PDA

View Full Version : Diverse genetic origins of medieval steppe nomad conquerors



Afshar
12-18-2019, 08:12 AM
Considering we dont have any Oghuz ancients yet, these samples might also come in handy because they are more western than the samples we have now.
Lets hope the data will be good quality.

Diverse genetic origins of medieval steppe nomad conquerors

Alexander Mikheyev, Lijun Qiu, Alexei Zarubin, Nikita Moshkov, Yuri Orlov, Duane Chartier, Tatiana Faleeva, Igor Kornienko, Vladimir Klyuchnikov, Elena Batieva, Tatiana V Tatarinova

doi: https://doi.org/10.1101/2019.12.15.876912

Abstract

Over millennia, steppe nomadic tribes raided and sometimes overran settled Eurasian civilizations. Most polities formed by steppe nomads were ephemeral, making it difficult to ascertain their genetic roots or what present-day populations, if any, have descended from them. Exceptionally, the Khazar Khaganate controlled the trade artery between the Black and Caspian Seas in VIII-IX centuries, acting as one of the major conduits between East and West. However, the genetic identity of the ruling elite within the polyglot and polyethnic Khaganate has been a much-debated mystery; a controversial hypothesis posits that post-conversion to Judaism the Khazars gave rise to modern Ashkenazim. We analyzed whole-genome sequences of eight men and one woman buried within the distinctive kurgans of the Khazar upper (warrior) class. After comparing them with reference panels of present-day Eurasian and Iron Age populations, we found that the Khazar political organization relied on a polyethnic elite. It was predominantly descended from Central Asian tribes but incorporated genetic admixture from populations conquered by Khazars. Thus, the Khazar ruling class was likely relatively small and able to maintain a genetic identity distinct from their subjugated populations over the course of centuries. Yet, men of mixed ancestry could also rise into the warrior class, possibly providing troop numbers necessary to maintain control of their large territory. However, when the Khaganate collapsed it left few persistent genetic traces in Europe. Our data confirm the Turkic roots of the Khazars, but also highlight their ethnic diversity and some integration of conquered populations.[/QUOTE]

Onur Dincer
12-19-2019, 08:29 PM
What is then the "Anatolia_Ottoman:MA2195" individual, who is almost half East Eurasian and half West Eurasian in genetics and not radiocarbon-dated and so could well be from Seljuk Anatolia rather than Ottoman Anatolia considering her apparent lack of native Anatolian admixture, other than Oghuz/Turcoman? But I concede that we need more Oghuz/Turcoman ancient genome results from the relevant time periods, preferably both from Anatolia and from the Oghuz Yabghu State lands in what is now Kazakhstan, to have a good idea about Oghuz/Turcoman genetics.

As for the Khazar ancient DNA paper, genetically one of the individuals is almost totally East Eurasian, another one is 80% East Eurasian, yet another one is 65% East Eurasian according to the autosomal results. I did not expect to see individuals with so high East Eurasian among Khazars considering the western geographic location of the Khazar lands. I wonder how non-elite Khazars were genetically. At least one of the individuals almost certainly lacks any Khazar ancestry considering the lack of East Eurasian ancestry in that individual in contrast with the other individuals.

Onur Dincer
12-23-2019, 04:49 PM
Please delete this.

Onur Dincer
12-23-2019, 04:58 PM
Mountain, you are doing the mistake of representing the original Oghuz/Turcoman ancestry using only one individual. There are already tested medieval Turkic individuals quite similar in genetics to Anatolia_Ottoman:MA2195 and can thus be used in addition to her to represent the genetics of the Oghuz/Turcomans coming to Anatolia and environs. Here below one analysis I have made:

35577

As you can see, all those Anatolian Turkish samples have perfectly fitting levels of distance except maybe the Adana Turkish sample because of the existence of some Arab individuals in that sample. There is no overfit.

Now, when we add Azeris to test the robustness of the analysis, here is what we find:

35578

Azeris have too high distances as they have high levels of Iranian ancestry (normal given their geography) and require Iranian populations for fitting distances.

