search this blog

Friday, January 21, 2022

Yamnaya is from Europe, but it's really from Asia


I was about to post a comment under a new preprint at bioRxiv, but the comment section isn't there anymore. Hopefully, this is just a temporary glitch.

The preprint in question is titled Reconstructing the spatiotemporal patterns of admixture during the European Holocene using a novel genomic dating method [LINK]. It's co-authored by Harvard/Broad MIT scientist Nick Patterson who occasionally comments at this blog.

My impression is that the authors see the people associated with the Yamnaya culture as Asians who simply used "far" Eastern Europe as a springboard to expand into other parts of Europe.

If so, they're dead wrong.

There are at least three arguments why the Yamnaya population should be seen as quintessentially European:

- its home was initially and overwhelmingly the Pontic-Caspian steppe, which is entirely located within the present-day borders of Europe

- Yamnaya genomes are clearly different from those of older populations native to nearby parts of Asia, and, in fact, these differences show a very strong correlation with the present-day borders between Europe and Asia

- the Yamnaya people weren't a new population in Europe by any stretch, but must have been overwhelmingly derived from the very similar Eneolithic peoples of the Pontic-Caspian steppe and/or the nearby forest steppe, both of which are located in Eastern Europe.

And yet, this is what the preprint claims:

The beginning of the Bronze Age was a period of major cultural and demographic change in Eurasia, accompanied by the spread of Yamnaya Steppe Pastoralist-related ancestry from Pontic-Caspian steppes into Europe and South Asia (16).

In fact, what really happened at this time was that Yamnaya steppe pastoralist-related ancestry spread from Eastern Europe to other parts of Europe, as well as to Central and West Asia.

The preprint does eventually explain that present-day South Asians derive their Yamnaya-related ancestry from a later eastward expansion of the European Corded Ware culture (CWC), but it completely ignores the fact that the Afanasievo culture was the result of the initial eastward expansion from Europe to Asia. That is, the ancestors of the Afanasievo people were recent migrants from the Pontic-Caspian steppe to Central Asia and Siberia.

There's also this:

Over the following millennium, the Yamnaya-derived groups of the Corded Ware Complex (CWC) and Bell Beaker complex (BBC) cultures brought Steppe pastoralist-related ancestry to Europe.

Seriously? Both the CWC and BBC, just like the Yamnaya culture, were from Europe. In fact, as per above, the descendants of the CWC expanded into Asia.

And this:

The second major migration occurred when populations associated with the Yamnaya culture in the Pontic-Caspian steppe expanded to central and western Europe from far eastern Europe.

The authors basically admit here that Yamnaya came from Eastern Europe, but they call it "far" Eastern Europe. Perhaps they know something I don't, but as things stand, there's no evidence that Yamnaya came from "far" Eastern Europe. In fact, the emerging consensus based on ancient DNA, including pre-publication data, is that Yamnaya may have originated in what is now Ukraine. In my opinion, Ukraine isn't located in "far" Eastern Europe, but more or less in the middle of it.

Inexplicably, this is what they say about the genetic origins of the Yamnaya and Afanasievo peoples:

These groups were likely the result of a genetic admixture between the descendants of EHG-related groups and CHG-related groups associated with the first farmers from Iran (8, 22, 36).

...

Thus, we combined all early Steppe pastoralist individuals in one group to obtain a more precise estimate for the genetic formation of proto-Yamnaya of ~4,400 to 4,000 BCE (Figure 2). These dates are noteworthy as they pre-date the archeological evidence by more than a millennium (37) and have important implications for understanding the origin of proto-Pontic Caspian cultures and their spread to Europe and South Asia.

Not really.

Like I said, the Yamnaya population was overwhelmingly derived from the Eneolithic peoples of the Eastern European steppe and/or forest steppe. And these Yamnaya-like Eneolithic peoples were spread out across a vast area of Eastern Europe by at least ~4,500 BCE. Some of their genomes have been available for several years, and many more are on the way.

It is possible that the Yamnaya and Afanasievo genotype formed in 4,400-4,000 BCE, but if so, then this was due to mixing between the Eneolithic steppe peoples and nearby European farmers. That's because the difference between the Yamnaya and Eneolithic steppe genotypes is minor (~15%) European farmer admixture in the former.

The really interesting puzzle is exactly where and when the peculiar Eneolithic steppe genotype came into being. Any ideas Dr Patterson?

See also...

Matters of geography

Understanding the Eneolithic steppe

146 comments:

MaxT said...

Hope upcoming study about Enolithic_steppe will address how Y-DNA J got into EHG. I don't think we have full picture of how Eneolithic steppe were formed. I'm more or less certain CHG admixture in steppe is older than current estimates.

Rob said...

- ''the other in southern Europe with WHG–GoyetQ2 ancestry (23)''

maybe should specify it as southwestern Europe. Doesn't seem that Goyet-Q2 plays a big role in Italy or Balkans (also in 'southern Europe'')

- ''Over the following millennium, the Yamnaya-derived groups of the Corded Ware Complex (CWC) and Bell Beaker complex (BBC) cultures brought Steppe pastoralist-related ancestry to Europe.'

Yeah, they should qualify 'Europe'' to 'central- & western Europe''

-''The most recent dates of ~7,500 BCE were inferred in eastern Europe in Ukraine HGs,''
Should specify that as Ukraine_N HGs

Vladimir said...

I have the impression that with their dates of 4400-4000 BC they got into the culture of Sredniy Stog II, Rassamakin calls it Stogovskaya. That is, they did not determine the date of mixing EHG and CHG, but the date of formation of the Yamnaya EHG/CHG/WHG/EEF profile, where EEF is present in a minimal amount. The subsequent increase in EEF, for example, at Fatyanovo, occurred later closer to 3200 BC.

Steppe said...

