Admixture (Ref3 K=11) HRP0121-HRP0130

Here are the admixture results using Reference 3 for Harappa participants HRP0121 to HRP0130.

You can see the participant results in a spreadsheet as well as their ethnic breakdowns and the reference population results.

Here's our bar chart and table. Remember you can click on the legend or the table headers to sort.

If the above interactive charts are not working, here's a static bar graph.


  1. Zack, thank you for putting out the results earlier than planned, I was eagerly awaiting the Keralaite Nair's results. His figures seem fairly typical for a South-Indian Brahmin, at least the Onge fraction does. I am not in a position to state to what extent the European component in these Ref3/K=11 runs comparable to the NEU in the standard K=10 analyses of Dodecad and the European component in the old HAP K=12, as you are the one putting the figures out there for interpretation, so feel free to weigh in. The old K=12 European seemed to be more exchangeable with the component that is usually modal among Northern Europeans, whereas this one seems more generically European-specific, albeit with a North to South cline (in Europe). In the Dodecad K=10, the Kerala Nair had no Northern European admixture, which is otherwise found among South-Indian Brahmins at around 4-7% IIRC. He did however, and surprisingly so, have Southern European admixture. Over here, their SW Asian (9%) is slightly more than their European (7%) component. This is consistent with Thorfinn's theories that the Dravidians were most likely a case of elite dominance and were exemplified mainly by the South-West Asian component, while the Aryans an eclectic mix of European and SW Asian. I mention this, as some consider the Nairs a socially dominant but traditionally Shudra caste. Some others, Kshatriya warriors. The original Nairs were said to be a military body, holding lands (along with the Namboothiri Brahmins in Kerala) and serving as a militia, but the present day Nair castes includes persons who, by hereditary occupation, are traders, artisans, oil-mongers, palanquin bearers, and even barbers and washermen. Hence their caste identity is unclear, per se. Let us assume for now they are a socially dominant, traditionally Shudra caste who were the elites of the area prior to the advent of Brahmanical newcomers from the North. There is one thing to consider - some of the South Indian Brahmin participants are slightly more SW Asian than European (HRP72 and HRP67 for example). Even the difference between EU and SWA is marginal for some North-West Indians; and the Sindhi reference data-set is more SW Asian than European. I am a little skeptical as to what extent we should use this specific level of K to make robust inferences about the ancestral components of the ancient groups in question. I recall Razib expressing his concern for something similar i.e a selective pattern specific to this level of K. More specifically, the Ref3K=11 runs don't seem to have components which are specific to i) West-Asia ii) Northern and Southern Europe. The first two are the main exogenous West-Eurasian elements in South-Asia - would you consider furnishing a new (perhaps a test-run) ADMIXTURE exercise, incorporating references for the three aforementioned components, but retaining the Onge component? I am sorry if I have misunderstood the way ADMIXTURE works and if this can't quite be done with ease.

    There are some other interesting participants. The Wikipedia page for Gujarati Ghanchis says that there are both Hindus and Muslims in this community. Many of their clan names seem very reminiscent of Rajput clan names (Solanki, Parmar, Chauhan, Rathod, Halai; etc), implying an overlap of some sort, just as the Rajputs do with the Gurjars and the Jatts. The Ganchi participant seems more similar to what you label Gujarati-B than the typical Gujaratis (-a).

  2. Another interesting aspect is the European-SW Asian scores of the Uttar Pradesh Brahmins. They seem to be heavily biased towards EU as opposed to the SWA component..

  3. what about the genetic impact of brahmins on nairs through their consort-system?

    • This doesn't seem implausible on an individual level for HRP124. His top matches on 23andMe's Compare Genes tool are all almost exclusively South Indian Brahmins (most being >76%), with a Namboothiri Brahmin at this top match. I don't think that is a mere coincidence.

  4. Thank you for these fascinating results!

  5. There is a Khatri surname (a clan name, rather); Nayyar or Nayar which also happens to be borne by Rajput Kshatriyas. OP Nayyar, famous music director who was of Punjabi origin was a Nayyar Khatri. Perhaps the presence of this clan/surname is an indicator of the south Indian Nair caste's potential Vedic origins?