search this blog

Thursday, June 2, 2016

Ancient mtDNA from medieval Hungary


There's a new preprint at bioRxiv looking at the maternal "ancestry and legacy" of early Hungarians and Avars in Europe. The problem is that the Hungarian and Avar expansions were in all likelihood male dominated events. Also, the data doesn't include full mito genomes. So what the authors are really testing are the broad maternal ancestry and affinities of the early Hungarians and Avars. Still very interesting, but maybe somewhat oversold.

Abstract: The ancient Hungarians originated from the Ural region in today's central Russia and migrated across the Eastern European steppe, according to historical sources. The Hungarians conquered the Carpathian Basin 895-907 AD, and admixed with the indigenous communities. Here we present mitochondrial DNA results from three datasets: one from the Avar period (7-9th centuries) of the Carpathian Basin (n = 31); an almost four-fold enlarged dataset from the Hungarian conquest-period (n=101); and one from the contemporaneous Hungarian-Slavic contact zone (n = 23). We compare these mitochondrial DNA hypervariable segment sequences and haplogroup results with other ancient and modern Eurasian data. Whereas the analyzed Avars represents a certain group of the Avar society that shows East and South European genetic characteristics, the Hungarian conquerors' maternal gene pool is a mixture of West Eurasian and Central and North Eurasian elements. Comprehensively analyzing the results, both the linguistically recorded Finno-Ugric roots and historically documented Turkic and Central Asian influxes had possible genetic imprints in the conquerors' genetic composition. Our data allows a complex series of historic and population genetic events before the formation of the medieval population of the Carpathian Basin, and the maternal genetic continuity between 10-12th centuries and modern Hungarians.

Csosz et al., Maternal Genetic Ancestry and Legacy of 10th Century AD Hungarians, bioRxiv, posted June 2, 2016, doi: http://dx.doi.org/10.1101/056655

22 comments:

Huck Finn said...

"Still very interesting, but maybe somewhat oversold."
Why? This is a good study and the results are very sensible.

Kristiina said...

I think that this is important:
"The conqueror population has a similar haplogroup composition to present-day Central Asians and Finno-Ugric populations, which is also supported by Ward type clustering. While Avars showed rather European connections, the contact zone population had a Near Eastern type haplogroup composition."

"Prehistoric Central Asian, south central Siberian (Minusinsk Hollow) and Baraba populations were not significantly different from the populations of the Carpathian Basin, and these affinities are also reflected on the clustering tree."

By the way, I cannot find the supplementary tables. Are they available anywhere or are they coming later on?

Kristiina said...

This is also interesting:
"We selected 23 modern populations from the GDM, MDS, and PCA datasets, which possibly had increased lineage-sharing with the conquerors and we compared them using a modern SHA (Supplementary Table S12). Populations speaking Uralic languages are not well studied for mtDNA, therefore we could only use Khantys, Mansis, Nenets, and Komis as references for Uralic peoples. The ancient conquest-period population had the highest lineage sharing with the Tatars in Russian Tatarstan, Nenets, and Komi groups (42–36%). They are followed by Hungarians, Russians in Bashkortostan, and three populations of almost identical percentages; Ukrainians, Khanty and Mansi population, and Szeklers. When counting lineages, rather than the number of sequences, Csangos, Khantys and Mansis, and the population of the Russian Bashkortostan Republic were the third, fourth and fifth populations with the highest lineage-sharing".

It is also worth to be noted that Khantys and Mansis are the closest linguistic relatives of the Hungarians and they do show mtDNA similarities with ancient Hungarians.

Rob said...

i think supp data isn't made avail until formal publication ?

It is interesting that the "border zone" groups are still near eastern, which means that until Slavs came (more or less contemporary with the avars and the final collapse of their pax), the Balkans was still very near eastern

It's also interesting to note that the avars are so "European", what would throw away the idea that their JuanJuan decsendants from Mongolia. Of course, y DNA and autosomes are needed to confirm this

Onur said...

