search this blog

Thursday, April 25, 2019

Some myths die hard


Ancient DNA tells us that the Bronze Age wasn't kind to the indigenous populations of Central Asia. It seems to have wiped them out totally. Indeed, Central Asia might well be the only major world region in which native hunter-gatherers failed to make a perceptible impact on the genetics of any extant populations.

Before the Neolithic transition, much of Central Asia was home to hunter-gatherers closely related to those of nearby western Siberia. During the Neolithic, agriculturalists and pastoralists from the Near East gradually moved into the more arable parts of southern and eastern Central Asia, eventually giving rise to the Bactria Margiana Archaeological Complex, or BMAC, and other similar communities.

It's not clear what their relationship was like with the native hunter-gatherers in these areas. But they did mix with them in varying degrees. This is obvious because genome-wide genetic ancestry characteristic of the Botai people, who hunted and eventually domesticated horses on the Kazakh steppe during the 4th millennium BCE, and were probably the archetypal Central Asians for their time, is found at significant levels in a number of later samples from Central Asian farmer and pastoralist sites, such as Dali, Gonur Tepe and Sarazm.

Thus, even though the Neolithic transition did have a big impact on Central Asia, and clearly led to large scale population replacements in some parts of the region, this was just the beginning of these population shifts. Moreover, in some cases the expanding farmer and pastoralist populations seem to have acquired significant indigenous Central Asian ancestry and spread it with them.

The precise geographic extent of the relatively unique Botai-related ancestry in prehistoric Eurasia is still something of a mystery. But to give you a general picture of where it was found from around 6,000 BCE to 2,000 BCE, here's a map with info about samples with significant levels of this type of ancestry from a wide range of sites in space and time.


Going by this map, I'd say it's safe to infer that the Botai-related ancestry was a major feature of practically all forager populations living between the Caspian Sea and the Altai Mountains. It was also present in the Early Bronze Age (EBA) pastoralist population associated with the Steppe Maykop archeological culture of Eastern Europe, so it may have already been in Europe as early as 3,800 BCE, because that's when the Steppe Maykop culture first appeared.

It's an interesting question where the ancestors of the Steppe Maykop herders came from. I once simply assumed that they were closely related to the Maykop people who lived in the Caucasus Mountains. But it's now clear that the populations associated with these two similar cultures were starkly different, with the Maykop people being basically of Near Eastern origin and lacking any discernible Botai-like ancestry. My guess for now is that the Steppe Maykop herders were in large part the descendants of the Kelteminar culture population from just east of the Caspian Sea, but we'll see about that when more ancient DNA comes in.

The other great mystery is what eventually happened to the Steppe Maykop people. Around 3,000 BCE, their culture vanished from the archeological record and their particular genetic signature disappeared from the steppe ancient DNA record. Where did they go? Did they migrate back east?

I don't know, but at about that time other Eastern European steppe herders, those associated with the Yamnaya and Corded Ware archeological cultures, began to stir and migrate in big numbers in basically all directions, including into Steppe Maykop territory. Indeed, unlike the Steppe Maykop population, these groups weren't closely related to any contemporaneous or earlier Central Asians. But they ended up moving into Central Asia, and in a big way too.

Their impact all the way from the Ural Mountains to what are now China and India was profound. For instance, not only did they end up totally replacing the Botai people, but also their horses. For more details on this topic check out the Youtube clip here. I have a strong suspicion that the same sort of thing happened to the aforementioned Steppe Maykop people. In other words, they may have been forced out from the Eastern European steppe, and perhaps sought shelter in the Caucasus Mountains?

Admittedly, I'm not offering anything new here. I just wanted to emphasize a few key points, because I'm still seeing some confusion online about the population history of Central Asia, and especially how it relates to the population history of Europe, and also the Proto-Indo-European homeland question. Make no mistake, thanks to the ancient DNA already available from Central Asia, we can confidently infer the following:

- the chance that the ancient European populations associated with the Yamnaya, Corded Ware and other closely related archeological cultures formed as a result of migrations from Central Asia is zero

- the chance that the Proto-Indo-European homeland was located in Central Asia is zero

- the chance that present-day Europeans, by and large, derive from any ancient Central Asian populations is zero

See also...

Central Asia as the PIE urheimat? Forget it

The Steppe Maykop enigma

Late PIE ground zero now obvious; location of PIE homeland still uncertain, but...

201 comments:

1 – 200 of 201   Newer›   Newest»
Synome said...

Agree with the conclusions here.

I think the best candidate for a population that retains some Botai/Kelteminar-related Neolithic central Asian ancestry is the Burusho community. I'm not closely familiar with analysis of their autosomal components, but from what I've seen it might be plausible to model them with this type of ancestry.

If anyone's tried I'd be happy to see the results.

Grey said...

you can see why the more sensitive souls prefer the idea of pots not people - the reality of human competition is pretty grim.

Davidski said...

@Synome

Burusho
IRN_Shahr_I_Sokhta_BA3 0.710±0.029
KAZ_Botai 0.048±0.024
RUS_Sintashta_MLBA 0.242±0.028

chisq 14.509
tail prob 0.206105
Full output

Burusho
IRN_Shahr_I_Sokhta_BA3 0.718±0.028
RUS_Sintashta_MLBA 0.243±0.028
RUS_West_Siberia_N 0.039±0.023

chisq 14.182
tail prob 0.223101
Full output

Kristiina said...

David, would you mind assessing the Botai ancestry in Uralics? Botai ancestry could be found in particular in Ugric-speaking Mansis and Khanties and Samoyedic speaking Nenets and Selkup, as well as in Udmurts in Volga-Kama. Moreover, I have seen modelings where Bolshoy Oleni Ostrov people are c. 10% Botai.

Leron said...

I believe these Botai-like people were responsible for the spread of IE, Uralic, Altaic and Dravidian into their respective regions. Not as speakers of these languages but as wedge in between them that pushed them out but eventually it collapsed on them. I see them as a proto version of the Huns, with Q males eagerly kidnapping foreign wives and growing large clans. Eventually the people who surrounded them got fed up and took over the structure away from them. The Xiongnu being one of their last significant descendants and the Ket people as all that remains.

Davidski said...

@Leron

Almost every time that I happen to read a comment of yours I get very angry, so angry in fact that I want to punch something.

We need to sort this out in some way, because it can't be a healthy thing. The ball's in your court.

music lover said...

This post is factually incorrect on many fronts:
1) The Botai and Dali populations are almost entirely ANE, much more so than the Yamnaya, and the ANE are not native to Europe, but is a population native to an area that ranges from the central steppe all the way to the forest zone of central asia. The Yamnaya, and later Corded Ware cultures are largely ANE derived, as are the modern populations of Central Asia. So unlike in Europe, where large proportions of WHG ancestry have been almost completely wiped out, Central Asians continue to have the majority of their ancestry from a source that has been present there since the Upper Paleolithic
2) Most of the populations of Europe are descended from the ANE via the Yamnaya, and therefore derive a huge portion of their ancestry to central asia
3) Proto Indo European didn't suddenly appear on a map handed to the Yamnaya by aliens, and likely derive from the EHG ancestry which is almost exclusively ANE derived. So if you follow the usual chain of logic, IE languages are native to Asia and not to Europe.
4) The definition of "Europe" as being part of regions East of the Don cannot be determined by anyone. Neither can Asia being a region west of the Don. These boundaries are fuzzy. But it is near certain that present day Kazakhstan and Ukraine can be described as Central Asia and Europe respectively.

Leron said...

Davidski: I'm just trying to think out loud for a bit and see if the linguistic research fits with the genetic data. There's these languages certain linguists are able to pair up across north Eurasia, but Yeniseian stands apart from them. Doesn't this Botai-related ancestry, once prevalent but later almost completely replaced, fit the situation of Yeniseian?

Xiongnu language and Yeniseian
https://www.researchgate.net/publication/290544045_Huns_and_Xiongnu_Identified_by_Hungarian_and_Yeniseian_Shared_Etymologies

Euroasiatic languages (including IE and Altaic)
https://www.pnas.org/content/early/2013/05/01/1218726110.abstract

Samuel Andrews said...

@music lover,

"So unlike in Europe, where large proportions of WHG ancestry have been almost completely wiped out, Central Asians continue to have the majority of their ancestry from a source that has been present there since the Upper Paleolithic"

Europeans do have a lot of WHG ancestry. Assuming Yamnaya was 20% WHG, then modern northern Europeans range from 25-35% WHG. WHG from late hunter gatherers ranges from 15-25% in northern Europe. Also, their Anatolian & CHG ancestors have common ancestry with UHG.

Central Asians do have a lot of ANE ancestry. But No central Asian shave signifcant ancestry from late hunter gatherers who lived in Central Asia. They descend from new groups who arrived in the last 4,000 years, from Europe (Andronovo), Afghanistan (BMAC), and eastern Asia (Mongolia area?). There is contribution from late hunter gatherers of Central Asia but it is low.

E. Donovan said...

The Nganassans appeared to prefer Botai and West Siberia HG/N versus other ANE sources when I ran them in nMonte months ago.

Samuel Andrews said...

@music lover,
"Proto Indo European didn't suddenly appear on a map handed to the Yamnaya by aliens, and likely derive from the EHG ancestry which is almost exclusively ANE derived. So if you follow the usual chain of logic, IE languages are native to Asia and not to Europe."

Yamnaya's main ancestor was ANE, but, their ANE ancestors arrived in Europe 10,000 years before Yamnaya culture emerged. So, Yamnaya had no recent central Asian ancestry. Indo European language formed 5,000-6,000 years ago somewhere in southern Russia/Ukraine which is squarely in Europe. Afanasievo, Andronovo introduced Indo European speech into central Asia.

Bob Floy said...

@Music lover

"Most of the populations of Europe are descended from the ANE via the Yamnaya, and therefore derive a huge portion of their ancestry to central asia"

About 15% on average is "huge"? Europe's ANE on the whole is about the same as it's WHG, which according to you was "wiped out".

Hugh Capet said...

This is an excellent post! I had one inquiry and that is, why do you think that the Steppe pastoralists failed to exterminate the people of the BMAC culture, and instead absorbed them into their genepool? Was this due to the fact that they made up a large segment of the population in southern and eastern Central Asia? I know that the steppe pastoralists replaced much of the BMAC population, but not all of it. Hence, the presence of BMAC ancestry in the early Iranian from Turkmenistan and the populations of Central Asia today.

Matt said...

Off topic, but on the off chance Ryu or anyone else interested in Sino-Tibetan is browsing and didn't already see - https://www.nature.com/articles/s41586-019-1153-z - "Phylogenetic evidence for Sino-Tibetan origin in northern China in the Late Neolithic" - "Consistent with the northern-origin hypothesis, our Bayesian phylogenetic analysis of 109 languages with 949 lexical root-meanings produced an estimated time depth for the divergence of Sino-Tibetan languages of approximately 4,200–7,800 years BP, with an average value of approximately 5,900 years BP."