I don't understand that either, Yamnaya and CWC both originated in Eastern Europe, geographically Europe ends in the Urals and in the Caucasus, just like I wrote with the professor for archaeo- and paleogenetics Johannes Krause, who published a great book but that alone The assertion “spread of the Asian steppe riders” is totally wrong, Eastern European steppe herdsmen or West Eurasian steppe herdsmen would be better. Just as he describes the Afanisevo culture as conquerors on horseback? However, the bearers of the Afanassjevo culture (also related to Tarim groups in the Dzungarian Basin/ Tochars), who were an eastern offshoot of the Yamnaya culture, did not, in my opinion, ride horses, nor did the Corded Ware and Bell Beakers, for the first time according to archaeological Evidence of the modern domestic horse ridden in the Srubna culture and later further developed by the eastern offshoot of the Corded Ware (Sintaschta culture) for warlike actions "The chariot" but only in the Andronovo culture horizon was riding done on a large scale. He wrote back to me that Yamnaya and the Afanisevo people probably rode Botai horses (Przewalski's horse) well maybe. I wrote him that the CWC and its offshoots were much more influential, he wrote he was mostly interested in genetics and he coordinated with other colleagues who had given him information that the Afanisevo people had ridden and sent me a couple of publications from Nature who already knew .

Davidski said...

Well, I can't comment any hearsay in regards to Johannes Krause. But if he really claimed that Yamnaya and Afanasievo people rode Botai horses, then geez.

However, I did once see a Youtube clip in which he falsely claimed that Yamnaya came from Central Asia.

I think that the extremely poor understanding of geography among people working with ancient DNA is a very serious problem.

Bob Floy said...

Definitely a pattern taking shape here, with these incorrect peer reviewed papers lately. Hard to see how anyone could deny it.

Davidski said...

This paper hasn't been peer reviewed yet, but yeah, it's like Europe ends east of Poland for most of these people.

I wouldn't be surprised if some of them thought that Poland was in Asia.

vAsiSTha said...

Yes, the iranian component which mixed with ehg between 4400-4000bce, was sitting in Europe. Sure.

vAsiSTha said...

@vladimir

"That is, they did not determine the date of mixing EHG and CHG, but the date of formation of the Yamnaya EHG/CHG/WHG/EEF profile, "

This is not how DATES works.

Davidski said...

Do you really believe that this Yamnaya-like individual was one of the first people of his type?

Progress 2 PG2001 4336-4178 BCE

Judging by this and similar samples from Vonyuchka, Khvalynsk, and Sredny Stog, we're dealing with a population that formed at least 4,500 BCE.

And that's a very conservative estimate.

John Thomas said...

David, it's all political.

Anti-whiteism is now de rigueur in academe, just like it is everywhere else.
It's the trendy - and therefore rewarded - cause to espouse.

Steppe said...

You described that well, but I can remember a friend from childhood days at school who thought Poland is in Asia and you live in tents, then we went to Poland and went to the weekly market with a few other friends and I had to interpret and I laughed at him too and said it's an Indo-European sorach like German and Poland is far away from Asia, he didn't understand the word "Indo" and he was very overwhelmed 😩😂, but I have to honestly say that the level at German schools is very poor in terms of history and geography.

Rob said...

For me, the question isn’t about what we call this population Caucasian, Caspian or Iranian; it’s why did HGs from near-eastern Europe started marrying women from that group en masse ?

Bob Floy said...

@Davidski

I bet it clears peer review.

@John Thomas is correct.

Palacista said...

You would think that someone at Harvard could read a map.

vAsiSTha said...

The reservoir effect on the dates of these steppe eneolithic sample must be ascertained. Same for khvalynsk.

Another Question is the validity of the assumption of 28yrs per generation. It was calibrated in one of Moorjanjs papers with paleolithic samples, but who knows how accurate it is.

Draft Dozen said...

@Davidski

In interview to the Deutsche Welle he also said that the russians are eurasian genetic medley:
"In Russia there is a growing interest in genetics, including at the state level, and much has been written about the Russian genome. As it can be described in terms of archaeogenetic?

No typical Russian genome does not exist. The people of Russia have almost all the genetic diversity found today outside of Africa. This is the same medley as in Europe. In the last 5000 years, Russia was one of the most dynamic regions of the planet from a genetic point of view. Passing on her vast territory of the steppe belt that connects Eastern Europe and Asia, was a kind of motorway on which is very often moved mobile groups, whether Scythians, Huns, Avars, Mongols. All this led to large-scale genetic exchange.
Significant genetic changes have occurred in the recent history of Russia. In the 20th century - the 30s, 40s, 50s years, and then, in the 60s and up to 90-years - on its territory there was a particularly strong mobility, to put it politically neutral. This is all very changed by the local population. So in today's Russia there is no clear genetic boundaries, they do not on the world map. All mixed up. No genetic differences between Russian residents from the other inhabitants of Eurasia is not fixed"
And also according to his words:"we are here in Germany for a long time and work very closely with various specialists in Moscow, St.-Petersburg, Novosibirsk, Stavropol, and in fact throughout Russia. So we have so much knowledge about this region"
So, such assessments from him, are not accidental, he is biased.

Matt said...

Off topic: https://twitter.com/mathiesoniain/status/1483963852928790532?cxt=HHwWiMC5leHQjJgpAAAA

Academic genetics will soon reinvents our use of PCA plus nMonte?

(Though Vahaduo is the superior algorithm with better runtime, for reasons Vahaduo has explained, and huijbregts never thought nMonte was to be used for PCA until we all proposed here that it should be. The history is it was used by huijbregts for ADMIXTURE components, then we proposed using it to fit f3 statistics, then moved on to using it for PCA to take advantage of the projection of smartPCA. Can't remember who proposed what stages of course.).

Florian Prive (start of the twitter thread) has issued a bunch of papers (https://privefl.github.io/com.html) whewre he has examined the relationship between euclidean distance (and ^2 distance) on PCA and formal measures of differentiation (Fst) and proposed using distance measures on PCA to define populations. Finding that ^2 euclidean distance basically relates well to Fst. So it's no surprise that he should make the "leap" in an "independent" way to using this for ancestry inference.

