November 20, 2012

U7 in Rostov Scythians

I found it quite interesting that in terms of mtDNA, the Rostov Scythians studied by der Sarkissian resembled closely the Shugnans of Tajikistan, who speak an eastern Iranian language. The author finds links between the Scythians and the "Central Asian Corridor", in particular with respect to mtDNA haplogroup U7.

This "Central Asian Corridor" sensu der Sarkissian (Iraq, Iran, Pakistan, India) seems to touch Frachetti's Inner Asian Mountain Corridor (shown below) in the region of the Pamirs.



Interestingly, the Sughnans belong, anthropologically to the Pamir-Ferghana type, which was also called Central Asian interfluvial type, the rivers in question being the Oxus and Jaxartes (Amu Darya and Syr Darya). And, of course, between these two rivers was the heartland of the Bactria Margiana Archaeological Complex, which I have previously linked with the Indo-Iranians.

Wells et al. studied Y-chromosomes of Sughnans, Yagnobis and other Iranic survivals of Tajikistan more than 10 years ago, and it will be very well worth revisiting them with newer methods. The area east of the Caspian and west of the IAMC intersects so much history, that any data from from it (new or ancient) would be extremely useful.

In my own experiments there has been an unambiguous "South Asian" genetic component in almost all Iranic peoples, even the westernmost Kurds. While the interpretation of this component is not easy, it does point to a genetic relationship between its possessors and Central/South Asia, with notable contrasts between Kurds/Iranians and their non-Iranic Armenian/Anatolian/Caucasian neighbors.

The occurrence of mtDNA haplogroup U7 in the Rostov Scythians is also consistent with a link between the Iranian nomads who penetrated into Europe with the area east of the Caspian, and it is also, of course, consistent with the narrative of Herodotus who recorded the migration of the Scythians into Europe.

There is a widely held theory that the origin of the Indo-Iranians are to be sought in eastern Europe. That theory appears inconsistent both with the "South Asian" autosomal signal in Iranic groups, and with the mtDNA evidence. Consider, again, the evidence of der Sarkissian:


Now, if Rostov Scythians were primarily descended from Mesolithic West Eurasians or even Bronze Age ones, then we would expect them to cluster at the "top", approaching the northern Europeoid extrema of PWC and Bronze Age Altai (ALT-BA). On the contrary, their position is well to the "south" of all European Bronze Age groups, and intermediate between Europeans and Iron Age Asian groups from south Siberia and Kazakhstan (KUR-IA, KAZ-IA). Again, this is compatible with an east-west migration during the Iron Age.

It might be worth speculating on the possible autosomal history of the steppe, for which the mtDNA evidence complements others: I offer that the long-term trend will be one of diminishing "North European", increasing "West Asian" and "East Eurasian" influences across the Neolithic-Bronze-Iron Age boundaries. At the western end of the steppe, there may also be "Mediterranean"/Sardinian-like infusions from the Balkans and Central Europe, although these clearly did not influence Inner/South Asia (where Mediterranean components shrink to non-existence), and Europe proper was mostly the recipient rather than the emitter of populations to Asia. Hopefully, autosomal data to test this conjecture will be made available in the coming years.

17 comments:

wagg said...

"The occurrence of mtDNA haplogroup U7 in the Rostov Scythians is also consistent with a link between the Iranian nomads who penetrated into Europe with the area east of the Caspian, and it is also, of course, consistent with the narrative of Herodotus who recorded the migration of the Scythians into Europe."

We can't really be sure of anything with this U7 because U7 was also found (at least 2 times IIRC) in ancient North European gene pool (Denmark vikings) - and is also found in Population deep in the Urals (the Mansi IIRC), so all this is not completely clear yet, I would say.

"There is a widely held theory that the origin of the Indo-Iranians are to be sought in eastern Europe. That theory appears inconsistent both with the "South Asian" autosomal signal in Iranic groups"

It is widely held for a reason: Archeology allege the source of the Kurgan cultures are found in the Volga region during chalcolithic. The Scythians are also ultimately derived from these earlier groups apparently, whatever hypothetical influences they may have undergone subsequently.

The skeletal remains of Bronze age Kazakhstan are also (maybe in a lesser proportion though I don't remember) related to populations from the Pontic steppes and the Altai ("proto-europoid" in the Russian/Ukrainian terminology), though the more it goes south and the more the time elapse, and the more the population is admixed.

Anonymous said...

All of this would seem to support the "Out-of-Bractria" scenario of PIE origins, or some variation of it.

wagg said...