Having a look at the study, I will say it's odd that they don't use multiple varieties of written Chinese, and either of Old Tibetan or Old Burmese to calibrate their dating. Instead they use constraints on their model topology they believe are implied by these dates " Several time calibrations of known historical events were used to estimate the divergence time for Sino-Tibetan languages (Supplementary Table 2)". Seems a bit arguable - if their constraints push differentiation too low, then they'll get the wrong divergence date.

The claim is also that the Southwestern origin theory of Sino-Tibetan is inconsistent with the date ranges (e.g. that theory presupposes at least a time depth of 9kya), but I'm not sure it's actually the case that a 4-6kya Southwestern origin is out of the running.

E. Donovan said...

One can certainly imagine (Archaic) Proto-Chinese developing within the Yangshao culture on the Yellow River.

Aniasi said...

I'd be interested in seeing if greater sampling and higher resolutions might show some Botai descent in isolated tribal populations around the Pamirs or Mountain South Asia.

Samuel Andrews said...

That would be interesting, but I think we've learned isolated populations like that don't exist. Pretty much, all populations are connected to many other populations in a huge region. Isolates like that don't really exist. All Pamirs should be madeup of the same elements.

Aniasi said...

I disagree. Large areas are unsampled, and we lack ancient DNA in Asia from before the Bronze age.

The populations we have for South Asian samples are overwhelmingly from the plains, and the Pamirs tie into the Himalayas.

Caste endogamy means many silos, and we lack high resolution sampling on many of them.

EurDNA said...

Indeed an interevting phenomena, seem to have been important in early pastoralism, even made it West, however began to decline after 3000 BC, & certainly after 2000 BC, undoubtedly due to Andronovo groups pushing west.
Old news, but probably a fair enough intro in anticipation of new version of Narasimhan.

@ Hugh
'' why do you think that the Steppe pastoralists failed to exterminate the people of the BMAC culture''

Interesting choice of words. How are you around small animals ?

Aniasi said...

I meant South Asia

Davidski said...

@music lover

You're totally crap at this. Go and listen to some music.

Davidski said...

@Hugh Capet

I think numbers played a big part why the steppe people didn't replace the farmers of southern and eastern Central Asia. The ratio of farmers to steppe people was too big for this to be demographically possible.

Also, I suspect another factor was that, perhaps just like in the Swat Valley, the steppe people felt out of place outside of the steppe and weren't in a position to totally dominate and push out these populations.

Davidski said...

@Aniasi

Are you saying that more thorough sampling of present-day populations in southern Central Asia might reveal groups with significant Botai-related ancestry?

I doubt it.

Davidski said...

@Kristiina

I'm not able to model Bolshoy_Oleni_Ostrov with Botai ancestry when using my standard set of outgroups. I could try changing the outgroups significantly to try and make things work, but then I'd just be rigging the test to get a specific outcome.

I suspect the idea that Bolshoy_Oleni_Ostrov has perceptible Botai/West_Siberia_N ancestry is something of a Twitter myth. But if you have any ideas of how to model Bolshoy_Oleni_Ostrov and Uralic speakers with this type of ancestry, then let me know and I can try it.

These are the outgroups I'm using (I took out RUS_Karelia_HG and RUS_West_Siberia_N, because the former is a reference pop, and the latter too similar to Botai)

Mbuti
MAR_Iberomaurusian
Levant_Natufian
IRN_Ganj_Dareh_N
Levant_PPNB
Anatolia_Barcin_N
GEO_CHG
RUS_MA1
RUS_Ust_Ishim
RUS_Kostenki14
SRB_Iron_Gates_HG
BRA_LapaDoSanto_9600BP
WHG

If anyone else has any ideas in regards to this topic, feel free to chime in.

Hugh Capet said...

@Davidski Thank you.

When_in_Rome said...

I always thought that the ANE derive from a population related to Kostenki14 and the Sunghir samples, since they are prehistoric Eastern Europeans, but with gene flow from an East Eurasian population. According to the paper "Paleolithic DNA from the Caucasus reveals core of West Eurasian ancestry" (Lazaridis 2018), this seems to be the case. But it would be interesting to see if the EHGs and Yamna samples are closer to Kostenki14 and Sunghir than MA1 and AG3. I believe the Proto-Indo-European speakers were predominantly of European origins.

With that said, to play devil's advocate, if they are closer to MA1 than to Kostenki14, wouldn't this indicate Central Asian origins?

Davidski said...

@When_in_Rome

With that said, to play devil's advocate, if they are closer to MA1 than to Kostenki14, wouldn't this indicate Central Asian origins?

Why would it, since MA1 and all closely related ancients are from Siberia?

When_in_Rome said...

@Davidski

Sorry, I meant to say Siberia. So would you say that the Yamna and Proto-Indo-Europeans ultimately have Siberian ancestry?

Davidski said...

@When_in_Rome

Sorry, I meant to say Siberia. So would you say that the Yamna and Proto-Indo-Europeans ultimately have Siberian ancestry?

Sure, why not?

And if we go back far enough we can say that they have African ancestry too.

When_in_Rome said...

@Davidski

I've always had trouble fully understanding the ANE since there aren't many samples. If you don't mind, can you tell me what your theory on the formation and origins of ANE, what route they took to Siberia, and when their DNA arrived in Europe?

Tea said...

@Hugh Capet

As mentioned above numbers were probably a factor or the major factor. IMHO They might not have been seen as much of a threat compared to how useful they were.

@matt

I think it's pretty dubious to say the least. The relationship of the branches within the family and classification of some of the languages and even sub-families is as far as i know unresolved. I haven't read the paper yet but I'm very skeptical that using 109 languages 949 lexical root-meanings is a good set up because the northern and central languages are much better studied and i'm sure over-represented.

Looking past the fact that their data can't be representative of the whole family the idea that there are known events that can be used for dating seems at best speculative and at worst an ad-hoc method that is inherently biased when applied to a family as a whole.

Again looking past this I wonder how the roots were selected and verified as low level reconstruction has only been done on some subfamilies and not for the larger language family.

I'm not saying that their conclusion is wrong or that they can't get anything out of their models but it's hard to be convinced that it isn't just picking up the relative flow of language and people due to Sinicization and the related north -> south movement of people over time being more impactfull than other movements.

Davidski said...

@When_in_Rome

If you don't mind, can you tell me what your theory on the formation and origins of ANE, what route they took to Siberia, and when their DNA arrived in Europe?

I don't really have a solid theory or even a strong opinion about any of that. It's just too far back in time for me to build up a strong interest in it.

My interest is more in the population dynamics of the West Eurasian Eneolithic, Bronze Age and Iron Age.

But check out this comment section and the linked preprint. As far as I know, it's the current state of knowledge in regards to ANE and related topics, and more or less accurate, even if far from perfect.

Y-haplogroup P1 in Pleistocene Siberia


Ric Hern said...

@ When_in_Rome

What I could gather is that it looks like this: Yana +-38 000 Years ago > Mal'ta Buret 24 000 Years ago > Afantova Gora related > Urals. Apparently the ANE component in Botai and Yamnaya split around +-17 000 Years ago. And if I recall correctly Villabruna +-14 000 Years ago have very little (Yet Detectable) ANE and some "Proto-CHG" ancestry. So somewhere between 17 000 and 14 000 ANE picked up some CHG maybe near the Southern Urals or near the Lower Don River (Kammenaya Balka).....

Bob Floy said...

@When_in_Rome

Also, EHG, which accounted for about 50% of Yamnaya's makeup, was the main vector for the spread of ANE into continental Europe.

When_in_Rome said...

@Davidski
Thanks for the link and help.

@Ric Hern
So it looks like a route from Northeast Siberia to the more southern Baikal region, then towards the Urals where they may have become EHG. But if this is true, why is it that ANE appear more related to Paleo-Europeans than to Asians? Even the ANS population looks closer to Paleo-Europeans than Asians.

@Bob Floy
Thanks, I knew this. I am trying to figure out the timeframe of ANE -> EHG. Based on the paper, it looks as though after 17,000 years ago for the arrival in Europe, albeit along the borders.

Ric Hern said...

@ When_in_Rome

Because that ANS population got replaced or migrated out of that area and East Asian relatives moved into that area....The other thing could be the origins of Haplogroup K. We see GHIJ in the Caucasus and Anatolia so K probably came from thereabout. Maybe the K that gave origin to P1 split up somewhere near the Caucasus and moved Northwards and Eastwards and then back Westwards as R and Q....?

When_in_Rome said...

@Ric Hern
Thanks

Ric Hern said...

@ When_in_Rome

My Pleasure.

Ric Hern said...

@ When_in_Rome

I think the clue to a very early West to East migration lies in the Vindija Neanderthal related ancestry....

jan.t.andersson said...

Plague has recently evoked interest as a key component in neolithic decline and population replacements at the time of indo-european expansion. Rasmussen et al (published in Cell January, 2019) found evidence of Yersinia pestis in neolithic individuals in Sweden dating from 4900 years ago and traced it to the Trypilja culture in present Romania/Ukraine. Maybe also a cause for the demise of the Maykop people?

Davidski said...

Yep, the plague and other western diseases may have decimated Central Asian populations too.

Aniasi said...

@Davidski

Ladakh, Kailasa, and Manasarovar were Indo-Aryan well into the Early Middle Ages. However, these regions are today considered Tibetic. Similarly, we can see that the far Northeastern reaches of Himachal Pradesh are areas of transition between Indo-European and Tibetic populations.

I believe, that despite the population turnover, that high resolution and high quantity sampling of present populations will show small signals of ancient Inner Asian Mountain Corridor peoples. If we can find pre-medieval samples, I firmly believe that we may see even more discernible lineages.

Davidski said...

@All

This is how Kumtepe4 and Kumtepe6 (from Copper Age and Neolithic Anatolia) plot in my updated PCA of ancient West Eurasia. Interesting trajectory, don't you think?

Image Link


Grey said...

Hugh Capet
"why do you think that the Steppe pastoralists failed to exterminate the people of the BMAC culture, and instead absorbed them into their genepool?"

Already answered but i think the underlying mechanism is crop-farming was developed in the optimal latitudes and that's where they had the maximum population density so i imagine what happened was as they expanded north (or south) from those optimal latitudes the farmer's population density decreased until they reach the crop-farming limit where they transition into (or catalyze the creation of) herders - so at the farmer/herder frontier the difference in population density between the two groups is at its lowest but that difference in population density increases as the herders expand south.