Along with paper by Benjamin Peter which relates PCA distance to f-stats - https://www.biorxiv.org/content/10.1101/2021.07.13.452141v2

I hope that something does come of academics looking at those methods, as it does really need a proper academic investigation to compare against established methods like qpAdm and ADMIXTURE. This method is probably working really well at inference for ancient components but I worry about how ancient components are projected onto dimensions defined by perhaps recent drift and whether projection is accurate. As well as inference involving projection of samples that are contributing little to present day people.

Davidski said...

@vAsiSTha

That date is already calibrated for reservoir effect.

A radiocarbon doublet of charcoal and human bone revealed a strong reservoir-effect in the human bone date. Dating: human bone 4991-4834 calBCE (6012±28BP, MAMS-110564), charcoal 4336-4173 calBCE (5397±28BP, MAMS- 110563).

Sucked in troll.

vAsiSTha said...

4336-4173 fits perfectly into the 4400-4000bce admixture date.

Not sure why you're giddy

Matt said...

The paper is a new thing to read itself, and I think it's a shame that we have to discuss it again in the context of the thing about geography.

Would've suggested if they could've split the Czech sample for instance, into windows defined by 350 year slices, and then seen what happens with the DATES inference. Or "Central Europe with Steppe Ancestry 3000-2650 BCE", "Central Europe with Steppe Ancestry 2650-2400 BCE", etc. Rather than using population based models only, which slices samples tested together quite broad in time but also into smaller samples.

Re; the suggestion they detect their curve due to 10%ish EEF rather than 40%ish CHG, they could stand to cross-test using an EEF reference against EHG+CHG.

Their method seems to be compatible with 3-way population models as in the European example, so it would be very easy for them to test and prove / disprove that this is driving the effect. Hopefully that can move us on from proposing that the fit is just due to EEF and then never testing that. (IranN or CHG probably shouldn't matter too much as within the limits of the test population for this method? IranN they use because of the large sample size I guess, though being more of a distal population to the true one.)

An alternative would be that somehow there was some ongoing process of admixture from still extant relict CHG rich and EHG rich populations, that spanned from before the time we have Progress_En samples, and also somewhat after them, and that the data we have is sparse. Like if two populations formed that were 75% EHG, 25% CHG, and 25% CHG, 75% EHG, by >5000 BCE, and then there was slow admixture between these leading to the formation of proto-Steppe genotype, perhaps a composite of that range would be dated later.

In the post-CA Europe model, we generally seem to have a pulse around 3000-2700 BCE, and then further flow and homogenization after this, and this seems to explain the slightly later dates in some populations, being an average of this range but with a bias towards the major and early period of mixture. The method has some questions in interpretation in the presence of continuous ongoing models?

The formation of Latvia_BA and Hungary_EBA_Mako here are estimated at 2520 BCE and 2580 BCE, which seems plausible as a composite.

Davidski said...

@Matt

The paper is a new thing to read itself, and I think it's a shame that we have to discuss it again in the context of the thing about geography.

Well, they keep doing this, so I'll keep doing what I do.

Davidski said...

@vAsiSTha

Right, so according to you PG2001 was one of the first people from this supposed EHG/Iranian mixture.

That would be a miracle. Another miracle would be an Iranian farmer on the steppe in 4,400 BCE.

I don't believe in miracles. But you obviously do. High time to see a psychiatrist.

Matt said...

@Davidski, yeah, I guess that's fair enough to be honest.

vAsiSTha said...

from wang supplement, harvard anno file doesnt have charcoal dates.
PG2001 4336-4178 bce, charcoal
PG2004 4233-4047 bce
VJ1001 4332-4238 bce

Fits perfectly, none of these are before 4400BCE.

"Right, so according to you PG2001 was one of the first people from this supposed EHG/Iranian mixture."

Of course, any sane person will look at the data and the preprint and conclude the same. Is there some rule that the 1-10th generation offspring of an admixture event cannot be found by archaeology?

get yourself checked first.

Davidski said...

There are similar samples from Khvalynsk sites dated to a similar time.

And many more such samples are on the way from further west.

So like I said. It's a miracle!

Matt said...

Related to the end of David's post, if I had to guess also, Nick Patterson and the others have already used this method for the other steppe samples that are unpublished (the various Khvalynsk cline samples and Sredny Stog?) and this paper is doing some element of preparing the way for using that method in those papers when they're published? Those results might be complementary or a contradiction.

The extensvie set of Khvalynsk samples seem like they could be a better test case, because they lack EEF admixture. Though the objection could be there that in the opinion of many, they're not ancestral to Yamnaya/steppe expansion proper, and so any dates there would just reflect an admixture between EHG and the true population.

Davidski said...

@Palacista

You would think that someone at Harvard could read a map.

What about Max Planck? And practically all the labs.

I actually think it's funny.

vAsiSTha said...

Go put an article saying the preprint dates are wrong. Go ahead and challenge them lol.

It's a simple 50/50 admixture. No reason to guess their dates are wrong.

Also this does only means that the admixture in proto yamnaya and proto afanasievo occurred in 4400-4000bce.

IF khvalynsk is indeed earlier than this, then that would mean khvalynsk is not the ancestor of proto yamnaya or steppe_en or whatever.

Use logic and learn to interpret better.

EastPole said...

@Draft Dozen
“In interview to the Deutsche Welle he also said that the russians are eurasian genetic medley:

No typical Russian genome does not exist.”


Yes and the Germans, French, British are euroafrican-euroasian medely. No typical German, French, British genome does not exist.

Davidski said...

@Matt

Maybe Nick Patterson can chime in and clear up a few things.

But it seems to me like this manuscript was prepared a long time ago.

Matt said...

@Davidski, yeah that's what I mean, it was written ages ago (they've used a fairly old version of the Allen adna resource for example), but there's a reason they've uploaded it to biorxiv now.

EastPole said...

medley not medely

Simon Stevin said...

@Davidski

Do you know where one can find the recalibrated dates for Vonyuchka, Progress, and Khvalynsk? Even recalibrated, I0434 at least dates to 4750 BCE.

