search this blog

Saturday, January 20, 2018

The case of Chalcolithic fortresses in the Northwestern Caucasus (Kozintsev 2017)


It's a pity that we still don't have any decent ancient DNA data from the North Caucasus and nearby steppes, apart from, of course, those few intriguing mitochondrial genomes from Maykop burials (see here). This leaves us guessing about the genetic origins of the people who lived in this region across the millennia, and thus their genealogical relationships to near and far ancient and modern-day populations, which might eventually prove pivotal in the search for the Proto-Indo-European homeland.

The most nagging questions to be solved are whether Yamnaya, and other closely related Eneolithic/Bronze Age steppe herder groups, sourced the greater part of their so called southern ancestry from the North Caucasus, and if so, from who exactly: groups indigenous to the region, or mixed populations with significant ancestry from, say, Transcaucasia (the Southern Caucasus) or even Mesopotamia?

To make matters worse, the archeology of the North Caucasus is fairly poorly understood. It's generally assumed that there was indeed a colonization of the Northwestern Caucasus by various peoples from the south, including Uruk migrants from Mesopotamia. But even if so, did they leave a lasting impact on the populations of the Caucasus and, subsequently, the steppes? Despite some strong opinions on the matter, particularly in the comments at this blog, no one can say for sure at this stage.

However, as far as I can see, a fascinating new archaeological paper by A.G. Kozintsev in Archaeology, Ethnology & Anthropology of Eurasia suggests that one such group of southern migrants, who built a fortress at Meshoko, in what is now Southern Russia, during the Chalcolithic, was overrun by people more culturally "archaic" and indigenous to the region. If true, and this wasn't an isolated incident, then for obvious reasons it might help to explain the lack of Mesopotamian- and South Caspian-specific uniparental markers amongst the Eneolithic/Bronze Age steppe herder groups, which is an issue that I have discussed at length in the past (see here, here and here). Below is the abstract from Kozintsev's paper. Emphasis is mine:

A multivariate method for assessing cultural changes at stratified sites is proposed. The variables are technological properties of ceramics, and occurrences of various categories of flint implements. The method is applied to stratigraphic sequences of Chalcolithic fortresses in the northwestern Caucasus dating to the late 5th–early 4th millennia BC: Meshoko and Yasenova Polyana. The properties of ceramics include hardness (assessed on the Mohs scale), wall thickness, and frequency of fragments tempered with calcium carbonate. For Meshoko, S.M. Ostashinsky’s data on the occurrence of implements made of high-quality colored flint, splintered pieces, and the total number of segments, points, inserts, scrapers, and perforators were used as well. Each parameter undergoes regular changes from the lower to the upper units of the sequence: ceramics progressively deteriorate, whereas flint industry becomes more and more sophisticated. These changes occur in parallel. Data were subjected to principal component analysis. The first principal component is regarded as a generalized measure of cultural change. The results support the view of the excavators: changes were caused by the interaction of two cultures differing in origin. The earlier culture, associated with the constructors of the Meshoko fortress, shows no local roots, and was evidently introduced from Transcaucasia. The one that replaced it was significantly more archaic (a few copper tools notwithstanding), and reveals local Neolithic roots. It alone can be termed the culture of ceramics with interiorpunched node decoration. The ceramics of Yasenova Polyana, too, indicate cultural heterogeneity and two occupation stages; but cultural changes are more complicated there, probably because the site existed longer, and more than two cultural components were involved.

A.G. Kozintsev, A Generalized Assessment of Cultural Changes at Stratified Sites: The Case of Chalcolithic Fortresses in the Northwestern Caucasus, Archaeology, Ethnology & Anthropology of Eurasia 45 (1) 2017, DOI: 10.17746/1563-0110.2017.45.1.062-075

See also...

Steppe Maykop: a buffer zone?

Genetic borders are usually linguistic borders too

On the genetic prehistory of the Greater Caucasus (Wang et al. 2018 preprint)

116 comments:

Alberto said...

Yes, this too supports the idea that the North Caucasus (pre-Maykop) is probably going to show a good amount of Ukraine_Neolithic admixture (together with CHG and some AN/EEF), making them look close to those outliers from the Balkans chalcolithic.

But we'll have to wait to see some aDNA to really know.

a said...

Speculating that one group is more advanced than another is futile, as is speculating about IQ and winning battles.
Pca's combined with genetic history can give us better clues into history. As noted best fits for Yamnaya are pulling toward [Balkan-West Europe] Iron Gates. The same regional overlap with Peștera cu Oase" (The Cave with Bones) and the human mandible as "Oase 1"can be found.We can speculate that the same region we find stage 4 Neaderthal attributes larger cranial capacity- larger round eyes- receding chin- stockier build; introgressive hybridization between European Naderthals and modern day Europeans took place.
On the other end of the pca- Near East and African Neaderthal introgressive hybridization models took place with a different set of populations.

Rob said...

@ Dave

" the archeology of the North Caucasus is fairly poorly understood."

Dave you should speak for yourself. Its complex but not poorly understood. The archaeologists studying the region understand it well enough. You should understand it too, so when the data comes, you can make even better models to help qualify and quantify the the details.

@ Alberto
I agree. The links in this pre-Majkop fort phase are are both southern and with the North Black Sea region. The Majkop period looks quite different, it has new sets of southern and this time Central Asia links. No northern influences are present, but instead Majkop influences appear in the steppe and Balkans during the second half of Majkop.

Davidski said...

@Alberto

I think those natives "overrunning" the Meshoko and possibly other fortresses will be mostly CHG with some EHG.

@Rob

Maykop is irrelevant genetically to the formation of Sredny Stog, Khvalynsk and probably even Yamnaya, for starters because of those weird mitogenomes they had.

Rob said...

@ Dave

“Maykop is irrelevant genetically to the formation of Sredny Stog, ”

You’re just trying to debate for the heck of it , but end up making no sense
Majkop is 700 Years later that Sredny Stog , which is why nobody has suggested what you claimed !

Davidski said...

@Rob

Sredny Stog and Maykop overlap in some chronologies. It depends which dates one has most faith in.

Grey said...

fort apache

Davidski said...

Ah, crap, I need to proof read my posts a little better.

Changes to title and text made.

Matt said...

So "local Neolithic roots" = pottery using eneolithic groups? (As per Russian tradition where the advent of the Neolithic is pottery).

AWood said...

The old group arrived from the south Caucasus and had more advanced metallurgy, and was overrun by local neolithic groups who, by the sounds of it, were local in origin to the Caucasus.

This reads to me lots of J2a, G2a, J1 in the latter group as there are very distinct isolates who have been in the Caucasus a very long time.

It's possible the older fortress was built by migrants from the south who just so happened to be from other unique branches of the same macrogroups J2, J1, and G2. Who really knows without testing.

All that is certain is that R1b is not native to the Caucasus based on many studies, and really only shows up in a few groups in the region, most of which are IE speaking, or connected to the steppe in some way (ie: Azerbaijan)

Alberto said...

@Davidski

No, it seems you understood it the other way around. The earlier layers represent a culture from the south:

"Most researchers have noticed that the culture of Meshoko includes two very dissimilar cultural components. One of them, represented in the lower units of the stratigraphic sequence, is likely of southern origin, as mentioned both by the excavators and by other specialists (see, e.g., (Andreyeva, 1977: 44; Trifonov, 2001)). Being earlier than Maikop, this culture anticipates it in a sense. Like Maikop, it reveals no local roots."

These should be a CHG-heavy population from Transcaucasia.

And then a more "archaic" culture came in. The "native" culture is related to the North Pontic region and through it to the Balkans. So these "natives" would be a Ukraine_Neolithic-heavy population (maybe with some AN/EEF admixture from the Balkans - bringing copper with them).

Read the paper and the one I linked in the previous thread. Archaeologically there seems to be agreement. We now need to get aDNA to see if that matches too.

Salden said...

https://www.biorxiv.org/content/early/2018/01/19/250688

Mitogenomic data indicate admixture components of Asian Hun and Srubnaya origin in the Hungarian Conquerors

Endre Neparaczki, Zoltan Maroti, Tibor Kalmar, Klaudia Kocsy, Kitti Maar, Peter Bihari, Istvan Nagy, Erzsebet Fothi, Ildiko Pap, Agnes Kustar, Gyorgy Palfi, Istvan Rasko, Albert Zink, Tibor Torok

doi: https://doi.org/10.1101/250688

Abstract

It has been widely accepted that the Finno-Ugric Hungarian language, originated from proto Uralic people, was brought into the Carpathian Basin by the Hungarian Conquerors. From the middle of the 19th century this view prevailed against the deep-rooted Hungarian Hun tradition, maintained in folk memory as well as in Hungarian and foreign written medieval sources, which claimed that Hungarians were kinsfolk of the Huns. In order to shed light on the genetic origin of the Conquerors we sequenced 102 mitogenomes from early Conqueror cemeteries and compared them to sequences of all available databases. We applied novel population genetic algorithms, named Shared Haplogroup Distance and MITOMIX, to reveal past admixture of maternal lineages. Phylogenetic and population genetic analysis indicated that more than one third of the Conqueror maternal lineages were derived from Central-Inner Asia and their most probable ultimate sources were the Asian Huns. The rest of the lineages most likely originated from the Bronze Age Potapovka-Poltavka-Srubnaya cultures of the Pontic-Caspian steppe, which area was part of the later European Hun empire. Our data give support to the Hungarian Hun tradition and provides indirect evidence for the genetic connection between Asian and European Huns. Available data imply that the Conquerors did not have a major contribution to the gene pool of the Carpathian Basin, raising doubts about the Conqueror origin of Hungarian language.

MomOfZoha said...

It's too bad that there are no Ubykh speaking individuals remaining, nor any studies on Ubykh specific DNA, thanks to Russian purges followed by Turkish assimilation. So all I have to go off is my mom's third closest DNA match, which happens to be an Ubykh individual (following two unrelated Iraqi Kurds). Luckily, Mr. Ubykh did mtDNA and Y-DNA tests too and is y-hg G-FGC6669 with mtDNA K2a9... A combination eerily reminiscent of Iranian Late Neolithic remains (IDI1671)...

