January 17, 2013

Very skeptic on claim of Neolithic flow from India to Australia

I feel quite skeptic about the claims held by this paper but in any case it is worth mentioning.

Irina Pugach et al., Genome-wide data substantiate Holocene gene flow from India to Australia. PNAS 2013. Pay per view (6-month embargo, then freely accessible) → LINK [10.1073/pnas.1211927110 ]

Abstract

The Australian continent holds some of the earliest archaeological evidence for the expansion of modern humans out of Africa, with initial occupation at least 40,000 y ago. It is commonly assumed that Australia remained largely isolated following initial colonization, but the genetic history of Australians has not been explored in detail to address this issue. Here, we analyze large-scale genotyping data from aboriginal Australians, New Guineans, island Southeast Asians and Indians. We find an ancient association between Australia, New Guinea, and the Mamanwa (a Negrito group from the Philippines), with divergence times for these groups estimated at 36,000 y ago, and supporting the view that these populations represent the descendants of an early “southern route” migration out of Africa, whereas other populations in the region arrived later by a separate dispersal. We also detect a signal indicative of substantial gene flow between the Indian populations and Australia well before European contact, contrary to the prevailing view that there was no contact between Australia and the rest of the world. We estimate this gene flow to have occurred during the Holocene, 4,230 y ago. This is also approximately when changes in tool technology, food processing, and the dingo appear in the Australian archaeological record, suggesting that these may be related to the migration from India. 

The evidence for this claim is all derived exclusively by statistical inference on autosomal DNA. Suspiciously enough, even if the authors claim admixture levels of as much as 11% and as recent as a mere 4000 years ago, no patrilineage (Y-DNA) nor matrilineage (mtDNA) [correction: see update below] has been ever detected that could be associated with this purported migration. 

Additionally c. 4000 years ago Southern India, the alleged origin of the genetic flow, was already immersed in a flourishing agricultural economy and it looks very strange that the migrants, people who were exchanging crops with Africa for example, would not carry a single element of this new economy to the island continent. Of course this inconsistency could easily be fixed by merely arguing that the molecular clock estimates used tick too quickly, which is a general problem anyhow and therefore no real surprise.

If the hypothesized migration happened earlier, in the Epipaleolithic or Late Upper Paleolithic, then it would also be easier to explain that, with smaller populations, genetic drift could have caused the extinction of whatever Indian uniparental markers that the migrants carried with them initially. It still causes my eyebrows to rise instinctively. 

Even then, if this was the case, we should be able to identify some sort of techno-cultural elements that the migrants may have carried with them, like microlithic stone technologies or whatever. As far as I know nothing of the like exists. 

The only techno-cultural burden that the migrants might have brought with them to Australia would therefore have been the dingo, but this dog has lots of relatives in Island SE Asia, where the authors could not detect any significant Indian admixture.

So the hypothesis looks weak to me. Let's see the evidence they present:



Above we can see the ADMIXTURE K=4 result, probably not the optimal one (which would probably produce an Australian-specific cluster (mostly but not fully masked as Papuan) and surely two different Indian ones, partly masked as European and Onge affinity) but the one the authors decided to show us as evidence for their hypothesis.

Not only this is surely not the optimal clustering level but also Australian Aborigines are comparatively undersampled, while Indian weight is overwhelming. This is a clear example of how NOT to design a scientifically useful sampling strategy for ADMIXTURE-like comparisons like this (because oversampled populations tend to overshadow the rest just by the weight of numbers). 

As it is, this graph proves nothing but rather suggests that some Indian affinity is part of Australian Aborigine ancestral or founder specificity, when compared with Papuans. This may have many explanations first of which is a mere artifact by reason of a poor sampling and depth design of the experiment. ADMIXTURE is a powerful neutral tool, just a like a test tube or the Geneva particle accelerator, but what we do with it may well not be neutral, either by reason of mischievous manipulation or mere error.

In this case I find the test very poorly designed and executed. If I have some time later in the weekend, I may try to perform an alternative test according to my humble possibilities - I promise nothing however.

A complementary test that the authors perform used Tree Mix. As I have discussed elsewhere, TreeMix often produces very strange results and I do not consider it a reliable tool at all, but for whatever is worth here it is what they got:



While the purported migrations generated by the Tree Mix algorithm appear to suggest a secondary genetic flow from India to Australia (orange arrow at C) the data on which such result is based (D) only gives the most tenuous level (green) of extra genetic affinity between Southern Indians (DRA) and Australian Aborigines (AUA). Meanwhile the highly questionable algorithm identifies Dravidians and North Chinese (CHB) as being genetically very close (blue), when they are not in fact.

So what do I get from this paper? TreeMix' usual senseless noise and apparent mismanagement of ADMIXTURE, a powerful tool when used properly.

Less than inconclusive, I'd say. But your take of course.


Update: G Horvat (see comments) points me to Kumar 2009 (so far unchallenged at PhyloTree)  for a shared mitochondrial lineage between Australia and India, known as M42. This haplogroup has the following structure (each → indicates a coding region mutation according to PhyloTree, Kumar originally listed a few more):

    • → M42'74 
      • → M42 
        • →→→→ M42a (Australian Aborigines)
        • → M42b (India)
      • →→ M74 (South China, Vietnam, India)
This allows for a potential mtDNA backing of this purported connection, however it is a very small lineage and Kumar claimed that M42 coalesced long ago, in the context of the first colonization of Asia and Australasia by Homo sapiens:

The divergence of the Indian and Australian M42 coding-region sequences suggests an early colonization of Australia, ~60 to 50 kyBP, quite in agreement with archaeological evidences. 

Yet the relatively long stem leading to M42a does allow for a later time-frame of arrival to Australia. Neolithic anyhow looks still most unlikely to me.


Update (Jan 18): Dingo DNA:

An important element to consider here are the origins of the dingo as the Australian wild dog is known. This dog variant suffered a strong founder effect upon arrival to Australia described mainly by two variants of the haplotype A29. This lineage only links to East Asia however, having arrived almost without doubt, via Indonesia from mainland East Asia (either Indochina or China or both).

It is clearly not related to Austronesian expansion and could have arrived either within the early Neolithic of ISEA (arguably Austroasiatic in language) or even earlier. At least one of the papers I checked rather supports a pre-Neolithic introduction and certainly before the archaeologically supported age of c. 3000 years ago.

The Y-DNA of dingos also shows a strong founder effect (only two haplotypes, with overlapping but distinct distributions) and again the most obvious connections seem to be in SE Asia.

See (freely accessible):

Update (Jan 18): It is probably interesting also to mention that Australian Aborigines show no difference with Papuans in their overall amount of Denisovan ancestry. This also appears as contradictory with the idea of significant external admixture, which should have diluted at least minimally that Denisovan component (Indians have none).


Update (Apr 7): A new "working paper" has been published on this matter, sharing my critical stand towards the sloppiness of Puhach's team but still considering plausible a Holocene gene flow from India. I have commented in a new entry.

    122 comments:

    1. With regards to mtDNA links, there was this paper but the conclusion is not too certain, IMO, due to the frequent recurrence of mtDNA mutations:

      http://www.biomedcentral.com/1471-2148/9/173

      I've been trying to locate references to A29 dog mtDNA sequences in India. I thought the sequences closest to those of the Dingo were all in SE Asia.

      ReplyDelete
      Replies
      1. Good find, thanks. As far as I can check the claim has not been challenged so far and the survey was done on full mitochondrial sequences first submitted to the Anthropological Survey of India, 27 on 14-OCT-2008. All branches are described by one or more coding mutations, looking pretty solid to me.

        Refs. http://www.phylotree.org/tree/subtree_M.htm, http://www.ncbi.nlm.nih.gov/nuccore/FJ380216 (and others taken from the PhyloTree source).

        However notice please the extremely contradictory age estimates, with Kumar claiming that:

        The divergence of the Indian and Australian M42 coding-region sequences suggests an early colonization of Australia, ~60 to 50 kyBP, quite in agreement with archaeological evidences.

        Another element to consider is that the related lineage M74 is found in SE Asia (South China, Vietnam) as well as India (Nihal), according to the info we have been gathering in the wiki.

        This lineage does suggest a flow from South Asia to SE Asia and Australia (maybe together both branches or maybe not). However, judging from Kumar's estimates (or even shortening them somewhat) that would have happened within the wider wave colonization of East Asia and Australasia from South Asia. Even if we'd consider it (tentatively) a secondary "late" wave it does not look "recent", hanging M42'74 just one coding region (CR) mutation away from M and M42 one mutation downstream from M42'74. On the other hand, the Australian branch hangs downstream of 5 CR mutations (judging from Kumar'09) what allows for a longer coalescence, maybe still in South Asia (or half-way, in a "lost" SE Asian locality).

        Even in the latter case, I still find it extremely difficult to believe in a Neolithic time-frame for this arrival to Australia.

        As for the dingo I can't say. I thought that their DNA had been sequenced and compared but a search I performed earlier produced no results other than morphological comparisons with Island SE Asian and also Indian dogs. Both seem related but the closest ones are probably those from Indonesia.

        Delete
    2. Maju - Do a search for the following terms in Google Scholar: Dingo A29 mtdna. There has also been at least one Dingo Y chromosome study.

      ReplyDelete
      Replies
      1. Got the mtDNA ones. I'll add a mention as update, thanks again: very useful contributions.

        Delete
    3. When it comes to agriculture, immigrants not bringing it with them, this might have been a very small group of people ending up in Australia unintentionally without seeds etc and incapable of keeping up their way of life - if the immigration even happened.

      That said, last year archaeological evidence was found for Lapita culture in the Torres Strait Islands, dated at about 3000 years ago. It wouldn't have been a long step for them to get to Australia after that, so similar claims like the Indian connection might come up from that direction.

      ReplyDelete
      Replies
      1. Nobody questions that the agricultural peoples of SE Asia and Melanesia could reach Australia and in fact did... punctually and without other obvious long-term consequences than the introduction of the dingo. The question here is about a necessarily larger and more consolidated migration, which was able to influence Australian Aborigines as much as 11% of their ancestry (curiously without apparently diluting the "Denisovan" ancestry, which remains as high as among Papuans).

        Delete
      2. This comment has been removed by the author.

        Delete
      3. After some 2400 years even a very small group could have had a that kind of effect, especially if the population it was contact with was a relatively small. This study also is not about all the Aboriginals of Australia, just from the NW and the exact population history there is relatively unknown. The aboriginal populations were hardly static, so we could make - for argument's sake - a totally theoretical assumption that later immigration from other parts of Australia could have greatly influenced the population that was in contact with the immigrants, possibly diluting their effect on the NW population and also helping to keep the Denisovan heritage on the level that it is.

        Delete
      4. Where are the other less slippery data: the Y-DNA, the mtDNA, the archaeological evidence... this is a mere artifact, almost a scientific prank. Your proposal seems so unlikely to me that I cannot take it even half-seriously without some clear evidence.

        "... diluting their effect on the NW population and also helping to keep the Denisovan heritage on the level that it is".

        That's simply not possible (after all we are talking of all AAs almost evenly sharing that component, no variations we can spot). The only reasoning that could be applied to the lack of dilution of the Denisovan component would be that it was once higher in AAs than Papuans, so dilution did happen but exactly to the Papuan reference level... but that would be a most strange coincidence.

        Your suggestions sound to me as quasi-religious rationalizations ex-post-facto of data that is simply wrongly generated. If you want to have faith in the tools and the researchers' ability to use them properly and without trickstery, then you have to reach out to such unlikely explanations, of course. But I find much easier simply not to have such faith.

        I find much more reasonable to discard TreeMix produce as noise (more often than not at least, in my experience), ADMIXTURE's one as a very likely mismanagement by the authors (we are not presented with the proper list of K-depths so we can see if the appearance of an Indian component is real or mere artifact caused by the 60Ka ago origin in that area, where most diversity is preserved, nor we are told of a cross-validation that supports that particular slice as optimal - also the already mentioned oversampling issue that normally causes heavy distortions in ADMIXTURE results) and the molecular-clock-o-logical estimates as the usual nonsense that sadly surrounds this matter.

        So all the data we are offered is at best dubious, hence the conclusion is also dubious. Said that, it has one use, stated in the paper: to reject the hypothesis that most AAs carry European admixture.

        There seems to be some extremely thin India-Australia link in some haploid DNA (including Afghanistan and maybe other parts of Eurasia) but a simpler explanation is that it belongs to a late wave within the Middle Paleolithic process of Eurasian expansion or something like that. Not Neolithic.

        Delete
    4. Agree with all of the above, plus, a significant incursion of neolithic people from India would result in linguistic evidence, which has not been detected in Australia.

      In contrast, the Chipewyans of Canada, who speak a Na-Dene language, have only 10 percent of their genes from the later wave of migrants bearing Na-Dene culture yet were linguistically wholly transformed.

      ReplyDelete
    5. To really look at this, more data from in-between places (Indonesia, Timor, other little islands) and other Aboriginals from different parts of Australia plus more Papuans and other Melanesians are needed.

      There are always early adopters and late adopters or even Luddite type people for any new toolkit. That goes for the latest cell phone, fashion, and music in a modern context. Same applies for bigger techno-cultural packages.

      What if some of the Aboriginals were "Luddites" who had been more in touch with Island S.E. Asia until a new toolkit was introduced that they rejected. They then left or got pushed south until they eventually appear as immigrants, maybe bringing their own "Luddite" toolkit and some Indian admixture to Australia.

      Food for thought: http://en.wikipedia.org/wiki/Wandjina

      ReplyDelete
      Replies
      1. What should be done is to make (at least) a tripartite India, Papua, AA comparison, with roughly equal-sized samples, with ADMIXTURE set with cross-validation. This tells you how many clusters are most realistic (and which K values are very unrealistic). Most probably the realist result will show no appearance of admixture but a distinct homogeneous AA cluster, maybe with minor Papuan admixture but no Eurasian one.

        The DIY process is described pretty well by Razib at this post: http://blogs.discovermagazine.com/gnxp/2011/03/analyzing-ancestry-with-admixture-step-by-step/ - which includes all the necessary tools to start toying around with ADMIXTURE and a good global dataset, that you can reform to your likes. Sadly that dataset does not include Australian Aborigines and I fear I am not techy enough to be able to locate and insert a sample in the dataset.

        So I will not do it for the time being.

        The "luddite" hypothesis sounds totally far fetched to me. Even if AAs would have been radical "luddites", nothing should have prevented the Indian immigrants from establishing their own successful farmer colonies - after all, according to the hypothesis proposed by Pugach they were at some point in enough numbers to be 10% of the population and AAs were organized in small hunter-gatherer bands that could pose no effective resistance against such a large number of farmers. Also, even in the worst case, some animals like goats or maybe some renaturalized crops could have prospered even after abandonment by their human masters (as presumably did the dingoes).

        Similar scenarios can be reconstructed for example in the Neolithic of many parts of Europe but there the 10-20% Neolithic immigrants (with haplogroups like Y-DNA E-V13, G2a, etc.) were indeed able to establish themselves as soon as they arrived, no matter how "luddite" were the aborigines (aDNA shows them still distinct some time after arrival).

        Not sure what you mean with your wondjina link but I discussed recently an observed paleo-climatic phenomenon that seems directly linked to that mythology: http://forwhattheywereweare.blogspot.com.es/2012/12/megadrought-may-have-affected-nw.html

        Delete
    6. I am in general skeptical about these model based programs ability to detect real admixture between old world populations, especially those that may already be closely related whether due to geography or ancient demographics. It is very hard for these programs to discern if reported results are due to common ancestry or really admixture, and this is because of the recombining nature of Autosomal DNA, the chromosomes simply can not be traced back to a single common ancestor.

      ReplyDelete
      Replies
      1. Mostly agreed. If you look from many angles (different sampling strategies and such) and the results repeat in general terms (the proportions will surely vary somewhat) you can infer that those repeating patterns are saying something, looking at Fst distances (a complementary measure) and other values like cross-validation, you can get a decent idea. But it's not rocket science at all.

        Delete
    7. http://en.tengrinews.kz/science/Gene-study-settles-debate-over-origin-of-European-Jews--16087/

      ReplyDelete
      Replies
      1. Discussed here: http://forwhattheywereweare.blogspot.com.es/2012/08/ample-analysis-of-genetics-of-european.html

        See also: http://forwhattheywereweare.blogspot.com.es/search/label/Jews

        Delete
    8. "I feel quite skeptic about the claims held by this paper"

      I know you'll find it surprising but I'm inclined to agree with you. Regarding the dingo:

      "This lineage only links to East Asia however, having arrived almost without doubt, via Indonesia from mainland East Asia"

      That is a huge problem for the authors and perhaps shows that the dingo is a separate issue.

      "Even then, if this was the case, we should be able to identify some sort of techno-cultural elements that the migrants may have carried with them, like microlithic stone technologies or whatever. As far as I know nothing of the like exists".

      We do have ther Pirri points. Unfortunately all the references I could find are to extracts from books, but try this one:

      http://www.archaeologywordsmith.com/lookup.php?category=&where=headword&terms=Pirri+point

      The small tool tradition does match the date suggested reasonably closely.

      "The divergence of the Indian and Australian M42 coding-region sequences suggests an early colonization of Australia, ~60 to 50 kyBP, quite in agreement with archaeological evidences".

      That eliminates M42'74 from having anything to do with the movement suggested in the paper.

      "This lineage does suggest a flow from South Asia to SE Asia and Australia (maybe together both branches or maybe not)".

      Or possibly to both India and Australia from South China.

      "We also detect a signal indicative of substantial gene flow between the Indian populations and Australia well before European contact"

      For some reason the authors automatically assume such gene flow would be from west to east.

      "even if the authors claim admixture levels of as much as 11% and as recent as a mere 4000 years ago, no patrilineage (Y-DNA) nor matrilineage (mtDNA) [correction: see update below] has been ever detected that could be associated with this purported migration".

      We do have a patrilineage that suggests movement from near Australia back to India: Y-DNA K1. Most K-M9 is east of Wallace's Line. Just K1 is from west of the line, in South Asia. Could it be that it is an indication of westward movement across Wallace's Line a mere 4,230 years ago? But that leaves the problem of 11% in common. K1 is a very rare lineage.

      ReplyDelete
      Replies
      1. The Pirri Point tech is curious (looks totally like an MSA variant btw) but without knowing IF and how it might relate with South Asia (where mode 4 microlithism was mainstream since c. 38 Ka ago) we can't judge. On first impression (google for images) it looks a "primitive" Middle Paleolithic Levallois-based tech to me, much like MSA. If this diagnostic is correct then we should reject even more any idea of human or cultural waves of any relevance arriving to Australia in recent times. Sadly I do not know enough about the prehistoric techno-cultures of Australia to judge but my first impression is rather as negative evidence than positive one for the Pugach hypothesis.

        "That eliminates M42'74 from having anything to do with the movement suggested in the paper".

        Not necessarily. Remember that age estimates are just educated guesses without any weight as evidence.

        "For some reason the authors automatically assume such gene flow would be from west to east".

        That's what the shown Tree Mix and ADIMXTURE results (as I say questionable both) suggest. We see no "Papuan" genetics in India (while describing, by affinity, 90% of the Australian sample). Whatever one may think there's no way one can deduce an Australia → India flow from that data.

        "We do have a patrilineage that suggests movement from near Australia back to India: Y-DNA K1".

        And all P as well (P is South Asian at origin beyond reasonable doubt).

        But neither is found in Aboriginal Australia as such and your "near Australia" is probably mainland SE Asia (maybe Sundaland but hardly farther to the SE). The current knowledge of MNOPS [aka K(xLT)] is surely too shallow to justify your claim because is for sure that MNOPS holds several (now hidden) dual branches underneath and is likely that one of them includes all the Oceanian lineages (not sufficiently researched as of now).

        Even if we'd accept the current 7-parts division (which as I say is surely just an approximation to the real thing), three branches are still continental (Eurasian) and not more than four Oceanian (and none of them Australian). But most likely all four Oceanian branches are just one (to be described as of now).

        Delete
    9. http://www.australianstamp.com/Coin-web/feature/history/abdream.htm has some Wondjina stories.

      Take out the magical-mythological association with bringing rain, and the Wondjina sound like a foreign group of people. The Wonjina rock art shows up in the Kimberly around 3800 B.P., around the time of this study's Indian gene flow.

      ReplyDelete
      Replies
      1. I would not dare to interpret the Wondjina as "people" much less as foreign people. Spirits appear in dreams, visions.

        Delete
    10. "without knowing IF and how it might relate with South Asia (where mode 4 microlithism was mainstream since c. 38 Ka ago) we can't judge".

