"Here is a something from Dienekes using Iraqi Kurdish samples from Xings et al.
http://dodecad.blogspot.de/2010/12/structure-in-west-asian-indo-european.html"
I heard this argument a lot. Let me briefly explain why this post of Dienekes is a bad example/evidence.
There is a very good reason why Dienekes never used the Xing et al dataset for comparison again. Kurti was referring to Dienekes' plot which is very noisy caused by the low number of compared SNPs.
It messed up the position of Kurds in the plot and all other populations in the plot. For example, just take a look at the Turks merging with Armenians in the plot. We now know this is not correct. There are major differences but the plot with the low K value could not see it in the relatively low number of SNPs. Dienekes never repeated any analysis with Xing et al because of that. However, Razib Khan, another science blogger started to use this noisy plot of Dienekes for his simplistic argument that Turks are just assimilated Armenians.
The Xing et al data are based on 250,000SNPs, but most of these SNPs are not part of other genetic studies, so a comparison is nearly impossible. Zack from the Harappa Ancestry project is trying to use Xing et al., but he has to reduce the number of SNPs down to 29,000 SNPs!
Zack wrote:
"This dataset is valuable because it contains several South Asian, Central Asian, Southeast Asian and Caucasian groups. However, it does not have a good SNP overlap with 23andme and the other datasets. It has only about 29,000 SNPs in common with 23andme v2 data. Combining HapMap, HGDP, SGVP, Behar et al and Xing et al with 23andme data leaves us with 25,000 SNPs. Due to that, I'll be using Xing et al data for only a few analyses."Zack also wrote:
"Do note that the Xing results were computed with a smaller number of SNPs and thus might be noisy."
Thanks for the clarification but what do you mean by this exactly
ReplyDelete"Kurti, one of the readers of this blog, mentioned that Kurds and Iranians are genetically different. As example/evidence, he presented this:"
Again it gives the impression as there is no difference.
So is there no difference between Kurds and Iranians?
If no could you please point out the similarities?
If yes could you please tell me which are the main differences. On Mc Donald´s maps Kurds usually appear between Iranians and Turks, Iranians and Georgians, Iranians and Armenians or once even Iranians and Adygai. Where would you plot the Kurds if you had such a map.
And about the part, Turks being acculturated Armenians. I have never believed this. It sounds to me ridiculous while historically it appears clearly that a huge part of Turks are in fact acculturated Kurds. In Yozgat for example according to Ottoman data. at least 50% of the population is of Kurdish origin. And there where also much more marriages between Kurds and Turks (both being Muslim populations and allies at least in the Ottoman empire).
ReplyDeleteThere is a small difference but not as big as in the figure of Dienekes you linked.
ReplyDeleteThis PCA plot is way better:
http://1.bp.blogspot.com/-UCP5T1pduGU/TzpBa9QbK3I/AAAAAAAAEe4/_uWuqnnb1zQ/s1600/1_2.png
from
http://dodecad.blogspot.com/2012/02/chromopainterfinestructure-analysis-of.html
I previously made this map from Dodecad K12b data. The map needs to be updated though.
https://maps.google.com/maps/ms?msid=212724382842737328910.0004bad4c0b64571b68c5&msa=0
"There is a small difference but not as big as in the figure of Dienekes you linked."
ReplyDeleteThe Plot shows actually very similar results as that figure I linked. However this Plot shows better the differences between the Armenians and Turks.
About these "small differences" If I remember well West Asian ethnic groups appear all very close to each other and its such "small differences" which make them appear distinct. So its all relative. However what I see on this plot is that Kurds seem to cluster as "distinct" to Iranians as they do to Armenians or Turks. While surprisingly Iranians seem to be closer to Turkic Groups this time.
Another thing. Do you know from where these "Iranian 19" samples come from. I cant believe that they are Iranian/Persian and suspect they might be Kurdish. Just look at them they cluster perfectly with the Kurdish samples and appear distinct from the Iranian_D samples. This might also be the reason why Dienekes dont use this samples anymore.
Between here is a new Dodecad Calculator (K10a)
https://docs.google.com/spreadsheet/ccc?key=0ArAJcY18g2GadC1kRjhxcHNfSGhPYlUxbEI0VVZPR0E#gid=0
"Another thing. Do you know from where these "Iranian 19" samples come from."
ReplyDeleteThese are the Iranians (N=20) from the Behar study and they are still in use, check row118 in your provided link.