Grey said...

When_in_Rome said...
"With that said, to play devil's advocate, if they are closer to MA1 than to Kostenki14, wouldn't this indicate Central Asian origins?"

there's two aspects to truth:
1) is a set of words technically true
and
2) do those words reveal more than they conceal.

ANE is part of the ancestry of multiple populations and language families so if the region where all the strands for PIE came together was the region north of the Black Sea then saying IE came from Siberia or Central Asia conceals more than it reveals.

epoch said...

Kumtepe4 is just at the right place in the right time. It is really, really unfortunate that its such a low res sample.

I emailed the corresponding author of the paper to ask if a resample would be at all possible, and if so, if it was considered. No response yet. If they could and would resamnple, it would settle the debate.

epoch said...

The ADMIXTURE run in the Anatolian HG paper has Kum4 include, see supplementary figure 1. Especially K=12 shows clear steppe admixture.

https://static-content.springer.com/esm/art%3A10.1038%2Fs41467-019-09209-7/MediaObjects/41467_2019_9209_MOESM1_ESM.pdf

epoch said...

@When_in_Rome

Mal'ta-Buret' culture was completely dependent on mammoth hunting. Gravettian culture, while not so exclusively dependent also revolved around mammoth hunting.

This is an article on the so called "Mammoth steppe". It shows the population dynamics of Mammoths from 44.000 years to 14.000 years ago. I have seen no detailed idea worked out yet on how Western ancestry arrived in Eastern Siberia and the resulting ANE arrived into Eastern Europe. But apparently something similar happened to Mammoths. And since hunter-gatherers follow herds rather than drive them, I think the fact that a certain Mammoth ancestry prevailed in Eastern Europe as well as in Beringia basically shows the driving force of how ancestry got dispersed.

http://eurogenes.blogspot.com/2016/05/following-mammoth-herds.html

Mind you, there is a 14.000 year old cave, Kapova Cave, in the Urals with pictures of Mammoths.

http://www.visual-arts-cork.com/prehistoric/kapova-cave-paintings.htm

Ygor C.S. said...

Didn't some of that indigenous Siberian-like Central Asian ancestry survive in the modern populations due to their Minor absorption by Neolithic Iranian-like farmers and perhaps even part of the BA Indo-European herders? Didn't the South-Central Asia paper also detect some minor West Siberian-like ancestry in the Indus Periphery genetic structure? I have consistently found minor signs of West Siberian-like ancestry in the models for the genetic makeup of the Kalash that I have fone, even as much as 6% Botai-like ir 2% Bolshoy Oleni Ostrov-like. Could it be just a deceiving artifact? I have also got some minor percents of West Siberian_N or Botai in a few pastoralist samples associated with Andronovo and Sarmatian contexts. What do you think they mean?

Leron said...

If steppe is just EHG+CHG+WHG I imagine pseudo-steppe signals appearing at lower resolutions with those possessing similar mixture.

Samuel Andrews said...

@Leron,

That issue has been looked at it. I got clear cut results using David's G25 PCA.

https://docs.google.com/spreadsheets/d/1LPWAEC3dbAEDu8aBAAcxIOa5CQjuflt0f0cvhCpZ_ME/edit#gid=1497568895

Samuel Andrews said...

KumTepe4 (4000bc??), may have been roughly similar to Myceneans & Bronze/Iron age Balkans in general. He might have same amount of CHG/IranNeo, Anatolian, Steppe.

EurDNA said...

Tocharian data is also sorely needed, because the Afansievo link with Tocharians really needs to be confirmed rather than speculated.

Grey said...

EurDNA
"The extent of Collapse seen across the Balkans & western Anatolia precludes it anything to do with raiders from the Khvalynsk culture"

could "anything to do with" include displaced near-steppe populations moving away?

Davidski said...

@EurDNA

Tocharian remains have been sampled and I suppose we'll see the results at some point.

Davidski said...

Ygor C.S.

Didn't some of that indigenous Siberian-like Central Asian ancestry survive in the modern populations due to their Minor absorption by Neolithic Iranian-like farmers and perhaps even part of the BA Indo-European herders? Didn't the South-Central Asia paper also detect some minor West Siberian-like ancestry in the Indus Periphery genetic structure? I have consistently found minor signs of West Siberian-like ancestry in the models for the genetic makeup of the Kalash that I have fone, even as much as 6% Botai-like ir 2% Bolshoy Oleni Ostrov-like. Could it be just a deceiving artifact? I have also got some minor percents of West Siberian_N or Botai in a few pastoralist samples associated with Andronovo and Sarmatian contexts. What do you think they mean?

There was definitely some Botai-related ancestry in both the early farmers and Sintashta-related steppe groups of Central Asia, but it wasn't enough to survive past the Bronze Age. That's probably why the main Sintashta cluster, without any of this type of admix, can be used successfully as a proxy for steppe ancestry in modern Asian populations.

The other groups you mention, like the Kalash, are actually from outside of Central Asia, and beyond the scope of this blog entry.

Big Momma said...

Somewhat irrelevant, but somebody really needs to get Y DNA SNP calls for I2181, a Y DNA R from Bulgaria around 4500 BCE. I can't believe it hasn't been done already (unless I'm missing something), so I'm posting here just to get attention.

EurDNA said...

@ Davidski
Yes I was quite surprised to see that initial paper on Xiahoe mammies suggesting they are R1a, but then x Z93.
Although it was pre-Golden Age technology, so I'm not sure how reliable it was.

@ Grey

''could "anything to do with" include displaced near-steppe populations moving away?''

I'm not quite sure what the question is, but the curious thing is a relative depopulation occurred earlier in Anatolia, then hit the Balkans, which is why some of the paleo-ecologists from Germany have been positing a Rapid Climate change senario (6.2 ky event). Whatever happened, it hit both the flowering Balkan Chalcolithic societies, such as Varna, but also the Suvorovo chiefs, who appeared to have been profiting via their dealings with them. According to Manzura, Govedarica & Rassamakin, there is some (still difficult to quantify) lag between the disappearance of elite Suvorovo burials and the subsequent appearnece of steppe-like groups in northern Balkans (Cernavoda, etc), and the latter appear more austere compared to earlier steppe chiefs. But yes, if proto-Yamnaya was starting to push west, then some of the nearer-steppe gropus (post-Stog, etc) might have themselves move into the relatively vacant lands in Bulgaria & Romania; that would make sense

Samuel Andrews said...

**Kazakhstan
2000bc (and before).
Kazakhstan inhabited by people mostly ANE with significant East Asian (related to Mongolians & Mesolithic Lokomotiv).
1500-2000bc.
Europeans (Andronovo), are main inhabitants in Kazakhstan. They and native groups live side by side & admix frequently.
500bc-0ad.
Iranian-speaking. Various different kinds of Andronovo+Asian mixes. All have very minor Middle Eastern/BMAC ancestry, which could be because that is where Iranian language originated.
500ad-1000ad.
Turkic-speaking. 30-50% descended from Iranian-speakers similar to ones who lived in TianShan/western China. 30-50% new East Asian admix. 10-20% new Middle East admix.

**Tian Shan, Tajickstan, Kyrgzstan.
2000bc (and before).
Inhabited mainly by Middle Eastern BMAC. They mostly derive from a 'recent' wave from Iran not a Neolithic wave from Iran. They have significant Caucasus/Geogrianlike-ancestry.
1500-1800bc.
Andronovo, arrives. Unknown what happened. We just know lots of admixture happened.
500bc-0ad.
Iranian-speaking. Kangju, Wusun. 50% Andronovo, 30% BMAC, 20% Asian (East Asian, ANE).

**European Russia.
2000bc.
Inhabited mostly by Srubanya (Andronovo's brother).
500bc-0ad.
Iranian-speaking. 50% Central Asian (largely Andronovo), 50% Srubanya?

Samuel Andrews said...

There is lots of survival of Kazkahstan/central Asian hunter gatherers in historical Scythians who lived there.

But, there's no indication the Turkic groups in Kazakhstan had much ancestry from the previous Scythians. Their West Eurasian ancestry, seems to come from Tian Shan/western China Iranians not the ones who lived in Kazakhstan.

capra internetensis said...

@Big Momma

I2181 has shit coverage, I2340 is from same site and time but much better quality. He is R1b-V88 according to the Sardinia preprint.

JuanRivera said...

There is some West_Siberia_N in Khvalynsk and Progress. Khvalynsk tends to fit better as Progress+EHG+West_Siberia_N than as only Progress+EHG. Progress itself models best as Vonyuchka+EHG+West_Siberia_N. However, in all the models it's always below 8%, which is too low to cause cultural change. Most logically it comes from Caspian HGs.

JuanRivera said...

And from Middle Volga-Southern Urals HGs. The former would be variable mixtures of Piedmont, EHG and West_Siberia_N, whereas the latter variable mixtures of EHG and West_Siberia_N.

Samuel Andrews said...

@All,

All, ancient Iranian-speakers have BMAC ancestry. Would, it makes sense, that first Iranian speakers lived in SC Asia & were a BMAC/Andronovo mix?

This is interesting, "Iranian" is a big phenomenon in Eurasian history because they were very successful in spreading their language. Andronovo, in general I guess was an advanced, formidable ancient society that gave rise to Aryans & Iranians.

@JuanRivera,

One possibility, is Yamnaya's CHG ancestor had more ANE than currently published CHG.

Davidski said...

@Samuel Andrews

All, ancient Iranian-speakers have BMAC ancestry. Would, it makes sense, that first Iranian speakers lived in SC Asia & were a BMAC/Andronovo mix?

Srubnaya is generally seen as Proto-Iranian.

Samuel Andrews said...

@Davidski,
That's surprising. What's the traditional view of origin of Cimmerians. Continuation of Srubnaya?

JuanRivera said...

Looked again, it's indeed extra ANE, though on the EHG side. Odd how models pick Karelia_EHG over Samara_EHG, with ratios of 2.05:1 for each.

Davidski said...

@Samuel Andrews

There are all sorts of theories about Cimmerians, including that they were either Iranic or Thracian. But they're generally thought to be from north of the Caucasus.

Srubnaya was mostly pegged as the Proto-Iranian culture because it showed direct continuity with later steppe cultures, especially the Scythians. And Andronovo was usually associated with the Proto-Indo-Aryans.

That may or may not have been the correct way of thinking about things, but also there are late Srubnaya sites on the border between present-day Turkmenistan and Iran, and Iranian languages are first attested in that region during the Iron Age.

I'm sure you've seen this blog entry already, but anyway...

An early Iranian, obviously

JuanRivera said...

It's telling that the second closest distance to Khvalynsk is the Karelia_EHG UzOO77 sample.

Andrzejewski said...