      From what I've read over the years no connection with any SE Asian technology. But that doesn't mean it wasn't 'influenced' by something SE Asian.

      "I would not dare to interpret the Wondjina as 'people' much less as foreign people. Spirits appear in dreams, visions".

      Yes, I saw that link while searching for 'Pirri points' and decided exactly the same.

      ReplyDelete
    11. "But neither is found in Aboriginal Australia as such and your 'near Australia' is probably mainland SE Asia (maybe Sundaland but hardly farther to the SE)".

      Extremely likely to be so. But 'Sundaland' is a lot closer to Australia than it is to South Asia.

      "Even if we'd accept the current 7-parts division (which as I say is surely just an approximation to the real thing), three branches are still continental (Eurasian) and not more than four Oceanian (and none of them Australian)".

      I see there has been a re-arrangement at ISOGG although the old nomenclature survives at Wikipedia. The old K1-M174 has dissappeared, so that eliminates the South Asian K. The other three Ks have been reduced by one digit but none can be considered 'continental', although K-P79 is found in Indonesia as well as Melanesia and K-P261 is Balinese. Haven't you consistently maintained that greatest diversity equals origin? So taking the latest ISOGG at face value we therefore have 5/7 haplogroups from Wallacea/New Guinea/Australia. South Asian K has dissappeared so that leaves just two outside the Wallacean region: NO in East asia and P probably in South Asia. That's a lot of diversity in comparison to anywhere else.

      ReplyDelete
      Replies
      1. I won't allow you to hijack the thread. You're warned, Terry.

        "Haven't you consistently maintained that greatest diversity equals origin?"

        Roughly yes. I still do. It's just that the apparent diversity is most unlikely to be real but an artifact of unequal research levels by region.

        Delete
    12. This is fascinating. I've found where K1-M174 has dissapeared to:

      http://www.isogg.org/tree/ISOGG_HapgrpK.html

      Quote:

      "M147 is downstream of M256. Listed 6 August 2012".

      As far as I can discover that makes the old K1 part of MNOPS. So yes, we do have just two MNOPS haplogroups east of Wallace's line and three west of it.

      ReplyDelete
      Replies
      1. M256 does not seem to exist. Is it maybe P256? Typo? Please double-check and, if confirmed, send email to Alice Fairhurst.

        It is a small private lineage but M-P256 has previously been reported in Afghanistan, so it's becoming the less Melanesian of all Melanesian lineages in fact - whatever that means. K* (i.e. excluding a few locally common lineages like R or NO) is found with some frequency in Eurasia but nobody pays it much attention. Maybe you should.

        Delete
      2. The Afghanistan reference: http://forwhattheywereweare.blogspot.com.es/2012/03/y-dna-from-afghanistan.html

        Delete
    13. Karafet et al. 2010

      Southeast Asia (China=383 (Han=165, Yao=60, Miao=58, She=51, Tujia=49), Vietnam=70, Taiwanese Aboriginals=48, Philippines=48, Malaysia=32)

      C-RPS4Y (n=43), including 26 C3-M217 and 17 C-RPS4Y(xC2-M38, C3-M217)
      Variance = 0.676

      O-M175 (n=462)
      Variance = 0.550

      MNOPS*-M526 (n=24), including 22 from the Philippines and 2 from Malaysia
      Variance = 0.381

      C-RPS4Y(xC2-M38, C3-M217) (n=17)
      Variance = 0.179


      Western Indonesia

      MNOPS*-M526 (n=12), including 5 Batak Toba (Sumatra), 5 Borneo, and 2 Java
      Variance = 0.720

      O-M175 (n=804)
      Variance = 0.547

      C-RPS4Y (n=40), including 39 C-RPS4Y(xC2-M38, C3-M217) and 1 C3-M217
      Variance = 0.165

      Eastern Indonesia

      MNOPS*-M526 (n=62), including 24 Flores, 14 Sumba, 9 Alor, 5 Moluccas, 4 Mandar (Sulawesi), 3 Lembata, and 3 Timor
      Variance = 0.698

      C-RPS4Y (n=456)
      Variance = 0.650

      O-M175 (n=166)
      Variance = 0.597

      M-P256 (n=121)
      Variance = 0.587


      Oceania (n=182), including 44 Maewo (Vanuatu), 33 PNG Highland, 24 Tahiti, 16 Micronesia, 15 PNG Coastal, 12 Tonga, 12 Western Samoa, 10 Nasioi, 10 Rapa Nui, 6 American Samoa

      O-M175 (n=31), including 18 O3a2-P201, 5 O3-M122(xO3a2-P201), 4 O1a*-M119(xP203, M110), 2 O-P203, 1 O-M110, and 1 O2b-SRY465(xO2b1-47z)
      Variance = 1.031

      O3-M122 (n=23), including 18 O3a2-P201(xO3a2c1-M134) (5 Tonga, 3 Micronesia, 3 Tahiti, 2 Vanuatu-Maewo, 2 Western Samoa, 1 PNG Coastal, 1 PNG Highland, 1 American Samoa), 3 O3a2c1-M134 (from Tahiti), and 2 O3a-P197(xO3a1c-JST002611, O3a2-P201) (1 Vanuatu-Maewo, 1 Tahiti)
      Variance = 0.981

      MNOPS*-M526 (n=20), including 7 Micronesia, 6 Maewo (Vanuatu), 5 PNG Highland, 1 PNG Coastal, and 1 Tahiti
      Variance = 0.898

      O3a2-P201(xO3a2c1-M134) (n=18)
      Variance = 0.792

      M-P256 (n=46)
      Variance = 0.706

      C-RPS4Y (n=59), including 40 C2a1-P33 (15 Tahiti, 9 Rapa Nui, 8 Western Samoa, 4 American Samoa, 4 Tonga), 5 C2a-M208(xC2a1-P33) (2 PNG Coastal, 1 Vanuatu-Maewo, 1 Tahiti, 1 Western Samoa), 12 C2*-M38(xC2a-M208) (9 Vanuatu-Maewo, 2 PNG Highland, 1 Micronesia), 2 C-RPS4Y(xC2-M38, C3-M217) (1 PNG Coastal, 1 Micronesia)
      Variance = 0.460

      C2-M38 (n=57)
      Variance = 0.377

      Maju, MNOPS*-M526(xM-P256, NO-M214, P-M45, S-M230) is clearly of ancient (original?) presence in Austronesia-Oceania (including Sahul, Wallacea, the Philippines, and marginally Sundaland). The distribution of MNOPS*-M526 and two of its close relations, M-P256 and S-M230, is largely limited to populations of this region, and MNOPS*-M526 is the most diverse Y-DNA haplogroup throughout Indonesia. It would also be the most diverse Y-DNA haplogroup in Oceania if it were not for the strangely high diversity of O-M175 (and O3-M122 in particular) Y-DNA in this pool of Oceanian samples. If one excludes the three O3a2c1-M134 Tahitian individuals, the O3a-P197(xO3a1c-JST002611, O3a2-P201) individual from Tahiti, and the O3a-P197(xO3a1c-JST002611, O3a2-P201) individual from Maewo, Vanuatu on the grounds of being suspect of recent Asian admixture, then we are left with a variance of 0.792 for the 18 O3a2-P201(xO3a2c1-M134) Y-chromosomes that are more likely to be of ancient Austronesian ancestry. This, again, is lower than the variance of the MNOPS*-M526 samples from Oceania.

      ReplyDelete
      Replies
      1. Not sure what you understand from what I said, Ebizur, but all I mean is that:

        1. I think most unlikely that MNOPS migrated from Wallacea or Australasia to continental Eurasia (instead I imagine a continental SE Asian origin, maybe as far as Sundaland but not beyond Wallace Line).

        2. This issue of MNOPS belongs to the early colonization of Eurasia and Australasia c. 60-40 Ka ago and not to any "Neolithic" or otherwise recent flows.

        What you say about C* or O3 does not seem related to Australian Aborigines and their hypothetical relations with India. I don't understand why you even mention them in this topic unless you want to conflate two distinct discussions, what I beg you do not.

        Delete
    14. Western Indonesia
      MNOPS-M526 (n=876)
      Variance = 0.594

      C-RPS4Y (n=40), including 39 C-RPS4Y(xC2-M38, C3-M217) and 1 C3-M217
      Variance = 0.165

      Eastern Indonesia
      MNOPS-M526 (n=484)
      Variance = 0.738

      C-RPS4Y (n=456)
      Variance = 0.650

      Oceania
      MNOPS-M526 (n=121)
      Variance = 0.982

      C-RPS4Y (n=59), including 40 C2a1-P33 (15 Tahiti, 9 Rapa Nui, 8 Western Samoa, 4 American Samoa, 4 Tonga), 5 C2a-M208(xC2a1-P33) (2 PNG Coastal, 1 Vanuatu-Maewo, 1 Tahiti, 1 Western Samoa), 12 C2*-M38(xC2a-M208) (9 Vanuatu-Maewo, 2 PNG Highland, 1 Micronesia), 2 C-RPS4Y(xC2-M38, C3-M217) (1 PNG Coastal, 1 Micronesia)
      Variance = 0.460

      Southeast Asia (China=383 (Han=165, Yao=60, Miao=58, She=51, Tujia=49), Vietnam=70, Taiwanese Aboriginals=48, Philippines=48, Malaysia=32)
      C-RPS4Y (n=43), including 26 C3-M217 and 17 C-RPS4Y(xC2-M38, C3-M217)
      Variance = 0.676

      MNOPS-M526 (n=519)
      Variance = 0.604

      However, C-RPS4Y(xC2-M38, C3-M217) (n=17)
      Variance = 0.179

      As you can see, even limiting our scope of consideration to the major clades MNOPS-M526 and C-RPS4Y, which together account for practically all Y-DNA in all populations of the region in question, MNOPS-M526 is more diverse than C-RPS4Y in all sample sets except the China/Southeast Asia sample set. However, the higher variance of haplogroup C in the China/Southeast Asia sample set is due to the combined presence of two deeply divergent subclades of C-RPS4Y, one being C3-M217 and the other being a not-yet-defined subclade of C-RPS4Y that has rather little internal variance.

      ReplyDelete
      Replies
      1. Is Australasian MNOPS more SRY-diverse than continental Eurasian one? The correlation with C (or rather lack of any correlation at all) is trivial here.

        Delete
    15. Maju wrote,

      "What you say about C* or O3 does not seem related to Australian Aborigines and their hypothetical relations with India. I don't understand why you even mention them in this topic unless you want to conflate two distinct discussions, what I beg you do not."

      I have responded to your suggestions that the apparent diversity of MNOPS in Oceania is "most unlikely to be real but an artifact of unequal research levels by region" and that the K-M147 and M-P256 singletons that have been observed rarely in some samples from South Asia deserve to be given greater weight than they have been in the past in the consideration of the region of origin of MNOPS-M526. You have not presented any evidence to support your suggestions; therefore, they reek of a sort of Eurocentrism (specifically, attempting to "whitewash" your Y-DNA haplogroup R ancestor by placing his tribe as close to Europe as possible).

      Maju wrote,

      "Is Australasian MNOPS more SRY-diverse than continental Eurasian one? The correlation with C (or rather lack of any correlation at all) is trivial here."

      According to the data from Karafet et al. 2010, the answer would be "yes":

      Oceania (n=182)
      MNOPS-M526 (n=121)
      Variance = 0.982

      Eastern Indonesia (n=957)
      MNOPS-M526 (n=484)
      Variance = 0.738

      Southeast Asia (n=581, including China=383 (Han=165, Yao=60, Miao=58, She=51, Tujia=49), Vietnam=70, Taiwanese Aboriginals=48, Philippines=48, Malaysia=32)
      MNOPS-M526 (n=519)
      Variance = 0.604

      Western Indonesia (n=960)
      MNOPS-M526 (n=876)
      Variance = 0.594

      The variance of MNOPS is high in the pool of samples from Oceania, intermediate in the pool of samples from Wallacea, and low in the pools of samples from China, Vietnam, Malaysia, the Philippines, and western Indonesia.

      ReplyDelete
      Replies
      1. "I have responded to your suggestions that the apparent diversity of MNOPS in Oceania is "most unlikely to be real but an artifact of unequal research levels by region""...

        Apparent basal diversity. I use to work with that for mtDNA but in Y-DNA it may be misleading.

        "According to the data from Karafet et al. 2010, the answer would be "yes"".

        OK. sure. What does "Oceania" mean in Karafet's context? Melanesia, Polynesia and maybe some parts of Micronesia. Where does that diversity come from? From the excess diversity of introducing Asian O sublineages most likely.

        Also I asked about Euriasia in general not just East Asia. I know that you do not have the data but that is what we would like to contrast: all MNOPS(xNO,P) of Australia and Near Melanesia (include Wallacea if you wish but NO and P must be excluded because we know for sure they are recent arrivals East and South of Wallace Line) and all MNOPS of continental Eurasia, which is mostly NO and P. That way we can estimate if, prior to Neolithic, there was more MNOPS diversity East or West of Wallace Line.

        The question must be answered properly, carefully, seriously. Otherwise we are onto the paradoxic situation of Brazil being (may well be) more genetically diverse than Portugal... but that's not because Portugal was colonized from Brazil but because Brazilian diversity drank of many different sources. Careful there. When we compare the genetic fraction which must have arrived from Portugal to Brazil and only that, the result must be that Brazil is less diverse than Portugal for that fraction of the genome (because it's impossible that all Portuguese lineages fed Brazilian roots, there's always some level of "bottleneck", of founder effect, even if mild, in any such colonization).

        There's room for noise, confusion and, I guess, manipulation (if one is into that crap) in this but let's try not to sow confusion willingly, OK? I believe that you can perfectly discern the nuances of this population genetics' little problem, can't you?

        Delete
    16. "M256 does not seem to exist. Is it maybe P256? Typo?"

      I suspect it's a typo. Perhaps M147 is downstream of P256, in which case K1 is a subclade of M. Either way K1 seems to be a product of back movement from Wallacea.

      "The variance of MNOPS is high in the pool of samples from Oceania, intermediate in the pool of samples from Wallacea, and low in the pools of samples from China, Vietnam, Malaysia, the Philippines, and western Indonesia".

      Thanks. That supports the conclusion I came to when examining the 2005 McDonald maps. Maju has vehemently opposed my occasional suggestions that such could be the case.

      "You have not presented any evidence to support your suggestions; therefore, they reek of a sort of Eurocentrism (specifically, attempting to 'whitewash' your Y-DNA haplogroup R ancestor by placing his tribe as close to Europe as possible)".

      Maju is not the only one with a desire to place European ancestry as close to Europe as possible (and I seem to remember Maju saying he was J). I see the problem as arising from an inability to see human expansion as other than unidirectional, probably related to primitive beliefs in some Garden of Eden.

      "Roughly yes. I still do. It's just that the apparent diversity is most unlikely to be real but an artifact of unequal research levels by region".

      Still at it.

      "What does 'Oceania' mean in Karafet's context? Melanesia, Polynesia and maybe some parts of Micronesia. Where does that diversity come from? From the excess diversity of introducing Asian O sublineages most likely".

      And there he goes again. Karafet presumably includes all those regions in 'Oceania', but O is not particularly common except in Polynesia and Micronesia. And it seems from Ebizur's first set of data that Karafet has excluded O from MNOPS.

      "I think most unlikely that MNOPS migrated from Wallacea or Australasia to continental Eurasia"

      Why? Is it that Melanesians are too primitive to have done so?

      "include Wallacea if you wish but NO and P must be excluded because we know for sure they are recent arrivals East and South of Wallace Line"

      We don't actually know that. It is true of O and R, but NO and P may have originated near Wallace's Line. And you wrote earlier:

      "but M-P256 has previously been reported in Afghanistan, so it's becoming the less Melanesian of all Melanesian lineages in fact"

      So we know that Y-DNA from Wallacea moved a huge distance.

      "That way we can estimate if, prior to Neolithic, there was more MNOPS diversity East or West of Wallace Line".

      A side issue. My bet would be that MNOPS coalesced either side of Wallace's Line, in Wallacea itself. Any population capable of developing the boating technology required to cross wallace's Line would have a sufficiently superior boating technology to be able to expand back through regions already occupied, exploiting the environment in new ways.

      ReplyDelete
      Replies
      1. "You have not presented any evidence to support your suggestions; therefore, they reek of a sort of Eurocentrism (specifically, attempting to 'whitewash' your Y-DNA haplogroup R ancestor by placing his tribe as close to Europe as possible)".

        I missed this quote. Who wrote that? Whoever did, (s)he's very wrong.

        Regardless of the origins of MNOPS, it is extremely clear for anyone looking at the data that P is from South Asia (where the three basal subclades, P*, R and Q exist with deep roots - Q may be from Iran but not too far) as is probably the case of R and maybe R1 and R1a, Q instead seems to be from West/Central Asia. This is beyond question.

        MNOPS seems to originate farther to the East but a backflow from Oceania at the levels required for the impact of Y-DNA P and NO, which are the largest ones by far in Eurasia, would imply a much clearer genetic signal, like more much more distributed Denisovan genetics and Oceanian mtDNA lineages, etc.

        In general I think that Oceania is a cul-de-sac and whatever goes in does not go out. Why? Because anything arriving to Oceania must have originated (at least before Modern Age) at Eurasia, where technology would be therefore similar or better and where populations numbers were also high. Only very strange circumstances could allow for a back-migration from Oceania in the terms you speculate with. That could be for example the Toba catastrophe... but we do not see the signature anyhwere.

        Mind you that I considered long ago that K (MNOPS was not yet known) might have originated in Melanesia for the very same simplistic reasons you're arguing here but I was quickly talked out of such idea by people of Chinese ethnicity. The logic is the very same I'm arguing here: that Melanesian K/MNOPS is not well studied and will probably some day be shown to be a single joint clade. This is a good hypothesis, not confirmed yet but not proven wrong either.

        Also I think that Y-DNA K/MNOPS and mtDNA R are related somehow, and in Melanesia there's only so much mtDNA R (P actually). Part of my reasoning is that if only one clade of mtDNA R arrived to Melanesia, it's plausible that it was also only one clade of Y-DNA MNOPS which did on the patrilineal side.

        So in general I do expect that all Melanesian MNOPS will converge to a single lineage MS.

        "I seem to remember Maju saying he was J"

        Wrong. I do not know. I never bothered testing: I'm interested in the genetics of peoples not individuals.

        "... but NO and P may have originated near Wallace's Line".

        Not for all I know. P looks like original from the North of South Asia, maybe towards Bengal but not further East, where all presence of P is anecdotal and derived. NO looks original from Southern China or somewhere very close. So the whole thing looks like originating from SE Asia and I have even argued that MNOPS (after factoring the Melanesian aspect) might have originated at Sundaland (what you, Terry, opposed, can't recall why, some bias against Sundaland you have - also apparent in the O1 debate).

        "Any population capable of developing the boating technology required to cross wallace's Line would have a sufficiently superior boating technology to be able to expand"...

        Problem is that we see no evidence whatsoever of any such expansion: neither in the material record, nor in the mtDNA one, nor the autosomal one (this data suggests only India→Australia, never Sahulland→Eurasia has been claimed on autosomal DNA: nothing, nada, rien de rien), nor in the issue of Denisovan genetics. The claim you make regarding K/MNOPS is a burning nail and should not hold the test of time.

        Also it's probable that your mystified Wallaceans actually arrived there on boat most of the way from Africa. Finally it is unclear whether this or that tech should give superiority. Personally I lean for the dog, which was domesticated in SE Asia... what a coincidence (not).

        Delete
    17. You guys can't take just one side of the K problem. K(xL,M,N,O,P,T) was found in many populations of India, for example, at low but non-negligible frequencies: 0-9% (most often: 2-6%). This is probably not K1 (a private lineage) and can or not be within MNOPS or whatever (but looks like not LT) → http://www.krepublishers.com/06-Special%20Volume-Journal/T-Anth-00-Special%20Volumes/T-Anth-SI-03-Anth-Today-Web/Anth-SI-03-31-Trivedi-R/Anth-SI-03-31-Trivedi-R-Tt.pdf.

      Less clearly there are many cases, in Europe for example, in which K* is not properly tested for some minor clade markers, like T (formerly K2) but the doubt remains. For example it was reported at Rootsweb (http://archiver.rootsweb.ancestry.com/th/read/GENEALOGY-DNA/2011-08/1312559170) a case in which Alsacians and Auvernians appear with some K(xL,N1c,P,T). They look single individuals (one each) but they still weight more than 1%.

