Page 1 of 58 1231151 ... LastLast
Results 1 to 10 of 572

Thread: Upcoming K15 Ancient Cultures Calculator at GenePlaza

  1. #1
    Banned
    Posts
    5,171
    Sex

    Lightbulb Upcoming K15 Ancient Cultures Calculator at GenePlaza

    Here is a sneak preview of the upcoming K15 Ancient Cultures calculator at GenePlaza.com

    The motivation behind this calculator was the recent publication of dozens of higher quality ancient genomes in Mathieson et al, 2018; The Genomic History Of Southeastern Europe and in Olaide et al, 2017 The Beaker Phenomenon And The Genomic Transformation Of Northwest Europe.

    These genomes greatly added to our understanding of the population demography of Europe and also western Asia.To increase accuracy of the results, and SNP overlaps between the customer and the calculator population references, only genomes with the highest average read depth coverage were carefully chosen to source the component allele frequencies. The calculator algorithm used is detailed at the Eurasian DNA. Thus I believe that this is the most accurate ADMIXTURE based calculator to date based on ancient references.

    This calculator uses higher coverage ancient genomes from the aforementioned as well as previous studies to represent the various Neolithic, Chalcolithic, and Bronze Age cultures, stretching from western Europe all the way to central Asia and Siberia, and which contributed to the genetic makeup of the various modern populations currently residing across Eurasia.

    Since we are missing actual ASI samples, I created an ASI composite based on various modern populations. We don't currently have an adequate quantity of high coverage Natufian and Levant N samples to create a reasonably reliable SW Asian ancient component. Thus SSA would include any SSA substructure accounted for within modern SW Asian components.

    Srubna, Iran-N, and CHG is underestimated due to the limited number of samples and marker overlaps.

    Scythian-C includes the Altai IA samples as well as the Volga sample. EHG (Karelian HG + Ukraine N and Ukraine M) and WHG (includes Iron Gates HGs) should be interpreted as excess not accounted for by other components.

    Anticipated launch of the calculator at Gene Plaza is about a week from now.

    Here are some sample results;


    [IMG][/IMG]

    EDIT: Diffusion via geographically intermediate populations also account for admixture percentages seen in more distant samples

    EDIT2: C and E European Beaker samples were not used

    EDIT3: There is considerable genetic overlap between Karasuk and Scythian sample, and undoubtedly Scythians are substantially Karasuk derived


    EDIT4:

    On the one extreme you have the Harvard folks that are taking the position that all E Asian in Indians is derived from AASI, then on the other extreme there is you who I think is saying that Indians didn’t get their E Asian from AASI , I take a middle position, I do believe that Indians got a bunch of their E Asian from AASI, but I also think and this is reflected in the calculator that Indians got some of their E Asian from C Asia, Tibeto-Burmans, Austro-Asiatics, and in the case of E Indians and Bangladeshis, simply from modern SE Asians.

    You say that modern calculators give you 25% E Asian, why do you think that 12.6% E Asian + 16.7% AASI + 12.3% E Scythian can’t account for your 25% ? Don’t you believe that AASI shares a huge amount of alleles with E Asians ?

    Don’t compare E Asian levels in MODERN W Eurasians shown in modern calculators with E Asian levels here, because:
    1- Modern calculators model you in terms of other moderns such as Caucasians and E Europeans, who themselves have low levels of underlying E Asian due to Turkic, Hunnic, Scythian, and Uralic migrations from the east over the past couple of 1000 years. This calculator measures your E Asian in reference to ENF, Beakers, and Steppe groups who were not the subject of admixture with the aforementioned groups migrating from the east.
    2- There is also the issue of the ancient genomes being haploid. I haven’t seen anyone online discuss the effect on ADMIXTURE calculators. Nick Patterson of Broad Institute has mentioned the phenomena of “ancient attraction” on a PCA, meaning the ancients are attracted to other ancients. I believe I have conveyed to Nick in one of my numerous discussions with him that this has to do with haploid genotyping of aDNA.