Aren’t Cimmerians part East Asians?

Davidski said...

@Andrzejewski

How does that prevent the Cimmerians from having being Iranic or Thracian speakers?

Andrzejewski said...

Okunevo are Botai-like ancestry. I believe that they played some role in the ethnogenesis of the Tarim Basin Mummies, who were already determined to have had some “Siberian type” ancestry. Seems like Cimmerians and Saka had bunch of Botai/Okunevo DNA. One thing said about Okunevo is their similarity to Native Americans, probably because they were mostly ANE with a significant share of ENA.

Andrzejewski said...

They may have been Iranian or Thracian speakers but their European Strubna/Andronovo generics is admixed with Mongolian-like genes.

Andrzejewski said...

If I remember correctly there was a fairly recent paper (Narasimhan 2018?) discounting the amount of BMAC ancestry in Andronovo or Sintashta to a very negligible ratio. Concepts like the influence of BMAC religion on Avesta and Rig Veda, previously prevailing among some researchers were largely discarded too.

Andrzejewski said...

@Samuel Scythians, Wusun and Tocharians are depicted in Chinese sources as largely resembling modern Eastern and Northern European populations. Which means that neither BMAC nor Botai were in large enough ratios or propertions to significantly alter their gene pools. Around the time of Christ and until roughly 500 AD huge swaths of Asia were populated by Europoid looking Steppe/CWC derived tribes or kingdoms largely descended from Srubnaya/Poltava/Andronovo/Sintashta and/or Afanasievo cultures until they were pushed back by Turkic and Mongolian groups.

Andrzejewski said...

Steppe Maykop May have been absorbed by contemporary Caucasus populations. There are three language isolate groups in India and Nepal: Burusho, Nihili and Kusundra and one of them may have some Botai like DNA.

Andrzejewski said...

I’ve been trying to figure out where the Ruan Ruan people came from, who gave rise to Altaic groups such as Turkic and Mongolians.

Samuel Andrews said...

@Andre,
"Around the time of Christ and until roughly 500 AD huge swaths of Asia were populated by Europoid looking Steppe/CWC derived tribes or kingdoms"

I agree, Central Asians looked different back then but it wasn't an eastern extension of Europe. They were very mixed. Few, were over 50% European.

Kazkhstan was an Asian-European mix. Most Scythian samples from Kazakhstan are more Asian (East Asian, ANE) than European. One was more European than Asian. One is only 25% European. *Some*, Scythians looked east Asian.

Wusun, Kangju were 50% European/Andronovo, 25% West Asian/BMAC, 20% Central Asian (ANE, East Asian). The only group to remain overwhelmingly Andronovo was Tagar in Siberia. Tagar had descendants Tashtyk culture living there till 400ad.

Andrzejewski said...

@Samuel Cue Tarim Basin Mummies, very Euro looking, including Cherchen Man, Ur David and “beauty of Lulan”

Davidski said...

It seems to me that some people are confusing a few issues here.

The main thing to understand is that the waves of Sintashta-related groups that moved into Central Asia from the northwest replaced and, in part, absorbed the native Central Asian peoples to such an extent that they ceased to exist as coherent populations, even if some of their ancestry persisted for a while in the region.

But of course the population shifts in Central Asia continued throughout the Bronze Age and into the Iron Age and Medieval period, with a lot of gene flow also coming in from the east. And this is also what has caused the disappearance of the indigenous Botai-related ancestry in Central Asia.

Some populations outside of Central Asia, like in Siberia and South Asia, might still carry small amounts of Botai-related ancestry, but that's beyond the scope of this blog entry, because, well, they live outside of Central Asia, and it's not clear yet how they acquired it.

Davidski said...

And by the way, Botai were not ANE. They were very far from ANE, because unlike AG3/MA1, which are ANE, they had EHG-like and more pronounced East Asian ancestry.

HAUMAVARGĀ said...

@Andrzejewski

Michael Witzel had written some really good articles about the BMAC influence on Indo-Iranians. What are your sources to discard the BMAC effect?

Samuel Andrews said...

Look, at this evidence the Caste system in India was set up by Aryan invasion.

Kalash=Good proxy for Bronze age South Asian Aryans. Kalash themselves might be only 30% Andronovo.

Shahr_I_Sokhta_BA3=Harrappan immigrant in bronze age BMAC culture. Malayan=most ASI Indians.

1.8111"

Gujarati

Shahr_I_Sokhta_BA3,54.5
Kalash,28.7
Malayan,16.8

1.795"

Brahmin_Gujarat

Kalash,56.8
Shahr_I_Sokhta_BA3,24.5
Malayan,15.2
Brahui,3.4

2.9181"

Bengali

Malayan,48.7
Kalash,26.1
Shahr_I_Sokhta_BA3,20.2
Ami,5

1.8806"

Brahmin_West_Bengal

Kalash,54.3
Malayan,41.9
Shahr_I_Sokhta_BA3,3.7
Brahui,0.1

Davidski said...

@Samuel Andrews

Look, at this evidence the Caste system in India was set up by Aryan invasion.

Not necessarily. Something like the caste system may have already existed in South Asia before the arrival of the Indo-Europeans, and they may just have used it to their advantage.

HAUMAVARGĀ said...

@Samuel Andrews

If you sent your last comment as a reply to me I don't see how it proves anything since Kalash people themselves can be modeled with BMAC.

Big Momma said...

@capra internetensis Well has anyone checked? Someone on here must know how to check the Y DNA of people like I2181 and I2430. They could be pre-L51 for all we know, V88 doesn't seem associated with Steppe ancestry so much.

Big Momma said...

Someone really needs to sort that out, how do we still not know what subclade of Y DNA R brought Steppe ancestry early on to the Balkans. I'd be surprised if it wasn't M269 really.

Matt said...

Antiquity of Caste is a pretty difficult question; Caste / jati ideas certainly develop by mid-late first millennium BCE in at least some populations in South Asia (with mature "synthetic" Hinduism), but is there evidence for them before that, including from the first introduction of Indo-Aryan languages to South Asia?

The largest sample sets we have so far of South Asia are the _IA sets, reasonably late, and they tend to cluster reasonably locally in terms of ancestry, certainly with respect to AASI fraction*, without much evidence of outliers that suggests a very structured local population. Other than that one outlier at Saidu_Sharif_IA. However there may be some sort of burial bias or other concerns there, so this is not conclusive in either direction.
See: https://imgur.com/a/hFJzcey

There was some suggestion that the heterogenity of Indus_Periphery outliers found in AASI ancestry may represent a structured population.... but since these are all conceptualized as migrants out of IVC, it is reasonably possible they could all just be like this as migrants from different places that were not in themselves Caste structured.

*More heterogenity wrt steppe related ancestry? I6893 Saidu_Sharif_IA woman ("DA-SIM0317-111, Grave 7, Individual 1") looks more similar to Rors than to other SS_IA population for'ex. Saidu_Sharif_IA seems more diverse and described as Buddhist site (https://en.wikipedia.org/wiki/Saidu_Sharif_Stupa - pilgrims?). Nothing special about I6893's grave. Neither seems anything special about grave of I7722 male outlier ("DA-SIM0317-077, Grave 2"), the high AASI sample. It may be interesting to look at the individual Iron Age samples in terms of position on intra-South Asian clines over time.

Andrzejewski said...

Narasimhan 2018, which also confirmed India’s founding Steppe lineage

Andrzejewski said...

Kalash people look more Europoid than other Northern Indian populations, hence Andronovo must be higher

Andrzejewski said...

Interesting to find out who the descendants of the Scythians, Cimmerians and Thracians are amongst modern day nations

Matt said...

OK, if anyone's interested, some time vs ancestry plots for the South Asian Indus_Periphery outliers and Iron Age samples so far in G25 and Narasimhan 2018 supplement: https://imgur.com/a/GwVphkG

Matt said...

Btw, Davidski, not sure if I6888 should be Saidu_Sharif_IA. In supplements for Narasimhan (p90), she's "Aligrama Village, Swat Protohistoric Graves, Pakistan - DA-ALI0317-001, T 29 (I6888): Date of 1000-500 BCE. ". They treat this as a separate cultural period that the other Aligrama Village samples. Note she has a different skeletal id code than everyone else put in the SS_IA group, but they give her the same latitude and longitude in the spreadsheet supplement?

So not the same site as others in her G25 classification. Doesn't seem like one of the Saidu Sharif Stupa burials?

Graphics with her in a separate category of her own: https://imgur.com/a/j7e7Yhd.

The IA samples, including the ones from around 1000 BCE and Saidu Sharif and Butkara later seem fairly constrained in their degree of ASI ancestry, probably more in the Aligrama_IA village and Buktara set for some reason. Looks like mild trend of increases in Indo-Aryan and ASI related ancestry? Saidu Sharif site looks most diverse in the degree of steppe related ancestry, poss 'cos. religious site pulling in folk from farther afield?

Richard Rocca said...

@Big Momma, I2181 does have poor quality. He is positive for an M269 equivalent:

R-P280
R1b1a1a-CTS9018
R1b1a1a2-PF6452

...however, he is negative for these three haplogroup P equivalents that all M269 men are positive for:

P-CTS3813
P-CTS10081
P-PF5867

So, like most junk-like samples, we can say noting definitive for this sample.

Davidski said...

@Matt

I6888 is marked as Saidu_Sharif_IA in the dataset and spreadsheet accompanying the paper. But yeah, these might be errors.

The paper is coming very soon, along with a new dataset, so let's wait and see.

Big Momma said...

@Rocca But if he's positive for those M269-related subclades does that not mean he's necessarily at least P297? I know nothing about this, but surely you can deduce somebody has ancestor subclades if they test positive for any of the descendent subclades of that missing subclade right? Unless testing for SNPs isn't black-and-white, but again, I know nothing.

I'll just quietly assume he's M269 pre-PF7562 for now!

Big Momma said...

Thanks for doing the work there by the way, even if I could learn I don't have Windows.

capra internetensis said...

@Big Momma

I looked at I2430's 1240 K file and he is R1b-L754(xP297), so there's no reason to think Marcus et al got it wrong.

Mesolithic foragers (majority EHG) on the Ukrainian steppe had R1b-V88, as did the Iron Gates foragers in Serbia and Romania, so it is not too surprising to find it in Bulgaria, with or without any Steppe ancestry.

SNPs can be positive due to damage or sequencing error etc, just one downstream SNP isn't good enough to make a call. False positives are common. It is not easy to tell whether a call is any good the sample quality is so low.

Gabriel said...

@Samuel Andrews
The Tien Shan Sakas and Tien Shan Huns were what mix ?

Richard Rocca said...

@Big momma said...I'll just quietly assume he's M269 pre-PF7562 for now!