Behar 2010 study:
http://dienekes.blogspot.com/2010/06/genome-wide-structure-of-jews-behar-et.html
Dienekes excluded one Iranian, he is working with 19 "Behar" Iranians.
"While surprisingly Iranians seem to be closer to Turkic Groups this time."
This is just one 2-D plot. The next one can look completely different.
A 2-dimensional plot can never show more than 2 dimensions. For K10a, 45 MDS 2-D plots are needed to show the whole complexity of relationships.
Behar Iranians are most probably from South and South west Iran this is because some of the individuals seem to have a higher South west asian component and some have african mixture they are probably from regions such as Bandar Abbas and Ilam.
ReplyDeleteNorth west iranians and west iranians would be the most similar to Kurds, where as Iranians from the east and south to south east would differ.
This comment has been removed by the author.
ReplyDelete"Dienekes excluded one Iranian, he is working with 19 "Behar" Iranians."
ReplyDeleteThan these are obviously misplaced on this map because they do even differ more from Kurds than the "Iranian_D" samples of which many are Big Lor and Azeri mixed Persian individuals.
"Behar Iranians are most probably from South and South west Iran"
The Behar Iranians are most probably "average" Iranians and sampled from Southwest, South and North/Northwest Iran. Its actually the Iranian_D samples which are biased towards North and Northwest (due many of the 9 tested individuals being Azeri, Big Lor/Kurdish etc. mixed)
"seem to have a higher South west asian component and some have african mixture they are probably from regions such as Bandar Abbas and Ilam."
Bandaris are highly mixed Arabs, Africans and Persian they would get much different results. And Ilam is very different from Bandar and actually a Kurdish region. That you mistake both of this regions as being very similar shows me that you dont have much knowledge about Iran after all. Ilamis are Feyli Kurds and cluster with Kurds from Kirmashan. While Bandaris are a complex mix of real Arab, African and Iranic components.
This are people from Bandar Abbas
http://www.bbc.co.uk/radio3/worldmusic/onlocation/media/iran21.jpg
And this Kurds from Ilam
http://rudaw.net/kurdish/files.php?file=cwan_268473355.jpg
The difference is huge. So assuming that both of them have similar genetics or that this Bandaris have such a West Asian genetic make up as the "Iranian" individuals of Behar makes no sense.
I am sure the Behar Iranians are samples token from Teheran, Fars, Yazd, Isfahan, Gilan and Mazandaran. While the Iranian_D samples are biased towards West/Northwest and North Iran.
I believe the Behar "Iranian" samples are misplaced on the plot because Dienekes tried to show the affinities them having more with Southwest Asian groups. And so he moved them more towards Assyrian and Armenian samples and ignored simplistically other differences. I mean just look at it, they appear on this plot the as distant as two different ethnic Groups would. So they ended up being between the Kurdish samples. Here I agree with Palisto a 2-D plot can never show perfectly the relationship of ethnic Groups.
ReplyDeleteNo they are not you have no evidence to say they that they were taken from all around Iran. The Dodecad Iranians are not biased so please do not talk of nonsense, their are no Kurds included in the Dodecad Iranians and only 1 lur is known.
ReplyDeleteJust like many samples such as Xing Kurds from iraq,and Kurds_Y from the Kazakhsatan who are all from the same region of group and same can be said about Iranians samples from Behar and likely were taken from one region, I doubt they travelled all around iran for samples.
What about the Arab region south west iran? Due to the african mixture it is safe to assume that the samples were taken from south iran even the south west asian cline of behar points to this. Also you are assuming they sampled the african group from bandar, their are iranians living all around bandar abbas who look normal iranian and likely have some mixture from those people or even Iranians with semetic ancestry from south Ilam. Becuase their are Kurds in Ilam does not mean that every group in Ilam would be the same.
Also Lors identify themselves as Iranians and even many as Persian so them included in the Dodecad is wrightful and no bias. They are geographly more closer to Pesians and even their language leans more towards Persian.
ReplyDelete"No they are not you have no evidence to say they that they were taken from all around Iran. The Dodecad Iranians are not biased so please do not talk of nonsense, their are no Kurds included in the Dodecad Iranians and only 1 lur is known."
ReplyDeleteAre you kidding? out of 9 individuals in Iranian_D samples I know at least two which are half Azeri. One of them was the User Rhino from Anthroscape.
