Total Pageviews

Tuesday, July 31, 2012

DNA tribes presents Kurdish data

DNATribes is a DNA testing company that I personally cannot recommend because there are other testing opportunities available that are better and for free (Dodecad, Harappa, Eurogenes, gedmatch, Interpretome, etc.).
However, I was notified that DNATribes recently added Kurds to their reference samples, so it is worth to take a look at their data presented in this pdf. In the below image the most important information about Kurds is summarized:

DNATribes is calling the most prominent component in Kurds the "Persian" component (or "world region") regardless of the fact that it is peaking in Kurds (39.5%) and not in Iran (22.0%). Turkmen also have a high portion of this component (32.6%). The description of this component states: "Lake Urmia, Zagros and Elburz Mountains." Lake Urmia is not Persian, the Zagros Mountains are not Persian, either. The name "Persian" is totally misleading for this component, it should be called "Iranian" to be fair. Some people still seem to have problems to distinguish between "Iranian" and "Persian".  This is as ignorant as putting Germans and Austrians into one group...oh wait, DNATribes actually did that as well.


Here are populations with more than 6% of the "Persian" component based on DNATribes:




Persian
Kurdish
39.50%
Turkmen
32.60%
Persian Qatar 25.60%
Iran
22.00%
Assyrian
16.80%
Turkey
16.10%
Jewish Iraq 14.30%
Makrani Pakistan 12.90%
Lezgin Caucasus 12.70%
Jordan
11.80%
Armenian
11.80%
Balochi Pakistan 10.90%
Tajik
10.90%
Lebanon
10.00%
Syria
9.70%
Palestinian Israel 9.60%
Kumyk
9.40%
Druze Israel Carmel Israel 8.80%
Yemen
7.70%
Georgia Caucasus 6.90%
Brahui Pakistan 6.80%
Adyghe
6.70%
Bedouin Negev Desert 6.30%
Jewish Morocco 6.10%














The closest to Kurds at DNATribes (based on Euclidean distance) are:



1 Kurdish
2 Iran
3 Persian Qatar
4 Turkmen
5 Turkey
6 Jordan
7 Syria
8 Armenian
9 Assyrian
10 Lebanon
11 Palestinian Israel
12 Druze Israel Carmel Israel
13 Tajik
14 Jewish Iraq
15 Nogay
16 Sephardic Jewish Bulgaria Jewish
17 Jewish Morocco
18 Yemen
19 Ashkenazi Jewish Europe Jewish
20 Kumyk

29 comments:

  1. "DNATribes is calling the most prominent component in Kurds the "Persian" component (or "world region") regardless of the fact that it is peaking in Kurds (39.5%) and not in Iran (22.0%). Turkmen also have a high portion of this component (32.6%)"

    Thats the first thing I thought when I saw that. "Holy.. the "Persian" component peaks in Kurds and is not even that strong among Iranians". Which is one of the main reasons why I try to avoid the term "Iranian" and prefer to call Kurds an Iranic ethnic group, because Persian has almost become a synonym for Iranian. This so called "Persian" component is probably an (West) Iranian and East Caucasian component which was introduced to Eastern Caspian areas from West Asia.

    ReplyDelete
  2. yet I have to admit I find DNATribes new analyses very interesting and good because they distinguish between the many different elements. The only thing What disturb me a bit was that they named this component, which extends from the Southern Caucasus, North Mesopotamia throughout North Iran all the way into Eastern Caspian borders, "Persian". They should have used a different more suiting name like "Iranic" or "Eastern West Asian".

    ReplyDelete
  3. It is best to name the components based on geography, and not based on ethnic groups. Because the components predate the existing ethnic groups.

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

    ReplyDelete
  5. I wonder the reason behind the "East Asian" component average of 1.1% in the DNA Tribes Greek population (see 8 Continents Admixture). Is it because there are some Anatolian Greeks among them and those Anatolian Greeks have above noise levels of "East Asian" component as a legacy of the Turkic invasion of Anatolia?

    ReplyDelete
    Replies
    1. Northern Greece has a large Turkish community, so it could be from them. But the weird thing about this is that the Greek sample set at DNATribes does not have any Northern Asian, it is "East Asian" alone, which is uncommon for Turks.

      Delete
    2. As far as I can see from the Dodecad Project, pure Balkan Turks totally lack the Mongoloid components. There are some Balkan Turk-Anatolian Turk hybrids in the Dodecad Project and they invariably have less amounts of Mongoloid components than the average pure Anatolian Turk. Only one Turkish member of Dodecad seems to be a pure Balkan Turk, and he/she seems to completely lack the Mongoloid components as far as I can see.

      Delete
    3. As for the "East Asian" alone issue, that depends on the genetic distance between the components. The "East Asian" and "North Asian" components of the DNA Tribes SNP analysis may be too close to each other to allow for substitution between each other, especially when they are minority components.

      Delete
    4. "Only one Turkish member of Dodecad seems to be a pure Balkan Turk, and he/she seems to completely lack the Mongoloid components as far as I can see."

      There are several Balkan Turks at Dodecad and they do have some of the East Asian component in K12b, e.g. DOD435 (4.6% East Asian and 1.4% Siberian), DOD611 (1.1% East Asian and 0% SIberian), and DOD835 (0.3% East Asian and 3.6% Siberian).