Khvalynsk_Eneolithic I0433, Samara Oblast, Russia, 4697-4539 calBCE, mtDNA: U5a1i, Y-DNA: R1a-M459*, MathiesonNature2015

Khvalynsk_Eneolithic I0122, Samara Oblast, Russia, 4936-4730 calBCE, mtDNA: H2a1, Y-DNA: R1b-V1636*, MathiesonNature2015

Khvalynsk_Eneolithic I0434, Samara Oblast, Russia, 5198-4853 calBCE, mtDNA: U4d, Y-DNA: Q1, MathiesonNature2015

a said...

@manuscript, prepared a long time ago" they were saying this back in 2012 before steppe samples were released, back in the Grugni et al days. Truly amazing.

Davidski said...

Not sure what you mean by recalibrated.

You just need to check the latest papers and Reich anno files and see what's there.

vAsiSTha said...

@simon

Progress and vonyuchka dates are in the Wang 2019 supplement excel.

https://www.nature.com/articles/s41467-018-08220-8

Use the charcoal date wherever available.

Simon Stevin said...

@Davidski

Sorry I meant the carbon dating that takes into account the reservoir effect. I checked the Reich lab anno files, and they still published the non-charcoal dating for PG2001. But the WangNatureCommunications paper from 2019, published the charcoal dating for PG2001, in addition to the bone dating. Furthermore, according to the Harvard anno files, the dating of the Khvalynsk men also likely suffers from a heavy reservoir effect. I’m guessing that some of the unpublished Khvalynsk samples charcoal date to around the time of PG2001, or perhaps even earlier. I0434 appears to be the oldest we have. CHG or Dzudzuana-like ancestry seems to have been in Eastern Europe for a long time. J1 is in EHGs from Karelia, in Eneolithic Khvalynsk, and in Afanasievo. Why is Iran their go to when we have CHGs and Dzudzuana right below the steppe in the Caucasus (and they’ve been there for at least 30k years)?

Rob said...

@ Simon
Those dates for Khvalysnk aren’t catered for reservoir effect (a/p Ian Mathieson)
They’re simply recalibrated from BP to BCE

Matt said...

Another thing I'd say about this paper, as well as the dates being a composite, they also represent that vast majority of ancestry.

Like the paper could be read as indicating that in Britain there was no admixture at all from WHG. However we know that in Britain there are HG rich outliers in Orkney (which could've happened on the mainland), and across the water in Ireland, there are samples I think Lara Cassidy found that share a specific excess of haplotypes with the Irish Mesolithic Hunter-Gatherer they sampled.

Given the sample set is pretty small, it would seem unlikely that we've found unusual samples who had unusual marriage patterns.

But it's simply the case that these populations were likely too small to make an overall impact. Instead the major impact of HG happened when the founding populations of the European Neolithic were relatively small and local HG populations were relatively large. Most of the HG ancestry in the West European EEF is probably from the Southern or Central French HGs who were in the right time and place to contribute ancestry that was large relative to the incoming population and probably a set of male haplogroups that drifted to become dominant?

So culturally it doesn't eliminate that a founding group could admix with local people, but that in-marriage coming from a neighbouring group could dilute this over time, if that group is large.

...

Also on my above comment that perhaps their date in this paper reflects an ongoing breakdown of barriers between CHG-EHG admixed people, rather than a single pulse between unadmixed CHG-EHG, there was a related comment about unpublished findings by David Anthony last month.

Anthony is definitely to be taken with a pinch of salt on genetics as its not his speciality, and I know a lot of people are very cynical of his communications, but I thought I'd offer it up anyway.

On this podcast podcast again: https://www.youtube.com/watch?v=O67ct6qd5Ik

15:40: "Variety of different groups north of Black and Caspian Sea" (before Yamnaya expansion) "At least 4 sources of genetic ancestry "(?)"active in steppes", then "admix around 4500 BCE".

If that were correct, it's not necessarily a breakdown of their model if they find a signal like that. Some different groups reflecting a CHG like source and EHG source could've formed in the Mesolithic, then barriers break down.

(Also further talk at 32:50 about Relatedness of Yamnaya.)

As I say though its not his speciality and there are some moments of obvious confusion in there. So the above could relate to that confusion. Particularly he gets confused on Western Europe. One odd moment at 13:10 talks about Mesolithic hunter gatherers "still there in Britain, making things like Stonehenge"? And I think he's still a little confused about HG resurgence in EEF, thinks its due to HG adopting farming then admixture happens, but not totally sure that really maps what happens. Seems a lot more going on since EN and it just doesn't tend to be the groups that demographically boom, until late in expansion. Generally seems somewhat confused on the NW, also talks at 37:00 about 98% replacement in British Isles, where it seems more like 8/9 or 9/10 from Olalde's work (e.g. https://imgur.com/a/3q0usNa), talks about monumentalism and megalithism as a reaction to steppe expansion (which definitely doesn't seem very well supported) and seems to get confused between whether Bell Beaker is the invasive group or the local group.

Davidski said...

Honestly, until I get all the samples and run them myself, I'm not taking anyone seriously.

Sct said...

Exactly to the point! Modern (western) scientists have extremely poor knowledge of European geography. Their perception of Eastern Europe is dominated by a 300 year old (no joke!) imperial construct, a myth, perfectly described in the book by Larry Wolff
"Inventing Eastern Europe: The Map of Civilization on the Mind of the Enlightenment".

My impression is, even when facts are obviously in contradiction with the myth of extremely distant but not worthy of research nor even a visit, exotic yet very well-known primitive boring
Eastern Europe, the myth wins instantly. No further research needed.

As an illustration, I adore German political maps of Europe with extremely dence western countries and with a whole Ukraine being depicted almost blank, with missing whole big rivers, having only 5 cities, 2 of which are actually villages!

Sct said...

Western Europe's imperial history is unfortunately not over, it wasn't processed and thus Eastern Europe is still mostly seen as a distant resourceful territory of otherwise inferior nature (=colony). So actually the similar perception to Russian view on the region. Bitterly, this perception caused catastrophic events of XXth century (see "Bloodlands" by Timothy Snyder), and the whole does not stop right know :(
EU's leading countries are almost silent again in the face of a highly probable catastrophe.