The thing about Ubykhs is that they were possibly the only "semi-nomadic, horseback" Circassian people, speakers of a now extinct Northwest Caucasian language. According to the Caucasus wikia, their "language still contains a finely differentiated vocabulary related to horses and tack."

And, this particular Ubykh person who knows his lineage inside out (they have kept track of several generations) just happens to have maternal and paternal haplogroups similar to a late Neolithic Iranian sample...

Perhaps, Chad the Pithy Viking will emerge victorious at the end of the day, after many more samples arise...

Shahanshah of Persia said...

David, wouldn't these samples be similar to Caucasus Hunter Gatherers?

Rob said...

The new tables in R1b project are interesting

Villabruna is now P297*
Khvalynsk is same as Kalavan Cave branch P389 xP297
El Partalon still the earliest M269

So it seems at some point R1b-pre-M269’migrated to the steppe

Rob said...

Alberto
Exactly . The abstract dave quoted is confusing because the Black Sea shell tempered pottery isn’t “native” to the Caucaus either - it arrives in 4500 BC with the Balkan - Sredny Stog exchange route (basically Gimbutas wave I in reverse). There was nothing in the north Caucasus before 4500 BC- still an Archaeological blank


Davidski said...

@Alberto

No, it seems you understood it the other way around. The earlier layers represent a culture from the south.

The earliest layers at the fortress site are indeed from the south, and that's because southern migrants built the fortress.

However, the fortress was apparently overrun by more arhaic groups with links to the local Northwest Caucasian Neolithic.

So which group do you think was more CHG-rich? The archaic, Neolithic derived locals, or the migrants from the south?

I'd say the locals were basically a mix of CHG and EHG, while the migrants from the south were more like Armenians, no?

Davidski said...

@MomOfZoha

And, this particular Ubykh person who knows his lineage inside out (they have kept track of several generations) just happens to have maternal and paternal haplogroups similar to a late Neolithic Iranian sample...

Take a look at the links in my post in regards to the sharing of uniparental markers between ancient Iran farmers and Bronze Age steppe groups.

Rob said...

Dave
The “locals” are Sredny Stog. In other words, not “local “ to the north Caucasus at all.
There’s *nothing* in the north Caucasus before 4500 BC
So as I said, those “local pottery groups” will be EHG-WHG , I2 and R1 folk

Davidski said...

@Rob

Doesn't sound like anything related to Sredny Stog.

"The one that replaced it was significantly more archaic (a few copper tools notwithstanding), and reveals local Neolithic roots."

Sounds more like something related to Prikaspiiskaya culture, no?

http://eurogenes.blogspot.com/2017/09/two-starkly-different-neolithic.html

Davidski said...

And indeed, Kozintsev makes a distinction between the "natives" of the Northwestern Caucasus and the steppe.

"Whereas the people who built the Chalcolithic fortresses in the southern Kuban drainage might be regarded as the first wave of migrants from the south, the Maikop people were the second wave. Like their predecessors, they had to maintain complex relationships with the natives of the northwestern Caucasus and with the steppe tribes—relationships that we don’t yet understand properly."

Of course, those Caucasus "natives" may have actually come from farming communities of the Lower Don, but whatever.

Like I said in the Prikaspiiskaya culture thread, there were two different Neolithic traditions on the steppe before Meshoko and Maikop, and both couldn't have originated in EHG populations. One was from the south, and probably very heavy in CHG.

Rob said...

@ Dave

Let's start with basics - there are no real 'Natives' in NW Caucasus. https://i.imgur.com/LvLDKKd.png
The earliest occupation in North /NW Caucasus forts begins in 4500 BC which is Copper Age, or perhaps 5000 BC at best (Late Neolithic).

The 'native Neolithic tradition' Kozintsev refers to is just Ukraine- Mariupol like people drifting south toward the Kuban
https://www.e-anthropology.com/Katalog/Arheologia/STM_DWL_ZkpP_5IXdtVbpqWaz.aspx

The earliest strata of the forts he attributes to 'southerners' - whom he attributes a certain type of ceramic and flint industry. 2 or 3 levels later appears ceramics with shell temper and internal punched node decoration, with clear analogies to steppe ceramics, Cucuteni C ware, etc.

So Meshoko (pre-Majkop) is a fusion of southerners (whom other Russian archaeologists - eg Trifonov - feel confident enough to link with the Darkveti culture in Georgia) and Black Sea Mariupol like people.

The north caspian scenario is different - a different set of '2 traditions'. The first Neolithic' tradition is simply the adoption/ local adaptation of pottery by local foragers. The second is the adoption of productive economy - domesticates which Vybornov suggestively links to North Caucasus or lower Don.

Chad Rohlfsen said...

Since the conversation has gone over to here, I'll just leave these here. The first graph doesn't involve Ust-Ishim, or BEu. The fit is less than desired and doesn't want to be improved.

https://drive.google.com/file/d/1qlEpkf6Fa_aZap_reJWTbYhbALKczxIS/view?usp=sharing

The second involves Ust-Ishim and BEu. Both are based on over 109k transversion sites. This second on has a pretty excellent fit and makes the Yamnaya about half Khavalynsk, and the rest a pretty even mix of Armenia_EBA and CHG.

https://drive.google.com/file/d/1dByIP5ougseDO8Pq52r_YqVfU2wAM31f/view?usp=sharing

If Armenia_ChL is more your flavor, here you go.

https://drive.google.com/file/d/19TsLLuHDy3ZiDQHn9DQMir3pvHwKE-Mt/view?usp=sharing

Davidski said...

@Chad

What you're demonstrating here is that your models don't have the power to discriminate against the Chalcolithic/Bronze Age West Asians.

My models do have the power because they clearly discriminate against them. Simple as that. It makes no difference what our opinions are about this or that sample as an outgroup.

So what you need to explain is why my models discriminate against the West Asians?

Davidski said...

Sure Rob, let's start with the basics:

- there were two different Neolithic traditions on the steppe, one a pottery Neolithic practiced by EHG foragers and the other a real producing economy Neolithic with links to the Lower Don and West Asia

- thus, even though Meshoko and Maikop people were migrants from the south to the North Caucasus, a southern Neolithic population already existed on the steppe

- it appears that Meshoko migrants were overrun or at least subsumed by this southern Neolithic-derived population

- this southern Neolithic-derived population did not come from Sredny Stog, although it may have had contacts with Sredny Stog, see here...

http://eurogenes.blogspot.com/2015/12/mixed-marriages-on-early-eneolithic.html

And if you disagree with these points, then can you explain how Sredny Stog acquired significant CHG admixture?

Matt said...

Davidski: Like I said in the Prikaspiiskaya culture thread, there were two different Neolithic traditions on the steppe before Meshoko and Maikop, and both couldn't have originated in EHG populations. One was from the south, and probably very heavy in CHG.

If I'm reading the linked paper correctly Prikaspiiskaya has pottery and tool kit similar to Khavlynsk Culture ("All of these features are similar to the material from Khvalynskaya culture"), who are the later culture on the overlapping territory ("It is located in the same area as Khvalynskaya culture.") and is food producing in the sense that "Domesticated sheep bones were found alongside the bones of wild species". Isn't it possible that they were autosomally pretty much like Khavlynsk Culture (or less CHG like, assuming further admixture over time)?

(What is the cultural / time boundary between the P and K? As far as I can tell from paper Prikaspiiskaya seems largely similar to Khavlynsk in its economy and material culture but lacking chalcolithic artefacts and cultural accretions.. but it's late here).

(Of course autosomally, the Khavlynsk samples are distinguished from EHG by variable levels of CHG like ancestry... but this does not entail that this actually came from a group that was persisting in the NC throughout the hiatus Rob has talked about?).

Chad Rohlfsen said...

How do you figure? I use the groups needed. I'm also matching f3 stats. I'm not wrong here, sorry. A worst z of 1.6. It matches all around.

Davidski said...

@Matt

Yes, Prikaspiiskaya was similar to Khvalynsk, and that's because Khvalynsk was probably a mixture of Prikaspiiskaya and forager (pottery Neolithic) populations.

If so, then obviously Khvalynsk people would have acquired their CHG admix from their Prikaspiiskaya ancestors and their EHG admix from their forager ancestors.

And even though the ratio of EHG to CHG appears to be very high in Khvalynsk overall, it may have been much lower in steppe populations further south, closer to where the Prikaspiiskaya population came from. I think this is the key quote from the linked paper...

The origin of Prikaspiiskaya culture is reckoned to be connected with the Lower Don region. Some migration from Western Asia could also have occurred. Thus, the Prikaspiiskaya sites in the Lower Volga region represent the second Neolithisation model proposed for this area. The model is connected with the appearance of a producing economy in the milieu of Prikaspiiskaya culture.

Davidski said...

@Chad

How do you figure? I use the groups needed. I'm also matching f3 stats. I'm not wrong here, sorry. A worst z of 1.6. It matches all around.

Please explain why my models discriminate against the West Asians.

It's a fact that they do, so why? Any ideas?

Rob said...

Dave youre not understanding and you’re confusing the Volga region with the Caucasus

But here’s a simple sanity check for your theory
If there were already CHG people in the azov area , as you contend, why is there no trace of them in the Dnieper rapid samples as late as 5000 BC ?
Foragers were mobile and had broad contacts
If they existed they’d show up in Samara EHG and Dnieper HG. But they don’t , until 4200 BC.
It means no such group existed

Chad Rohlfsen said...

I think you should use Khavalynsk and not EHG. They were gone from the Volga long before Yamnaya.

Davidski said...

@Chad

Can you answer the question?

Matt said...

@Davidski, the thing is to me the paper basically seems to be describing Prikaspiiskaya as having more or less the same material economy, pottery tradition, tool kit as Khavalynsk. Doesn't seem to be describing as some transplanted Neolithic economy that *had* to fuse with the other material cultures under description to become Khavalynsk.

