PDA

View Full Version : Caucasus Results Comparison



Apex n Harmony
08-08-2017, 08:49 PM
It appears there isn't a thread dedicated to a comparison of results from users with roots in the Caucasus (Armenia, Azerbaijan, Georgia, and the North Caucasus) and nearby areas (e.g., Iran, Iraq, Kurdistan, Turkey). Let's use this thread to post our results from various vendors/websites. This will allow us to better understand the differences between vendors/websites (particularly with respect to this part of the world), and hopefully it will allow us to better understand our results.

Also, please share which GEDMatch calculators you consider most relevant for our purposes and why.

All four of my grandparents were born in Iran. My known ancestry on paper is very roughly 37.5% Gilaki, 25% Persian, 25% North Iranian, and 12.5% Kurdish.

AncestryDNA results: 91% Caucasus; 4% South Asia; 3% Middle East; 2% European Jewish.

MyHeritage results: 90% West Asia; 5.4% Middle East; 3.7% South Asia; 0.9% Ashkenazi Jewish.

DNA.Land results: 65% Central Indoeuropean; 13% Mediterranean Islander; 10% Indus Valley; 9% Indo-Iranian; 2.2% Kalash.

Saba123
08-08-2017, 10:47 PM
Nice you have one of the highest Caucasus results I've seen for an Iranian. Heres my results btw my known ancestry is 37.5% Bakhtiari 25% Qashqai 25% Fars Iranian 12.5% Unknown Iranian.
ANCESTRYDNA
Caucasus 73% Asia South 14% Middle East 10% Africa Southeastern Bantu 2% European Jewish < 1%
DNALAND
Central Indoeuropean 38%, Indo-Iranian 28%, Mediterranean Islander 29%, Indus Valley 2%, Kalash 1.7%, Ambiguous 1.7%
FTDNA
Asia Minor 71% Central Asian 21% East Middle East 3% West African 2% East Central Africa < 1% North and Central America < 1%West Middle East < 1%
MyHeritage
West Asia 89.2% South Asia 8.4% Ashkenazi Jewish 1.5% Middle Eastern 0.9%

asm
08-08-2017, 10:58 PM
I'm from Quba region of Azerbaijan.My family are mostly native Caucasians (Lezgins+ probably from other little Caucasian tribes)
I only know the family of my great-great grandfather (mother's side) came from Iran,for now don't know which city.I learnt about it last year.My family is fully Sunni muslims.Do you know can be Sunnis among Iranian Azeris or Persians in Iran?Maybe they are from those ethnic Caucasians who were exiled to Iran during the time of Nadir Shah?Or they are converts from Shia to Sunnism?What is your opinion?
Other fact that i know about my non-Caucasus origin is that it's said my father's paternal side are from Levant/Sham.But in my results i see almost no Arabic and Levant origin.

FTDNA

Asia Minor -74%
Eastern Europe -12 %
South Central Asia - 10%
British Isles - 2%
Siberia - 2%

MyHeritage

West Asian 64,8%
South Asian 6,3%
Greek 15,6%
Irish, Scottish, and Welsh 10,3%
Italian - 3,0%

DNALand

Central Indoeuropean 70%
Indo-Iranian 10%
Kalash 2.5%
Indus Valley 2.1%
Balkan 4.3%
South/Central European 3.8%
North Slavic 3.9%
Ashkenazi 2.2%
Northwest European 1.2%

Apex n Harmony
08-09-2017, 12:31 AM
I'm from Quba region of Azerbaijan.My family are mostly native Caucasians (Lezgins+ probably from other little Caucasian tribes)
I only know the family of my great-great grandfather (mother's side) came from Iran,for now don't know which city.I learnt about it last year.My family is fully Sunni Shafii muslims.Even my that great-grandfather was Shafii sheikh.Do you know can be Sunnis among Iranian Azeris or Persians in Iran?I don't think they can be Kurds.Maybe they are from those ethnic Caucasians who were exiled to Iran during the time of Nadir Shah?Or they are converts from Shia to Sunnism?What is your opinion?
Other fact that i know about my non-Caucasus origin is that it's said my father's paternal side are from Levant/Sham.But in my results i see almost no Arabic and Levant origin.

FTDNA

Asia Minor -74%
Eastern Europe -12 %
South Central Asia - 10%
British Isles - 2%
Siberia - 2%

MyOrigins

West Asian 64,8%
South Asian 6,3%
Greek 15,6%
Irish, Scottish, and Welsh 10,3%
Italian - 3,0%

DNALand

Central Indoeuropean 70%
Indo-Iranian 10%
Kalash 2.5%
Indus Valley 2.1%
Balkan 4.3%
South/Central European 3.8%
North Slavic 3.9%
Ashkenazi 2.2%
Northwest European 1.2%

There are certainly Sunnis throughout Iran, particularly in the Baluch, Turkoman, and Kurdish communities. I believe there are even some Sunnis among Iranian Azeris and particular Persian communities in the South. But there's no way I can tell what your ancestor's roots were. Conversion is certainly a possibility too, I'd imagine.

When you wrote "MyOrigins," did you mean to write "MyHeritage"? It's a bit interesting to me to see 6.3% South Asia. Iranians very often show South Asian in these tests. I wonder if it's typical of North and South Caucasians, too.

Apex n Harmony
08-09-2017, 12:39 AM
Nice you have one of the highest Caucasus results I've seen for an Iranian. Heres my results btw my known ancestry is 37.5% Bakhtiari 25% Qashqai 25% Fars Iranian 12.5% Unknown Iranian.
ANCESTRYDNA
Caucasus 73% Asia South 14% Middle East 10% Africa Southeastern Bantu 2% European Jewish < 1%
DNALAND
Central Indoeuropean 38%, Indo-Iranian 28%, Mediterranean Islander 29%, Indus Valley 2%, Kalash 1.7%, Ambiguous 1.7%
FTDNA
Asia Minor 71% Central Asian 21% East Middle East 3% West African 2% East Central Africa < 1% North and Central America < 1%West Middle East < 1%
MyHeritage
West Asia 89.2% South Asia 8.4% Ashkenazi Jewish 1.5% Middle Eastern 0.9%

Very cool. For what it's worth, I've noticed that Iranians on AncestryDNA most often show greater "Caucasus" when their roots are in the north and/or Caspian Sea area, and those same Iranians tend to show lower percentages of Middle East, South Asia, and Italy/Greece than Iranians from further south.

My father scored 96% Caucasus on AncestryDNA (2% Middle East; 2% South Asia). His family is mostly from Gilan.

asm
08-09-2017, 12:47 AM
When you wrote "MyOrigins," did you mean to write "MyHeritage"?

I corrected


It's a bit interesting to me to see 6.3% South Asia. Iranians very often show South Asian in these tests. I wonder if it's typical of North and South Caucasians, too.

Yes even aboriginal Caucasians score some South Asian.

https://www.nytimes.com/2014/02/14/science/tracing-ancestry-team-produces-genetic-atlas-of-human-mixing-events.html

jesus
08-09-2017, 01:52 AM
Very cool. For what it's worth, I've noticed that Iranians on AncestryDNA most often show greater "Caucasus" when their roots are in the north and/or Caspian Sea area, and those same Iranians tend to show lower percentages of Middle East, South Asia, and Italy/Greece than Iranians from further south.

My father scored 96% Caucasus on AncestryDNA (2% Middle East; 2% South Asia). His family is mostly from Gilan.

It's probably because Gilakis are similar to the reference Caucasus population used by AncestryDNA. You should download his data to GEDmatch. There no Gilaki samples on GEDmatch, only few Mazandaranis and Gorganis were tested from the caspian region.

jesus
08-09-2017, 01:57 AM
I corrected



Yes even aboriginal Caucasians score some South Asian.

https://www.nytimes.com/2014/02/14/science/tracing-ancestry-team-produces-genetic-atlas-of-human-mixing-events.html

It's due to shared ancestry with south asians. South asians have a lot of ancestry from Neolithic Iran and Caucasian hunter gatherers.

Saba123
08-09-2017, 03:26 AM
Yes this is true, it would only make sense considering the geographic proximities and considering that many Northerns are Azeri or some other type of Caucasian. It still kinda crazy how much we vary in Caucasus. My third Cousin, from my Bakhtiari side but he is only about 25% Bakhtari the rest Khuzestan Iranian, scored the lowest Caucasus percentage I have ever seen for an Iranian. His result: 53% Caucasian , 20% Asian South , 17% Middle Eastern, 5% Italian/Greece ,4% European Jew and 1% Great Britain.

kush
08-09-2017, 04:16 AM
This might be slightly off topic but I've seen a few afghan AncestryDna and 23andme results online and most of them get 70-80% south asian, and the iranians get around 5-10%. Funny thing is afghanistan is considered Central asia on ancestrydna, yet they get majority south asian. On gedmatch calculators, majority of afghans cluster closer to iranians than even most northwest south asian groups such as sindhis and punjabis. I'm just confused as to why afghans get majority south asian on those tests? shouldn't they have similar components as iranians with slightly higher south asian than iranians?

XooR
08-09-2017, 04:39 AM
First off all thank you Apex n Harmony for starting this thread. I was always curious about Caucasus component in results and the correlation between different vendors.

My all know ancestry is Laz from good old Lazistan. I pretty much tested myself with all major companies. At the same time I am trying to gather data specifically from the Lazistan region.

I will start typing up my personal results first and then I will attach a photo for collected results in excel format.

23andMe
Middle Eastern 89.1% Italian 7.9% Broadly European 2.3% Unassigned 0.5% Yakut 0.2%

ANCESTRYDNA
Caucasus 70% Italy/Greece 25% Middle East 4% European Jewish < 1%

DNALAND
Central Indoeuropean 81%, Mediterranean Islander 17%, Ashkenazi 1.7%

FTDNA
Asia Minor 100%

MyHeritage
West Asia 78.2% Greek 11.4% Italian 10.4%

DNATribes Regional Clusters
Pontic West Caucasus 35.9% Azeri-Kurdish-Persian-Turkish 23.2% Lebanese-Cyprus East Mediterranean 18.3% Ashkenazi Jewish 8.8% Aegean-Balkan-Italian 5.5% Mideastern Jewish 4.4% Bedouin-GulfArabs 3.8%

Gedmatch MDLP K-16 Modern Results

Admix Results (sorted):

# Population Percent
1 Caucasian 63.09
2 Steppe 14.00
3 Neolithic 12.18
4 NearEast 6.13
5 Indian 4.13


Finished reading population data. 517 populations found.
16 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Armenian_Armenia @ 4.930754
2 Armenian_Armenia @ 5.056004
3 Georgian_Laz @ 6.289100
4 Georgian_Megrelia @ 6.510499
5 Turk_Trabzon @ 6.649474
6 Georgians_Zugdidi @ 7.321535
7 Georgian_Georgia @ 7.367646
8 Georgian_Kakheti @ 8.264682
9 Georgian_Svanetia @ 8.399694
10 Abhkasian_Abkhasia @ 8.558256

I would like to see other members MDLP K-16 Modern Results as well for comparison.

And below is collected Turkish (mostly Laz) results from various vendors.


18038

Other than these there is only one Laz result from Geno 2.0 and his result was 100% Asia Minor.

Apex n Harmony
08-09-2017, 04:55 AM
It's probably because Gilakis are similar to the reference Caucasus population used by AncestryDNA. You should download his data to GEDmatch. There no Gilaki samples on GEDmatch, only few Mazandaranis and Gorganis were tested from the caspian region.

I agree that it's likely that AncestryDNA's reference panel for the Caucasus includes Gilakis. However, I've also read that Gilakis and Mazandaranis (who are closely related) likely have roots in the South Caucasus.

I did upload my results and my father's results to GEDmatch and I ran them through virtually every calculator. But I don't know which calculators are most relevant, and I don't know which ones are telling me recent ancestry as opposed to deep ancestry. (I know that some calculators, like Eurogenes K36, explicitly focus on deeper ancestry.)

Would you mind telling me which calculators are best suited for Iranians, and whether they focus on recent or deep ancestry? I'd be happy to share results.

Apex n Harmony
08-09-2017, 05:18 AM
It's due to shared ancestry with south asians. South asians have a lot of ancestry from Neolithic Iran and Caucasian hunter gatherers.

I'm aware of this shared ancestry, but I have a question that hopefully you can answer (I know nothing about genetics or ancestry, so bear with me): If Caucasians share ancestry with South Asians, shouldn't that shared ancestry be present in both Caucasians and South Asians? If not, why not? And if so, why would AncestryDNA match that segment of your DNA to South Asia and not the Caucasus, if it's shared and thus present in both communities?

XooR
08-09-2017, 12:08 PM
I agree that it's likely that AncestryDNA's reference panel for the Caucasus includes Gilakis. However, I've also read that Gilakis and Mazandaranis (who are closely related) likely have roots in the South Caucasus.

I did upload my results and my father's results to GEDmatch and I ran them through virtually every calculator. But I don't know which calculators are most relevant, and I don't know which ones are telling me recent ancestry as opposed to deep ancestry. (I know that some calculators, like Eurogenes K36, explicitly focus on deeper ancestry.)

