The big South Asian ancient DNA paper is finally out (in pre-print form) after many months of waiting and publication delays (purportedly for political reasons). UPDATE: This is not actually the long awaited Indian DNA paper, although it answers many of the same questions by using different ancient DNA data. END UPDATE.
The big news that we've been awaiting is the twelve ancient DNA samples from a cemetery in Pakistan. The gender, mtDNA, Y-DNA and Y-DNA classification of those samples follows.
There are also hundreds other ancient DNA samples over a wide range of time from the vicinity included. I will update my analysis and the findings as time permits. The abstract, some key figures and lots of discussion in the comments is available at Eurogenes. Razib does a nice job of dissecting the pre-print.
| M | R30b1 | L1a | South Asian Private | |
| M | .. | L1a | South Asian Private | |
| M | W3a1b | R1b1a1a2a1a1c2b2b1a2 | R-M269/U106/Z381/S499 | |
| M | W3a1b | R2a | South Asian Private | |
| M | U2e1'2'3 | C1b1a1a1 | SW Asia, Central Asia and South Asia | |
| M | M4 | L1a | South Asian Private | |
| M | M5a | Q1b2 | SW Asia | |
| M | U7a | .. | Female | |
| F | T2g1 | .. | Female | |
| F | U2c1 | .. | Female | |
| F | U3b1a1 | .. | Female | |
| M | T2g1 | L1a | South Asian Private | |
It strongly appears that Y-DNA L1a, R2a and Q1b2 al derive from Iranian farmers.
UPDATE April 4, 2018: I left this comment at Razib's blog:
“Also, the polytomy at the base of the eastern Eurasian human family tree, where all the major lineages diverge rapidly from each other, makes me wonder about gene flow vs. admixture. It seems possible that the polytomy may mask a phylogenetic tree topology which had gradually bifurcating nodes, if periodically a single daughter population replaced all its sister lineages in a local geographic zone. Much of history in human meta-populations may be characterized by isolation-by-distance and gene flow, erased by the extinction of most lineages and expansion of a favored lineage.”
This is a really interesting observation, and both possibilities, naively, seem pretty plausible.
It is almost unthinkable that there was complete genetic continuity and stasis in South Asia from prior to 65,000 BP to 6000 BP. All of the examples we have point to very sustained stasis as very rare, even though turnover may have happened less frequently among hunter-gatherers than among Holocene people at the continental level. The late modern history of hunter-gatherer tribes of Native Americans, for example, suggests that moderately long distance folk wanderings and exterminations of whole tribes were relatively common even on time scales of 1000 years or so.
There were probably at least two significant waves of migration and expansion after the one that gave rise to the Papuans in mainland Asia that greatly interrupted HG genetics there.
Y-DNA D people were probably a mid Upper Paleolithic Northern route arrival in Asia (although pre-LGM given Y-DNA D in the Andamanese) and were possibly male dominated (given that the Onge autosomally are close to AASI), that migrated to South Asia from Tibet to the South ultimately reaching Burma and the Andamans (since phylogeny-wise the Y-DNA D of India and the Andamans is closer to Siberian and Tibetan Y-DNA D than to Japanese Y-DNA D which splits at a very basal point from other Y-DNA D). Other Y-DNA D people take the Northern route to become the founding Jomon people of Japan. Most Y-DNA D people of Northern Asia in between are wiped out in the LGM.
Y-DNA C is remarkably rare and quite low if phylogenetic diversity in South Asia. This could simply mean that the coastal route theory for Y-DNA C is wrong, and that instead it took a clockwise northern route to reach East Asia, mainland SE Asia and Island SE Asia, and the lack of phylogenetic diversity of Y-DNA C in South Asia tends to support that reading of the data. But, another possibility, given the proportionately high level of C-M130* in South Asia relative to other Y-DNA C haplotypes is that Y-DNA C differentiated from Y-DNA CF in India, with lots of Y-DNA C people migrating east, but a few remaining, and that Y-DNA F people (including sister Y-DNA clade Y-DNA H people) subsequently wiped out most of the original Y-DNA C right population of South Asia, and that a lot of Y-DNA C people in India today are associated with a Y-DNA C1b1 back migration later in the Upper Paleolithic. The fact that autosomal ASI ancestry in India is pretty much proportional to Y-DNA C proportions in India, even though the proportions are low, also points to the antiquity of Y-DNA C in India, followed by later events.