I mean, it seems like the description of the Prikaspiiskaya is that they already have the same pottery tradition that is diagnostic of pottery Neolithic from southern Russia anyway. It seems like they are describing a pottery Neolithic with elements of domesticated animals that simply replaces the other cultural traditions they speak of.

So why would Khavalynsk need separate EHG-forager ancestors? (No admixture from EHG, Khavalynsk = Prikaspiiskaya+further gene flow from south?).

@Chad, Steppe2Full suggests, if Steppe=EHG, then EHG in Steppe_EMBA=0.86*0.64*0.7)=0.385 (38.5%)?

Drawback of these models seem that although you account for all the structure in the ancient samples in the structure (e.g. for Steppe2Full - Steppe_Eneolithic, Armenia_CHL, Steppe_EMBA, CHG, Iran_CHL, Ust Ishim) there is no use of information on how samples relate to populations outside of the phylogenic mode (e.g. WHG, Barcin_N, Levant_N, etc.) Lots of pre-basal ghosts in this phylogeny.

Davidski said...

@Rob

Dave youre not understanding and you’re confusing the Volga region with the Caucasus

I'm not confusing anything. The Neolithic producing economy in the Volga region is reckoned to be intrusive from the south (Lower Don).

Yes or no?

If there were already CHG people in the azov area , as you contend, why is there no trace of them in the Dnieper rapid samples as late as 5000 BC ?

They may have been too far west or isolated culturally. Obviously, DNA doesn't diffuse and spread via osmosis. It only spreads when people mix.

Foragers were mobile and had broad contacts
If they existed they’d show up in Samara EHG and Dnieper HG. But they don’t , until 4200 BC.
It means no such group existed


You're making too many assumptions here, and ignoring the fact that CHG was present in Sredny Stog.

So archeology and DNA both show that such a group existed.

Davidski said...

@Matt

There were two very different Neolithic traditions in the Lower Volga. One seems to have been intrusive from the south, and the other local to the region.

Khvalynsk looks like a recent mixture of two genetically distinct populations, one intrusive from the south and the other local to the region, because of the variable ratios of CHG and EHG in the three Khvalynsk samples.

Hence, doesn't it make sense to link the CHG admixture in Khvalynsk to the intrusive group from the south, and at least much of the EHG admixture in Khvalynsk to the local population?

Chad Rohlfsen said...

I think it has something to do with Khavalynsk, or just the pops you're using. EHG didn't exist in the Volga valley anymore, so why use it? Using Khavalynsk, f3, my Graphs, and qpAdm agree that Armenians are important. Another thing to consider is that Armenia EBA/ChL and Iran ChL are essentially similar in relationship to WHG and MA1, but different in relationship to Ust-Ishim, making BEu important. So, using Armenia ChL, Armenia_EBA, or Iran_ChL in the same run come up with high errors. One of the three needs to be in the pright to figure it out. Better yet would be to just use the appropriate in time and geography.

If that doesn't answer your questions, perhaps you can clarify what you want me to comment on specifically rather than just discriminating between West Asians.

Here's validation for my models and graph...mix of Armenia and CHG.

result: Steppe_ChL Armenia_EBA Steppe_EMBA -0.002890 0.000902 -3.203 477
result: Steppe_ChL Iran_ChL Steppe_EMBA -0.003186 0.000880 -3.620 449
result: Steppe_ChL CHG Steppe_EMBA -0.004878 0.001094 -4.460 679

See how Armenia and CHG are relatively equal, and so is Iran_ChL? An intermediary pop is responsible. This didn't come from Europe.

My bet; An Armenian-like pop moved into the North Caucasus and mixed with a pop that was a little more CHG-like than Khavalynsk. Nothing fanciful or exotic required. Just matching archaeology.

Davidski said...

@Chad

I think it has something to do with Khavalynsk, or just the pops you're using. EHG didn't exist in the Volga valley anymore, so why use it?

Obviously because we only have three Khvalynsk samples, and they show variable ratios of CHG and EHG. Using such recently admixed populations as references is a big problem because your models will be restricted and likely highly biased due to sampling bias.

In other words, it's more practical at this stage to use CHG and EHG than Khvalynsk.

And by using CHG and EHG, no real assumption is made that pure CHG and EHG groups existed during the Chalcolithic, or even Neolithic. Rather, the assumption made is that the Khvalynsk and other groups that mixed to form Yamnaya were mixtures of CHG and EHG in whatever ratios.

Rob said...

@ Dave

"I'm not confusing anything. The Neolithic producing economy in the Volga region is reckoned to be intrusive from the south (Lower Don).

Yes or no?"

West Dave. The Don is west of the Volga, not south
And this was c. 5000 BC (Late Neolithic, pre-Eneolithic).

"They may have been too far west or isolated culturally. Obviously, DNA doesn't diffuse and spread via osmosis. It only spreads when people mix."

So the entire zone from north Balkans to Latvia to the Urals was a veritable mix of WHG and ANE, R1 and I2, but for some reason the CHG people of the Azov were paralaysed and social outcast and simply remained in their (imaginary) communities ?
I am tempted to give you a point for creativity, but I ended up giving yo minus points.


* "You're making too many assumptions here, and ignoring the fact that CHG was present in Sredny Stog.
"

Indeed. The earliest evidence of CHG admixture in Russia -Urkaine is from the Sredny Stog- Khvalynsk horizon, which dates from 4500 - 3800 BC, and even then patchily (mostly in the haplogroup Q individual).

This is Copper Age, not Mesolithic.


But hey, name a specific site and burial where the crypto CHGs dwelt I ll get it sampled tomorrow ?

Anonymous said...

There's a great ADMIXTURE graph in the Supplement of the paper "Extensive farming in Estonia started through a sex-biased migration from the Steppe"
Go to page 11 (marked as 12).

https://www.biorxiv.org/highwire/filestream/32767/field_highwire_adjunct_files/0/112714-1.pdf

Really, to me there's no mystery regarding the continuation of the CHG-type ancestry in EHG samples, it was slow and constant.

Now, what surprises me is the relative absense of CHG components in Motala_HG, suggesting something strange actually.

Rob said...

@ Dave

" it appears that Meshoko migrants were overrun or at least subsumed by this southern Neolithic-derived population"

You have it backwards. The Meshoko migrants were progressively taken over by northerners.
Geez ...

Davidski said...

@Rob

West Dave. The Don is west of the Volga, not south

So what?

How does this change the fact that Khvalynsk looks like a mixture of two distinct Neolithic cultures, one from the south (Lower Done and West Asia) and one local to the steppes north of the Caspian?

So the entire zone from north Balkans to Latvia to the Urals was a veritable mix of WHG and ANE, R1 and I2, but for some reason the CHG people of the Azov were paralaysed and social outcast?
I am tempted to give you a point for creativity, but I ended up giving yo minus points.


Well Rob, I'd say that if you actually dig hard enough, you'll find that foragers and indeed early farmers were often highly adapted and thus restricted to very specific ecosystems.

Wouldn't this explain the lack of widespread admixture between the steppe-dwelling EHG and, say, the highland dwelling CHG until the Chalolithic/Early Bronze Age?

But hey, name a specific site and burial where the crypto CHGs dwelt?

They weren't crypto CHG, they were overwhelmingly of CHG origin, and they may have lived, and already been mixing with EHG, at several points in the vast area between the Sea of Azov, North Caucasus and the Caspian.

That's what the ancient DNA models and uniparental markers suggest.

Davidski said...

@Rob

You have it backwards. The Meshoko migrants were progressively taken over by northerners.
Geez ...


As I said above, Kozintsev makes a clear distinction between the steppe peoples and Northwest Caucasian natives.

You obviously don't, but don't pretend that you understand Kozintsev's text better than I do.

The one that replaced it was significantly more archaic (a few copper tools notwithstanding), and reveals local Neolithic roots.

Whereas the people who built the Chalcolithic fortresses in the southern Kuban drainage might be regarded as the first wave of migrants from the south, the Maikop people were the second wave. Like their predecessors, they had to maintain complex relationships with the natives of the northwestern Caucasus and with the steppe tribes—relationships that we don’t yet understand properly.

Rob said...
This comment has been removed by the author.
Rob said...

@ Dave

"So what?

How does this change the fact that Khvalynsk looks like a mixture of two distinct Neolithic cultures, one from the south (Lower Done and West Asia) and one local to the steppes north of the Caspian?"

What matters is the chronology. You're attempting to prove a Mesolithic presence of CHG people north of the Caucasus on arguements from the Copper Age. Which is why your suggestions are unconvincing.


* "They weren't crypto CHG, they were overwhelmingly of CHG origin, and they may have lived, and already been mixing with EHG, at several points in the vast area between the Sea of Azov, North Caucasus and the Caspian.'

Again, please name the culture and individual sites where these secluded CHG foragers lived ?
Your only hope is Crimea. I admit that a possibility

"That's what the ancient DNA models and uniparental markers suggest""

No, the genetic data suggests that CHG started creeping in after 4500 BC, and only significantly in 1 individual, who just happens to have an 'exotic haplogroup Q, which possibly came from the Kelteminar culture or proto-Botai. Id bet when he is specifically carbon dated and corrected for reservoir effects, he could date as late as 3800 BC.

Rob said...

