ANI/ASI Admixture Dating

Via Razib, here's an interesting abstract from the International Congress of Human Genetics by David Reich's group:

Estimating a date of mixture of ancestral South Asian populations.

Linguistic and genetic studies have shown that most Indian groups have ancestry from two genetically divergent populations, Ancestral North Indians (ANI) and Ancestral South Indians (ASI). However, the date of mixture still remains unknown. We analyze genome-wide data from about 60 South Asian groups using a newly developed method that utilizes information related to admixture linkage disequilibrium to estimate mixture dates. Our analyses suggest that major ANI-ASI mixture occurred in the ancestors of both northern and southern Indians 1,200-3,500 years ago, overlapping the time when Indo-European languages first began to be spoken in the subcontinent. These results suggest that this formative period of Indian history was accompanied by mixtures between two highly diverged populations, although our results do not rule other, older ANI-ASI admixture events. A cultural shift subsequently led to widespread endogamy, which decreased the rate of additional population mixtures.

I would be very interested in reading that paper. Also, I wonder how many new samples did they genotype beyond the ones in Reich et al' Reconstructing Indian Population History and if I could get my hands on the new data.

I have a feeling that ANI (Ancestral North Indian) captures a bunch of different migrations and conquests etc, so I am not sure if it can be equated to Indo-European language movement.

I wonder if I can use HAPMIX or StepPCO to get similar admixture dating.

Related Posts:

38 Comments.

  1. Check this out, too. Very similar age estimates -

    Phylogeography of R1a1 Y-chromosomal haplogroup and genetic history of Indo-Europeans. V. A. Stepanov, V. N. Kharkov Institute for Medical Genetics, Tomsk, Russian Federation.

    Recent discussion of the prehistoric spreading of the Indo-European language group has generally concentrated on two alternative hypotheses: so-called “Kurgan Culture” hypothesis, which places the homeland of proto-Indo-Europeans to the Steppe of Eastern Europe, and alternative hypothesis of the spread of farmers from the Near East (Anatolia) to Europe in the Neolithic times. Y-chromosomal haplogroup R1a1, lineage is thought to have originated in the Eurasian Steppes north of the Black and Caspian Seas, seems to be associated with the Kurgan culture. Three geographic areas with the highest frequency of R1a1 haplogroup were revealed: Eastern Europe; Southern Siberia and Hindustan where the highest diversity of microsatellite haplotypes was observed. Phylogenetic analysis of microsatellite haplotypes demonstrates the presence of three corresponding major clusters with the age of the generation of haplotytic diversity of 7.2-12.5 ky. The highest diversity in Hindustani is related to the presence of haplotypes of Indo-Pakistani and Southern Siberian clusters in the population from India and Pakistan, probably due to relatively recent migrations from Central Asia. The age of the cluster admittedly brought to Hindustan from Central Asia / Southern Siberia is 3,9 +/- 1,3 ky. Probably, the primary center of the generation of diversity and expansion of R1a1a was the territory of the Eastern European Steppe. With the spread of of R1a1 carriers, secondary centers of genetic diversity and population expansions were formed in the Southern Siberia and Hindustan.

    • well, there has been a debate about r1a1 estimates for a decade. doesn't seem like we're that much closer to resolving it. i think ancient DNA is our best bet at solving this issue.

    • The time-line is not clear but the direction of movement of R1a1 (R-M417) is most likely from Europe to the east. Passarino* very early had posited this (out of Ukraine) idea but papers like that were soon replaced by STR diversity analysis.

      *"the combined analysis of the Eu19 biallelic markers with the 49a,f system has provided interesting information concerning past human migrations in Eurasia. The 49a,f system has allowed a further characterization of the Eu19 lineage by a new marker, the 49a,f Ht11. On the other hand, the Eu19 lineage has provided stable markers to identify all those haplotype variants deriving from the 49a,f Ht 11. These variants, in turn, by subclassifying the Eu19 lineage, have given insights on its geographic subdivision and a new hint that Ukraine may be the area where the Eu19/Ht 11 spread from."

      SNP tree and data - http://www.box.net/shared/tgtnvajp4rzx05o4st7n
      https://docs.google.com/spreadsheet/ccc?authkey=CMjR3ssG&key=0AjnH-bSBYrYTdHV0WWR1ZF80anFmbGh0VmNvbGh6d1E&hl=en_US&authkey=CMjR3ssG#gid=0

  2. Well, this paper concludes south asian R1a1a7 as older than eurasian.
    http://www.nature.com/ejhg/journal/v18/n4/full/ejhg2009194a.html

    • That paper concludes no such thing.