      Even if we'd be able to have all the relevant data properly collected and filed (a nightmare on its own right), we would not be able to evaluate in any realistic way the basal diversity or general diversity of K, whose rare variants are clearly scattered all around Eurasia and Oceania.

      ReplyDelete
    18. Maju,

      The K(xL,N1c,P,T) individuals from France probably belong to some sort of haplogroup N. N(xN1c) has been reported from Italy, Slovenia, Czechia, etc., so I see no reason why some of it could not have settled in France.

      The finding of K(xL,M,N,O,P,T) Y-DNA in India seems more interesting to me, but I would like to see this finding reproduced by an independent set of researchers.

      ReplyDelete
      Replies
      1. Look, Ebizur: I can't bother searching for days or weeks for all instances of strange K* found around the World. I'm just stating that it does exist (and these examples are not the only ones, just the ones I could most easily find - you are the collector of haplogroups, you should know). It might be N*? Of course: that or K-other, that's how it goes. K1 might also be M, or K3 be S or all them be part of a yet undescribed MS lineage, as I speculate... we just don't know.

        We therefore have to shrug and accept the limitations of this single-angle approach and look at complementary evidence like archaeology or mtDNA or even autosomal DNA for whatever is worth.

        And none of those seem to support a migration out of Australasia. So that's extra reason to doubt yours and Terry's simplistic reading of the apparent diversity of K-MNOPS in the region. Time will tell in any case.

        Delete
    19. "Who wrote that? Whoever did, (s)he's very wrong".

      Ebizur wrote it, and I believe he is correct. Otherwise why would you be so hostile to the idea that MNOPS coalesced somewhere other than in South Asia?

      "In general I think that Oceania is a cul-de-sac and whatever goes in does not go out. Why? Because anything arriving to Oceania must have originated (at least before Modern Age) at Eurasia, where technology would be therefore similar or better"

      False assumption. Do you really believe that Melanesians were incapable of innovation?

      "Problem is that we see no evidence whatsoever of any such expansion"

      Yes we do. The expansion of P and NO. Not to mention mt-DNA R. Using your logic it would be impossible for R to have expanded either, even from South Asia as you so passionately believe.

      "I considered long ago that K (MNOPS was not yet known) might have originated in Melanesia for the very same simplistic reasons you're arguing here but I was quickly talked out of such idea by people of Chinese ethnicity".

      Makes sense. They like to believe they are the superior people in the region. And they certainly don't want to accept that they may descend from the totally inferior Melanesians.

      "The logic is the very same I'm arguing here: that Melanesian K/MNOPS is not well studied and will probably some day be shown to be a single joint clade".

      It has certainly been studied a lot more since I first became interested in the subject. The fact that K1 has been subsumed by some other clade does show more will be discovered but I doubt you are totally correct in your Hypothesis. I'm inclined to the idea that MNOPS may be shown to have branched progressively, just as IJK has been shown to have.

      "So in general I do expect that all Melanesian MNOPS will converge to a single lineage MS".

      Who's jumping to conclusions with no eveidence now?

      "Also it's probable that your mystified Wallaceans actually arrived there on boat most of the way from Africa".

      Based on what evidence?

      "Part of my reasoning is that if only one clade of mtDNA R arrived to Melanesia, it's plausible that it was also only one clade of Y-DNA MNOPS which did on the patrilineal side"

      You're ignoring the fact that 3 clades of M arrived in New Guinea/Melanesia as well, and actually 4 clades within P from Australia. Admittedly P is just one clade of R but other clades of mt-DNA N crossed Wallace's Line to Australia, along with several non-Melanesian M clades. So it is quite possible that several Y-DNA clades reached New Guinea/Melanesia.

      ReplyDelete
      Replies
      1. "Otherwise why would you be so hostile to the idea that MNOPS coalesced somewhere other than in South Asia?"

        SE Asia but this side of Wallace Line. You have not read what I wrote, I'm not going to read what you write. Next time answer read first, answer later, please.

        Delete
      2. "Do you really believe that Melanesians were incapable of innovation?"

        No. Who stands for Melanesian principal in the Oceanic expansion issue? That's not you but me.

        I just say that, everything else equal, consolidated populations hold their ground everywhere. The exceptions (few but some) must be quite obvious anyhow (for example North Africa, where African specific mtDNA is only 25%).

        "And they certainly don't want to accept that they may descend from the totally inferior Melanesians".

        Junk! We all descend from totally cool ancient Homo sapiens: Melanesians and us are close cousins. You are defaming people just because it suits your polemicist strategy.

        It's a mere matter of physics: consolidated populations normally don't let others go through, much less in great numbers.

        ...

        I don't want to argue with you about every single detail. Either you focus or no debate.

        Delete
    20. "K(xL,M,N,O,P,T) was found in many populations of India, for example, at low but non-negligible frequencies: 0-9% (most often: 2-6%). This is probably not K1 (a private lineage) and can or not be within MNOPS or whatever (but looks like not LT)"

      In the link you provided it's not K(xL,M,N,O,P,T). I notice the paper was published before T was defined and the K-M70 is actually T1a. But what the K* represents is still a mystery because they say:

      "none of our studied samples showed the presence of haplogroup K3-M147"

      So K* is not what until very recently we've been calling K1. Not present? Perhaps the presence of K* is a result of minor movement west from Wallacea. It does seem to be most common in the northeast. And an aside from the present subject in the link:

      "The present analysis showed none of the deletions (DYS390.1 or DYS 390.3) associated with Australian or Polynesian C* chromosomes, contesting the
      claims of link between India and Australian aboriginals"

      So there goes another theory. And another aside:

      "Haplogroup D, a monophyletic branch of M168 lineage, defined by an Alu insertion and M174C mutation, on the other hand, was restricted in Bhutia and Tharu tribal groups. Its presence among them is most likely due to gene flow from Tibet, where this haplogroup has earlier been reported".

      So that's goodbye to any movement through South Asia for the haplogroup.

      "Even if we'd be able to have all the relevant data properly collected and filed (a nightmare on its own right), we would not be able to evaluate in any realistic way the basal diversity or general diversity of K, whose rare variants are clearly scattered all around Eurasia and Oceania".

      You are absolutely determined to avoid conceding that MNOPS may have coalesced in Wallacea. What is your problem?

      ReplyDelete
      Replies
      1. "So K* is not what until very recently we've been calling K1".

        Nice that you could contrast that. It may still be K(xMNOPS,LT) and be more related to the origin of K as such than to that of MNOPS [aka K(xLT)] but the matter is that "the asterisk zone" is muddy everywhere, not just in Melanesia.

        If K(xLT,MNOPS), then it would support my idea of K originating in South Asia and migrating eastward only to go back as P or proto-P. This is also supported by the relative geography of F (SA), IJK(SA/WA) and MNOPS (SEA) and by similar patterns of tropical flow and backflow SA→SEA→SA→WEA quite obvious in the mtDNA.

        In any case we are going to the Great Eurasian expansion and not to a recent flow of Neolithic times.

        Delete
      2. "So that's goodbye to any movement through South Asia for the haplogroup".

        Nobody knows how proto-D arrived to SE Asia. SA is just the most parsimonious (shorter, less obstacles) route. But whatever happened the evidence is lost to us by now.

        "You are absolutely determined to avoid conceding that MNOPS may have coalesced in Wallacea. What is your problem?"

        First of all, K/MNOPS is minor in Wallacea. It's an area dominated by C2, a lineage that definitely never made it back to Eurasia in any meaningful numbers. So it'd be a Papuan and not Wallacean origin if anything (or in the Near Melanesian islands NE of Papua or even Filipino Negrito but NOT WALLACEAN).

        Second, we see nothing of what they have in the mtDNA or autosomal sides of the equation that you so dearly avoid discussing.

        Third, it is unlikely by the very geography that sets them apart: Wallace Line. That is a major biogeographical barrier added to the basic tendency of populations to stay put and either absorb or reject newcomers, acting as filters, buffers, and not as channels for migrations.

        But in any case: you should have a holistic theory, including archaeology, autosomal DNA, mtDNA and not just the slippery Y-DNA. If you cannot think about a problem from all angles, I do not need to even consider your "solutions" seriously because they lack the basics: holism.

        Delete
    21. "In any case we are going to the Great Eurasian expansion and not to a recent flow of Neolithic times".

      I agree, but I will make a few more comments.

      "If you cannot think about a problem from all angles, I do not need to even consider your 'solutions' seriously because they lack the basics: holism".

      Maju. You have consistently failed to think about a problem from all angles yourself. You have formulated an hypothesis and then looked at the evidence with the aim of proving that hypothesis. For example:

      "Look, Ebizur: I can't bother searching for days or weeks for all instances of strange K* found around the World. I'm just stating that it does exist"

      As a result of your failure to think about the problem from all angles you concentrate on the very few individuals through Eurasia typed as unresolved K*-M9 while ignoring the far greater number of unresolved K*-M9 individuals in Australia and Melanesia. Consequently you are able to conclude that the haplotype moved from the region of very little presence to the region of much greater presence. You are obliged then to invoke 'founder effect' as an explanation, because that is the only way you can make the evidence fit your hypothesis. You also conveniently ignore the opposite possibility because it doesn't fit your hypothesis.

      "I just say that, everything else equal, consolidated populations hold their ground everywhere".

      I tend to agree, but no human population has spread out like ink through blotting paper. They occupy regions they are most used to at first, leaving regions less familiar to them to be exploited by later arrivals. After all in most single regions of the world haplogroups from several different regions are mixed today.

      "Nobody knows how proto-D arrived to SE Asia. SA is just the most parsimonious (shorter, less obstacles) route. But whatever happened the evidence is lost to us by now".

      It is surely unlikely that D would have been completely wiped out through greater India by any later human expansion. Today many haplogroups survive in regions later over-run by another population. On the other hand Neanderthal Y-DNA is no longer with us.

      "First of all, K/MNOPS is minor in Wallacea".

      It's very prominent either side of it though. Apart from NO and P more plentiful east of it than west.

      "It's [Wallacea] an area dominated by C2, a lineage that definitely never made it back to Eurasia in any meaningful numbers".

      I agree. In fact that alone suggests that Y-DNA K reached Wallacea from the Philippines, not via Nusa Tenggara.

      "Second, we see nothing of what they have in the mtDNA or autosomal sides of the equation that you so dearly avoid discussing".

      I have brought mt-DNA up up many times. You've got a very short memory Maju. I'll discuss it any time.

      ReplyDelete
      Replies
      1. You are intently confusing "looking at the problem from all angles", i.e. to think holistically, on one side and not to be willing to do more work, i.e. to be lazy, tired or whatever you wish. Think vs. work.

        If you do the tiresome job... I'll be glad to think about the data you provide. Thanks in advance.

        ... "no human population has spread out like ink through blotting paper".

        Not quite but almost so. I already offered some maths in a previous discussion in which it was made evident that 100 founders create a population of millions (if conditions allow) in few centuries. It's exponential and, without resistance, in optimal conditions, each woman can have a good number of reproducing offspring, doubling, tripling or more each generation.

        And here we are usually talking of spans of many millennia, so more than enough time.

        "It is surely unlikely"...

        Your usual logic: if it includes "surely", "unlikely", "certainly" and nothing else, it's Terry stating his unfounded opinion.

        "... that D would have been completely wiped out through greater India by any later human expansion".

        It seems obvious then (assuming your "logic" is correct, what happened to Toba for example?) that it migrated through huge mobs.

        "Apart from NO and P"...

        Which make some 99% of the extant MNOPS, at least in numbers, and are two of the most widespread Y-DNA haplogroups on Earth. You are removing 99% of the evidence on an most unlikely trust that the current ISOGG tree is absolutely real (it has been amended once and again and will be in the future).

        I cannot share that faith.

        Delete
    22. Maju wrote,

      "First of all, K/MNOPS is minor in Wallacea. It's an area dominated by C2, a lineage that definitely never made it back to Eurasia in any meaningful numbers. So it'd be a Papuan and not Wallacean origin if anything (or in the Near Melanesian islands NE of Papua or even Filipino Negrito but NOT WALLACEAN)."

      Actually, MNOPS-M526 comprises the majority of Y-DNA in the Karafet team's samples of populations from Wallacea:

      Eastern Indonesia (Karafet et al. 2010)

      145/957 = 15.2% C-RPS4Y(xC2-M38, C3-M217) [115/394 = 29.2% Flores, 21/92 = 22.8% Lembata, 6/54 = 11.1% Mandar (West Sulawesi), 1/9 = 11.1% Timor, 1/28 = 3.6% Alor, 1/350 = 0.3% Sumba, 0/30 Moluccas]
      313/957 = 32.7% C2*-M38(xC2a-M208) [201/350 = 57.4% Sumba, 4/9 = 44.4% Timor, 12/28 = 42.9% Alor, 10/30 = 33.3% Moluccas, 19/92 = 20.7% Lembata, 61/394 = 15.5% Flores, 6/54 = 11.1% Mandar (West Sulawesi)]

      13/957 = 1.4% F-P14(xG-M201, H-M69, I-P19, J-M304, L-M20, MNOPS-M526, K3-P261) [3/54 = 5.6% Mandar from Sulawesi, 4/92 = 4.3% Lembata, 6/394 = 1.5% Flores]
      1/957 = 0.1% J2-M172(xJ2b-M12) [1/54 = 1.9% Mandar from Sulawesi]

      62/957 = 6.5% MNOPS*-M526 [3/9 = 33.3% Timor, 9/28 = 32.1% Alor, 5/30 = 16.7% Moluccas, 4/54 = 7.4% Mandar from Sulawesi, 24/394 = 6.1% Flores, 14/350 = 4.0% Sumba, 3/92 = 3.3% Lembata]
      10/957 = 1.0% M-P256(xM1a-P34, M1b-P87, M2-M353, M3-P118) [10/394 = 2.5% Flores]
      111/957 = 11.6% M1a-P34 [72/394 = 18.3% Flores, 5/30 = 16.7% Moluccas, 4/28 = 14.3% Alor, 11/92 = 12.0% Lembata, 17/350 = 4.9% Sumba, 2/54 = 3.7% Mandar (West Sulawesi), 0/9 Timor]
      7/957 = 0.7% O3*-M122(xO3a-P197) [7/350 = 2.0% Sumba]
      34/957 = 3.6% O3a2-P201(xO3a2b-M7, O3a2c1-M134) [9/54 = 16.7% Mandar (West Sulawesi), 1/9 = 11.1% Timor, 3/30 = 10.0% Moluccas, 18/394 = 4.6% Flores, 3/350 = 0.9% Sumba, 0/28 Alor, 0/92 Lembata]
      30/957 = 3.1% O1a*-M119(xO1a1-P203, O1a2-M110) [2/394 = 0.5% Flores, 7/54 = 13.0% Mandar (West Sulawesi), 19/350 = 5.4% Sumba, 2/92 = 2.2% Lembata, 0/9 Timor, 0/28 Alor, 0/30 Moluccas]
      41/957 = 4.3% O1a1-P203 [36/394 = 9.1% Flores, 1/30 = 3.3% Moluccas, 1/92 = 1.1% Lembata, 3/350 = 0.9% Sumba, 0/9 Timor, 0/28 Alor, 0/54 Mandar (West Sulawesi)]
      27/957 = 2.8% O1a2-M110 [4/54 = 7.4% Mandar (West Sulawesi), 22/350 = 6.3% Sumba, 1/30 = 3.3% Moluccas, 0/9 Timor, 0/28 Alor, 0/92 Lembata, 0/394 Flores]
      26/957 = 2.7% O2a1-M95(xO2a1a-M111) [7/54 = 13.0% Mandar (West Sulawesi), 18/394 = 4.6% Flores, 1/350 = 0.3% Sumba, 0/9 Timor, 0/28 Alor, 0/30 Moluccas, 0/92 Lembata]
      1/957 = 0.1% R2a-M124 [1/54 = 1.9% Mandar (West Sulawesi)]
      1/957 = 0.1% R1a1a-M17 [1/54 = 1.9% Mandar (West Sulawesi)]
      8/957 = 0.8% S*-M230(xS1-M254) [2/92 = 2.2% Lembata, 3/350 = 0.9% Sumba, 3/394 = 0.8% Flores, 0/9 Timor, 0/28 Alor, 0/30 Moluccas, 0/54 Mandar (West Sulawesi)]
      127/957 = 13.3% S1-M254(xS1d-M226) [29/92 = 31.5% Lembata, 59/350 = 16.9% Sumba, 5/30 = 16.7% Moluccas, 29/394 = 7.4% Flores, 2/28 = 7.1% Alor, 3/54 = 5.6% Mandar (West Sulawesi), 0/9 Timor]

      485/957 = 50.7% MNOPS-M526 total [165/957 = 17.2% O-M175 total, 135/957 = 14.1% S-M230 total, 121/957 = 12.6% M-P256 total, 62/957 = 6.5% MNOPS*-M526, 2/957 = 0.2% R-M207 total]

      458/957 = 47.9% C-RPS4Y total

      14/957 = 1.5% F-P14(xG-M201, H-M69, I-P19, L-M20, MNOPS-M526, K3-P261) total

      ReplyDelete
    23. Granted, the difference in total frequency between MNOPS-M526 and C-RPS4Y in Wallacea is insignificant, and the raw frequency of C-RPS4Y is actually somewhat higher than that of MNOPS-M526 when O-M175 is excluded on the basis of being more closely related to N-M231, which is basically found only on continental Eurasia (though it has been found in a few individuals from at least Borneo and Fiji as I recall from my reading of other studies), but the difference is not really significant in either case, and the fact remains that even Australasian-specific subclades of MNOPS-M526 exhibit Y-STR variance that is approximately equal to or greater than that of C-RPS4Y in Australasia:

      Western Indonesia
      MNOPS-M526 (n=876)
      Variance = 0.594

      O-M175 (n=804)
      Variance = 0.547

      C-RPS4Y (n=40), including 39 C-RPS4Y(xC2-M38, C3-M217) and 1 C3-M217
      Variance = 0.165

      Eastern Indonesia
      MNOPS-M526 (n=484)
      Variance = 0.738

      C-RPS4Y (n=456)
      Variance = 0.650

      O-M175 (n=166)
      Variance = 0.597

      M-P34 (n=111)
      Variance = 0.590

      M-P256 (n=121)
      Variance = 0.587

      Oceania (n=182), including 44 Maewo (Vanuatu), 33 PNG Highland, 24 Tahiti, 16 Micronesia, 15 PNG Coastal, 12 Tonga, 12 Western Samoa, 10 Nasioi, 10 Rapa Nui, 6 American Samoa
      O-M175 (n=31), including 18 O3a2-P201, 5 O3-M122(xO3a2-P201), 4 O1a*-M119(xP203, M110), 2 O-P203, 1 O-M110, and 1 O2b-SRY465(xO2b1-47z)
      Variance = 1.031

      MNOPS-M526 (n=121)
      Variance = 0.982

      M-P256 (n=46)
      Variance = 0.706

      C-RPS4Y (n=59), including 40 C2a1-P33 (15 Tahiti, 9 Rapa Nui, 8 Western Samoa, 4 American Samoa, 4 Tonga), 5 C2a-M208(xC2a1-P33) (2 PNG Coastal, 1 Vanuatu-Maewo, 1 Tahiti, 1 Western Samoa), 12 C2*-M38(xC2a-M208) (9 Vanuatu-Maewo, 2 PNG Highland, 1 Micronesia), 2 C-RPS4Y(xC2-M38, C3-M217) (1 PNG Coastal, 1 Micronesia)
      Variance = 0.460

      S-M230 (n=23)
      Variance = 0.428

      If the variance of M-P256 is not significantly less than the variance of C-RPS4Y (which includes C2-M38) in Australasia, then there is no way to support the hypothesis of an earlier expansion of C-RPS4Y into Australasia followed by a later expansion of the MNOPS-M526 group that would become the ancestor of M-P256. Extant C-RPS4Y and M-P256 Y-chromosomes in Australasia reflect expansions that have bequeathed approximately the same amount of Y-STR variance to modern populations in the region, so they should be assumed to have begun their expansion in the region at approximately the same time without any evidence to the contrary (à la O-M175).

      ReplyDelete
    24. @Ebizur: indeed you are right and I stand corrected on my previous claim re. K in Wallacea. Still there's a lot of other reasons to question the origin of MNOPS [K(xLT)] beyond Wallace line, for example the lack of 'Denisovan' admixture in mainland Eurasia (but very marked in non-admixed Aboriginal Australasians) or the much more clear to gauge diversity of mtDNA R (or whatever other mtDNA-focused viewpoint you wish to adopt), never mind the archaeological issues.