Look Dave

Sorry, but just take this in

The North-Eastern Black Sea Cultural Province in the VI - II millennium BC
Viktor Trifonov (Institute for the History of Material Culture St. Petersburg, St. Petersburg)

"The North-Eastern Black Sea Cultural Province of the VI – II millennium BC is a cultural phenomenon in the Western Caucasus which developed in the context of sub-continental contacts. It was one of few a cultural provinces on the Black Sea rim. Before the Ist millennium BC, all of them had predominantly inland rather than coastal or maritime contacts and never formed any kind of Circum-Pontic cultural formations. The history of the NE Black Sea province begins from the end of regional cultural gap between epipaleolithic – mesolithic sites and “neolithic” sites. The farming in the NW Caucasus
was adopted in the relatively short period between the VI – V millennium BC, when the introduction of a new subsistence pattern, based on pig-breeding and gardening was stimulated by climate improvements, and the increasing spread of the deciduous forest from the NE Anatolia to NW Caucasus. That was a secondary colonization of a considerably depopulated area. In the IVth millennium BC the foothills settlers of the Maikop culture and the highlanders of the Darkveti-Meshoko culture coexisted until the end of both traditions and the arising of the Dolmen culture, which was responsible partly for the origin of the Novosvobodnaya aspect and the wide use of dolmens until the end of the IInd millennium BC.'

Davidski said...

@Rob

What matters is the chronology. You're attempting to prove a presence of CHG people north of the Caucasus on arguements from the Copper Age. Which is why your suggestions are unconvincing.

My chronology and indeed geography are just fine. They fit with recent archaeological papers on the topic...

Mixed marriages on the early Eneolithic steppe

Three-way

You should just cut your loses and admit this.

Again, please name the culture and individual sites where these secluded CHG foragers lived ?
Your only hope is Crimea. I admit that a possibility


CHG foragers lived in the Caucasus. CHG-rich Neolithic farmers probably lived in a wider area, as far north as the Lower Don.

There's a map that you might find useful at that second link. Note the Neolithic sites on the Lower Don, just a hop and a skip away from the North Caucasus.

Davidski said...

@Rob

Sorry, but just take this in.

I took it in.

I don't find it convincing because it contradicts other papers on the topic (see my links above) and ancient DNA data.

You can't just pick what you like Rob. You need to look at all of the evidence.

Rob said...

Flawless victory. Fatality

Davidski said...

@Rob

Flawless victory. Fatality

Only in your rich imagination.

But alas I'm only interested in facts and data.

Rob said...

yes I know the papers well, but seems like you need a reminder :

“To conclude, the archaeobotanical investigations conducted by the author have provided evidence that cereal cultivation began in eastern Ukraine and south-west Russia around the second half of the 5th millennium calBC, as demonstrated by the analysis of pottery impressions from the Zanovskoe and Ra- kushechny Yar sites. The earliest evidence of cereal cultivation in eastern Ukraine and southwest Russia comes from the Sredny-Stog culture sites at Zanov- skoe and Rakushechny Yar, where cereals and their chaff impressions were identified by the author in the layers dated to the second half of the 5th to the first half of the 4th millennium calBC. A variety of cereal species were identified from the grain and chaff impressions, comprising hulled and naked barley, spelt wheat, probably flax, and broomcorn millet species.“

Late 5th and early 4th millennium
“Hop skip and jump” away from 7th Millenium Mesolithic ?
Maybe if you’re superman

Arza said...

@ Rob
If
the genetic data suggests that CHG started creeping in after 4500 BC
then what is the the source of J1 Y-DNA in I0211 (5500-5000BCE)?

Davidski said...

@Rob

Have you already forgotten this?

A production economy is a particular feature of the second group of sites (in the Lower Volga), which can be dated to the end of the 6th millenium BC. This is one of the earliest pieces of evidence of the existence of domesticated species in Eastern Europe.

The origin of Prikaspiiskaya culture is reckoned to be connected with the Lower Don region. Some migration from Western Asia could also have occurred. Thus, the Prikaspiiskaya sites in the Lower Volga region represent the second Neolithisation model proposed for this area. The model is connected with the appearance of a producing economy in the milieu of Prikaspiiskaya culture.


http://eurogenes.blogspot.com/2017/09/two-starkly-different-neolithic.html

Wouldn't groups like these from the Lower Don be the source of CHG admixture in both Khvalynsk and Sredny Stog?

Or are you going to keep denying this possibility?

Rob said...

Yes because that date is problematic - too old than it really is . How can it date to 6th century when it’s alleged source- lower Don- only became productive in the 5th ?

And if they were the source of CHG, why is any CHG missing until 4000 BC ?

Davidski said...

@Rob

And if they were the source of CHG, why is any CHG missing until 4000 BC?

CHG is found in Sredny Stog and Khvalynsk samples.

So unless you're proposing that CHG-related people raced to Sredny Stog and Khvalynsk sites around 4,000 BC and the Sredny Stog and Khvalynsk samples we have just happen to be the first on the steppe with CHG ancestry, then it's a safe bet that CHG diffused onto the steppe well before 4000 BC.

It's just common sense.

Rob said...

Arza- yep I mentioned the late glacial expansion of Imeretian culture north to Crimea, north Caucasus , even Moldavia
But it disappeared from the Caucasus at exactly that period (c. 7000 yBp); this Karelian is a remarkable vestige in the extremity

Rob said...

@ Dave

“Race around “?


You seem intent on pretending that you’re unaware that only one of 3 Khvalynsk has significant CHG; meaning the process was just occurring at that very period

Seems you just have some kind of phobia of south Caucasian impact ; leading you to make silly theories which just waste everyone’s time
Anyway- as I said; find me that CHG groups in say 5000 BC . Sample it.

Davidski said...

@Rob

Only one of 3 Khvalynsk has significant CHG.

As per formal models, all three Khvalynsk samples have significant CHG ancestry, as in statistically significant relative to EHG.

And Sredny Stog samples have CHG too.

Seems you just have some kind of phobia of south Caucasian impact ; leading you to make silly theories

There's nothing silly about my analyses and interpretation of the data.

Anyway- as I said; find me that CHG groups in say 5000 BC. Sample it.

Pretty sure they've already been sampled. I'll just sit back and wait a few months, like I did to see the Eneolithic M417 on the steppe, which I also predicted by reading the data correctly, unlike you.

Rob said...

No Dave
I correctly stated that R1a came via the forests and seeped down to the steppe periodically
And CWC isn’t from Yamnaya
Seems you still can’t grasp basic tenets

Davidski said...

@Rob

I correctly stated that R1a came via the forests and seeped down to the steppe periodically

Right, and the first male Eneolithic sample from a major steppe archaeological culture just happens to be R1a.

What are the chances?!

And CWC isn’t from Yamnaya

We don't know that yet, but even if it's not, then R1a-M417 is still from the steppe and from a population very similar and closely related to Yamnaya.

This is what many of us have been saying here for years, and you were arguing against it like it was the strangest thing in the world.

Not so strange now.

Rob said...

You mean Dereivka in the forest steppe ?
What about the R1a in CCC with CHG dating to 3000 BC ?

Davidski said...

@Rob

You mean Dereivka in the forest steppe?

No, I mean the Dereivka site at Alexandria on the steppe. Here's a map. The steppe/forest steppe boundary is clearly marked.

https://i.pinimg.com/originals/57/33/3c/57333c0609d0e31b273fc437db3a83ab.jpg

What about the R1a in CCC with CHG dating to 3000 BC?

Oh crap, you were right all along and obviously Corded Ware derives from CCC north of the steppe.

Ah, no, wait, Corded Ware has a shitload of CHG and uniparental markers from the steppe, not just R1a-M417.

Corded Ware is definitely from the steppe.

Chad Rohlfsen said...

@ David,

"Obviously because we only have three Khvalynsk samples, and they show variable ratios of CHG and EHG. Using such recently admixed populations as references is a big problem because your models will be restricted and likely highly biased due to sampling bias.

In other words, it's more practical at this stage to use CHG and EHG than Khvalynsk."

I could obviously say the same about only having three EHG samples, one of which is drifted towards WHG. Why is it more practical to use EHG and CHG? Neither were around in a pure form. It's like trying to use Anatolia and WHG to describe the transition from Iberia_EN to MN. Why not just use Iberia_EN, which we have samples for? Same logic applies with using Khavalynsk. It works and the statistics I've shown prove that.

Sofia Aurora said...

Guys how can someone purchase the article?
There is no link for downloading!!!
What's all about that?
Why does someone has an online journal without clarifing HOW can we download the bloody articles?

Rob said...

I’m not saying it didnt; but the evidence suggests that the north was the R1a diversification area; and CHG came from the east , at least in part ; because it is found all the way up in CCC

Rob said...

The article is freely available in his academia page

Davidski said...

@Chad

I could obviously say the same about only having three EHG samples, one of which is drifted towards WHG.

You can't. They form a statistically homogeneous population in terms of their ANE/WHG ratios.

Why is it more practical to use EHG and CHG? Neither were around in a pure form.

They were around in mixed form, with highly variable ratios of CHG and EHG, so it makes good sense to use them in unadmixed form to let the algorithm vary the CHG and EHG ratios.

It's like trying to use Anatolia and WHG to describe the transition from Iberia_EN to MN. Why not just use Iberia_EN, which we have samples for?

You can if you want to. I've done it myself. But since Iberia_EN samples are homogeneous then we can use Iberia_EN.

Same logic applies with using Khavalynsk.

It doesn't because Khvalynsk is a recent mixture with highly variable CHG and EHG ratios, so we don't know if the three samples we have average out precisely to resemble the population that was ancestral to Yamnaya. But I'd say that the chances of this are slim.

It works and the statistics I've shown prove that.

Your models show that it works with the outgroups that you used, but my models show that it's much less likely, or in fact even that it doesn't work with other outgroups.

So now you have to prove that it works with the outgroups that I used.

Davidski said...

Actually let me rephrase that last part:

So now you have to prove that it works and works better than Blatterhole_MN, Koros_HG etc. with the outgroups that I used.

Sofia Aurora said...

Thanks Rob

Anthro Survey said...

@David

"It might help to explain the lack of Mesopotamian- and South Caspian-specific uniparental markers amongst the Eneolithic/Bronze Age steppe herder groups, which is an issue that I have discussed at length in the past"

I think it's relatively simple: the efflux of CHGs northwards towards Ukraine from the Caucasus preceded an influx of Ubaid/Uruk-related migrants from northern Mesopotamia who likely brought ANF-related ancestry to the region and some sort of a "proto-urban package". That's probably why CHG is the best stand-in for steppe's southern ancestry in your models, not Iran_Chl or contemporary Georgians or smth.