For a visual representation of the situation, see:

35579

In short, even if Anatolian Turks have some Iranian admixture, it is not as high as in your overfitting models.

Dorkymon
12-23-2019, 06:30 PM
Mountain, you are doing the mistake of representing the original Oghuz/Turcoman ancestry using only one individual. There are already tested medieval Turkic individuals quite similar in genetics to Anatolia_Ottoman:MA2195 and can thus be used in addition to her to represent the genetics of the Oghuz/Turcomans coming to Anatolia and environs. Here below one analysis I have made:

35577

As you can see, all those Anatolian Turkish samples have perfectly fitting levels of distance except maybe the Adana Turkish sample because of the existence of some Arab individuals in that sample. There is no overfit.

Now, when we add Azeris to test the robustness of the analysis, here is what we find:

35578

Azeris have too high distances as they have high levels of Iranian ancestry (normal given their geography) and require Iranian populations for fitting distances.

For a visual representation of the situation, see:

35579

In short, even if Anatolian Turks have some Iranian admixture, it is not as high as in your overfitting models.

Any reason why you haven't used Kaz_Turk and Kaz_Kipchak? Or are they there in the model but simply not picked up?

Onur Dincer
12-23-2019, 07:28 PM
Any reason why you haven't used Kaz_Turk and Kaz_Kipchak? Or are they there in the model but simply not picked up?

Because they are not genetically close to the Anatolia_Ottoman:MA2195 individual as much as the Turkic samples I used. KAZ_Turk is especially distant, it is from the earlier Gokturk era hence the name.

Onur Dincer
12-23-2019, 07:35 PM
You are right I made the mistake of not looking at the other medieval individuals. I will try to look at them later when I have more time.

But I'm still not convinced because with Azeri's you can create fits low like 0.0167 by only using medieval Turkic individuals, Armenians, and Assyrians without reflecting their Iranian ancestry, which is more confusing now.

Target Distance Armenian Assyrian KAZ_Karluk KAZ_Kimak
Azeri 0.01665795 43.6 39.4 10.0 7.0

Azeris have different native ancestries depending on the region, not just Iranian. Just check the genetic differences between the Azeri sample and the Azeri_Dagestan sample and you will see.

Onur Dincer
12-25-2019, 07:07 PM
Mountain, I do not know why you use a very mixed population such as modern Turkmens to examine the genetics of the Oghuz/Turcoman nomads coming to Anatolia and environs. Modern Turkmens are not a simple mix of the Oghuz/Turcoman nomads coming to what is now Turkmenistan and environs and the Iranian natives of those lands, they also have more recent admixture from the Iranian populations to their south and Kazakhs and Karakalpaks to their north due to their raiding activities as nomads and slave trade. They are surely not a remnant population such as Central Anatolian Greeks or many island Greeks and have little to offer if we want to examine the genetic picture of 1000 years ago.

I would not say the medieval Karluk, Karakhanid, Kimak and Kipchak samples are too eastern-shifted in their West Eurasian composition compared to Anatolia_Ottoman_MA2195, that would be an overstatement. Their eastern shift compared to Anatolia_Ottoman_MA2195 is not so high overall and, more importantly, it requires more ancient DNA testing to have a good idea about the West Eurasian composition of Oghuz/Turcomans. For Anatolia_Ottoman_MA2195 herself we do not even have radiocarbon dating or isotope data and we do not know if she has Anatolian admixture or not.

Nowhere have I stated that modern Anatolian Turks do not have any Iranian admixture from the last 1000 years, but I state that your Iranian admixture estimations for Anatolian Turks are very likely overestimations and point to the reasons why.

