Saturday, April 23, 2011

K=11 ADMIXTURE results for selected participants

In Dienekes' Anthropology Blog, I have created the most comprehensive analysis of genetic structure in West Eurasians so far.

Using a new technique to frame the region of interest (West Eurasia and North Africa) by meta-population controls representing Sub-Saharan Africans, East Eurasians, and South Asians, I was able to extract several robust components that track admixture in unprecedented detail.

In comparison to previous analyses, the North European component has been split into Northwest and Northeast European, while the South European one has been split into Sardinian and Basque. Additionally, the inclusion of the above-mentioned controls, as well as a great variety of West Eurasian populations has probably improved the accuracy of the results for all the other components substantially.

You can read the details, and obtain detailed data downloads from this study in my other blog. But, it's good to remember that it is only possible for me to do things like this because of the project volunteers.

Because of the anthropological nature of the study, only individuals who belonged to particular populations could participate, and they can all see their individual results here. So, check the figure on the left to see if your Dodecad population is included, and then look at the spreadsheet for your individual results.

16 comments:

  1. Could you make Fst, MCLUST and MDS/PCA analyses of these populations?

    ReplyDelete
  2. My results for this run were quite different than the standard K=10.
    Dodecad Ancestry Project ID : DOD464
    South Asian - 67.6 %
    Northern European - 4.1%
    East Asian - 2.6%
    West Asian - 25.7%

    As opposed to Genetic structure of West Eurasian exercise at K=11;
    South Asian - 98.74%
    East Asian - 0.65%
    Sardinian - 0.61

    Pardon my ignorance, but is this vast difference in results due to the specificity and diversification in the West Eurasian K's/ancestral populations fed into the program? Also, the minor exogenous West Eurasian admixture is now Sardinian instead of West Asian (The reference samples being ethnic groups from the Caucasus, I assume). Is the East Asian and Sardinian just noise, and could it be that I am in reality 100% South Asian?

    Another BGA (Dr. Doug McDonald's) had a similar result for me (i.e close to 100% South Asian, reference population = Dehli Brahmins + Sub-cluster of Singapore Indians).

    Thanks for your time, Dienekes.

    ReplyDelete
  3. This comment has been removed by the author.

    ReplyDelete
  4. Dienekes,

    Fantastic analysis, great work!!!

    One small request, could you freeze the top row - with the population labels - in the spreadsheet, so that scrolling down, one can easily see what they refer to - thanks.

    ReplyDelete
  5. Thank you very much, Dienekes! I've been anticipating another admix run and here we have one centered more squarely on Europe itself.

    Northern Europe and Southern Europe each get split (NW vs NE & Basque vs Sardinian) but East/NE Asia get conflated for a net addition of one component. It is now that I have E. African precipitate out. Very interesting. Perhaps King Belus did debark on Argolid soil after all!

    On behalf of the non-single-ethnic folks, I ask if, now that you have the 11 components defined, it is possible to analyze the rest of the database.

    ReplyDelete
  6. One small request, could you freeze the top row - with the population labels - in the spreadsheet, so that scrolling down, one can easily see what they refer to - thanks.

    Good idea. Dieneke, could you freeze the top rows of other spreadsheets (including the ones in the other blog) as well?

    ReplyDelete
  7. Hi Dienekes, would you be willing to answer my query (the comment which hasn't been approved) privately? A short statement would be fine : vasishta.anthropology@gmail.com - DOD464 here..

    ReplyDelete
  8. Very detailed. Too bad I wasn't on it. I would have loved to see how much Basque I have...

    ReplyDelete
  9. I looked at my third and fourth cousins on the Relative Finder Function of 23 & me and have found most of them to be of Basque/Celt haplogroups, so I would be interested in seeing how much Basque I have.

    ReplyDelete
  10. Hi Dienekes, would you be willing to answer my query (the comment which hasn't been approved) privately?

    Vasishta, as Dienekes published your comment, however late, I am for an on blog response from Dienekes, as I am curious about that issue too.

    ReplyDelete
  11. Pardon my ignorance, but is this vast difference in results due to the specificity and diversification in the West Eurasian K's/ancestral populations fed into the program?

    In this experiment I have used the diverse Indian_D set as a way to "frame" West Eurasia, rather than use 1-2 populations (Kannadi and Gujarati) as I did before. Such populations are even more "southern" than you, and hence they define the "South Asian" pole. As there are no Indian tribals or South Indian low castes in Indian_D, the South Asian pole of variation is formed by South Indian Brahmins.

    ReplyDelete
  12. I am DOD 372, of Souther Italian and Sicilian ancestry. My Sardinian component is 32.33 % - the highest component for me (and also one of the highest Sardinian components in the project.) My West Asian component is 30.58%. Could anyone explain this Sardinian component for me as I have no known Sardinian ancestry. I assume this means my Italian ancestors shared a common ancestor with some element of the Sardinian population. Any way to tell if this is relatively recent or considerably older.
    Thanks.

    ReplyDelete
  13. Archduke, (such a name for an Italian!) "Sardinian" refers to an entire cluster representing aboriginal southern europeans who are best epitomized by Sardinians because they have less membership in other clusters. You did not object to being ~31% West Asian even though you likely have no recent Anatolian or Caucasic ancestry.

    Notice also that Sicilian, North Italian, Greek, Austrian and other highly specific places are not and cannot be among the eleven categories.

    Moreover, check the spreadsheet to see what other people get. You'll notice that all Italians have substantial "Sardinian" scores. Of course! Where else would they be?

    ReplyDelete
  14. When Dr. McDonald ran my autosomal, I came up to 30% N African. When Dodecad ran my autosomal, I came up to 26% S Euro and 2.7% NW African. I wonder if I will turn up 30% N African if I am run under the N African classification in Dodecad...

    ReplyDelete
  15. That's impossible, as the population labels are not used in any way by ADMIXTURE; it only looks at the raw genotype data.

    ReplyDelete