As to whether this was an isolated incident or not? We can't know for sure, but it's clear, on the whole, that the ANF-like ancestry was there to stay.

Indeed, the appearance of gracile characteristics in the south and north Caucases during the Chalcolithic has been contrasted with earlier presence of more archaic skulls and correlated with an influx of putative southern migrants.
This north-bound movement of people to the Caucasus MAY explain in part why Georgians and N. Caucasians, to a lesser extent, clearly overlap with other MEs in looks yet why steppe-rich(thus CHG-rich) populations like Swedes and Russians do not.
(Subsequent south-bound Kura-Araxes related movements---as far as Palestine, evidenced by Khirbet-Kerak Ware---"homogenized" the ME further, it seems.)

Alberto said...

@Davidski

By the end of the day, no matter how you want to interpret the different groups involved: Southern, Local and North Western (The “pre-Maikop” culture is an essentially Chalcolithic phenomenon related to the western and northern Black Sea region by means of a network of “prestige exchange”. Numerous artefacts, which are diagnostic for this network, are securely dated to the late 5th mill. BC through their presence in Tripolie B1 contexts.), it seems likely that when we get samples from that area ca. 4000 BC, they are going to show a good amount of Ukraine_Neolithic and CHG admixture, and probably some amount of AN/EEF admixture. And that would put them close to the outliers from the east Balkans.

Or at least, this is what we should expect with the current data available from neighbouring regions and the archaeological interpretations. So I still doubt that this population was a major part of Yamnaya. Unless we change Ukraine_Neolithic for EHG (that would be the locals??) and then we could have Yamnaya right there. But that seems unlikely.

Rob said...

@ Alberto:

But even though from the Meso to Neolithic in Ukraine there was a slight EHG-> WHG shift, even some ANF, there'd arguably still be EHG there. Indeed all the I2a2a1b is probably from Ukraine Enelithic. So all we'd need is a slight shift back EHG-ward as part of the package of CHG coming along with that, from wherever exactly this was..

Davidski said...

@Alberto

You always seem to view everything with the hope that in the end the ancient data will show a Bronze Age migration (of a Pathan-like population?) from Central Asia to the Pontic-Caspian steppe.

But that won't happen, because Yamnaya is basically a mixture of Eastern European and Caucasus foragers, and there are obviously good reasons why they're called Eastern European and Caucasus foragers, as opposed to Central Asian foragers.

The North Caucasus and nearby steppe area is a vast region which could potentially throw up some very interesting results.

But chances are that all we'll find there is a CHG-like, albeit more northerly shifted, population during the Late Neolithic/Chalcolithic, and it will prove to be the perfect fit for Sredny Stog, Khvalynsk, Yamnaya etc.

Although I won't be overly surprised if something close to the Yamnaya genotype, but without the Blatterhole_MN-like western shift, is shown to exist as early as the Mesolithic in the North Caucasus.

I refuse to entertain the notion that such a population arrived in Eastern Europe from east of the Caspian Sea, let alone Central Asia, and with R1a-M417 and R1b-M269. You really need to consider seriously that you're going to look very silly after pushing so adamantly such a weak theory for as long as you've been here.

Alberto said...

@Rob

I don't think the shift is so slight. Ukraine Neolithic is around 45% WHG + 45% EHG + 10% CHG.

Yamnaya is about 44% EHG + 44% CHG + 6% AN + 6% WHG.

At 4500-4000 BC, where would we find 2 populations, one being 88% EHG + 12% WHG and another one being 88% CHG and 12% AN? (or similar combinations of those components)? I think it must be east of the Black Sea, because the samples we have from around the Black Sea at that stage have much more WHG and AN ancestry.

@Davidski

You're projecting your own bias onto mine. It's you who have a strong preference here, not me. Anything that disagrees with that preference is someone else's bias.

You disagree with me even when you actually do agree.

I have no preference whether Yamnaya formed in the North Caucasus or the Lower Volga-Ural region, nor if it was further east (or south east). I'm showing the numbers and the possibilities that are realistic, but I'm always careful to stress that only aDNA will tell the truth. i just hope it comes soon, because otherwise these debates get tiring.

Matt said...

@Davidski: Hence, doesn't it make sense to link the CHG admixture in Khvalynsk to the intrusive group from the south, and at least much of the EHG admixture in Khvalynsk to the local population?

I wish there were some material on the intrusive Prikaspiiskaya culture (I can't find any if there is!), because as far as I can tell it is closest to Khavalynsk without copper goods.

It seems to me that effectively Prikaspiiskaya is, here: a group that precedes Khvalynsk, has a similar pottery tradition, toolkit, overlaps the same area, and begins to shows some traces of the domestic animals, and to which Khavalynsk looks to add further signs of influence and interaction from the south (copper).

Just from that I would not conclude from such a description that P is a CHG-heavy culture originating from the North Caucasus. Seems more likely that K should be autosomally similar to the predecessor with further accumulation of CHG ancestry from ongoing interactions with the south (hence some variation in CHG ancestry) while Prikaspiiskaya would be autosomally between CHG and EHG (EHG with a very low CHG fraction, lower than the Khvalynsk samples)...

Matt said...

(Relevant excerpt from "Initial stages of two Neolithisation models": "Until recently, the appearance of a producing economy in the Low Volga was connected with Khvalynskaya culture sites (Karahuduk, Kairshak VI, Kombakte), which are distributed throughout the whole of the steppe zone.

The lithic industry of this culture differs from the flint industry of preceding cultures, as it includes heavy flakes produced by enhanced pressure knapping. Scrapers, knives, points on large blades, and triangular arrowheads with a truncated base predominated. There are also insets on narrow blades. Pottery was made from clay tempered with crushed freshwater mollusk shells. The upper part of the round-bottom vessels is thickened (Fig. 3). The decoration and technique of applying it to vessels differs from that found in the preceding Kairshak and Orlovskaya cultures (Vasiliev 2003).

Domesticated sheep bones were found at Khvalynskaya sites, along with the bones of wild species (kulan, saiga, tarpan). Detailed statistical data can be found in a number of publications (Kuzmina 1988; Vybornov et al. 2015). Khvalynskaya culture sites date to 4900–4600 BC (Tab. 1).

There are two hypotheses about the origin of this culture. It has been suggested that it was an autochthonous culture formed on the basis of the preceding Prikaspiiskaya culture. Other researchers suppose that it could have appeared as the result of the migration of tribes from the southern region of the Trans-Caspian area.

Remains of domesticated animals were also recently found at Prikaspiiskaya sites, preceding Khvalynskaya culture (Vybornov et al. 2015). It is located in the same area as Khvalynskaya culture. Evidence of a quarzite stone industry has been found at Prikaspiiskaya sites (Kurpezhe-molla and Oroshaemoe). Heavy flakes were used as preforms for tools produced by an enhanced pressure technique. Tools included scrapers, knives and points on large blades, ‘fish-form’ arrowheads, as well as insets on narrow flint blades. Flat-bottom pottery was made from clay artificially tempered with shells. The upper parts of the vessels have a thickened external side, called the ‘neck-piece’. Pottery was decorated with comb impressions and traced lines (Fig. 4). All of these features are similar to the material from Khvalynskaya culture (Vasiliev 1981; Vybornov et al. 2015; 2016). Domesticated sheep bones were found alongside the bones of wild species (kulan, saiga, tarpan) (Vybornov et al. 2015). The Prikaspiiskaya sites are dated to 5500–4800 BC (Tab. 1).

This culture is attributed to the Neolithic period, as no copper artefacts have been found. However, no traces of transition from the Kairshakskaya or Orlovskaya cultures to Prikaspiiskaya culture could be identified.

The origin of Prikaspiiskaya culture is reckoned to be connected with the Lower Don region. Some migration from Western Asia could also have occurred.


The description of toolkit and pottery seems to indicates similar subsistence to Khvalynsk and cultural continuity, and only "some migration from Western Asia" is alluded to in the origin. Not like a fully Neolithic food producing pastoralist culture which was later absorbed into a pottery using forager culture to create a hybrid with more emphasis on hunting and foraging. My read is that they represent an introduction of food producing (a new Neolithic form/model), but I can't see that the paper indicates a difference in subsistence is stark enough to lead me towards a transplanted pastoral economy of mostly southern autosomal origin. Obviously I could be wrong and more papers on the topic that I haven't read, if they exist, could flesh this out.)

Anonymous said...

..."Corded Ware is definitely from the steppe."

The Dereivka culture is from the forest step. Corded Ware culture rather was from the forest step also, or border from forest step and forest(?). Lexicon of PIE was forest, not pure steppe.

Grey said...

"I mean, it seems like the description of the Prikaspiiskaya is that they already have the same pottery tradition that is diagnostic of pottery Neolithic from southern Russia anyway. It seems like they are describing a pottery Neolithic with elements of domesticated animals that simply replaces the other cultural traditions they speak of."

iirc Ertobolle culture was a sedentary pottery using wetlands forager culture which added animal husbandry as a result of contact with LBK or Funnelbeaker farmers - which unlike other foragers they could do relatively easily as they were pre-adapted for farming (through already being sedentary)

if there was a sedentary pottery using wetlands culture around the shores of the black sea maybe something similar happened and a specifically wetlands forager group merged with a sheep herder group from the south?

Grey said...

also specifically wetlands foragers would by definition have a restricted geographical range around lakes, seas etc

i'd imagine this idea could be tested if there was evidence of the original group having a largely marine based diet?

Anonymous said...

Alberto said: "Ukraine Neolithic is around 45% WHG + 45% EHG + 10% CHG.

Yamnaya is about 44% EHG + 44% CHG + 6% AN + 6% WHG."

It is not true.

Ukraine Neolithic from Mathieson et al. 2017 not even close to this.