For the sampled Avars they write these:

"The Avar dataset originates from a regional group of the Avar society, who buried their dead in catacomb graves (26). Furthermore, anthropological results showed that this part of the Avar population represents mostly Europid, local morphological characters, and therefore it cannot be used as a proxy of the whole Avar population of the Carpathian Basin."

"One further point to note is that the Avar samples belong to a micro region of the Avar Khaganate, and therefore they do not represent the whole Avar population of the Carpathian Basin."

Rob said...

Onur

Yes, quite right

Shaikorth said...

The supplementary figures are released and the tables are listed in the supplementar pdf. Missing tables is most likely a mistake, if it is an omission it's a baffling one.

Not using autosomal DNA or even complete MtDNA sequences is naturally a letdown in any ancient DNA studies these days.

Onur said...

Not using autosomal DNA or even complete MtDNA sequences is naturally a letdown in any ancient DNA studies these days.

Exactly.

Karl_K said...

@Onur

Exactly.

epoch2013 said...

@Rob

"It's also interesting to note that the avars are so "European", what would throw away the idea that their JuanJuan decsendants from Mongolia. Of course, y DNA and autosomes are needed to confirm this"

Authors of those days distinguished between "real Avars", apparently from the Turkic central Asian stock and "Pseudo-Avars". Ethnonyms are a mess.

https://en.wikipedia.org/wiki/Pannonian_Avars

Samuel Andrews said...

@Rob,
"It is interesting that the "border zone" groups are still near eastern, which means that until Slavs came (more or less contemporary with the avars and the final collapse of their pax), the Balkans was still very near eastern "

Still Near Eastern? When were the Balkans ever Near Eastern? If anything the Balkans went from WHG, to mostly EEF, to largely Steppe, to largely near Eastern. near Eastern admixture came last.

Ariele Iacopo Maggi said...

Samuel Andrews

Farmers came from the near east...

Samuel Andrews said...

The Near east iim referring to is the one which behaves on PCA, ADMIXTURE, and D-stats like modern Near Easterners. Swedes might have more EEF than Greeks and most in the Balkans, and the Balkans began to lose its EEFness by at least 3000 BC IMO. So by the time of Slavic expansion it wasn't very EEF.

batman said...

@ Ariele

So they say - but did they really?

What prooof actually exists that substain this old hypothesis?

Rob said...

off topic

They've got a genome from Iran Neolithic (female). Basically CHG like

http://smbe-2016.p.asnevents.com.au/days/2016-07-07/abstract/35146

Abstract only, atm

Rob said...

@ EPoch 2013


"Authors of those days distinguished between "real Avars", apparently from the Turkic central Asian stock and "Pseudo-Avars". Ethnonyms are a mess."

No no, Don;t read Wikipedia ;)
The Avars are Avars. There were no other "Avars" in the world.
The 'Pseduo-Avar' names was invented by the GokTurks to diminish the pride of the Avars, with whom they were enemies. So by calling them Psedo-Avars to the Byzantines, they wished to suggest that they're frauds, phoney's and that the Byzantines should abandon the treaty they just signed with them.

Rob said...

@ Samuel Andrews

* "Still Near Eastern? When were the Balkans ever Near Eastern? If anything the Balkans went from WHG, to mostly EEF....."


-> As Ariele pointed out, Near Eastern as in the sense of Early Farmers (because farmers came via Anatolia, which is in the "Near East"), not necessarily modern Near Eastern people like Syrians, etc

So I wonder what the authors meant by the 'margin zone' groups (presumably near old Roman border) being close to near eastern ?

* "If anything the Balkans went from WHG, to mostly EEF to largely Steppe, to largely near Eastern. near Eastern admixture came last."

-> ? largely steppe ? When ?

We've not yet seen any Balkan Bronze Age samples. The Hungarian late BA (immediately north of the Balkans) sample showed some steppe admixture, but considerably less than CWC or BB, and was still largely EEF. Of course, samples from actually Kurgan graves in Bulgaria or Serbia might very / entirely steppic, but the actual *Balkan* Bronze Age cultures like Ezero, Sitagroi, Vucedol will imaginably be like the BA _Hungary, but even more EEF.