"
What about the Arab region south west iran? Due to the african mixture it is safe to assume that the samples were taken from south iran even the south west asian cline of behar points to this. Also you are assuming they sampled the african group from bandar, their are iranians living all around bandar abbas who look normal iranian and likely have some mixture from those people or even Iranians with semetic ancestry from south Ilam. Becuase their are Kurds in Ilam does not mean that every group in Ilam would be the same."
My Friend significant African admixture was not only found in one individual but many. And I remember a study showing significant African admixture as far as in Mazandaran in North. I I find this study I will post it.
The Arab individuals of Kuzhestan are genetically very similar to Southwest Iranian groups and just linguistically assimilated.
"Also Lors identify themselves as Iranians and even many as Persian so them included in the Dodecad is wrightful and no bias. They are geographly more closer to Pesians and even their language leans more towards "
Where did I say they can not identify as Iranians? Are you some kind of Iranian advocate? I simply stated that Results of samples which where almost half of them taken from West/Northwest, while the large majority of Iranians live in North (Teheran), Isfahan Fars and Yazd, is biased.
Because some of the Iranian samples has some Azari ancestry does not mean it will change the results much. Their many I bet who are proper iranians. Some of the Kurds such as me and the other Kurd from my region of some non Kurdish ancestry, his 1/4 turkish I am 1/8 Circassian has this had massive impact on the Kurdsih average?
ReplyDelete""The Arab individuals of Kuzhestan are genetically very similar to Southwest Iranian groups and just linguistically assimilated""
So you believe the arab group had no genetic influence from the arabs? just language change? give me a break, they even look more arab then any Iranian I have seen and I am sure they had some genetic influence from Arabs.
I am no iranian advocate as I just state what I see. On Dodecad iranian the african mixture is lower, even the south west asian mixture is lower. Where as it higher among the Iranians of Behar, which I again I will say it safe to assume they came from the southern regions.
Iranians from Yazd and tehran would not have such high south west asian cline as it gets lower the more east you go.
Also I think Palisto might know where some of the Dodecad Iranians came from as he did something on ABF once to compare to Kurds, maybe he can say something about it.
"Also I think Palisto might know where some of the Dodecad Iranians came from as he did something on ABF once to compare to Kurds, maybe he can say something about it."
DeleteBoth DOD126 and 127 are mixed West-East Iranians from the same family (Hamadan and Khorasan).
Iranians seems to be equally related to Kurds and other Iranians. Iraqi Kurds have a lower distance to each other and are equally distant to Kurds from Turkey and Iranians.
Kurds from Turkey are more or less equally distant to Kurds from Iraq and Iranians.
Last year, I calculated distances based on Dodecad V3 (with less individuals):
Iranian Ir_Kurd Trk_Kurd
Iranians 6.3 6.9 8.2
Ir_Kurds 6.9 4.2 8.4
Trk_Kurd 8.4 8.4 4.1
Thanks for the share Palisto. They both being a mix between Western (Hamadan) and Eastern/Northeastern (Khorasan) Iran should come close to be representative for Iranians (excluding Kurds, most Azeris and Balochis in Southeast). And by looking at their data I feel somehow confirmed that most of the Iranian_D samples are biased towards North and West or mixed. And the Behar Iranian samples (excluding the two idividuals with above average African admixture) seem to come closest to Iranian average.
Deletecompare Iranian_D and Behar Iranian average with DOD 127 (mix of West and East Iranian).
DeleteDOD 127
West Asian 45.6%
Mediterranean 22.7%
Atl Baltic 12%
Red Sea 5.5%
South Asian 9.4%
Siberian 2.6%
Southeast Asian 2.1%
Iranian Behar
Iranians
West Asian 51%
Mediterranean 25.6%
Atl_Baltic 6.8%
Red Sea 5.8%
South Asian 8%
Siberian 1.1%
East Asian 0.3%
Southeast Asian 0.3%
Sub Saharan African 0.9%
Palaeoafrican 0.2%
Iranian_D
West Asian 52.1%
Mediterranean 25.3%
Atl Baltic 8.3%
Red Sea 4.4%
South Asian 6.8%
Siberian 1.5%
East Asian 0.8%
Southeast Asian 0.3%
Sub Saharan African 0.5%
By looking at the East and West Iranian mixed individual, its more than obvious to me that the Iranian_D samples are strongly biased toward North and Western Iran. While the Behar Iranian samples are probably taken as usual from somewhere in Teheran Isfahan or maybe Yazd and would come closest to average when we exclude the two samples which show above average African admixture.