Xdzyn said...

It's very sad that the damned political-ideological preferences have put their grip on ancient DNA studies (nothing against it, each person believes what they want), but putting their personal narratives above proven facts, I think this is very dangerous. it's not good...

MaxT said...

David Anthony in his book claims earliest EHG/CHG mix is found in unpublished Ekaterinovka sample which is 3% CHG. This is northern most sample with CHG according to him.

Davidski said...

David Anthony also claimed that Yamnaya was 70% CHG because he looked at an Admixture graph from Wang et al.

So yeah, like I said, I'll wait for all the samples.

Andrzejewski said...

@MaxT “ Hope upcoming study about Enolithic_steppe will address how Y-DNA J got into EHG. I don't think we have full picture of how Eneolithic steppe were formed. I'm more or less certain CHG admixture in steppe is older than current estimates.”

What most researchers and average people forget to mention is that CHG is also an ANE derived population, not just the EHG.

Andrzejewski said...

@Rob “ 'The most recent dates of ~7,500 BCE were inferred in eastern Europe in Ukraine HGs,''
Should specify that as Ukraine_N HGs”

Are Ukraine_N HG the Bus Dniester or the Dnieper Donets? Did some of these group get assimilated into Yamnaya and Sredny Stog? Maybe it was subsumed in the 15% European Farmer admixture in Yamnaya?

Andrzejewski said...

@Davidski “ o you really believe that this Yamnaya-like individual was one of the first people of his type?

Progress 2 PG2001 4336-4178 BCE

Judging by this and similar samples from Vonyuchka, Khvalynsk, and Sredny Stog, we're dealing with a population that formed at least 4,500 BCE.

And that's a very conservative estimate.”

The big question is - was Yanyuchka and Progress ancestral to Sredny, Yamnaya and CWC (Lower Don Cultures), or vice versa? What was the vector of transmission?

Andrzejewski said...

@Rob “ For me, the question isn’t about what we call this population Caucasian, Caspian or Iranian; it’s why did HGs from near-eastern Europe started marrying women from that group en masse?”

Or man, if the J in EHG from Karelia was any indication. Maybe it points to an ANE marker present at one time in both EHG and CHG.

Like I mentioned in an earlier comment: CHG and EHG are *both* derivatives of ANE.

Olympus Mons said...

Steppe freaks...

"...Following the Corded Ware culture, from around 2,800 to 2,300 BCE, Bell Beaker pottery became widespread across Europe (39). Using 19 Chalcolithic and Bronze Age samples, including ten associated with Beaker-complex artifacts, we inferred the dynamics of the spread of the Beaker complex across Europe. We inferred the oldest date of Steppe pastoralist-related admixture was ~3,200 BCE (3600–2800 BCE) in EBA Mallorca samples from Iberia. We note the EBA Mallorca sample is not directly associated with Beaker culture, but qpAdm modeling suggests that this individual is clade with the small subset of Iberian Beaker-complex-associated individuals who carried Steppe pastoralist-related (40). M..."

Indeed regarding elephants in rooms...

Davidski said...

Yeah, Bell Beakers are from Armenia, or something.

Great theory.

Rob said...



''Or man, if the J in EHG from Karelia was any indication. Maybe it points to an ANE marker present at one time in both EHG and CHG.''

Everyone brings it up, but I don't think the one or 2 Y-hg J in northern EHGs like Popovo explains much per se about the dynamics of EHG <-> CHG interaction in the Volga-Caucasus steppe. It's merely a marker which alerts us to the fact that Caucasus-related groups were in contact with EHGs, but this occurred in specific phases since the Epipaleolithic, not a continuous & constant flow. Hence the key is figuring out what was happenening circa 4500 BC
The rise in CHG on the steppe reflects the encroachment of C-T farmers toward the Dnieper & emergent Sredni Stog network rather than an imagined migration from south of the Caspian, but based on pre-existing sporadic links between HGs in the Don & Causcasus regions



“ Like I mentioned in an earlier comment: CHG and EHG are *both* derivatives of ANE”

Don’t think so. EHG obviously has 40-60% ANE; but Im not seeing a huge amount of ANE in CHG. They might be closer to Natufians, but not 100% sure yet



''Are Ukraine_N HG the Bus Dniester or the Dnieper Donets? Did some of these group get assimilated into Yamnaya and Sredny Stog? Maybe it was subsumed in the 15% European Farmer admixture in Yamnaya?''


We don;t have any Bug-Dniester data. Whilst relevant in the very early phase of things, they are not relevant for Sredni Stog, because their territory was swallowed up by C-T after ~ 4800 BC. My guess is they might belong to Y-hg C, R1b-V88 and non-Yamnaya variants of I2a.

Technically, Vovnigi & Volnienski are Sursk stage 2, coincident with Azov-Dnieper stage Ib, whilst Dereivka I is Dnieper- Donets. I think the classic DD-II (rich in R1b-V88 in Dereivka_N) in the forest-steppe wasn't contributory to Sreni-Stog (''poto-Yamnaya''), but the Sursk clearly were. To evaluate their role at a G-W level, we need 55-4500 data, we cannot base it on Yamnaya genomes. But, yeah their lineages constitute something up to 50% of Balkan Yamnaya , and likely earlier steppe related groups like Cernavoda.

Also the Donecs group in the Dnieper-Don region, is considered part of DDII, but often marked separately. Earlier, they had links all the way to Poland, but at some point during the Neolithic they switched toward the Caucasus as major focus


Simon Stevin said...

This is the direct carbon date of the Mallorca EBA sample I4329: 2468-2294 calBCE (2383 calBCE). That high end dating of 3200 (3195) BCE comes from DATES. This is DATES according to the recent Chintalapati and Patterson paper: “While several methods exist for dating admixture in contemporary populations, they are not suitable for sparse, low coverage data available from ancient specimens. To overcome this limitation, we developed DATES that leverages ancestry covariance patterns across the genome of a single individual to infer the timing of admixture. By performing simulations, we show that DATES provides reliable results under a range of demographic scenarios and outperforms available methods for ancient DNA applications.” So this is a theoretical model, not direct, reservoir calibrated, carbon dating of an individual specimen, unless I’m mistaken.