Would you mind telling me which calculators are best suited for Iranians, and whether they focus on recent or deep ancestry? I'd be happy to share results.

Apex you can start sharing your and your fathers MDLP K16 Modern and Near East Neolithic K13 results?
Thank you.

Apex n Harmony
08-09-2017, 03:32 PM
Apex you can start sharing your and your fathers MDLP K16 Modern and Near East Neolithic K13 results?
Thank you.

Sure. Would you please briefly explain in lay terms what these two particular calculators aim to reveal exactly? What is their purpose?

MY MDLP K16 MODERN RESULTS:

Admix Results (sorted):

# Population Percent
1 Caucasian 50.54
2 Indian 16.83
3 Steppe 14.41
4 Neolithic 7.48
5 NearEast 6.18
6 Siberian 1.15
7 NorthEastEuropean 1.02

Least-squares method.

Using 1 population approximation:
1 Iran_Lor_Lor @ 6.241778
2 Iranian_Fars @ 6.313917
3 Kurd_Syria @ 6.324459
4 Zoroastrian_Iran @ 6.471915
5 Iranian_Iran @ 6.564443

Using 2 populations approximation:
1 50% Georgian_Kakheti +50% Iran_Mazandrani_Mazandrani @ 3.894845

Using 3 populations approximation:
1 50% Armenian_Armenia +25% Georgian_Svanetia +25% Pathan_Punjab @ 2.418469

Using 4 populations approximation:
1 Armenian_Lebanon + Armenian_Armenia + Brahui_Baluchistan + Chechen_Chechnya @ 2.129056


MY FATHER'S MDLP K16 MODERN RESULTS:

Admix Results (sorted):

# Population Percent
1 Caucasian 53.44
2 Indian 17.69
3 Steppe 13.38
4 Neolithic 7.06
5 NearEast 6.17

Least-squares method.

Using 1 population approximation:
1 Iran_Mazandrani_Mazandrani @ 7.475444
2 Iran_Lor_Lor @ 7.632030
3 Iranian_Fars @ 7.756958
4 Iranian_Iran @ 7.765945
5 Azerbaijani_Baku @ 7.784318

Using 2 populations approximation:
1 50% Georgian_Kakheti +50% Iran_Mazandrani_Mazandrani @ 2.309063

Using 3 populations approximation:
1 50% Georgian_Kakheti +25% Georgian_Svanetia +25% Makrani_Pakistan @ 2.038242

Using 4 populations approximation:
1 Armenian_Armenia + Brahui_Baluchistan + Georgian_Kakheti + Georgian_Svanetia @ 1.675165


MY NEAR EAST NEOLITHIC K13 RESULTS:

Admix Results (sorted):

# Population Percent
1 IRAN_NEOLITHIC 36.06
2 CHG_EEF 22.33
3 NATUFIAN 12.22
4 ANATOLIA_NEOLITHIC 8.75
5 EHG 7.67
6 ANCESTRAL_INDIAN 5.54
7 SE_ASIAN 3.01
8 PAPUAN 1.55
9 SIBERIAN 1.31

Least-squares method.

Using 1 population approximation:
1 Iranian @ 4.332438
2 Iranian_Shirazi @ 6.471955
3 Azeri @ 6.623166
4 Iranian_Lori @ 6.632074
5 Kurd_C @ 8.738378

Using 2 populations approximation:
1 50% Azeri +50% Iranian @ 4.304070

Using 3 populations approximation:
1 50% Assyrian +25% Kalash +25% Kumyk @ 3.102138

Using 4 populations approximation:
1 Assyrian + Brahui + Chechen + Turkish_Kayseri @ 2.706299


MY FATHER'S NEAR EAST NEOLITHIC K13 RESULTS:

Admix Results (sorted):

# Population Percent
1 IRAN_NEOLITHIC 36.79
2 CHG_EEF 23.83
3 NATUFIAN 11.66
4 ANATOLIA_NEOLITHIC 9.73
5 EHG 7.61
6 ANCESTRAL_INDIAN 5.95
7 KARITIANA 1.65
8 SIBERIAN 1.21

Least-squares method.

Using 1 population approximation:
1 Iranian @ 4.912882
2 Iranian_Lori @ 6.859886
3 Iranian_Shirazi @ 7.255656
4 Azeri @ 7.385113
5 Kurd_C @ 8.693030

Using 2 populations approximation:
1 50% Iranian +50% Iranian @ 4.912882

Using 3 populations approximation:
1 50% Armenian +25% Chechen +25% Kurd_SE @ 2.751393

Using 4 populations approximation:
1 Assyrian + Balochi + Chechen + Turkish_Trabzon @ 2.399197

XooR
08-09-2017, 05:03 PM
Sure. Would you please briefly explain in lay terms what these two particular calculators aim to reveal exactly? What is their purpose?

Most of the time creator of the calculator include some basic info and explanation regarding their purpose. I assume all of the DNA companies and calculator creators aim to create a perfect admixture (ethnicity breakdown) that works for everyone.

Below is what MDLP 16 says, also the blog for this project is found here: http://magnusducatus.blogspot.com/

" The K16 model of Admixture K16 focuses primarily on 16 worldwide basic, distich components of modern human ancestry, which have been discovered and detailed in Haak et al. (2014), Lazaridis et al. (2016). These components were empirically 'learned' in ADMIXTURE software from allele frequencies of learning merged dataset (116463 SNPs) of human populations, which, in their turn, had been converted into 'synthetic groups of individuals', with each of them representing one (of 16) ancestral populations. Then, the rest of dataset ('reference') was projected unto those ancestral populations with SNPWEIGHTS software: the results of component projection were averaged per each modern group of human population, and taken as 'population reference values' of component membership in 2xOracle and OracleX4 "


And below is Near east Neolithic K13, as far as I know this project focuses primarily Eurasian (especially West Asian) and ancient DNA. There is no website, but you can contact the creator at [email protected]

" The Near East Neolithic 13 calculator is based on the recently recovered ancient genomes from the Near East. The genomes were described in a recent 2016 paper, <93>The genetic structure of the world's first farmers<94> by Losif Lazaridis et al. (http://biorxiv.org/content/early/2016/06/16/059311). For further information please reference the paper.

The allele frequencies for the calculator's components are based on a combination of both extant samples, as well as ancient genomes described below. "



I find Gedmatch calculators more accurate because first, they don't have any concern about money and most of the calculators are created by academics. Secondly, it allows you to analyze your data and compare with others & reference populations as well. I pick these two because MLDP 16 happen to be the latest and Neolithic K13 breaks down CHG and Iran Neolithic so we can see differentiation between the region which we can not see through commercial company results because they give you a rough region and percentages, on the other hand Oracle function in gedmatch attempts to pinpoint your origins to a more specific population or region. Even though results are speculative (because we do not have ancient DNA from every region yet), I have found they do make sense, and in some cases, can be remarkably accurate.

Apex n Harmony
08-09-2017, 05:14 PM
Most of the time creator of the calculator include some basic info and explanation regarding their purpose. I assume all of the DNA companies and calculator creators aim to create a perfect admixture (ethnicity breakdown) that works for everyone.

Below is what MDLP 16 says, also the blog for this project is found here: http://magnusducatus.blogspot.com/

" The K16 model of Admixture K16 focuses primarily on 16 worldwide basic, distich components of modern human ancestry, which have been discovered and detailed in Haak et al. (2014), Lazaridis et al. (2016). These components were empirically 'learned' in ADMIXTURE software from allele frequencies of learning merged dataset (116463 SNPs) of human populations, which, in their turn, had been converted into 'synthetic groups of individuals', with each of them representing one (of 16) ancestral populations. Then, the rest of dataset ('reference') was projected unto those ancestral populations with SNPWEIGHTS software: the results of component projection were averaged per each modern group of human population, and taken as 'population reference values' of component membership in 2xOracle and OracleX4 "


And below is Near east Neolithic K13, as far as I know this project focuses primarily Eurasian (especially West Asian) and ancient DNA. There is no website, but you can contact the creator at [email protected]

" The Near East Neolithic 13 calculator is based on the recently recovered ancient genomes from the Near East. The genomes were described in a recent 2016 paper, <93>The genetic structure of the world's first farmers<94> by Losif Lazaridis et al. (http://biorxiv.org/content/early/2016/06/16/059311). For further information please reference the paper.

The allele frequencies for the calculator's components are based on a combination of both extant samples, as well as ancient genomes described below. "



I find Gedmatch calculators more accurate because first, they don't have any concern about money and most of the calculators are created by academics. Secondly, it allows you to analyze your data and compare with others & reference populations as well. I pick these two because MLDP 16 happen to be the latest and Neolithic K13 breaks down CHG and Iran Neolithic so we can see differentiation between the region which we can not see through commercial company results because they give you a rough region and percentages, on the other hand Oracle function in gedmatch attempts to pinpoint your origins to a more specific population or region. Even though results are speculative (because we do not have ancient DNA from every region yet), I have found they do make sense, and in some cases, can be remarkably accurate.

Yeah, I'd read the summaries for the calculators but I wasn't entirely sure what they meant. In short, do both of these calculators focus on ancient ancestry as opposed to recent ancestry? I think the "Modern" in MDLP K16 Modern is throwing me off...

XooR
08-09-2017, 05:19 PM
Below are MDLPK16 and Neolithic K13 Laz results

18042

18043

XooR
08-09-2017, 07:26 PM
Yeah, I'd read the summaries for the calculators but I wasn't entirely sure what they meant. In short, do both of these calculators focus on ancient ancestry as opposed to recent ancestry? I think the "Modern" in MDLP K16 Modern is throwing me off...

As far as I know they are not primarily focused on ancient ancestry, they use ancient dna data as reference to construct their calculator or depending on their goal they combine data from ancients and moderns to create a perfect admixture. Like I said before results are speculative no one can claim that they have 100% definite results (yet). In MDLP K16 case, Caucasian is defined as - a major component of ancestry of modern inhabitants of Caucasus, Iran and northern Indian : it was derived from genomes of mesolithic Caucasian Hunter-gatherers: a major ancestral component linked to CHG was carried west and east by migrating herders from the Eurasian Steppe.

There are specifically constructed calculators that shows your affinity with ancient dna references. For example;

The puntDNAL K12 ancient calculator utilizes public data from the Reich Lab in Harvard, the Estonian Biocentre, and data from my personal collection in order to create ancient components, which include the following: Anatolian neolithic farmer, European hunter-gatherer, and Caucasus hunter-gatherer. The European hunter-gatherer component is a newer component that is composed of all three European hunter-gatherers, such as Western hunter-gatherer, Eastern hunter-gatherer, and Scandinavian hunter-gatherer. Furthermore, I divided this calculator into two versions: the puntDNAL K12a version utilizes ancient oracle and the puntDNAL K12m version utilizes modern oracle. Questions and comments about this calculator should be directed to Abdullahi Warsame at [email protected]

or recently became available Neolithic & Bronnze Age K11 Admixture Calculator; (not in Gedmatch it is available in http://www.geneplaza.com)

The above K11 calculator based on rarer alleles (previously discussed in other threads) is now available at http://www.geneplaza.com along with a comprehensive explanation of the calculator and its components. Since the algorithm for this calculator is different from other ADMIXTURE based calculators, such as those on Gedmatch.com, your results will also differ.

SSA and E Asian levels should be closer to your actual levels since they are measured relative to the less SSA and E Asian admixed ancient references, whereas in your typical ADMIXTURE calculator, those levels are relative to modern populations which have received pulses of SSA and E Asian admixture post Bronze Age/Neolithic.

jesus
08-09-2017, 09:05 PM
There are certainly Sunnis throughout Iran, particularly in the Baluch, Turkoman, and Kurdish communities. I believe there are even some Sunnis among Iranian Azeris and particular Persian communities in the South. But there's no way I can tell what your ancestor's roots were. Conversion is certainly a possibility too, I'd imagine.

When you wrote "MyOrigins," did you mean to write "MyHeritage"? It's a bit interesting to me to see 6.3% South Asia. Iranians very often show South Asian in these tests. I wonder if it's typical of North and South Caucasians, too.

Some Talysh people are Sunni as well.

Apex n Harmony
08-10-2017, 01:54 AM
This might be slightly off topic but I've seen a few afghan AncestryDna and 23andme results online and most of them get 70-80% south asian, and the iranians get around 5-10%. Funny thing is afghanistan is considered Central asia on ancestrydna, yet they get majority south asian. On gedmatch calculators, majority of afghans cluster closer to iranians than even most northwest south asian groups such as sindhis and punjabis. I'm just confused as to why afghans get majority south asian on those tests? shouldn't they have similar components as iranians with slightly higher south asian than iranians?

This is really just a guess, but I would chalk this up to reference panels. AncestryDNA only includes 26 samples from the entire Asia Central region, which (as you correctly point out) includes Afghanistan. Afghanistan is extremely ethnically diverse, as is that entire region. Who knows where AncestryDNA's 26 samples came from? Who knows which (or how many) ethnic communities in Afghanistan were actually included in that set of samples? It's possible that the Afghan results you're referring to derive from Afghans whose ethnic communities were not included in the Central Asia reference panel.

