Genetic odds & ends

At my other weblog I report on evidence that a sample from Cambodia dated to 100 to 300 AD seems to have considerable Indian ancestry. This is not a result in isolation. Lots of evidence points to non-trivial Indian gene flow. The devil is now in the details of when/who.

Second, there is lots of talk about “person X looks like population Y, so perhaps they have ancestry from population Y.” This is almost certainly wrong in most cases.

Looking at Indian populations there tends to be far more variation in physical appearance within a population than the variation of total ancestry. In other words, some Tamil Brahmins look like South Indian Tribal people and other Tamil Brahmins look like West Asians. But in terms of total ancestral components, there’s no difference.

The theoretical explanation for what’s going on is that the genetic loci which control “physical appearance” are much smaller in number than the whole genome (on the order of dozens of loci). As such, the sample variance is rather large (the N denominator is small).

South Asian populations differ across each other, but there is usually a quite large within-population variation on genetic variants implicated in physical characteristics. This means that there are a large range and quite a bit of variation.

Though a lot of the discussion involves Muslims, I have heard from multiple non-Muslim people of Northwest Indian stock (e.g., Pandits) that they must have “Persian ancestry” because they look so Persian. The genetics refutes this rather strongly. Rather, modern Persians and many Northwest Indians share deep ancestry which diverged after the Last Glacial Maximum 20,000 years ago.

0

49 Replies to “Genetic odds & ends”

  1. Indians from the subcontinent migrated to South Asia and set up Hindu kingdoms and certainly mixed with the native poulations of the region. Hence the gene flow.

    1+
  2. Off topic question:
    Considering that majority middle eastern people are fair skinned (and caucasoid) despite the sunny hot desert environment, are the people from middle east originally people who migrated from Europe? And vice versa?

    0
    1. I think Egypt could be a good example. If you look at the lot of Pharaoh’s time imagery i think they are described as brown-ish. But post Alexander, Rome and Mamluks (slaves from central asia,Armenia etc ) , their gene pool is all mixed up with Caucasian folks. Just my guess.

      1+
    2. Only majority of Levantine Arabs, folks from Transcaucasia and Iran can be considered light-skinned; gulf Arabs are quite Dark, I travelled a few gulf countries and noticed a fair share of local Arabs had my skin tone(medium brown) and the majority have my dad’s skin tone(olive).
      The light skin was brought to Europe by Anatolian farmers as Mesolithic Europeans were dark-skinned. My guess, the light skin evolved in Caucasus mountain range, Anatolian plateau and Iranian plateau?
      DaThang in other post analysed that Gulf Arabs like Saudis have a considerably higher Natufian, a notably lower Anatolian(aka Early European farmer), slightly lower Caucasus_HG ancestry, and roughly similar Iran_N ancestry compared to Levantine Arabs. In my opinion, Natufians and Iran_HG weren’t as light-skinned as Anatolians.

      We can continue here the discussion from the other post:
      “Then you must be Sylheti. I saw 7 or 8 individuals in Sylhet who looked completely non-south asian.”

      No, I’m not a Sylheti. Allover Bangladesh you can find some super west Asian looking individuals.

      “So, the logical deduction is that some Sylhetis have some west asian ancestry from west asians who settled there.
      Once Tsar Razib Khan also said that he found some west asian ancestry (may be he was talking about some sylhetis)”

      I dunno what are you talking about. All the Sylheti samples in South Asian genotype project look regular Bengali. Only one individual scores 2% Yemenite jew, which could be noise. Let’s assume a few Sylhetis have 2-3% Arab from XII century missionaries, that definitely not going to affect phenotypes rather the Iran_HG and Steppe combined make 50-60% of Bengali genome, so those west Asian look more likely is from local genes. Mind you Bengali Brahmins are more ANI shifted than any regular Bengali/Sylheti, but you’ll find more west Asian looking Bengali Muslims than brahmins, why? Cuz Muslims are way more numerous than brahmins.

      The Bengali woman in this video has a typical Bengali look:
      https://youtu.be/8SLKgtHOkYo
      Yet a few people in comment section compared her look with Israeli actress Gal Gadot. Not surprised that some people can observe west Eurasian traits even among mixed people.
      Overall, I think most Bengalis shows 60-70% Iran_hg/Steppe/Cro-Magnon-like traits, 20-30% AASI traits and 10-20% East Asian traits. So, the total combination of these distinct traits produces a Bengali look.