About my former post, above:

"We can't really be sure of anything with this U7 because U7 was also found (at least 2 times IIRC) in ancient North European gene pool (Denmark vikings)"

Sorry, one of the mtDNA U7 Viking sample I was thinking of was Actually from Norway (Holck 2006) but there was a U7 in aDNA samples from the early Christian period of denmark (http://www.ncbi.nlm.nih.gov/pubmed/15838837, the abstract mentions that U7 is also found among western Siberian tribes, incidentally).

Davidski said...

The U7 can be explained by contacts between the Scythians and civilizations of the south.

That lineage was probably carried by some royal from the south who was either kidnapped by the Scythians or moved to Russia as part of an arranged marriage between elites.

Nirjhar007 said...

''in my own experiments there has been an unambiguous "South Asian" genetic component in almost all Iranic peoples, even the westernmost Kurds. While the interpretation of this component is not easy, it does point to a genetic relationship between its possessors and Central/South Asia, with notable contrasts between Kurds/Iranians and their non-Iranic Armenian/Anatolian/Caucasian neighbors.''
well dienekes the presence of south asian component clearly says what kazanas have pointed in this article-
http://www.omilosmeleton.gr/pdf/en/indology/Vedic_and_Avestan.pdf
and as i gave you in a link before that there is ample linguistic support also for both indo-iranian and P.I.E homeland to be in south central asia ( around sapta sindh)and as ASI ANI mix is 4000 yo old as per to moorjani et al and as the collpse of ivc is environmental-
http://news.yahoo.com/huge-ancient-civilization-collapse-explained-123449804.html
which caused the mix as harappans had to move both east and west
there is no doubt in from where the the pie spread from the land which didnt face any intrusion from 4500 to 600 bc as per archaeology is concerned.
stay well.

valeryz2001 said...

Dienekes, there are many lineages other than U7 supporting the theory of early contacts between East Europeans/Uralic people and ancient Central Asians - those located between Iran and India. Particularly, N2a, N1a, several clades within W, R1a (mt*R1a of course), mt*R2 - those are rare lineages everywhere, so using them as a proof has greater power itself.

valeryz2001 said...

That theory appears inconsistent both with the "South Asian" autosomal signal in Iranic groups, and with the mtDNA evidence.

of course, this "SA" seems relevant in terms of mtdna only if the suggested ancient population was somehow isolated from India and even from Indus Valley: the prevalent direction of maternal gene flow seems to be from South Central Asia to India and not vice versa, otherwise we'd see the variety of Indian lineages in Central Asia while their total frequency is now limited everywhere in Central Asia.

Dienekes said...

of course, this "SA" seems relevant in terms of mtdna only if the suggested ancient population was somehow isolated from India and even from Indus Valley

India was the recipient of the same population element, which, combined with "ASI" in varying proportions, became the modern range of populations. Of course, the question exists how far north and west ASI extended; Reich et al. (2009) estimated ~1/4 of it in Pakistan, and Indian-like mtDNA (M subclades) extends well outside the Indian subcontinent (e.g., in Iran).

There is probably no unadmixed descendant population of "early Indo-Iranians". In the west, they appear to have the added influence of "Levant/Mediterranean"-type elements; in India with ASI -although the ancient limits of ASI are unknown as mentioned above-, and in Central Asia of East Eurasians.

Va_Highlander said...

And, of course, between these two rivers was the heartland of the Bactria Margiana Archaeological Complex, which I have previously linked with the Indo-Iranians.

It may or may not have been the heartland and linking the BMAC with the Indo-Iranians is problematic. All such claims known to me rest upon the work of Viktor Sarianidi and Sarianidi I think fails to prove his case. In fact, I don't think he even comes close to proving his case.

As for the BMAC heartland, I should very much expect to find it south of the Oxus and quite possibly far to the south.

Naren Palepu said...

"All of this would seem to support the "Out-of-Bactria" scenario of PIE origins, or some variation of it."

Bactria, BMAC, Mehergarh, Indus Valley-- How far are they for each other?. Why only BMAC? It's a tiny subset of that area.

The largest civilization spread on the Globe in plains before 5K BC is this area. ( Except China).

Between mountains of BMAC and old plains of Indus-Mehergarh, More population is from Plains.

When weather patterns changed, Populations expanded in all the directions.

eurologist said...

All of this would seem to support the "Out-of-B[r - sic]actria" scenario of PIE origins, or some variation of it.

Ezr,

How, so? I do not see this, at all. I am confused by your statement. At the least, it is late by several millennia.

Rob said...