In particular, the Y-DNA C people in India were probably marginalized by the expansions of Y-DNA F*, F1, F3 and H in India in the pre-Neolithic period, with other clades of F and daughter clades derived from F expanding into both West Eurasia and East Eurasia where the expanding clades became dominant. It is hard to know what gave the Y-DNA F/H people a decisive advantage over the Y-DNA C people in India and elsewhere, although forced to supply my best guess, I might suspect dog domestication or perhaps mastering how to turn wild grains into flour (flour predates the Neolithic revolution by at least ten or twenty thousand years).
It is also worth noting that ancient DNA suggests that in parallel with these developments in India, that Y-DNA C was once much more common than it is today in Europe, which definitely reflects Neolithic and Steppe driven replacement of remaining European HGs with Y-DNA C, but which may also reflect Mesolithic era replacement.
Similarly, in East Asia and SE Asia, Y-DNA O which is also a remote descendant of Y-DNA F, also sweeps those regions even before the Neolithic revolution.
If Indian nationalists want to discuss their basal and formative influence on the rest of the world, they would be well advised to de-emphasize the Bronze Age and to instead focus on how, on one hand, Y-DNA F is the dominant ancestor of modern Eurasian Y-DNA clades and that it probably originated in India (or at least had its first major expansion there), and how, on the the other hand, in the Iron Age, Buddhism, which also has its origins in India, came to be a profound and arguably dominant religious influence in East Asia.
Of course, the problem is that Indian Nationalism today is Hindu rather than Buddhist, which is a religious movement that India didn’t heavily export and which outside Bali didn’t have much staying power where it was exported, and which isn’t entirely home grown, even though much of it has local roots.
Similarly, the expansion of Y-DNA F people to become the predominant people of Eurasia (especially West Eurasia) is so remote and thinly attested archaeologically that it is hard to identify with those ancient hunter-gathers.
UPDATE April 4, 2018: I left this comment at Razib's blog:
“Also, the polytomy at the base of the eastern Eurasian human family tree, where all the major lineages diverge rapidly from each other, makes me wonder about gene flow vs. admixture. It seems possible that the polytomy may mask a phylogenetic tree topology which had gradually bifurcating nodes, if periodically a single daughter population replaced all its sister lineages in a local geographic zone. Much of history in human meta-populations may be characterized by isolation-by-distance and gene flow, erased by the extinction of most lineages and expansion of a favored lineage.”
This is a really interesting observation, and both possibilities, naively, seem pretty plausible.
It is almost unthinkable that there was complete genetic continuity and stasis in South Asia from prior to 65,000 BP to 6000 BP. All of the examples we have point to very sustained stasis as very rare, even though turnover may have happened less frequently among hunter-gatherers than among Holocene people at the continental level. The late modern history of hunter-gatherer tribes of Native Americans, for example, suggests that moderately long distance folk wanderings and exterminations of whole tribes were relatively common even on time scales of 1000 years or so.
There were probably at least two significant waves of migration and expansion after the one that gave rise to the Papuans in mainland Asia that greatly interrupted HG genetics there.
Y-DNA D people were probably a mid Upper Paleolithic Northern route arrival in Asia (although pre-LGM given Y-DNA D in the Andamanese) and were possibly male dominated (given that the Onge autosomally are close to AASI), that migrated to South Asia from Tibet to the South ultimately reaching Burma and the Andamans (since phylogeny-wise the Y-DNA D of India and the Andamans is closer to Siberian and Tibetan Y-DNA D than to Japanese Y-DNA D which splits at a very basal point from other Y-DNA D). Other Y-DNA D people take the Northern route to become the founding Jomon people of Japan. Most Y-DNA D people of Northern Asia in between are wiped out in the LGM.
Y-DNA C is remarkably rare and quite low if phylogenetic diversity in South Asia. This could simply mean that the coastal route theory for Y-DNA C is wrong, and that instead it took a clockwise northern route to reach East Asia, mainland SE Asia and Island SE Asia, and the lack of phylogenetic diversity of Y-DNA C in South Asia tends to support that reading of the data. But, another possibility, given the proportionately high level of C-M130* in South Asia relative to other Y-DNA C haplotypes is that Y-DNA C differentiated from Y-DNA CF in India, with lots of Y-DNA C people migrating east, but a few remaining, and that Y-DNA F people (including sister Y-DNA clade Y-DNA H people) subsequently wiped out most of the original Y-DNA C right population of South Asia, and that a lot of Y-DNA C people in India today are associated with a Y-DNA C1b1 back migration later in the Upper Paleolithic. The fact that autosomal ASI ancestry in India is pretty much proportional to Y-DNA C proportions in India, even though the proportions are low, also points to the antiquity of Y-DNA C in India, followed by later events.