      “Yes, possible. But if that is the case then groups with much higher Iran HG should exhibit west asian(persian?) phenotypes in higher amount.”

      Not necessarily, but that’s true for most cases. NW Indians have more west Asian looking people than Peninsular Indians as both groups are big, but there’s more west Asian looking Bengali Muslim than Brahmins, and I explained the reason.
      In the case of Gujaratis, they aren’t homogeneous, but most of the time have similar phenotypes, and their situation might be similar to Bengali one.
      Iran_HG is quite high even in South Indian middle castes, but their steppe is minimal and has higher AASI than NW Indians. I noticed mid-caste south Indians sometimes looks Sindhi or even Balochi people.

      1+
      1. @Son Goku
        >Overall, I think most Bengalis shows 60-70% Iran_hg/Steppe/Cro-Magnon-like traits, 20-30% AASI traits and 10-20% East Asian traits.

        The tentatively hypothesized Levantine cro-magnon is supposed to be a part of AASI itself, I wouldn’t make any claims about a specific amount yet. Looking at only Iran HG, steppe and AASI ancestries, plain old global25 gives me this:
        “sample”: “Bengali_Bangladesh:Average”,
        “fit”: 3.7925,
        “CustomGroup_Simulated_AASI”: 55,
        “IRN_HotuIIIb_Meso”: 23.33,
        “RUS_Sintashta_MLBA”: 21.67

        Adding east Asian:
        “sample”: “Bengali_Bangladesh:Average”,
        “fit”: 2.3553,
        “CustomGroup_Simulated_AASI”: 49.17,
        “RUS_Sintashta_MLBA”: 25,
        “IRN_HotuIIIb_Meso”: 18.33,
        “Dai”: 7.5

        Using more recent, composite ancestral inputs gives me this:
        “sample”: “Bengali_Bangladesh:Average”,
        “fit”: 1.2726,
        “IRN_Shahr_I_Sokhta_BA2”: 50,
        “CustomGroup_Simulated_AASI”: 25,
        “RUS_Sintashta_MLBA”: 15,
        “Dai”: 10
        15% Sintashta and an unknown amount of Iran HG in BA2 since the new BA2 is old BA2 + old BA3. Old BA2 was 18% AASI and old BA3 was over 40% AASI and I don’t know the number of samples in old BA2 and BA3 so I can’t really approximate the AASI in new BA2. Assuming it is equal, the result would be new BA2 being 30% AASI and 70% Iran HG. So Iran HG ancestry in Bengali Bangladesh would be 35%
        10% Dai + 15% Steppe + 35% Iran HG + 40% AASI.

        2+
        1. Yeah the AASI must capture the hypothesised cro-magnon type Palaeolithic west Asian. Something we gonna discover in the future. By AASI traits I meant only australoid-like traits.
          I would add 2-3% more east Asian lowering the same %age of AASI for East Bengalis, for west Bengalis that looks good ATM.

          0
      2. No. I have never seen the phenotypes in Dhaka or elsewhere that i saw in Sylhet. (Their skin colour,face,nose etc)
        But the vast vast majority of Sylhetis didn’t look like that. So, majority Sylhetis may not have west asian ancestry. I think those features were Persian. North indians or even pakistanis dont look like them.

        Since such phenotypes cannot be seen outside of Sylhet, so most probably those features were the result of west asian ancestry

        0
        1. I have never heard of such thing before. I’ve been to Sylhet two times and never noticed anything like that. Didn’t noticed in British Sylheti community either. Some light-skinned sharp-featured Sylhetis I know look South Asian first but can pass as west Asian too. But how possible that there are a few Sylhetis that don’t look South Asian but only west Asian? The simple answer would be they aren’t Sylheti at all or only part Sylheti; they must have lots of recent west Asian ancestry then. Do you personally know them? If yes, you might convince them to take a genetical DNA test.
          The lightest person in my family is one of my maternal uncles, and he has the same skin tone as Qasem Soleimani, so obviously, Iran_HG or Sintashta genes can produce such skin tone in Bengalis. The nose shape and types in both west and south Asians very. As I said before it’s the AASI genes that make south Asians distinct from West Asians.