      As for the variance issue you mention, I find not very useful to compare variance between different haplogroups. Personally I think it's plausible that C arrived first or that C and MNOPS arrived together, having diverse founder effects depending on region. The main reason to imagine the C first model is that MNOPS in Eastern Greater Eurasia (i.e. East Asia plus Near Oceania) appears overall as a wedge between the two main C zones (NE Asia and Australasia-cum-Wallacea) but I reckon that there are other more randomized possibilities to end up with such distribution - so I leave this issue somewhat open.

      Still, looking again at the mtDNA angle, Y-DNA MNOPS in Eurasia can be somewhat strongly correlated with mtDNA R (R11-F and B especially in East Asia but many other clades in SE Asia and Australasia, and yet many others in South Asia and West Eurasia) and this may support the two successive waves model I am tempted by. They'd be both very early in time, maybe one pre-Toba and the other post-Toba for example - or, if a bit later, one previous to the domestication of dogs in SEA and the other after it (the dogs would give the competitive edge in this case).

      ReplyDelete
    25. "Not quite but almost so.[no human population has spread out like ink through blotting paper]"

      Incorrect. Even in Polynesia people occupied the coast first and only moved further inland as it became necessary through population growth. And even today in Africa various groups tend to occupy similar ecological niches while other groups occupy different ones.

      "100 founders create a population of millions (if conditions allow) in few centuries".

      Yes, but they generally don't. Habitat constraints kick in. And that 'few centuries' allows all sorts of other developments.

      "Which make some 99% of the extant MNOPS, at least in numbers, and are two of the most widespread Y-DNA haplogroups on Earth".

      So now you're claiming what you used to accuse me of doing. Particular haplogroups coalesced everywhere at once. What have the bare 'numbers' in each haplogroup got to do with 'origin'? According to your logic Y-DNA R probably coalesced in North America.

      "the current ISOGG tree is absolutely real (it has been amended once and again and will be in the future)".

      The fundmanental tree has not been substantially altered for years now. Mostly just new subclades have been added. The last major alteration was the discovery of MNOPS. I agree that K may have some phylogeny yet to be resolved.

      "Actually, MNOPS-M526 comprises the majority of Y-DNA in the Karafet team's samples of populations from Wallacea"

      Thanks yet again for your most informative collection of data.

      "MNOPS in Eastern Greater Eurasia (i.e. East Asia plus Near Oceania) appears overall as a wedge between the two main C zones (NE Asia and Australasia-cum-Wallacea)"

      Not entirely true. I hope Ebizur can provide data but I'm sure undifferentiated C*-RPS4Y is reasonably common around the South China Sea.

      "for example the lack of 'Denisovan' admixture in mainland Eurasia (but very marked in non-admixed Aboriginal Australasians)"

      I think the whole matter of Denisova admixture is far from clear. I'm sure that further research will reveal much.

      "Still, looking again at the mtDNA angle, Y-DNA MNOPS in Eurasia can be somewhat strongly correlated with mtDNA R (R11-F and B especially in East Asia"

      Especially in SE Asia. I've been trying to get you to see that for years. In fact check the distribution of mt-DNA in the region: No basal M or N in Wallacea or Southern Indonesia. Just downstream haplogroups. On the other hand we find R23 straddling Wallace's Line in Bali and Sumba, R22 in the Lesser Sunda Islands and R14 in the Lesser Sunda Islands (as well as in New Guinea and the Nicobars, the latter definitely a later arrival). We also have P in Australia (along with N-derived N13, N14, S and O), R12'21 in Australia and the Malay Negritos). And R11-F and B are fairly close by although basal M and N haplogroups share the region with them. To me is is more than 'just possible' that mt-DNA R and Y-DNA MNOPS coalesced and expanded from very much the same region. The second of the waves you than support would be from east to west, through South Asia.

      ReplyDelete
      Replies
      1. "Habitat constraints kick in".

        That may apply for really cold regions or arid ones but hardly for tropical ones: people take what they need and I just don't believe in your arbitrary micro-niche differences between coast and inland or savanna and jungle, differences do exist but not at the level of making some areas effectively impassable, that's an absurd belief.

        "What have the bare 'numbers' in each haplogroup got to do with 'origin'?"

        Nothing. Just that 99% of numbers may hide much diversity. That was my point.

        You persuade me with the mtDNA, the Denisovan DNA and the archaeology (or at least two of these three) and I'll believe in your Wallacean hypothesis. Meanwhile not: it's just an artifact caused by irregular research levels.

        "I'm sure undifferentiated C*-RPS4Y is reasonably common around the South China Sea".

        Undifferentiated? You mean unclassified subclades. There's nothing undifferentiated about "asterisk" clades but our vision (or lack of it thereof).

        Whatever the case, when I described my impression on C being at the periphery of MNOPS that's obviously based on numbers, so a few rear-guard survivors won't change that.

        But well, enough. Please tell me how all this correlate with data that, at best, can only be interpreted as SA→AA direction and never vice versa.

        Delete
    26. "You persuade me with the mtDNA, the Denisovan DNA and the archaeology (or at least two of these three) and I'll believe in your Wallacean hypothesis. Meanwhile not: it's just an artifact caused by irregular research levels".

      Ebizur has just presented a very convincing piece of research yet you're determined not to accept the evidence he provided. I would have thought this settled it:

      "62/957 = 6.5% MNOPS*-M526 [3/9 = 33.3% Timor, 9/28 = 32.1% Alor, 5/30 = 16.7% Moluccas, 4/54 = 7.4% Mandar from Sulawesi, 24/394 = 6.1% Flores, 14/350 = 4.0% Sumba, 3/92 = 3.3% Lembata]"

      So there you have it. That MNOPS is not M, NO, P or S). How do you explain it? To me it tells us that the islands forming a triangle with the sides: Sumba, Timor and Alor/Lembata/Flores have had a pivotal role in the origin of modern human Y-DNA. This triangle lies between Sunda and Sahul.

      "looking again at the mtDNA angle, Y-DNA MNOPS in Eurasia can be somewhat strongly correlated with mtDNA R (R11-F and B especially in East Asia but many other clades in SE Asia and Australasia, and yet many others in South Asia and West Eurasia)"

      Yes. And the haplogroup evidence can easily be interpreted as indicating that mt-DNA R also originated somewhere very near that island triangle. But you absolutely refuse to even consider the possibility, for some reason that completely escapes me. Racism?

      "Undifferentiated? You mean unclassified subclades".

      OK, split hairs if you must. 'Unclassified subclades' then. But those unclassified subclades are not C1, C2, C3, C4, C5, C6 or the newly discovered(?) C7. And they are far beterr represented around the South China Sea than they are in South Asia. Once more, as in the example of K*, you are able to conclude that the haplotype moved from the region of very little presence to the region of much greater presence. You are obliged then to invoke 'founder effect' as an explanation, because that is the only way you can make the evidence fit your hypothesis. You also conveniently ignore the opposite possibility because it doesn't fit your hypothesis.

      ReplyDelete
      Replies
      1. I'm asking for mtDNA, autosomal DNA and archaeological data and you come again with Y-DNA. Please!

        Look: MNOPS was discovered first as NOP, i.e. studying major continental clades. Only later they realized that all the K in Oceania was inside it also. I'm pretty sure that, as soon as someone finds the marker (that's the hardest task), it will be shown that all Oceanian MNOPS is a single subclade of MNOPS (would be MS) all this discussion will be shown pointless (but you won't admit, just move one without comment).

        Why I feel so certain about this? Because of mtDNA, etc. Because I look at the matter from all angles and not just Y-DNA. So please spare me your Y-DNA-only conjectures. Thanks.

        "But those unclassified subclades are not C1, C2, C3, C4, C5, C6 or the newly discovered(?) C7".

        Not necessarily so. It depends what they have tested for in each study. But anyhow, we agree that C probably originated in SE Asia, just that I tend to think of the mainland (coast?, I can't say) and you gamble with something a bit more ambiguous.

        "... they are far beterr represented around the South China Sea than they are in South Asia".

        We already saw the other day that there's also quite a bit of C* in India. This allows for some uncertainty and we must wait for more data or just shrug and write "Tropical Asia" or whatever.

        "You are obliged then to invoke 'founder effect' as an explanation"...

        Founder effect is THE EXPLANATION in all cases I can think of everywhere on Earth: someone had to be the founder (several people of course but maybe related or whatever) and they invariably have A MAJOR EFFECT on what came after them.

        It's not "invoking": it's how things are in fact in population genetics. As soon as you have a punctual migration, you have a founder effect or several.

        Delete
    27. What about this study:

      http://www.cell.com/current-biology/retrieve/pii/S0960982202007893

      Its from 2002, but the conclusion is the same. When we add the fact with mitochondrial M42 and also the lingvistic support (the lingvists claim, that the similarity between south India and Pana-nyungan consonants system can not be juts a coincidence), I am not skeptic about the migration despite I was when I read the mentioned 2002 paper.

      ReplyDelete
      Replies
      1. This issue arose in one of the many recent hot discussions on Oceanian genetics, maybe in this one.

        The case is that the C* found in that paper was resolved in another paper as C4 and quite distinct in the haplotype structure (more markers usued probably). I'll see if I can find the relevant paper later.

        Delete
      2. Hammer 2005, fig. 4-d. C4 had not yet been described but C1, C2 and C3 were. Per the info provided at fig. 2 C* in fig. 4-d is C(xC1,C2,C3), but C1 is drawn as well to the left of the | mark (per fig. 4 legend, because it's a Japan-focused paper).

        So we see now, three years after Redd 2002 (surely with more SRY markers being used) that the Oceanian C* (C4 from Australia, then still undescribed) hangs from SE Asian C* and not anymore from the Indian C* (part of which must be C5, then also not yet described). Per the "materials and methods" section, some of their sample are from Redd 2002b (your link).

        So we can safely discard the appearance of agglomeration of Asutralian and Indian C* as per Redd 2002 as an artifact caused by some limitations of the data. Also Australian C* (in both papers) is now described as C4, a distinct haplogroup within C.

        Delete
    28. 'Founder effect is THE EXPLANATION in all cases I can think of everywhere on Earth: someone had to be the founder"

      Of course. But you use it all the time, whenever the evidence cannot be manipulated to fit your pre-existing belief, not just where we're considering the first inhabitants. And often in cases where it is by no means necessary to consider 'founder effect' as being the explanation at all.

      "I'm pretty sure that, as soon as someone finds the marker (that's the hardest task), it will be shown that all Oceanian MNOPS is a single subclade of MNOPS (would be MS)"

      I'm fairly sure it won't be. You 'desire' it to be so to make the evidence fit your belief. And it seems now that what we would at present call K1-P60, K2-P79, K3-P261 and K*-M9 are all part of what you refer to as 'MNOPS'. That's a whole lot more South Wallacean/Melanesian Y-DNA haplogroups.

      "Because of mtDNA, etc. Because I look at the matter from all angles and not just Y-DNA".

      Why don't you just take the time to check out the distribution of mt-DNA R?

      "We already saw the other day that there's also quite a bit of C* in India".

      Did we really? Isn't most Indian Y-DNA C in fact C5-M356? As you say, 'Indian C* (part of which must be C5, then also not yet described)'.

      ReplyDelete
      Replies
      1. "But you use it all the time, whenever the evidence cannot be manipulated to fit your pre-existing belief"...

        I think that's a diversion you use instead of discussing the facts for what they are. If you go to each particular case, we can discuss them in detail, but if you just make generalizations... then nothing good can come out, just imprecission and confusion.

        "I'm fairly sure it won't be".

        My logic is based on several facts:

        1. NO and MNOPS (initially NOP) markers have been already found, substantially altering the K tree.
        2. Y-DNA ultimately is always branched in pairs (and MNOPS does not follow yet this).
        3. mtDNA (nor autosomal one) suggest an "out of Wallacea/Australasia" process at all.

        "Why don't you just take the time to check out the distribution of mt-DNA R?"

        What's the "problem" with mtDNA R? Be specific, please: doing otherwise is just . So far it looks as originating in South Asia, where basal diversity is maximal (alternatively you can argue for an Indochina origin, highly dubious, but not beyond Wallace Line).

        "Did we really?"

        Yes we did.

        "Isn't most Indian Y-DNA C in fact C5-M356?"

        Many tribes have C* as dominant. Browse the blog back please.

        Delete
    29. "1. NO and MNOPS (initially NOP) markers have been already found, substantially altering the K tree".

      The information Ebizur supplied regarding MNOPS in Wallacea indicates that all the K in that region contains the M525 mutation that defines MNOPS. The K could be K1, K2 or, unlikley, K3. So at least most 'K' is MNOPS.

      "2. Y-DNA ultimately is always branched in pairs (and MNOPS does not follow yet this)".

      I agree that finer detail may reveal NOP branched off first but we don't yet 'know' that, so in the meantime we should use the data as we have it instead of making things up.

      "3. mtDNA (nor autosomal one) suggest an "out of Wallacea/Australasia" process at all".

      I firmly disagree. The phylogeny of mt-DNA R certainly does fit with an out of Wallacea/Australasia' process. It's just that you refuse to see it.

      "Many tribes have C* as dominant. Browse the blog back please".

      I'm sorry. You'll have to lead me to it again. As far as I'm aware most Indian 'C*' has now gone, replaced by C5.

      ReplyDelete
      Replies
      1. 1. Not all K. At least LT is not. Anyhow, it is not really relevant for my hypothesis: all Oceanian K/MNOPS would be part of a local subclade (would be "MS" for lack of a better name).

        2. "I agree that finer detail may reveal NOP branched off first"...

        Not at all what I meant. Than NOP was discovered first (and only then found to be actually the larger MNOPS) only speaks of the greater level of research dedicated to these larger continental clades. It says nothing of which branched out first. In my hypothesis it'd be (maybe) MS (including all Oceanian K).

        "... so in the meantime we should use the data as we have it instead of making things up".

        You can't restrict yourself to Y-DNA and the rest of the data does not suggest AT ALL that there was any "out of Wallacea" migration. Would it be otherwise, I would have agreed with you long ago - but nope.

        3. "The phylogeny of mt-DNA R certainly does fit with an out of Wallacea/Australasia' process".

        Not at all: only one (out of 16) sublineages is from Oceania. Maybe there's some other in Wallacea (can't recall right now) but the vast majority is from continental Asia, the plurality from South Asia.

        "I'm sorry. You'll have to lead me to it again".

        I have less time than I used to have but we had a lenghty debate back then, so you should know what I'm talking about. Right now I'm out of home and therefore do not have my bookmarks at hand, sry.

        Delete
    30. "The information Ebizur supplied regarding MNOPS in Wallacea indicates that all the K in that region contains the M525 mutation that defines MNOPS. The K could be K1, K2 or, unlikley, K3. So at least most 'K' is MNOPS".

      The map in this paper shows the K in Wallacea as simply 'K'. We can now be sure it is MNOPS:

      http://www.sciencedirect.com/science/article/pii/S0960982209020673

      ReplyDelete
      Replies
      1. I do not question that at all: what we don't know is how it is organized phylogenetically downstream of MNOPS, of the M525 mutation. My hypothesis (to be tested by whatever research when it comes) is that all Oceanian MNOPS is part of a distinct subclade of MNOPS, let's call it MS or MNOPS1 or whatever you wish.

        In other words that K(xLT,xNO,xP) is a single subclade, at least in Oceania. But we will have to wait till someone researches it: I simply lack the means.

        Delete
    31. "I have less time than I used to have but we had a lenghty debate back then, so you should know what I'm talking about. [Many tribes have C* as dominant]".

      I think I remember it now. Our main argument centred on whether Munda had come into India from further east. The tribals with C* in them were all Munda- or Austro-Asiatic-speaking. The C* had come in from further east with O2a1a.

      "My hypothesis (to be tested by whatever research when it comes) is that all Oceanian MNOPS is part of a distinct subclade of MNOPS"

      It could be. But the only reason you're suggesting that is because that is what you want it to be. How would it be if Australian K*-M9 is shown to have branched off first?

      "1. Not all K. At least LT is not"

      Certainly LT is not part of MNOPS. But The K*-M9 in Auistralia is MNOPS. As far as I'm aware the K*-M9 near the LT memebrs has not been phylogenetically placed. It may belong to some pre-KMNOPS haplogroup and related to LT or it may not.

      "Than NOP was discovered first (and only then found to be actually the larger MNOPS) only speaks of the greater level of research dedicated to these larger continental clades".

      Yes. And that makes it unlikely there will be any major changes in the near future. I await developments of course.

      "only one (out of 16) sublineages is from Oceania [phylogeny of mt-DNA R]".

      Who said anthying about 'Oceania'?

      "Maybe there's some other in Wallacea (can't recall right now)"

      Open your eyes, Maju.

      ReplyDelete
      Replies
      1. Re. Y-DNA C* in India:

        It was as easy as searching just a bit into the 'South Asia' label in this blog: THIS ENTRY!!!.

        I wrote then:

        Haplogroup C

        As the authors note, 90% (66/74) of all the C-M130 samples belong to C5 (M356), while the rest (8/74) tested negative for both C5 and C3 (M217), so I guess we are here before at leas one other subhaplogroup of C (because the likelihood of being Japanese C1 or Australasian C2 or C4 is practically zero).

        The eight C* individuals are scattered (table S1) among several groups (all of which also display C5, as well as F*) but notably concentrated among the Piramalai Kallar (4/5 within C), which are a DLF group.

        Besides C*, which may well be a remnant of either the early Eurasian expansion or of the first backflows from SE Asia (a likely not-so-likely candidate for the origin of macro-haplogroup C), the very notable presence of C5 among tribals and some farmers may well indicate that the origin of C5 is in South Asia, even if the clade also has some presence in Central and West Asia.

        Haplogroup C has a high variance in this study (0.80), greatest among DLF (0.89) and HTF (0.81)
        .

        That entry also includes, in the appendix, the two aforementioned older C trees of Hammer 2006 and Redd 2002. You may therefore want to add to your list of "favorites" or "bookmarks" for later reference on Y-DNA C.

        Delete
      2. "How would it be if Australian K*-M9 is shown to have branched off first?"

        It would be very interesting to know about it. Sincerely I am as interested as you are or more about the real, yet to be properly described, phylogeny of K/MNOPS in Oceania. It does not exist yet, so I have to base my inferences on other data (archaeology, autosomal DNA and mtDNA) and none of these suggest even slightly a back-migration into mainland Asia from that part of the World.

        "Who said anthying about 'Oceania'?"

        Oceania, Wallacea... same thing for Prehistory purposes. Wallacea is actually intermediate between Oceania and Asia and is at least occasionally included in the Oceanian region of Melanesia, so you should have no qualms about I including it in Oceania.


        Delete
    32. OK. I'll do it for you. I know you've seen this:

      http://dispatchesfromturtleisland.blogspot.co.nz/2013/01/munda-as-intrusive-to-india.html

      Perhaps not my comment there:

      "interestingly the mt-DNA R-derived haplogroups R7 and R8 are both said to be especially associated with Munda-speaking populations".

      That supports the analysis of mt-DNA R I made at Wikimedia:

      http://ourorigins.wikia.com/wiki/Mt_R_east_to_west

      Where I wrote:

      "So here we have the pattern of R's expansion through India revealed by the geographic distribution of the basal haplogroups: Basal R entered Northeast India from somewhere further east. R7 formed somewhere round that entry point and, after a period of 8 mutations, moved south through eastern India. Basal R did the same. R8 coalesced from basal R in Orissa, and carried on, after 5 mutations, as far south as Sri Lanka and up the Godavari River".

      "only one (out of 16) sublineages is from Oceania"

      I list 18 R lineages in the above link. But we now have the resources to let us see which part of India has the greatest basal R diversity. And compare it to basal diversity in regions of comparable extent. After all we would expect greater diversity to be at least partly a function of greater extent. We can now be reasonably sure that R7 and R8 coalesced east and west of the Ganges mouth respectively and were carried west to their present geographic margin by the expansion of Munda-speaking people. So they are 'East Indian' haplogroups. R6 is also present along much of the east coast of India but is spread along the Ganges River and up it as far as Kashmir. That's three haplogroups coalescing in East India. That leaves the three other R mt-DNAs as 'West India': R30, R5 and R31. Outscored by Southwest Asia: U, R2'JT, R1, R0 and R3. I agree that region may be larger than either of the Indian halves.