    Here is from me to you for the 1st time, and you won’t see this online or in any book. When ancients are haploid genotyped, we lose quite a bit of variation in their genome, because hetrozygosity is replaced with homozygosity in perhaps 30% of the genome. For example, say you have a position rs1234 that in reality is GT in an MLBA steppe population, however it is genotyped as GG because you have more reads supporting the G allele. Also suppose that if the ancients from that population were not haploid genotyped, 70% of them would be GT. Now suppose that by coincidence 60% of E Asians have at least one T allele at rs1234. Guess what, had it not been for haploid genotyping those steppe MLBAs as GG, you would have 70% of them who would have had at least 1 T allele, and thus GT at rs1234 would have been assigned steppe, however, because of the haploid genotype of GG in steppe, it is now assigned as E Asian.

    So back to Nick Patterson’s phenomena of ancient attraction. I believe it is more appropriately termed repulsion of moderns by ancients, where moderns are distanced from ancients to the aforementioned haploid genotyping of aDNA. I believe this is responsible for SOME of the misassignment , and MAY act to inflate E Asian slightly. The majority though I believe is due to paragraph 1 above.
    Last edited by Onur Dincer; 05-02-2018 at 08:48 PM. Reason: The last edit written by the OP and inserted to this post by the forum administration at the request of the OP.

  2. The Following 35 Users Say Thank You to Kurd For This Useful Post:

     ADW_1981 (03-06-2018),  Aha (03-07-2018),  ajc347 (03-10-2018),  Alain (03-18-2018),  Angoliga (03-07-2018),  bmoney (03-06-2018),  Cascio (04-30-2018),  Dibran (03-06-2018),  Hanna (03-07-2018),  J1 DYS388=13 (03-06-2018),  Jessie (03-06-2018),  JFWinstone (03-07-2018),  jortita (03-06-2018),  Judith (03-25-2018),  kingjohn (03-06-2018),  lukaszM (03-07-2018),  Mixed (06-27-2018),  Modernancientdna (03-06-2018),  MonkeyDLuffy (03-06-2018),  Nikiforov (03-09-2018),  Nive1526 (03-06-2018),  NK19191 (05-06-2018),  okarinaofsteiner (05-06-2018),  oz (04-30-2018),  parasar (03-06-2018),  pnb123 (03-06-2018),  poi (03-06-2018),  Pribislav (03-06-2018),  Raskolnikov (05-03-2018),  Reza (03-06-2018),  SaltyShanker (06-02-2018),  sktibo (03-31-2018),  Theconqueror (03-06-2018),  Varun R (03-06-2018),  ZephyrousMandaru (03-07-2018)

  3. #2
    Bronze Class Member
    Posts
    1,567
    Sex
    Ethnicity
    Assamese Sutiya and Tai
    Nationality
    Indian
    Y-DNA (P)
    R1a1
    mtDNA (M)
    M13'46'61

    India China Tajikistan Mongolia Russian Federation Thailand
    If I understand correctly ASI will be proxy for Ancestral Southeurasian as you have not included any East Asian ancestry population in the calculator

  4. #3
    Banned
    Posts
    5,171
    Sex

    Quote Originally Posted by jortita View Post
    If I understand correctly ASI will be proxy for Ancestral Southeurasian as you have not included any East Asian ancestry population in the calculator
    Yes, it is best to minimize the use of modern populations in ancient calculators as they conflate ancient percentages. The types and number of various samples were carefully chosen to approximate ASI. They include Jarawans, a couple of SE Asian pops, and a few of Indian tribals in varying quantities. It becomes significant 5%+ in Iraqi Kurds and is under 20% for most Pakistani Pashtuns as well as many Punjabis.
    Last edited by Kurd; 03-06-2018 at 04:12 AM.

  5. The Following 6 Users Say Thank You to Kurd For This Useful Post:

     bmoney (03-06-2018),  jortita (03-06-2018),  khanabadoshi (03-06-2018),  kingjohn (03-06-2018),  poi (03-06-2018),  Reza (03-06-2018)

  6. #4
    Banned
    Posts
    4,088
    Sex
    Ethnicity
    karius the black hair
    Nationality
    east med

    dear kurd ,
    kudos for all your work
    you have my raw data
    can you run me so i can get some general picture ?
    with kind regards
    adam

  7. The Following User Says Thank You to kingjohn For This Useful Post:

     timberwolf (03-06-2018)

  8. #5
    Registered Users
    Posts
    356
    Sex
    Ethnicity
    Chauhan
    Y-DNA (P)
    R2a2b1b
    mtDNA (M)
    U7a

    Pakistan United States of America Azad Kashmir Jammu and Kashmir
    Thank you for your work Kurd!
    Kashmir

  9. The Following 4 Users Say Thank You to noman For This Useful Post:

     Angoliga (05-06-2018),  Cascio (03-07-2018),  khanabadoshi (03-06-2018),  kingjohn (03-06-2018)

  10. #6
    Registered Users
    Posts
    286
    Sex
    Omitted

    Looks very interesting.
    Beakers in an Assyrian? Beakers everywhere!