          1+
      3. Agree with most of your phenotypic commentary bro but Bengali Brahmins do not have the clear E Asian shift most Bangadeshis and WB mid and low castes have and IMO Bangladesh and WB mid/low castes are overall definitely less ‘West Asian’ looking. You need to see more Bengali Brahmins. They overall look like West UP people which matches their genetics, with some NW types but overall they are very distinct to Bangladeshis and ive seen quite a lot of the latter and even been to a couple of Boishaki melas

        0
    3. The middle eastern populations come from varying amounts of the following ancestral populations:
      -Natufians
      -Anatolians of the neolithic
      -Iranians of the neolithic
      -CHG
      -Varying amounts of both sub saharan and steppe/Indo-European ancestries (generally low in both cases, but the steppe component is something around 10% to 20% in modern Iranians).
      A picture is worth a thousand words so here you go: https://imgur.com/a/bka4DaT

      Caucasoid traits would have come from either west Eurasians or a combo of west Eurasians and basal Eurasians. So it would be fairly fundamental for the middle east going back to well before the neolithic. The spread of light skin into the middle east on the other hand would have mostly been an Anatolian thing as far as the A5 light skin variant is concerned. Early Iranian cereal farmers like those from Tepe Abdul Hosein weren’t as fair skinned as Anatolian farmers from Boncuklu. IIRC even Ganj Dareh pastoralists were fairer than the Abdul Hosein farmers, so idk exactly what was going on over there.

      3+
      1. Iranian Bandari has the highest Iran_N and Sintashta among west Asians, not surprised that they sometimes get compared with South Asians. They’re also the darkest among Iranians(excluding Sistan-Balochistan) IIRC.

        1+
    1. The ancient DNA is from late dynastic and hellenic period though. I think that early to mid dynasty Egyptians would probably have more Natufian-like ancestry than less of the recently intrusive Anatolian + CHG ancestry of Levant.

      0
        1. Please do sir, regardless i am least concerned on “where we came from”

          We are here to stay, and to rule India (thru of course, our King in the North, Monsieur Modi) 😛

          0
          1. put it this way – Modi – who likely genetically scores/similar to a Patel (assuming more AHG% than a Patel since he is lower caste), scores more like me/has more ancestors in common with me, a high Inpe South Indian assuming you’re not a non-Brahmin from W (coastal Maharashtra, Konkan, Gujarat). If you are though, in that case we’d all share high levels of InPe ancestry not seen in those ratios in the Gangetic region (excl Haryana and former IVC geographic spread areas)
            The irony is Modi is probably one of the most genetically relatively unmixed IVC descended individuals than most in South Asia yet unknown to him, he buries the story of his Rakhigarhi ancestors to protect Indo-Aryan high caste culture which his ideology believes in
            as the glue for South Asia; to add to that he is a low caste who historically would have been discriminated against and excluded from Aryan oral high culture education that say even Dravidian Brahmins like Shankaracharya were entitled to for the most of post IA migration South Asia history. Im not sure if his community has a gotra system, traditionally a method of IA paternal lineage awareness and preservation

            0
          2. “The irony is Modi is probably one of the most genetically relatively unmixed IVC descended individuals than most in South Asia yet unknown to him, he buries the story of his Rakhigarhi ancestors to protect Indo-Aryan high caste culture which his ideology believes in
            as the glue for South Asia;”

            Well isn;t it truth (glue ) really? I mean what else do parts of India’s mainland have in common apart from their religion? Why would a Tamil Nadu coexist with a Bihar?

            On Modi being un-mixed, i highly doubt that, but still people who are even more “un-mixed” than Modi (ie S-Indians, E-Indians) made the religion of the Aryans their own. So i am not sure who “buried” their ancestors more.

            0
          3. On Modi being un-mixed, i highly doubt that, but still people who are even more “un-mixed” than Modi (ie S-Indians, E-Indians) made the religion of the Aryans their own. So i am not sure who “buried” their ancestors more.

            I think it is the other way around, the Indo Aryan incorporated much of the AASI beliefs.