      But including haplogroups to the east of India in Southeast Asia and Australia the score is:

      East India: 3. R7, R8 and R9.
      West India: 3. R5, R6 and R30.
      Southwest Asia: 5. R0, R1, R3, U and R2'JT.
      Australia: 2. P and shares R12'21 with Malay Negritos.
      Lesser Sunda Islands: 3. R22, R23 and R14 (the last in New Guinea and the Nicobar Islands.
      Throughout island SE Asia and into the mainland: 2. R11'B and R9.

      So what boundaries do you wish to use when considering 'basal diversity'?

      ReplyDelete
      Replies
      1. It is very difficult for me to follow your logic here. All I can say is that while I do appreciate a relatively notable duality between North and South in South Asia, possibly caused by an arid Deccan in the Pleistocene (and/or arguably Neolithic inflows later on), I don't really perceive a clearly cut West-East divide in either subregion, at most some gradation.

        Besides your division is totally unreal and arbitrary re R subclades.

        Hence I have no idea why you would split "India" in West and East. My count is as follows (reference):

        South Asia: 6 (R5, R6, R7, R8, R30, R31)*
        West Asia: 2 (R0, R3)
        Both: 3 (R1, R2'JT, U)
        [total West: 11]

        Mainland SE Asia (incl. South China): 2 (R9/F, R11'24'B)
        NE Asia: nothing specific but also found branches of the two above
        Island SE Asia (incl. Malaysia and Wallacea): 2 (R22, R23)
        Oceania: 1 (P)
        ISEA & Oceania: 2 (R12'21, R14)
        [total East: 7]

        Total clades 18. Estimated centroid somewhere in South Asia, probably in the course of the Narmada or Ganges (my best hunch is lower Ganges).

        *Notice that most of these South Asian lineages are found East and West equally but rather to the North - details from the wiki:

        → R5 Across Central India except SE Coast. Especially SW coast and Sri Lanka. Present only in the most westerly of the Madhya Pradesh tribals: the Bhil.
        → R6 Uttar Pradesh, Kashmir and SE coast. Also Pakistan. Present in all three Madhya Pradesh tribals.
        → R7 Especially in NE India and Andhra Pradesh. Frequent in Munda speakers. Not present in the most westerly Madhya Pradesh tribals, the Bhil, but present in the other two.
        → R8 Present in Andhra Pradesh, Orissa, Maharashtra. Also especially in South India and Sri Lanka. Perhaps originated in Orissa. Not recorded in the Sahariya, the northernmost of the Madhya Pradesh tribals, but present in the other two.
        → R30 Not listed in Madhya Pradesh tribals but widespread in South Asia, especially NW and Central. Uttar Pradesh, Punjab, Rajasthan, Gujarat, Maharashtra.
        → R31 Of the Madhya Pradesh tribals recorded only in the Bhil, the most westerly. Present in Rajasthan, Southern India and Sri Lanka.

        Most simply look too widespread to be considered as part of any arbitrary subregion of the South Asian subcontinent.

        Delete
    33. "As the authors note, 90% (66/74) of all the C-M130 samples belong to C5 (M356), while the rest (8/74) tested negative for both C5 and C3 (M217)"

      So C* is common' in India, you say. Basically confined to the Piramalai Kallar, who could have come from anywhere into India.

      "the real, yet to be properly described, phylogeny of K/MNOPS in Oceania. It does not exist yet"

      Perhaps the Oceania set has yet to be defined but K as a whole is particularly interesting. Haplogroups have peeled off it all the way from just out of Africa to Melanesia.

      "Oceania, Wallacea... same thing for Prehistory purposes".

      What are you on? There is no overlap between the two regions. Wallacea is quite a small region within modern Indonesia. Oceania is the region beyong New Guinea. Much of Oceania was uninhabited until long after people had reached Wallacea.

      "Wallacea is actually intermediate between Oceania and Asia and is at least occasionally included in the Oceanian region of Melanesia"

      I have never ever heard of 'Melanesia' being included in 'Wallacea'. Where did you get that idea from? And we're talking specifically 'Southern Wallacea' in relation to mt-DNA R. That is the thin line of islands that stretch from the Malay Peninsula to northwest Australia centred on the little triangle formed by Sumba, Timor and Flores/Alor. In area the whole lot is far smaller than the whole of India, even if you include all that water. In that thread of islands we have 5 basal R haplogroups: P at one end, R12'21 at either end and R14, R22 and R23 in the middle. Besides which we have R9 and R24'B extremely close by.

      "It is very difficult for me to follow your logic here".

      Is that on purpose?

      "I don't really perceive a clearly cut West-East divide in either subregion, at most some gradation".

      Look again then. And the division you offer is completely irrational. You combine south and west Asia (presumably to increase your numbers) and then divide the Far East into four separate regions (presumably to limit the numbers). That is hardly a logical way to go about things.

      "Most simply look too widespread to be considered as part of any arbitrary subregion of the South Asian subcontinent".

      Because you are reluctant to divide India into separate regions because it would expose your faulty logic. I agree that R6 is particularly widespread but R7 and R8 definitely have an eastern India origin. R5 and R31 are almost certainly not of eastern Indian origin. That leaves R30. Widespread but especially in central India and the northwest. Yet you can say, 'Most simply look too widespread to be considered as part of any arbitrary subregion of the South Asian subcontinent'. Open your eyes.

      ReplyDelete
    34. Sorry. I've just noticed this comment:

      "*Notice that most of these South Asian lineages are found East and West equally but rather to the North"

      Doesn't that absolutely scream, 'migration via the Ganges'? Exactly as I suggested at Wikimedia.

      ReplyDelete
    35. No, C* is not confined to just that population at all. You are not even looking at the data. Do things right, please.

      K has not "peeled off" either. It split in two clades, one centered in Pakistan and the other maybe in SE Asia. Etc.

      There's no reason to think that Wallacea was populated long befores Oceania (Sahul). Oceania includes Australia, etc., for your information.

      It is notMelanesia the one included, sometimes, in Wallacea but vice versa.

      I did not combine WEA and SA to increase anything but to emphasize the perception of geographic distribution and, therefore, the logic behind the centroid: almost double number of basal subclades West than East of the Ganges Delta.

      ReplyDelete
    36. "C* is not confined to just that population at all. You are not even looking at the data".

      Maju. The data says just 8 of 72 Y-DNA C's were C*. And those 8 were 'concentrated among the Piramalai Kallar'. That doesn't leave very many for anyone else. And all of the C* were 'scattered (table S1) among several groups (all of which also display C5, as well as F*)'. I note that Ebizur pointed out on your earlier post that the C* were all associated with aristocratic groups and could have arrived from the northeast with the Neolithic. It which case they would be related to C% without actually having C5's defining mutation.

      "K has not 'peeled off' either. It split in two clades, one centered in Pakistan and the other maybe in SE Asia. Etc"

      OK. So you're excluding IJ from K. IJ formed in SW Asia when IJK split in two. The LT formed in Pakistan when K split in two. The, as far as we can tell from out current state of knowledge, K(xLT) split all to pieces in SE Asia. To me it is simple to comprehend.

      "There's no reason to think that Wallacea was populated long befores Oceania (Sahul)".

      I could go along to some extent with the northern Solomon Islands portion of Oceania, but it's doubtful it was settled until some time after people had reached Australia. The remainder of Oceania, beyond the southern Solomon Islands was definitely not settled until Lapita/Austronesian times, some three or four thousand years ago.

      "I did not combine WEA and SA to increase anything but to emphasize the perception of geographic distribution"

      And then you ignored the geographic distribution in India. Why?

      "almost double number of basal subclades West than East of the Ganges Delta".

      And the Ganges Delta is a significant boundary in the distribution of mt-DNA R because ...?

      ReplyDelete
      Replies
      1. ... "the C* were all associated with aristocratic groups and could have arrived from the northeast with the Neolithic".

        No!!!

        You are already slaving me again (do your damn data-mining work instead of ranting so much!!!): C* in Tamil Nadu is found:
        → Paliyan n=1 (foragers, also lots of C5)
        → Thoda n=1 (foragers)
        → Vanniyar NTN n=1 (agriculturalists, also lots of C5)
        → Cape Nadar n=1 (agriculturalists)
        → Piramallai Kallar n=4 (agriculturalists)

        None of them is "aristocratic" at all: the foragers belong all to the most genuinely local "class" (HTF, froagers of Tamil or Malayalam language) and the agriculturalists to the oldest and more locally rooted of all them (DLF - dry land farmers). The "aristocratic" Brahmin-related irrigation farmers and Artisan-Warrior classes are lacking this paragroup and not rich in C5 either (although some specific groups do have this one).

        Delete
      2. "So you're excluding IJ from K".

        Per ISOGG. Indeed.

        If you meant IJK, say so. You could also say F for all I know as well.

        "And then you ignored the geographic distribution in India. Why?"

        Because I don't agree with it. I think it's arbitrary in most cases.

        Even if it would be half-correct, it would only help to more precisely positioning the phylogenetic centroid of R, which I already said should be in the Lower Ganges.

        Visualization: as I described the matter the scales have 11 and 7 weight units, so it leans to the West. To get them balanced, we have to move two weights to the "East" platter. So just pick your favorite most eastern-leaning South Asian R basal subclades and voilá: 9:9. Where is the centroid then, the edge of the rebalanced scales? In Central-East India (same result as I suggested above more or less).

        Now, Terry, you will have to play thimblerig again but no matter how much you move the cups, the ball is stuck to that geography.

        But I beg you to memorize this well and don't bother me again with this issue unless NEW DATA changes things.

        "And the Ganges Delta is a significant boundary in the distribution of mt-DNA R because ...?"

        The Ganges Delta (its Easternmost edge in fact) is the significant geographic and ethnic boundary between South and SE Asia. Look at any map: Bengal (Bangladesh) borders Burma/Myanmar.

        Why would you even ask this?

        Delete
    37. "C* in Tamil Nadu is found:
      → Paliyan n=1 (foragers, also lots of C5)
      → Thoda n=1 (foragers)
      → Vanniyar NTN n=1 (agriculturalists, also lots of C5)
      → Cape Nadar n=1 (agriculturalists)"

      And that proves C* is widespread and common in India? Come on.

      "If you meant IJK, say so".

      Semantics, yet again. IJK is part of K. IJK is the name of the K branch that first split from F. It later gave rise to LT and still later to MNOPS. Simple?

      "it would only help to more precisely positioning the phylogenetic centroid of R, which I already said should be in the Lower Ganges".

      So we're half in agreement. You believe R originated in the Lower Ganges, I believe it entered South Asia from the Lower Ganges. The fact that R appears to be spread along the Ganges suggests the region was unexploited before R either colesced there or arrived. According to your usual logic that should have been impossible. The first arrivals should have dominated.

      "as I described the matter the scales have 11 and 7 weight units, so it leans to the West. To get them balanced"

      Why do we have to get them 'balanced'? Simply to fit your belief.

      "unless NEW DATA changes things".

      We don't need 'new data'. It should be sufficient for you to properly examine the old data.

      "The Ganges Delta (its Easternmost edge in fact) is the significant geographic and ethnic boundary between South and SE Asia".

      But not for mt-DNA R. Check it out.

      ReplyDelete
    38. [sarcasm]
      It demonstrates whatever you wish.

      Confusing IJK and K is mere "semantics", sure, and not another example of your non-existant lack of any interest for science or even keeping your interlocutor informed of what you actually mean. Of course your interlocutor is to blame because he/she should have been able to imagine that when Terry says K he means IJK or maybe MNOPS depending on the circumstance but he is always right in any case. How could he be wrong?! Isn't he the great man who spelled Wallacea on the map?! The one who insists on having people migrating through Siberia naked and without even adapting their skin pigments to low radiation needs? The one for whom all humans are so extremely equal that a Denisovan a Neanderthal and a Sapiens... are exactly interchangeable like the cups of a thimblerig, all equally empty.

      Precision? What for. Terry is always right, loosely so.

      Scales, balance? What is that for? Naturally Terry will gladly pay you 6 for the value of 24. Don't insist in getting all your bill paid... he is always right no matter what maths might say. Instead of deducing where things may have happened, what for?!, Terry paints on the map inspired not just but God Itself but by Car Sagan's Garage Dragon and the Flying Spaghetti Monster as well... his ink blots may be capricious and based on almost nothing at all but dare not to question them because they are inspired by the Holy Trinity of Fairyland themselves.

      If Maju's disbelief still gets him counting the weights, it is no doubt because he keeps believing wrongly in the science of weight. To the pyre with him, we do not need data much less to ponder it carefully.

      And Maju burns in this pyre of words while a crowd of imps all miniature copies of Terry dance around crying: "Data?! We don't need any data!"
      [/end sarcasm]
      [/end discussion]

      ReplyDelete
    39. "Confusing IJK and K is mere "semantics", sure, and not another example of your non-existant lack of any interest for science or even keeping your interlocutor informed of what you actually mean".

      OK. To avoid confusion I will in future refer to the haplogroup in question as 'Y-DNA F(xF*,F1,F2,F3,G,H) from now on. I'm sure you really know exactly what I mean but I'll use that terminology from now on. Would you be content with F-M522?

      "Isn't he the great man who spelled Wallacea on the map?!"

      You had never heard of it, so stop complaining.

      "The one who insists on having people migrating through Siberia naked"

      I have never claimed they were 'naked'.

      "without even adapting their skin pigments to low radiation needs?"

      East asians today have hardly adapted 'their skin pigments to low radiation needs', so you're talking rubbish yet again.

      ReplyDelete
      Replies
      1. Why don't you just use the name given by ISOGG? We all do, with the very debatable exception of MNOPS/K(xLT).

        Secondly, I prefer to discuss the particulars of IJK after discussing F and K. Otherwise it is very difficult to understand. It's very important to understand F very particularly: its diversity, its global relevance, its obviously South Asian origins...

        "East asians today have hardly adapted 'their skin pigments to low radiation needs'"...

        Actually they are extremely pale in many cases, even if they live in high-radiation zones at the same or lower latitude than Mediterranean Europe. Also their known pigmentation alleles are different in almost all cases from those of West Eurasians, what clearly indicates two northwards migrations, as if all the other DNA did not speak enough volumes on their own right.

        I'm tired of your irrational fanaticism. Very much tired! I believe I deserve better, sincerely.

        Delete
    40. "Why don't you just use the name given by ISOGG? We all do, with the very debatable exception of MNOPS/K(xLT)".

      To everyone else it would have been obvious that my original comment (but K as a whole is particularly interesting) refered to was the basal F haplogroup called 'IJK' in ISOGG. Surelky 'IJK' includes 'K'. That is why the momenclature includes the letter 'K'. It was you who chose to confuse matters.

      "Secondly, I prefer to discuss the particulars of IJK after discussing F and K."

      Why? If we consider the whole IJK haplogroup we see immediately that, as I said originally, 'but K as a whole is particularly interesting. Haplogroups have peeled off it all the way from just out of Africa to Melanesia'.

      "It's very important to understand F very particularly: its diversity, its global relevance, its obviously South Asian origins..."

      F has by no means an obviously South Asian origin. There is no reason at all to assume that either Y-DNA G or F3 originated there. Neither is it true of IJK when IJ and K (per ISOGG) most likely split to the west of South Asia.

      "Actually they are extremely pale in many cases"

      In 'some' cases. Evenks, Mongolians, Tungus and Inuit can hardly be described as 'pale'.

      "I'm tired of your irrational fanaticism".

      I think the above shows you are accusing the wrong person of 'irrational fanaticism'. Try looking in a mirror.

      ReplyDelete
      Replies
      1. "F has by no means an obviously South Asian origin".

        Count the basal subclades and put them on a map. Then dare to repeat that before a mirror without laughing, you little hypocrite.

        As for IJK: it has two subclades: one apparently coalesced in West Asia (IJ) and the other from South Asia or farther East (K), so again I feel strongly inclined to locate its global origins at South Asia (because of parental weight)

        K would probably South Asian as well for the same reasons as above but reversed in direction.

        MNOPS (unless it can be proven that P diverged first, what for now is just a conjecture) appears to be from further East than South Asia but so far its substructure is poorly understood so we can still get important surprises.

        Whatever the case all those clades clearly hang from South Asian F and are scattered in or around SA.

        "Evenks, Mongolians, Tungus and Inuit can hardly be described as 'pale'".

        Japanese and Koreans are often paler than Europeans, just not even a fraction as reddish (they seem to lack enough pheomelanin). They live at Mediterranean-like latitudes however.

        There's nothing comparable to Northern Europe in terms of warmth combined with low radiation input (because of latitude but also cloudiness) and that's why the most extreme cases of stable quasi-albinism are found in those parts of Europe, and only there: they needed it for long term survival.

        You should know all this: everyone interested in anthropology should know this.

        Delete
    41. "The Ganges Delta (its Easternmost edge in fact) is the significant geographic and ethnic boundary between South and SE Asia".

      At last we are getting somewhere. You have always maintained that humans would easily have been able to cross that region on their way to australia by using the boats they were easily able to manufacture. I see now that you agree with me: the Ganges Delta was a considerable obstruction to eastward movement.

      Regarding your comment at the other blog where we are currently arguing. I see very few examples of conditional support for any statements you have made here either.

      ReplyDelete
      Replies
      1. "You have always maintained that humans would easily have been able to cross that region on their way to australia by using the boats they were easily able to manufacture".

        I still think that way. Stop trying to manipulate my words, please.

        "... the Ganges Delta was a considerable obstruction to eastward movement".

        You are totally misinterpreting my ideas, on purpose surely. The Ganges Delta, itself included, is where South Asia seems to end ethnically. It is a very densely populated region however and has signs of occupation in spite of its sedimentary nature (which tends to hide everything) since at least the Upper Paleolithic.

        [Terry: oh "UP" I can take this piece of info and use it as weapon for my argumentation just by removing "at least" and declaring it "absolute" by divine grace]

        [Maju: shit! I can foresee he will do that but I can't do much other than partake with everybody of my vision... so they are warned beforehand]

        As I was saying: SE East Asia or the transition region leading to it begins not in the Ganges Delta but East of it. The hill country or whatever... I don't know enough. All I know is that they are less populated than the two Bengals and I can only imagine that such a fertile region was also highly populated in the past, even if there was no farming yet (why not: basic farming is just replacing natural fertility by domesticated one).

        Why would I need to explain all this again? Good question.

        Delete
    42. "I still think that way. Stop trying to manipulate my words, please".

      I'm manipulating nothing. But I'm confused. What do you actually think. 'humans would easily have been able to cross that region on their way to australia by using the boats they were easily able to manufacture' or that 'The Ganges Delta (its Easternmost edge in fact) is the significant geographic and ethnic boundary between South and SE Asia'. You can't have it both ways.

      "You are totally misinterpreting my ideas, on purpose surely".

      You have not made yourself very clear as to what you actually believe. You change what you believe depending on what position you are currently taking.

      "The Ganges Delta, itself included, is where South Asia seems to end ethnically".

      And genetically.

      "It is a very densely populated region however"

      Very densely populated today, yes. Ever heard of agriculture?

      "has signs of occupation in spite of its sedimentary nature (which tends to hide everything) since at least the Upper Paleolithic".

      In the Delta? I don't think so. Evidence?

      "I can only imagine that such a fertile region was also highly populated in the past"

      Imagine whatever you wish. That doesn't prove your imagined scenario correct.

      "SE East Asia or the transition region leading to it begins not in the Ganges Delta but East of it. The hill country or whatever..."

      And virtually all connection between SE Asia and northeast India is 'from' that hill country, not 'into' it.

      "Count the basal subclades and put them on a map".

      That might have some relevance if you were prepared to compare regions of comparable extent. You are against such consideration, for some strange reason.

      "As for IJK: it has two subclades: one apparently coalesced in West Asia (IJ) and the other from South Asia or farther East (K), so again I feel strongly inclined to locate its global origins at South Asia (because of parental weight)"

      What 'parental weight'? Two haplogroups in 'West Asia' and one in 'South Asia'. IJ could easily be the 'parental haplogroup'.

      "MNOPS (unless it can be proven that P diverged first, what for now is just a conjecture)"

      Yes, and as far as I know you are the only one to conjecture such.

      "so far its substructure is poorly understood so we can still get important surprises".

      We may do, but unless the phylogeny is hugely altered it is reasonably evident that the haplogroup diversified considerably and rapidly somewhere near Wallacea.

      "Whatever the case all those clades clearly hang from South Asian F and are scattered in or around SA".

      Which clades? MNOPS? come off it. IJK hangs of F, and looks to hang off it to the west of India. As do the basal F haplogroups G and F3.