Onur Dincer
12-26-2019, 03:43 AM
On Turkmens' raiding activities and their role in slavery and slave trade, see for instance:

The reason why the Turkomans are more inhuman than
the other barbarous wandering tribes, cannot be attributed, I
believe, to any other cause, than that they are slave-dealers.
Their daring forays are usually undertaken with a view to
carry away captives, whom they retain in chains, until re-
deemed by their relations, or sell them in the bazars of Khiva,
if the ransom money fails to arrive in time, or proves insuffi-
cient to satisfy their cupidity. The prisoners are sometimes
retained for their own use, and sent to tend their flocks in the
desert, or employed in field works. Thus they are the terror
of their neighbours — the Persians of Mazanderan, Asterabad,
and Khorassan, who are obliged to be always on their guard
against the sudden attacks of the Turkomans.

As the Persians are of the Shia sect, and the Turkomans
of the Sunni, the latter justify themselves on the ground that
to seize on a Persian and sell him is lawful ; others, however,
are more sincere, and own that if the Persians, instead of
Shia, had been Sunni, then they themselves must have turned
Shia, — as the circumstance of being of the same religion might
have interfered perhaps with their present lucrative trade.
These religious scruples do not prevent them, however, from
capturing persons of their own religious persuasion and their
own tribe, with whom they happen to be in enmity, and fixing
enormous prices for their release.

The Turkomans observe a difference between their children
from Turkoman mothers, and those from the Persian female
captives whom they take as wives, and the Kazakh women
whom they purchase from the Uzbeks of Khiva. The Tur-
komans of pure race enjoy full privileges, while the others are
not allowed to contract marriages with Turkoman women of
pure blood, but must choose themselves wives among the
half-castes and Kazakh captives.

As there exists a great animosity between the Yamuds and
Goklans they do not intermarry, although they reckon them-
selves of equally noble lineage. The same hatred is extended
to the Tekke Turkomans, whom the Goklans and Yamuds,
moreover, look upon as their inferiors, being, according to their
genealogies, the descendants of a slave-woman, whilst they are
the posterity of a free-woman.

The more intimate connection of the Astrakhan and Kazan
Tartars with the Mogols can be traced in their features ; with
the Nogay it is less visible. In like manner, the Turkomans
further off in the desert, and the Uzbeks of Khive, have more
of the Mogol expression than the Turkomans who encamp
near the Persian frontier. The frequent intercourse of the
Nogay, in latter years, with the Cherkess, seems to have im-
proved their race ; and notwithstanding the enmity that exists
between the Turkomans and the Persians, it is still not un-
likely that their close vicinity should have produced on the
former a similar effect in a lapse of several centuries. The
fact we have seen, that the Turkomans marry Persian women,
when they take them as prisoners.

From "On the Yamud and Goklan Tribes of Turkomania" by Baron Clement Augustus de Bode (1844)

Onur Dincer
12-26-2019, 10:52 PM
Is there any phenomenon where averages consisting of more samples produce better fits? could that be reason why some populations produce better fits?

If we use source populations that are known or likely to have contributed to a target population or source populations genetically very close to such source populations in our model, usually the more individuals used for the source populations the better the fits simply because the genetic averages of the source populations come closer to the true values with more sampling. This is just simple statistics.

Coming back to Anatolia_Ottoman_MA2195, the safest way to learn if she has native Anatolian admixture or not is to do ancient DNA tests on Oghuz/Turcomans from the lands of the Oghuz Yabgu State in what is now Kazakhstan. If Oghuz/Turcomans from those lands consistently show less CHG and Anatolia Neolithic leanings and more Iran Neolithic leaning and also more East Eurasian contribution in their genetics than Anatolia_Ottoman_MA2195, that would be a strong indication that Anatolia_Ottoman_MA2195 has some, even if minor, Byzantine Greek ancestry from Anatolia. So far we have no Oghuz/Turcoman ancient genome results from the Oghuz Yabgu State lands and this deficiency is disappointing given that we already have ancient genome results from Karluks, Karakhanids, Kipchaks, Kimaks and Western Gokturks, all from the territory of what is now Kazakhstan. There should have been more efforts to extract DNA from Oghuz/Turcoman skeletons by the researchers working in Kazakhstan due to their higher importance for the later history.