            Take the Rg Veda as unadulterated Indo-Aryan.
            The Rg veda has no mention of Shiva, Kali Ganesh (I doubt there were elephants on steppe) and other animistic, totemic gods.

            Add to that caste marks on body, face forehead that look no different from Australian Aborogine markings for corroboree or Africans face and body paint, clinches the argument.

            1+
          4. If you know much about Hinduism the glue is certainly not Brahmin/high-caste culture, mass-adopted to lower castes now as Hinduism via recent religious innovations such as the Bhakti movement whereas previously they were exclusive to Dvija castes. ie if your ancestors were not Kshatriya or Brahman they would not even know what Vedic texts contained so whats left that’s actually authentically Vedic Hindu for the majority of North Indians (non-Brahmin/Rajput)? Aryan influenced folk religion picked up by the umbrella defined as Hinduism mixed with preexisting native tradition

            Aryan culture at the Indo-European stage has adopted several IVC and other native traditions upon entering South Asia. Krishna, Shiva, even Indra (BMAC) have no cognate in other Aryan IE religions proving they were picked up on the way, nothing even Indo-Iranian stage about them, check the Avesta. Further to your assumptions on Hindu purity Vegetarianism and other innovations such as Ascetism were likely picked up and copied into mainstream priestly Hinduism from non-Vedic/Aryan origin Agamic/Sramana traditions such as Jainism/Buddhism as they are missing from the Vedic religion: https://en.wikipedia.org/wiki/Historical_Vedic_religion
            If youre Gangetic, why dont you pay respects to the native HG tribals in your area who resemble Jharkhand tribals (haplogroup ydna H and mtdna M have a large prevalence in the Gangetic plains as in the South and IVC ancestry is relatively lower compared to the rest of South Asia) that you are genetically the most related to (likely 2-3x your Aryan ancestry) and worship their culture? A UP Brahmin is only 30% or so actually Aryan (IE) with much lower stats for lower castes.
            Re Modi genetics, South Brahmins generally have higher IE ancestry than Patidars and lower castes from the Ganges, so do Eastern Bengali Brahmins so genetically its more nuanced. Some Patel results come out as 0% steppe which is astonishing for a IA speaking group. Modis largest ancestral group infering from Patel results with a more South shift is certainly something that resembles Narasimhans Indus Periphery ie the urban farming civilisation who were native to NW South Asia who were a mixture of Iran N and AASI. He would be distinct to both Gangetic (from Haryana east & South) and South-Indian low castes due to his high Iran N/IVC ratio and his lower steppe ancestry compared to Gangetic and NW castes.
            Basically read Narasimhans paper and the history of Hinduism

            0
          5. A Kerala/Karnataka Brahman who practices Srauta traditionally is more ‘Aryan’ culturally than the majority of non-Brahman North Indians and likely genetically as well

            0
  3. I would really like to have a word* with the “Pandits” (I presume you mean Kashmiri ones) who posit Persian ancestry. I have to say this is *really* odd in my own experience. But then jokers exist everywhere.

    (Btw what I won’t be surprised about in Kashmiri Pandits is E Asian ancestry. Another reason why Nepal is a good modern analogue of what Kashmir would’ve been like without Islam)

    [*] Just kidding, couldn’t care less.

    1+
    1. .”what I won’t be surprised about in Kashmiri Pandits is E Asian ancestry. Another reason why Nepal is a good modern analogue of what Kashmir would’ve been like without Islam”

      May be this is due to a substrate ancestry from Khas tribes that Kashmiris, Nepalis and many other Himalayan people share?

      https://en.wikipedia.org/wiki/Khas_people

      I don’t know much about Khas people other than what is available in wikipedia. But it will be interesting to know more about these people. They are probably the source of all east asian ancestry in himalayan tribes.

      3+
  4. “I report on evidence that a sample from Cambodia dated to 100 to 300 AD seems to have considerable Indian ancestry. This is not a result in isolation. Lots of evidence points to non-trivial Indian gene flow. The devil is now in the details of when/who.”

    Must be centuries of elite Indian injection primarily from the south, going by Khymer art, architecture and the pallava grantha script.