"By looking at the East and West Iranian mixed individual, its more than obvious to me that the Iranian_D samples are strongly biased toward North and Western Iran."
DeleteKurti, you will hardly find Iranians from Central or East Iran because Central and East Iran is mostly desert and thus sparsely populated. In fact most Iranians are from Northwestern Iran:
http://media-3.web.britannica.com/eb-media/82/72782-004-0D1724AD.gif
And Teheran is a melting pot of Iranians mostly from the Western and Northwestern region.
"Kurti, you will hardly find Iranians from Central or East Iran because Central and East Iran is mostly desert and thus sparsely populated. In fact most Iranians are from Northwestern Iran:"
DeleteI know, With Central Iran I mean the areas of Isfahan and Yazd. And with East Iran I mean the areas of Khorasan, Semnan and Golestan which has a total Population of ~8.5 Mio. With South Iran I mean Fars which has a total populatio of 4.5 Mio. The density in Northwest is higher but thats also because the provinces are smaller. Teheran is a melting pot between all of these, from South, North, Northwest and Northeast.
And another thing I forgot to mention The areas with the highest density in West and Northwest Iran are either Kurdish(West Azerbaijan, Kordestan, Kirmashan, Ilam) or Azeri(East Azerbajan, Ardabil). So they play no big role in determining the Genetic average of Iranians. Of course it depends if you count Kurds and Azeris as "Iranians".
Delete"The density in Northwest is higher but thats also because the provinces are smaller."
DeleteThe population density is population size/km2, so the size of the province does not effect the population density.
"And another thing I forgot to mention The areas with the highest density in West and Northwest Iran are either Kurdish(West Azerbaijan, Kordestan, Kirmashan, Ilam) or Azeri(East Azerbajan, Ardabil). So they play no big role in determining the Genetic average of Iranians. Of course it depends if you count Kurds and Azeris as "Iranians"."
I did not count Iranians, Behar et al did, and I am pretty sure that "Iranians" not only includes Persians. I cannot speak for the Dodecad Iranians. In general, Iran is a mosaic of many ethnic groups, a lot of Iranian people have mixed backgrounds.
Palisto, I believe we are discussing minor differences. Actually If the Iranian_D samples are accurate or not was not even the topic. The differences between Behar Iranians and Iranian_D are so small that it doesnt matter. We somehow slipped into this while my main "concern" was that Dienekes did misplaced the Behar Iranian samples on his map. Even though the Iranian_D samples are closer to Kurds the Behar samples somehow were place into the Kurdish samples.
DeleteBoth Behar Iranian and Iranian_D are very similar and because I know some of the participation, I just think the Behar Iranian samples might be closer to the average Iranian. But this is all subjective.
If I had to make a map showing the relationships between Kurds , Iranians, Turks, Azeris etc. it would look like this
http://img5.fotos-hochladen.net/uploads/plotbcz6u154kd.jpg
Zack from the Harappa Ancestry Project presented the Behar Iranians here:
ReplyDeletehttp://www.harappadna.org/2011/03/iranians/
I think it is reasonable to assume that a lot Iranians in the Harappa Ancestry Project are also in the Dodecad project.
Zack wrote:
"The big difference between Harappa Project Iranians and Behar et al Iranians is African admixture. Only one Harappa Iranian (HRP0046) has 1% African admixture while three Behar Iranians have more than 10%."
This comment has been removed by the author.
ReplyDeleteBecause some of the Iranian samples has some Azari ancestry does not mean it will change the results much. Their many I bet who are proper iranians. Some of the Kurds such as me and the other Kurd from my region of some non Kurdish ancestry, his 1/4 turkish I am 1/8 Circassian has this had massive impact on the Kurdsih average?"
ReplyDeleteYou are always one of the first to mention those tiny 1/4 or 1/8 Foreign admixture in some Kurdish individuals. Yet you play down the impact of two 1/2 Azeris and one 1/2 Lor in a sample size of 9! compared to a sample size of 13 of Kurd_D.
"So you believe the arab group had no genetic influence from the arabs? just language change? give me a break, they even look more arab then any Iranian I have seen and I am sure they had some genetic influence from Arabs."
How many Iranian Arabs have you seen in your live? They look as much Arab as Nouri al Maliki does.