In particular, the Y-DNA C people in India were probably marginalized by the expansions of Y-DNA F*, F1, F3 and H in India in the pre-Neolithic period, with other clades of F and daughter clades derived from F expanding into both West Eurasia and East Eurasia where the expanding clades became dominant. It is hard to know what gave the Y-DNA F/H people a decisive advantage over the Y-DNA C people in India and elsewhere, although forced to supply my best guess, I might suspect dog domestication or perhaps mastering how to turn wild grains into flour (flour predates the Neolithic revolution by at least ten or twenty thousand years).
It is also worth noting that ancient DNA suggests that in parallel with these developments in India, that Y-DNA C was once much more common than it is today in Europe, which definitely reflects Neolithic and Steppe driven replacement of remaining European HGs with Y-DNA C, but which may also reflect Mesolithic era replacement.
Similarly, in East Asia and SE Asia, Y-DNA O which is also a remote descendant of Y-DNA F, also sweeps those regions even before the Neolithic revolution.
If Indian nationalists want to discuss their basal and formative influence on the rest of the world, they would be well advised to de-emphasize the Bronze Age and to instead focus on how, on one hand, Y-DNA F is the dominant ancestor of modern Eurasian Y-DNA clades and that it probably originated in India (or at least had its first major expansion there), and how, on the the other hand, in the Iron Age, Buddhism, which also has its origins in India, came to be a profound and arguably dominant religious influence in East Asia.
Of course, the problem is that Indian Nationalism today is Hindu rather than Buddhist, which is a religious movement that India didn’t heavily export and which outside Bali didn’t have much staying power where it was exported, and which isn’t entirely home grown, even though much of it has local roots.
Similarly, the expansion of Y-DNA F people to become the predominant people of Eurasia (especially West Eurasia) is so remote and thinly attested archaeologically that it is hard to identify with those ancient hunter-gathers.
Wouldn't it be more plausible for L to be from IVC specifically rather than Iranian farmers?
ReplyDeleteNo. The conclusion that it probably comes from Iranian farmers is that substantial pool of ancient Iranian farmer DNA in the study.
ReplyDeleteNevermind, you're absolutely right.
ReplyDeleteI wonder what ASI's haplogroups were then. C and D only?
The biggest ASI Y-DNA haplogroup was probably H. https://en.wikipedia.org/wiki/Haplogroup_H_(Y-DNA)
ReplyDeleteY-DNA F*, F1 and F3 would be other strong candidates. https://en.wikipedia.org/wiki/Haplogroup_F-M89
Between them, they make up probably at least half of the Y-DNA in South Asia and their geography and caste presence is a decent match to ASI.
There is more Y-DNA C in ASI regions than in ANI regions, but it a really very sparse supporting the minority view that Y-DNA C made have made its way to the East via a Northern route, rather than a Southern one. https://en.wikipedia.org/wiki/Haplogroup_C-M130 If Y-DNA C did take a Southern route it was probably mostly replaced in the Upper Paleolithic era.
ReplyDeleteY-DNA D is almost surely late to the party. It very likely did take a Northern route (with the deep divergence between Japanese and non-Japanese Y-DNA D clades and the dates that Jomon habitation began in Japan limits the dates somewhat). In South Asia and Mainland SE Asia, Y-DNA D is closely related to that in Siberia and Tibet, and distant from that in Japan, and Y-DNA D is pretty much absent from every place except Eastern South Asia, Burma and the Andamans. So, in all likelihood, Y-DNA D in those places comes from migrants from the Tibetan refugia (there are temperate valleys there), down to the coast in places that were vacant of other intense human habitation to keep them out.
H does make the most sense. For some reason I was getting T and H confused.
ReplyDeleteThe story of Y-DNA T in India is a pet project of mine for which I need more data. I suspect that it may be the Y-DNA of the people who brought Sahel African domesticates to South India making the South Indian Neolithic possible and possibly providing a source for the Dravidian language.
ReplyDeleteT doesn't match the distribution of Dravidian very well though does it. The Sahel->Neolithic link I think is plausible but I'd bet on Dravidian being another Iranian Neolithic language.
ReplyDeleteIndian cows are present in Africa but not the reverse IIRC, so any exchange was likely two-way.
ReplyDeleteI'd be almost certain that Dravidian is not an Iranian Neolithic language (that discussion would deserve another post of its own). I wouldn't be surprised if there was a two way exchange.
ReplyDeleteThe T distribution matches the source location of the South Indian Neolithic and the inferred location of proto-Dravidian pretty well, just not the entire extent of it.