He should have been positive for those three haplogroup P SNPs and he wasn't. Given that, we can't even assume he belongs to haplogroup R let alone some kind of M269.

Samuel Andrews said...

@Geberial,


Saka_Tian Shan: 40% Andronovo/Srubnaya, 25% BMAC, 20% East Asian, 14% ANE.
Saka_Hun: 39% Andronovo/Srubnaya, 18% BMAC, 31% East Asian, 11% ANE

Single Hun genome from Hungary, is basically identical to Saka_Hun. In my opinion, some Huns were Tian_Shan Saka origin, but the elite of mostly East Asian origin.

Aniasi said...

Agreed. Higher Iran Neolithic also prevails in higher castes.

Samuel Andrews said...

@Aniasi,

Brahmins have a little more IranNeolithic. But, Brahmins have twice as Andronovo. So, the Aryans were of largely Andronovo origin & placed themselves at the top of the caste system.

rozenfag said...

@Samuel Andrews
Which Hun genome from Hungary are you talking about? Can you provide link?

Davidski said...

It looks like the big horse paper is coming next week, or soon anyway. Here's the abstract from the ENA. Does anyone know what the hell the part in bold imply? I assume the Siberian horse lineage is the Botai horse.

Horse domestication revolutionized warfare and accelerated travel, trade and the geographic expansion of languages. Here we present the largest DNA time-series for a non-human organism to date, including genome-scale data from 149 ancient animals and 129 ancient genomes (≥1-fold coverage), 87 of which are new. This extensive dataset allows us to assess the modern legacy of past equestrian civilisations. We find that two extinct horse lineages existed during early domestication, one at the far western (Iberia) and the other at the far eastern range (Siberia) of Eurasia. None of these contributed significantly to modern diversity. We show that the influence of Persian-related horse lineages increased following the Islamic conquests in Europe and Asia. Multiple alleles associated with elite-racing, including at the MSTN “speed gene”, only rose in popularity within the last millennium. Finally, the development of modern breeding impacted genetic diversity more dramatically than the previous millennia of human management.

https://www.ebi.ac.uk/ena/data/view/PRJEB31613

Davidski said...

OK, I think I get it. What they're trying to communicate is that there were several horse species during the early domestication period (which should be the Eneolithic), and the modern domesticated horse lineage didn't derive in any important way from the extinct Iberian or Siberian lineages.

But then it seems like there's a sentence missing, because they skip to the historic period, when Arabian and Turkomen horses significantly impacted on modern breeds, without actually saying where the main modern domesticated lineage came from.

They're probably being obtuse on purpose. I'll try and fill in the missing part...

We find that two extinct horse lineages existed during early domestication, one at the far western (Iberia) and the other at the far eastern range (Siberia) of Eurasia. None of these contributed significantly to modern diversity, which largely derives from a horse species native to the Pontic-Caspian steppe in Eastern Europe. We show that the influence of Persian-related horse lineages increased following the Islamic conquests in Europe and Asia.

Dragos said...

This is the end
Beautiful friend
This is the end
My only friend
The end

Davidski said...

@Dragos

Yep, big fan...

Apocalypse Now intro: The Doors, The End {1979}

This movie's awesome too.

Dragos said...

Davidski
Yep. How are you with horse genomes ?

Davidski said...

Not very good. I'll be waiting with great anticipation for the paper. Hopefully it doesn't take too long.

Davidski said...

This should be pretty good. They've got Mesolithic, Neolithic and Copper Age samples from the forest steppe and steppe in Eastern Europe and Central Asia, and a Copper Age sample from Iran.

If Kokorevo_Rus3_14450 (from the forest steppe, and presumably dating to 14,450 BP) is shown to be ancestral to the modern domesticated horse lineage that'll be awesome.

Altata_NB31_Neolithic
Belkaragay_NB13_CopperAge
Belkaragay_NB15_CopperAge
CaminoDeLasYeseras_CdY2_4678 (Bell Beaker horse?)
Derkul_NB2_Neolithic
Derkul_NB4_Neolithic
Kokorevo_Rus3_14450
LebyazhinkaIV_NB35_Neolithic
PotapovkaI_1_3900
Sagzabad_SAGS27_3117
Sintashta_NB44_3577
Sintashta_NB45_3577
TepeHasanlu_1140_2682
TepeHasanlu_2327_2352
TepeHasanlu_2529_2352
TepeHasanlu_2689_2352
TepeHasanlu_3394_2808
TepeHasanlu_3398_2352
TepeHasanlu_3459_2667
TepeHasanlu_3461_2930
TepeHasanlu_368_2896
TepeMehrAli_Trj12x31_CopperAge
Yerqorqan_YER28_2853

rozenfag said...

Kokorevo seems to be in Siberia, close to Krasnoyarsk. That's quite far away from Pontic-Caspian steppe. May be it is a dead end lineage?

Davidski said...

Not sure, maybe I got the wrong Kokorevo?

https://geographic.org/geographic_names/name.php?uni=-4105329&fid=5516&c=russia

rozenfag said...

There are several places with that name in Russia. But when I searched for Kokorevo archaeology, most of the results were about place in Siberia: https://www.academia.edu/31523281/The_large_mammals_of_North-Minusinsk_basin_in_the_Last_Glacial_period

Davidski said...

Yeah, it might be an extinct Siberian horse.

Ric Hern said...

@ Davidski

The Iberian early horses are most probably the Soraia breed and the Siberian horses related to Botai...

Ric Hern said...

Tracking of the female lines maybe would have thrown some light on survivors of early domestication into Modern Times since most Males are now related to the Arabian horse and Akhal Teke...

Ric Hern said...

Maybe some Early domesticated Steppe Male or Female lines survived among the Fjord Horse and or Icelandic Horse....

Davidski said...

@Ric

Almost all domestic horse males carry the same Y-chromosome lineage as Scythian horses (Y-HT1), so this is an early steppe line. The only exceptions are some Siberian and Mongolian horses. See here...

https://www.nature.com/articles/s41598-019-42640-w/figures/4

The position of variant fBOI (indicative for Y-HT1 in Wutke et al.77) is marked by an arrow.

Ric Hern said...

Yes, Scythian is the key word here. Let's wait and see what the paper comes up with. As far as I know the Scythain horses were Akhal Teke like. And we know of later imports and crosses especially with Turkic Horses. So even if Y-HT1 were the early branch it is not to say that maybe ancient lines of Y-HT1 in Europe survived the Y-HT1 imported in Historical times...just like R1b V88 did not survive R1b M269 in most parts of Europe...

EastPole said...

@Davidski
“We find that two extinct horse lineages existed during early domestication, one at the far western (Iberia) and the other at the far eastern range (Siberia) of Eurasia. None of these contributed significantly to modern diversity, which largely derives from a horse species native to the Pontic-Caspian steppe in Eastern Europe.”

Probably there were many horse domestications around the world. Globular Amphora Culture in Poland also domesticated horses and buried them in graves with other domesticated animals. But GAC horses were small forest horses. Later CWC introduced large steppe horses.
So yes, modern horse species came from the steppe. The big question, important for IE problem, is from which steppe, from Sredny Stog or from Yamnaya?

Davidski said...

@Ric

Yeah, it gets confusing, because even though Y-HT1, and its sister clade Y-HT4, are probably originally from the steppe, and Y-HT1 is actually first attested in Bronze Age samples from Slovakia, almost all European horse males carry Y-HT1 lines from Near Eastern founders.

This has caused many people to assume that the modern domesticated horse breed, or at least the Y-HT1 lineage, originated somewhere in Asia.

So this paper is going to be very important in clearing this up, and I don't think it's a coincidence that the authors sampled so many ancient Iranian horses. They probably went out of their way to do that, so that they could establish the direction of the multiple gene flow events between Iran and Europe across space and time.

And, for obvious reasons, it's going to be a big paper for the Proto-Indo-European (PIE) homeland debate, maybe even bigger than any human ancient DNA paper to date.

Davidski said...

@EastPole

So yes, modern horse species came from the steppe. The big question, important for IE problem, is from which steppe, from Sredny Stog or from Yamnaya?

I'm pretty sure that they kept the same type of horse. Basically the same one that Sintashta had.

Ric Hern said...

@ Davidski

Indeed.

old europe said...



regardless of whom domesticaed the horse fact is the horseheaded sceptres with a complete horse figure are older in the western steppe. That means that the horse simbology and hence the importance of the horse was already alive and kicking in the big agricoltural societies of the western pontic steppe. They ( the sceptres) spreaded later to the Volga ( see Govedarica). So chances that the first horse tamers were the steppe people ( Sredni Stog or Skelya culture as more correctly rasamakin defines it) that more closely interacted with Cucuteni and company ( and were heavily farmer mixed). Anthony wrote that some Sredni Stog elite and clans were incorporeted in the eastern Tripolie societies. I bet individual I6561 and I4140 were among them. Sredni Stog as I said many times ( read Anthony, Rassamakin, Telegin and Kotova, Manzura) looks like an offshoot of farmers culture east of the Dneper. Add to this that Yamanaia was born out of a strong influence of SS. Ukraine eneolithic genetic signal enters in the Yamanaia genetic profile.

old europe said...


Contact zone between farmers and dneper/ donets locals is the place of birth of first kurgans and corded pottery ( there is a debate going on if corded pottery were of farmers coinage too).
I think we are moving from a scenario in which the PIE genetic marker was EHG + CHG like to ome in which the real marker is EEF + Steppe eneolithic.

Davidski said...

@old europe

The important issue isn't where horses were first domesticated, but where the modern domestic horse breed was domesticated, because this is the breed that expanded with Indo-European cultures, and Indo-European cultures share a horse cult that probably traces back to their shared homeland.

Aram said...

In the Wang et al. paper there was a unpublished sample from a LBA-IA Koban culture in Caucasus.
Based on the Amixture he had a lot of Siberian/ East Asian ancestry and could be similar to those Cimmerians that we have seen in Moldova.

Matt said...

Re; horses, OK, let's have a look at sorting all these listed samples by date:

https://imgur.com/a/EOcB17R

Note not all horses - taxid319699 are mules, taxid9793 are donkeys and taxid9794 are your onagers. taxid9796 are E. Caballus horses

Will have to have a closer look and see what samples they have. We know that there were horses down through Central Europe and the Near East and some horse trading, but the abstract only calls out the (S)Iberian populations as extinct. So does this rule out a fairly extended domestication horizon or if not that, early domestication bolstered by wild breedstock from an extended region? Or a limited domestication and breedstock horizon only?

Davidski said...

@Matt

Whoops, yeah, some of those "horses" from Iran in that paper are actually onagers and donkeys.

In regards to an extended horse domestication horizon, from earlier research we already know that it wasn't overly extended.