"Iranians from Yazd and tehran would not have such high south west asian cline as it gets lower the more east you go."
You sure about that? And what about Iranians from Fars and Isfahan?
The thing is that in the Behar Iranian samples there is not samples with really huge African admixture but the African is distributed evenly between all of them. What only shows me there cant have been Bandari samples among them, because real ethnic Bandaris would score at least 15-20% African.
And something more about the Lors. I dont know how much you know about them that you call them "closer to Persians". Neither historically, traditionally (even their dances are similar to Kurds) are they more similar to Persians. They are more likely Kurds who have adopted so much Persian substratum on their language that it became Southwest Iranic.
Here this map is just about the mtdna( I couldnt find the other study) but it as well gives you an Idea about the differences in Iran. As you can see you find African mtdna as far as North in Mazandaran.
http://img472.imageshack.us/img472/2081/iranmtdna2cu.jpg
Like I said I dont know how accurated the Iranian_D samples are but what I know for sure is that the Behar Iranian samples are not from Bandar Abbas.
Population portrait of Behar Iranian samples.
ReplyDeletehttp://img5.fotos-hochladen.net/uploads/admixtureirani19cpmjvfu3.png
The samples look relatively similar with just two individuals having ~5% more African admixture than the other. But the rest of their components are relatively the same as other West Asian groups. So it makes not much sense to assume they are of Bandari origin. I would rather believe that weak African admixture is something common among Iranians.
This comment has been removed by the author.
ReplyDeleteI am on my mobile, so I am linited. Having circassian and Turkish ancestry is not same as Azaris are just turkified iranics. Also notice that some Anatolian Kurds seem to lean towards Armenians which indicates Anatolian mixture. Look at Yezidi Kurd his actually comes even more closer to the behar Iranians then any of the kuds. When people consider yezidis the least mixed of the Kurds.
ReplyDeleteAlso why on earth would Iranians have low African? There is no historical reason for this apart from the Arab mixture from the south and Africans from Bandar.
like I said before north Iranians and west Iranians would be similar to Kurds and there would be some differences from other parts due to other mixtures and geography. Persians are just a mixture of native people and the iranics.
Have you looked at the map I posted which shows African presence even as far North as Mazandaran?
ReplyDeleteThe Yezidi Individual was just one. You need more to make an accurate statement also MC Donald positions the Kurds generally more Northeastern as they are. If you take all the samples and position them just a bit more Southwestern, interestingly you get almost a map of Kurdistan. Palisto did a better placing for the Yezidi individual which is somewhere in West Iran. It all depends on the Population you use as reference for an area. So the only reason Kurds tend to come out more Northeastern as they really are is because Mc Donald uses Armenians and Assyrians as only reference for Iraqi and Anatolian Kurdistan.
All Anatolian Kurds tested so far are Alevi (Alevi=oral stories about Khorasan).
And turkified "Azeri" is almost the same as a Turkified Circassian. Both are Caucasian groups by genetics.
Even some Kurds get some African around <1%. The african presence among the behar is unusualy high and that some of the individuals cline towards Arabs, shows that the samples were likely taken from southern iran where majority of the arab influence will be, the african part came from the Arabs, as I doubt the Bandar africans had much gentic influence on iranians. As I ask again why would you expect all Persians to have the african mixture? What historical reason for this?
ReplyDeleteYes all the Kurds of Antolia are Alevi. But I doubt all of them came from Iran, as some believe it is only the Kurds of Maras/Malatya are the ones who actually arrived from Iran due to their dialect of Kurmanji been more closer to that of Kurmanjis of Iran. Also we need some Sunni Kurds from Anatolia, but so far none of tested yet, I doubt a Kurd from Hakkari would differ much from a Kurd from Northern Iraq and Urmia, only the more west you go there might be some difference do to geographical location and mixture.
The Azaris of Iran are not same as the Azaris of the North. Azaris of iran descend from the ancient Iranics of NW iran, they spoke the ancient Iranic Azari language. Where as the North Azaris were known as Caucasian albanians untill they got Turkified.
And about the part, Turks being acculturated Armenians. I have never believed this. It sounds to me ridiculous while historically it appears clearly that a huge part of Turks are in fact acculturated Kurds. In Yozgat for example according to Ottoman data. at least 50% of the population is of Kurdish origin. And there where also much more marriages between Kurds and Turks (both being Muslim populations and allies at least in the Ottoman empire).
