 'Basal Eurasians' lived within the last 100,000 years will also challenge a lot of people's definitions of what is "African" and what is "Eurasian".

    I would suspect that there was a lot more ancient population structure in North and East Africa. A lot of this has been homogenized more recently (the last 15,000 years) by migrations within Africa and including back-migrations from Eurasia.

    The data from this paper seems to rule out any admixture between Sub-Saharan Africa and ancient Natufians. That is the first step towards getting a solid understanding of the region in ancient times.

    ReplyDelete
  6. Another finding that is interesting for pop culture arguments (like which actors 'should' be playing ancient Middle Easterners in films), is that one of the Neolithic Levantine samples (Ain Ghazal, Jordan. Sample AG83_5) seemed to have both of the major 'European' light skin alleles.

    It will be very interesting when we finally get DNA from ancient Egypt, because this person lived only ~600 kilometers from Giza and several thousand years before the well known Pharaohs.

    Even one of the Natufians had the light skin allele of SLC24A5 (from Raqefet Cave, Israel) around 11,000 BCE.

    It looks to me like the high latitude vitamin D hypothesis for lighter skin is not matching up with the ancient genomics very much. I think it was something else in the diet that caused the selection, not simply the lack of vitamin D.

    ReplyDelete
    Replies
    1. "Another finding that is interesting for pop culture arguments (like which actors 'should' be playing ancient Middle Easterners in films), is that one of the Neolithic Levantine samples (Ain Ghazal, Jordan. Sample AG83_5) seemed to have both of the major 'European' light skin alleles."

      Yes, I was somewhat surprised to see that one of the samples had the derived allele for SLC45A2:

      rs16891982

      Levant Neolithic I0867 CC
      Levant Neolithic I1701 CC
      Levant Neolithic I1707 GG

      But this same sample doesn't have the derived allele for SLC24A5:

      rs1426654

      Armenia Early Bronze Age I1633 AA
      Iran Copper Age I1661 AA
      Iran Neolithic I1290 AA
      Iran recent I1955 AA
      Levant Bronze Age I1706 AA
      Levant Bronze Age I1730 AA
      Levant Neolithic I0867 AA
      Levant Neolithic I1707 GG
      Natufian I1072 GG

      The other Neolithic Levantines do have the A allele for SLC24A5 though, which implies there was some clear de-pigmentation here but it's interesting that this one Natufian Genetiker (https://genetiker.wordpress.com/2016/06/22/phenotype-snps-from-the-ancient-near-east/) shared a result for is GG for SLC24A5. It's interesting though that the derived variant of SLC45A2 only really appears among Eastern European Hunter-Gatherers (EHGs) when looking at various pre-historic populations in the Ice Age Europe study:

      https://drive.google.com/file/d/0B2ARnUeK-Y8WdmM1aF9VdS1rRTg/view?usp=sharing

      But it's not surprising to find the derived allele for SLC24A5. I'd be shocked if it didn't appear at all, quite frankly. Even Horn Africans like Somalis and Beta Israels have it:

      http://anthromadness.blogspot.ae/2016/01/some-horn-african-phenotypes-supposed.html

      "It looks to me like the high latitude vitamin D hypothesis for lighter skin is not matching up with the ancient genomics very much. I think it was something else in the diet that caused the selection, not simply the lack of vitamin D."

      Oh, yeah... That whole "high latitude" hypothesis has seemed pretty invalid for a while now.

      Delete
  7. Another great post by you my lord Sheikh! :D

    Out of topic, I notice that a lot of West-Central African and East African Nilotic populations show very tiny negligible amounts (around 1-4%) of the Cushitic component of K=14 in tables S8 and S9 in the Supplementary figures and tables of Tishkoff 2009:http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2947357/#SD2

    I wonder whether West-Central Africans and East African Nilotes really have extremely low negligible amount of Cushitic admixture or is it just statistical noise. In my opinion, it might be either case :P

    Another intereesting thing that I notice is that the Cushitic component peaks at around 78.8 percent in Iraqw, who are South Cushitic speakers. Do you know how much African and Eurasian are the Iraqw? :P

    ReplyDelete
    Replies
    1. "Another great post by you my lord Sheikh! :D"

      Lol, thanks, Khuuri.