Regardless, DATES has earlier or contemporaneous steppe admixed samples:

Sweden_BattleAxe.SG, (WHG + Anatolian_farmers)/ (Yamnaya + Afanaseivo), 2537 BCE (average), DATES_mean: 3209 BCE

Denmark_MN_B.SG, (WHG + Anatolian_farmers)/ (Yamnaya + Afanaseivo), 2619 BCE (average), DATES_mean: 3151 BCE

Estonia_CordedWare.SG, (WHG + Anatolian_farmers)/ (Yamnaya + Afanaseivo), 2539 BCE (average), DATES_mean: 2931 BCE

Germany_CordedWare, (WHG + Anatolian_farmers)/ (Yamnaya + Afanaseivo), 2340 BCE (average), DATES_mean: 3068 BCE

Poland_CWC.SG, (WHG + Anatolian_farmers)/ (Yamnaya + Afanaseivo), 2733 BCE (average), DATES_mean: 3013 BCE

Lithuania_LN, (WHG + Anatolian_farmers)/ (Yamnaya + Afanaseivo), 2672 BCE (average), DATES_mean: 3512 BCE

DATES vs direct carbon dates. Checkout these disparities:

Spain_BA_merged, N: 10, (WHG + Anatolian_farmers)/ (Yamnaya + Afanaseivo), 1388 BCE (average), DATES_mean: 2508 BCE

Portugal_MBA.SG, N: 4, (WHG + Anatolian_farmers)/ (Yamnaya + Afanaseivo), 1599 BCE (average), DATES_mean: 4231 BCE

Sweden_LN.SG, N: 2*, (WHG + Anatolian_farmers)/ (Yamnaya + Afanaseivo), 1990 BCE (average), DATES_mean: 3306 BCE

Denmark_LN_BA.SG, N: 1, (WHG + Anatolian_farmers)/ (Yamnaya + Afanaseivo), 1888 BCE (average), DATES_mean: 4072 BCE

Davidski said...

@Simon Stevin

DATES doesn't date samples, it dates admixture events.

Simon Stevin said...

@Davidski

Correct, which is what I was trying to point out to Olympus the steppe denier. The fact there is such a disparity between the direct carbon dates, and DATES itself, seems suspect to me. One is hard data, the other is a theoretical model.

Davidski said...

@Matt

...there's a reason they've uploaded it to biorxiv now.

Yes, of course, they uploaded it so that they can cite it.

Andrzejewski said...

@Rob “ The rise in CHG on the steppe reflects the encroachment of C-T farmers toward the Dnieper & emergent Sredni Stog network rather than an imagined migration from south of the Caspian”

But why would Tripolye bring CHG rather than LBK-related Anatolian Barcin admixture into the forest Steppe?

Olympus Mons said...

“Yeah, Bell Beakers are from Armenia, or something.”

No. The admixture date to form steppe component, as I told you hundreds of times, occurred by 4500BC when Shulaveri-Shomu (Iran CHG) admixes in the steppes with EHG.
That is what I told you endless times, with early Svobodnoe, Zakubanye culture, early meshoko and the rest is we will see. - So, years later, I am closer than ever from being right. Eventually and hypothetical.

And yes, From Shulaveri to bell beakers. Thought the steppes most likely. Or west black sea (Boian Culture or something).

Now, earliest steppe in Mallorca, a bloody island in the Mediterranean its calls for a little bit of steppist , CWC, humbleness a reasonable person would say.

Davidski said...

@All

I won't allow any more political or quasi-political commentary here.

Olympus Mons said...

Davidski, and just to finish this.
My “demented” assertion at those days (2016?) was that some of the people arriving to Iberia, by 3500BC, when Iberia went from a neolithic wasteland to a very populated region had steppe admix already (and were already R1b…) …. In which sense is Patterson stating Iberia steppe admix is oldest found does not reinforce me?

Indeed.

Rob said...

@ Andrze

''But why would Tripolye bring CHG rather than LBK-related Anatolian Barcin admixture into the forest Steppe?''

obviously Tripolje did not themselves bring CHG, the classic (pre-late Tripolje) variants are Anatolian farmers with ~ 15% WHG. However, C-T & Varna stimulated trade networks which were run by steppe / forest-steppe hunter-fishers from the Dnieper-Don region. This projected the network back as far as the lower Volga and the Caucasus, which was a major source of obsidian. Upon arrival to the Caucasus mountains, the Dnieper-Don HGs married CHG-related women and took them back home (~ simply put). But these Caucasus HGs were not the isolated archaic Kotias-type. They had already been in contact with EHGs and even Central Asia. Hence Progress Eneolithi score ~ 20% Iran-N & WSHG like ancestry in addition to CHG & EHG


Davidski said...

@Olympus Mons

DATES provides admixture dates, with no information where they happened.

The admixture that you think happened in Iberia probably took place in Central Europe, still within the Corded Ware horizon.

And no, Eneolithic steppe doesn't have anything to do with Shulaveri Shomu, and neither does Yamnaya or Bell Beakers.

Rob said...

Funny to see OM still deliriously claiming that BB came via Northern Africa

Davidski said...

Blast from the past. This was almost seven years ago. Not a bad conclusion for its time...

My view at the moment is that a population very similar to the teal samples formed in Central Asia or the North Caucasus during the Neolithic as result of admixture between MA1-like and Near Eastern groups. This population, I believe, then expanded into the Pontic-Caspian steppe by the onset of the Eneolithic.

https://eurogenes.blogspot.com/2015/04/the-teal-people-did-they-actually-exist.html

Simon Stevin said...