My modeling Yamnaya with probability of every coefficients calculating

Yamnaya_Samara
Coefficients:
Estimate Std. Error t value Pr(>|t|)
EHG 0.58412 0.11644 5.016 3.22e-05 ***
CHG 0.41442 0.08392 4.938 3.95e-05 ***
distance 0.001465346

Estimate Std. Error t value Pr(>|t|)
EHG 0.57221 0.11689 4.896 4.88e-05 ***
CHG 0.41098 0.08487 4.843 5.60e-05 ***
WHG 0.01540 0.01503 1.025 0.315
distance 0.00140385

Yamnaya_Samara
Estimate Std. Error t value Pr(>|t|)
EHG 0.58015 0.23227 2.498 0.0198 *
CHG 0.17503 0.26062 0.672 0.5083 (poor)
WHG 0.00000 0.05505 0.000 1.0000
Anatolia_N 0.24376 0.39108 0.623 0.5390 (poor)
distance 0.001062141

Signif. codes: (best) 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 (poor)

Other sets of statistics give the same thing.


Yamnaya_Samara
Coefficients:
Estimate Std. Error t value Pr(>|t|)
EHG 0.59125 0.11764 5.026 3.14e-05 ***
CHG 0.40874 0.08258 4.949 3.84e-05 ***
distance 1.483573e-05

Estimate Std. Error t value Pr(>|t|)
EHG 0.57931 0.11805 4.907 4.74e-05 ***
CHG 0.40596 0.08361 4.856 5.42e-05 ***
WHG 0.01471 0.01387 1.061 0.299
distance 1.416827e-05

EHG 0.58896 0.23263 2.532 0.0183 *
CHG 0.18509 0.26387 0.701 0.4898 (poor)
WHG 0.00000 0.05294 0.000 1.0000
Anatolia_N 0.22594 0.38355 0.589 0.5613 (poor)
distance 1.095934e-05

Signif. codes: (best) 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 (poor)

Yamnaya_Kalmykia
Estimate Std. Error t value Pr(>|t|)
EHG 0.52809 0.10402 5.077 2.75e-05 ***
CHG 0.47191 0.09306 5.071 2.79e-05 ***
distance 3.551554e-06

Yamnaya_Kalmykia
Estimate Std. Error t value Pr(>|t|)
EHG 0.5278879 0.1061858 4.971 4.01e-05 ***
CHG 0.4718602 0.0949014 4.972 4.01e-05 ***
WHG 0.0002484 0.0067867 0.037 0.971
distance 3.551366e-06

Yamnaya_Kalmykia
Estimate Std. Error t value Pr(>|t|)
EHG 0.52776 0.11735 4.498 0.000149 ***
CHG 0.44486 0.12715 3.499 0.001849 **
WHG 0.00000 0.01846 0.000 1.000000
Anatolia_N 0.02737 0.11883 0.230 0.819786
distance 3.495135e-06

Davidski said...

@Matt

My read is that they represent an introduction of food producing (a new Neolithic form/model), but I can't see that the paper indicates a difference in subsistence is stark enough to lead me towards a transplanted pastoral economy of mostly southern autosomal origin.

You seem to be suggesting that Khvalynsk formed when an EHG steppe population somehow acquired a food producing economy related to the Neolithic of West Asia, perhaps along with minor West Asian admixture, and then acquired more West Asian admixture via later contacts with southern groups.

This is possible, I suppose, but it's not very parsimonious in the light of genetic data from the Neolithic from other parts of West Eurasia.

I'd say a much more parsimonious model is that a food producing economy was introduced onto the steppe via a migration of a CHG-rich population from the Caucasus, and Khvalynsk eventually formed when these migrants mixed with an EHG pottery Neolithic population native to the steppe.

I don't think that the similarity between the Prikaspiiskaya culture and Khvalynsk makes the latter scenario less parsimonious than it is based on what generally happened during the Neolithic in Europe.

Matt said...

@Davidski, yep, that's more or less what I'm saying. Although without the "somehow"- it doesn't seem so mysterious or questionable that Prikaspiiskaya could acquire domestic animals for food production without becoming majority CHG, since that's... er... exactly what the Khvalynsk that followed them were, and the vast majority of the material culture was identical between these two cultures which overlapped in time and space, except for the elements that were Chalcolithic and couldn't be.

I can see how you'd say that it is parsimonious to assume an introduction of a Neolithic migration layer which persisted as a separate culture for some time, but then if we were to follow the Neolithic from the European Early Farmer examples, then we'd also have had to assumed the Khvalynsk Culture itself would be expected to be majority CHG in ancestry, with Yamnaya being a resurgence of EHG ancestry....

Rather than the process we actually have in the region, where EHG->Khvalynsk->Yamnaya is an accumulation of more CHG ancestry. I'm not sure if it's more parsimonious to assume an back extrapolation of the trend we already have in place on the steppe (to an intermediate layer between EHG->Khvalynsk), or to assume a model from the EEF which is roughly the opposite.

(In any case this is all a bit of a sidetrack on the topic of whether there even any CHG groups about in the NC during the Mesolithic, or whether that ancestry comes from a southern region where it would've had more opportunity to pick up the Neolithic package).

Davidski said...

@Matt

I can see how you'd say that it is parsimonious to assume an introduction of a Neolithic migration layer which persisted as a separate culture for some time, but then if we were to follow the Neolithic from the European Early Farmer examples, then we'd also have had to assumed the Khvalynsk Culture itself would be expected to be majority CHG in ancestry, with Yamnaya being a resurgence of EHG ancestry....

Based on the fact that a food producing economy arrived in Europe via demic diffusion from Anatolia, I think it's sensible to assume that the early Prikaspiiskaya population was from the Caucasus and thus largely CHG, and it became largely EHG as it turned into Khvalynsk.

Yamnaya is a very different issue, but it may well represent a resurgence of CHG ancestry on the steppe. I don't see a problem with such a scenario, considering that we know that such resurgences in genetic components happened, and there's no rule, as far as I can see, preventing it from being an CHG resurgence in this particular case. Indeed, early indications are that there my have been a resurgence of EEF ancestry in Central Europe during the Iron Age.

Onur Dincer said...

@Davidski

Indeed, early indications are that there my have been a resurgence of EEF ancestry in Central Europe during the Iron Age.

That EEF resurgence may have happened during the Roman and medieval times. Comparisons of Roman British, Anglo-Saxon and modern British genetic results hint at that, modern Brits seem to show some increase in EEF-related ancestry during the late medieval times (due to the Norman French migrations?)

Rob said...

@ Onur

“That EEF resurgence may have happened during the Roman and medieval times. “

Nonsense
Wielbark culture has nothing to do with Rome or the medieval period

Matt said...

Well, I mean, at this point, I guess I'm stuck at point of just wanting to see further how the archaeologists have described it and if they go into those questions. (Was it fully eneolithic, pastoral Neolithic?).

There seems like nothing else in English, so if any Russian readers want to read up on the культура Прикаспийская.

I can put stuff through google translate - https://sites.google.com/site/civilizacium/home/arheologia/prikaspijskaa-kultura but it's not so illuminating.

Though (and this will be obvious to those that, unlike me, were paying much more attention) Pri-kaspiiskaya Culture, just translates as Pre-Caspian Culture. So English language literature references to the "North Caspian Culture" and "Pre-Caspian Culture" referring to the same cultural layer. (And of course Orlovskaya is simply "Orel Culture" and so on, and translations such as this will be generally more fruitful for chasing up English language references).

Cites by Mallory on the "Pre-Caspian culture" (from a fairly random website - http://s155239215.onlinehome.us/turkic/btn_Archeology/Mallory/JMalloryEneolothBronzeAgeEn.htm) as follows - "5000-3000 BC The Pre-Caspian culture to the south" (of the Samara Culture) "is very poorly known with little more than twenty sites identified. These are generally on heavily eroded dune surfaces where material from different periods has been mixed together. The sites are, as a rule, situated along the shores of dry lakes and are composed of flat-based ceramics, quartzite tools and occasionally animal bones. Both the Samara and Pre-Caspian cultures are synchronized with the Dnieper-Donets culture to the west."

A little more on the "North Caspian Culture" from https://en.wikipedia.org/wiki/Samara_culture - "Related sites" (to Samara culture sites) "are Varfolomievka on the Volga (5500 BCE), which was part of the North Caspian culture,[clarification needed] and Mykol'ske, on the Dnieper". Anthony describes the Varfolomievka site here - https://books.google.co.uk/books?id=nLIufwC4szwC&lpg=PA188&ots=DFIm96yU7i&dq=Varfolomievka&pg=PA188#v=onepage&q=Varfolomievka&f=false, but it has multiple layers and I don't have any clear picture from either of these.

Onur Dincer said...

@Rob

Nonsense
Wielbark culture has nothing to do with Rome or the medieval period


Check this:

http://journals.plos.org/plosgenetics/article/figure/image?size=medium&id=info:doi/10.1371/journal.pgen.1006852.g004

The recent EEF increase is even more salient in Scandinavia.

Matt said...

Another paper - https://revije.ff.uni-lj.si/DocumentaPraehistorica/article/download/42.3/5018 - "The origin of farming in the Lower Volga Region" - by Vybornov looks to have some more information on when and where domestication appears in Caspian culture sites. (Not read much of it, though, honestly, so far).

Rob said...

Onur
There’s nothing salient about Wielbark
It is dominated by I1, G2 and I2a2

Onur Dincer said...

@Rob

I was talking based on the autosomal results.

Chad Rohlfsen said...

Why do you use this list? Some redundant stuff here that just kills SNP coverage.

Davidski said...

Take a look at the output I posted: "numsnps used" is always very high. And high enough to clearly discriminate against the West Asian pops.

So what's the problem?

Chad Rohlfsen said...

There's no necessity to use many of them. It's overdone. The results show that. There is no evidence of a CHG pop after 7000BCE. It would be like me trying to force Magdalenian ancestry in everyone and not stopping until I find the cluster of pright that make it work. Your infeasible in the Iran ChL and CHG model falls on CHG not being needed because it's not. I would steer clear of it. Stick with what there is evidence of existing during this time.

