Most of the “East Asian” in East Bengalis is not from the Munda


I did some more data analysis. Added Tibetans, etc. Since some readers have more opinions than I do I’ll leave commentary up to them. Two notes

1) The “Northeast Indian” group includes populations like Mizos (I know that from the ID codes). They seem different from Nagas, who are more Tibetan

2) No idea why Bangladeshis are showing so much “South Chinese” signal in admixture. Perhaps it is artifactual, or, we’re missing some donor population? There is clearly some Munda admixture in a few individuals, but it doesn’t seem to be the dominant contributor of East Asian ancestry.

Hoping for some illuminating comments.

3+

23 Replies to “Most of the “East Asian” in East Bengalis is not from the Munda”

  1. A British anthropologist categorized us as “Mongolo Dravid”.
    https://secureias.com/racial-classification-of-india-by-herbert-hope-risley/
    But vast majority of Bangladeshis dont look very mongoloid though blood tests suggest a strong mongoloid influence in us. some look east asian for sure.Yes,I have relatives who look very mongoloid. But I dont think even more than 10 percent Bengalis in Dhaka look mongoloid. However, undoubtedly Bangladeshis are phenotypically diverse. If all Bangladeshis belong to the same cluster then why does this diversity exist?
    Also note that even Bengali Brahmins like Mamata Banerjee can look Mongoloid. Please do some research on whether some/many bangladeshis have Portuguese(via pirates?),Central Asian,west Asian and African ancestry or not. Because i have many relatives who look very Caucasoid and also very few African looking friends.

    Another thing i would mention is that biracial people(for example, Neil deGrasse Tyson ) sometimes can look very south asian. Why is this so?

    0
    1. Also,a hindu family i know is completely Mongoloid. i dont know if they are bengali or not. i think they are tribals. as far as i remember, their surname is “verma”

      0
      1. Wasn’t there a pre Ahom kingdom (in Assam) where the King’s title was Verma. In Hindu texts they were described as tribals ( before they joined Hinduism)

        0
        1. that surprises me. I was expecting like .7 or .8. I have to look into that paper. I wonder if the group used craniometric measurements for “African features.” It would be hard to be objective about facial phenotype without some sort of good measurement system.

          1+
        2. Shriver et al Hum Genet. 2003 Apr;112(4):387-99.

          We present details of a panel of 34 AIMs* and demonstrate how such studies can proceed, by using skin pigmentation as a model phenotype.

          In the two African population samples, we observed significant correlations between estimates of individual ancestry and skin pigmentation as measured by reflectometry (R(2)=0.21, P<0.0001 for the African-American sample and R(2)=0.16, P<0.0001 for the British African-Caribbean sample).

          *Ancestry informative markers (AIMs) are genetic loci showing alleles
          https://www.ncbi.nlm.nih.gov/pubmed/12579416

          =====
          Parra and Kittles, Nature Genetics 36(11 Suppl):S54-60 · December 2004
          We studied the relationship between pigmentation and ancestry in five populations of mixed ancestry with a wide range of pigmentation and ancestral proportions (African Americans from Washington, DC; African Caribbeans living in England; Puerto Ricans from New York; Mexicans from Guerrero; and Hispanics from San Luis Valley). The strength of the relationship between skin color and ancestry was quite variable, with the correlations ranging in intensity from moderately strong (Puerto Rico, rho = 0.633) to weak (Mexico, rho = 0.212).
          https://www.researchgate.net/publication/8209215_Implications_of_correlations_between_skin_color_and_genetic_ancestry_for_biomedical_research

          0
  2. Interesting PCA!

    Re: your selection of components – why both Tamil and Telegu? And which of those sub populations did use for them ie Tamil Brahmin (with a bit of steppe MLBA), Velama for Telegu (IranN + AASI)?

    I like the idea of three different east asian components – but does Han South not overlap alot with Cambodian and Tibetan?

    The Tibetan signal is relatively steady in Bengalis, but the Han S:Cambodian seems to fluctuate wildly. Which Bengali samples did you use? Is this a regional difference? Or just noise and components bleeding into one another?

    I commented on this more in the last post, but I had always imagined the E Asian in Bengalis to be something like – proto Bengali + x munda/AA + y TB the further East they moved.

    Even on this new PCA, some Bengalis are shifted towards the Munda – S Indian cline. But the main direction is towards Garo (local TB group) suggesting a bigger component.

    The problem with modelling with the Munda is that they are something like 25% SE Asian 75% AASI. In my mind, comparing Bangladeshis to local Indic populations like Biharis, we couldn’t have packaged a 1:3 SEA:AASI to explain all the E Asian anyway.

    Historically, there is mention of indigenous groups called Kirat in the region – do they have modern day descendants? Could they represent a missing vector instead of Munda, with Khasi and other now neighbouring TB groups more recent phenomena that Bangladeshis admixed with?

    Just speculation.