What is the difference between Iran_N and CHG? Iran_N seems to have more ANE than CHG, but both are mostly Dzudzuana related correct? Dzudzuana is right below the steppe (same as Satsurblia and Kotias Klde), so CHG-like ancestry has been proximate to the steppe for 30k years at the least. WSHG has East Asian admixture, so the additional ANE affinity in Progress, Vonyuchka, and Khvalynsk, has to be older. Perhaps there are Mesolithic/Neolithic, proto-WSHG groups without the East Asian affinity. Aren’t WSHs closer to CHG anyway? I don’t really see the point or need for admixture from Iran, when Caucasia is right there. A mixture between ANE, Dzudzuana, and Epigravettian HGs, could have occurred on the Pontic Caspian steppe itself, and Central Asia no?

Steppe said...

I found another mistake in Krause's book, the Iranian farmers brought their genes across the Caucasus to the Eurasian steppe in this region, they mixed with the local hunters and gatherers within that mixed population, Yamnaya arises, one mistake after the other, geographically and genetically Unfortunately, the problem is that people who don't have access to Eurogenesblog or Anthrogenica believe that. The book is good in many things, but rather flawed in the area of ​​the origin of the steppe cultures.

Davidski said...

Years ago the Iranian farmer scenario was possible. It's just that now it looks like a very low probability bet.

So it depends when Krause penned his manuscript. Maybe he started writing it years ago, and didn't bother updating.

KingOfGondor said...

There's a reason why the Pontic-Caspian steppe is named so. The Caspian sea lies on the border present-day Asia and present-day Europe. The Yamnaya horizon stretched from (what is today) Ukraine to (what is today) Kazakhstan, so they can't be considered European by today's definition nor by the definition of Europe in any past era either.

Also, the terms Europe and Asia had zero meaning in the 4th and 5th millenium BCs, so labeling people who lived during those eras as such is a special kind of stupid. Even if we limit the Yamnaya population to those who resided in the Ukraine at the time, your labeling makes no sense. By that standard, the European residents descended from farmers who migrated from the Levant to Europe during previous millenia (i.e., EEF) should be called Middle-Easterners or Levantine. Since their genome dominates both the hunter-gatherers and Yamnaya in southern Europe, perhaps southern Europeans should not be called European any more, eh?

Davidski said...

@KingOfGondor

You're a moron.

Ukraine is in Europe, and so are the nearby parts of Russia north of the Caucasus and west of the Urals.

So Yamnaya is a European archeological culture.

Davidski said...

Matters of (basic) geography

https://eurogenes.blogspot.com/2017/06/matters-of-geography.html

Haha.

Steppe said...

The book "Hybris" is from the end of 2021 and mostly up to date genetically but not renewed or misrepresented in some aspects, he also said to me he is mainly only interested in genetics, archeology and geography, he talks to colleagues, it's a shame that some scientists don't broaden their horizons a bit, I personally think it's good to bring archaeology, archaeogenetics and history together to have a better understanding of these things.

Gaska said...

This paper is the last attempt by Patterson and colleagues to keep their steppe theory alive. The disaster of the dating of the CWC sites in Bohemia (3,000 BC), and the appearance of the Narva signal in all R1b-L51 males made Papac et al had to liquidate the Yamnaya culture as the origin of this lineage and look for an alternative in the forest steppe. Then they had to resort to Heyd (2022) to justify the possible archaeological linkage of the CWC with Ukraine and to Patterson to make the steppe signal in mainland Europe more ancient, i.e. this paper only tries to shore up a building that is still collapsing.

The funniest thing is that many people had previously noticed the existence of this signal on the CWC. It was detected by Fernandes in Poland, Narasimhan in Czechia and also in Germany and Sweden. By the way, I also remember Davidski modeling the CWC with this Baltic signal and wondering why it disappears later on BBC- They were not wrong, however in those days nobody gave importance to the existence of that signal. Why now it is the main argument to liquidate Yamnaya, when it is evident (both by the male uniparental markers - R1a, Q, Z2103 and female-exact coincidence of mitochondrial lineages, Autosomal DNA) that the emigration to Bohemia had its origin directly in the steppes? Very simple, because that migration did not involve L51 and they have had to look for a moderately credible alternative (steppe forest)

The Yamnaya disaster was already known to my friend Davidski and me more than two years ago (You can read the comments in Is Yamnya overrated?), all the kurganist alarms went off because the data were very difficult to reconcile with the steppe theory. The Czechs have taken more than two years to publish the work, have increased the number of samples analyzed and have looked for a reasonable explanation for what seemed inexplicable. Well, at the end of the day, that is what every good scientist does, trying to prove his theory with all the arguments within his reach. But the battle continues, because now (forgotten Yamnaya, thank God), they have to find L51>L151>P312 in the Ukrainian forest or in any deposit of C-T, Dereivka, S Stog etc, and this is still far from being achieved.

I think they will not find it and if they do it will be related to the WHGs of the Baltic (look at the autosomal composition of the Ukrainian HGs), it is evident that there were population movements to the easternmost regions (Volosovo may be the solution to the dilemma).



Davidski said...

What's with these surprise cameo appearances after years of banishment?

Gaska said...

Because you know that the ban (at least mine) was unfair and unjustified and I must be a masochist because I miss these discussions

huijbregts said...

I was surprised to find Matt yesterday explaining the place of nMonte in genetic history. I was not aware of this all.
Hopefully David will permit me a few supplementary remarks.
I wrote nMonte as an experiment because I was curious whether a simple random walk algorithm could identify relevant groups.
I was pleasantly surprises when it did. Actually it identified a single set of non-unique samples; next it used the naming labels within this set to infer distances to well known predefined groups.
I am still surprised that the simple trick worked so well.
By then (about 2015) I knew next to nothing about mathematical genetics.
By now I better understand the problems with this kind of algorithms.
In the first place there is the problem of overfitting. The ancients division of Global25 contains some five thousand samples. This permits no more 12 binary choices (because 2^12 = 4096).
And if you select a subset, the number is still smaller. So if you use 25 dimensions, you are are heavily overfitting.
Many guys have tried to repair this by by what they call "scaling" the data, but which is really an anti-scaling and does not solve the problem of too many dimensions. It is much better to truncate the dimensions at a much lower value than 25.
In nMonte3 I have limited the damage of overfitting by applying a penalty on larger distances. Unfortunately I also offered the opportunity to switch off this penalizing by using the option pen=0. In spite of my repeated warning not to use this, if you do not perfectly understand what you are doing, many users interpreted this as an opportunity to prove there expert status :(

If I were younger, I might try to improve nMonte by using a Bayesian algorithm.
As it is, I incidentally use nMonte as a quick and dirty simple method.
But mostly I am content with visualizing the data with algorithms like UMAP, which I think is underestimated, especially at this forum.