      "Out of topic, I notice that a lot of West-Central African and East African Nilotic populations show very tiny negligible amounts (around 1-4%) of the Cushitic component of K=14 in tables S8 and S9 in the Supplementary figures and tables of Tishkoff 2009:http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2947357/#SD2

      I wonder whether West-Central Africans and East African Nilotes really have extremely low negligible amount of Cushitic admixture or is it just statistical noise. In my opinion, it might be either case :P"

      Depends on the populations in question. If we're talking about Nilo-Saharan speakers in Southeast Africa then some of them definitely do have such admixture either negligibly or substantially. If we're talking about ones in Sudan; they've historically lived in the same general area as North-Cushitic (Beja being the only surviving member of this branch) speakers and populations like Nubians and Sudanese Arabs who seem to have "Somali-like" ancestry in them so it's plausible. But, for West-Central Africans... Depends on the ones in question. I.e. It's plausible that some Central Africans in areas like Congo have such admixture but if we're talking about Western Africans proper like Yorubas; it's most likely just noise.

      "Another intereesting thing that I notice is that the Cushitic component peaks at around 78.8 percent in Iraqw, who are South Cushitic speakers. Do you know how much African and Eurasian are the Iraqw? :P"

      No, Tishkoff et al. 2009 did a poor job of telling us how West Eurasian the Iraqw were, if I recall correctly. Apologies, my friend.

      Delete
    2. Thanks Ya Great Sheikh!

      I forget to mention I also saw the Indian component in very negligible amounts in many African populations in the same K=14 tables in tables S8 and S9. I wonder if is the Indian component is genuine or is it just noise. What do you think?

      Another question: I notice in this supplementary table 4 (from another study): http://3.bp.blogspot.com/-7OA6NeHHdxo/VJTdh-XNzKI/AAAAAAAAB4Y/VgvWohRDC18/s1600/105r589.jpg

      I notice that Fulas here are only 10-12% Eurasian on average which is rather intriguing for me. From what I read and seeing some results, most Fulas are around 15-30% Eurasian (correct me if I am wrong no the ancestry percentage). Do you know where the Fula results are from?

      Also I was surprised to find out some very small Eurasian amount in Senegambian populations like Mandinka, Wolof and very negligible, almost non-existent in the Diola. I think these groups likely get their Eurasian from mixing with the Fula or similar groups. What I wonder are there Mandinka, Wolof, Diola who don't have any Eurasian like most other West-Central Africans? Maybe we need more samples.

      Delete
    3. "Thanks Ya Great Sheikh!"

      You're welcome again, ya Khuur!

      "I forget to mention I also saw the Indian component in very negligible amounts in many African populations in the same K=14 tables in tables S8 and S9. I wonder if is the Indian component is genuine or is it just noise. What do you think?"

      I wouldn't say it's real at all. It's either noise or somewhat of a stand-in for that broadly Eurasian affinity I told you about a while back.

      "I notice that Fulas here are only 10-12% Eurasian on average which is rather intriguing for me. From what I read and seeing some results, most Fulas are around 15-30% Eurasian (correct me if I am wrong no the ancestry percentage). Do you know where the Fula results are from?"

      I don't know where those samples are from if that's what you mean but that figure is from this paper:

      http://www.nature.com/nature/journal/vaop/ncurrent/full/nature13997.html

      Fulanis can be rather heterogeneous, at least from what I've noticed. I'm no "expert" on them.

      "Also I was surprised to find out some very small Eurasian amount in Senegambian populations like Mandinka, Wolof and very negligible, almost non-existent in the Diola. I think these groups likely get their Eurasian from mixing with the Fula or similar groups. What I wonder are there Mandinka, Wolof, Diola who don't have any Eurasian like most other West-Central Africans? Maybe we need more samples."

      Yes, we might need more samples... I'm sure there's some variability among some of these populations and some people might be totally "Non-Eurasian" given how close to being so the admixed groups already are. But again, I'm not "expert" on West-Central African population genetics but I came to a similar conclusion as you regarding the source of their West Eurasian admixture. :-)

      Delete
  8. Sorry for very late reply. So you also conclude that the Fulanis are a likely source for their minor negligible West Eurasian admit?

    ReplyDelete