      "Japanese and Koreans are often paler than Europeans"

      But usually much darker. And I must confess I don't recall seeing any 'paler than Europeans' when I spent three weeks there in the 1980s. Of course they may have changed since then.

      "There's nothing comparable to Northern Europe in terms of warmth combined with low radiation input"

      Certainly not in 'warmth'. Although Japanese may 'live at Mediterranean-like latitudes' the climate is far from being 'Mediterranean'. Anyway, what has warmth got to do with pale skin?

      ReplyDelete
      Replies
      1. "What do you actually think. 'humans would easily have been able to cross that region on their way to australia by using the boats they were easily able to manufacture' or that 'The Ganges Delta (its Easternmost edge in fact) is the significant geographic and ethnic boundary between South and SE Asia'. You can't have it both ways".

        I do. The buffer zone appears to be mostly East of the Ganges in what they call NE India (Assam, etc.) and maybe some areas of Burma. There's no obvious genetic transition zone (if not extremely mild and diluted) in the Ganges Delta, that's because it was (most likely) always much more densely populated than neighboring regions of SE Asia, so any random inputs were easily diluted to quasi-homeopathic levels. The "border", the buffer zone, is the "hill country" not the Ganges Delta.

        "Ever heard of agriculture?"

        You're talking to a former gardener, please. Seriously: you seem to somehow imagine that a fertile area that now hosts lust orchards could not in the past, when the ecological balance had not even been disrupted yet, host lust jungles and swamps that provided a lot of food for Paleolithic standards.

        ""Count the basal subclades and put them on a map".

        That might have some relevance if you were prepared to compare regions of comparable extent".

        It does not matter: just use a map without regions: a blank map of Asia... locate the dots and estimate the centroid produced (you can use a geometrical method if you wish to be more precise). You cannot reject the laws of geometry and they produce a centroid in Central-East India invariably.

        → IJK: It is not clearly centered in West Asia. Even if you only consider LT (obviously centered in Pakistan) and IJ (coalesced in Iran plausibly), you still are between the two regions. And then the rest of K (MNOPS) pulls to the East very clearly and strongly (as does their ancestor F), so the origin of F, IJK and K are all probably in South Asia. G, IJ and MNOPS are "prodigal sons" so to say - albeit quite successful ones admittedly.

        You understand it better looking at the whole phylogeny than at particular lines in isolation in any case... unless you wish to entrench yourself in particular pre-conceived models.

        [→ P as possible "first son" of MNOPS] "... as far as I know you are the only one to conjecture such".

        And as far as I know too, don't get too hot about that. Still plausible, considering the geography of related mtDNA clade R. But time will tell in any case.

        "... unless the phylogeny is hugely altered"...

        Just well defined. MNOPS has to split (like every Y-DNA clade if we'd knew all the info hidden in the male chromosome) in pairs, which is the phylogenetic order of those pairs remains to be determined but the star-like structure we know as of now is for sure not ultimately correct (not correct enough, that's it), not good enough to say anything, much less with the emphasis you do.

        Why? Because you have no support from other lines of evidence, notably mtDNA.

        "I don't recall seeing any 'paler than Europeans'"

        Paler than some Europeans? I bet you do. Either that or you don't know enough the real European diversity.

        "Anyway, what has warmth got to do with pale skin?"

        Nothing, not directly. Solar radiation does. And at high latitudes in winter Solar radiation is too low for children's brains (and other organs) to develop properly because of lack of vitamin D synthesis in the skin, if that skin is too dark. It is the most obvious evolutionary adaption in humans since the OoA, you should be more familiar with it, especially as I have written on it in the past (and you were a reader, or at least a furious commenter, sometimes I doubt you read enough, of this blog already).

        Delete
    43. Maju wrote,

      "Japanese and Koreans are often paler than Europeans"

      Often paler than what Europeans? As terryt has commented already, I do not recall having met any Japanese or Korean who is paler than I, and I am an individual of European ancestry who interacts with members of these ethnic groups on a daily basis. A translucent complexion essentially does not occur among Mongoloids (though it might be found in some individuals from some predominantly Mongoloid populations who have significant recent Caucasoid admixture). Even the palest Mongoloids have an ivory or creamy white, opaque complexion.

      terryt wrote,

      "Certainly not in 'warmth'. Although Japanese may 'live at Mediterranean-like latitudes' the climate is far from being 'Mediterranean'. Anyway, what has warmth got to do with pale skin?"

      Most of Japan is subtropical or practically so (e.g. one can grow oranges in one's backyard up to about the border between the Kanto and Tohoku regions). The mountains facing the Sea of Japan coast of Honshu receive a great deal of snowfall, but it is still not especially cold. On the Pacific coast, snow is rare, with perhaps a few centimeters of it falling once a year if at all. On the other hand, the northern island of Hokkaido has a notably more continental, temperate climate.

      However, Japan receives most of its precipitation in the summer, and the difference in air temperature and humidity between summer (hot and humid) and winter (dry and moderately cold) is great, so it is definitely not a Mediterranean climate. I suppose one might say that the climate in Japan is similar to, but generally more mild and marine-influenced than, the climate along the eastern seaboard of the United States.

      Korea, like Hokkaido, generally has a much more continental and temperate climate than most of Japan, but the island of Jeju off the southwest coast of the peninsula is rather subtropical, and a lot of citrus is grown there, too.

      ReplyDelete
      Replies
      1. Well, if you're Northern European you may well not be the standard by which Europeans are measured, much less if we look to a time previous to the Medieval Agricultural Revolution (in which advances like the heavy plow allowed for a demographic explosion in Northern Europe, while the Mediterranean remained mostly unaffected). In any case, I will agree that the shade of white is not the same, but it is white anyhow. You emphasize a "translucence" that I can hardly identify (I emphasized pheomelanin's reddish or "anti-yellowish" tendency instead - note: human skin tones are all yellow-orange by hue but vary in the exact hue, some being redder and others yellower, they also vary in darkness, of course, brown or 'black', and paleness, beige, cream or 'white'). Maybe we are both right (never thought of "translucence" as a skin quality, admittedly).

        These differences are caused because, genetics dixit, both populations had apparently separate, distinct, evolutionary paths to paleness, what reinforces the idea of two distinct South to North migrations, one in the East and one in the West. It does not seem like either population had the need to incorporate (even via introgression) the genes evolved by the other for light skin tone. We are therefore before a case of convergent evolution.

        "Anyway, what has warmth got to do with pale skin?"

        Nothing. Solar radiation has and therefore latitude.

        Delete
    44. "I do [have it both ways]. The buffer zone appears to be mostly East of the Ganges in what they call NE India (Assam, etc.) and maybe some areas of Burma".

      Yes. The combined ganges/Brahmaputra Delta and the jungle-clad mountains of Northeast India combine to form a formidable barrier.

      "'The "border', the buffer zone, is the 'hill country' not the Ganges Delta".

      Not so. It is the combination of the two that forms the barrier.

      "that's because it was (most likely) always much more densely populated than neighboring regions of SE Asia"

      Very unlikely to be the case. Far more likely is that it was more recently colonised from nearby regions as boating improved and farming was introduced.

      "so any random inputs were easily diluted to quasi-homeopathic levels".

      Doesn't fit the evidence. The Ganges Delta does not have any specifically 'Ganges' haplogroups. That is surely unlikely to be the case if it has been long occupied.

      "host lust jungles and swamps that provided a lot of food for Paleolithic standards".

      Jungles and swamps are not very productive for humans with a Paleolithic lifestyle. Clear the jungle or drain the swamp and grow crops: yes, productive.

      "Why? Because you have no support from other lines of evidence, notably mtDNA".

      The mt-DNA does provide evidence. It's just that you have made up your mind not to see it.

      "just use a map without regions: a blank map of Asia... locate the dots and estimate the centroid produced"

      The 'centroid' is not relevant if the movement has been predominately in one direction. And when we place the dots for both Y-DNA MNOPS and mt-DNA R we find those in SE Asia, especially in Southern Wallacea, are particularly close to each other.

      "IJK: It is not clearly centered in West Asia. Even if you only consider LT (obviously centered in Pakistan) and IJ (coalesced in Iran plausibly), you still are between the two regions".

      Again you're willfully ignoring the phylogeny. IJ branched off before LT did. There is absolutely no reason to believe IJ and LT formed in the same region. That's just another example of your Garden of Eden Syndrome.

      "then the rest of K (MNOPS) pulls to the East very clearly and strongly"

      And that clade didn't form until some time after both IJ and LT had at least started their development some way to the west of where MNOPS coalesced. You can't use MNOPS to tell you where either IJ or LT coalesced.

      "MNOPS has to split (like every Y-DNA clade if we'd knew all the info hidden in the male chromosome) in pairs"

      I certainly will not be surprised if it is eventually shown that NO and P branched off some little time before the remainder of the haplogroup reached Wallacea. But even that would place its origin no further west than somewhere in Sundaland.

      "the origin of F, IJK and K are all probably in South Asia. G, IJ and MNOPS are 'prodigal sons' so to say - albeit quite successful ones admittedly".

      One 'fact' we know is that F has an ultimate origin, through CT, in Africa. Now, what would we expect to see in a haplogroup that had moved from Africa to Australia? Precisely what we see in Y-DNA F and IJK. A progressive branching along the route.

      "You understand it better looking at the whole phylogeny"

      Which you insist on constantly ignoring.

      "unless you wish to entrench yourself in particular pre-conceived models".

      Yes. And it is very apparent that you have entrenched yourself in a particular model: some sort of Garden of Eden theory where everything emerged from a single region at a single time.

      ReplyDelete
      Replies
      1. Most of your comment is one-liner opinions. Your problem: you know what I think and therefore I won't comment further.

        Some of them however appear to show confusion re. what I mean. For example:

        "IJ branched off before LT did. There is absolutely no reason to believe IJ and LT formed in the same region".

        LT is just one of two sequentially mentioned references for K, which is the phylogenetic "brother" of IJ. As I said above, the other reference is MNOPS, which appears to have branched further East, so the centroid of K is between Pakistan and SE Asia and therefore the centroid of IJK may be estimated (with uncertainty) to be near Pakistan.

        "The 'centroid' is not relevant if the movement has been predominately in one direction".

        Debatable if the centroid should be corrected for that (something I argued for in the past and you rejected vehemently, I must say - you swing a lot and very capriciously). But first of all in any case we must determine the centroid, which if needs correction must be argued starting from that point: the raw geometrical centroid.

        "And [MNOPS] didn't form until some time after both IJ and LT"...

        We can't say that. It is perfectly possible, in pure theory at least, that pre-IJ and pre-LT lingered as small ("private") lineages before expanding and forming the haplogroups as we know them now. I won't lean strongly on one or the other direction because I lack clear evidence (looking only at the Y-DNA phylogeny) but, looking at other evidence (mtDNA especially), I'd say that this scenario is quite possible at least for IJ and G (and maybe also LT). Unsure in any case.

        "Now, what would we expect to see in a haplogroup that had moved from Africa to Australia? Precisely what we see in Y-DNA F and IJK. A progressive branching along the route".

        I would not expect that (unless you cling to the "very rapid coastal migration" hypothesis in its fastest and most purely coastal form, what I believe you do not) and that is not what we see in F in any case: we see a clear expansion not in successive ordered branches from West to East but first of all in South Asia and only then around it.

        End of what is worthy discussing.

        Delete
    45. "I am an individual of European ancestry who interacts with members of these ethnic groups on a daily basis".

      That's interesting. Where do you live? I have quite a bit to do with several young Koreans here in New Zealand so I too am familiar with their skin colour.

      "Often paler than what Europeans?".

      Japanese women and men who work in offices will be paler than European men who work in the fields. I admit I saw people in Andalucia who were as dark as many Japanese, but these people usually claimed to be Gypsies.

      "And at high latitudes in winter Solar radiation is too low for children's brains (and other organs) to develop properly because of lack of vitamin D synthesis in the skin"

      So you believe that Inuit and Chukchis are intellectually inferior.

      "you should be more familiar with it, especially as I have written on it in the past"

      I have been aware of the theory for years, and I remain unconvinced. After all today people in the northern latatudes wear clothing yet don't suffere extremely from lack of vitamin D.

      "You emphasize a 'translucence' that I can hardly identify"

      Have you ever actually seen a person from Japan, Korea or even China?

      "The mountains facing the Sea of Japan coast of Honshu receive a great deal of snowfall, but it is still not especially cold".

      I was there in Winter, and I found it cold. Mind you I'd just come from Australia. But the region I lived in there had Winter snow.

      "They live at Mediterranean-like latitudes however".

      New Zealand is at the same latitude south as Japan and Spain are north, and geologically much the same as Japan. That is one of the main reasons I visited that country, to compare it to NZ. I live exactly opposite Cadiz. The climate here is not 'Mediterranean'.

      ReplyDelete
    46. "End of what is worthy discussing".

      It is obvious you still don't understand basic genetics, nor do you have any wish to understand the subject.

      "LT is just one of two sequentially mentioned references for K, which is the phylogenetic 'brother' of IJ".

      LT is very much a 'younger brother' to IJ. Check ISOGG. Haplogroup K underwent a further series of mutations (M9, P128, P131 and P132) after IJ and before LT. The mutation M526 then led to K(xLT) before that haplogroup split into K1, K2, K3, M, NO, P and S. I grant that this may not represent a star-like expansion, but it surely represents a fairly rapid one.

      "We can't say that [[MNOPS didn't form until some time after both IJ and LT]. It is perfectly possible, in pure theory at least, that pre-IJ and pre-LT lingered as small ('private') lineages before expanding and forming the haplogroups as we know them now".

      Perfectly possible. But less so in the case of LT. It has just the downstream mutations L298/P326 and L811 from K9x(IJ) but IJ has awhole string of them, and a further string within each of I and J. But as 'private lineages' there is no evidence they moved anywhere from where they first coalesced before they 'expanded'. The only reason you demand that they did is to make the evidence fit your pre-conceived belief. And there is certainly no evidence at all to support your seeming belief that IJ migrated out of South Asia at some time.

      "the centroid of K is between Pakistan and SE Asia and therefore the centroid of IJK may be estimated (with uncertainty) to be near Pakistan".

      The 'centroid of K' is goint to tell you nothing. 'K' is a series of haplogroups with a series of centroids. To get any realistic view you have to consider the centroid of IJK separately from the centroid of KLT (or whatever you care to call the haplogroup K(xIJ) and then the centrioid of MNOPS, taking care to give all 7 of the haplogroups within that group equal weight.

      "Debatable if the centroid should be corrected for that (something I argued for in the past and you rejected vehemently, I must say - you swing a lot and very capriciously)".

      I am extremely doubtful I have ever done that. Care to remind me where I did it?

      "I would not expect that (unless you cling to the 'very rapid coastal migration' hypothesis in its fastest and most purely coastal form"

      Absolute rubbish. A rapid migration would hardly leave time for new haplogroups to arise along the route. We would expect a string of haplogroups to be more likely to appear if the migration was particularly slow.

      "and that is not what we see in F in any case: we see a clear expansion not in successive ordered branches from West to East but first of all in South Asia and only then around it".

      Once more that is absolute rubbish. You can only claim to see the expansion as having started in South Asia if you have made up your mind in advance that that is what you want to see. We actually do see successive ordered branches from West to East. F is scattered along the same route as K. From west to east: G, IJK, F3, (all three west of India) H, F1, F3, (all three in India) and F2 (in East Asia). I would have thought that was simple, and obvious.

      The other day you wrote:

      "MNOPS has to split (like every Y-DNA clade if we'd knew all the info hidden in the male chromosome) in pairs"

      Why did you decide to confine that idea to just the Y-DNA? Surely the same should apply to mt-DN. Or might that conflict with one of you pet beliefs about mt-DNA M?

      ReplyDelete
      Replies
      1. "LT is very much a 'younger brother' to IJ".

        LT is not a "brother" to IJ, K is. LT would be a "nephew" (but sometimes nephews can be younger than uncles, more so in our context).

        "[LT] has just the downstream mutations L298/P326 and L811 from [Kx(IJ)]"...

        No. We do not know that: we only know of those SNPs but there could be many others yet to be researched. You are over-deducing from data that is necessarily incomplete.

        "The 'centroid of K' is goint to tell you nothing"...

        I am of the opposite opinion.

        And then you go with "extremely doubtful", "absolutely rubbish" and other opinionated and surely unfounded claims that are such a drain to me that I refuse to even read them. Ciao.

        Delete
    47. Maju wrote,

      "LT is not a 'brother' to IJ, K is. LT would be a 'nephew' (but sometimes nephews can be younger than uncles, more so in our context)."

      I think you must have intended to note that "sometimes uncles can be younger than nephews." This is correct, and it reminds me of a certain peeve of mine regarding people who argue for "multiple colonizations" of regions like Australasia or East Asia because of the co-occurrence in populations of those regions of clades that are widely separated on the present phylogenetic trees.

      For example, many people seem to be fond of the idea that populations whose Y-DNA belonged exclusively to D-M174 or C-M130 (or both) had spread throughout East Asia at a very early date and were subsequently replaced by populations whose Y-DNA belonged exclusively to O-M175. The only motivation that I can imagine for such an argument is the idea that O-M175 must have originated more recently than D-M174 or C-M130 because O-M175 is a sister clade of N-M231 and both these clades are downstream of F-M89, K-M9, and MNOPS-M526.

      However, as I have pointed out in one of my recent comments on this blog, the Y-STR variance of Australasian-specific subclades of MNOPS-M526 is equal to or greater than the variance of Australasian C-M130. Likewise, considering O2a1-M95 or O3-M122, subclades of O-M175, each exhibits far more variance than D-M174 or C-M130 in East Asia. O-M175 is also the most widespread and frequently occurring Y-DNA haplogroup in East Asia, so it should be considered as the best candidate for the first modern humans to reach East Asia in any such "multiple waves" model, not as recent colonists. (Of course, another subclade of O-M175, O2b-SRY465, exhibits much less Y-STR variance, and thus probably has expanded relatively recently from an ancestral population that has been bottlenecked somehow.)

      In order to argue otherwise, one must accept at least one of the following two propositions:

      (1) All measures of Y-STR variance (and hence all estimates, even relative ones, of TMRCA based on the molecular clock theory) are utterly useless.

      (2) Haplogroup O-M175 somehow has managed to accumulate Y-STR variance in some undetermined location outside of East Asia, maintain that Y-STR variance throughout the course of a massive population movement and colonization of East Asia, and subsequently be completely eradicated from its erstwhile homeland.

      ReplyDelete
      Replies
      1. "I think you must have intended to note that "sometimes uncles can be younger than nephews".

        Right. :D

        "For example, many people seem to be fond of the idea that populations whose Y-DNA belonged exclusively to D-M174 or C-M130 (or both) had spread throughout East Asia at a very early date and were subsequently replaced by populations whose Y-DNA belonged exclusively to O-M175".

        You are right in questioning this if presented as a supposed "fact" and in such simple terms. We cannot easily gauge that. I personally lean towards such hypothesis on various logical grounds (for example O sits between the two main areas of C distribution: Australasia and NE Asia, what could be well explained by this model, similarly it sits between the three main areas of D: Tibet, Japan and Andaman) but I admit I cannot be sure about it, certainly not on "age" reasoning alone.

        "In order to argue otherwise, one must accept at least one of the following two propositions:

        (1) All measures of Y-STR variance (and hence all estimates, even relative ones, of TMRCA based on the molecular clock theory) are utterly useless".

        I generally disdain TRMCA estimates, as you probably know. As for the variance, I'm not sure it can be properly gauged in "absolute" terms, i.e. comparing it between haplogroups with only remote relationship, but still may be of some use (at least preliminary estimation) within a particular clade. Why? Because simply one haplogroup, for whatever reasons (maybe an ancient bottleneck) has less of such diversity. Another reason may be that for random causes or hidden genetic internal logic the STR markers relevant in one or another clade are different (and as most STRs have been first tried in the most widespread clades, such as R, O, J or E, they work poorly for rarer ones).

        So I find your comparison of STR variance between C and O in Oceania to be an intriguing curiosity but with unclear meaning, if any at all. I hope that you understand my viewpoint.

        Delete
      2. Actually, considering what I say in my reply below to Terry re. this tree (based on 1000 genomes project samples), C and NO look contemporary (c. 50 Ka with my re-calibration of the chronology), with O being only slightly more recent (c. 45 Ka). So they may well have expanded in a similar window and process. D looks more recent (c. 25 Ka). I'm willing to swing in this case although the evidence would still need some refinement, as only R1b, O3 and E1b have really large samples.

        Delete
    48. "LT is not a 'brother' to IJ, K is".