Kurd
08-10-2017, 02:18 AM
This might be slightly off topic but I've seen a few afghan AncestryDna and 23andme results online and most of them get 70-80% south asian, and the iranians get around 5-10%. Funny thing is afghanistan is considered Central asia on ancestrydna, yet they get majority south asian. On gedmatch calculators, majority of afghans cluster closer to iranians than even most northwest south asian groups such as sindhis and punjabis. I'm just confused as to why afghans get majority south asian on those tests? shouldn't they have similar components as iranians with slightly higher south asian than iranians?

Any test that shows Afghans as 70-80% S Asian, bases the S Asian component on Afghan references in addition to Indian references So in effect the test is saying that the Afghan is 70-80% Afghan, which of course is not very informative to genetic substructure if you happen to be an Indian or Afghan.

For example, I can make a Kurd 70-80% S Asian if I made Kurds also S Asian references. Wouldn't make sense of course but that is certainly possible. With ADMIXTURE based calculators such as those on Gedmatch the problem is of a different nature. Those calculators underestimate differences between divergent populations, and overestimate similarity between populations in close geographic proximity sharing SOME drift. As an example, I can have W Asians score about 10% -15 % E Asian when I have only a couple of testers in the run, however, as soon as I throw in some E Asian testers, those same W Asians will score only 0-3% E Asian. In very simplistic terms, they are basically expelled from the E Asian comp by the E Asian test subjects.

That is the reason I put together calculators based on ancients sources. The reality is there is high genetic continuity all the way from NW India to Kurdistan. Differences are based on a very small number of alleles, which comprise a tiny percentage of alleles derived in Eurasians.

Apex n Harmony
08-13-2017, 05:59 PM
Nice you have one of the highest Caucasus results I've seen for an Iranian. Heres my results btw my known ancestry is 37.5% Bakhtiari 25% Qashqai 25% Fars Iranian 12.5% Unknown Iranian.
ANCESTRYDNA
Caucasus 73% Asia South 14% Middle East 10% Africa Southeastern Bantu 2% European Jewish < 1%
DNALAND
Central Indoeuropean 38%, Indo-Iranian 28%, Mediterranean Islander 29%, Indus Valley 2%, Kalash 1.7%, Ambiguous 1.7%
FTDNA
Asia Minor 71% Central Asian 21% East Middle East 3% West African 2% East Central Africa < 1% North and Central America < 1%West Middle East < 1%
MyHeritage
West Asia 89.2% South Asia 8.4% Ashkenazi Jewish 1.5% Middle Eastern 0.9%

I see that in another thread you mentioned that you've seen many Iranians show "Arab" in their DNA.Land results. That's very surprising to me. DNA.Land's map shows that no samples were taken from Iraq, which is where I always imagined Iranians' "Middle East" oriented results derived from. I figured that would explain the absence of any "Arab" showing in my DNA.Land results, my brother's DNA.Land results, and my parents' DNA.Land results, even though we all shows varying degrees of "Middle East" in our Ancestry and MyHeritage results (anywhere from 2% to 8%).

Also, I know "Mediterranean Islander" on DNA.Land has thrown many people off. For what it's worth, I fell smack in the middle of my parents' results: my dad showed 1.4%, my mom showed 29%, and I showed 13%.

Again, my dad's family almost entirely comes from Gilan (Caspian Sea area), while my mom's family is likely a mix of Caucasian, Iranian Azeri, Kurdish, and Persian.

ancestryfan1994
08-13-2017, 06:05 PM
I see that in another thread you mentioned that you've seen many Iranians show "Arab" in their DNA.Land results. That's very surprising to me. DNA.Land's map shows that no samples were taken from Iraq, which is where I always imagined Iranians' "Middle East" oriented results derived from. I figured that would explain the absence of any "Arab" showing in my DNA.Land results, my brother's DNA.Land results, and my parents' DNA.Land results, even though we all shows varying degrees of "Middle East" in our Ancestry and MyHeritage results (anywhere from 2% to 8%).

Also, I know "Mediterranean Islander" on DNA.Land has thrown many people off. For what it's worth, I fell smack in the middle of my parents' results: my dad showed 1.4%, my mom showed 29%, and I showed 13%.

Again, my dad's family almost entirely comes from Gilan (Caspian Sea area), while my mom's family is likely a mix of Caucasian, Iranian Azeri, Kurdish, and Persian.

The Arab ancestry in Iran could come from various sources. Mostly its Iraqi/Bedouin like you said. In my families case, its mostly due to recent Egyptian ancestry. Theres a few Arab tribes in Ahwaz, but also a Lebanese minority in Iran from centuries back mostly located in and around Qom.

Apex n Harmony
08-13-2017, 10:47 PM
The Arab ancestry in Iran could come from various sources. Mostly its Iraqi/Bedouin like you said. In my families case, its mostly due to recent Egyptian ancestry. Theres a few Arab tribes in Ahwaz, but also a Lebanese minority in Iran from centuries back mostly located in and around Qom.

You're certainly right that there are Arabs in Iran. But in my comment I was referring to Iranians who don't identify as Arab, don't speak Arabic as a first language, and have no known Arab ancestry. Those Iranians still show some "Middle East" or "Arab" in many tests, and I figured that's a result of matching samples taken from present-day Iraq (using the DNA.Land results I shared as a basis for my conclusion). I could be wrong, though.

Saba123
08-14-2017, 05:52 AM
I see that in another thread you mentioned that you've seen many Iranians show "Arab" in their DNA.Land results. That's very surprising to me. DNA.Land's map shows that no samples were taken from Iraq, which is where I always imagined Iranians' "Middle East" oriented results derived from. I figured that would explain the absence of any "Arab" showing in my DNA.Land results, my brother's DNA.Land results, and my parents' DNA.Land results, even though we all shows varying degrees of "Middle East" in our Ancestry and MyHeritage results (anywhere from 2% to 8%).

Also, I know "Mediterranean Islander" on DNA.Land has thrown many people off. For what it's worth, I fell smack in the middle of my parents' results: my dad showed 1.4%, my mom showed 29%, and I showed 13%.

Again, my dad's family almost entirely comes from Gilan (Caspian Sea area), while my mom's family is likely a mix of Caucasian, Iranian Azeri, Kurdish, and Persian.

Yeah your are my fellow 0% Arab lol, I seriously have no clue what kind of Arab DNA they have that caused them to show up as Arab while we have 0%. It could be that our population avoided Arab invasions/intermixing better or something, I have heard that Shomal held off Arabs the longest so that seems like a reasonable explanation. But my Bakhtiari family has lived near Iraq for hundreds of years and Qashqais had tribes that were of Arab origin, so my results still hard to explain but seem kinda consistent with the fact that I always come up pretty low for Arab DNA. And my very high Med Islander is another oddity that I do not know the answer to, my only guess is that it has something to with Cyprus and maybe my Turkish ancestry?

Alkaevli
08-19-2017, 12:46 PM
My father

FTDNA
https://i.hizliresim.com/3qnb8O.png

MyHeritage
https://www.myheritage.com/dna/ethnicity/intro/EPT5IK34CPOMUIRA88O5EL9JE9554LBK9COJ6PA68OR7GIRB99 SN6LAFDH1M2LQ9D9MJ4GQKD98LAK2A6HOKOPRKE1576E9BC9PK 2Q0
https://i.hizliresim.com/JljZVB.png

DNA.Land
https://i.hizliresim.com/7N48qL.png (https://hizliresim.com/7N48qL)


CHG admixture (puntDNAL K12)

Admix Results (sorted):

# Population Percent
1 Caucasus_HG 40.24
2 Anatolian_NF 23.68
3 Near_East 11.86
4 European_HG 10.22
5 Siberian 4.86
6 South_Asian 3.98
7 East_Asian 3.91
8 Amerindian 0.42
9 South_African_HG 0.42
10 Oceanian 0.29
11 Sub-Saharan 0.07
12 Beringian 0.05

Single Population Sharing:

# Population (source) Distance
1 Turkish 5.59
2 Kurdish 8.92
3 Kumyk 9.49
4 Turkish_Kayseri 10.06
5 Iranian 10.86
6 Turkish_Aydin 11.23
7 Balkar 12.2
8 Laz 12.82
9 Chechen 13.15
10 Turkish_Trabzon 13.48
11 Assyrian 13.9
12 Adygei 13.98
13 Nogai 14.03
14 Georgian_Jew 14.29
15 North_Ossetian 14.54
16 Armenian 14.77
17 Iranian_Jew 15.6
18 Lezgin 18.66
19 Iraqi_Jew 18.74
20 Georgian 19.16

Apex n Harmony
08-20-2017, 08:13 PM
Has anyone else found that his/her Dodecad k12b and HarappaWorld results are almost identical? Here are my results (Iranian):

HarappaWorld
43.47 % - Caucasian
25.91 % - Baloch
11.71 % - Southwest Asian
7.03 % - Northeast European
4.58 % - Mediterranean
3.45 % - South Indian
1.27 % - Siberian

Dodecad K12b
41.33 % - Caucasus
26.40 % - Gedrosia
11.96 % - Southwest Asian
7.66 % - North European
6.12 % - Atlantic Mediterranean
4.04 % - South Asian
1.69 % - Siberian

Also, does anyone know whether these calculators might exaggerate the Gedrosia/Baloch component for non-South Asians? (I know HarappaWorld's focus is South Asians.)

Eurogenes k12b appears to give similar results, although the Gedrosia/Baloch component is swapped for West Central Asian, which likely encompasses different peoples:

Eurogenes K12b
38.12 % - Caucasian
21.78 % - West Central Asian
12.99 % - Mediterranean
11.78 % - Southwest Asian
6.03 % - North European
4.80 % - South Asian
2.33 % - West European
1.80 % - Siberian

Kurd
08-21-2017, 12:19 AM
Has anyone else found that his/her Dodecad k12b and HarappaWorld results are almost identical? Here are my results (Iranian):

Also, does anyone know whether these calculators might exaggerate the Gedrosia/Baloch component for non-South Asians? (I know HarappaWorld's focus is South Asians.)


The Gedrosia/Baloch signal is nothing more than shared mutations at a few positions in Baloch/Brahui/Makranis. The fact that these mutations are shared with Kurds/W Iranians at a substantial subset of positions and the fact that frequencies of alleles for Baloch/Brahui/Makrani overlap allele frequencies for many positions For Kurds and W Iranians suggest not too distant a past when Kurds/W Iranians - Baloch/Brahui/Makrani diverged. What diffrentiates Kurds/W Iranians and Baloch is a little higher ASI type signal in the latter which was picked up after arriving at their present locations.

The genetic connections between Kurds/W Iranians and NW Indians are very substantial. In addition to both populations being substantially Neolithic Iran N (recovered in Kurdistan region) derived, they also share more recent substructure likely mediated by west to east migrations of groups such as proto-Kurd/W Iranian derived groups such as ancestral Baloch over the past couple of thousand of years

Saba123
08-21-2017, 01:34 AM
HarrapaWorld
# Population Percent
1 Caucasian 39.18
2 Baloch 28.33
3 SW-Asian 12.31
4 Mediterranean 5.18
5 S-Indian 5.06
6 NE-Euro 4.77
7 American 1.68
8 W-African 1.63
9 SE-Asian 0.63
10 E-African 0.5
11 Pygmy 0.3
12 Siberian 0.23
13 Beringian 0.14
14 San 0.06

Single Population Sharing:

# Population (source) Distance
1 iranian (behar) 3.48
2 iranian (harappa) 3.99
3 kurd (harappa) 4.66
4 kurd (xing) 6.4
5 kurd (yunusbayev) 7.02
6 turkish (harappa) 11.24
7 azeri (harappa) 11.58
8 uzbekistan-jew (behar) 11.78
9 armenian (harappa) 13.1
10 iraqi-arab (harappa) 13.44


Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 86.6% kurd (xing) + 13.4% up-muslim (harappa) @ 2.24
2 88.8% kurd (xing) + 11.2% bengali-brahmin (harappa) @ 2.26
3 90% kurd (xing) + 10% caribbean-indian (harappa) @ 2.27
4 89.3% kurd (xing) + 10.7% up-kshatriya (metspalu) @ 2.27
5 90% kurd (xing) + 10% sri-lankan (harappa) @ 2.28
6 89.2% kurd (xing) + 10.8% vaish (reich) @ 2.28
7 89.3% kurd (xing) + 10.7% karnataka-brahmin (harappa) @ 2.29
8 89.7% kurd (xing) + 10.3% ap-brahmin (xing) @ 2.29
9 89.9% kurd (xing) + 10.1% bihari (harappa) @ 2.31
10 90.4% kurd (xing) + 9.6% dharkar (metspalu) @ 2.31


Dodecad K12B
# Population Percent
1 Caucasus 37.49
2 Gedrosia 28.76
3 Southwest_Asian 13.42
4 North_European 6.02
5 South_Asian 5.07
6 Atlantic_Med 4.88
7 Sub_Saharan 2.25
8 Southeast_Asian 1.05
9 East_Asian 0.66
10 Siberian 0.34
11 East_African 0.07

Single Population Sharing:

# Population (source) Distance
1 Iranians (Behar) 4.08
2 Iranian (Dodecad) 4.57
3 Kurd (Dodecad) 5.73
4 Kurds (Yunusbayev) 6.72
5 Uzbekistan_Jews (Behar) 12.78
6 Turkmens (Yunusbayev) 14.88
7 Turks (Behar) 17.09
8 Iranian_Jews (Behar) 18.29
9 Azerbaijan_Jews (Behar) 18.3
10 Georgia_Jews (Behar) 18.84


Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 89% Kurds (Yunusbayev) + 11% Cochin_Jews (Behar) @ 3.16
2 90.1% Kurds (Yunusbayev) + 9.9% Brahmins_from_Tamil_Nadu (Metspalu) @ 3.19
3 91% Kurds (Yunusbayev) + 9% INS30 (SGVP) @ 3.2
4 90.7% Kurds (Yunusbayev) + 9.3% Iyengar (Dodecad) @ 3.2
5 90.2% Kurds (Yunusbayev) + 9.8% Indian (Dodecad) @ 3.2
6 90.5% Kurds (Yunusbayev) + 9.5% Iyer (Dodecad) @ 3.2
7 90.9% Kurds (Yunusbayev) + 9.1% GIH30 (Dodecad) @ 3.22
8 91.3% Kurds (Yunusbayev) + 8.7% Kanjars (Metspalu) @ 3.23
9 91.2% Kurds (Yunusbayev) + 8.8% Muslim (Metspalu) @ 3.23
10 91.4% Kurds (Yunusbayev) + 8.6% Velamas (Metspalu) @ 3.25


The components percentages are slightly different for me but I find the oracles are pretty damn similar. Which is nice because I do find these calculators are pretty accurate for certain people.

