search this blog

Friday, January 22, 2016

Y-HG J2 has a deep and complex history in South Asia

Open access at Nature Scientific Reports:

The global distribution of J2-M172 sub-haplogroups has been associated with Neolithic demic diffusion. Two branches of J2-M172, J2a-M410 and J2b-M102 make a considerable part of Y chromosome gene pool of the Indian subcontinent. We investigated the Neolithic contribution of demic dispersal from West to Indian paternal lineages, which majorly consists of haplogroups of Late Pleistocene ancestry. To accomplish this, we have analysed 3023 Y-chromosomes from different ethnic populations, of which 355 belonged to J2-M172. Comparison of our data with worldwide data, including Y-STRs of 1157 individuals and haplogroup frequencies of 6966 individuals, suggested a complex scenario that cannot be explained by a single wave of agricultural expansion from Near East to South Asia. Contrary to the widely accepted elite dominance model, we found a substantial presence of J2a-M410 and J2b-M102 haplogroups in both caste and tribal populations of India. Unlike demic spread in Eurasia, our results advocate a unique, complex and ancient arrival of J2a-M410 and J2b-M102 haplogroups into Indian subcontinent.

Singh et al., Dissecting the influence of Neolithic demic diffusion on Indian Y-chromosome pool through J2-M172 haplogroup, Scientific Reports 6, Article number: 19157, (2016) doi:10.1038/srep19157


Nirjhar007 said...

Thanks for posting this.

Gill said...

I've mentioned it before here I think. Check out the YFull J2 tree and look at the TMRCA estimations for when the popular branches diverged. There's a flurry of activity around the Neolithic and then around 5000-6000ybp, and then the ~3000ybp mark representing the birth of the "modern" post-IE tribal order.

And then look at the history of pre-Harappan and Harappan civilization on Wikipedia.

I must say though, looking at Y-STRs for J2 in South Asia is completely useless. These branches within South Asia separated upwards of 5000ybp, how could Y-STR even begin to address the challenge (FWIW, none of the STRs fit mine... I'm YF02959 on YFull's tree under J-Y978*).

At least with SNPs, there are two major clades... Z2437 and Y978. And under Z2437, the Z2443 branch (in GIH and PJL) has had a Saudi and Malay turn up, and is probably the main clade for the Pakistan area, including likely for Kalash/Burusho/Sindhi/Pathan/Afghans.

Imagine if they bothered to test for just those two SNPs.

Nirjhar007 said...

There were two activities, take the note :
1. Around 4500-3800 BC when the Chalcolithic really started to emerge in Mehrgarh with signals of New peoples arrival, most likely from N Iran area.

2. Around 600 BC, which was most likely triggered by Iranians again.

So, to really show that there were movements between those periods into the Subcontinent which can be considered significant, the only way is aDNA.
So lets hope that the Haryana aDNA will tell us with most accuracy, what really happened.

Martin Clifford Styan said...

I already looked at Singh et al.’s article on J2 in India after it was mentioned in earlier comments on this blog a few days ago. Around a year ago I spent some time studying J2 mainly on the basis of the 111 STR haplotypes given in the Family Tree DNA projects. I copied a large number of these into a file on my computer and attempted to sort out the different branches mainly on the basis of a selection of 34 of the most stable STRs. J2 is probably the most complex and deeply divided of the major West Eurasian Y-haplogroups. As with Y-haplogroups in general, the commercial testing organizations and their amateur clients have done great work in identifying numerous sub-divisions, but they have not provided good information on their geographical distribution. On the other hand, the supposedly professional academic researchers have tended to ignore the discoveries of the commercial and amateur sectors, and to simply repeat what was done in the early days of Y-chromosome research 10-15 years ago. They provide good information on geographical distribution, but only for a very limited range of groups. Singh et al.’s paper is an outstanding example of this tendency. Thirteen SNPs and 17 STRs were tested. This means that the SNPs for most of the known branches of J2 were not included and the range of STRs is not sufficient to add much. In spite of this, they have still provided some very useful information.

J2a1b M67 is not found in India, although it forms the majority of J2a in much of the Middle East and Europe.

J2a1a M47 was not found by this study, but one example was found from India in a previous study. This group is found in various Middle Eastern countries (en.wikipedia).

J2a1h L24 is also one of the main branches of J2a in the Middle East and Europe. The FTDNA India Project has nine 67 STR J2a haplotypes from India and Pakistan. From these 4 are directly placed in J2a1h L24 and a total of 8 have DYS 450 = 9, which generally distinguishes this group from other branches of J2a. However, three of these appear to come from the same family. Unfortunately, Singh et al. did not test for J2a1h L24. They tested for J2a1h1 M158 and did not find it. However, it appears from the yfull tree that members of J2a1h L24 mostly belong to J2a1h2 L25 and not to the M158 branch.

A good point in the study is that they tested for Z2396 and found it in a considerable number of Indians. The yfull tree reveals that Z2396 is an alternative to PF5197. These two SNPs identify an important brother clade to J2a1h L24. According to the FTDNA J2 Project, the L24 and PF5197 are united by the SNP Z2221. The Z2221 group is the brother clade of the PF5116 group, which unites most of the rest of J2a, including J2a1b M67 and J2a1a M47. I found a number of Middle Eastern and European examples of the PF5197/Z2396 group in the FTDNA projects. This group is itself quite deeply divided. SNPs have been found for some of its sub-clades and there are significant differences in the figures for the more stable STRs. Without the SNP evidence, I think it would not be clear that these sub-clades are more closely related to each other than to other members of J2a.

Singh et al. found a number of examples of J2a1c M68 especially in the south. My collection of J2 haplotypes includes only one example of J2a1c M68 and it comes from Iraq. However, the FTDNA J2 Project includes the M68 group in a larger group identified by the SNP Z6058. This group is a brother clade to a group identified by PF5087, which is divided by PF5116 and Z2221 into groups which include the majority of J2a as mentioned in the last paragraph. I collected 23 examples of the Z6058 group from the FTDNA projects. They come from the Arabian Peninsular and from various European countries from Portugal to Russia. They have fairly consistent STR figures, distinguished from the rest of J2a especially by DYS 490 = 13 (other branches almost always have 490 = 12).

Martin Clifford Styan said...

Both the basic branches of J2b are found in India and with different distributions. J2b1 M205 is found mainly in the north-west, especially in Rajasthan. J2b2 M241 is found mainly in the east, including among the Austro-Asiatic tribals, but also in the south. J2b2 M241 is also common in the Balkans, especially among the Albanians.

Is it possible that the branches of J2 found in India (J2a1h L24, J2 Z2396/PF5197, J2a1c M68 and J2b) are associated with the Caucasus Hunter Gatherer admixture component, while the J2 branches found in the Middle East but not India (especially J2a1b M67) are associated with the South-West Asian component?

mooreisbetter said...

Oh, I love this line:

Contrary to the widely accepted elite dominance model

Hope folks are paying attention.

According to my models, not only J2, but also R1b spread with much more mundane mechanisms.

And I am confident that serious scientific papers will put an end to the pseudoscience of elite dominance models, within the next two years.

More at

Gill said...

The founder effect among J2b2-M241 in East India along with the relative scarcity of it back in the Indus region, where it presumably once was more common, can possibly suggest an exodus to the east after the collapse of the IVC. This could also represent the tribal origin of modern Austroasiatic/Tribals in Southeast India.

Oh and over on the European L283 side, there's a slew of Italian samples from different subclades, some of which split apart around ~5000+ya.