    • I agree with Onur that this is not a claim that this paper makes. If this were true, it would, of course, be quite a boost to those who continue to adhere to some silly Out-of-India Indo-European migration hypothesis. In addition, not detecting Eastern Europe-specific R1a markers in India has no relevance to Indo-European migrations into or out of India. The region of interest remains the steppe zone extending from the Altai mountains to the Volga River basin, not Eastern Europe.

      • Quick clarification: Though the Volga River basin is technically in Eastern Europe, it was not sampled in this paper specifically. What I meant to say was that contemporary R1a marker prevalence in Poland and the Czech Republic has little bearing on Bronze Age movements into or out of India.

        • If Nirjhar had said "this paper concludes South Asian R1a1a as older than Eurasian" instead of saying "this paper concludes South Asian R1a1a7 as older than Eurasian", I could agree with him. According to the results of that paper, the coalescent times of R1a1a is older in South Asia than in the rest of the world. But according to the same paper, R1a1a7 is entirely confined to Europe and its immediate surroundings such as Anatolia (thus it is absent in South Asia and in almost all of the continent of Asia) and its oldest coalescent time is in Poland and its surroundings.

          As for the Indo-European migrations, it is clear from genetics that Indo-European-speaking people came to South Asia from West Asia and not from Europe.

        • BTW, it is not true that the Volga River basin wasn't sampled specifically in that paper. They sampled all of Eastern Europe including the easternmost areas such as the Volga and Ural regions. And in all of those easternmost European regions they found R1a1a7, however small in quantity.

          • They sampled all of Eastern Europe including the easternmost areas such as the Volga and Ural regions. And in all of those easternmost European regions they found R1a1a7, however small in quantity.

            I mean the Native peoples of those areas.

          • Nirjhar did make that (typographical?) error, and Underhill et al does find the coalescent age for R1a1a highest in western India. Sengupta, Sharma, and Mirabal had also come to similar conclusions. The problem appears to be that they are all wrong as diversity is proving to be a unreliable measure. Recent population expansion in Europe must be swamping the underlying lineage diversity there. R1a1-M417 has the earliest branches in Europe with subsequent movement to West Asia and South Asia. http://www.gwozdz.org/R1a.pdf

    • Here is a quotation from the abstract of the Underhill et al. paper.

      "Importantly, the virtual absence of M458 chromosomes outside Europe speaks against substantial patrilineal gene flow from East Europe to Asia, including to India, at least since the mid-Holocene."

  3. Onur you are a eurocentrist and I am a 0 centrist, I didnt say a single word one OIT but u did.
    Both zhivotovsky and genealogical methods are wrong and ancient DNA (FARMANA skeletons) and use of much trustable UEP should be the path to know the truth!
    And talk about your K.H. and A.H. they have one thing better than the "stupid" OIT and its academic support.
    The trying of putting Genetics and linguistic together is good 4 academics but to be truthful it will have no clinical base at all as linguistics have NO SCIENTIFIC VALUE on judging totally unseen yoke.
    P.S. truth will alone win.

    • Whoa! How you arrive at those conclusions about me from what I write is beyond me. Your replies to me are so unrelated to what I write that I am thinking that you confused me with someone else. I just corrected your error, which is, as Parasar said, likely a typographical error.

      Onur you are a eurocentrist and I am a 0 centrist

      Where did I say anything Eurocentrist?

      I didnt say a single word one OIT but u did.

      Where did I say any single word on OIT?

      Both zhivotovsky and genealogical methods are wrong and ancient DNA (FARMANA skeletons) and use of much trustable UEP should be the path to know the truth!

      Does that have anything to do with what I wrote above?

      And talk about your K.H. and A.H. they have one thing better than the "stupid" OIT and its academic support.

      What are K.H. and A.H.?

      The trying of putting Genetics and linguistic together is good 4 academics but to be truthful it will have no clinical base at all as linguistics have NO SCIENTIFIC VALUE on judging totally unseen yoke.

      If you mean by "putting genetics and linguistic together" equal or similar treatment of them in inferring ancestry, no serious person (including me of course) takes seriously such an idea. Linguistics should be used by genetics only in an auxiliary way, as linguistics by itself can say nothing about ancestry.