      You're squirming away from considering the evidence, again. My comment was in reference to your comment the other day, 'LT is just one of two sequentially mentioned references for K, which is the phylogenetic 'brother' of IJ'. We can surely agree that LT and IJ are relations, and LT formed later than did IJ. Call LT 'brother', 'nephew' or whatever. That is hardly the point.

      "LT would be a 'nephew' (but sometimes nephews can be younger than uncles, more so in our context)".

      I agree that the final mutation that gave rise to IJ could be later than the final one that gave rise to LT. But the mutation that separated IJ from K is obviously earlier than the one that separated LT from KMNOPS. It is reasonabl;e to suppose that the original separation occurred in two separate regions. The only reason you would insist on them having separated from the line that was to give rise to KMNOPS in the same region as each other is the belief in some Garden of Eden scenario of human expansion.

      "We do not know that: we only know of those SNPs but there could be many others yet to be researched. You are over-deducing from data that is necessarily incomplete"

      A typically ridiculous comment from you. Would you have been satisfied if I'd written 'at least' in front of 'downstream mutations'? I doubt it. You are absolutely committed to avoiding the issue. My comment as to the relationship between IJ and LT with regard to the series of mutations leading to each haplogroup still stands.

      "I am of the opposite opinion [The 'centroid of K' is goint to tell you nothing]"

      I presume you hold the same opinion regarding the centroid of Y-DNA C2. And what about T? Does you opinion hold for all haplogroups?

      ReplyDelete
      Replies
      1. Yo cannot simply count SNPs, unless the whole Y chromosome (of a sizable sample for each clade) has been studied (not just sequenced but all the phylogenetically relevant SNPs found and named). Y-DNA genetics may be heading that way in the near future but it has not yet achieved such precision at all. By the moment we are just contemplating a small sample of all the accountable SNPs in all lineages, but in some (better studied) lineages larger than in others.

        A namesake of you (main page) has used a (very limited in sample size) such full-sequence count approach and has reached to this tree (whose chronology should be re-calibrated so CF splits c. 80-74 Ka. and CF'DE, aka CT, splits c. 90-125 Ka.). IF the results are valid (what we can't be sure of all cases yet because the haplogroups are only so-well represented but looks like a reasonable progress particularly for this endless debate), then the order of age of coalescence of haplogroups within F is (ages in brackets according to my recalculation after calibrating CF to 80Ka):

        1. K
        2. P and IJ (55 Ka)
        3. NO (50 Ka)
        4. O (45 Ka)
        5. J, I, I2, O2, O3, R (40 Ka)
        6. Q, R1b, LT, N (25-20 Ka)

        It's not yet "rocket science" but in what regards to age estimates it is indeed the most serious attempt I have ever seen, failing only in the calibration. I presume that T.D. Robb is concerned about, for example, the very old age that A0 would get if applying my kind of calibration, but it'd be still close to the overall age of the species (I get 250 Ka and the generally acknowledged age of H. sapiens is c. 200 Ka - it may also be a barely pre-Sapiens lineage reincorporated upon the migration to West Africa, assuming that the age estimate is correct).

        Whatever the case, that is a serious attempt at Y-DNA age estimation based on ALL SNPs (he's using the 1000 genomes project) and not what you're doing with only the ISOGG reported SNPs.

        "I presume you hold the same opinion regarding the centroid of Y-DNA C2. And what about T? Does you opinion hold for all haplogroups?"

        Of course it does and in fact I do accept that the origin of C2 (but not C2a) is surely at Wallacea. I have never considered the possible centroid of T but in preliminary view it looks Pakistani.

        Delete
    49. "The only motivation that I can imagine for such an argument is the idea that O-M175 must have originated more recently than D-M174 or C-M130"

      It is perfectly possible for a haplogroup to expand long after it has coalesced. But usually in that case the haplogroup will have a long tail of mutations. I realise Maju disagrees vehemently with that idea in that he believes long tails develop while a haplogroup is on the move. I think that scenario very unlikely. Haplogroups tend to diversify while on the move. On the other hand I actually agree with Maju here, 'simply one haplogroup, for whatever reasons (maybe an ancient bottleneck) has less of such diversity'.

      "the Y-STR variance of Australasian-specific subclades of MNOPS-M526 is equal to or greater than the variance of Australasian C-M130".

      As I recall that diversity of C did not include Australian C4-M347. That haplogroup's inclusion would almost certainly have increases C's diversity in the region. I'll point out that I'm not claiming necessarily that C and K did or did not arrive in Australia together.

      "Likewise, considering O2a1-M95 or O3-M122, subclades of O-M175, each exhibits far more variance than D-M174 or C-M130 in East Asia".

      But that variance in O as compared with C may be a product of more recent expansion of the first and a much smaller population base for the other two.

      "Haplogroup O-M175 somehow has managed to accumulate Y-STR variance in some undetermined location outside of East Asia"

      I doubt that anyone seriously considers that O originated anywhere other than in East Asia. However it may have accumulated its considerable diversity in a reasonably small region with a large population before expading greatly through the remainder of East Asia.

      "O sits between the two main areas of C distribution: Australasia and NE Asia, what could be well explained by this model"

      Believe it or not. I agree with Maju here. Mind you I doubt that C was widespread through East Asia before O began its expansion. So O hasn't actually 'replaced' C, but mainly moved into previously unexploited regions within East Asia. The same situation would apply to D.

      ReplyDelete
    50. Maju wrote,

      "Actually, considering what I say in my reply below to Terry re. this tree (based on 1000 genomes project samples), C and NO look contemporary (c. 50 Ka with my re-calibration of the chronology), with O being only slightly more recent (c. 45 Ka). So they may well have expanded in a similar window and process. D looks more recent (c. 25 Ka). I'm willing to swing in this case although the evidence would still need some refinement, as only R1b, O3 and E1b have really large samples."

      Which subclades of C-M130 have been represented in this data set? As I have noted previously, the East Asian subclades C3-M217 and C1-M8, each of which has a recent TMRCA compared to most major subclades of O-M175 (but rather similar to the TMRCA of O2b-SRY465), seem to be more closely related to each other (and to South Asian C5-M356) than either C3-M217 or C1-M8 is related to Australasian C2-M38. Perhaps C-M130 as a whole might exhibit internal variance equal to or greater than that of MNOPS-M526, but the fact is that the major subclades of C-M130 in East Asia, C3-M217 and C1-M8, seem to coalesce to a common ancestor much more recent than the common ancestor of all extant derivatives of O3-M122, for example. In other words, the common ancestor of C3-M217 and C1-M8 (and probably also C5-M356 IMHO) was a single individual in a single population somewhere (also in East Asia?) long after O3-M122 had diversified into a large number of subclades (implying that O3-M122 was already present in significant numbers over a wide area).

      The only way to harmonize this fact with a scenario in which C-M130 has preceded O-M175 in the course of the spread of modern humans into and across East Asia is to assume that all other branches of C-M130 have been wiped out (perhaps as a direct result of the spread of populations bearing subclades of O-M175), a single branch of C-M130 somehow has managed to survive (either in isolation or by admixing into another population), and the descendants of this single C-M130 individual have diversified into modern C1-M8, C3-M217, etc. (However, to the detriment of the scenario of survival of C-M130 in East Asia by minor admixture into a larger population, why does C3-M217 occur alongside representatives of N-M231 or Q1a-MEH2 in Siberia and alongside representatives of Q1a3-M346 in the Americas, where no representatives of O-M175 are to be found in the vicinity?)

      Basically, the diversity of C-M130 in modern East Asian populations is inconsistent with a hypothesis of assimilation of diverse and widespread "East Asian aborigines" bearing C-M130 by a later wave of expansion by O-M175-bearing populations. The diversity of C-M130 in modern East Asian populations is rather consistent with a hypothesis of "introgression" from a now essentially extinct population of Out-of-Africa humans that existed in at least one part of East Asia at a very ancient date (other relatives of the population that served as the source of this introgression must have been utterly obliterated).

      ReplyDelete
      Replies
      1. "Which subclades of C-M130 have been represented in this data set?"

        I can't say for sure but I know it is based on the 1000 genomes project, which lacks any Australian Aborigine sample, so I guess that C4 is absent but not C1, C2 and C3 (and maybe others like C5 and C*). Ask the author directly better.

        "why does C3-M217 occur alongside representatives of N-M231 or Q1a-MEH2 in Siberia and alongside representatives of Q1a3-M346 in the Americas, where no representatives of O-M175 are to be found in the vicinity?"

        Possibly Q1a met C3 (as well as the main NA matrilineages) in far NE Asia, where O is mostly absent. Even I'd dare suggest that C3 is a secondary arrival to America, probably related to the Na-Dene expansion (restricted to parts of N. America which largely overlap with Na-Dene areas), what allows for even a more belated incorporation of C3 to the Native American genetic pool.

        In any case I think that, if O and C expanded in about the same process and chronology, present distribution may just reflect a variety of founder effects, whose causes are effectively random for us. Notice also that in nearly all scenarios Y-DNA has some tendency to concentrate into few lineages, what may be due to sexual selection in the broadest sense (i.e. particularly successful men can have more offspring than women, causing a cumulative effect through time, which results in more male lineages being "drifted out" than female ones in conditions of low population densities - generally these pruned or reduced lineages tend to be the less common ones to begin with: nothing intrinsic about the lineages, just their initial numbers determine their chances of long term survival).

        In any case the extremely low densities of Siberian populations in the Paleolithic must be considered as allowing extreme genetic drift (pruning of clades, mostly the rarest ones), so maybe C3 (and others?) were first incorporated only to be pruned some time later (for example at the LGM, which must have been very challenging for those peoples).

        Some more notes:

        ... "implying that O3-M122 was already present in significant numbers over a wide area".

        Why not? If O and C expanded in the same process it would seem obvious that O dominated in numbers since the beginning. Not sure about the "single individual" issue however: it seems a bit exaggerated, but a question of initial numbers or proportions surely yes.

        Delete
    51. Logically, there is no way to escape the conclusion that C-M130 in East Asia reflects one of the following:

      (1) low-level introgression from a now-extinct population of prehistoric East Asia as described above (the "introgression hypothesis" -- note that this is the only hypothesis that is consistent with the proposition that C-M13O Y-DNA in East Asia reflects assimilation of a local population that preceded the now-dominant O-M175 populations in East Asia)

      (2) an instance of minor gene flow from a C-M130-bearing population into a widespread, well-established population of East Asia that mainly bore some subclade of O-M175 due to an immigration event (the "miscegenation" or "(Australasian?) immigrant hypothesis")

      (3) an expansion subsequent to near-extinction of C-M130 in a polymorphic population due to a stochastic process (the "genetic drift hypothesis")

      In any of these cases, why are representatives of C3-M217 also found in Siberia and the Americas as I have mentioned above? Multiple independent migrations from an isolated C3-M217-bearing source population onto three different substrata: one bearing O-M175 subclades in East Asia, one bearing N-M231 or Q1a-MEH2 subclades in Siberia, and one bearing Q1a3-M346 subclades in the Americas? Who were these C3-M217-bearing men, where on Earth did they emigrate from, and why were they so popular among foreign women? Considering both the centroid of the geographical distribution of modern representatives of C3-M217 and the present location of their distant relatives in C1-M8, the common ancestor of C3-M217 and C1-M8 should be located in or near pre-Jōmon Japan, with the emigration of some C3-M217 individuals to East Asia, Siberia, or the Americas being roughly contemporaneous with the incipient Jōmon Period.

      ReplyDelete
      Replies
      1. Assuming that Robb's findings are correct and therefore that C and NO expanded in a similar time-frame, I lean to explanation #3. The others may also make some sense and be even complementary, not exclusive, but I can't say.

        Again assuming that Robb is correct, I'd say that pre-C, like mtDNA pre-N and Y-DNA pre-D appear to be originally low sized survivors (maybe there were some others) from the primary early Eurasian genetic pool, which was dominated by Y-DNA F and mtDNA M. However there was a swing at a second moment in the relevance of mtDNA (N, and especially R, gaining much importance for some reason, probably in relation to Y-DNA MNOPS' own success), which is not mirrored by Y-DNA C and D (although they clearly kept enough numbers somewhere to eventually have their own localized expansions in the East Eurasian outskirts).

        A possible cause of pruning may have been the Toba supervolcano, which must have caused major difficulties to survivors, marking a before-and-after boundary in the Eurasian genetic pool without destroying it altogether. But regardless of the particulars I reconstruct the following:

        1. An initial population A dominated by yF and mtM.
        2. A secondary major population B near SE Asia (Bengal?), which may have benefited from Toba somehow, dominated by yMNOPS and mtN and especially mtR, later incorporating some other lineages here and there (notably yIJ in the West, more diverse ones in the East).

        Y-DNA C and D would be also Eastern survivors (mostly or totally pruned in South Asia by Toba?) which were incorporated to the wider MNOPS (mostly O in East Asia) expansion, succeeding only at the margins of that expansion and being reduced in the core instead.

        Delete
    52. "Yo cannot simply count SNPs, unless the whole Y chromosome (of a sizable sample for each clade) has been studied"

      This is simply your love of molecular clockology kicking in. My conclusions are not based on the number of SNPs and I am certainly not using them in any estimation of the age of the various haplogroups. I am merely looking at the order in which the ancestors of the various haplogroups branch off what became the 'main' line. I am interested only in the presence of SNPs. I do not date haplogroups by the number of SNPs as I accept the accumulation of mutations at the base of a haplogroup is dependent ot on mutliple factors.

      "Of course it does and in fact I do accept that the origin of C2 (but not C2a) is surely at Wallacea".

      The trouble with that suggestion is that C2's 'centroid' is somewhere in the wider Pacific, a region where C2 cannot possibly have originated. And T's centroid is in Egypt which again contradicts everything we assume about T's origin. I'm sure there are many more examples where the centroid cannot possibly coincide with the region of origin.

      "C3-M217 and C1-M8, each of which has a recent TMRCA compared to most major subclades of O-M175 (but rather similar to the TMRCA of O2b-SRY465), seem to be more closely related to each other (and to South Asian C5-M356) than either C3-M217 or C1-M8 is related to Australasian C2-M38".

      Now that's interesting as maju has previously postulated (without evidence) that NOP was the first haplogroup to split off from KMNOPS. So, using slightly more evidence than Maju has I will suggest that C3, C5 and C1 split from basal C before C2 or C4 did.

      "The only way to harmonize this fact with a scenario in which C-M130 has preceded O-M175 in the course of the spread of modern humans into and across East Asia is to assume that all other branches of C-M130 have been wiped out (perhaps as a direct result of the spread of populations bearing subclades of O-M175)"

      Or the constriction of habitat during the ice age. I agree with maju in that I think absolute genocide is rather rare in human history, although very long-term it seems to happen through drift.

      "a single branch of C-M130 somehow has managed to survive (either in isolation or by admixing into another population)"

      Certainly the most likely explanation to me.

      "However, to the detriment of the scenario of survival of C-M130 in East Asia by minor admixture into a larger population, why does C3-M217 occur alongside representatives of N-M231 or Q1a-MEH2 in Siberia and alongside representatives of Q1a3-M346 in the Americas, where no representatives of O-M175 are to be found in the vicinity?"

      Subsequent population migration and mixing? In other words I agree with maju, 'Possibly Q1a met C3 (as well as the main NA matrilineages) in far NE Asia, where O is mostly absent. Even I'd dare suggest that C3 is a secondary arrival to America, probably related to the Na-Dene expansion'.

      ReplyDelete
      Replies
      1. We are already around the 100th commentary in this thread, most of which are yours and my replies to you, so I believe it is much more than enough. Neither I nor the readers of this blog are here to pay attention to you and your ideas.

        I have no idea where the centroid of T lays (surely not in Egypt): we'd need to have a decent understanding of the phylogeny and a decent, roughly equal, sampling of global populations. It's totally off-topic also, so I decline to discuss that, as well as most of the rest you say. Let's see if I can find a last bit that deserves discussion...

        "maju has previously postulated (without evidence) that NOP was the first haplogroup to split off from KMNOPS".

        I speculated (not postulated) that it might have been P in fact (there's no NOP as of now and my speculative idea was that MNOPS would have first split near Bengal into P and MNOS and later this one in SEA into NO and MS). Your namesake's T.D. Robb data may support it oddly enough but only P and NO data are available. It's just a speculation, don't drown in a glass of water, Mr. Polemicist, OK?

        Delete
    53. "The diversity of C-M130 in modern East Asian populations is rather consistent with a hypothesis of 'introgression' from a now essentially extinct population of Out-of-Africa humans that existed in at least one part of East Asia at a very ancient date"

      I'm very much in agreement with that hypothesis. In fact I suspect that such a population is the source of the 'Mongoloid phenotype'.

      "(1) low-level introgression from a now-extinct population of prehistoric East Asia as described above (the 'introgression hypothesis' -- note that this is the only hypothesis that is consistent with the proposition that C-M13O Y-DNA in East Asia reflects assimilation of a local population that preceded the now-dominant O-M175 populations in East Asia)"

      Obviously I think this is the most likely hypothesis.

      "Who were these C3-M217-bearing men, where on Earth did they emigrate from, and why were they so popular among foreign women?"

      My suspicion is that they were from at least slightly further west than Japan. They had come to occupy very cold conditions so it wasn't that they were 'popular among foreign women' but that they were able to procure such women from neighbouring populations. I strongly suspect their original companions were members of mt-DNA A.

      "and especially R, gaining much importance for some reason, probably in relation to Y-DNA MNOPS' own success"

      And you are well aware of my thoughts on the source of that 'success'.

      ReplyDelete
    54. "I believe it is much more than enough".

      OK. Last comment from me on this thread:

      "I have no idea where the centroid of T lays (surely not in Egypt)"

      Sorry. Check out the distribution shown here:

      http://dienekes.blogspot.co.nz/2009/11/y-chromosome-diversity-human-expansion.html

      And here, were the centroid seems to be the Horn:

      http://en.wikipedia.org/wiki/Haplogroup_T-M184_(Y-DNA)

      I'm sure that in neither diagram does the centroid coincide with where it is most likely T coalesced. And C2's centroid would be pulled far to the east by the expansion of C2a. Now, don't say I shouldn't include C2a's distribution in calculating C2's centroid because that is exactly what you have done when you include all downstream clades within IJK's distribution to determine its centroid.

      "so I decline to discuss that"

      You refuse to discuss it mainly because it opens most of your claims so much that it is possible to drive a huge truck through them.

      "I speculated (not postulated) that it might have been P in fact (there's no NOP as of now and my speculative idea was that MNOPS would have first split near Bengal into P and MNOS and later this one in SEA into NO and MS)"

      Splitting hairs yet again. OK. So I 'postulated' that C3, C5 and C1 split from basal C before C2 or C4 did. If you feel free to postulate then surely others of us shoul;d be free to do so as well.

      "Your namesake's T.D. Robb data may support it oddly enough but only P and NO data are available".

      Taking his data as seriously as you appear to be doing is really strange after your comment, 'Yo cannot simply count SNPs, unless the whole Y chromosome (of a sizable sample for each clade) has been studied'. But I see the obvious explanation: you accept 'molecular-clockology' when it fits your belief but not when it contradicts your belief.

      ReplyDelete
      Replies
      1. You are not wanting to understand how I determine the phylogenetically pondered centroids: it has nothing to do with frequency but with the mapping of basal subclades.

        We are not discussing T here anyhow.

        "Taking his data as seriously as you appear to be doing is really strange after your comment, 'Yo cannot simply count SNPs, unless the whole Y chromosome (of a sizable sample for each clade) has been studied'".

        That's exactly what Robb did with the 1000 Genomes data, except that many haplogroups have still relatively small samples and may need more study. Also his calibration for the linear chronology is blatantly wrong.

        Even if some clades' estimates are not still reliable enough IMO, others are fine because if you get R1b and R1a, then R1 (not R) is most likely correctly measured, and if you get E, D, F and C, the "CT" (CF'DE) is most likely correctly measured, including all relevant diversity in it.

        Why don't you try to understand things before fighting against them?

        Delete
    55. Maju wrote,

      "Possibly Q1a met C3 (as well as the main NA matrilineages) in far NE Asia, where O is mostly absent. Even I'd dare suggest that C3 is a secondary arrival to America, probably related to the Na-Dene expansion (restricted to parts of N. America which largely overlap with Na-Dene areas), what allows for even a more belated incorporation of C3 to the Native American genetic pool."