ReplyDeleteThere was no significant Kurdish population in Anatolia west of the eastern parts of Marash and Sivas and north of the southern parts of Erzurum and Ardahan during the Ottoman times and before, only some small enclaves that date largely to the very late Ottoman times. OTOH, Turkish-speaking parts of Anatolia was mostly Greek-speaking before the Turkification, and Armenians constituted a significant minority there already during the Byzantine times. Also, Ottomans frequently mention mixing with the invaded Christian natives, converting them to Islam in the process, in their accounts of invasions. Ottomans were just repeating an already established practice: external contemporary historical sources recount that Turkmens were mixing with and converting (by various ways) Christians in Anatolia in the locales they invaded. In addition, Anatolia stayed under Islamic rule long enough to permit conversion of bulk of its Christian population to Islam. Combining all these facts with genetics and what we know about the Turkmen migration to Anatolia from what is now Kazakhstan during the Seljuq times, your whole argument becomes invalid, anachronistic and untenable. But, as I said, Armenians are just a part of the whole story. As Dienekes succintly and successfully summarized: "Turkish speaking Anatolians are mostly Anatolian Greeks, except in the eastern provinces where they are Caucasian, Armenian, Iranic (Kurdish mainly), and Semitic (Assyrian-Arab)."
http://dienekes.blogspot.com/2011/09/who-are-anatolian-turks-reappraisal-of.html?showComment=1315082522236#c3358763458640138292
"As Dienekes succintly and successfully summarized: "Turkish speaking Anatolians are mostly Anatolian Greeks, except in the eastern provinces where they are Caucasian, Armenian, Iranic (Kurdish mainly), and Semitic (Assyrian-Arab).""
DeleteThe "average Turkish" admixture is best described as following (excluding Turks as reference using Oracle):
Distance %Population1 and %Population2
4.7945 57.1% Cypriots and 42.9% Turkmens_Y
5.3758 74.1% Armenian_D and 25.9% Uzbeks
5.4306 60.4% Assyrian_D and 39.6% Nogais_Y
5.6132 61.4% Azerbaijan_Jews and 38.6% Nogais_Y
5.8490 81.4% Azerbaijan_Jews and 18.6% Chuvashs
6.1464 56.5% Iraq_Jews and 43.5% Nogais_Y
However, the genetic variation is huge among Turks, so there is no "average Turk". Thus, some Turks appear genetically as Anatolian people, some as Balkan people, some as Caucasus people and a few as Kurdish or Assyrian people.
BTW, none of the Turkic-speaking populations you use are genetically representative of the original Turkmen population that invaded Anatolia from what is now Kazakhstan during the Seljuq times. I estimate that they genetically fell somewhere between Kazakhs and Uyghurs.
Delete"none of the Turkic-speaking populations you use are genetically representative of the original Turkmen population that invaded Anatolia from what is now Kazakhstan during the Seljuq times."
DeleteI agree, the original Turkmen population was different.
On the genetic side, Anatolian Turks are genetically too western to have acquired most or a big chunk of their genes from Armenians, Assyrians and/or Kurds. Anatolian Turks genetically position somewhere between Anatolian Greeks and Armenians, reflecting their ancestry from both.
ReplyDeleteAnatolian Turks genetically position somewhere between Anatolian Greeks and Armenians, reflecting their ancestry from both.
ReplyDeleteAnatolian Turks genetically position closer to the Anatolian Greek side rather than the Armenian side I'd say, based on the scant material available on the genetics of Anatolian Greeks.
Regarding the Anatolian Greeks I have not seen any genetic data, so I cannot really say that Turks tend to be more like Anatolian Greeks. It would be interesting to see genetic data of unmixed Pontic Greeks for this. My prediction is that Pontic Greek DNA is comparable to a mix of Cypriots and Armenians but a prediction does not replace data.
DeleteThere are no pure Anatolian Greeks in the Dodecad Project, but there are some Greeks who are Balkan Greek-Anatolian Greek hybrids, and they clearly lean to Anatolian Turks in autosomal analyses. If there was a pure Anatolian Greek in Dodecad, he/she would no doubt show up even closer to Anatolian Turks. Pontic Greeks are only a part of Anatolian Greeks and there may be some significant genetic variation among Anatolian Greeks. We'll see as more Greeks are sampled for genetic analyses.
Delete