      "The "East Asian" and "North Asian" components of the DNA Tribes SNP analysis may be too close to each other to allow for substitution between each other"

      Yes, I agree. The Dodecad project had the same problems (see above in Balkan Turks).

      Delete
    5. There are several Balkan Turks at Dodecad and they do have some of the East Asian component in K12b, e.g. DOD435 (4.6% East Asian and 1.4% Siberian), DOD611 (1.1% East Asian and 0% SIberian), and DOD835 (0.3% East Asian and 3.6% Siberian).

      Are they full Balkan Turks, or are they recent Balkan Turk/Anatolian Turk hybrids? From what I understand reading the occasional ethnic and geographical details provided by Dienekes about the Turkish project participants used in his analyses, the latter is the case.

      Delete
    6. "Are they full Balkan Turks, or are they recent Balkan Turk/Anatolian Turk hybrids?"

      DOD435 is at least 7/8 from the Balkans. DOD435 maternal side is Bulgarian Turk, the paternal side of DOD435 is not clear, some kind of mix of Macedonian Albanian (Paternal grandfather), the father of the paternal grandmother is not known (Orphan), this could be Anatolian Turk.

      The paternal side of DOD107 (1.2% East Asian and 1.8% Siberian) is from the Balkans, the mother of DOD107 is DOD155 (1.4% East Asian and 2.4% Siberian), father's is not part of Dodecad.

      I don't know the details of the others.

      Delete
    7. Palisto, may I ask you where you got all these info from?

      In the Dodecad ancestry thread, of the Turkish participants you mentioned information exists only about DOD435 and DOD611.

      http://dodecad.blogspot.com/2010/11/information-about-project-samples.html

      What do DOD435 and DOD611 tell about themselves in the ancestry thread? In contrast to what you say, DOD435 says that all of his/her ancestry is Turkish (he/she surely means the known ancestry). He/she does not provide any information about his/her geographical origins, so it is not clear whether he/she is descended from Balkan Turks and how much. As for DOD611, he/she reports that most of his/her known ancestry is from the Balkans and Galicia, Ukraine (Crimean Tatar ancestry from Ukraine?), but adds that he/she may have some Anatolian roots as well. No info is available for DOD835, DOD107 and DOD155 in the Dodecad ancestry thread.

      Delete
    8. BTW, as far as I know, Ottomans did not settle in Galicia. So, the Galicia ancestry of DOD611, if Turkic (most probably so based on his/her inclusion in the Dodecad Project), is most probably from the Crimean Tatars there.

      Delete
    9. I don't know where you got this information about Galicia for DOD611.
      DOD611 is not mentioned at
      http://dodecad.blogspot.com/2010/11/information-about-project-samples.html


      If you had clicked the "Load more..." link at the bottom of the ancestry thread, you would have seen DOD611 and the info about his/her Galician roots.

      Delete
    10. So many of the "Turkish" samples in Dodecad are actually from recent Balkan origin. I am quite curious how Turkish individuals with less or no recent Balkan heritage would come out like. Do we have some samples with no recent Balkan heritage (where at least all four of the Grandparents are known to be from Anatolia)?

      Delete
    11. Okay, thank I see now. DOD611 is mixed: Turkish roots from paternally Bosnia, Northern Greece and Galicia/Ukraine and maternally Bosnia.

      Delete
    12. This comment has been removed by the author.

      Delete
    13. Kurti,

      Most Dodecad Turks are exclusively from Anatolia, as is clear from the fact that most of them fall in exclusively Anatolian clusters in cluster analyses. BTW, the Behar et al. 2010 Turkish samples are exclusively from the region of Anatolia historically called Cappadocia.

      Delete
    14. Palisto,

      I am still not convinced there is more than one Dodecad participant who is full Balkan Turk.

      Delete
    15. For instance, as I pointed out above, the Galicia ancestry of DOD661 most probably denotes Crimean Tatar ancestry.

      Delete
    16. "So many of the "Turkish" samples in Dodecad are actually from recent Balkan origin. I am quite curious how Turkish individuals with less or no recent Balkan heritage would come out like. Do we have some samples with no recent Balkan heritage (where at least all four of the Grandparents are known to be from Anatolia)?"

      I don' know. If I have to guess I would say:
      DOD433

      Delete
    17. I don' know. If I have to guess I would say:
      DOD433


      He/she is clearly more Mongoloid-admixed than the average full Anatolian Turk, even excluding the Black Sea region.

      Delete
    18. So DOD 433 with less East Asian input comes close to Turkish average?

      Delete
    19. "He/she is clearly more Mongoloid-admixed than the average full Anatolian Turk, even excluding the Black Sea region."

      "So DOD 433 with less East Asian input comes close to Turkish average?"

      I responded here:
      http://kurdishdna.blogspot.com/2012/08/anatolian-turks.html

      Delete
    20. This comment has been removed by the author.

      Delete
    21. So DOD 433 with less East Asian input comes close to Turkish average?

      Kurti, my answer is in the new thread that Palisto linked as well.

      Delete
    22. This comment has been removed by the author.

      Delete
    23. Kurti, I have now added a second reply post in the same thread, which continues from the first one.

      Delete
  6. Now thinking about it and its distribution THE BEST fitting name for this so called "Persian" component would be "Caspian"! since it obviously follows a root from Western to Eastern Caspian.

    ReplyDelete