      C3-M217 Y-DNA has been found in at least Colombia (Wayuus) and Ecuador in addition to being found throughout North America (and not only among speakers of Na-Dene languages, as is often mistakenly reported, though some Na-Dene-speaking populations do exhibit unusually high frequencies of lineages belonging to the C3b-P39 clade).

      In any case, the utter lack of O-M175 Y-DNA in indigenous Americans makes it that much more likely that the migration(s) that introduced C3-M217 into the Americas did not include males from populations of continental East/Southeast Asia, who exhibit O-M175 Y-DNA with extremely high frequency and variance. On the flip side, Q-M242 exhibits great variance and is widespread (though with low frequency) in modern populations of East/Southeast Asia, so it is plausible that an ancient Q-M242-bearing population might have admixed with an ancient C3-M217- or proto-C1&C3-bearing population somewhere in eastern Asia prior to the population of the Americas by modern humans.

      ReplyDelete
      Replies
      1. Well, then the first model: they met in Far NE Asia, meaning Mongolia or further North and went to America hand by hand (but in a Q-dominated population). In fact Mongolia and some nearby parts of China are where the "mode 4" tech is first found (according to present day archaeological knowledge), however this contact area may have extended further North into the southern Siberian strip or even much of Eastern Siberia, where Archaeology does not seem to be informative as of yet.

        ... "makes it that much more likely that the migration(s) that introduced C3-M217 into the Americas did not include males from populations of continental East/Southeast Asia, who exhibit O-M175 Y-DNA with extremely high frequency and variance".

        I thought this was as a matter of course. All Native American lineages appear to be Siberian with the partial exception of mtDNA B, which has only relatively low frequencies among Mongols (even less among Buryats, Altaians) and seems very rare to unheard of in Siberia proper. But that's what the concept "founder effect" is for, right? It's not like it was not close to Siberia if it was near Beijing at the time of the proto-Amerindian transit (and therefore incorporation of most of the mtDNA pool).

        "Q-M242 exhibits great variance and is widespread (though with low frequency) in modern populations of East/Southeast Asia"...

        I have no idea where you get that from. If it said "Central/West Asia", that would be correct but Q is rare in East Asia other than the Northeast. Alternatively you may be thinking of O.

        Delete
    56. This comment has been removed by the author.

      ReplyDelete
    57. Maju wrote,

      "In any case the extremely low densities of Siberian populations in the Paleolithic must be considered as allowing extreme genetic drift (pruning of clades, mostly the rarest ones), so maybe C3 (and others?) were first incorporated only to be pruned some time later (for example at the LGM, which must have been very challenging for those peoples).

      Some more notes:

      ... "implying that O3-M122 was already present in significant numbers over a wide area".

      Why not? If O and C expanded in the same process it would seem obvious that O dominated in numbers since the beginning. Not sure about the "single individual" issue however: it seems a bit exaggerated, but a question of initial numbers or proportions surely yes."

      You do accept the theory of singular origin of haplogroups, correct? If you do not accept this fundamental theory, then you should not be discussing the topic of haploid genetics at all.

      In regard to the molecular clock theory, you seem to be fond of calibrating the clock so that it produces older dates, but you do not seem to be averse to the idea that Y-STR variance is proportional to age, since you seem to accept the relative ages of clades as explained according to molecular clock-type analyses.

      If you accept both the singular origin of haplogroups and the hypothesis that intraclade Y-STR variance is proportional to age of the clade, then you must accept that a single C3-M217-bearing man lived at a time when O3-M122 had already diversified into several subclades (thus requiring a large population that could sustain such genetic diversity).

      Likewise, if the Y-STR variance of the set {C1+C3} is lower than the Y-STR variance of the set {O3}, then you must accept that all modern C1-M8 and C3-M217 individuals descend in the direct patrilineage from a single individual (proto-C1&C3) who lived at a time when carriers of O3-M122 were already numerous.

      It is implausible that the same subset of C3-M217 should have been pruned in multiple populations (one population whose other Y-chromosomes belonged to O-M175, another population whose other Y-chromosomes belonged to N-M231, and another population whose Y-chromosomes belonged to Q-M242) in such a manner that the entire set of extant C3-M217 (including representatives from all the aforementioned hypothetical populations: {C3+O}, {C3+N}, {C3+Q}) should exhibit lower Y-STR variance than the representatives of O, N, or Q from each of those polymorphic populations.

      In the simplest scenario, the relative variance of these clades is what naturally should be expected if carriers of C3-M217 had colonized or influenced through gene flow several substrate populations belonging to subclades of O-M175, N-M231, or Q-M242, but no one seems to want to consider this simplest scenario (perhaps because of "haplogroup nepotism"?). Researchers have noticed this discrepancy early on, but they have tried to write it off as being the result of a lower population size for C-M130 (or D-M174) in a restricted geographical area as opposed to larger population size and living area for O-M175 under the assumption of a Neolithic origin for the latter clade. However,
      there is absolutely no way that the origin of the O-M175 clade (or even any of its most populous subclades, O3-M122, O2a1-M95, or O1a-M119) is as recent as the Neolithic under anyone's calibration of the molecular clock. It seems odd then to explain the excess of Y-STR variance of Paleolithic time depth in O-M175 relative to C-M130 in East Asia as being due to the Neolithic lifestyle of populations carrying O-M175 (and each of its subclades for that matter, since each is clearly older than the Neolithic), especially so when one considers the recency of the generally accepted onset of the Neolithic in East/Southeast Asia.

      ReplyDelete
      Replies
      1. "You do accept the theory of singular origin of haplogroups, correct?"

        I have never seen that "theory" formally expressed and demonstrated (can you direct me to a relevant paper?) and I'm not even sure of what you mean by it. If you mean that each haplogroup represents a distinct ancestral population, that is not necessarily correct and if you go around assuming that then, citing your own words, "you should not be discussing haploid genetics at all".

        Haplogroups may or not be from different populations: seldom a population is as extremely homogeneous as to have only one haplogroup. Of course it depends of the particular population and of what you operatively define as "population" (larger, smaller, etc.) but in most cases several lineages coexist and tend to coexist in dynamic interaction in the long run because populations are almost never pure isolates, where drift can act to that extreme.

        So both possibilities coexist like life and death in Schrödinger's cat and must be evaluated, unless we have some other evidence (that we usually do not) as roughly equiprobable.

        I believe that it is I who should be upset that people go to dissect genetics with a deterministic (obsolete Newtonian-reductionist) approach and not understanding even the basics of the Science of Chaos (or even the basics of Holism). Unobserved things are not this or that exclusively (.XOR.) but they are this, or maybe that, or maybe both or maybe something else. If you think in black and white, in false dichotomies, you can't really understand reality and will always have a (correct) feeling of confusion. But, well, I'm going off-topic (necessarily so, I believe).

        Delete
      2. "you do not seem to be averse to the idea that Y-STR variance is proportional to age"...

        In truth I prefer to pass of "molecular clock" hunches and most of the time I instinctively ignore whatever refers to that pseudoscientific mathematical artifact. A well revised and properly demonstrated MC theory may in the future be of some help, and I consider Terry Robb's attempt as a clear step in the right direction. But STR study seems biased or even altogether wrong, producing results that are very much unrealistic.

        Unlike C14 and other scientific age measuring methods, MC has never been demonstrated just postulated and parroted scholastically. That is not empirical science.

        So first of all is wrong to even bring the results as if they'd be "facts" or "data" to consider and not a mere educated guess. However there may be a seed of truth in the concept of MC basis but for that we need data that is not subject to researchers' bias (like selection of these but not those STR markers, crazy claims of Pan-Homo split being less than 8 Ma, a systematic bias in all MC speculations, etc.) That could be done counting all SNPs in the Y chromosome (not STRs but SNPs) as Robb does.

        It does not work that well for mtDNA because mtDNA SNPs do not happen in every generation, so the dominant haplotype will almost always drift out the novel ones, but in Y-DNA all haplotypes of each new generation are novel and unique, I understand, hence the movement never stops and can be measured properly. But all the Y chromosome must be considered and for SNPs, because STR markers are not too reliable re. phylogeny, so they won't be reliable re. MC calculations either.

        That system, now costly but every day less so, fulfills the theory and just needs proper calibration (like age(CF)=80Ka), but even if promising it still need to be demonstrated with empirical confirmation via aDNA or maybe some other method that I cannot imagine. C14 age estimation was not accepted because someone wrote a paper and the maths appeared correct: it was accepted after being used to empirically calculate the age of ancient objects, whose age was known by other methods as well.

        Empiricism is the mother of Science. A mathematical theory that has not been proven empirically is nothing and pretending it is "science" is mere charlatanism.

        Delete
      3. "... if the Y-STR variance of the set {C1+C3} is lower than the Y-STR variance of the set {O3}, then you must accept that all modern C1-M8 and C3-M217 individuals descend in the direct patrilineage from a single individual (proto-C1&C3) who lived at a time when carriers of O3-M122 were already numerous".

        I am not at all certain that it means that. First of all, while plausible, we still lack any empirical (SNP) evidence of C1 and C3 sharing a common ancestor other than C-root itself; secondly: you have not provided the evidence implicit in the sentence above; thirdly: the STR markers used for O and other K-derived lineages may not work well for C (careful there!), STR variance (which is just suggestive, not proving, of whatever you read in it) should not be compared across such distant haplogroups at all (that's something you insist to do but that is clearly wrong).

        "It is implausible that the same subset of C3-M217 should have been pruned in multiple populations"...

        Agree. Sharing an ancestor implies a sharing a founder effect of some sort. I do not think I implied that by any means: I suggest a C3 founder effect just north of the O3 area of expansion (say Mongolia and China respectively for simplicity. There was a pre-C3 among the NEA seed population but that was drifted out; the C3 we see now comes all from a single ancestor one of several/many that carried the minor pre-C3 lineage (C1'3?) in the early colonization of NEA. If your hypothesis of C3 and C1 sharing an ancestor is correct (looks reasonable), then C3 would be one and C1 another survivor.

        "Researchers have noticed this discrepancy early on, but they have tried to write it off as being the result of a lower population size for C-M130 (or D-M174) in a restricted geographical area as opposed to larger population size and living area for O-M175 under the assumption of a Neolithic origin for the latter clade".

        I'm not sure it applies at all. But if it does, then the "smaller clade at origin" theory looks a reasonable explanation and is consistent with what I'm saying of C and D expanding as part of the MNOPS wave, more or less, as product of minor lineages within that wave (there may still have been some sort of ethnic differences but we cannot understand them well, so assuming they were all part of the same pop. seems a good starting point).

        By drift (I presume, alternatively by force) O became dominant in all SEA and middle EA but related and unrelated lineages survived in the periphery: N, D variants, C variants and also Q in the Far North. These lineages either were ab origin part of the O expansion (clearly the case of N and not the case of Q) or were pushed out by the O expansion (or a mixed complex scenario - after all we are discussing many millennia in a few lines).

        "... there is absolutely no way that the origin of the O-M175 clade (or even any of its most populous subclades, O3-M122, O2a1-M95, or O1a-M119) is as recent as the Neolithic under anyone's calibration of the molecular clock".

        Never said otherwise. The most recent age estimate I ever read for O3 is c. 30 Ka. But guess I don't read as much molecular-clock-o-logy as you do. IMO the presence of B4'5 near Beijing c. 40 Ka ago, suggests to me that the O expansion was already almost finished by then (and that is coincident with Robb's full Y-DNA SNP new clock if properly re-calibrated at age(CF)=80Ka).

        Delete
    58. Maju wrote,

      "I have never seen that 'theory' formally expressed and demonstrated (can you direct me to a relevant paper?) and I'm not even sure of what you mean by it. If you mean that each haplogroup represents a distinct ancestral population, that is not necessarily correct and if you go around assuming that then, citing your own words, 'you should not be discussing haploid genetics at all'."

      I really cannot endure such dissimulation.

      It should be obvious to anyone (even by context alone) that I have referred to the assumption that underlies all research on haploid genetics: that every haplogroup has a single origin (necessarily in a single individual) and therefore represents a monophyletic clade.

      ReplyDelete
      Replies
      1. Alright. That is correct: each haplogroup comes from a single ancestor (SNPs are extremely rare and cannot statistically happen twice in Human timeframe, much less if there is more than just one SNP defining a lineage).

        It was not "dissimulation": I truly thought you were proposing something that Terry T. has insisted upon once and again: one haplogroup = one population, which I strongly disagree with, at least as preliminary assumption.

        Delete
    59. Aplogies. Me again.

      "I truly thought you were proposing something that Terry T. has insisted upon once and again: one haplogroup = one population"

      Here we go again. 'Dissimulation'. I have never claimed any such thing. I have claimed, which is probably where you're getting mixed up, that haplogroups have not all originated in the same place and at the same time. That is a theory you seem to accept without question and has led to my coining the phrase 'Garden of Eden Syndrome'.

      "You do accept the theory of singular origin of haplogroups, correct? If you do not accept this fundamental theory, then you should not be discussing the topic of haploid genetics at all".

      Many times I have had doubts about Maju's acceptance of such. For example:

      "I have never seen that 'theory' formally expressed and demonstrated'

      The fact he needs to have it 'formally expressed and demonstrated' is alarming. But finally, revelation:

      "Alright. That is correct: each haplogroup comes from a single ancestor"

      At last. And along with your recent realisation that Northeast India/South China and Baluchistan/Afghanistan represent barriers to free migration are you now prepared to accept that IJK separated into IJ and K at the Baluchistan/Afghanistan barrier and that LT and MNOPS separated at the Northeast India/South China barrier?

      "is consistent with what I'm saying of C and D expanding as part of the MNOPS wave, more or less, as product of minor lineages within that wave"

      Garden of Eden Syndrome. I find that very unlikely. MNOPS is largely southern while C3 is northern. D is somewhere in between. I suspect that all three lineages expanded independently, and at different times. Of course I accept that some downstream clades of one haplogroup combined with downstream clades of another at times during the various expansions.

      "In truth I prefer to pass of 'molecular clock' hunches and most of the time I instinctively ignore whatever refers to that pseudoscientific mathematical artifact".

      Except in the case of the T.D. Robb data. And he certainly does 'simply count SNPs' and includes examples where 'the whole Y chromosome (of a sizable sample for each clade)' has not 'been studied'.

      "and not only among speakers of Na-Dene languages, as is often mistakenly reported, though some Na-Dene-speaking populations do exhibit unusually high frequencies of lineages belonging to the C3b-P39 clade"

      But that doesn't preclude an original association with Na-Dene languages. Y-DNA can travel ahead of any language expansion.

      "you must accept that all modern C1-M8 and C3-M217 individuals descend in the direct patrilineage from a single individual (proto-C1&C3) who lived at a time when carriers of O3-M122 were already numerous".

      That numerical superiority need not indicate a particularly wide distribution though. Just fairly ideal conditions for the presence of a sizable population.

      "there is absolutely no way that the origin of the O-M175 clade (or even any of its most populous subclades, O3-M122, O2a1-M95, or O1a-M119) is as recent as the Neolithic under anyone's calibration of the molecular clock".

      I think we can assume, though, that the Neolithic began in a region already reasonably well populated. And certainly some clades do seem to have expanded as recently as at least the early Neolithic. O2a1a, O3a1c, O3a2c1 and possibly O3a2b.

      ReplyDelete
      Replies
      1. Sorry, Terry, but many times you have argued that different haplogroups or seeds of haplogroups could not have traveled together to their same or nearby final destination, that each one must have migrated separately. I actually recall one such discussion on East Asian lineages but you have leaned towards that idea more often, although maybe you abandoned it at some point.

        "Many times I have had doubts about Maju's acceptance of such".

        Spare us your demagogy, Mr. Polemicist.

        Delete
    60. This is not actually the place for this but it was embargoed for some time and it is the first I've seen the whole paper. Although it does deal with molecular-clockology and is a year old I'm sure you will find it interesting, perhaps even blog about it:

      "The hypothesized Gulf Oasis appears to be the most likely locus of the earliest branching of haplogroup N, including the three relict basal N(xR) haplogroups studied here, as well as the major Eurasian haplogroup R".

      http://www.sciencedirect.com/science/article/pii/S0002929711005453

      "We show that haplogroups N1 and N2, as well as haplogroup X, are ancient relicts of the dispersal of modern humans from Africa 160 ka ago and can provide insight into the earliest stages of the process".

      The authors do seem to make the mistake of calling both N1 and N5 basal N haplogroups. This is presumably to make the evidence fit the fashionable 'southern' migration:

      "they are most likely relicts from the first modern-human settlement in Arabia during the earliest stage of the southern coastal dispersal from the Horn of Africa to the rest of the world".

      But they later say:

      "In Southern Asia, only one new extant N(xR) branch, N5, arose; N5 is extremely rare and has also been recently described in Iran, raising the possibility that this lineage could also have arisen in Southwest Asia".

      Interesting. That seems to wipe out any 'via India' route, unless you're going to stick with R in South Asia. On that subject the authors say:

      "This suggests that haplogroup N had an origin immediately outside Africa, most likely in the Arabian Peninsula. This region was probably also the cradle for haplogroup R, dating to 159 ka ago".

      So they have R originating in SW Asia as well. But the study does not include R so their claim is not well-supported here.

      "Other N(xR) haplogroups include N5 in Southern Asia, N9 and A in Eastern Asia, N21 and N22 in Southeast Asia, and O and S in Australasia".

      For a start N13 and N14 (which they leave out) along with O and S are confined to the 'Australian' portion of Australasia. And the authors don't mention N8, N10 or N11 either, all South Chinese. And N5 is part of N1'5. Of course I agree with this statement:

      "Strictly speaking, haplogroups N and R could have arisen anywhere in the region between Southwest Asia and Australasia; basal branches are found in this region today. However, given L3's genesis in eastern Africa, the most parsimonious location for their origin is in the vicinity of the Arabian Peninsula".

      That has long been my position. In summary:

      "Whereas distinct episodes of dispersal and expansion into and within Europe (and indeed America) can be detected, demographic expansion in Southwest Asia appears as a continuous phenomenon from the Late Glacial period through to the Neolithic period. This genetic evidence is consistent with archaeological interpretations of the expansion of sedentary Natufian hamlets in the Levant during the wet phase between 15 and 13 ka ago. Such expansion lead to the techniques of agriculture and domestication under the harsher conditions of the Younger Dryas and ultimately to the Neolithic cultures of the early Holocene"

      ReplyDelete
      Replies
      1. Yeah, not the place. You might want to use your own forgotten blog and then drop a line, instead of the whole article, thanks.

        Delete
      2. Anyhow, the study does not demonstrate in any way that N or R spread from West Asia, instead the authors declare their belief that it was that way on alleged grounds of "parsimony" (wrong: unsupported by the actual distribution of basal subclades) and then they go on to study Western lineages of N(xR) and that's it. I'm positive that it does not introduce any new element that could change my viewpoint much less their belief is demonstrated in any way. It's quite a pointless study in that aspect, although of course it may have some interest for data collection (but nothing else).

        Delete
    61. "wrong: unsupported by the actual distribution of basal subclades"

      'In your opinion'. The authors spend much of the paper showing that in fact they are dealing with basal N clades.

      "I'm positive that it does not introduce any new element that could change my viewpoint"

      I'm sure that is correct. It would take a huge bomb to make you change your opinion. Evidence alone is not going to do it.

      "Sorry, Terry, but many times you have argued that different haplogroups or seeds of haplogroups could not have traveled together to their same or nearby final destination, that each one must have migrated separately".

      Where the archeology supports such a conclusion. On the other hand you have consistently had huge difficulty accepting that the SE Asian 'Austonesians' at various times before they emerged into the Pacific included Y-DNAs O1, O3 and C2 along with various clades of B4 and B5.

      ReplyDelete
      Replies
      1. As far as I saw most N subclades are ignored and only the three N(xR) ones with main distribution in the West (N1, X, W) are dealt with. If you are only interested in these clades (3 out of 14, or is it 18?), then the paper is somewhat informative. But absolutely misleading in regards to N as a whole.

        I don't know why the bring about Austronesians (just to drive me crazy, I bet) but I disagree with your claim about my opinion. I also disagree with your use of the word "emerged". But trying to avoid getting driven off-topic again.

        Delete
    62. Update: new "working paper" on the matter: http://forwhattheywereweare.blogspot.com/2013/04/questioning-india-australia-link-and.html

      ReplyDelete

    Please, be reasonably respectful when making comments. I do not tolerate in particular sexism, racism nor homophobia. Personal attacks, manipulation and trolling are also very much unwelcome here.The author reserves the right to delete any abusive comment.

    Preliminary comment moderation is... ON (your comment may take some time, maybe days or weeks to appear).