There's a single modern domesticated clade with little admixture from other clades, and two main Y-chromosome lineages that reach high frequencies during the Bronze Age (and then one later goes extinct).

The problem is that, up to this paper, it wasn't clear where this clade came from exactly, and especially whether it's native to Eastern Europe or West Asia. Hopefully this paper clears that up.

Matt said...

It doesn't look like they have *that* many samples to test that - in the frame of 4000-3000 BCE when horse domestication is supposed to have place, maybe the Lebyazhinka (Samara?) horses are relevant, while Belkaragay / Derkul / Altata are Kazakhstan I think and not really Western Steppe or Near East? Everything else looks too late or early if I'm reading the dates correctly.

Davidski said...

@Matt

Derkul seems to be on the Pontic-Caspian steppe, even though it's in Kazakhstan. It's close enough anyway.

So the samples from Russia and nearby parts of Kazakhstan should be informative about whether the ancestors of the modern domesticated horse clade lived in that area originally.

Conversely, the Iron Age (and Late Bronze Age?) samples from Iran, and there are a fair number of them in the paper, should be informative about whether Eastern European horses appeared in the Near East during the metal ages and whether they totally replaced local horses or were mixed with them.

I suppose that without pre-Iron Age samples from Iran there will always be some doubts about what those Iron Age Iranian horses really represent, and what their relationship is exactly with Eastern European horses, but it shouldn't be too difficult to test whether they show a very recent and direct, as opposed to a much deeper, relationship with Eastern European horses.

Davidski said...

By the way, Altata is in Russia south of Samara.

https://www.google.com/maps/place/Altata,+Saratov+Oblast,+Russia,+413465/@51.2151585,39.747699,5z/data=!4m5!3m4!1s0x416d0c9af03054fb:0x2146ada854c3fb64!8m2!3d51.100792!4d48.718582

Matt said...

Possibly Derkul could be Derkul River in Ukraine as well?

Davidski said...

Might be. That's close to Alexandria where the (human) R1a-M417 sample is from.

Gaska said...


Late quaternary horses in Eurasia in the face of climate and vegetation change- Michela Leonardi, Francesco Boschin (25 julio 2.018).Climate controlled the distribution of horses, and they inhabited regions in Europe and Asia with different climates and ecosystem productivity, suggesting plasticity to populate different environments. Their decline in Europe during the Holocene appears associated with an increasing loss and fragmentation of open habitats. Europe was the most likely source for the spread of horses toward more temperate regions, and we propose both Iberia and central Asia as potential centers of domestication”. The suitability of both Iberia and eastern Europe appears constant throughout the entire post-LGM period, in line with these regions being hotspots of genetic diversity and, possibly, the refugia sources for the recolonization of the continent. While the Pontic-Caspian region appears not suitable for European horses around the time when horses where first domesticated some 5.5 ka ago), part of this region appears suitable for the Asian horses (with the Caspian Sea as the westernmost boundary). This may suggest that horse domestication started from a population background related to an Asian ancestry and that the further spread of the domesticated horses in Europe involved either adaptation to novel niches (possibly through selective breeding) or the application of domestication techniques to local horse populations pre-adapted to these environmental conditions.

+ Tracing the origins of the domestication of the horse in Iberia: ancient DNA and the evidence of Atapuerca. Jaime Lira (2015). The analysis of mitochondrial DNA of a large number of individuals of different races distributed throughout the world has allowed to know that in the process of domestication of the horse were involved a large number of maternal lines, reaching to differentiate, at least, up to 17 haplogroups. The presence of Neolithic horses and the Bronze Age in haplogroup C typical of Lusitanian horses showed that, at some point, wild Iberian horses had become part of the peninsular domestic populations. Our research confirms that, at least, Iberian wild mares had contributed with their DNA to make up Iberian domestic populations.

That is, I do not know what they mean by extinct lines because the Lusitanian horses have been in Iberia for thousands of years and were domesticated in situ.

Davidski said...

@Gaska

That study is based on low resolution mtDNA. Best wait for the new paper which has full genomes from ancient Iberian horses.

Gaska said...

@Davidski

Yes we will see. In Iberia there are indications of domestication since the late Neolithic, but only for meat consumption and to work in the field. Evidences of horse riding in male skeletons have been detected in the Bronze Age, specifically in the village of Almoloya (Argar culture).

Jaime Lira-Neolithic and Bronze Age sequences grouped in other clusters, one of which
(Lusitano group C) is exclusively represented by modern horses of Iberian origin.
Moreover, Bronze Age Iberian sequences displayed the lowest nucleotide diversity values
when compared with modern horses, ancient wild horses and other ancient domesticates
using nonparametric bootstrapping analyses. We conclude that the excessive clustering of
Bronze Age horses in the Lusitano group C, the observed nucleotide diversity and the local
continuity from wild Neolithic Iberian to modern Iberian horses, could be explained by
the use of local wild mares during an early Iberian domestication or restocking event,
whereas the D1 group probably was introduced into Iberia in later historical times.

They are not bad samples or of bad quality, but it will undoubtedly be interesting to check if the horses of the steppes arrived in Spain. If I had to bet, I would say no.

Davidski said...

@Gaska

They are not bad samples or of bad quality, but it will undoubtedly be interesting to check if the horses of the steppes arrived in Spain. If I had to bet, I would say no.

All modern horses, except Przewalski's horses from Mongolia, belong to the same lineage that dates back to the Eneolithic.

So either the ancestors of the horses that are in Spain now arrived from somewhere else after the Eneolithic, or all modern horses except Przewalski's horses are from Iberia.

But of course we already know that native Iberian horses went extinct.

Gaska said...

@davidski

I suppose you will refer to the males because haplogroup C females have been here since the Neolithic. They are like Spanish women (I hope that women who read this are not offended). Everyone is determined to extinguish everything that is native to Iberia, men, horses ...


By the way, today begins the second Reconquista in Spain.

Gabriel said...

@Samuel Andrews
Thanks. Do modern Kazakhs approximate these results ?

Matt said...

On the horse topic, random links of possible interest and some highlighted suggestions of interest:

https://www.cell.com/trends/genetics/fulltext/S0168-9525(19)30035-6?rss=yes - "The Promise of Paleogenomics Beyond Our Own Species" - Reich and Catherine Bruson's recent April '19 article on animal adna. On horses: Case Study #2: Ancient Horse Genomes Reveal Multiple Episodes of Domestication Domestication involves mutualistic relationships between humans and domesticated plant and animal species; it occurs on a continuum ranging from minimally managed species requiring only limited human investment, to fully domesticated species that rely on humans for all aspects of their survival and reproduction [41]. Paleogenomic technology is changing our understanding of domestication as it gives us the opportunity to witness the process as it unfolds instead of being limited to studying the end product. Arguably the most impressive research in this area has been on horses, the species in which the quality of paleogenomic data and population genomic analysis has been most similar to, and in some cases has exceeded, that in humans. The startling findings of these studies highlight the potential of aDNA studies of non-humans.

The earliest archaeological evidence for horse domestication comes from sites in the Pontic-Caspian steppe and Kazakhstan, with the Eneolithic Botai culture dated to around 5500 years ago [42–45]. In the most ambitious aDNA study of non-humans to date, Gaunitz et al. [8] sequenced over 40 ancient horse genomes and found evidence for multiple horse domestication processes and an unanticipated feralization of the Przewalski’s horse. The authors compared genomes from 20 horse samples associated with the Botai culture sites with genomes from 20 horse samples excavated from other locations in Europe and Central Asia, covering the past 5000 years. They also compared these ancient genomes with previously published ancient and modern domestic horse genomes, and a 19th century Przewalski’s horse genome [43,46–48]. Przewalski’s horses have long been considered to be the only living wild horse [47].

However, Gaunitz et al. found that the Przewalski’s horses clustered together with the ancient domestic Botai horses, which is a pattern consistent with descending directly from them, while all other domestic horses clustered into a separate monophyletic group. The simplest explanation for this is that Przewalski’s horses are a feral population descended from the horses that were first domesticated at sites like Botai. Modern domestic horses likely resulted from a separate later domestication episode that occurred sometime before 4000 years ago, from a different lineage of wild horses before the time that all wild horses went extinct.
(tbc.)

Matt said...

(cont.)"The analysis of Gaunitz et al. also showed that there was admixture between various horse populations. This is another important piece of paleogenomic evidence that interbreeding between closely related populations is common. In some cases where ancient admixture has been identified, interbreeding appears to have introduced adaptive alleles into ancient populations [49–52]. For domesticated species, introgressive capture of wild animals and admixture between domestic and wild stocks was an important feature of animal exploitation [53–56].

Admixture with wild populations plausibly increased herd health by re-introducing genetic diversity to domestic populations and may have given domesticated species advantages when living in new environments [54,56–59].

The surprising findings of Gaunitz et al. about horses reveal that animal domestication can be a complex process, characterized by feralization, population turnover events, and interbreeding between wild and domestic stocks [53–56,60].

These findings underscore the point that past population dynamics cannot always be inferred from present-day DNA variation. This is especially the case for horses. All living Przewalski’s horses descend from just over a dozen founder individuals captured during the turn of the 20th century and saved from extinction through captive-breeding programs [61]. Przewalski’s horses and modern domestic horses may appear to be from unrelated lineages when modern DNA is analyzed, but paleogenomics reveals that they did not evolve in complete isolation, and that the history of horses may have involved multiple experiments in human management of wild populations that are difficult to detect without the insight provided by ancient genomes. Paleogenomics has also revealed the existence of archaic lineages of now-extinct horse populations that would have remained


So Reich does I guess, favour some degree of introgression and combination, based on Gaunitz, even though it may be quite likely that there was a genetic "main source" of ancestry for the main domestic horse lineage.

Matt said...

Turning to Gaunitz (https://science.sciencemag.org/content/360/6384/111), I think we should note (and for me this was a bit lost in the furore of the main difference between Botai horses and the present day domesticated lineage):

"Outgroup f3- and D-statistics also revealed that Dunaújváros_Duk2 (Duk2), the earliest and most basal specimen within DOM2, was divergent to all other DOM2 members. This is not due to sequencing errors, because the internal branch that splits from Duk2 and leads to the ancestor of all remaining DOM2 horses is long (Fig. 2B). This suggests instead shared ancestry between Duk2 and a divergent ghost population. We thus excluded Duk2 in admixture graph reconstructions (16) to avoid bias due to contributions from unsampled lineages (Fig. 3).".

Further, in the supplement Optimizing f2 parameters revealed that the genetic drift from Dunaujvaros_Duk2_4077 to the rest of ancient domesticates is large (f2 = 0.00026) with regards to the limited period of time separating the age of this sample and the node ancestral to all remaining members of the DOM2 lineage...The divergence of Dunaujvaros_Duk2_4077 within the DOM2 lineage is rather suggestive of genetic ancestry from an unknown ‘ghost’ population within this sample. .