    The Brahmin element must have been disproportionate. Again this is attested in the local history, particularly of the royalty who have strong links with Brahmins and Brahmin related dynasties like the Pallavas.

    Many of the Khymer kings were Sanskrit and vedic scholars.

    https://en.wikipedia.org/wiki/Banteay_Srei

    https://en.wikipedia.org/wiki/Yajnavaraha

    The Kaundinya origins of the Khymer kingdom tallies with the Kaundinya gotra Brahmins also prominent in the south.

    The Angkor region is truly the pinnacle of Indic religious architecture, far surpassing anything in India or the rest of asia (or indeed the world).

    It is actually awe inspiring to see such an understated nation create such magnificent temples.

    Truly the peak of Indian diaspora influence in terms of art and architecture.

    0
  5. Razib,
    I have seen that PANTA KAPU( telugu speaking) show a lot more indus derived ancestry. PANTA literally means crop in telugu, as in growing/protecting crops. Could that be something related to origin of farming in indus Valley? There are about 10 different kinds of KAPUs and all of those occupied (s) some important role in the village. Many KAPUs use Reddy, Naidu titles. It is told that some KAPUs diverged to form different castes like KAMMA, VELAMA and am not sure how true it is.

    0
  6. Razib writes – “Though a lot of the discussion involves Muslims, I have heard from multiple non-Muslim people of Northwest Indian stock (e.g., Pandits) that they must have “Persian ancestry” because they look so Persian. The genetics refutes this rather strongly.”

    I would associate this with the desire of North Indian elites who were ruled for centuries by fair skinned (Muslim) elites who were mostly Turko-Persian-Afghan. Being ruled by fair skinned “Persians” may have contributed greatly to skin color obsession that so many desis suffer from.

    1+
    1. Right on, this ‘white disease’ was no wonder propagated by ‘turko-iranians’. I say ‘turko-iranians’ instead of ‘turko-persians’ because Persians are ‘less white looking ‘ compared to other Iranian ethnicities. You can see that in Bollywood today. Usually Muslims are proud of their ‘white skin’. However, it was only exacerbated during colonialism. My pakistani friend couldn’t believe that Dr Sivan (ISRO chief) could exist in India.

      0
      1. “My pakistani friend couldn’t believe that Dr Sivan (ISRO chief) could exist in India.”

        In what way? Was he shocked that Dr Sivan was not cleaning the toilets like the people who look like him in Pakistan?

        3+
  7. I would disagree.India colorism has a longer past than turko mongols. Areas which did not have turko mongol rule also have this preference.

    1+
    1. White supremacy in Mahabharata:
      https://www.sacred-texts.com/hin/m12/m12b015.htm#fn_62
      “…The complexion the Brahmanas obtained was white; that which the Kshatriyas obtained was red; that which the Vaisyas got was yellow; and that which was given to the Sudras was black.”

      In Bangla language, a word for fair skin is “gouro-borno” which literally means “skin colour of the master”. This word is most probably derived from Sanskrit.

      So, i think that this fair-skin supremacy in this subcontinent is connected to Aryan arrival.

      PS: I have a Bangla Mahabharata which is very different (and short) from the online Mahabharata

      3+
      1. LOL, what’s up with every ethnicity getting their own Mahabharata.? Like I heard Tamils have some 100 different types of them. And that is the land of Periyar. Dont tell me, like the revisionist Bengali Ramayana , Duryodhan is some sort of tragic Hector type character there .

        In hindsight, we should have copyrighted the damn thing. 😛

        But on to the point i think there is some evidence that colorism was not as prevalent (in Aryan times) as it has become today. Krishna, after all , is the dark god , while Kali is the darkest.

        2+
        1. May be Krishna and Kali were pre-Aryan deities.
          Popular fictions colour Krishna as blue instead of black. May be there are some fair-skinned versions of Kali also.
          Also, black Kali is a negative character ig.

          I dont understand how can there be different versions of Mahabharata whereas it is just a single book (or collection of books?) by a single person.

          0
          1. Krishna blue color just means he is un-earthly , so black with a different shade, no where close to white or else he would described as such,just like his brother is. Pretty sure Kali is dark in all her avatars though

            I think where does pre aryan ends and aryan starts is difficult to say. Shiva is a mountain god, and can be characterized as pre aryan. A popular Indian writer made him Tibetan 😛 Either way there is hardly any evidence to conclude one way or the other.