XooR
08-31-2017, 12:51 PM
HarappaWorld

Admix Results (sorted):

# Population Percent
1 Caucasian 57.14
2 Baloch 17.66
3 SW-Asian 10.24
4 Mediterranean 9.33
5 NE-Euro 4.79


Dodecad K12b

Admix Results (sorted):

# Population Percent
1 Caucasus 56.42
2 Gedrosia 16.51
3 Southwest_Asian 10.94
4 Atlantic_Med 9.44
5 North_European 6.36

Eurogenes K12b


Population
Western European 2.64
Siberian -
East African -
West Central Asian 8.71
South Asian -
West African -
Caucasus 54.09
Finnish -
Mediterranean 19.31
Southwest Asian 11.76
North European 3.49
East Asian -

Saba123
09-02-2017, 12:52 AM
http://i.imgur.com/luQEJ2k.jpg
My Aunt's results. She is decently less Caucasus and more Arab then I am, Her results are very typical of a Southern Iranian.

Alkaevli
09-02-2017, 01:24 PM
HarappaWorld

Admix Results (sorted):

# Population Percent
1 Caucasian 57.14
2 Baloch 17.66
3 SW-Asian 10.24
4 Mediterranean 9.33
5 NE-Euro 4.79


Dodecad K12b

Admix Results (sorted):

# Population Percent
1 Caucasus 56.42
2 Gedrosia 16.51
3 Southwest_Asian 10.94
4 Atlantic_Med 9.44
5 North_European 6.36

Eurogenes K12b


Population
Western European 2.64
Siberian -
East African -
West Central Asian 8.71
South Asian -
West African -
Caucasus 54.09
Finnish -
Mediterranean 19.31
Southwest Asian 11.76
North European 3.49
East Asian -

From what I've seen the main difference between the Laz and Georgians is that the former have more Mediterranean/Anatolia_Neolithic admixture than the latter.

Georgians can score up to 70% Caucasus.

Saba123
09-03-2017, 01:44 AM
Here are some of my Aunt's GEDmatch results. I might I have spoken too soon on her having average Southern results lol. She seems to consistently get East European in pretty much every calculator, I can not tell of this comes from actually Russian heritage or some sort Caucasus group like Adygei or Georgian, I suspect the latter though.
Eurogene K36
Arabian 5.44
Armenian 8.91
Central_African 0.27
East_Central_Euro 0.99
East_Med 14.65
Eastern_Euro 2.83
Near_Eastern 20.03
North_Atlantic 0.74
North_Caucasian 10.95
South_Asian 0.40
South_Central_Asian 27.34
West_African 0.94
West_Caucasian 6.47

HarappaWorld
1 Caucasian 36.46
2 Baloch 27.18
3 SW-Asian 13.7
4 NE-Euro 6.68
5 S-Indian 5.25
6 Mediterranean 4.8
7 E-African 1.52
8 W-African 1.35
9 American 0.75
10 San 0.72
11 Beringian 0.69
12 Siberian 0.48
13 Papuan 0.42

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 91.9% iranian (behar) + 8.1% haryana-jatt (harappa) @ 2.18
2 95.9% iranian (behar) + 4.1% lithuanian (behar) @ 2.38
3 95.2% iranian (behar) + 4.8% mordovian (yunusbayev) @ 2.4
4 95.5% iranian (behar) + 4.5% russian (hgdp) @ 2.4
5 88.6% iranian (behar) + 11.4% pashtun (harappa) @ 2.41
6 96.3% iranian (behar) + 3.7% finnish (1000genomes) @ 2.45
7 95.5% iranian (behar) + 4.5% belorussian (behar) @ 2.46
8 95.5% iranian (behar) + 4.5% russian (behar) @ 2.48
9 95.2% iranian (behar) + 4.8% ukranian (yunusbayev) @ 2.5
10 88.2% iranian (behar) + 11.8% tajik (yunusbayev) @ 2.53
11 92.5% iranian (behar) + 7.5% punjabi-jatt-sikh (harappa) @ 2.55
12 95% iranian (behar) + 5% chuvash (behar) @ 2.56
13 95.6% iranian (behar) + 4.4% orcadian (hgdp) @ 2.58
14 95.4% iranian (behar) + 4.6% n-european (xing) @ 2.59
15 95.4% iranian (behar) + 4.6% utahn-white (hapmap) @ 2.59
16 93.7% iranian (behar) + 6.3% nepalese-a (xing) @ 2.6
17 92.2% iranian (behar) + 7.8% punjabi-khatri (harappa) @ 2.62
18 94.9% iranian (behar) + 5.1% slovenian (xing) @ 2.62
19 95.5% iranian (behar) + 4.5% british (1000genomes) @ 2.64
20 95.5% iranian (behar) + 4.5% utahn-white (1000genomes) @ 2.64

Dodecad K12
1 Caucasus 35.66
2 Gedrosia 26.55
3 Southwest_Asian 14.09
4 North_European 8.17
5 South_Asian 6.03
6 Atlantic_Med 4.64
7 Sub_Saharan 2.21
8 Siberian 1.03
9 East_African 1.02
10 Northwest_African 0.6

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 72.6% Uzbekistan_Jews (Behar) + 27.4% Pathan (HGDP) @ 3.68
2 93.1% Iranians (Behar) + 6.9% Russian_B (Behar) @ 4.07
3 92.7% Iranians (Behar) + 7.3% Ukranians (Yunusbayev) @ 4.08
4 93.7% Iranians (Behar) + 6.3% Lithuanian (Dodecad) @ 4.13
5 93.1% Iranians (Behar) + 6.9% Russian (HGDP) @ 4.13
6 93.3% Iranians (Behar) + 6.7% Belorussian (Behar) @ 4.14
7 92.9% Iranians (Behar) + 7.1% Mordovians (Yunusbayev) @ 4.15
8 93% Iranians (Behar) + 7% Mixed_Slav (Dodecad) @ 4.16
9 94.1% Iranians (Behar) + 5.9% Lithuanians (Behar) @ 4.18
10 77% Uzbekistan_Jews (Behar) + 23% Jatt (Dodecad) @ 4.19
11 94.2% Iranians (Behar) + 5.8% Finnish (Dodecad) @ 4.23
12 94.1% Iranians (Behar) + 5.9% FIN30 (1000Genomes) @ 4.25
13 93.2% Iranians (Behar) + 6.8% Polish (Dodecad) @ 4.28
14 75% Uzbekistan_Jews (Behar) + 25% Burusho (HGDP) @ 4.33
15 92.7% Iranians (Behar) + 7.3% Chuvashs (Behar) @ 4.36
16 90.2% Iranians (Behar) + 9.8% Romanians (Behar) @ 4.53
17 63.8% Georgia_Jews (Behar) + 36.2% Pathan (HGDP) @ 4.53
18 89.9% Iranians (Behar) + 10.1% Bulgarian (Dodecad) @ 4.55
19 87.1% Kurds (Yunusbayev) + 12.9% Brahmins_from_Uttar_Pradesh (Metspalu) @ 4.57
20 90.4% Iranians (Behar) + 9.6% Bulgarians (Yunusbayev) @ 4.6

MDLP K23b

1 Caucasian 33.36
2 South_Central_Asian 27.28
3 Near_East 11.56
4 South_Indian 6.34
5 European_Early_Farmers 4.87
6 European_Hunters_Gatherers 3.77
7 North_African 3.74
8 Ancestral_Altaic 2.92
9 Subsaharian 2.26
10 Australoid 1.32
11 Arctic 0.94
12 Khoisan 0.41
13 Archaic_African 0.32
14 East_African 0.31
15 Tungus-Altaic 0.25
16 Amerindian 0.2
17 East_Siberian 0.09
18 Archaic_Human 0.04

Single Population Sharing:

# Population (source) Distance
1 Iranian ( ) 5.17
2 Kurd_South ( ) 7
3 Kurd_East ( ) 8.74
4 Kurd ( ) 12.25
5 Kurd_North ( ) 12.56
6 Azeri ( ) 13.11
7 Iraki ( ) 14.02
8 Ain_Touta_WGA ( ) 15.73
9 Uzbek_Tashkent ( ) 16.18
10 Baku_WGA ( ) 16.25

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 87.8% Iranian ( ) + 12.2% Adygei ( ) @ 2.33
2 91.4% Iranian ( ) + 8.6% Abkhasian ( ) @ 2.34
3 88.4% Iranian ( ) + 11.6% North_Ossetian ( ) @ 2.37
4 92% Iranian ( ) + 8% Georgian_Svan ( ) @ 2.37
5 86.1% Iranian ( ) + 13.9% Chechen ( ) @ 2.38
6 85% Iranian ( ) + 15% Kumyk ( ) @ 2.38
7 92.4% Iranian ( ) + 7.6% Georgian_Tbilisi ( ) @ 2.39
8 91.6% Iranian ( ) + 8.4% Georgian ( ) @ 2.39
9 92% Iranian ( ) + 8% Georgian_Megrelia ( ) @ 2.39
10 88.8% Iranian ( ) + 11.2% Ossetian ( ) @ 2.4
11 88.3% Iranian ( ) + 11.7% Balkar ( ) @ 2.4
12 91.7% Iranian ( ) + 8.3% Adjara ( ) @ 2.41
13 86.4% Iranian ( ) + 13.6% Kabardin ( ) @ 2.41
14 92.5% Iranian ( ) + 7.5% Georgian_Imereti ( ) @ 2.42
15 85.7% Iranian ( ) + 14.3% Cirkassian ( ) @ 2.46
16 93% Iranian ( ) + 7% Georgian_Laz ( ) @ 2.47
17 88.5% Iranian ( ) + 11.5% Circassian ( ) @ 2.47
18 90.3% Iranian ( ) + 9.7% Kakheti ( ) @ 2.49
19 82% Iranian ( ) + 18% Azeri_Dagestan ( ) @ 2.52
20 91.4% Iranian ( ) + 8.6% Turk_Trabzon ( ) @ 2.6

Eurogene K15
1 East_Med 31.64
2 West_Asian 31.29
3 South_Asian 12.28
4 Eastern_Euro 7.98
5 Red_Sea 5.15
6 Baltic 4.03
7 Sub-Saharan 2.51
8 Atlantic 2.32
9 West_Med 1.13
10 Amerindian 0.8
11 Northeast_African 0.65
12 Oceanian 0.22

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 95% Iranian + 5% MA-1 @ 3.65
2 96.8% Iranian + 3.2% Erzya @ 4.14
3 98.1% Iranian + 1.9% Mbuti_Pygmy @ 4.15
4 98.1% Iranian + 1.9% Bantu_N.E. @ 4.15
5 98.2% Iranian + 1.8% Biaka_Pygmy @ 4.15
6 98.1% Iranian + 1.9% Luhya @ 4.15
7 98% Iranian + 2% San @ 4.16
8 98.3% Iranian + 1.7% Bantu_S.E. @ 4.16
9 97.2% Iranian + 2.8% Lithuanian @ 4.16
10 98.3% Iranian + 1.7% Bantu_S.W. @ 4.16
11 96.8% Iranian + 3.2% Ukrainian_Belgorod @ 4.16
12 96.8% Iranian + 3.2% Southwest_Russian @ 4.16
13 97.1% Iranian + 2.9% Estonian_Polish @ 4.18
14 97% Iranian + 3% Kargopol_Russian @ 4.18
15 98.4% Iranian + 1.6% Mandenka @ 4.18
16 98.5% Iranian + 1.5% Yoruban @ 4.19
17 96.9% Iranian + 3.1% Ukrainian @ 4.19
18 97.1% Iranian + 2.9% Russian_Smolensk @ 4.19
19 97.2% Iranian + 2.8% Belorussian @ 4.2
20 97% Iranian + 3% Ukrainian_Lviv @ 4.22

Apex n Harmony
09-03-2017, 03:48 PM
Here are some of my Aunt's GEDmatch results. I might I have spoken too soon on her having average Southern results lol. She seems to consistently get East European in pretty much every calculator, I can not tell of this comes from actually Russian heritage or some sort Caucasus group like Adygei or Georgian, I suspect the latter though.
Eurogene K36
Arabian 5.44
Armenian 8.91
Central_African 0.27
East_Central_Euro 0.99
East_Med 14.65
Eastern_Euro 2.83
Near_Eastern 20.03
North_Atlantic 0.74
North_Caucasian 10.95
South_Asian 0.40
South_Central_Asian 27.34
West_African 0.94
West_Caucasian 6.47


I can't imagine it'd be related to Adygei or Georgian ancestry; otherwise the North and/or West Caucasian components would be much higher, I would think.