* "near Eastern admixture came last"

Not sure, without aDNA. If by "Near Eastern' here you are referring to the "CHG" aspect, then my guess will be sometime between 4000 & 2000 BC. Otherwise, there is no near eastern migration to Balkans after that.

Samuel Andrews said...

@Rob,

"The Hungarian late BA (immediately north of the Balkans) sample showed some steppe admixture, but considerably less than CWC or BB, and was still largely EEF"

Most models I've seen puts BA Hungary at 30% Yamnaya. I'm very confident that when Slavs arrived the Balkans weren't very EEF.

"Not sure, without aDNA. If by "Near Eastern' here you are referring to the "CHG" aspect, then my guess will be sometime between 4000 & 2000 BC. Otherwise, there is no near eastern migration to Balkans after that."

By "Near Eastern" I mean something like Syrians or Palestinians or Cypriots. I'm confident it's an important part of the ancestry of Balkanerers. It could be the source of their Y DNA J2 and E1b-V13.

Rob said...

@ Sam


"Most models I've seen puts BA Hungary at 30% Yamnaya. I'm very confident that when Slavs arrived the Balkans weren't very EEF. "

That doesn't make much sense to me, because modern Balkan groups still are high in EEF (30%); not as high as Italians & Spaniards, but still higher than Poles or Swedes.

Now, considering the Slavs came from western Ukraine/ southern Poland / northeastern Romania, you'd expect them to have similar EEF levels to Poles & Ukrainians, ie not very high. So if pre-Slavic Balkans had low EEF, and the Slavs had even lower EEF, then where'd the extra 20% come from ?

Obviously, there's a fatal flaw in that scenario.
But what does make sense is that Bronze Age Balkans remained low in steppe admixture, and high in EEF, with some additional CHG during the Early Helladic/ Bronze Age - with arriving J2 people. Moreoever, the amount of steppe admixture in BA _ Hungary is probably closer to 15% than 30%. This would drop to 10% in Balkans propper, all in line with what ive stated.


"By "Near Eastern" I mean something like Syrians or Palestinians or Cypriots. I'm confident it's an important part of the ancestry of Balkanerers. It could be the source of their Y DNA J2 and E1b-V13."

I doubt that Syrians are the source of J2 and E-V13, because they & their neighbours don't have any of the said haplogroups. Instead, Syrians, Palestinians are mostly J1, not J2, and certainly not the J2b fond in the Balkans. The J2 in the Balkans is similar to that found in Italy & southern Spain, which appear to form a group of their own which split off a few thousand years ago, and is somewhat more distantly related to that found in Modern Iranians. (But i think J2 research is lagging a little in population studies so its not too clear).

And the E - V13 common there now appears to have arrived in the Neolithic, It has been found in Neolithic SPain & Hungary, and is a distinct haplogroup to the E now found in modern Syrians, etc. (I think this was explained to you 3 years ago ?)

epoch2013 said...

@Rob

I'm pretty sure Paul the Deacon mentioned something similar. But then again, even if the Lombards betrayed the Gepids in their fight against the Avars, that didn't mean Lombards and Avars were exactly friends so that could be a slur too. Paul also describes Avars as exceptionally cruel.

Mind you, it's an interesting question: How much of what we know from literature is actually politically motivated slander? Atilla's fatal nose bleed? The heathen Heruli being gay is no doubt a slur, considering that other, older and non-Christian authors said Germans rejected homosuxuality.

Rob said...

@ Epoch

I cant recall PtD talking about the Avars, althugh he does mention Bulgars

I don't think any investigator has ever demonstrated convincingly that there were any other Avars elsewhere , have they ? Some have like to link to Ruruan & Hephthalites, but no source ever called them "Avars".

"How much of what we know from literature is actually politically motivated slander? Atilla's fatal nose bleed? etc "

Most of it, Id say. Even those posing as 'ethnographers' like Tacitus definitely had political undercurrents. It was inseparable. Even the professional anthropologists of today are swayed by current geo-political and academic flavours.

epoch2013 said...

@Rob

I skimmed to it. Paul does mention Avars, quite a lot of times, especially te accouns of Grimoald. But indeed nothing on the Pseudo-Avars. My bad.