Your discrimination between West Asians is only CHG vs something EEF/Anatolian, which is easy enough to accomplish. All those pright do is take up space, and lower SNP counts that make it harder to distinguish closely related West Asians, like Armenia and Iran ChL.

I'm pretty certain if you asked Nick or David if they think some elusive CHG pop is running around in the Late Neolithic spreading genes and pastoralism you'll hear a highly unlikely.

All the pright needs is Mota, Ust-Ishim, MA1-AG3, WHG, Kostenki14, Anatolia_N, Levant_N, Iran_N, and you can add Native Americans or an East Asian if you like. All this Goyet, Vestonice, Elmiron stuff won't help a bit with Steppe pops and it will drop you tens of thousands or even up to a couple hundred thousand of lost SNPs.

Chad Rohlfsen said...

Just give it a try and see.

Davidski said...

@Chad

There's no necessity to use many of them. It's overdone. The results show that.

The results clearly show that they're necessary. Look at my output more carefully.

There is no evidence of a CHG pop after 7000BCE.

Irrelevant, because my models are only implying that Yamnaya has ancestry from a Neolithic population rich in CHG.

This population may have been a mixture of CHG, eastern EEF and even EHG. That's not only possible, it's very likely.

All those pright do is take up space, and lower SNP counts that make it harder to distinguish closely related West Asians, like Armenia and Iran ChL.

Well, Armenia_EBA is feasible, while Iran_ChL isn't, so I don't understand your point.

https://drive.google.com/file/d/1Zt0Y-cGDYV0LpQ1xr7T0Bok4Fe8j3TgS/view?usp=sharing

https://drive.google.com/file/d/1hgh3d50SVxHvPPstifn662T6TtBoBIFD/view?usp=sharing

I'm pretty certain if you asked Nick or David if they think some elusive CHG pop is running around in the Late Neolithic spreading genes and pastoralism you'll hear a highly unlikely.

As per above, this is a straw man.

All this Goyet, Vestonice, Elmiron stuff won't help a bit with Steppe pops and it will drop you tens of thousands or even up to a couple hundred thousand of lost SNPs.

My set up is very strong. There's no need to tweak it to make the West Asians fit better. That would be highly unscientific.

Alberto said...

@Chad

I do agree that probably (from my experience observing and using D-stats based datasheets) all the UP (except MA1) don't add much, except some noise. I think the same about Mota, while other outgroups like Yoruba or Mbuti don't add much noise, but don't add information either. A carefully selected set of outgroups is what works best (but there you have to take into account how sensitive formal stats are to the genetic diversity of each population, so you can make the right balance. For example, the higher diversity of AN and EHG vs. WHG and CHG clearly shifts the proportions towards the former if you use those samples in both pright and pleft).

All that said, I don't see why the output of Davidski's models is doubtful. In the end, most of the information is taken from the pright pops that actually provide that information. And all the models use the same outgroups, so it seems like a fair comparison to me.

Maybe it would be more helpful if you post models that prove your point? For example, 2 models with the following pleft pops:

Yamnaya_Samara
EHG
CHG
Koros_HG
Koros_EN

Yamnaya_Samara
EHG
Iran_Chalcolithic (or Armenia_Chl/BA if you prefer)

with your preferred set of pright pops and show that the second model works better?

(We're interested in knowing the best theoretical model to see the different realistic options about the possible origin of the populations involved in the genesis of Yamnaya, and then we can make assumptions as to whether those populations actually existed in those places or not).

Anonymous said...

The Prikaspiyskaya culture evolves from local Orel culture. The Lower Don culture of Rakushechny Yar culture. There had spread Altatinskiy type having a distinct origin from the forest area, it survived to the Yamnaya time.

Productive economy went into these cultures rather from Central Asia (Jebel/Eastern Caspian) than from the West. This is indicated by the fact that the main animals in these areas were sheep and goats, pottery have not got link with the West. The Caucasus is excluded.

Matt said...

It doesn't seem totally consistent to argue that there's "limited data" from ancient West Asia and at the same time that we should stick with only what is in existence at this time, which is likely to underrepresent the diversity that we could represent with freer mixes of less admixed populations like CHG. The limited sampling is a reason to use CHG.

Re: correct pright populations, the impression I have from Davidski's recent PCA run on the Martiniano data (and generally) is that overlapping West Asian populations in PC1 and PC2 were distinct in a CHG dominated PC3. (Steppe samples also). This implies to me that we may be missing data by not having a representative of CHG in the pright... This is perhaps something which various UP groups are capturing between them?

(Maybe not resolvable; as a test compromise pright, perhaps (Ethiopia_4500BP, Ust_Ishim, Tianyuan, MA1, AfontovaGora3, ElMiron, Villabruna, WHG, SHG, Iran_N, Natufian, Levant_N, Barcin_N, Tepecik_Ciftlik_N). Omit the UP groups, but include a group which appears to be directly related to Antolia_N+CHG+Levant_N?)

...

@super nord, thank you for the comment.

Davidski said...

@supernord

Domesticated sheep, and probably goats too, spread from the Caucasus to the western steppe, and then from there to Central Asia. Here's the latest on that...

http://www.mdpi.com/genes/genes-08-00165/article_deploy/html/images/genes-08-00165-g001.png

http://www.mdpi.com/2073-4425/8/6/165/htm

Davidski said...

@Matt

Using your set of pright pops...

https://drive.google.com/file/d/1D4l797pbP1TRmBEtUKJzZ_csnkxQm4wj/view?usp=sharing

https://drive.google.com/file/d/1GRsbHO3J8aleHZLHDmrXUceLzlMZI_PM/view?usp=sharing

https://drive.google.com/file/d/1nw5AcCF3yRfa3vSBoFJgET3zYdTu0q_5/view?usp=sharing

https://drive.google.com/file/d/1iU2ie6XSvE3BoNxLGVoiAzNYLFL0F6rZ/view?usp=sharing

Anonymous said...

@Davidski said...

"Domesticated sheep, and probably goats too, spread from the Caucasus to the western steppe, and then from there to Central Asia. Here's the latest on that..."

It's impossible. In Jebel (and Central Asia), they are distributed directly from Iran, in essence, back in the Mesolithic era. There is no connection with the Caucasus, fundamentally no. This article is dedicated to the Pamir mountains only Bronze Age, not the Neolithic. And a word to believe not because they completely ignored the data from Central Asia (Eastern Caspian), it is easy to see. Writing about Central Asia it is not taking data from it, it's just the audacity. Not good issue for Central Asia, the Pamirs only, that is eastern border of the Central Asia, but not Central Asia.

Davidski said...

@supernord

Goats and sheep made it into the Caucasus from the northern fertile crescent around 8000 years ago. Then they made it onto the steppe from the Caucasus. And then they were taken from the western steppe to Central Asia by the ancestors of the Afanasievo people.

This is the consensus. That's why it's the model shown on this map.

http://www.mdpi.com/genes/genes-08-00165/article_deploy/html/images/genes-08-00165-g001.png

You're talking about the spread of goats and sheep to the east Caspian, and to South Central Asia. This did indeed take place from what is now Iran, and, again, it's shown on the above map.

However, you're also claiming that goats and sheep made it onto the steppe from Iran via the east Caspian. Maybe they did, in part, but this is not the consensus. The consensus is that they spread onto the steppe from the Caucasus. So you're the one arguing against the consensus here, not me.

And please keep in mind also, that the author of that paper on the Prikaspiyskaya culture places its origins in the Lower Don region, near the North Caucasus, not east of the Caspian. See here...

http://eurogenes.blogspot.com/2017/09/two-starkly-different-neolithic.html

Rob said...

In a nutshell

1) ceramics adopted independently in south Russia / Ukraine. A smimilar ceramic repertoire from Bug-Dniester to Caspian , from 6000 BC ->
No migrations implied.!

2) Elshanka pottery different; clear influences from kelteminar etc

3) first adoption of domesticates in precaspian culture, ? Hypothesised to be from lower Don; first sheep , then also cattle pigs etc by the time of Khvalynsk, during which time contacts with West even more manifest (copper and other prestige symbols)

Anonymous said...

Okay, let the spread of productive economy via the Caucasus is possible, but it is not a consensus. This is a new opinion appeared recently. Look at these new study 2014 that they referenced http://www.oxfordhandbooks.com/view/10.1093/oxfordhb/9780199935413.001.0001/oxfordhb-9780199935413-e-13.

South Dagestan "In the Neolithic layer (C) at Chokh, a large stone building with a corridor-like entrance yielded abundant material (Amirkhanov, 1987). Continuity is obvious in the lithic material from the Mesolithic layers: scalene triangles still predominate and small blades become frequent. Grinding stones and pottery (mineral-tempered ware with flat bases) also appear, and a sherd decorated with two knobs evokes the Aratashen-Shulaveri-Shomutepe culture. Bone sickle handles decorated with incised diamond shapes closely parallel the culture of Sialk I (sixth millennium bc) on the Iranian plateau (Wechler, 2001).

Based on the presence of domesticated animals (sheep)

The Neolithic level of Chokh in Dagestan probably(!!!) belongs to the sixth millennium bc, based on its parallels in cultural material with the cultures of Aratashen-Shulaveri-Shomutepe and Sialk-1. Moreover, the presence of fully domesticated cereals and the absence of wild varieties found in this layer suggest an import. There is no support for the hypothesis of a local development of agriculture."


Nowhere in any study I have not seen this way or this assertion, but in Central Asia such was recognized by all, there are these animals even older than the Caucasus. That Central Asia was consesual.

Alogo said...

My impression is similar to David's in that regard, that the Caucasus route is generally more proposed. But these matters seem to have been debated at least since the 70s (as far as my amateur self is aware) and they haven't been fully settled. Supernord might be more up to date in what the local archaeologists generally and currently think but clearly both areas have been postulated in the literature as far as I can tell. So it's basically hoping we get samples from the whole area between the Kuban and Djeitun at some point to potentially really settle matters in all these issues.