So the earliest Czech horse (Vatya Cultural context, 2077 BCE) is not just basal within its clade, but has some introgression of ancestry which the others in the sample do not have, or least have in a diluted / different ratio... (The other earliest seems to be from Sintashta culture at 2023 BCE).

It's interesting that some of these dates in Gaunitz for expansion of today's domestic horse group and given by Reich tend to be around 4000-4500 YBP (2000-2500 BCE) in this paper, and not any earlier at 3500 BCE or so.
(E.g. under their Bayesian skyline on mtdna, growth phase hits around 2500 BCE following a sharp decline before this - https://imgur.com/a/fbTq6hW).

That's not to say the timing is that precise, necessarily, but it's worth a bit of consideration, since these dates are fairly important (e.g. what does it mean if the growth of the domesticated horse population really only got going on the steppe during the mature Sintashta / Andronovo phase derived from CW and not Yamnaya?).

(Outram last month described the domestication of the main domestic horse as "later" than the Botai - https://astanatimes.com/2019/03/botai-horses-in-north-kazakhstan-were-first-to-be-domesticated-says-uk-professor/. However he may well just be describing that in terms of introduction to Kazakhstan.)

I'd also note that Gaunitz states evidence of Sporadic introgression of Botai ancestry into multiple DOM2 herds occurred until 1000 years ago. This gene flow was mediated not only through females, because 15 Botai-Borly4 individuals carried mitochondrial haplotypes characteristic of DOM2 matrilines (figs. S12 and S13), but also through males, given the persistence of Botai-Borly4–related patrilines within DOM2 (figs. S15 to S18)

Lastly, although the genetic load of PH and Botai-Borly4 genomes was equivalent until ~118 years ago, it drastically increased in modern animals (Fig. 4C). This accumulation of deleterious variants was thus not associated with PH feralization but with the recent introgression of deleterious variants from modern domesticates and demographic collapse, which hampered purifying selection.

So although contributing in very limited amount to the main lineage, the Botai horse lineage did not go extinct immediately or reduce in population size sharply, but must have been around in feral / non-feral form for quite a while, somewhere.

Matt said...

Oops, Hungary, not Czech!

Davidski said...

@Matt

It'll be interesting to see what the new paper says about the date of the domestication of the non-Botai lineage.

I'll be surprised if there aren't any domesticated samples from the Pontic-Caspian steppe dating to before 3,000 BCE, but hey, I love surprises.

In any case, it seems like we're dealing here with at least two quite distinct domestication sources, and then a quick expansion of the horses from the non-Botai source, after which a lot of complex things probably happened, including introgressions from disparate lineages.

For me, the most important thing is which wild horses are ancestral to the main domesticated lineage that we already know was present in Sintashta, Scythian, Gallic cavalry, Iron Age Iranian etc. burials, because that will also tell us where the early Indo-Europeans got their horses. And it's unlikely that they got them very far from their homeland.

Davidski said...

@All

Apparently domestic horses and Przewalski's horses aren't different species. They're different sub-species.

My mistake.

Matt said...

@Davidski: I'll be surprised if there aren't any domesticated samples from the Pontic-Caspian steppe dating to before 3,000 BCE

Hopefully they'll be able to say something pretty clear about whether any samples they find are domesticated or not.

Going from the Gaunitz 2018, genetic signs of domestication would be pretty inconclusive at this stage (it seems?) but if there's some complex admixture process in the genesis of early domestication that just can't be explained any other way, or other changes in population size which are strongly suspect to be related to domestication*, they might be able to refine something there. (Otherwise they're falling back on skeletal measurements as index of domestication as in Gaunitz 2018, though they also thought they saw something in coat color variants).

*The overall load in early domesticated horses, at Botai/Borly4 and among the "Ancient domestic" clade seems roughly the same (unlike the present-day domestic and Przewalski samples post-118 years ago which have quite heavy load), but maybe they will find some signal looking through very early horses?

Ric Hern said...

@ Davidski

That Salzmünde Tabiano horse still needs explaining....

olga said...


Actually in the North of Spain exist small horses that look like those painted in the caverns. The Potoka in Euskadi (Basque Contry) Asturcon in Asturias and Losinos. Some of them are now kept free and protected in the Cantabrian mountains. The Potokas used to work in the mines, carrying materials, coal and iron.
I am very curious to know if they are autocthonous as people thinks or not.

Gaska said...

@Olga

I do not know any DNA study of pottokas and asturcones, although they are very similar to the horses painted in the Cantabrian caves (Ekain, Santimamiñe, Altamira....). do you know any?

Un saludo

Davidski said...

@All

Does anyone know where Frank is hiding? He went MIA in the last comment thread.

I need to talk to him about the distribution of Botai-related ancestry in Eurasia across space and time.

bellbeakerblogger said...

@Matt

Wilkens, Wrangham and Fitch hypothesized a unified "domestication syndrome". They see a single mutation in mammals that reduce migrating neutral crest cells in fetal development. They hypothesize this retardation acts on many of the characteristics associated with neoteny, such as coloration and behavior.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4096361/

If they're right, then it sounds like at some point you'd be able to create a domestic test(s) for mammals.

https://www.americanscientist.org/blog/the-long-view/the-genes-behind-domestication

olga said...

@Gasca

Right now I don´t know of any genetic research about Pottoka, the basque horse. I´ll do my home work and look in google.In any case is a very good subjet related with human history, the domestication of Pyrenean horses, sheeps and some other animals still in use in the old Baserris (basque farm).

https://www.youtube.com/watch?v=Njya9V0yHek

In this youtube, you may see how similar are Pottokas to the horses painted in Ekain and some other caves.
Slds

JuanRivera said...

Could anyone run f3 analyses using West_Siberia_N in khvalynsk and progress? And using Karelia_EHG? I neither know how to use github nor I have the necessary equipment.

Davidski said...

@JuanRivera

The strength, but also weakness, of the f3 mixture test (or qp3pop) is that it doesn't need the real mixture sources to test whether the target pop is of mixed origin. It can use much more distant proxies for the real mixture sources.

And guess what? Although West_Siberia_N is clearly distinct from Karelia_HG (and EHG in general), it's similar enough so that it can be used successfully with qp3pop as the proxy for the mixture source for Khvalynsk_En, Progress_En, and Vonyuchka_En.

So the only way to test with formal stats if these Eneolithic steppe groups carry ancestry closely related to West_Siberia_N, that isn't simply EHG, is with qpAdm.

As you can see below, Khvalynsk_En, Progress_En, and Vonyuchka_En can all be modeled with minor West_Siberia_N ancestry, but the standard errors are either as high or higher than the admixture estimates. So what this suggests to me is that they might have very minor West_Siberia_N ancestry, but then again they might not have any, and it's possible that all we really need here are better mixture proxies than CHG and Karelia_HG, perhaps with a little more ANE?

RUS_Progress_En
GEO_CHG 0.542±0.020
RUS_Karelia_HG 0.383±0.041
RUS_West_Siberia_N 0.075±0.041
chisq 8.039
Tail prob 0.530245
Full output

RUS_Vonyuchka_En
GEO_CHG 0.608±0.026
RUS_Karelia_HG 0.345±0.053
RUS_West_Siberia_N 0.047±0.052
chisq 13.582
Tail prob 0.137998
Full output

RUS_Khvalynsk_En
GEO_CHG 0.226±0.021
RUS_Karelia_HG 0.763±0.045
RUS_West_Siberia_N 0.011±0.047
chisq 13.481
Tail prob 0.142016
Full output

Also, it's important to note that the lack of proximate mixture source seems to be usually a bigger problem for the Global25/nMonte method. So in these sorts of relatively distal models of deep ancestry it's likely to perform worse than qpAdm. That's probably because Global25/nMonte needs more recent drift to latch onto to perform at its best.

Bob Floy said...

@David

Which paper is Progress_En described in?

Davidski said...

Wang et al...

The tracer dye

Bob Floy said...

Thanks.

GoldieZP said...

@ Davidski I do not see where I can email you for your help. After viewing my mother's Eurogenes EUtest V2 K15 Admixture Proportions chart, I sought you out here. I came across your page http://bga101.blogspot.com/2018/03/if-youre-using-my-tools-to-find-jewish.html

and saw the Global25 test, which I will take. My only DNA test for my mother, sister, aunt and myself was through myheritage, which I uploaded to genesisgedmatch. I showed no Jewish DNA, and I was told my mother's 1.9% was noise. HOWEVER, my matches are with cousins (that are placed with connection on my tree) with Jewish DNA of close to 50% a Levy family from Poland where my ethnicity takes me (Poland, Ukraine, Romania Austria, Germany, Denmark, Finland, Norway,Sweden, United Kingdom) also with the surname Cohen which I have not made connection on tree as of yet.

With all that said, can you take a look at my kit number of me and my mother because on genesis.gedmatch I see Jewish percentages that are very sizeable. But alas, I am just now starting to learn te ropes from the steps you laid out for your readers to dig deeper. Can you be of assistance now just to clarify what I am looking at on gedmatch when I select "jew" thanks much

Carlos Aramayo said...

@Davidski,

Do you have some comment on this new paper?


Jeong et al. 2019. "The genetic history of admixture across inner Eurasia"

Abstract

The indigenous populations of inner Eurasia—a huge geographic region covering the central Eurasian steppe and the northern Eurasian taiga and tundra—harbour tremendous diversity in their genes, cultures and languages. In this study, we report novel genome-wide data for 763 individuals from Armenia, Georgia, Kazakhstan, Moldova, Mongolia, Russia, Tajikistan, Ukraine and Uzbekistan. We furthermore report additional damage-reduced genome-wide data of two previously published individuals from the Eneolithic Botai culture in Kazakhstan (~5,400 bp). We find that present-day inner Eurasian populations are structured into three distinct admixture clines stretching between various western and eastern Eurasian ancestries, mirroring geography. The Botai and more recent ancient genomes from Siberia show a decrease in contributions from so-called ‘ancient North Eurasian’ ancestry over time, which is detectable only in the northern-most ‘forest-tundra’ cline. The intermediate ‘steppe-forest’ cline descends from the Late Bronze Age steppe ancestries, while the ‘southern steppe’ cline further to the south shows a strong West/South Asian influence. Ancient genomes suggest a northward spread of the southern steppe cline in Central Asia during the first millennium bc. Finally, the genetic structure of Caucasus populations highlights a role of the Caucasus Mountains as a barrier to gene flow and suggests a post-Neolithic gene flow into North Caucasus populations from the steppe.


https://www.nature.com/articles/s41559-019-0878-2

Andrzejewski said...