  4. Oh yeah it was some outing,
    1. SW asian origin of aryas?
    You are a dienke fan arent u?
    Just because there is Sw components in some indians doesnt mean swa was our homeland as RV do not mention any alien home than sapta sindh, the case of avestan airnyam vajeh is dubious and not reliable.
    2. If indian R1a1* is 5200 year old then it does correlates with the emergence of IVC and language of the sindh basin and i can accept the date but not the baseless 1500-1200 b.c.
    3. After 800 b.c there were massive intrusions by persians, sakas, greeks,huns, yuzhes,arabs,pakthas,muslims and at last british.
    Forget the last two though J2b can be related to muslims can you imagine all these people coming and not making any impact specially in you guys gurus dodecad components?
    Yes there are many r1a carriers came here but not indo-aryas atleast after 3000 b.c.
    There is a recent trying to make j2 as a indo-european intrusion to india but unfortunately it has a problem as we brahmins do not have next to none of it and i think with giacomo that J2 is a farming mesopotemian intrusion to india starting from mehrgarh and going with sapta sindhs R1a1a* vedic people to make the great IVC/SSC which was quite advanced than its cotimed ones. 4. NO zhivotovsky or genealogic but ancient dna and as dienke have mentioned UEP, and talk about ancient dna farmana is going to be a crucial breakthrouth for our true and unbiassed history.
    5. THE indian genome is made by three main groups from base:
    1. Original ANI.
    2.Original ASI.
    3. South East asian mtdna from archaic times.
    Now as Rajib predicted the original ANI is pseudofied due to the massive intrusions, The case of ASI is also same but yet not clarified, the reason of having the east asian like component is the archaic main mtdna of us M and its daughter R which also the main mother of european genome.
    6. BMAC can be related to the pre-saka people not aryas.
    At the end i want to say try to be more baseful and judge the bases, do not go with any chaps theory but findings.
    Satyameva Jayate.
    think with giacomo that J2 is a farming mesopotemian intrusion to india starting from mehrgarh and going with sapta sindhs R1a1a* vedic people to make the great IVC/SSC which was quite advanced than its cotimed ones. 4. NO zhivotovsky or genealogic but ancient dna and as dienke have mentioned UEP, and talk about ancient dna farmana is going to be a crucial breakthrouth for our true and unbiassed history.
    5. THE indian genome is made by three main groups from base:
    1. Original ANI.
    2.Original ASI.
    3. South East asian mtdna from archaic times.
    Now as Rajib predicted the original ANI is pseudofied due to the massive intrusions, The case of ASI is also same but yet not clarified, the reason of having the east asian like component is the archaic main mtdna of us M and its daughter R which also the main mother of european genome.
    6. BMAC can be related to the pre-saka people not aryas.
    At the end i want to say try to be more baseful and judge the bases, do not go with any chaps theory but findings.
    Satyameva Jayate.

  5. My instincts given the archaeology are to think that ANI autosomally includes within itself both a Harappan layer and an Indo-European layer that strongly overlap geographically and were largely disjoint with ASI until Indo-European expansions with a Harappan substrate.

    I'm not sure that the models that are being fit to in order to set that date are sufficiently robust to remain accurate given that more complex model of what probably really happened.

    • Your idea that that the Indo-European layer will hardly be distinguishable from the Harappan layer must surely be correct. The Neolithic arrived started in Pakistan at least 9000 years ago as attested by Mehrgarh.

      http://en.wikipedia.org/wiki/Mehrgarh

      Agriculture spread very soon to all of Pakistan and resulted in a population larger than in Egypt or Mesopotamia. This population must have been largely West Eurasian. If Indo-Europeans arrived from the West and imposed their language on Pakistan, this must have been by a process of Elite Dominance unaccompanied by much genetic change. It would have been logistically impossible to get enough people across the barren land of Afghanistan to significantly affect the genetics of the huge population in Pakistan once agriculture was established.

      It is not surprising that Sengupta et al. find, “Therefore, our data do not support models that invoke a pronounced recent genetic input from Central Asia to explain the observed genetic variation in South Asia. “ Similarly Underhill et al. state, “"Importantly, the virtual absence of M458 chromosomes outside Europe speaks against substantial patrilineal gene flow from East Europe to Asia, including to India, at least since the mid-Holocene."

      The Harappan culture remained largely in Pakistan for many millenia and did not spread to the rest of the Subcontinent because Pakistan is separated from India by the Thar desert and has different ecological conditions. India remained largely ASI all this time. After the collapse of the Harappan civilization around 2000 B.C., people form Pakistan speaking Indo-European languages migrated to India and created the cline of ANI that is now observed.

      • If India was largely ASI till 2000bc who were these occupants of the Ganga valley in the as early as the beginning of Holocene -

        "They look like European Cro-Magnon skeletons!" 'Cro Magnon' forager type with prominent chin, deep nasal notch, and slight prognathism. http://books.google.com/books?id=W6zQHNavWlsC&pg=PA226

        The above was supplemented by more finds at Damdama -
        "Mesolithic sites near Allahabad (Uttar Pradesh) have yielded abundant well-preserved human skeletons permitting a bioarchaeological approach to past life-ways. Prior research on human remains from Sarai Nahar Rai and Mahadaha is supplemented by this analysis of human skeletal variation in 47 specimens from Damdama."
        http://muse.jhu.edu/login?uri=/journals/asian_perspectives/v042/42.2lukacs.pdf

        • The author is comparing these people of the Ganga plain to Cro-Magnons because they were tall and strongly built. However the Ganga people had some prognathism and alveolar prognathism that the Cro-Magnons lacked. There is nothing in the descriptions of the skeletons that would preclude them from being ASI. It would be great if it would be possible to recover some DNA from these skeletons.

          • On page 231 of Kennedy's book, he says, "There is a superficial resemblance of Gangetic mesolithic people to European upper paleo
            lithic fossils, but these phenotypic features are limited. There is no reason to hypothesize that the lake culture people are a branch of Cro-Magnon hunters who migrated eastward from the Durdogne valley, but phenotypic parallels may reflect similarities of lifeway within the hunting-foraging and lithic technological survival strategies of some prehistoric populations."

          • I would agree that there is no reason to hypothesize an eastward movement. On the same page he asks "why do populations like [these] ... no longer exist in south Asia?"

            Ancient DNA from south Asia would definitely clarify.

            We know that:
            1. The Upper Paleolithic European from Kostenki was U2 mtDNA;
            2. U2 is currently present in South Asia and in Europe;
            3. the morphology of the Kostenki European was tropical.

            "EARLY UPPER PALEOLITHIC MAN FROM MARKINA GORA (KOSTENKI XIV) ... ... typical of tropical populations. The trait combination links the cranium with those of Papuans and Melanesians."

            Therefore it appears that mesolithic Europeans are a mix of these two types, a Cro Magnon type and a tropical type.

          • That U2 is shared by South Asians and Europeans was observed more than 12 years ago by Kivisild etsl. "Deep common ancestry of Indian and western Eurasian mitochondrial DNA lineages. Current Biology 9: 1331-1334".

            http://jorde-lab.genetics.utah.edu/elibrary/Kivisild_1999.pdf

            This connection is very ancient (~50,000 years). There is not much evidence for significant recent (Holocene) genetic flow from Europe, West Asia or Central Asia to South Asia.

            In fact Kivisild writes. "One of the major topics of my research has been genetic diversity in South Asia and its role in the settlement of Eurasia. Have the diverse populations of India emerged as a result of admixture of East and West Eurasians as the language distributions would suggest? The study of mtDNA and Y chromosome variation has revealed that, instead, a large majority of the genetic lineages present in India – both among the caste and tribal groups – are unique to the subcontinent and likely trace their common ancestry with West and East Eurasian lineages back to the single and rapid out of Africa migration approximately 40 to 60 thousand years back with relatively minor element of later admixture across subcontinental borders."

            http://www.human-evol.cam.ac.uk/Members/Kivisild/Kivisild.htm

          • Kivisild was dividing using 16129C which is the one in Europe.
            It is rare in India but present:
            http://www.biomedcentral.com/1471-2156/9/86/table/T5
            U2e in Thogataveera and Reddy - Pg 7/972 The Thogataveera samples are U2e1
            http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1182158/pdf/AJHGv75p966.pdf

            That Brahmin type found in Northeastern Mongolia was R1a1, U2e1:
            "This is the first genetic evidence that a male of distinctive Indo-European lineages (R1a1) was present in the Xiongnu of Mongolia. This might indicate an Indo-European migration into Northeast Asia 2,000 years ago ... The genetic evidence of U2e1 and R1a1 may help to clarify the migration patterns of Indo-Europeans and ancient East-West contacts of the Xiongnu Empire ... "Calculation by DNAVIEW shows that the autosomal profile of MNX3 West Eurasian male is 14 times more probable from a Brahmin Indian than from a modern Caucasian (Table 6)." http://volgagermanbrit.us/documents/Kim_et_al.pdf

          • The skeleton in the Mongolian grave probably is that of a Buddhist missionary monk from India.

            U2e is present at low frequency in South Asia as well as in Europe. U2a, U2b and U2c are present at high frequency in South Asia and not at all in Europe. The simplest explanation for this pattern is that U2 differentiated into its sub-components in the Subcontinent about 50,000 years ago and only U2e migrated to Europe. So the upper paleolithic Kostenki man may be descended from migrants from the Subcontinent.

            Metspalu et al. in their recent paper seem open to the possibility "that West Eurasian diversity is derived from the more diverse South Asian gene pool."

  6. The billion SNPs of Alba? | Gene Expression | Discover Magazine - pingback on April 18, 2012 at 9:31 am
  7. The billion SNPs of Alba? | Biology News by Biologged - pingback on April 18, 2012 at 1:29 pm
  8. Facing the ocean | Gene Expression | Discover Magazine - pingback on April 29, 2012 at 1:33 am
  9. Facing the ocean | Biology News by Biologged - pingback on April 29, 2012 at 7:29 am
  10. ANI-ASI Admixture Dating | Harappa Ancestry Project - pingback on June 25, 2012 at 5:26 am