            On Mahabharat the guy who is supposed to have written it is himself as much a mythical character as the characters in the epics. So its not really a surprise.

            1+
          2. Mahabharata has been added to by several authors over the centuries. There are so many later interpolations and different recensions. In fact probably the most conservative of all chapters is found in Indonesia not India.
            In Bali you will be shocked at how huge Mahabharata is to the local culture. All the roads are named after characters such as Pandavas, Krishna, Draupadi etc. Locals have names like Arjun and Partha.
            The Kurushektra war imagery can be seen all over the places from shops to hotels.

            1+
          3. To me one of the interesting things in Bali was how structures are dedicated to relative minor obscure figures of Hinduism. Like Garuda etc. Perhaps their Hindu epics dedicate more to these characters.

            The other observation was how submissive Bali Hinduism really is. Its been encroached by Indonesian Muslims and the locals really resent it, but somehow resigned to their fate. It was something which reminded me of my early years growing up in N-India.

            0
      2. White-symbol of peace (doves)
        Red-symbol of aggression (bloody)
        Black-symbol of bad things (dark, gloomy days when nothing grows)
        Not sure where yellow fits in. How’s this racist? Stop imposing late 19th or early 20th century pseudo scientific racial theories.
        Infact, research needs to be done about origins of white supremacy? My bet is on the middle east.

        0
    2. The cave frescos in Ajanta accurately depict skin tone for monks and nobles.

      https://www.pinterest.com/pin/444308319466348843/

      I cannot imagine modern India depicting brown people in this manner.

      In modern India any depiction of a holy figure is portrayed with a light skin tone that is atypical in the region by default . Even the ones that canonically have dark or black skin are often portrayed as light blue.

      The obsession with light skin seems to post date the Buddhist period, based on artwork.

      0
      1. Sumit says – “In modern India any depiction of a holy figure is portrayed with a light skin tone that is atypical in the region by default . Even the ones that canonically have dark or black skin are often portrayed as light blue.”

        The best illustration of this are Amar Chitra Katha comic books which have contributed so much to the stereotyping of upper and lower castes, as well as the depiction of (most) Muslim personalities.

        Color consciousness predated Turko-Persian-Afghan rule. The extended rule by the Turko-Persian-Afghan elites likely contributed to its reinforcement. And then came English colonial rule……

        0
  8. “I report on evidence that a sample from Cambodia dated to 100 to 300 AD seems to have considerable Indian ancestry.”

    Has anyone analysed this sample and compared it with contemporary Indian castes?

    0
  9. “Though a lot of the discussion involves Muslims, I have heard from multiple non-Muslim people of Northwest Indian stock (e.g., Pandits) that they must have “Persian ancestry” because they look so Persian. The genetics refutes this rather strongly. Rather, modern Persians and many Northwest Indians share deep ancestry which diverged after the Last Glacial Maximum 20,000 years ago.”

    Razib – a lot of NW ethnonationalists believe they are direct descendants of Iranians but my belief is they just have a lot less AMH admix diluting their archaic Indus_Periphery type genetics making them seem distinct when they are really just on the continuum. Only tribes like the Baloch/Brahui, Western Pashtuns or Shia or Parsi migrants have actual migratory plateau Iranian genetics as seen by their oversized ANF to steppe ratio

    NW Iran_N is the same old ANEoid flavour of Central Asian West Siberian N influenced Iranian Farmers possibly of the R2/J2a type and devoid of the ANF. This is seen all the way down to Sri Lanka diluted by AMH. This is in contrast to the majority of the modern Iranian genepool typified by the Lur people which was heavily ANF from genesis as highlighted by the latest Reich lab paper whereas most of the ANF in the NW comes via Sintashta type people

    East Iranic tribes have always had a distinct steppe and Iran N substrate compared to modern Iranians and places like Herat which are much more influenced by the ANF/Iran N Iranian farmer

    Have you got any thoughts on this or consider deep-diving with a post? Could you in general provide your thoughts on the ANF influence in South Asia and whether some of it trickled through before or separate to steppe migrations?

    0

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.