So perhaps the ANE-like contribution in modern Caucasus pops dates to post-Neolithic BA Steppe migrations?

Simon_W said...

@GoldieZP

I fear Davidski no longer cares about what's goin on Gedmatch. I think if you're looking for a very decent test assessing your ethnic origins within the last ~500 years, then 23andme is going to be a good choice. They also test for Ashkenazi Jewish ancestry.

Their biggest flaw IMHO is that they no longer phase parents on their children, because, allegedly they found that this doesn't increase the "resolution" of the results, lol. Whatever this means, phasing obviously increases the reliability and accuracy of their results, but they want to skip this step to save time and money and they have hoped their customers wouldn't notice.


But for people who can get phased on one of their parents, i.e. if they still have a parent around who can get tested as well, this is no problem.

GoldieZP said...

@Simon_W Thank you I will look into this. Do look at the pics of myheritage matches with Cohen and Levi. Interesting!

Davidski said...

@Carlos

We've discussed the prerpint here...

More Botai genomes (Jeong et al. 2018 preprint)

Davidski said...

@GoldieZP

See here...

If you're using my tools to find Jewish ancestry please read this

Matt said...

@Davidski, OT, is it easy for you to run up a Fst matrix on the new ancient / HO dataset? Not much going on at the moment, so I would like to have a quick look and see if anything else going on in those that didn't show up in G25.

Davidski said...

@Matt

Here's an Fst matrix based on the dataset that I'm now using for all Admixtools related stuff.

https://drive.google.com/open?id=1XDLLioGzCxQp-ePlnbH3ttTIiAFxcuLM

Matt said...

@Davidski, thanks. That set doesn't seem to work too well though - the numbers for between modern population look normal, but it looks like the ancients are all hugely inflated (Anatolia_Isparta_EBA-Abkhasian : 0.217, Anatolia_Barcin_N-Anatolia_Boncuklu_N : 0.482 vs Papuan-Yoruba : 0.243). Is this a matter of setting a flag or something?

Davidski said...

OK, hang on, let me try it with inbreed: YES.

Davidski said...

@Matt

Try this. Same link...

https://drive.google.com/open?id=1XDLLioGzCxQp-ePlnbH3ttTIiAFxcuLM

EastPole said...

horse paper:

https://www.cell.com/cell/fulltext/S0092-8674(19)30384-8

Ric Hern said...

So Icelandic Horse was basically the Pre-Modern look....

EastPole said...

The main conclusion of this paper is:

"the origins of the modern domestic horse remain open"

capra internetensis said...

Thanks Eastpole

Ric Hern said...

So I think that Cave Painting at Kapova Cave, Southern Urals makes sense. Looks like todays large ponies or small horses in the range of Bashkirs, Icelandics and Highlands and some Mountain Horses of Iberia and the Balkans....

EastPole said...

But there is hope for future work:

"Future work must focus on mapping genomic affinities in
the 3rd and 4th mill. BCE, especially at other candidate regions
for early domestication in the Pontic-Caspian (Anthony, 2007)
and Anatolia (Arbuckle, 2012; Benecke, 2006)."

Davidski said...

So Sintashta horses (Halvai & Sintashta) are the most basal horses within the modern domesticated clade that lack significant Native Iberian (IBE) admix?

Dunaujvaros Duk2 and ElsVilars UE4618 both have significant IBE-related admix, so that's why they're more basal than Sintashta horses.

It's very interesting to see IBE-related admix in Duk2, a horse from the Carpathian Basin. This has got to be linked in some way to the Bell Beakers. See page 10 of the paper.

EastPole said...

David Reich talks about the insights ancient DNA provides into the spread of languages at a seminar at Berkeley:


"DR: spread of steppe ancestry to the east likely a result of spread of yamnaya descendents, since yamnaya were dead. High genetic similarity to corded ware people."


https://twitter.com/jgschraiber/status/1124043200677675008

Davidski said...

What's the phylogenetic position of Altata_NB31_Neolithic?

That's a Neolithic horse from the Pontic-Caspian steppe.

Dragos said...

Crap coverage

Dragos said...

This is actually funny . Carlos has figured out that DUK-2 from Hungary spoke NW Indo-aeuropean
https://indo-european.eu/2019/05/yamna-the-likely-source-of-the-modern-horse-domesticates-the-closest-sample-yet-from-bell-beaker-hungary/

He should probably just stick with his story -writing
https://indo-european.eu/proto-indo-european-and-proto-uralic-dialects-and-peoples/

Davidski said...

Of course, the problem with the theory that DUK-2 was an Indo-European speaker are twofold...

- he was a horse, so he didn't speak anything

- he had admixture from native Iberian horses, which is lacking in the modern domesticated horse lineage that spread with attested Indo-European speakers

Ric Hern said...

How can it be that Scandinavian Seafarers spread the North European Horse while Pre-Viking Pictish horses of this group were already in Britain before any Scandinavian migration ?

Huck Finn said...

It's pretty interesting that Treemix in Fig 3 connects the horses from Iron Age Ridala fortified settlement in Saaremaa island to the horses of a Pazyryk burial in Olon Kurin Gol of Mongolian Altay. If the result is to be trusted, some horse lineages were apparently brought from Ural area to Baltic area by the migrating Uralic speakers?

Besides, after looking at Jeong et al it seems that not a lot has changed since Wong et al of 2017. Khanty/Mansi minus Evenki is apparently the "Nganasan related meta population" which seem to connect Uralic speakers: a ANE/WSHG/EHG/WHG-heavy population which as such does not exist any more. Apparently the same model applies to Nganasan: Ngagasan minus Evenki is the Proto Uralic group, which explains the Uralic language of Nganasans.

According to unpublished but already shown results fex Iron Age Finns seem to have harboured lots of that HG-based genetic heritage, apparently coming from the areas next to Ural mountains, Western Siberia in the neigbourhood probably being also included.

Matt said...

Catching up on comments:

@Bell Beaker Blogger: Yep, in theory you're right, and they could develop such a test if there's a clear signal of domestication (especially a trans-species signal). Difficulty would be identifying the right locii, and then having coverage of them in an ancient set, having a good sample count to establish differences statistically, and it being likely that the ancient samples are not necessarily fully domesticated at early points.

@Davidski, thanks, yes inbreed=YES fst works better. Haven't found anything new yet and had to prune set quite a lot as still lots of samples with unusual high fst due to low / weak coverage, errors, etc. Using a limited cut down subset of PCoA: https://imgur.com/a/VmjXIL2

@Davidski:What's the phylogenetic position of Altata_NB31_Neolithic?

None, in any of the autosom. trees. Looking at presence of some in uniparental trees (and f3 matrix, etc), are some of them from their ENA list y / mtdna only? Really unlucky for them to only get mtdna out of the Copper Age / Neolithic specimens that would be likely to tell us most about early domestication. Poor guys, probably not what they aimed to do when putting together the study. The mtdna tree doesn't seem to tell a particularly clear story of branching in horse groups.

Davidski said...

@Huck Finn

Is there any relationship between the Iron Age Ridala site and the Tarand Grave site that has produced the earliest N1c so far in the East Baltic?

Huck Finn said...

It seems to me that "BC" or something is missing in the horse study, concerning the timing of Ridala, as Ridala apparently is a Bronze Age/early Iron Age Fortified settlement, see fex here, "Estonian Archeology 3" , p. 65:

http://www.oapen.org/search?identifier=423939

Kunda tarands on the other hand apparently are those of Kunda Hiiemägi, see fex here, p. 13:

http://kirj.ee/public/Archaeology/2016/issue_1/arch-2016-1-3-32.pdf

Timingwise, both Ridala and Kunda therefore are from the same period, late Bronze Age-Early Iron Age. Both Ridala Horse ("Pazyryk connection", i.e. Scythian connetion related to Ural area?) and men in the Tarands (isotope wise not locals, N1c) have foreign features.

However, even if tarands have been found in Saaremaa island, no Tarands with or without N1c, have been found inside Ridala settlement itself. It might not have been impossible, as sort of similar mortuary houses have been found in Dyakovo fortified settlements of the Oka area. I'm of course personally convinced that both Kunda burial and Ridala settlement represent the same immigration wave from the Upper Volga area.

Davidski said...

@Huck Finn

Do you know which graves in this paper the remains came from that had their Y-haplogroups tested?

http://kirj.ee/public/Archaeology/2016/issue_1/arch-2016-1-3-32.pdf

Huck Finn said...

The two guys with a foreign isotype, which also have N1c, are from Kunda Hiiemägi tarand burial.

Davidski said...

Are the burials with the R1a samples described in that paper?

Huck Finn said...

Saag & Tambets have told, in an abstract, if I recall it right, that three out of five tarand samples they have for the time being are N1c. It may therefore be that the two non-N1c-samples are R1a and possibly/probably also from Kunda tarands. However, that remains to be seen. At least I don't have that information.

Research groups both in Estonia and in Finland have been leaking information during the last couple of years but the studies themselves are unluckily still in the pipeline.

Davidski said...

Do you have the link for the Youtube clip outlining the preliminary aDNA results for the remains from the stone-cist and tarand graves?

Huck Finn said...

Youtube clip? No, I don't, unluckily.

Davidski said...

I found it...

https://www.youtube.com/watch?v=0x_wUQKvosM&feature=youtu.be&t=6515

Andrzejewski said...

Is it likely that Okunevo Culture in the proximity of the Altai Mountains were displaced Botai? Would the fact that Przawelaki horses could be found in Mongolia lend credence to this theory? What about Okunevo sharing Botai’s both ANE and ENA ancestries (similar to American Indians BTW)?

Davidski said...

@Huck Finn

Did you get anything new out of that Youtube clip?

Huck Finn said...

No, I didn't and actually I thought that there was some other clip than this one now in Youtube. However, based on that what I've seen, some time in the late Bronze Age/early iron Age WSHG?/EHG/WHG-biased N1c seem to come to the Baltic area, maybe from Upper Volga area. Even in Finland there are still local biases: one maybe related to Scandinavian Bronze Age (more Farmer?), one based on Leväluhta type of Pre Saami people (more BHG type of Siberia?) and one somehow biased towards ANE based HG-groups. I'd guess that the last one is the original West Uralic feature, but that remains to be seen.

Grey said...

"So Icelandic Horse was basically the Pre-Modern look...."

shetland ponies?

https://www.youtube.com/watch?v=yZdsf_uyWnA

Grey said...

missed a bit

dunno if true but at 3:00 in to the video i just linked it's said shetlands are "pound for pound" the strongest horse and can pull twice their body weight.

«Oldest ‹Older   1 – 200 of 201   Newer› Newest»