    1+
  3. Dear Razib,
    I have been an avid follower of yours for several months and have read lots of your articles, specially on the Bengali Genetics. Thank you for such amazing information on a topic which is of great interest to me.
    I am from Sylhet, Bangladesh. My earliest known male ancestor, named Hazi Yousuf came her with non other than the famous sufi saint Hazrat Shah Jalal (RA) himself. My family still maintains and operates his tomb and the mosque located in Sylhet. I am the 25th generation descendant from Hazi Yousuf, who was a companion of Shah Jalal (RA). He assisted the Saint while he lived and after his death, Hazi Yousuf was the first in line of my ancestors who looked after his tomb complex. He supposedly married a Brahmin Hindu woman from Sonargoan near Dhaka or a relative of the Sultan Shamsuddin Firuz Shah. My own mother is descendant from Syed Nasiruddin. supposedly a Turk from Delhi, the commander of the force of 360 pilgrims who followed Shah Jalal to Sylhet.
    This is what I have known all my life as I grew up. However, none of it can be proved with 100% accuracy as I have no genetic records to prove a Yemeni, Turk or Bengali Brahmin ancestry. I would love to know my genetic ancestry if possible.
    I wish there was a way I could connect with you and somehow delve more into my genetic history which would satisfy my own interests and will probably help in the ongoing research on Bengali Genetics.

    1+
  4. I wish lot of genetic ancestry research could be done in Bangladesh. It has become cheap enough and also most Bangladeshis are not too hung up about where did their forefathers came from. Maybe until 30-40 years ago, Middle Eastern, Persian ancestry was something to be ostentiously proud about but social undercurrents have changed a lot. Middle East and by default the Western parts of SOuth Asia are not seen as neither good Muslims nor successful societies. Also, construction of a proud Bangladeshi identity have advanced a lot. So, people will not be too alarmed to know where his/her forefathers came from.

    A personal observation was that South East Asian features (I dislike words Mongoloid, Caucasoid, Ne… ) are more prominent among Hindus of Bangladesh than Muslims. Admittedly, most of the hIndus left in Bangladesh after decades of migration to India are from lower castes and poorer sections. However, it would be really interesting to know how much structural differences poor Bangladeshi Hindus and Muslims show after living side by side for many hundreds of years.

    2+
    1. That’s an interesting point.

      Alot of the higher caste Hindus have left Bangladesh, so maybe more representative of lower caste Hindu groups vis-a-vis a broader spectrum of middle class Muslims.

      Bengali Brahmins clearly do score alot less E Asian, but I don’t think that would be true for non-Brahmins.

      I do think there is more structure to Bangladeshis (albeit on a smaller scale) – would be really interesting to tease out what is geography, social stratification and religious affiliation.

      1+
    2. “A personal observation was that South East Asian features (I dislike words Mongoloid, Caucasoid, Ne… ) are more prominent among Hindus of Bangladesh than Muslims.”

      @shafiq I came across many Bangladeshi expatriates, most of them looked pan south Asian, especially Gujarati and Upites. Some also looked Punjabi and south Indian. A few even looked central Asian turanid. I don’t know what %age of Bangladeshis have east Asian influenced features, perhaps 15-20%. Here is a study of nasal parameters between Bengalis and Chakmas:
      https://www.researchgate.net/publication/322070500_Study_of_Nasal_Parameters_Between_Bengali_and_Chakma_School_Going_Children_of_Bangladesh

      The result revealed most Bengalis have leptorrhine/Narrow nose(definitely western eurasian type) but the chakmas have mesorrhine nose.

      0
    1. Dear Razib,

      May I know more details on the Sample Collection plan in Bangladesh. Is it anyway possible for me to be a part of it? thanks

      0
  5. I do think there is more structure to Bangladeshis (albeit on a smaller scale) – would be really interesting to tease out what is geography, social stratification and religious affiliation.

    i think there is structure geographically. but i don’t think there is structure on social stratification as much. that looks like endogamous communities, and i don’t see that in the 1K bangladeshi samples (the exception are a small # of individuals collected together who lack east asian)

    1+
    1. Agree that geography would trump other aspects. As you say, no endogamous communities as such. But within Hindu non-brahmin communities? The Muslim population is probably inherently more mixed given a less segregated class system and conversion from all backgrounds.

      The 1k samples are essentially blinded in terms of background bar 4 Bengali grandparents (which may not actually have been the case given your identification of the ?SC outliers). They could represent a similar socioeconomic (and geographical) background – though I accept that similar PJL sample collection was the opposite in terms of homogeneity.

      It would useful to use other Bengali samples to compare. Academic and otherwise. It’s the regional differences I’m most interested in.

      1+
  6. All this new bengali commentators on the post just shows that Bangladeshi folks (of Brown pundits) were just lurking while we (Indians and Pakistanis) have been going hammer and tongs at each other.

    Is this how you folks are trying to be the best country in S-Asia? LOL. Smart strategy.

    3+
  7. @Razib Is there an Iran/IVC-Steppe cline in Bangladesh? I’ve noticed some Bangladeshis have higher steppe, but lower Iranian and vice versa although overall western Eurasian was still the same. The same goes for AASI and east Asian. Perhaps Western Bengalis have more Munda type east Asian and AASI and less East Asian than eastern Bengalis? Does the same apply for Odias and Assamese?

    0

Comments are closed.