Apex n Harmony
09-03-2017, 04:38 PM
Any test that shows Afghans as 70-80% S Asian, bases the S Asian component on Afghan references in addition to Indian references So in effect the test is saying that the Afghan is 70-80% Afghan, which of course is not very informative to genetic substructure if you happen to be an Indian or Afghan.

For example, I can make a Kurd 70-80% S Asian if I made Kurds also S Asian references. Wouldn't make sense of course but that is certainly possible.

But aren't the results of the commercial tests telling us something more? My understanding is that they (or at least AncestryDNA) don't compare your DNA directly to each person in their reference panel. Instead, they survey the samples they've taken from around the world (which are, of course, already admixed, but I'll leave that aside for now), identify allele frequencies that happen to be more prevalent in one region than another, and then create their reference populations accordingly. If that's the case, aren't the results telling us something more than "congrats, you're an Afghan who's matched our Afghan sample in our South Asia group, and thus, you have a high South Asia score"? Isn't it more like, "you're an Afghan who displays an overwhelming frequency of alleles that we've identified as most common to the South Asia region, and thus, you have a high South Asia score"?

I'll use a personal example that might make more sense. I'm Iranian and I scored 91% "Caucasus" on AncestryDNA. My father scored 96%. From what I've seen, Iranians typically score between 60 and 80 percent "Caucasus" on that test. Would you simply chalk this up to Ancestry's inclusion of an Iranian in their "Caucasus" reference panel that has a similar ancestral background to mine? Would it not be more accurate to conclude that my family displays an overwhelming frequency of alleles that Ancestry has identified as most common to the Caucasus region (which would, of course, be more insightful than the former)? Similarly, it would seem wrong to conclude than an Iranian who scores 65% "Caucasus" has simply matched an Iranian individual included in Ancestry's Caucasus reference panel at a rate of 65%.

As always, I apologize in advance if I'm overlooking something obvious, stating something obvious, and/or displaying my incredible lack of knowledge on this subject.

DMXX
09-03-2017, 08:06 PM
Saba,

Your results aren't unusual for an Iranian and they don't automatically suggest there is some cryptic modern NE European ancestry in your heritage. Unless, of course, you have a reason to suspect NE European heritage (oral tradition perhaps).

The "Iranian" reference population used in these older calculators are from Behar et al. The samples appear to be pan-Iranian, but have a couple individuals with surplus Near-Eastern and African admixture compared to the others within the set.

The most likely explanation for your results is that you happen to have a slightly stronger affinity with modern European populations relative to the average of that dataset. This isn't unusual (I display the same pattern quite consistently - My HarappaWorld, Dodecad K12 and Eurogenes K15 Mixed Mode results for comparison, posted in order).

A reasonable interpretation is that any excess NE European in Iranians and Kurds relative to the appropriate reference populations reflects an excess in LNBA steppe ancestry.



# Primary Population (source) Secondary Population (source) Distance
1 92.3% iranian (harappa) + 7.7% ashkenazi (harappa) @ 1.44
2 92% iranian (harappa) + 8% ashkenazy-jew (behar) @ 1.5
3 94% iranian (harappa) + 6% tuscan (1000genomes) @ 1.5
4 95% iranian (harappa) + 5% italian (hgdp) @ 1.5
5 96% iranian (harappa) + 4% spaniard (1000genomes) @ 1.54
6 94.6% iranian (harappa) + 5.4% romanian-a (behar) @ 1.6
7 94.5% iranian (harappa) + 5.5% bulgarian (yunusbayev) @ 1.61
8 96.1% iranian (harappa) + 3.9% french (hgdp) @ 1.62
9 96.8% iranian (harappa) + 3.2% spain-basc (henn2012) @ 1.69
10 96.9% iranian (harappa) + 3.1% basque (hgdp) @ 1.7
11 93% iranian (harappa) + 7% morocco-jew (behar) @ 1.7
12 92.5% iranian (harappa) + 7.5% sephardic-jew (behar) @ 1.73
13 96.5% iranian (harappa) + 3.5% utahn-white (1000genomes) @ 1.74
14 96.5% iranian (harappa) + 3.5% n-european (xing) @ 1.74
15 96.1% iranian (harappa) + 3.9% hungarian (behar) @ 1.74
16 96.6% iranian (harappa) + 3.4% british (1000genomes) @ 1.75
17 96.2% iranian (harappa) + 3.8% slovenian (xing) @ 1.76
18 96.8% iranian (harappa) + 3.2% orcadian (hgdp) @ 1.79
19 89.3% iranian (harappa) + 10.7% turk-aydin (hodoglugil) @ 1.79
20 96.7% iranian (harappa) + 3.3% sardinian (hgdp) @ 1.83

# Primary Population (source) Secondary Population (source) Distance
1 69.7% Uzbekistan_Jews (Behar) + 30.3% Tajiks (Yunusbayev) @ 2.58
2 59.7% Azerbaijan_Jews (Behar) + 40.3% Tajiks (Yunusbayev) @ 2.64
3 58.7% Assyrian (Dodecad) + 41.3% Tajiks (Yunusbayev) @ 2.67
4 59.6% Georgia_Jews (Behar) + 40.4% Tajiks (Yunusbayev) @ 2.86
5 94.6% Iranian (Dodecad) + 5.4% Ukranians (Yunusbayev) @ 3.11
6 95% Iranian (Dodecad) + 5% Russian_B (Behar) @ 3.13
7 94.9% Iranian (Dodecad) + 5.1% Mixed_Slav (Dodecad) @ 3.14
8 95% Iranian (Dodecad) + 5% Russian (Dodecad) @ 3.15
9 95.1% Iranian (Dodecad) + 4.9% Belorussian (Behar) @ 3.15
10 94.9% Iranian (Dodecad) + 5.1% Polish (Dodecad) @ 3.16
11 95% Iranian (Dodecad) + 5% Russian (HGDP) @ 3.16
12 95.5% Iranian (Dodecad) + 4.5% Lithuanian (Dodecad) @ 3.17
13 95.8% Iranian (Dodecad) + 4.2% Finnish (Dodecad) @ 3.2
14 95.7% Iranian (Dodecad) + 4.3% Lithuanians (Behar) @ 3.2
15 95.7% Iranian (Dodecad) + 4.3% FIN30 (1000Genomes) @ 3.21
16 94.9% Iranian (Dodecad) + 5.1% Mordovians (Yunusbayev) @ 3.21
17 59.3% Iranian_Jews (Behar) + 40.7% Tajiks (Yunusbayev) @ 3.22
18 92.2% Iranian (Dodecad) + 7.8% Romanians (Behar) @ 3.22
19 91.9% Iranian (Dodecad) + 8.1% Bulgarian (Dodecad) @ 3.23
20 87.5% Kurds (Yunusbayev) + 12.5% Uzbeks (Behar) @ 3.23

# Primary Population (source) Secondary Population (source) Distance
1 92.4% Iranian + 7.6% Swedish @ 3.06
2 92.4% Iranian + 7.6% North_Swedish @ 3.09
3 92.5% Iranian + 7.5% Norwegian @ 3.1
4 92.3% Iranian + 7.7% La_Brana-1 @ 3.11
5 92.7% Iranian + 7.3% West_Norwegian @ 3.11
6 92.4% Iranian + 7.6% Danish @ 3.14
7 92.5% Iranian + 7.5% North_Dutch @ 3.16
8 91.6% Iranian + 8.4% West_German @ 3.17
9 92.3% Iranian + 7.7% Finnish @ 3.17
10 92.1% Iranian + 7.9% North_German @ 3.18
11 92.3% Iranian + 7.7% Southwest_Finnish @ 3.18
12 92.6% Iranian + 7.4% Irish @ 3.19
13 92.7% Iranian + 7.3% West_Scottish @ 3.2
14 92.7% Iranian + 7.3% Orcadian @ 3.2
15 92.5% Iranian + 7.5% Southeast_English @ 3.2
16 91% Iranian + 9% Hungarian @ 3.25
17 92.6% Iranian + 7.4% Southwest_English @ 3.27
18 91.9% Iranian + 8.1% South_Dutch @ 3.29
19 91.9% Iranian + 8.1% East_German @ 3.33
20 91.2% Iranian + 8.8% Austrian @ 3.35

Saba123
09-03-2017, 09:15 PM
Saba,

Your results aren't unusual for an Iranian and they don't automatically suggest there is some cryptic modern NE European ancestry in your heritage. Unless, of course, you have a reason to suspect NE European heritage (oral tradition perhaps).

The "Iranian" reference population used in these older calculators are from Behar et al. The samples appear to be pan-Iranian, but have a couple individuals with surplus Near-Eastern and African admixture compared to the others within the set.

The most likely explanation for your results is that you happen to have a slightly stronger affinity with modern European populations relative to the average of that dataset. This isn't unusual (I display the same pattern quite consistently - My HarappaWorld, Dodecad K12 and Eurogenes K15 Mixed Mode results for comparison, posted in order).

A reasonable interpretation is that any excess NE European in Iranians and Kurds relative to the appropriate reference populations reflects an excess in LNBA steppe ancestry.



# Primary Population (source) Secondary Population (source) Distance
1 92.3% iranian (harappa) + 7.7% ashkenazi (harappa) @ 1.44
2 92% iranian (harappa) + 8% ashkenazy-jew (behar) @ 1.5
3 94% iranian (harappa) + 6% tuscan (1000genomes) @ 1.5
4 95% iranian (harappa) + 5% italian (hgdp) @ 1.5
5 96% iranian (harappa) + 4% spaniard (1000genomes) @ 1.54
6 94.6% iranian (harappa) + 5.4% romanian-a (behar) @ 1.6
7 94.5% iranian (harappa) + 5.5% bulgarian (yunusbayev) @ 1.61
8 96.1% iranian (harappa) + 3.9% french (hgdp) @ 1.62
9 96.8% iranian (harappa) + 3.2% spain-basc (henn2012) @ 1.69
10 96.9% iranian (harappa) + 3.1% basque (hgdp) @ 1.7
11 93% iranian (harappa) + 7% morocco-jew (behar) @ 1.7
12 92.5% iranian (harappa) + 7.5% sephardic-jew (behar) @ 1.73
13 96.5% iranian (harappa) + 3.5% utahn-white (1000genomes) @ 1.74
14 96.5% iranian (harappa) + 3.5% n-european (xing) @ 1.74
15 96.1% iranian (harappa) + 3.9% hungarian (behar) @ 1.74
16 96.6% iranian (harappa) + 3.4% british (1000genomes) @ 1.75
17 96.2% iranian (harappa) + 3.8% slovenian (xing) @ 1.76
18 96.8% iranian (harappa) + 3.2% orcadian (hgdp) @ 1.79
19 89.3% iranian (harappa) + 10.7% turk-aydin (hodoglugil) @ 1.79
20 96.7% iranian (harappa) + 3.3% sardinian (hgdp) @ 1.83

# Primary Population (source) Secondary Population (source) Distance
1 69.7% Uzbekistan_Jews (Behar) + 30.3% Tajiks (Yunusbayev) @ 2.58
2 59.7% Azerbaijan_Jews (Behar) + 40.3% Tajiks (Yunusbayev) @ 2.64
3 58.7% Assyrian (Dodecad) + 41.3% Tajiks (Yunusbayev) @ 2.67
4 59.6% Georgia_Jews (Behar) + 40.4% Tajiks (Yunusbayev) @ 2.86
5 94.6% Iranian (Dodecad) + 5.4% Ukranians (Yunusbayev) @ 3.11
6 95% Iranian (Dodecad) + 5% Russian_B (Behar) @ 3.13
7 94.9% Iranian (Dodecad) + 5.1% Mixed_Slav (Dodecad) @ 3.14
8 95% Iranian (Dodecad) + 5% Russian (Dodecad) @ 3.15
9 95.1% Iranian (Dodecad) + 4.9% Belorussian (Behar) @ 3.15
10 94.9% Iranian (Dodecad) + 5.1% Polish (Dodecad) @ 3.16
11 95% Iranian (Dodecad) + 5% Russian (HGDP) @ 3.16
12 95.5% Iranian (Dodecad) + 4.5% Lithuanian (Dodecad) @ 3.17
13 95.8% Iranian (Dodecad) + 4.2% Finnish (Dodecad) @ 3.2
14 95.7% Iranian (Dodecad) + 4.3% Lithuanians (Behar) @ 3.2
15 95.7% Iranian (Dodecad) + 4.3% FIN30 (1000Genomes) @ 3.21
16 94.9% Iranian (Dodecad) + 5.1% Mordovians (Yunusbayev) @ 3.21
17 59.3% Iranian_Jews (Behar) + 40.7% Tajiks (Yunusbayev) @ 3.22
18 92.2% Iranian (Dodecad) + 7.8% Romanians (Behar) @ 3.22
19 91.9% Iranian (Dodecad) + 8.1% Bulgarian (Dodecad) @ 3.23
20 87.5% Kurds (Yunusbayev) + 12.5% Uzbeks (Behar) @ 3.23

# Primary Population (source) Secondary Population (source) Distance
1 92.4% Iranian + 7.6% Swedish @ 3.06
2 92.4% Iranian + 7.6% North_Swedish @ 3.09
3 92.5% Iranian + 7.5% Norwegian @ 3.1
4 92.3% Iranian + 7.7% La_Brana-1 @ 3.11
5 92.7% Iranian + 7.3% West_Norwegian @ 3.11
6 92.4% Iranian + 7.6% Danish @ 3.14
7 92.5% Iranian + 7.5% North_Dutch @ 3.16
8 91.6% Iranian + 8.4% West_German @ 3.17
9 92.3% Iranian + 7.7% Finnish @ 3.17
10 92.1% Iranian + 7.9% North_German @ 3.18
11 92.3% Iranian + 7.7% Southwest_Finnish @ 3.18
12 92.6% Iranian + 7.4% Irish @ 3.19
13 92.7% Iranian + 7.3% West_Scottish @ 3.2
14 92.7% Iranian + 7.3% Orcadian @ 3.2
15 92.5% Iranian + 7.5% Southeast_English @ 3.2
16 91% Iranian + 9% Hungarian @ 3.25
17 92.6% Iranian + 7.4% Southwest_English @ 3.27
18 91.9% Iranian + 8.1% South_Dutch @ 3.29
19 91.9% Iranian + 8.1% East_German @ 3.33
20 91.2% Iranian + 8.8% Austrian @ 3.35


Well considering my family is Southern Iranian(Bushrer, Fars, and Khuzestan) and also has excess African and Near Eastern so having excess NE European is pretty surprising. Considering you are half Azeri having excess NE European does not seem unusual. My family had long been suspicious that we had some sort of Eastern European or Caucasus heritage because of the appearance of my Great-Grandmother and because she came from a town near Aspas, Fars. "In Thomas Herbert's Travels in Persia: 1627-1629', he writes that in Aspas some 40,000 transplanted Christian Circassians and Georgians were living, as he was travelling through various regions of the Safavid Empire". My Great-Grandmother was very oddly pale, naturally blonde hair, and blue eyes and was named Katherine but prefered Katayoon. She also always refused to tell of her family history, for some unknown reason, so we could never draw any definite conclusions. My aunt's is fairly closely related (15 cm -8 cm) to many pure Lezigans and other North Caucasians. Also she is relate to quite a few Finns and Russian including a Russian sample from Physical Anthropology.

DMXX
09-03-2017, 09:26 PM
I wouldn't use physical features to make those sorts of calls. Practically every Iranian, Kurd and Afghan seems to have at least one relative with lighter pigmentation. An entire branch of my paternal side's family is auburn-to-mousey brown haired and blue eyed, but Irishmen, they are not!

The segment sharing with various Europeans is also nothing new to report here. Again, most Iranians, Kurds and Afghans have a surprisingly high number of segments shared with Europeans on 23andMe. The origin of those shared segments in most cases is, per Occam's razor, something common, ancient and quite likely bidirectional (as apposed to all of us having individual, recent European introgression).

Georgians actually don't appear significantly more steppe-derived than Iranians, but Circassians are. Your only conclusive means of determining whether the extra NE European-related scores via ADMIXTURE is a surplus of the generic LNBA steppe ancestry Iranians possess approximately 15-25% of, or genuine North Caucasian ancestry, is through segment analysis. Like yourself, I have an oral tradition of descent from the Caucasus, which was confirmed partly through IBD. We have Adyghe/Circassian samples, so that'll be straightforward to confirm or rule out.

Saba123
09-03-2017, 10:25 PM
I wouldn't use physical features to make those sorts of calls. Practically every Iranian, Kurd and Afghan seems to have at least one relative with lighter pigmentation. An entire branch of my paternal side's family is auburn-to-mousey brown haired and blue eyed, but Irishmen, they are not!

The segment sharing with various Europeans is also nothing new to report here. Again, most Iranians, Kurds and Afghans have a surprisingly high number of segments shared with Europeans on 23andMe. The origin of those shared segments in most cases is, per Occam's razor, something common, ancient and quite likely bidirectional (as apposed to all of us having individual, recent European introgression).

Georgians actually don't appear significantly more steppe-derived than Iranians, but Circassians are. Your only conclusive means of determining whether the extra NE European-related scores via ADMIXTURE is a surplus of the generic LNBA steppe ancestry Iranians possess approximately 15-25% of, or genuine North Caucasian ancestry, is through segment analysis. Like yourself, I have an oral tradition of descent from the Caucasus, which was confirmed partly through IBD. We have Adyghe/Circassian samples, so that'll be straightforward to confirm or rule out.

Yeah honestly you're probably right, I think I'm just clinging on to the fact that a hypothesis I had was kinda proven and seems to answer the question of why I look non-Iranian to most Iranians. But as you stated phenotype is a very flimsy way of guessing someone's heritage. I would like to confirm if I have any sort Caucasus descent though. I seem to get Lezgin mixture on some calculators and my Aunt matches with some Lezgins, so I think a good start would be to compare to Lezgin.

DMXX
09-03-2017, 10:41 PM
Iranians come in all shapes and sizes; there's some Iranian-specific phenotypes that pop up across the entire plateau, whereas others are either rare or specific to certain areas. I can empathise with your experience (most Iranians here, for whatever reason, don't think I'm "one of them" from an ethnic perspective at first glance).

Yes, it's worth investigating. I do not mean to appear dismissive of the possibility. Rather, there are some regular patterns which could be observed from the results of Iranians. Said patterns have to be borne in mind moving forwards. That is the only intent behind my insertion in this discussion. Keep us abreast regarding any new findings you may discover, baradaram!

Saba123
09-04-2017, 01:36 AM
Iranians come in all shapes and sizes; there's some Iranian-specific phenotypes that pop up across the entire plateau, whereas others are either rare or specific to certain areas. I can empathise with your experience (most Iranians here, for whatever reason, don't think I'm "one of them" from an ethnic perspective at first glance).

Yes, it's worth investigating. I do not mean to appear dismissive of the possibility. Rather, there are some regular patterns which could be observed from the results of Iranians. Said patterns have to be borne in mind moving forwards. That is the only intent behind my insertion in this discussion. Keep us abreast regarding any new findings you may discover, baradaram!

Yeah its kinda of disheartening when people tell me theres no way I am Iranian because of my coloring and facial features, even though I speak Farsi fluently and know more about Iranian history than them. It would be nice to have nice specific answer as to why I do not look Iranian but alas not every thing has a specific answer. I will look into this matter though and see if I can reach some sort of clarity in where the excess NE Euro comes from, keeping mind the regular patterns seen in Iranians. I will definitely let you know if I find anything new

XooR
09-04-2017, 05:16 AM
From what I've seen the main difference between the Laz and Georgians is that the former have more Mediterranean/Anatolia_Neolithic admixture than the latter.

Georgians can score up to 70% Caucasus.

If you look at it in individual level yes there are some Georgians score +- 70%, but Harappa Georgian average is around %60 (you can see attached Harappa Georgian average calculated from 29 individuals. Also I calculated Harappa Laz averages (from 17 individuals). From what I see Laz people have significant amount of SW Asia and Mediterranean compare to Georgians and Georgians have higher Baloch and NE Euro.

In addition there is a Laz individual scores 81.8% Caucasian in MDLP K23b calculator but Laz average is around 60%

18543

18544

vatan
09-04-2017, 05:39 AM
Yeah its kinda of disheartening when people tell me theres no way I am Iranian because of my coloring and facial features, even though I speak Farsi fluently and know more about Iranian history than them. It would be nice to have nice specific answer as to why I do not look Iranian but alas not every thing has a specific answer. I will look into this matter though and see if I can reach some sort of clarity in where the excess NE Euro comes from, keeping mind the regular patterns seen in Iranians. I will definitely let you know if I find anything new

Could be from ancient iranics who migrated to Iran. Wouldn't be surprised.

surbakhunWeesste
09-04-2017, 09:02 PM
Yeah its kinda of disheartening when people tell me theres no way I am Iranian because of my coloring and facial features, even though I speak Farsi fluently and know more about Iranian history than them. It would be nice to have nice specific answer as to why I do not look Iranian but alas not every thing has a specific answer. I will look into this matter though and see if I can reach some sort of clarity in where the excess NE Euro comes from, keeping mind the regular patterns seen in Iranians. I will definitely let you know if I find anything new

so they mistook you for someone from the sub-continent or England(: is my city)?
umm like whatever happened to Afro-Iranians or a random Iranian who looked like Bjork they are 'abundant' here in California esp, Socal. You can look like a swede or a Bengali or a Dinka warrior but if you speak Farsi and tell them you are one of them, they won't bat an eye: thank you mixed heritage. Mad world.

Apex n Harmony
09-04-2017, 11:41 PM
Yeah its kinda of disheartening when people tell me theres no way I am Iranian because of my coloring and facial features, even though I speak Farsi fluently and know more about Iranian history than them. It would be nice to have nice specific answer as to why I do not look Iranian but alas not every thing has a specific answer. I will look into this matter though and see if I can reach some sort of clarity in where the excess NE Euro comes from, keeping mind the regular patterns seen in Iranians. I will definitely let you know if I find anything new

Anyone who thinks you don't look "Iranian" doesn't understand (among other things) the incredible ethnic diversity of Iran.

Saba123
09-05-2017, 12:34 AM
so they mistook you for someone from the sub-continent or England(: is my city)?
umm like whatever happened to Afro-Iranians or a random Iranian who looked like Bjork they are 'abundant' here in California esp, Socal. You can look like a swede or a Bengali or a Dinka warrior but if you speak Farsi and tell them you are one of them, they won't bat an eye: thank you mixed heritage. Mad world.

usually they think I am Ashkenazi Jewish, Latino, White American or some incredibly random mix of ethnicities. Well I live Socal as well in a pretty Iranian populated and I constantly have people insisting I am not Iranian or atleast I must be half.

Saba123
09-05-2017, 12:40 AM
Anyone who thinks you don't look "Iranian" doesn't understand (among other things) the incredible ethnic diversity of Iran.

Yeah honestly and alot of them are actually recent immigrants from Iran and come from big cities like Tehran, so I would expect them to have some understanding of the ethnic diversity in Iran but apparently not lol. I think most of them are just thrown off by my coloring and by how American I act, but alot of them tell me my facial features are not typical of an Iranian. I do not take insult to it anymore though because it happens every single day pretty much hahaha.

Kurd
09-05-2017, 04:57 AM
But aren't the results of the commercial tests telling us something more? My understanding is that they (or at least AncestryDNA) don't compare your DNA directly to each person in their reference panel. Instead, they survey the samples they've taken from around the world (which are, of course, already admixed, but I'll leave that aside for now), identify allele frequencies that happen to be more prevalent in one region than another, and then create their reference populations accordingly. If that's the case, aren't the results telling us something more than "congrats, you're an Afghan who's matched our Afghan sample in our South Asia group, and thus, you have a high South Asia score"? Isn't it more like, "you're an Afghan who displays an overwhelming frequency of alleles that we've identified as most common to the South Asia region, and thus, you have a high South Asia score"?

I'll use a personal example that might make more sense. I'm Iranian and I scored 91% "Caucasus" on AncestryDNA. My father scored 96%. From what I've seen, Iranians typically score between 60 and 80 percent "Caucasus" on that test. Would you simply chalk this up to Ancestry's inclusion of an Iranian in their "Caucasus" reference panel that has a similar ancestral background to mine? Would it not be more accurate to conclude that my family displays an overwhelming frequency of alleles that Ancestry has identified as most common to the Caucasus region (which would, of course, be more insightful than the former)? Similarly, it would seem wrong to conclude than an Iranian who scores 65% "Caucasus" has simply matched an Iranian individual included in Ancestry's Caucasus reference panel at a rate of 65%.

As always, I apologize in advance if I'm overlooking something obvious, stating something obvious, and/or displaying my incredible lack of knowledge on this subject.

It's nothing but an illusion. It sucks to be a calculator reference because individuals from populations who have members included are calculator references don't really learn much. So if Laghmanis, Jalalabadis or Peshawris are included as S Asian references, then E Pakhtuns will turn up close to 100% S Asians and Kandhari Pashtuns perhaps 60% S Asian....however, if Pashtuns are NOT included in the S Asian reference, AND the test has a W Asian component with Kurd/Iranian references, then a Laghmani could end up 25% S Asian, and a Kandhari 15% S Asian. So then what gives?

In my latest K12 test, the baselines for references are much more informative and meaningful for describing genetic substructure than some AncestryDNA or 23andMe test, because I use ancestral populations which are not as admixed as the modern references they use. If you check the results posted under the thread or the spreadsheet that a couple of members were kind enough to generate, you'll notice the only difference between Kurds/Iranians and Afghans is perhaps 7-10% higher SE Eurasian/ASI related and about 10% E Steppe related.

Apex n Harmony
09-09-2017, 04:58 PM
It's nothing but an illusion. It sucks to be a calculator reference because individuals from populations who have members included are calculator references don't really learn much. So if Laghmanis, Jalalabadis or Peshawris are included as S Asian references, then E Pakhtuns will turn up close to 100% S Asians and Kandhari Pashtuns perhaps 60% S Asian....however, if Pashtuns are NOT included in the S Asian reference, AND the test has a W Asian component with Kurd/Iranian references, then a Laghmani could end up 25% S Asian, and a Kandhari 15% S Asian. So then what gives?


So, given that my family scores well into the 90s in "Caucasus" on AncestryDNA (and my father a whopping 96%), I presume there are Iranian references included in the reference panel. However, if we're simply matching Iranian samples, why would we still show traces of South Asia and Middle East, when the Iranian samples themselves surely already contain South Asia and Middle East substructure? Am I missing something obvious here?

ArmoSlavLady
09-09-2017, 05:05 PM
Hello, My ancestry is 75% Armenian and 25% Russian. AncestryDNA was able to actually show that down to the percentile, if you consider the Italian/Greek DNA Armenian.

AncestryDNA:
Europe 38%
Europe East 17%
Italy/Greece 13%
Low Confidence Region
Europe West 5%
Finland/Northwest Russia 3%
West Asia 62%
Caucasus 59%
Low Confidence Region
Middle East 3%

DNA.Land showed some very strange results, which do not correspond at all with my known heritage.

DNA.Land
West Eurasian 100%
Central Indoeuropean 42%
Includes: Abkhasian in Abkhazia/Georgia; Armenian in Armenia; Georgian/Megrels in Georgia; Iranian in Iran; Druze in (Carmel) Israel; Balkar, Chechen, Kumyk, Lezgin, North Ossetian and Adygei in (Caucasus and 5 other sites) Russia and Turkish in (Adana, Aydin, Balikesir, Istanbul, Kayseri, Trabzon and 1 other site) Turkey
South/Central European 33%
Includes: Italian/Bergamo, Italian/Tuscan and Toscani in (Bergamo, Tuscany and 1 other site) Italy
Balkan 1%
Includes: Albanian in Albania; Bulgarian in Bulgaria and Greek in (2 sites) Greece
Northwest European 8.7%
Includes: Scottish Argyll_Bute_GBR and British in England; Icelandic in Iceland; Norwegian in Norway and Orcadian in Orkney Islands
Indo-Iranian 7.6%
Includes: Balochi, Brahui and Makrani in Pakistan
Kalash 1%
Includes: Kalash in Pakistan
Finnish 5.3%
Includes: Finnish in Finland
North Slavic 1.8%
Includes: Belarusian in Belarus; Estonian in Estonia; Lithuanian in Lithuania; Mordovian and Russian in Russia and Ukrainian in (East) Ukraine

Super odd, but interesting. Not sure how the two could be so different.

MonkeyDLuffy
09-09-2017, 07:29 PM
So, given that my family scores well into the 90s in "Caucasus" on AncestryDNA (and my father a whopping 96%), I presume there are Iranian references included in the reference panel. However, if we're simply matching Iranian samples, why would we still show traces of South Asia and Middle East, when the Iranian samples themselves surely already contain South Asia and Middle East substructure? Am I missing something obvious here?

I think a good explanation for that would be at what age/time period the SA mix happened in Iranian samples. SA samples are already a mix of various East and west eurasian components. Their genetic structure changed with invasions. Lets say for example, the SA population that mixed with Iranians lacked the east eurasian or siberian kind of mix, which is now found in modern south asians. So if an Iranian takes a test where samples are based on modern population, the similar mixture will be absorbed while the later mixture will stand out as SA separately.

Kurd
09-09-2017, 11:26 PM
So, given that my family scores well into the 90s in "Caucasus" on AncestryDNA (and my father a whopping 96%), I presume there are Iranian references included in the reference panel. However, if we're simply matching Iranian samples, why would we still show traces of South Asia and Middle East, when the Iranian samples themselves surely already contain South Asia and Middle East substructure? Am I missing something obvious here?

That is a very high Caucasus %. For comparison I'm aware of only 1 Kurd from N Iraq who has tested with Ancestry, and I believe they had 76% Caucasus, 20% Middle Eastern, 3% S Asian, 1% European. Based on this my guess is that their Caucasus references include Armenians, and possibly a couple from the very NW of Iran. The 4% and 3% S Asian you and the N Iraq Kurd get indicates to me that you guys are a little more S Asian shifted than their Caucasian references.

If they had used references exclusively from N and C Caucasus then perhaps your SA would have been 10-15%. So in effect your SA % will fluctuate all over the place depending on how close their W Asian references is to you.

MonkeyDLuffy
09-10-2017, 12:06 AM
That is a very high Caucasus %. For comparison I'm aware of only 1 Kurd from N Iraq who has tested with Ancestry, and I believe they had 76% Caucasus, 20% Middle Eastern, 3% S Asian, 1% European. Based on this my guess is that their Caucasus references include Armenians, and possibly a couple from the very NW of Iran. The 4% and 3% S Asian you and the N Iraq Kurd get indicates to me that you guys are a little more S Asian shifted than their Caucasian references.

If they had used references exclusively from N and C Caucasus then perhaps your SA would have been 10-15%. So in effect your SA % will fluctuate all over the place depending on how close their W Asian references is to you.

Agree, Using modern population as an average is not the best option, since every individual has different admixture even within same ethnicity. Ancient works the best. E.g. My own community is my 8th match usually on oracles.

Hedda Gabler
09-13-2017, 12:00 PM
Please note that, "Southern Central Asian" haplogroup/marker L is often categorized as "South Asian". I think those %4 South Asia is mostly Y-DNA L. It is common among North Eastern Caucasians.

Kurd
09-13-2017, 07:39 PM
Please note that, "Southern Central Asian" haplogroup/marker L is often categorized as "South Asian". I think those %4 South Asia is mostly Y-DNA L. It is common among North Eastern Caucasians.

Incorrect, If you are suggesting that there is a correlation between haplogroups and admixture levels of S Asian within modern ethnic groups.

Haplogroups are irrelevant for describing admixture levels intra-ethnic group . They are relevant for describing historical migration routes and population movements .

Admixture levels generally follow geographical Clines.

PROOF:

Take a decent size sample with 8 great grand-parents belonging to a single ethnic group, whether Sorani Kurds, Feyli Kurds, Azeris, Saudis, etc, and you will find no correlation between admixture levels of S Asian and individual haplogroups

Anabasis
09-14-2017, 09:11 PM
Please note that, "Southern Central Asian" haplogroup/marker L is often categorized as "South Asian". I think those %4 South Asia is mostly Y-DNA L. It is common among North Eastern Caucasians.

The L among the north eastern turkey is under L1b-M317>Ph8 while all south asian L's are L1a1 and L1a2. L1b and L1a had common ancestry at least 18600 years ago according to yfull which is very huge timescale to make such a conclusion. What is interesting is L1a1 and La12 also found among some Levant and Mesopotamian population but the most eastern L1b samples are from eastern Iran which is a different branch of L1b>m317>Sk1414. Thus i am pretty sure that there arent any recent contact between south asia and NE Turkey occured in last 18k years old. While interpreting the frequencies of haplogroup you should never neglect the timescales between haplogroup ages. You need to consider the deepest subsclade of haplogroup to make conclusions related with difusion of haplogroups.

femekh
11-28-2017, 09:20 PM
AncestryDNA:
70% Caucasus, 19% Middle East, 11% Greece/Italy

DNA Land:
67% Central Indo-European, 21% Mediterranean Islander, 5.2% Arab/Egyptian, 3.6% South/Central European, 2.3% Indo-Iranian, 1.2% Ambiguous

I was born in Brazil, raised in the USA, and I'm Armenian on both sides. Never knew of any other ethnicities in my family history. But DNA Land's results seem more accurate than Ancestry's.

Blue Devils
12-07-2017, 07:54 PM
FTDNA

Middle Eastern
>Asia Minor - 53%

European
>British Isles - 18%
>East Europe -11%

Central/South Asia
>Central Asia - 16%

New World
>North and Central America - 3%

MyHeritage

West Asian - 58,2%
Irish, Scottish, and Welsh - 14.3%
Central Asian - 12.4%
Scandinavian - 9.4%
South Asian - 3.2%
Baltic - 1.3%
English - 1.2%

DNALand

Central Indoeuropean - 60%
Northwest European - 20%
Indus Valley - 13%
Finnish - 2.4%
Kalash - 2%
Ambiguous - 1.5%

maydonez
03-06-2018, 02:21 PM
I was tested with 23andme V5 chip. So, I can't upload my rawdata to either myheritage, nor ftDNA. And 23andme is really useless in my opinion for Caucasians, since they don't have a single reference population for North Caucasians.

Here are my results:

DNA.Land:
Central Indoeuropean: 68%
Mediterranean Islander: 16%
Southwestern European: 4.5%
Gujarati: 4.1%
Central Asian: 2.8%
Ashkenazi: 2.7%
South/Central European: 1.6%

Gedmatch K36:
North Caucasian: 36.57
East Mediterranean: 11.72
West Caucasian: 10.45
Armenian: 7.74
South Central Asian: 7.55
Near Eastern: 6.44
Italian: 6.09
Arabian: 5.08
North Atlantic: 3.75
Basque: 1.94
West Mediterranean: 1.06

Gedmatch K12b:
Caucasus 43.33
West Central Asian: 11.96
Mediterranean: 11.78
Western European: 10.92
Southwest Asian: 10.64
North European: 4.93
South Asian: 3.91
Siberian: 2.02
East African: 0.52
East Asian: -
Finnish: -
West African: -

MLDP K16 Modern:
This calculator was not perfect with population Oracles, but it was still close.
Caucasian: 50.87
Steppe: 16.58
Neolithic: 11.43
Indian: 7.60
Near East: 6.22
North East European: 4.13
Arctic: 1.14

puntDNAL K10:
CHG: 39.64
ENF: 31.52
WHG: 19.10
Oceanian: 3.08
ASI: 2.54
Amerindian: 2.21

puntDNAL K15:
Caucasian: 46.03
Mediterranean: 25.88
NE_European: 13.34
SW_Asian: 6.45
S_Indian: 3.35
Siberian: 2.76


I don't have any known ancestor outside the Caucasus.
I show high South Asian components in such calculators(also on Gedmatch, etc) but when Mephisto did a PCA for me, I get a spot in Adyghe and Chechen clusters, matching most closely to Adyghes, North Ossetians and Chechens.
I have no idea why I get high South Indian results.
Harappa was so irrelevant for me, I am not even bothered to write it down here.
I can provide any other calculator results if necessary.
(I wanted to shorten my answer with a spoiler but the button is somehow lost) :/

Tag Heuer
03-11-2018, 08:17 AM
3.35 of S. Indian is not high and it is typical for Caucasians to have it at range from 0 to approx 5.

My results for example:
1 Caucasian 48.06
2 NE_European 20.79
3 Mediterranean 19.14
4 S_Indian 4.36
5 SW_Asian 2.27
6 Siberian 2.04
7 Beringian 1.07

digital_noise
04-08-2018, 06:50 AM
My wife's results. She is from Tehran.

AncestryDNA:
Caucasus 74%
Asia South 12%
Middle East 9%
European Jewish 5%


puntDNAL K15 Oracle

Admix Results (sorted):

# Population Percent
1 Caucasian 46.21
2 Mediterranean 19.04
3 SW_Asian 9.82
4 S_Indian 9.56
5 NE_European 8.04
6 Horn_Of_Africa 2.56
7 Siberian 1.23
8 S_African 1.05


Finished reading population data. 157 populations found.
15 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Iranian @ 2.712312
2 Kurdish @ 8.344073
3 Azerbaijani @ 9.973098
4 Turk_Istanbul @ 11.687766
5 Kumyk @ 14.345983
6 Turk_Kayseri @ 15.357976
7 Armenian @ 15.612658
8 Balkar @ 16.383217
9 North_Ossetian @ 16.399874
10 Assyrian @ 17.064280
11 Chechen @ 17.328701
12 Lezgin @ 19.415741
13 Pashtun @ 19.984222
14 Turkmen @ 20.007998
15 Tadjik @ 20.025848
16 Abkhasian @ 20.074356
17 Lebanese @ 20.254656
18 Turk_Trabzon @ 20.279926
19 Iraqi_Jew @ 21.075518
20 Syrian @ 21.849354

Using 2 populations approximation:
1 50% Iranian +50% Iranian @ 2.712312


Using 3 populations approximation:
1 50% Iranian +25% Brahui +25% Cypriot @ 2.226547


Using 4 populations approximation:
++++++++++++++++++++++++++++++++++++++++++++++++++ +++++++++++++++++++++++++++++
1 Turk_Istanbul + Iranian + Brahui + Lebanese @ 1.399466
2 Turk_Istanbul + Iranian + Balochi + Lebanese @ 1.557250
3 North_Ossetian + Sindhi + Abkhasian + Bedouin_A @ 1.780527
4 Turk_Istanbul + Iranian + Brahui + Palestinian @ 1.851263
5 Turk_Istanbul + Iranian + Balochi + Iraqi_Jew @ 1.901537
6 Kumyk + Iranian + Brahui + Samaritian @ 1.903539
7 Iranian + Balochi + Azerbaijani + Lebanese @ 1.929104
8 Balkar + Sindhi + Georgian + Bedouin_A @ 1.932088
9 North_Ossetian + Sindhi + Georgian + Bedouin_A @ 1.946349
10 Turk_Istanbul + Iranian + Brahui + Iraqi_Jew @ 1.947340
11 North_Ossetian + Iranian + Balochi + Jordanian @ 1.964334
12 North_Ossetian + Kurdish + Brahui + Iraqi_Jew @ 1.975317
13 Iranian + Brahui + Azerbaijani + Lebanese @ 1.976286
14 Kumyk + Sindhi + Georgian + Bedouin_A @ 1.976651
15 North_Ossetian + Kurdish + Brahui + Jordanian @ 1.977220
16 Turk_Istanbul + Iranian + Balochi + Syrian @ 1.983867
17 Kumyk + Iranian + Balochi + Palestinian @ 1.994168
18 Kumyk + Brahui + Azerbaijani + Iraqi_Jew @ 1.995915
19 Kumyk + Balochi + Azerbaijani + Iraqi_Jew @ 1.996019
20 Kumyk + Kurdish + Balochi + Iraqi_Jew @ 2.021435

Done.

Elapsed time 1.3077 seconds.


MDLP K16 2xOracle and OracleX4

Admix Results (sorted):

# Population Percent
1 Caucasian 51.20
2 Indian 20.27
3 Steppe 11.42
4 NearEast 6.59
5 Neolithic 6.57
6 Siberian 1.48


Finished reading population data. 517 populations found.
16 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Iran_Mazandrani_Mazandrani @ 3.984978
2 Iranian_Iran @ 5.117136
3 Iranian_Fars @ 5.200183
4 Iran_Lor_Lor @ 5.721693
5 Zoroastrian_Iran @ 6.190014
6 Kurd_Syria @ 7.707963
7 Iranian_Kerman @ 8.452819
8 Azerbaijani_Baku @ 8.596276
9 Azeri_Azerbaijan @ 9.450549
10 Azerbaijani_Dagestan_Azerbaijan_Dagestan @ 10.068139
11 Azerbaijani_Dagestan_Azerbaijan_Dagestan @ 10.489834
12 Jew_Uzbekistan @ 10.625613
13 Azerbaijani_Azerbaijan @ 11.445561
14 Azerbaijani_Khachmaz_Azerbaijan_Khachmaz @ 11.928846
15 Azerbaijani_Agdzhabedi_Azerbaijan_Agdzhabedi @ 12.898784
16 Assyrian_Iraq @ 12.927719
17 Armenian_Martuni @ 12.942954
18 Armenian_Dprabak @ 13.029356
19 Assyrian_Turkey @ 13.112374
20 Armenian_Chambarak @ 13.165009

Using 2 populations approximation:
1 50% Iran_Lor_Lor +50% Iran_Mazandrani_Mazandrani @ 3.158137


Using 3 populations approximation:
1 50% Armenian_Armenia +25% Balochi_Baluchistan +25% Iran_Mazandrani_Mazandrani @ 1.767233


Using 4 populations approximation:
++++++++++++++++++++++++++++++++++++++++++++++++++ ++++++++++++++++++++++++++++++++++++++++++++++++++ +++++++++++++++++++++++
1 Armenian_Armenia + Brahui_Baluchistan + Georgian_Svanetia + Iran_Lor_Lor @ 1.413549
2 Armenian_Armenia + Balochi_Baluchistan + Georgian_Svanetia + Iran_Lor_Lor @ 1.438016
3 Armenian_Armenia + Georgian_Kakheti + Iran_Mazandrani_Mazandrani + Makrani_Pakistan @ 1.542434
4 Azerbaijani_Baku + Brahui_Baluchistan + Georgian_Kakheti + Georgian_Kakheti @ 1.614667
5 Armenian_Armenia + Brahui_Baluchistan + Georgian_Svanetia + Iranian_Iran @ 1.640866
6 Azerbaijani_Baku + Balochi_Baluchistan + Georgian_Kakheti + Georgian_Kakheti @ 1.651564
7 Armenian_Armenia + Brahui_Baluchistan + Georgian_Svanetia + Iran_Lor_Lor @ 1.668521
8 Armenian_Armenia + Balochi_Baluchistan + Georgian_Svanetia + Iranian_Iran @ 1.678542
9 Armenian_Armenia + Balochi_Baluchistan + Georgian_Svanetia + Iran_Lor_Lor @ 1.715279
10 Armenian_Armenia + Brahui_Baluchistan + Georgians_Zugdidi + Iran_Lor_Lor @ 1.727811
11 Georgian_Kakheti + Iran_Mazandrani_Mazandrani + Iran_Mazandrani_Mazandrani + Iran_Mazandrani_Mazandrani @ 1.729237
12 Armenian_Armenia + Brahui_Baluchistan + Georgian_Kakheti + Iran_Lor_Lor @ 1.755088
13 Armenian_Armenia + Balochi_Baluchistan + Georgian_Kakheti + Iran_Lor_Lor @ 1.755437
14 Armenian_Armenia + Brahui_Baluchistan + Georgian_Svanetia + Kurd_Syria @ 1.766955
15 Armenian_Armenia + Armenian_Armenia + Balochi_Baluchistan + Iran_Mazandrani_Mazandrani @ 1.767233
16 Armenian_Armenia + Brahui_Baluchistan + Georgian_Svanetia + Iranian_Fars @ 1.774321
17 Armenian_Armenia + Balochi_Baluchistan + Georgians_Zugdidi + Iran_Lor_Lor @ 1.784194
18 Armenian_Armenia + Brahui_Baluchistan + Georgian_Georgia + Iran_Lor_Lor @ 1.793256
19 Armenian_Armenia + Balochi_Baluchistan + Georgian_Svanetia + Kurd_Syria @ 1.801614
20 Armenian_Armenia + Balochi_Baluchistan + Georgian_Georgia + Iran_Lor_Lor @ 1.804562

JonikW
04-09-2018, 10:44 AM
Can anyone tell me whether these results look 25% Georgian? My wife is from Georgia; known ancestry 25% Georgia (Kakhetia), 75% Russia (Urals + Kursk region). Her 23and me results are just in but not yet phased with our son. She hasn't tested anywhere else. Does this level of Balkans and Italy look expected? I notice she has some cousin matches on the first couple of pages with Romanian surnames but also wonder whether the Balkans/Italy combination could be Greek or something else. Her Russian cousin matches are pretty much 100% Eastern Europe Russia.
Her Gedmatch Genesis kit: NS4188352 (I only have access to a smartphone and the formatting messed up when I tried to post oracles here.)

- European 87.3%

- Eastern European 38.6% Russia

- Balkan 27.9%

- Italian 7.8%

- Finnish 0.9%

- Sardinian 0.4%

- Broadly Southern European 5.9%

- Broadly Northwestern European 1.3%

- Broadly European 4.6%

- Western Asian & North African12.0%

- Western Asian10.8%

- Broadly Western Asian & North African1.2%

- Sub-Saharan African0.1%

- East African0.1%

- Unassigned0.5%

Fungene
04-09-2018, 11:38 AM
Just a thought:
I have very high Balkan in 23andme, and no connection to Georgia shows up in DNA Relatives.
The Western Asian probably captures the Georgian ancestry, but 38.9% is quite low Eastern European for someone who is 75% Russian.
It is quite possible that there is real Romanian ancestry, especially given the names that show up, I am presuming, in DNA relatives.
Here is something to consider, although it might not be relevant to your wife's ancestry, meaning the Romanian individual's presence in the former Soviet Union was voluntary, and his/her case had nothing to do with the population dispersal described below:
https://www.geopolintelligence.com/contact/
"Definitive figures are hard to assess, but the number of Moldovan deportees throughout the years of Soviet rule is considered to be around half a million. According to the 1958 edition of the British Encyclopedia (volume 15, p. 662), it was estimated that by mid-1955, the Soviet authorities had deported about 500,000 people from the MSSR. A corroborating indication is the fact that in 1979, according to Soviet statistics, there were 415,371 Moldovans living in Ukraine, over 100,000 in various parts of the USSR’s Russian Federation, including Siberia and the Russia’s Far East, over 33,000 in Soviet Central Asia and other distant places of the USSR."

JonikW
04-09-2018, 12:41 PM
Thanks so much. Are you Georgian? I can see my wife is getting Moldovan and Romanian on some calculators. If it really is Moldovan, might that also account for the Italian percentage? I was disappointed not to see any Georgian cousins. She has some full Georgian cousins back home so it would be good if I could persuade one to test.

Fungene
04-09-2018, 01:09 PM
I have mostly Romanian ancestry, with some 5th generation Greek, and some distant Eastern European ancestry. My 23andme and other results are in the "Balkan" subforum.
Yes, it is quite typical for many Romanians to get some Italian, at least, that's what I see in DNA Relatives in 23andme.
That Southern European (including Balkan, Italian, etc.) in your wife's results looks quite typical for a Romanian.
Not getting Georgian cousins? Probably because not many Georgians have tested, as you have guessed.

JonikW
04-09-2018, 02:35 PM
I'll dig out the late maternal grandmother's family notes and tree and see if they shed any light. We will write to the cousin matches with Romanian names too. I can see Papuc, Dobrescu, Popescu, Dobrescu and Christodouleas as possible candidates. Perhaps you can tell me if those are Romanian. My wife's grandmother certainly didn't look Slavic (dark eyes and olive complexion) although she had a Russian maiden name. She and her husband were from near Kursk. Sadly I never met him but have been told he looked very dark for a Russian. Coincidentally, we all went to Romania and Moldova recently so this all looks very interesting.

Fungene
04-09-2018, 02:55 PM
That certainly sounds like an interesting family history to explore.
Just a note: I would preserve DNA matches' anonymity. If they wish to share information on sites such as Anthrogenica, that's OK.
You've been to Moldova? That's interesting. I've not yet travelled to that country. I have been to Moldavia, in Romania, on my hunt for museums that show fabulous art from the Neolithic. I see online that some museums in Moldova have very important pieces from the Neolithic and Copper Age. I want to visit.

JonikW
04-09-2018, 03:14 PM
Thanks for your reply. I hope giving surnames only wasn't enough to disrespect anonymity or identify anyone. I certainly intended to be careful so hope I've done enough! Did you think at least three of those names were Romanian? Yes, Romania is a wonderful country. We stayed in Timisoara, which was beautiful. We didn't have any time in Moldova to explore unfortunately. We are certainly starting to rewrite the history of my wife's non Georgian side. Both parents are testing soon so we'll find out more then. My wife's father's surname is Kunelashvili and it will be interesting seeing what his Y haplogroup is.

digital_noise
05-01-2018, 08:01 PM
My Heritage Results

Me:
English 57.1%
Italian/Greek 30.3%
Eastern Europe 11.6%
Ashkenazi Jewish 1.0%

Wife:
West Asian 81%
South Asian 9.7%
Central Asian .8%
North African 5.2%
Ashkenazi jewish 3.3%

Daughter:
West Asian 52.1%
NW Europe 33.3%
Italian 12.7%
Ashkenazi Jewish 1.9%

My results state English, my daughters state Irish/Scottish/Welsh. In reality its all of the above depending on how far back you want to go so both could be correct, technically.

ABC
07-03-2018, 10:16 PM
I'm 3/4 laz-georgian (turkish) and 1/4 kurd

DNAland:
West Eurasian 100% - Central Indoeuropean 89%
South European 10%- Balkan 7.2% South/Central European -3%
Ambiguous 1.1%

For the rest of the calculators, I'm too lazy to search it all up again :)

XooR
07-26-2018, 12:51 AM
ABC which company you have tested from?

Dmitry
08-01-2018, 01:30 AM
I have some Afghan GEDmatch results if any members are interested, as they are related to Iranians and Kurds, and somewhat related to North Caucasus groups.

XooR
09-13-2018, 12:47 PM
My Ancestry results just updated and now I am 100% Turkey and the Caucasus

25882

Abd.H
09-13-2018, 11:42 PM
My Ancestry results just updated and now I am 100% Turkey and the Caucasus

25882

I see that you had 25% Italy/Greece before and now it disapeared
I had also 20% Italy/Greece before and now it disapeared
and my Caucasus increased from 38% to 46% Turkey and the Caucasus
and also I got a new region and it is 8% Iran
25897