Updated all haplogroup maps for Caucasus and Kurdistan

Maciamo

Veteran member
Admin
Messages
9,950
Reaction score
3,230
Points
113
Location
Lothier
Ethnic group
Italo-celto-germanic
The Caucasus region, one of the densest ethnic melting pot in the world, has long been an enigma for haplogroups. There were numerous studies (Wells et al. 2001, Nasidze et al. 2003 and 2004, Marchani et al. 2008) but all reported a lot of haplogroup F without testing for H or J1 (and indeed sometimes not even for G !), and K without testing for L or T. The same held true for Kurdistan. Fortunately a new study of the Caucasus by Balanovsky et al. (2011) as well as an in-depth analysis of J1 by Chiaroni et al. (2010) shed some light on the mystery. Most of the F that wasn't G was in fact J1, and most of it had nothing to do with the ubiquitous Arabic J1c3. The origin of J1 was placed in Kurdistan, around Lake Van.

You will find all the maps here.

You will see that I have put a lot of effort in painstakingly delimiting the areas for each ethnic group in the Caucasus and colourised the map for each of them one by one for every haplogroup.

I removed all I2a from the Caucasus, not knowing if there was any, as it seems that most of the Middle Eastern I is just I* or I2* (except in Kurdistan and Turkey). I checked for clues on the regional projects on FTDNA, but I could only find a substantial number of I members in the Armenian project, and out of 20 I's there was only one I2a2, but 14 I* or I2*. The 5 other members, very surprisingly, belonged to the Germanic I2b1c (P78+). I checked it up and noticed that this subclade was found all over Europe, from Russia to Ireland and to Spain, via Ukraine, Poland, Germany, France and Britain. It is also found in Greece and Turkey. I would be careful about the self-reported ancestry though since all five members were Armenians from Turkey (I would guess around Istanbul) or Crimea, not from Armenia itself. This probably explains everything (non-paternity event somewhere in the past). It also fits better with the distribution of I2b in regions settled by the Goths and Vikings.
 
I have seen the updated % of Haplogroups among Kurds from Turkish Kurdistan and I find this very realistic. But one thing is strange. The frequency of G2a has fallen drastically from 12 to 2%! I think this is really a bit low considering how high it is in the neighboring Populations and that one of the few tested Kurds on 23andme has came out as G2a5.

I think, from 9% of the "other". ~4% is R2a and the rest of the 5% are most probably G* and some N or R1b considering that Anatolian Kurds are closer to Caucasus and this would make it logical that they got more G than Iraqi Kurds (4%).

However this are only little things. I thought I should mention while there is a thread about it.

Overall the new percentages of Anatolian Kurds are very realistic. Another great work Maciamo.
 
Thank you very much for spending so much time on Kurdistan.
 
I have seen the updated % of Haplogroups among Kurds from Turkish Kurdistan and I find this very realistic. But one thing is strange. The frequency of G2a has fallen drastically from 12 to 2%! I think this is really a bit low considering how high it is in the neighboring Populations and that one of the few tested Kurds on 23andme has came out as G2a5.

I think, from 9% of the "other". ~4% is R2a and the rest of the 5% are most probably G* and some N or R1b considering that Anatolian Kurds are closer to Caucasus and this would make it logical that they got more G than Iraqi Kurds (4%).

However this are only little things. I thought I should mention while there is a thread about it.

Overall the new percentages of Anatolian Kurds are very realistic. Another great work Maciamo.

It is not a guess. I have the data from Nasidze et al.. I mistakenly placed the F samples under G before, but the Chiaroni study clearly confirmed that the most of the F could be classified as J1 (with perhaps 1% of H). Therefore, there is only 2.5% of G from Nasidze's study of Kurdistan. The 6% of so-called P1 are actually R2a, while the 5% of P* are probably Q.
 
It is not a guess. I have the data from Nasidze et al.. I mistakenly placed the F samples under G before, but the Chiaroni study clearly confirmed that the most of the F could be classified as J1 (with perhaps 1% of H). Therefore, there is only 2.5% of G from Nasidze's study of Kurdistan. The 6% of so-called P1 are actually R2a, while the 5% of P* are probably Q.

A question... could the Haplogroup "F" from the German Linear Pottery site be actually J1?
 
It is not a guess. I have the data from Nasidze et al.. I mistakenly placed the F samples under G before, but the Chiaroni study clearly confirmed that the most of the F could be classified as J1 (with perhaps 1% of H). Therefore, there is only 2.5% of G from Nasidze's study of Kurdistan. The 6% of so-called P1 are actually R2a, while the 5% of P* are probably Q.

your right now I see. I was just assuming that some of the "other" 9% which were not typed by Nasdize could actually be G2a*. But This is only guess gaming on my side.