Your reconstruction is a little simplistic. No on can be surprised that sth Asian lineages are found in Scythians, in light of the "corridoor" arguement. However, that the Scythians migrated EN BLOC as a "nation" from somewhere south to Pontic steppe is quite non-sensical in the light of how current Anglophone anthropologists and archaeologists formulate ethnogenesis theories. The fact that the Scythian groups does not cluster as tight as with European Mesolithic groups mean nothing: they (Euro Meso groups) were not a homogeneous whole.

Dienekes said...

However, that the Scythians migrated EN BLOC as a "nation" from somewhere south to Pontic steppe is quite non-sensical in the light of how current Anglophone anthropologists and archaeologists formulate ethnogenesis theories.

Their position appears to be intermediate between Iron Age groups and Europeans and to coincide with various mixed origin groups such as Shugnans, Tatars, Udmurts, etc. and to be distinct from Bronze Age groups.

The simplest explanation is that they were formed by admixture, which makes sense, because eastern nomads like the Scythians would partially take local (Cimmerian?) wives.

All this will become quite clear, I think when Neolithic, Eneolithic, Bronze, and Iron Age populations from the European steppe are sampled.

wagg said...

Dr Rob: "No on can be surprised that sth Asian lineages are found in Scythians"

Definitely, and I can't see what's so strange in the fact that south Asian autosomal DNA being found in Iranic population either, whatever theory we're thinking of (except one placing the Indo-iranian craddle in west Asia).

Dr Rob: "However, that the Scythians migrated EN BLOC as a "nation" from somewhere south to Pontic steppe is quite non-sensical"

I don't think anyone is proposing this and anyway we KNOW that they were admixed (they are generally described roughly as a mix of 3 major physical types IIRC: proto-europoid (north european, and they were indeed described mentionning that light hair and blue eyes being not rare among them by Greek and Roman authors), east asian and west-asian (I've read a description saying "middle-east-like" morphology, I assume it actually included Afghan-like type (implicationg southAsian/central asian hgs)).


Dienekes: "the question exists how far north and west ASI extended"

I assume it was quite spread at the origin. After all, the Kostenki man (more than 30,000 yrs old) near Voronezh, in Russia, was U2 (whose origin is very likely India) - likely the origin of U2e (found in Europe/north Eurasia), the only non-south Asian subclade of U2 AFAIk - and his skull had affinities to the Australoid populations.
A human group, deep in Russia and coming from paleolithic India, carriers of a mtDNA U (an haplogroup whose general distribution is in west Eurasia, even if U2 itself is centered in India but not particularly in the south AFAIK) having roughly Australian aborigine-like features, clearly we're looking at ASI-like substrate.

GailT said...

I don't think you can reliably place the origin of mtDNA U2 in India. The Kostenki U2 sample has none of the U2e mutations but it does have 4 unique mutations (542, 711, 13269 and 15262) so it appears to be an extinct sister clade to the other U2 subclades. Also, U2d is found mostly in the North Caucasus, the Near East and Europe.

I think a more likely scenario would be an origin of U2 in southwest Asia, followed by migration of U2a, U2b and U2c into India, and U2d, U2e and Kostenski U2 into the Near East, Europe and Russia.

Nirjhar007 said...

''There is probably no unadmixed descendant population of "early Indo-Iranians". In the west, they appear to have the added influence of "Levant/Mediterranean"-type elements; in India with ASI -although the ancient limits of ASI are unknown as mentioned above-''
dienekes use simple logic and accept what one of the fathers of aryan invasion theory max mueller found out 100+ years ago that
“Zoroastrians were a
colony from Northern India...[who] migrated westward to Arachosia and Persia”.
stay well.

George said...

Scythian and "SE-N" N.Ossetian mtdna
http://s019.radikal.ru/i618/1212/40/5c989117704a.jpg
http://digital.library.adelaide.edu.au/dspace/bitstream/2440/74221/1/02whole.pdf
For Iron age + post-Iron age, N.Ossetians are in the same line with Scythians, one other line for example are Saudis Iraqis and Jordanians.
And something else, the "Scythian" one sample "RD-2" C mtdna possibly isn't Scythian becouse it is in Budini-Scythian borders, and the C-mtdna line in that picture is pulling the Scythian samples to it self, and without that Scythians are in the same place with N.Ossetians.

The RD-2 is from Bokovsky district Razmetniy(Budini-Scythian borders)
http://s017.radikal.ru/i412/1211/84/701ae62aae8c.png
Herodotus wrote the northern part of the Tanais(Don) river is in the teritory of Budini people