  11. The Following User Says Thank You to Fungene For This Useful Post:

     kingjohn (03-06-2018)

  12. #7
    Banned
    Posts
    5,171
    Sex

    Quote Originally Posted by Fungene View Post
    Looks very interesting.
    Beakers in an Assyrian? Beakers everywhere!
    4000 years is plenty of time for Beaker related to diffuse via European populations to W Asia, but what we may also be seeing is some shared ancestral Yamna related. In other words, some ancestral shared signal of the Yamna type (yet unsampled Yamna) which contributed to both Beakers and W Asians.

  13. The Following 4 Users Say Thank You to Kurd For This Useful Post:

     Alain (03-18-2018),  Fungene (03-06-2018),  Judith (03-25-2018),  kingjohn (03-06-2018)

  14. #8
    Banned
    Posts
    5,171
    Sex

    It is probably just as informative to look at the group total percentages as the individual ones. For example, the 2 Pashtun individuals have the highest total of W + E Steppe, or approx 48%, and if one considers beakers as 50% W Steppe derived then a total of approx 53%, which is approximately the same as the 2 English individuals if one factors in Beakers; 50-55% W Steppe for the English.

    Also, interestingly, the British Neolithic samples have a little different sub-structure than the SE European Neolithic ones.


    POPULATION Kurd-Iraq1 Kurd-Iraq2 Assyrian English-SE English French-N Pashtun-Afg1 Pashtun-Pak1 Punjabi-Mixed Brahmin- S Indian
    WESTERN EURASIAN STEPPE 13.51% 21.45% 16.81% 36.59% 39.46% 30.09% 41.34% 39.07% 30.86% 26.28%
    Yamna 6.94% 8.94% 8.01% 15.19% 17.94% 15.01% 13.31% 8.27% 14.35% 11.42%
    Andronovo 0.00% 8.78% 6.50% 17.77% 18.82% 8.92% 24.00% 29.59% 13.39% 14.86%
    Srubna 6.56% 3.72% 2.30% 3.62% 0.86% 6.03% 4.03% 1.22% 3.12% 0.00%
    Eastern European Hunter Gatherers 0.00% 0.00% 0.00% 0.00% 1.83% 0.13% 0.00% 0.00% 0.00% 0.00%
    EASTERN EURASIAN STEPPE 20.38% 11.26% 7.39% 0.67% 0.00% 5.70% 7.04% 8.31% 10.61% 8.00%
    Karasuk 8.37% 8.10% 7.39% 0.00% 0.00% 5.44% 6.71% 8.31% 10.61% 0.00%
    Central Scythians 11.37% 3.15% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 7.90%
    Eastern Scythian 0.64% 0.00% 0.00% 0.67% 0.00% 0.26% 0.32% 0.00% 0.00% 0.10%
    CONTINENTAL EUROPEAN 2.65% 5.68% 7.61% 26.44% 27.35% 23.95% 8.31% 9.76% 5.67% 5.47%
    Bell Beakers 2.65% 5.68% 7.61% 20.96% 21.13% 17.81% 8.31% 9.75% 5.67% 5.47%
    Western European Hunter Gatherers 0.00% 0.00% 0.00% 5.47% 6.22% 6.13% 0.00% 0.00% 0.00% 0.00%
    NEOLITHIC FARMERS 36.40% 32.80% 42.14% 34.42% 30.09% 35.94% 10.90% 5.01% 9.02% 8.61%
    SE European-Neolithic 23.70% 22.90% 21.45% 16.40% 5.46% 14.80% 6.50% 2.85% 4.10% 3.87%
    British-Neolithic 11.50% 9.31% 18.02% 13.57% 22.71% 17.74% 1.75% 0.00% 3.03% 1.29%
    Iberian-Neolithic 1.19% 0.59% 2.68% 4.46% 1.92% 3.40% 2.65% 2.16% 1.89% 3.44%
    WEST & SOUTH ASIAN 24.59% 25.93% 24.10% 1.23% 3.11% 3.73% 30.76% 36.32% 41.37% 48.87%
    CHG-Iran-Neolithic 18.77% 19.42% 21.94% 1.23% 2.15% 3.73% 17.20% 18.16% 19.40% 14.13%
    Ancestral South Eurasian 5.81% 6.51% 2.16% 0.00% 0.96% 0.00% 13.56% 18.15% 21.97% 34.74%
    Sub-Saharan African 2.47% 2.88% 1.94% 0.65% 0.00% 0.58% 1.65% 1.54% 2.47% 2.78%