A question... could the Haplogroup "F" from the German Linear Pottery site be actually J1?

This is possible. I have heard that they found in Tests some J1 around regions like Saxony.
 
The Haplogroup I STR cluster that Armenians belong to (minus the one Armenian I2a which is obviously different) has tested positive for L596 and L597, as have a couple of more distant clusters, making the chief Armenian subclade certainly I2c (as defined by ISOGG).

An I2c map would probably only show low levels of membership in the Middle East, the Caucasus, and Crete... interesting because its center of diversity is close to Germany, it just happens to be at <1% levels there.

"I*" and "I2*" are currently empty designations per Ken Nordtvedt and per Bob May.
 
Hello Maciamo. Would you please add Assyrians and Yezidis?

"Genetic Testing of Language Replacement Hypothesis in Southwest Asia"
LEVON YEPISKOPOSIAN, ASHOT HARUTYUNIAN, ARMINE KHUDOYAN
Arya International University, Yerevan
Institute of Man, Yerevan
2006

n=x 106
hgx Assyrian
hg1 42%
hg2 1%
hg3 2%
hg7 0%
hg8 0%
hg9 38%
hg16 0%
hg21 3%
hg26 15%
hg28 0%
hg29 0%

n=x 196
hgx Yezidi
hg1 28%
hg2 5%
hg3 2%
hg7 0%
hg8 0%
hg9 42%
hg16 0%
hg21 16%
hg26 6%
hg28 1%
hg29 0%
 

Attachments

  • PC_YEPISKOPOSIAN (1).jpg
    PC_YEPISKOPOSIAN (1).jpg
    34.1 KB · Views: 41
Nasidze Report(2004): Darginians of Dagestan 58% and Abkhazians 33% of Haplogroup I


Balankovsky Report(2011): Dargins 0% and Abkhazians 0 % of Haplogroup I




Who has right? Nasidze or Balankovsky?
 
Nasidze Report(2004): Darginians of Dagestan 58% and Abkhazians 33% of Haplogroup I


Balankovsky Report(2011): Dargins 0% and Abkhazians 0 % of Haplogroup I




Who has right? Nasidze or Balankovsky?

Balanovsky et al. because their study tested deep subclades with SNP's while Nasidze relied essentially on approximate and unreliable STR's for haplogroup determination.
 
The Caucasus region, one of the densest ethnic melting pot in the world, has long been an enigma for haplogroups. There were numerous studies (Wells et al. 2001, Nasidze et al. 2003 and 2004, Marchani et al. 2008) but all reported a lot of haplogroup F without testing for H or J1 (and indeed sometimes not even for G !), and K without testing for L or T. The same held true for Kurdistan. Fortunately a new study of the Caucasus by Balanovsky et al. (2011) as well as an in-depth analysis of J1 by Chiaroni et al. (2010) shed some light on the mystery. Most of the F that wasn't G was in fact J1, and most of it had nothing to do with the ubiquitous Arabic J1c3. The origin of J1 was placed in Kurdistan, around Lake Van.

You will find all the maps here.

You will see that I have put a lot of effort in painstakingly delimiting the areas for each ethnic group in the Caucasus and colourised the map for each of them one by one for every haplogroup.

I removed all I2a from the Caucasus, not knowing if there was any, as it seems that most of the Middle Eastern I is just I* or I2* (except in Kurdistan and Turkey). I checked for clues on the regional projects on FTDNA, but I could only find a substantial number of I members in the Armenian project, and out of 20 I's there was only one I2a2, but 14 I* or I2*. The 5 other members, very surprisingly, belonged to the Germanic I2b1c (P78+). I checked it up and noticed that this subclade was found all over Europe, from Russia to Ireland and to Spain, via Ukraine, Poland, Germany, France and Britain. It is also found in Greece and Turkey. I would be careful about the self-reported ancestry though since all five members were Armenians from Turkey (I would guess around Istanbul) or Crimea, not from Armenia itself. This probably explains everything (non-paternity event somewhere in the past). It also fits better with the distribution of I2b in regions settled by the Goths and Vikings.

Maciamo, I just want to thank you for the hard work you do by collecting and updating all these data and bettering them- I had no honey to offer you, do accept these "dry" thanks - good evening
 

This thread has been viewed 16493 times.

Back
Top