Davidski said...

@supernord

Indeed, food production, including domesticated goats and sheep were introduced into the Caucasus from the Fertile Crescent around 6,000 BC. That's what the map that I linked to shows, and this is what is being confirmed with ancient DNA.

"The sites where these bones were found are the earliest agricultural settlements in the Caucasus. Other novel signs of agriculture and cultural artifacts also suddenly started to appear in what were hunting-and-gathering areas," lead author Seiji Kadowaki says. "This ties in well with the introduction of domesticated goats from the Fertile Crescent around the same time, suggesting that populations moved or indigenous hunter-gatherers in the Caucasus accepted agricultural lifestyles from the Fertile Crescent about 7,500-8,000 years ago."

https://www.sciencedaily.com/releases/2016/06/160613090635.htm

And Afanasievo people, who took goats and sheep to the Altai, derived almost half of their ancestry from Caucasus Hunter-Gatherers (CHG).

It's rather unlikely that they had any ancestry from the Iranian Plateau, because the uniparental markers don't match. Ancient and modern-day Iranian mtDNA is too exotic for the Bronze Age steppe, with typically South Caspian markers like U7, which are missing from the ancient steppe.

So there's no problem for the Caucasus route, but there is a problem for the east Caspian route.

Matt said...

@Davidski, so the results with Tepecik in the pright and no UP samples in pright (other than late El_Miron sample):

CHG:EHG:Blatterhole_MN = 0.406 : 0.437 : 0.157, tail p: 0.50477906
CHG:EHG:Koros_EN:Koros_HG = 0.399 : 0.440 : 0.086 : 0.076, tail p: 0.423959457
CHG:EHG:Armenia_EBA = 0.108 : 0.504 : 0.368, tail p: 0.0732311356
CHG:EHG:Iran_Chl = -0.266 : 0.594 : 0.672, tail: 0.0522131614

I don't have anything more to say about them, but thanks.

Ric Hern said...

Was there no way that the knowledge of Herding could have spread along the East Black Sea Coast and the Western Caucasus to the Pontic Steppe ? After all we see Otzi high up in the Alps....how many times did he cross the Alps before..?

Anonymous said...

It nice to see the tail probs. improving with each strategic advance in modeling. Still, Yamnaya's is still markedly lower than others, leading me to think there are some subtleties missing.

EHG + CHG are certainly the starting points for left pops which make up the bulk of their ancestry. European farmer as a small proportion makes perfect sense, and with that is going to come WHG. To get the most fine tuning, it would probably be best to have them separate (as opposed to a single high-WHG MN source), in the event that the EHG that contributed to Yamnaya was more Western, which I think is the case. Davidski, you have already attempted this as CHG + EHG + Koros_EN + Koros_HG, which turned out to be the best fit so far.

After reading the comments to this topic regarding the neolithic of the Caucasus, the 4-pop model of CHG+EHG+EEF+WHG as an end-all assumes 1 of 3 possibilities
1) Local development of neolithic traditions by CHG individuals
2) CHG existed outside of the Caucasus and developed a neolithic tradition wherever that was, carrying it to the Caucasus, demic replacement during the neolithic is thus indiscernible due to the extremely similar nature of the two groups
3) Neolithic traditions were brought to the Caucasus by people not related to CHG but had 0 demic impact.

Your attestation of "food production, including domesticated goats and sheep were introduced into the Caucasus from the Fertile Crescent around 6,000 BC" would logically cause you to dismiss option 1.

Option 2 is technically possible, but given our coverage of ancient samples, the options are extremely limited. In addition it assumes option 3 for the southern spread of this neolithic tradition.

Option 3: 0 demic impact, would not just be on later steppe populations. Arrows on the arrows here...
http://www.mdpi.com/genes/genes-08-00165/article_deploy/html/images/genes-08-00165-g001-550.jpg
show movements heading through Iran at 6000BC. We have samples from Western Iran at 8000BC and 5000BC. There is no shift towards CHG between the two time periods, but rather towards Central Anatolians.

If all three options are eliminated, then the model of CHG+EHG+EEF+WHG is not complete. A Central Anatolian population would probably be worth a shot [i]in addition[/i] to the aforementioned 4 pops.

Rob said...

Interesting diagram. Reminds me of Krause’s model . But given that the earliest domesticate (sheep) in Caspian (5500) is arguably earlier than the Don (4500 BC) ; there might be a need for arrow from east Caspian to lower Volga
But there are definitional and taphonymical issues with the entire chronology issue

Seinundzeit said...

All,

Very late to this conversation, but just for fun...

Since the shape of qpAdm output is quite contingent on the pright setup, it might be of some interest to look at an example of output which has been produced via a totally different pathway.

For that, Globe_10 can suffice.

People tend to get chaotic results with this data, but that's because they fail to apply eigenvalue scaling (as described by our very own Matt and Alberto; as always, by eigenvalue scaling, I don't mean to imply our old ideas concerning weighting of PCs).

When scaling is applied, the results are always sensible.

Anyway, the reference populations were these:

South_Africa_2000BP
Mende_Sierra_Leone
Mota
Jarawa
Mongola
Ulchi
WHG
ANE
EHG
CHG
Iran_Chl
Iran_M/N
Barcin_N
Baalberge_MN
Natufian
Levant_N
Levant_EBA

Afanasievo:

55% EHG + 2% WHG
35.9% CHG
7.2% Baalberge_MN

distance=0.1751

Yamnaya_Samara:

49.20% EHG + 8.50% WHG
39.75% CHG
2.55% Baalberge_MN

distance=0.3455

Despite the inclusion of Iran_Chal, and ancient Levantine populations, Steppe_EMBA populations do seem to derive nearly all of their Near Eastern ancestry from CHG-related populations.

For what it's worth, early Latvian Corded Ware and the Srubnaya_outlier fare rather differently, with respect to their non-Near Eastern ancestry.

Corded_Ware (Latvia):

31% WHG + 26.4% EHG
42.6% CHG

distance=0.8637

Srubnaya_outlier:

45.3% EHG + 26.8% ANE
26.2% CHG
1.7% Ulchi

distance=0.5063

The WHG/EHG/ANE proportions do differ greatly from Steppe_EMBA. Obvious hints of substructure among eastern European/steppe/Siberian hunter-gatherers. Also, this means that the Srubnaya_outlier is overwhelmingly ANE in terms of deep genetic ancestry (has a heavy shot of extra ANE, has EHG which is mostly ANE, and has CHG which is loaded with something ANE-related).

Regardless, looking at Scythian_ZevakinoChilikta, the Near Eastern ancestry is very different from Steppe_EMBA, early Latvian Corded Ware, and the Srubnaya_outlier.

Scythian_ZevakinoChilikta:

29.35% Ulchi + 5.70% Mongola
17.80% EHG + 11.15% ANE
9.95% Iran_Chl + 8.45% CHG + 3.55% Iran_M/N

distance=0.0456

So, Global_10 can discriminate between different sorts of Near Eastern ancestry, just as it can between different sorts of Mesolithic European/Siberian ancestry.

Based on this, EHG + CHG for Yamnaya/Afanasievo is probably close to reality, rather than a quirk with the modelling.

Anthro Survey,

I have not forgotten about your request.

In the free-time that's coming up, I'll be working with qpAdm and qpGraph, so if I see anything that hasn't been shown before, I'll keep you posted.

Davidski said...

@human443

After reading the comments to this topic regarding the neolithic of the Caucasus, the 4-pop model of CHG+EHG+EEF+WHG as an end-all assumes 1 of 3 possibilities.

There's a 4th possibility: goats and sheep were introduced into the Caucasus with migrants from the northern Fertile Crescent, who did make a strong genetic impact on many areas of the Caucasus, but next to nothing on the steppe.

If all three options are eliminated, then the model of CHG+EHG+EEF+WHG is not complete. A Central Anatolian population would probably be worth a shot [i]in addition[/i] to the aforementioned 4 pops.

Yamnaya Samara doesn't like Central Anatolians, at least relatively speaking.

https://drive.google.com/file/d/1F7_540CJCmcTtU6j8ZW2BbyP79EZ90j2/view?usp=sharing

https://drive.google.com/file/d/1g_2dUAytTL8JAtfGZ8x8yoSLhwdU-NsE/view?usp=sharing

Ric Hern said...

As far as I could remember goats were domesticated in the Zagros Mountains the earliest...

Anonymous said...

@Davidski

"Indeed, food production, including domesticated goats and sheep were introduced into the Caucasus from the Fertile Crescent around 6,000 BC."

Only around 5000-5500, it is not 6000 BC.

"That's what the map that I linked to shows, and this is what is being confirmed with ancient DNA."

Ancient DNA for only BRONZE AGE (< 2300 BC) was! It is not early.

"So there's no problem for the Caucasus route, but there is a problem for the east Caspian route."

No problem with east Caspian route, only geneticists fully ignore any data from east Caspian. Absolutely they do not test.

Anonymous said...

@Davidski

Thanks for running those two runs. What I find confusing is, in the 4-pop run this model is produced...
10.036 0.612773 0.438 0.436 0.062 0.064
With a solid tail prob of 0.61
But with the 5-pop run, where each pop in in turn zero'd as a possible combination, this happens?
12.819 0.382326 0.420 0.432 0.073 0.074 0.000
The proportions differ slightly and the tail-prob craps out to 0.38. Why does this happen with the same right pops in both runs? Shouldn't the 5-pop run just produce...
10.036 0.612773 0.438 0.436 0.062 0.064 0.000?

Davidski said...

@human443

I don't know. You should ask Nick Patterson about that.

But I never use qpAdm to model more than four streams of ancestry because the results generally look unstable.

The most practical strategy seems to be to break up the analysis as much as possible, and avoid very similar reference samples (i.e. those that correlate too closely with each other).

Ric Hern said...

Interesting is that the Mouflons original range stretched all the way to the Crimea where they apparently became extinct about 3000 years ago...