  15. The Following 6 Users Say Thank You to Kurd For This Useful Post:

     khanabadoshi (03-06-2018),  kingjohn (03-06-2018),  michal3141 (03-12-2018),  MonkeyDLuffy (03-06-2018),  Reza (03-06-2018),  Varun R (03-06-2018)

  16. #9
    Global Moderator
    Posts
    809
    Sex
    Location
    UK
    Ethnicity
    Bengali
    Y-DNA (P)
    R1a-L657
    mtDNA (M)
    M4b1

    Great work Kurd. But echoing jortita's point about ASE, in people with both ASI and East Asian, would be difficult to differentiate between those two components. Presumably some of the East Asian will go into E Scythian or Karasuk?
    Paternal; Y-DNA: R1a-L657> Y6> Y11> Y920*
    Paternal; mtDNA: M5b'c
    Maternal; Y-DNA: R1a-L657> Y6
    Maternal; mtDNA: M4b1
    Maternal Grandmother; Y-DNA: R1a-L657> Y6

  17. The Following User Says Thank You to Reza For This Useful Post:

     kingjohn (03-06-2018)

  18. #10
    Banned
    Posts
    5,171
    Sex

    Quote Originally Posted by Reza View Post
    Great work Kurd. But echoing jortita's point about ASE, in people with both ASI and East Asian, would be difficult to differentiate between those two components. Presumably some of the East Asian will go into E Scythian or Karasuk?
    Exactly, until we have relevant E Asian ancient sources that contributed a different type of E Eurasian signal to S Asians than Karasuk and Scythians, it’s best not to introduce modern E Asians as they would mess up the Karasuk and Scythian signals. For now Karasuk, Scythians, and other steppe groups are sufficient to account for Siberian and E Asian type admixture in W/SC/and S Asians west of Bangladesh. We just need many more higher coverage Karasuk and Scythian samples to more accurately account for Siberian and E Asian in those groups

    Edit: If you think about it ASE, Karasuk, Scythians do account for almost all Siberian and E Asian admixture in S Asians

    The component is more appropriately called ASE than ASI in the K15
    Last edited by Kurd; 03-06-2018 at 12:55 PM.

  19. The Following 8 Users Say Thank You to Kurd For This Useful Post:

     bmoney (03-07-2018),  khanabadoshi (03-06-2018),  kingjohn (03-06-2018),  poi (03-06-2018),  Reza (03-06-2018),  SpinosaurusN3H1 (06-15-2018),  tremaineharris (05-03-2018),  Varun R (03-06-2018)

Page 1 of 58 1231151 ... LastLast

Similar Threads

  1. Upcoming K36 Bronze/Iron Age Calculator GenePlaza
    By Kurd in forum Ancient (aDNA)
    Replies: 431
    Last Post: 11-07-2020, 06:48 PM
  2. Upcoming K53 Ancient Eurasia Calculator
    By Kurd in forum Ancient (aDNA)
    Replies: 84
    Last Post: 02-12-2019, 12:05 AM
  3. Upcoming Ancient Eurasia K6 Calculator
    By Kurd in forum Autosomal (auDNA)
    Replies: 143
    Last Post: 05-25-2018, 09:27 AM
  4. Replies: 15
    Last Post: 05-09-2018, 07:02 AM
  5. Upcoming K21 Calculator at GenePlaza
    By Kurd in forum Autosomal (auDNA)
    Replies: 190
    Last Post: 12-04-2017, 06:05 PM

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •