Detailed Scythian & Sarmatian analysis

[Zoro]

A very comprehensive detailed analysis on the demographic changes on the Iranian plateau since Chalcolithic



Using whole genome sequenced (WGS) ancient and modern DNA sanples, along with qpAdm and qpWave formal analysis [7] our passing admixture models for Kurds, subject to a set of very stringent protocols outlined in the last section, show that present day Kurds derive approximately 58% to 71% from a local West Asian pre-Indo-Iranian pastoralist stock, supplemented with 29% to 42% Scythian/Sarmatian/Parthian related Indo-Iranian admixture from Central Asia (fig 30-31). These admixture events around 2000-2700 years ago between the local pre-Indo-Iranic ancient sheep herders of the Zagros mountains and Iranian plateau, and the invading Scythian/Sarmatian/Parthians/Medes were likely responsible for the proliferation of Indo-Iranian languages in the region.

Herein we summarize the findings of those papers [1,2,3,4,5] and present it along with our own detailed genetic analysis using formal statistics [7].


Introduction of Indo-Iranian languages to the Iranian Plateau

At present Indo-Iranian languages are spoken from the Black Sea region to Xinjiang province in China (Sarikol spoken by Tajiks). Gernot Windfuhr who is professor of Iranian Studies, placed Kurdish under the Parthian branch, albeit with a Median substratum [9]. This is shown in fig 29. Based on our personal fluency of the Kurdish and Pashto languages, there appears to be dozens, if not hundreds of nouns and verbs which are found in both Kurdish and Pashto to the exclusion of Farsi. Since Parhtians originated from the present day Afghanistan region, this could possibly be an artifact of Parthian having a greater influence on the Kurdish and Pashto languages to the exclusion of Farsi, which is classified as a southwestern Iranian language.

It appears that Indo-Iranian became dominant in the Iran region during the Iron-Age with the Medes and Parthians around 2500 years ago. Unterlander, Palstra, and Lazaridis et al [1] showed that Kurds have an almost 100% probability of descent from Iron-Age Scythians (figs 12-14).


https://i.imgur.com/DD4FaIw.jpg

https://eurasiandna.com/new-studies-show-elevated-scythian-east-asian-admixture-in-kurds-as-compared-to-persians-and-ethnic-groups-in-west-asia/

[Zoro]

Using qpAdm [7], we were able to produce good genetic models for Kurds using a 3-way combination of Chalcolithic Iranians + Neolithic Levant + Iron-Age Scythian & Sarmatians.

We used many precautions conducting this genetic analysis using qpAdm and qpWave [7]. They include:

Only using the highest coverage/quality ancient samples from each source populations
Using whole genome sequenced Iraqi Kurd samples to maximize SNP overlap with ancient samples.
Using qpWave [7] to rule out the possibility that Kurds can simply be modeled using 2 streams of ancestry consisting of Chalcolithic Zagrosian Iranians and Neolithic Levant.
Using a high number of either WGS or diploid ancient pright reference populations to maximize detection of subtle genetic differences between Kurds and the 3 pleft sources.

The following pright references were used in the qpAdm analysis:

Jo-Hoan-Simmons
Devils-Gate-Neolithic-WGS
Iran-GanjDareh-N
Anatolia-Neolithic
EHG-I0061-DIPLOID
Morocco-Iberomaurusian
Loschbour-DIPLOID
Kolyma-Mesolithic-WGS
Russia-Sunghir6
Botai-EN-DIPLOID
Yana-UP-WGS

Depending on the Scythian/Sarmatian samples used we were able to maintain an overlap of 220,000 to 400,000 SNPs between the samples.

We used a p-value > 0.05 for passing models, with one model reaching a p-value of 0.99.

Using whole genome sequenced (WGS) ancient and modern DNA sanples, along with qpAdm and qpWave formal analysis [7] our passing admixture models for Kurds, subject to a set of very stringent protocols outlined in the last section, show that present day Kurds derive approximately 58% to 71% from a local West Asian pre-Indo-Iranian pastoralist stock, supplemented with 29% to 42% Scythian/Sarmatian/Parthian related Indo-Iranian admixture from Central Asia (fig 30-31). These admixture events around 2000-2700 years ago between the local pre-Indo-Iranic ancient sheep herders of the Zagros mountains and Iranian plateau, and the invading Scythian/Sarmatian/Parthians/Medes were likely responsible for the proliferation of Indo-Iranian languages in the region.

[vbnetkhio]

A very comprehensive detailed analysis on the demographic changes on the Iranian plateau since Chalcolithic


anything about Sytho-Sarmatian ancestry in East and Southeast Europeans? I wonder if there is any.

[Zoro]

anything about Sytho-Sarmatian ancestry in East and Southeast Europeans? I wonder if there is any.

I’m sure there’s. Why don’t you try qpAdm using the pright pops used in the study and use some Simmons or 1240k Europeans using various western sarmatians/scythians

1st use qpWave to make sure those Europeans can’t be modelled without sarmatians

[vbnetkhio]

I’m sure there’s. Why don’t you try qpAdm using the pright pops used in the study and use some Simmons or 1240k Europeans using various western sarmatians/scythians

1st use qpWave to make sure those Europeans can’t be modelled without sarmatians

I had some problems with qpadm. If i remember right I think I tried the same model twice and got wildly different results. Is it even supposed to be the same on each run?
I suspected the program was overheating my processor so it started giving random results.

i'll give it another shot definitely.

[Mejgusu]

I never saw a Kurdish result with 42% „Scythian/Samartian“, that would be around 13-16% north European on dodecad. They are around 15-30% steppe/early Iranic influenced.

[Kyp]

I never saw a Kurdish result with 42% „Scythian/Samartian“, that would be around 13-16% north European on dodecad. They are around 15-30% steppe/early Iranic influenced.


It's because he is using Scythians as the only source for Indo-Iranians and totally disregards earlier Indo-Iranian (Aryan) migrations.

[xripkan]

A very comprehensive detailed analysis on the demographic changes on the Iranian plateau since Chalcolithic






https://i.imgur.com/DD4FaIw.jpg

https://eurasiandna.com/new-studies-show-elevated-scythian-east-asian-admixture-in-kurds-as-compared-to-persians-and-ethnic-groups-in-west-asia/

Which Scytho-Sarmatian haplogroups have been found among Kurds?

[Zoro]

I had some problems with qpadm. If i remember right I think I tried the same model twice and got wildly different results. Is it even supposed to be the same on each run?
I suspected the program was overheating my processor so it started giving random results.

i'll give it another shot definitely.

I’ve never gotten 2 different results from the same run. Maybe you used a different dataset or a different number of samples from a population in the 2nd run

[Zoro]

It's because he is using Scythians as the only source for Indo-Iranians and totally disregards earlier Indo-Iranian (Aryan) migrations.

Well we know that Indo-Iranian was being used in the iran region around the Medes and Scythian time 2500 years ago but we don’t have evidence of it being used earlier such as in Sintashta’s time 4000 years ago in Iran.

You also have the MAX shift in genetics between Chalcolithic Iranians and present kurds on the E. Asian axis. This would not favor Sintashta. Also kurds form a much closer clade with sarmatians and Scythians than Sintashta/Andronovo using formal stats.

All this would not favor much admixture from a pure Sintashta/ andronovo population but some is possible if we find Indo-Iranian was introduced to Iran significantly before Medes

[Kyp]

Well we know that Indo-Iranian was being used in the iran region around the Medes and Scythian time 2500 years ago but we don’t have evidence of it being used earlier such as in Sintashta’s time 4000 years ago in Iran.

You also have the MAX shift in genetics between Chalcolithic Iranians and present kurds on the E. Asian axis. This would not favor Sintashta. Also kurds form a much closer clade with sarmatians and Scythians than Sintashta/Andronovo using formal stats.

All this would not favor much admixture from a pure Sintashta/ andronovo population but some is possible if we find Indo-Iranian was introduced to Iran significantly before Medes

I think Turkmenistan_IA is the best sample we have.

[Leto]

I think Turkmenistan_IA is the best sample we have.
Is it from the Yaz culture? 1500-500 BCE.

[Kyp]

Is it from the Yaz culture? 1500-500 BCE.

I don't know. It's the cleancut Steppe + BMAC mixed R1a-Z93 sample with close to zero East Eurasian.

[Leto]

I don't know. It's the cleancut Steppe + BMAC mixed R1a-Z93 sample with close to zero East Eurasian.
I know that sample. But also Kashkarchi is pretty good, it's a bit older than the TM one. 1200-1000 BC. Found in Fergana valley, Uzbekistan. Both Kashkarchi samples are R1a. They are deleted from Gedmatch now unfortunately.

[Kyp]

I know that sample. But also Kashkarchi is pretty good, it's a bit older than the TM one. 1200-1000 BC. Found in Fergana valley, Uzbekistan. Both Kashkarchi samples are R1a. They are deleted from Gedmatch now unfortunately.

I still have the TKM_IA Dodecad:

YDNA: R1a-z93

Gedrosian: 35.97
Siberian: 1.19
Northwest_African: 0.00
Southeast_Asian: 0.00
Atlantic_Med: 8.33
North_European: 32.42
South_Asian 2.23
East_African 0.00
Southwest_Asian: 0.00
East_Asian: 0.00
Caucasian: 19.16
Sub_Saharan: 0.71

Distance to: Turkmenistan_IA
12.18744846 Pamiri_Rushan
13.36913236 Pamiri_Shughnan
13.82942877 Yagnobi
17.06988283 Pamiri_Ishkashim
18.36052015 Tajiks
19.39978866 Tajik_Tajikistan
20.10360415 Dargin
20.79598519 Tabasaran
21.09303440 Avar
21.20582939 Lak
23.31007293 Pashtun_Kandahar
23.45186560 Lak2
23.52673586 Lezgin
25.17929109 Qumuq
25.52924206 Tajik_Herat
25.75928959 Tajik_Panjshir&Kapisa
25.89588577 Kho
27.07257653 Kalash
27.08688428 Pashtun_Uthmankhel
27.19150787 Pashtun_Tarkalani
27.29554359 Tajik_Balkh
27.32555031 Tajik_Kabul
27.90486517 Nuristani
29.85703602 Pashtun_Yusafzai
30.04058422 Chechen

[Leto]

I still have the TKM_IA Dodecad:

YDNA: R1a-z93

Gedrosian: 35.97
Siberian: 1.19
Northwest_African: 0.00
Southeast_Asian: 0.00
Atlantic_Med: 8.33
North_European: 32.42
South_Asian 2.23
East_African 0.00
Southwest_Asian: 0.00
East_Asian: 0.00
Caucasian: 19.16
Sub_Saharan: 0.71

Distance to: Turkmenistan_IA
12.18744846 Pamiri_Rushan
13.36913236 Pamiri_Shughnan
13.82942877 Yagnobi
17.06988283 Pamiri_Ishkashim
18.36052015 Tajiks
19.39978866 Tajik_Tajikistan
20.10360415 Dargin
20.79598519 Tabasaran
21.09303440 Avar
21.20582939 Lak
23.31007293 Pashtun_Kandahar
23.45186560 Lak2
23.52673586 Lezgin
25.17929109 Qumuq
25.52924206 Tajik_Herat
25.75928959 Tajik_Panjshir&Kapisa
25.89588577 Kho
27.07257653 Kalash
27.08688428 Pashtun_Uthmankhel
27.19150787 Pashtun_Tarkalani
27.29554359 Tajik_Balkh
27.32555031 Tajik_Kabul
27.90486517 Nuristani
29.85703602 Pashtun_Yusafzai
30.04058422 Chechen
Found one of the Kashkarchi

Kashkarchi_BA (1200-1000 BC)
N1a1a1
R1a1a1b

NE-Euro 52.21
Baloch 22.38
Mediterranean 14.72
Caucasian 3.86
American 1.93
S-Indian 1.76
E-African 1.2
Beringian 0.65
Siberian 0.63
SE-Asian 0.44
W-African 0.18

North_European 50.37
Gedrosia 21.01
Atlantic_Med 18.98
South_Asian 3.73
Caucasus 2.75
Siberian 1.83
Sub_Saharan 0.8
East_African 0.53

Both are on G25 of course.

[Zoro]

I think Turkmenistan_IA is the best sample we have.

Yes it's a 2700 year old decent quality sample (Early Mede period). It seems to be the only one of its kind. Perhaps it's ancestral to Parthians or perhaps similar to Medes.

Didn't I post some qpWave results using it a while back. I'll look at it again and I'll mention to Dilawer also to see if it's possible to model a modern Kurd or Azeri using it + Scythian. It maybe difficult if it's too similar to Scythians. Qpadm doesn't like 2 samples in the same run that are somewhat similar but my guess is it probably doesn't have enough E Eurasian to explain the E Asian shift in kurds compared to Iran-chl

[Aren]

Well we know that Indo-Iranian was being used in the iran region around the Medes and Scythian time 2500 years ago but we don’t have evidence of it being used earlier such as in Sintashta’s time 4000 years ago in Iran.

You also have the MAX shift in genetics between Chalcolithic Iranians and present kurds on the E. Asian axis. This would not favor Sintashta. Also kurds form a much closer clade with sarmatians and Scythians than Sintashta/Andronovo using formal stats.

All this would not favor much admixture from a pure Sintashta/ andronovo population but some is possible if we find Indo-Iranian was introduced to Iran significantly before Medes

Yet you are ignoring the early Iron Age R-Z93 from Turkmenistan, genetically 50% Sintastha-like with no East Eurasian. Iran was not populated with 100% Sintashta/Andronovo-like Indo-Iranians, but rather more similar to this IA Turanian.
The minor East Asian in Kurds is rather from recent Ottoman era contacts with Turks.

[Aren]

I think Turkmenistan_IA is the best sample we have.

I always assumed all of the Steppe in Kurds and Iranians is from Central Asia and from their Iranic side, but that might not be the case. We're up for a surprise with upcoming Bronze Age and pre-Iranic Iron Age samples from Western Iran and Mesopotamia. Steppe is gonna be found and along with it R1b. The source is probably from Northern Caucasus > Armenia > Upper Mesopotamia and Urmia plain.

[Kyp]

I always assumed all of the Steppe in Kurds and Iranians is from Central Asia and from their Iranic side, but that might not be the case. We're up for a surprise with upcoming Bronze Age and pre-Iranic Iron Age samples from Western Iran and Mesopotamia. Steppe is gonna be found and along with it R1b. The source is probably from Northern Caucasus > Armenia > Upper Mesopotamia and Urmia plain.

Might be true,

I think the Hasanlu sample also already had steppe ancestry pre-dating Iranian migrations and was R1b if I remember correctly.

[Mejgusu]

I always assumed all of the Steppe in Kurds and Iranians is from Central Asia and from their Iranic side, but that might not be the case. We're up for a surprise with upcoming Bronze Age and pre-Iranic Iron Age samples from Western Iran and Mesopotamia. Steppe is gonna be found and along with it R1b. The source is probably from Northern Caucasus > Armenia > Upper Mesopotamia and Urmia plain.

I really doubt that this steppe influence was immense. Armenians have a lot of R1b and have just small steppe influence. But this pre Iranic steppe influence could be true since first IE-speakers came from Northern Caucasus to Westasia.

[Zoro]

Might be true,

I think the Hasanlu sample also already had steppe ancestry pre-dating Iranian migrations and was R1b if I remember correctly.

No doubt that 2700 year old Hasanlu (early Mede period) has steppe ancestry because formal stats show that Kurds are considerably more steppe and east asian shifted when compared with Iran-Chl than when compared with Hasanlu. Therefore Hasanlu must have picked up steppe since Iran-Chl. It's possible that it can be used to represent Mede in some fashion.

[Zoro]

I always assumed all of the Steppe in Kurds and Iranians is from Central Asia and from their Iranic side, but that might not be the case. We're up for a surprise with upcoming Bronze Age and pre-Iranic Iron Age samples from Western Iran and Mesopotamia. Steppe is gonna be found and along with it R1b. The source is probably from Northern Caucasus > Armenia > Upper Mesopotamia and Urmia plain.

It could easily be so. Steppe and Central Asian admixture into Iran is certainly not a one time pulse but a continuous process. That's why when a Kurd is modeled as local W. Asian + Steppe this does not imply a one time admixture event but rather the steppe amount is the amount accumulated over a 3000 year period from multiple events.

When we get those genomes we will have to see how much of the E. Asian, S. Asian, and steppe shift in Iranics compared with Iran-Chl can be explained by those samples. My guess is they maybe able to explain some of the shift on 1 or 2 of those 3 axes but not all. It will be something to look forward to.

We also know that Parthians/Scythians/Sarmatians, Medes Turkics and Zoroastrianism were all things that affected W. Iranics to some degree.

[Zoro]

Iran was not populated with 100% Sintashta/Andronovo-like Indo-Iranians, but rather more similar to this IA Turanian.

Let's see if we have more of these types excavated in the region.


The minor East Asian in Kurds is rather from recent Ottoman era contacts with Turks.

A couple of things we should keep in mind:

If we find E Asian only in some members of a population that would indicate more recent admixture and within short order that admixture would be quickly diluted in a population

If we find E Asian widespread in a population with some regional variations. With a very populous group such as Kurds this would indicate older >1000 year old E Asian introgresion which is more resistant to dilution except if Kurds have hybridized or will hybridize with E Asian poor neighbors.

It maybe minor using a K11 or greater Admixture based calculator but that would not be the proper way to measure it. I'll illustrate with an example.

Let's say we both do extensive formal tests and determine we both have the same E Asian level let's say 10%.

Now suppose we both test with a K11 calculator which has 2 W Asian components whose references are genetically much more similar to me than you.

What will end up happening is I may get 2% E Asian but you get 5% E Asian with this calculator even though we both actually have 10%. Reason being is that those 2 W. Asian reference comps are genetically much more similar to me than you. Thus the overwhelming majoirty of my admixture gets assigned to those 2 comps unlike you. And if it's old E Asian then those references have almost as much as me, thus nearly all of my E Asian except for 2% gets absorbed into these 2 comps

[Zoro]

I found some old qpAdm runs where Turkmenistan_IA was used as an Indo-Iranian steppe proxy. There's no doubt that this 2700 year old Z93 sample has a special relationship with Kurds and other Iranics because whenever it's introduced into the run Iran-Chl drops in Kurds and the majority of admixture gets assigned to Turkmenistan-IA. However in spite of this model still needs some Scythian-Sarmatian to pass.

In fact when a Kurd is modeled as Iran-chl + Levant-N + Turkmenistan-IA + Scythian/Sarmatian, with some Scythians/Sarmatians, the Scythian/Sarmatian assignment is greater than Turkmenistan-IA and in those instances Levant-N is not as much affected but Iran-Chl drops off to very low levels.

This would suggest the majority of admixture in present day Kurds is invasive from the steppe

This sample certainly deserves more attention. I will try to examine it more at length from various angles including what percentage E Eurasian it is from a formal stat standpoint.

[PaleoEuropean]

It's because he is using Scythians as the only source for Indo-Iranians and totally disregards earlier Indo-Iranian (Aryan) migrations.

Scythians were also mutts, not like they were pure Asian. Scythians are just the the mixed central Asians that moved out of Iranian Central Asia, Kurds are Iranian mixed central Asians that moved out of Iran and Central Asia mostly during the Mongol Invasion's.

[PaleoEuropean]

How many people are in the source populations and how many are in the target populations?

[Mejgusu]

Scythians were also mutts, not like they were pure Asian. Scythians are just the the mixed central Asians that moved out of Iranian Central Asia, Kurds are Iranian mixed central Asians that moved out of Iran and Central Asia mostly during the Mongol Invasion's.

That only counts for eastern Scythians, western Scythians were, with some exceptions, predominantly West Eurasian. Westasian Iranics like Kurds, Iranians or Talysh are mostly descendants of the western Scythians, which were genetically like 75% Russian+ 25% Tadjik/Pashtun. East Eurasian influence was mostly visible among those who had contact with eastern scythia. If i am not wrong i once saw a Sassanid? sample which was quite similar with nowadays Iranians, just more steppe shifted. It is true that some Iranic always had east eurasian influence but nowadays Westasian Iranics got it from Turkic expansion. I can give some points for theory, for instance the most steppe influenced Kurd i have ever seen was from Central Anatolian Konya. Those Kurds were separated early during Sunni-Shia conflict were Qizilbash Turks massacred or went to Azerbaijan in 15/16th century and Kurds were settled to Konya. Those Kurds have nearly no-east eurasian, in the contrary some in east Anatolia show around 3% east eurasian(the highest was 5/6%). The ethnic structure in east Anatolia was far more complicated and diverse than to other parts of Anatolia, still there are cities with multiethnic makeup.

[Zoro]

How many people are in the source populations and how many are in the target populations?

This is the response I received regarding this:


Where the ancient population displayed a very large amount of genetic variation on PCA and IBD, and when there was a WGS or diploid sample available for the population, only the highest coveraged (quality) sample was used to represent the population. Thus for the majority of sources and references the best sample was used.
For the Kurdish WGS sample a Kurmanji Iraqi sample was used for which the non-WGS version showed that it fell within the normal amount of Kurdish variation. This can clearly be seen where the sample is a Kurd-C sample in our previous study consisting of several Kurds https://eurasiandna.com/impact-of-the-iron-age-saka-and-scythians-on-south-west-asian-demography/

This table which I copied from the website EurasianDNA.com indicates the sample ID for the sources

https://i.imgur.com/N2o8Znq.jpg

[Zoro]

That only counts for eastern Scythians, western Scythians were, with some exceptions, predominantly West Eurasian. Westasian Iranics like Kurds, Iranians or Talysh are mostly descendants of the western Scythians, which were genetically like 75% Russian+ 25% Tadjik/Pashtun.

To make sure there's no confusion, no Hungarian Scythian samples were used to represent W. Scythians in both the Scythian paper by various authors including Lazaridis and referenced in the study at https://eurasiandna.com/new-studies-show-elevated-scythian-east-asian-admixture-in-kurds-as-compared-to-persians-and-ethnic-groups-in-west-asia/

The Scythian paper mentioned used 3 samples to represent W. Scythians. 2 were from W. Kazakhstan and 1 from the Aral Sea area. In the EurasianDNA study only Kazakhstan, Kyrgistan and Mongolian samples were used. IDs are labeled in the above table


in the contrary some in east Anatolia show around 3% east eurasian(the highest was 5/6%)

If those numbers are based on a K7 or higher Admixture calculator then they should not be relied on because I can make an Admixture based calculator where Europeans, W. Asians, and C. Asians get 0% E Asian (something like 23andme). Here's how I can design it:

I would make 3 Eastern European components, 3 N.Caucasian comps, 3 Iranian comps, 3 Turkish comps, 3 S. Caucasian comps, 3 Afghan comps. All testers will pretty much be absorbed into their own clusters and will score near 0% E. Asian.

If you want to measure E. Asian based on Admixture calculator and not formal stats, that's fine, but you must design it so that:

- One comp is based on ENF (W Eurasian)
- One comp is based on WHG (W Eurasian)
- One comp is based on Devils-Gate-N (E Eurasian)
- One comp is based on Onge and Papuans (S Eurasian)
- One comp based on Pygmies (SSA)

I probably wouldn't add a Siberian comp because it may contain some W. Eurasian. If you have alot of Siberian, the Devils-Gate will pick it up because modern Siberians ( NeoSiberians) are substantially Kolyma-Mesolithic derived meaning a mix of ancient E Asian and ANE/Yana-UP

[Chelubey]

I see here a collision between historical/linguistic data and the genetic concept.
I think historical information / linguistic evidence should not be ignored in favor of a private genetic concept.
According to historical / linguistic / archaeological data, the Proto-Slavic ancestral home was located somewhere in the region of Poland / western Belarus two thousand years ago.
But try to imagine that some geneticist connects the origin of the Proto-Slavs with a hypothetical Central Asian component that appeared in the genetics of the Slavs 1000 years ago. Should we ignore historical / linguistic data and logic in this case? I think no.
I think the collision being discussed here can be solved very simply:
1) Zoro's models are possiblyy correct, the steppe component of the Kurds comes from the Turkic-Huns and Scythians-Sarmatians.
2) Turks-Huns-Scythians did not bring the Kurdish language to the Iranian plateau, they switched to the Kurdish language later.
(I do not think that serious historical linguists will agree that the Scythians spoke the Kurdish (Proto-Kurdish) language.

https://en.wikipedia.org/wiki/Achaemenid_Empire
https://en.wikipedia.org/wiki/Darius_the_Great#European_Scythian_campaign

[Demhat]

A very comprehensive detailed analysis on the demographic changes on the Iranian plateau since Chalcolithic






https://i.imgur.com/DD4FaIw.jpg

https://eurasiandna.com/new-studies-show-elevated-scythian-east-asian-admixture-in-kurds-as-compared-to-persians-and-ethnic-groups-in-west-asia/
Nice study,

sidenote The thing is, what I have been telling for some time now, when you mention the linguistic Indo_Iranian ancestors of Kurds like a mix of Parthian/Media/Scyhthian/Sarmatian, you haven't really understand the ethno_genetic origin of these groups and inner relationship of these to each other very well. With you I am not referring to you but most people talking about this subject in specific.

What Medes and Parthians are is different from what people understand. The Medes did not exist as one unified branch of Indi_Iranians but a tribal union of Indo_iranic and non Indo_Iranic tribes that formed in what is nowadays North/West Iran, Mesopotamia and Transcaucasus (Anatolia,South Caucasus).

1. According to Greek sources 2 to 3 of the initial/founding Median tribes among the 7, were actually of Scythian or Scythian derived origin.
2. On top of that Media was slightly later settled and ruled by another tribe of Scythians. This time the Royal Scythians who came from East Ukraine/South Russia.
3. We also have Greek sources that Cimmerians settled in territory of the Medes
4. What we understand under Parthians are literally a Median core that once again got enriched with additional Scythian admixture
5. The Parthian dynasty, according to Roman Sources, settled Alan/Sarmatian horse archers on the borders to the Roman Empire (Iraqi Kurdistan) to fight them off.

All in all. All these Scythian/Sarmatian elements were included among the Medes and Parthians and this tribal diversity of closely related groups is still mirrored in the modern Kurds.


And in my opinion this is also reflected here by the results. When you don't have Median/Parthian samples as proxy at hand. You will have Kurds be modeled as a mixture of Iran_CHL + a Scythian like source (since Scythians represent best the Steppe Iranic ancestry) and a little of other admixture (Levant_Neo/Anatolian_Neo).

If we had Median/Parthian samples at hand that would probably be how they would be modeled too.

[Demhat]

Using whole genome sequenced (WGS) ancient and modern DNA sanples, along with qpAdm and qpWave formal analysis [7] our passing admixture models for Kurds, subject to a set of very stringent protocols outlined in the last section, show that present day Kurds derive approximately 58% to 71% from a local West Asian pre-Indo-Iranian pastoralist stock, supplemented with 29% to 42% Scythian/Sarmatian/Parthian related Indo-Iranian admixture from Central Asia (fig 30-31).


This is the part on which I disagree because the authors make the assumption that Parthians genetically resembled the Scythians/Sarmatians. While historic/linguistic data indicates otherwise. We don't really have any Parthian sources so far. I assume Parthian samples from North and West Iran, which were basically Medes will overlap by 80-90%

[Zoro]

I see here a collision between historical/linguistic data and the genetic concept.
I think historical information / linguistic evidence should not be ignored in favor of a private genetic concept.
According to historical / linguistic / archaeological data, the Proto-Slavic ancestral home was located somewhere in the region of Poland / western Belarus two thousand years ago.
But try to imagine that some geneticist connects the origin of the Proto-Slavs with a hypothetical Central Asian component that appeared in the genetics of the Slavs 1000 years ago. Should we ignore historical / linguistic data and logic in this case? I think no.
I think the collision being discussed here can be solved very simply:
1) Zoro's models are possiblyy correct, the steppe component of the Kurds comes from the Turkic-Huns and Scythians-Sarmatians.
2) Turks-Huns-Scythians did not bring the Kurdish language to the Iranian plateau, they switched to the Kurdish language later.
(I do not think that serious historical linguists will agree that the Scythians spoke the Kurdish (Proto-Kurdish) language.

https://en.wikipedia.org/wiki/Achaemenid_Empire
https://en.wikipedia.org/wiki/Darius_the_Great#European_Scythian_campaign

Yes language gets a little messy on how much is picked up from neighbors and admixtures vs how much is changed from one generation to the other, but I tend to agree with the following in the article


Gernot Windfuhr who is professor of Iranian Studies, placed Kurdish under the Parthian branch, albeit with a Median substratum [9]. This is shown in fig 29

It seems that Parthians contained Scythians such as Parni and others.

Anyways, I think people have a valid point as to whether Indo-Iranian was originally brought into Iran by non-Samatian/Scythians such as 2700 year old Turkmenistan-IA and that if we model Kurds as

Iran-Chl + Levan-N + Turkmenistan-IA + Scythian/Sarmatian/Medieval Turk then Scythian/Sarmatian/Medieval Turk percentage will go down.

I have relayed this concern to EurasianDNA and they agree it's a valid one and they have been working on such models (above) for the past month.

I wasn't able to get too many details on the outcomes but they did convey that they will publish those models soon and they did get some to work depending on the Scythian sample. They also mentioned that with Turkmenistan-IA introduced, Iran-Chl falls off substantially and Scythian falls off a little but Turkmenistan-IA fails in certain models containing Scythians. We'll just have to wait for the results

[Demhat]

It's because he is using Scythians as the only source for Indo-Iranians and totally disregards earlier Indo-Iranian (Aryan) migrations.

That is not the reason principal component calculators are not very effective in this regard. The 29-42% Scythian like/derived ancestry is likely very accurate at the end of the day it is from professional scientists not semi amateur bloggers.

However what you probably mean is that this analysis is ignoring the likely possibility that allot of this Scythian ancestry is actually derived via the Median tribes. This is because we don't have Median/Parthian samples yet. So yes the 29-42% Scythian Sarmatian like ancestry is correct but as usual it depends on the point of definition and modeling.


I am sure this 29-42% Scythian/Sarmatian + Iran_CHL + some Levant/Anatolian_Neo ancestry could be defined as
~80-90% Median + 10-15% extra Scythian/Sarmatian + 5-10% other.

[Zoro]

This is the part on which I disagree because the authors make the assumption that Parthians genetically resembled the Scythians/Sarmatians. While historic/linguistic data indicates otherwise. We don't really have any Parthian sources so far. I assume Parthian samples from North and West Iran, which were basically Medes will overlap by 80-90%

Well we do know that Parthians were diverse and contained Scythians such as Parni and who knows what else. We also know they originated from the Turkmenistan/E Iran/Afghanistan area so they should have some DNA from that area. But let's wait for the samples to come.

The point with the modeling is that existing known samples are used. So if a model using 35% Saka-DA47 passes then it just means that Kurds have 35% admixture similar to Saka-DA47. The model doesn' t tell us whether that admixture was conveyed to Kurds by Parthians or Medieval Turks or whoever. It's up to us to pick the logical candidate based on history which sometimes is not easy to do.

[Zoro]

That is not the reason principal component calculators are not very effective in this regard. The 29-42% Scythian like/derived ancestry is likely very accurate at the end of the day it is from professional scientists not semi amateur bloggers.

However what you probably mean is that this analysis is ignoring the likely possibility that allot of this Scythian ancestry is actually derived via the Median tribes. This is because we don't have Median/Parthian samples yet. So yes the 29-42% Scythian Sarmatian like ancestry is correct but as usual it depends on the point of definition and modeling.


I am sure this 29-42% Scythian/Sarmatian + Iran_CHL + some Levant/Anatolian_Neo ancestry could be defined as
~80-90% Median + 10-15% extra Scythian/Sarmatian + 5-10% other.

You may be correct. Let's wait for them to publish, but to be clear they're using Turkmenistan-IA since that is available to them

As far as 2700 year old Hasanlu-IA that can also be used as long as we determine how Hasanlu-IA can be modeled because that sample has obvious steppe

[Demhat]

You also have the MAX shift in genetics between Chalcolithic Iranians and present kurds on the E. Asian axis. This would not favor Sintashta. Also kurds form a much closer clade with sarmatians and Scythians than Sintashta/Andronovo using formal stats.

All this would not favor much admixture from a pure Sintashta/ andronovo population but some is possible if we find Indo-Iranian was introduced to Iran significantly before Medes

That should be obvious but many amateurs and amateur bloggers still can't resist to make their models based on Sintashta/Andronovo samples. The Indo_Iranic ancestors of the Kurds would much more likely arrived from a Scythian/Sarmatian like source from Iron Age South Turkmenistan and/or North Caucasus/Western Steppes. Than directly from Sintashta/Andronovo. Trying to model the Kurds via Sintashta would be like trying to model the ancient ancestors of Greeks via Yamnaya/Srubna/Catacomb ancestry. What some amateurs actually did.

[Zoro]

That should be obvious but many amateurs and amateur bloggers still can't resist to make their models based on Sintashta/Andronovo samples. The Indo_Iranic ancestors of the Kurds would much more likely arrived from a Scythian/Sarmatian like source from Iron Age South Turkmenistan and/or North Caucasus/Western Steppes. Than directly from Sintashta/Andronovo. Trying to model the Kurds via Sintashta would be like trying to model the ancient ancestors of Greeks via Yamnaya/Srubna/Catacomb ancestry. What some amateurs actually did.

Agreed but the ones that say Kurds should be modeled as Iran-Chl + Levan-N + Turkmenistan-IA + Scythian/Sarmatian also have a point because we should see what steppe layer of pre-Scythian Kurds may have. IMO Medes (and Hasanlu-IA) contained that pre-Scythian layer of steppe

[Demhat]

Well we do know that Parthians were diverse and contained Scythians such as Parni and who knows what else. We also know they originated from the Turkmenistan/E Iran/Afghanistan area so they should have some DNA from that area. But let's wait for the samples to come.


The Parni were not really the Parthians. That is a mistake many people do. They confuse them or use it like synonyms. The Parni were basically a Nomadic tribe from modern Turkmenistan that spoke a language inbetween Scythian and Median. They were an early elite tribe and founders of the Parthian Dynasty, however the Parthian Core was predominantly made up by Medes.



The point with the modeling is that existing known samples are used. So if a model using 35% Saka-DA47 passes then it just means that Kurds have 35% admixture similar to Saka-DA47. The model doesn' t tell us whether that admixture was conveyed to Kurds by Parthians or Medieval Turks or whoever. It's up to us to pick the logical candidate based on history which sometimes is not easy to do.


Exactly!
And that is my point. I don't doubt for a second that Kurds have 30-40% Scytho_Sarmatian like/derived ancestry. However how that ancestry ended up among Kurds, via "proxy " sources (such as the Medes/Parthians) or directly (via Royal Scythians and Saka) that is up to debate. And imo it makes historically most sense that both is the case. And that is how I came up with my assumed model of ~80-90% Median + 10-15% extra Scythian/Sarmatian + 5-10% other (extra Caucasian, Anatolian, Levantine/Arabian, Turkic admixture).

[Kyp]

Agreed but the ones that say Kurds should be modeled as Iran-Chl + Levan-N + Turkmenistan-IA + Scythian/Sarmatian also have a point because we should see what steppe layer of pre-Scythian Kurds may have. IMO Medes (and Hasanlu-IA) contained that pre-Scythian layer of steppe

It's not even confirmed that Hasanlu_IA is Iranian. Ydna is also not Iranian but R1b similar to Armenians.. It also predates the Medes. It still might be Median but I would be a bit cautious with that sample.

As someone pointed out a third steppe-migration from the Caucasus route has to be considered here.

[Zoro]

The Parni were not really the Parthians. That is a mistake many people do. They confuse them or use it like synonyms. The Parni were basically a Nomadic tribe from modern Turkmenistan that spoke a language inbetween Scythian and Median. They were an early elite tribe and founders of the Parthian Dynasty, however the Parthian Core was predominantly made up by Medes.

Exactly!
And that is my point. I don't doubt for a second that Kurds have 30-40% Scytho_Sarmatian like/derived ancestry. However how that ancestry ended up among Kurds, via "proxy " sources (such as the Medes/Parthians) or directly (via Royal Scythians and Saka) that is up to debate. And imo it makes historically most sense that both is the case. And that is how I came up with my assumed model of ~80-90% Median + 10-15% extra Scythian/Sarmatian + 5-10% other (extra Caucasian/Anatolian/Arabian/Turkic admixture).

I was only suggesting that Parthians would have Parni among their ranks https://en.wikipedia.org/wiki/Parni. We won't know for sure until a dozen Parthians are excavated. We would be speculating at this point. But at this point they should be considered one of the logical proxies for that Saka


In the middle of the 3rd century BCE, the Parni invaded Parthia, "drove away the Greek satraps, who had then only just acquired independence, and founded a new dynasty",[5] that of the Arsacids.

[Demhat]

Is it from the Yaz culture? 1500-500 BCE.

No unfortunately not it is slightly more Northern located from Yaz. Yaz would probably be slightly more Iran_Neo shifted. But Turkemistan_IA is the closest thing we have to how YAZ might have looked like.

[Demhat]

I was only suggesting that Parthians would have Parni among their ranks https://en.wikipedia.org/wiki/Parni. We won't know for sure until a dozen Parthians are excavated. We would be speculating at this point. But at this point they should be considered one of the logical proxies for that Saka

I have collected allot of sources that shed light on the origin of Parthians and what Parthians really were I will share them the coming days. The thing is Parthians were literally. pred. Medes + admixture from another Scythian like source (Parni). So yes the Parni are most likely one of the Scythian like sources in modern Kurds. I just refuse to call them Parthians because that would indicate the Parthians resembled the Parni more than they did the Medes which is historically, linguistically and probably genetically very incorrect to say. Parni were a principal component of the Parthians but the core of the Parthians were local Medes.

[Zoro]

It's not even confirmed that Hasanlu_IA is Iranian. Ydna is also not Iranian but R1b similar to Armenians.. It also predates the Medes. It still might be Median but I would be a bit cautious with that sample.

As someone pointed out a third steppe-migration from the Caucasus route has to be considered here

Well it's pretty deep into Iran (south of Urmia and Tabriz) and contains alot of Iran-Chl and is not too different from Haji-Firuz-IA which is a bit more NW shifted.

Yes, it's early Mede but I think most think Medes were Indo-Iranian thus Indo-Iranians must have arrived in Iran before Medes (2700 years ago)

Hasanlu could have received steppe from that route or maybe a combination of sources

[Demhat]

It's not even confirmed that Hasanlu_IA is Iranian. Ydna is also not Iranian but R1b similar to Armenians.. It also predates the Medes. It still might be Median but I would be a bit cautious with that sample.

As someone pointed out a third steppe-migration from the Caucasus route has to be considered here.

You are making a msitake here my friend. You make the assumption that R1b did not exist among the Medes based on the fact that Sintashta samples or the IA_Turkmenistan sample are R1a. But Sintashta, not even IA_Turkmenistan are representative for Medes. IA_Turkmenistan might and possibly is one side of the coin (ancestral component) but the other side is the local tribes. So R1b might and most likely did exist in large frequency among the Medes. Also take in mind Turkmenistan is basically a hotspot of R1b too. Medes are not synonym with IA_Turkmenistan neither should they be identical to them.

We also found R1b among Sarmatian samples, despite the "lack of it" in Sintashta and Andronovo. Which brings us back to Srubna, Abashevo, Potapovka (Catacomb derived) cultures as source.

We need to get rid of this thinking scheme that just because a Haplogroup was not found in population A it doesn't mean it isn't found in pop B too.

In mathematics we call this set theory. If set A is within set B. B could still have Elements that are not found in set A at all.

[Kyp]

You are making a msitake here my friend. You make the assumption that R1b did not exist among the Medes based on the fact that Sintashta samples are R1a. But Sintashta, not even IA_Turkmenistan are representative for Medes. IA_Turkmenistan might and possibly is one side of the coin (ancestral component) but the other side is the local tribes. So R1b might and most likely did exist in large frequency among the Medes. Also take in mind Turkmenistan is basically a hotspot of R1b too. Medes are not synonym with IA_Turkmenistan neither should they be identical to them.

We also found R1b among Sarmatian samples.

We need to get rid of this thinking scheme that just because a Haplogroup was not found in population A it doesn't mean it isn't found in pop B too.

In mathematics we call this set theory. If set A is within set B. B could still have Elements that are not found in set A at all.

I just said that we shouldn't blindly assume it's Median.

Especially around Urmia plain/East Anatolia/Northern Mesopotamia there is still a lot of room for future surprises/speculation. I also don't think that's the Median heartland.

[Token]

That is not the reason principal component calculators are not very effective in this regard. The 29-42% Scythian like/derived ancestry is likely very accurate at the end of the day it is from professional scientists not semi amateur bloggers.

Dilawer is not a professional scientist and has never published a paper on genetics. He is proficient with Admixtools, but he is no more professional than Davidski and the likes. I'd describe him as a skilled amateur with quite a controversial take on some topics of Iranian ethnogenesis. I remember him some time ago denying Sintashta influence altogether, suggesting that Iran was iranicized by Iron Age nomads and posting a thread comparing Kurdish language to Scytho-Sarmatian.

[Demhat]

I just said that we shouldn't blindly assume it's Median.

Especially around Urmia plain/East Anatolia/Northern Mesopotamia there is still a lot of room for future surprises/speculation. I also don't think that's the Median heartland.

The Median "heartland" is a very small territory from where most of the linguistic forefathers of the Medes initially started out. That Hamadan, Teheran, Isfahan triangle. This was during the time when the "Medes" were still identifying themselves by their tribal groups and as Aryans according to Greek sources.

The Medes evolved when the Indo_Iranic tribes went into a alliance with each and other local tribes to counter the Assyrian rule. Interestingly all of Western Media (the territory from the Zagros-Alborz mountains into North Mesopotamia and the Transcaucasus/Anatolia) was identified and called, first by the Assyrians and later by the Achaemenids as Gutium (land of the Gutians). In fact it goes as far as the terms Gutian and Mede becoming synonym for each other. We still have no clue what or who the Gutians were. Their language is considered as an isolate. But I know one thing for certain. When linguists call a language an isolate it most often indicates that they don't have enough samples at hand and couldn't yet identify the source. For all we know the Gutians had pretty Indo_European (especially Indo_Iranian) names. My assumption is that this R1b sample could be from the Gutian tribes among the Medes.

Another theory of mine is, well it is actually a pretty acknowledged theory in the scientific world, that the ethnic term "Kurd" is derived from Gutian, who were in later stages called Qardu/qurdu by Assyrian/Syriac and Jewish sources.

The best description of the Medes is the merge of incoming Indo_Ianic and a more local older layer of tribes. Heck there are very strong theories that the first Median King, Daioces, was actually from the Mannaean tribe.

Elements of the Gutians/Mitanni, Mannaeans-Hurrians + incoming Indo_Iranic tribes from around Turkmenistan = Medes

slightly later these Medes got additional influx of Scythian_Sarmatian and Cimmerian genes via the North Caucasus and Steppe.

Another hundreds of years later: Medes + another Scythian source (Parni) = Parthian

And this ultimately leads us to the ethnogenesis of modern Kurds.

As Zoro pointed: when a Kurd is modeled as a mix of older local layer of W.Asian + Steppe this does not imply a one time admixture event but rather the steppe amount is the amount accumulated over a 3000 year period from multiple events..

And Kurds are most likely the result of this merge of tribes that formed the Median/Parthian confederation/dynasty, while still using the ethnic term of the mountainous territory they live in (Gutium/Kurti/Karda/Kardu).

[Demhat]

Dilawer is not a professional scientist and has never published a paper on genetics. He is proficient with Admixtools, but he is no more professional than Davidski and the likes. I'd describe him as a skilled amateur
A very skilled Amateur with a skill set more comparable to Dienekes imo



with quite a controversial take on some topics of Iranian ethnogenesis.


I remember him some time ago denying Sintashta influence altogether, suggesting that Iran was iranicized by Iron Age nomads and posting a thread comparing Kurdish language to Scytho-Sarmatian.

That is not really a conroversial theory, but the general take on events. direct Sintashta influence on Iran is small. The Indo_Iranic ancestry on the Iranian Plateau was brought by Iron Age nomads from people similar to the YAZ culture that is the most established theory from linguists to archeologists.
However these Iron Age linguistic forefathers being directly Scytho_Sarmatians that is up to debate. I agree with you here.

[Zoro]

Dilawer is not a professional scientist and has never published a paper on genetics. He is proficient with Admixtools, but he is no more professional than Davidski and the likes. I'd describe him as a skilled amateur with quite a controversial take on some topics of Iranian ethnogenesis. I remember him some time ago denying Sintashta influence altogether, suggesting that Iran was iranicized by Iron Age nomads and posting a thread comparing Kurdish language to Scytho-Sarmatian.

I agree with you that he's proficient with Admixtools, and that is what this latest article is based on but I don't agree with the comparison with Davidski simply because among other things unlike Dilawer Davidski doesn;t know how to diploid genotype ancient DNA sequences (infact he relies on others to even convert BAM files) and he isn't able to invent population genetics software on his own which is different from currently available software such as Admixture, Admixtools, etc. because this requires an understanding of biioinformatics on a higher level (Dilawer totally on his own invented SAPDA which is totally different software and basis than other out there. It's available at GenePlaza). Davidski is a good blogger and does understand various aspects of population history pretty well.

Anyways, this thread is not about who is best, but is to discuss the ethnogenesis of Kurds and Iranics.

I also don't think you're being fair regarding Sintashta because Dilawer did seriously consider Sintashta in his modelling of Kurds ( check his 2018 article https://eurasiandna.com/impact-of-the-iron-age-saka-and-scythians-on-south-west-asian-demography/) . Also he is currently working on models for Kurds as : Iran-chl + Levant-N + Turkmenistan-IA + Scythian/Sarmatian.

I think many people misunderstood what he said about Sintashta. He was simply suggesting that to get to present Kurds you would have to go through the Iron Age. In other words Sintashta genes would have to be conveyed to Kurds via Iron-Age proxies such as Sarmatians and Turkmenistan-IA

[Demhat]

Or how PaleoEuropean described the Scythians. The Medes/Parthians were mutts of older local layers and incoming Indo_Iranic tribes themselves.

[Aren]

You are making a msitake here my friend. You make the assumption that R1b did not exist among the Medes based on the fact that Sintashta samples or the IA_Turkmenistan sample are R1a. But Sintashta, not even IA_Turkmenistan are representative for Medes. IA_Turkmenistan might and possibly is one side of the coin (ancestral component) but the other side is the local tribes. So R1b might and most likely did exist in large frequency among the Medes. Also take in mind Turkmenistan is basically a hotspot of R1b too. Medes are not synonym with IA_Turkmenistan neither should they be identical to them.

We also found R1b among Sarmatian samples.

We need to get rid of this thinking scheme that just because a Haplogroup was not found in population A it doesn't mean it isn't found in pop B too.

In mathematics we call this set theory. If set A is within set B. B could still have Elements that are not found in set A at all.

R1b-Y23838, the subclade Hasanlu_IA belong to does not exist among Eastern Iranics, Baloch and Pashtuns, upstream is mostly Armenians, Georgians, Chechen, Dagestanis but no Iranians. Hajji Firuz_IA is even older and also belongs to R1b-Z2103. Medes if they had R1b is not from their Indo-Iranian side but an assimilated clade from the pre-Iranian natives of Western/NW Iran.

The R1b found in Scythians, Sarmatians and modern day Central Asians are of different subclades.

[Demhat]

R1b-Y23838, the subclade Hasanlu_IA belong to does not exist among Eastern Iranics, Baloch and Pashtuns, upstream is mostly Armenians, Georgians, Chechen, Dagestanis but no Iranians. Hajji Firuz_IA is even older and also belongs to R1b-Z2103. Medes if they had R1b is not from their Indo-Iranian side but an assimilated clade from the pre-Iranian natives of Western/NW Iran.

The R1b found in Scythians, Sarmatians and modern day Central Asians are of different subclades.

Good to know. This is why I indicated that the R1b might be from the local portion of their ancestors. Even if IA_Turkmenistan like people are representative for the linguistic ancestors of the Medes. There is still the other older layer of their local ancestors from whom they might have got that Haplogroup from. My point is just because a Haplogroup is not typical for Sintashta or other Steppe Indo_Iranic tribes it doesn't mean it is non existent among Medes, since Medes are most likely not identical to IA_Turkmenistan or Sintashta to begin with.


You are making a msitake here my friend. You make the assumption that R1b did not exist among the Medes based on the fact that Sintashta samples or the IA_Turkmenistan sample are R1a. But Sintashta, not even IA_Turkmenistan are representative for Medes. IA_Turkmenistan might and possibly is one side of the coin (ancestral component) but the other side is the local tribes. So R1b might and most likely did exist in large frequency among the Medes. Also take in mind Turkmenistan is basically a hotspot of R1b too. Medes are not synonym with IA_Turkmenistan neither should they be identical to them.

[Demhat]

double post

[vbnetkhio]

I agree with you that he's proficient with Admixtools, and that is what this latest article is based on but I don't agree with the comparison with Davidski simply because among other things unlike Dilawer Davidski doesn;t know how to diploid genotype ancient DNA sequences (infact he relies on others to even convert BAM files) and he isn't able to invent population genetics software on his own which is different from currently available software such as Admixture, Admixtools, etc. because this requires an understanding of biioinformatics on a higher level (Dilawer totally on his own invented SAPDA which is totally different software and basis than other out there. It's available at GenePlaza). Davidski is a good blogger and does understand various aspects of population history pretty well.

Anyways, this thread is not about who is best, but is to discuss the ethnogenesis of Kurds and Iranics.

I also don't think you're being fair regarding Sintashta because Dilawer did seriously consider Sintashta in his modelling of Kurds ( check his 2018 article https://eurasiandna.com/impact-of-the-iron-age-saka-and-scythians-on-south-west-asian-demography/) . Also he is currently working on models for Kurds as : Iran-chl + Levant-N + Turkmenistan-IA + Scythian/Sarmatian.

I think many people misunderstood what he said about Sintashta. He was simply suggesting that to get to present Kurds you would have to go through the Iron Age. In other words Sintashta genes would have to be conveyed to Kurds via Iron-Age proxies such as Sarmatians and Turkmenistan-IA

could you ask Dilawer to convert these fastq files to plink?

https://www.ebi.ac.uk/ena/browser/text-search?query=serbia%20human

(click on Samples (368) and then you'll see the 20 or so Bronze age samples)

[Zoro]

could you ask Dilawer to convert these fastq files to plink?

https://www.ebi.ac.uk/ena/browser/text-search?query=serbia%20human

(click on Samples (368) and then you'll see the 20 or so Bronze age samples)

I asked him about this for someone a while back and this was his response


Genotyping sequences is a very time consuming process and it takes me days to process a few sequences. That's the reason I only do it for my own research when I absolutely have no other choice such as when I need true diploid genotypes of ancient samples. My advise is to contact the authors for VCF or Plink files. Most of the time they will send them to you unless their paper is still in pre-print stage.

[Joachim]

Isnt jt just catacomb yamnaya getting mistaken for scythian Ancestry?