Matt said...

At this point I guess I find Gaska's theorizing on R1b-L51 pretty strange.

At the start, years ago, he was like "R1b-L51 is from SW Europe".

Then "It must be from Germany, where it was encountered by expanding steppe groups".

Then "It's from Narva related group from NE Europe or Western Russia. But definitely not the steppe. Even though shares a relatively recent common ancestry with Yamnaya-modal Z2103."

What is the point of this constantly moving origin story? At some point, at best - and this is still unlikely - some location in Western Russia that is not the PC steppe could be found for R1b-L51, or R1b-M269 before the steppe expansion, and then "Aha, gotcha!", even though this location will be far closer to the PC steppe than anything originally proposed, and will still have been brought to Western Europe by the same wave movement? Continue to propose these evolving scenarios if you will, but to combine this with the tone of "Aha, actually I was right all along and the academic conspiracy was wrong!" is just weird.

Copper Axe said...

@Davidski

I've started up my own cheeky little blog a while ago and I've made a post in collaboration with Altvred regarding this article:

https://musaeumscythia.blogspot.com/2022/01/when-did-western-steppe-herder-genetic.html

If Nick Patterson is reading these comments I hope that this reaches him as well. I would really like to see this points adressed during the peer review.

Cheers!

Copper Axe said...

"What's with these surprise cameo appearances after years of banishment?"

Genome Wars: The return of the Iberian refuge hypothesis

Davidski said...

@Copper Axe

https://musaeumscythia.blogspot.com/2022/01/when-did-western-steppe-herder-genetic.html

If Nick Patterson is reading these comments I hope that this reaches him as well. I would really like to see this points adressed during the peer review.


Yep, nice effort all round.

Matt said...

@huijbregts; Scaling is just using the eigenvalue data provided by the PCA the data to restore the property where distance computed from in the data to represent the underlying frequency differentiation by the PCA.

Unscaled data simply "throws away" this eigenvalue data, and distances in the data matrix do not represent the underlying frequency differentation captured by the PCA.

Prive, as I understand it, "scales" his data in his preprints where he relates distance across PCA dimensions to Fst, how it can be used to define subsets of participents in UKBiobank and other datasets, and where he describes how polygenic risk scores (PRS) accuracy falls off with increasing distance. Peter "scales" PCA data in the preprint where he relates PCA distances to f-stats.

If either had used unscaled data, there would be no or far reduced relationship of either of these well established measures of genetic differentiation to PCA distances.

If an academic consensus emerges on the relationship of distances on PCA to mainstream established population genetics measures of differentiation, it will not be based on unscaled distance.

If you have an interest in understanding these topics, I would invite you to read these preprints.

Gaska said...

NO Matt with all due respect, don't try to confuse the situation. I have written dozens of posts about L51 and I have always left open the possibility that its origin is in any European Neolithic culture (the case in Smyadovo is an example), in the Baltic HGs, even in the Iron Gates HGs. What I have always said is that I was certainly not in Yamnaya and it seems that I was right. The rest of the story doesn't matter to me as much, but I'm still curious about the exact origin of my lineage.

If you remember the discussion with Davidski about Volosovo and what we knew about R1b-M269 in Russia, it is clear that everyone accepted the possibility of finding that marker much further north than the steppes. It's definitely a possibility, but in my opinion if that's true then the move west has nothing to do with the CWC. The cases in Bohemia have no associated grave goods, only one sample (U106) has a type A battle axe, and I suppose you know the origin of that type of weapon, right? It is not bad to expose these doubts and debate them, I have always said that if I am wrong I will admit my mistake as CQ has done (to Davidski's satisfaction, I supose). Many other people are unable to admit that they were wrong and I think you are one of those people.

In the recent paper on the Argar culture, there is a sample of R1b-L51 dated to 2,800 BC, although Villalba (or rather Max Planck) has excluded it from the analysis saying that since it has a steppe signal we are facing an error (How do you say?, Are you serious? Is that a scientific argument?)-That site is a typical Neolithic site of the Spanish Mediterranean coast without traces of BB culture or anything like it. Doesn't that seem weird to you?

Now the steppe signal, according to Patterson, reached Mallorca in 3,200 BC???? really? Where did this mixture occur, in Central Europe? - The use of DATES to give more antiquity to the steppe signal is only an attempt to cover the backs for future cases of steppe signal in mainland Europe before 3,000 BC

By the way, you could tell Razib that he shouldn't insult anyone for disagreeing with his Harvard friends. I suppose you will agree with him, but I have never insulted anyone for having different opinions.

Matt said...

@Gaska, perhaps I recall incorrectly, but anyway, not much point contesting that and arguing over it. Anyway, I don't think it's likely that an expansion of R1b-L51 is related to an expansion of a hunter-gatherer culture from NE Europe that happens at the same time as the expansion of steppe ancestry and is undetected save as a subtle excess of HG ancestry in some early 3rd millennium samples with steppe ancestry, or some samples with probably a dubious date and context. I don't really talk to Razib except as by comments on his blog, so probably could not convey that message - maybe David would be so kind as to? IDK.

Vladimir said...

Maybe what they found 4400-4000 BC is the moment of mixing of the steppe Eneolite CHG/EHG with the Ukrainian Neolithic/Mesolithic WHG/EHG? If they write that they catch only the last mixing, then EHG is mixed twice, first with CHG, and then as part of WHG/EHG.

Simon Stevin said...

@Matt

He’s either lying or he can’t read. I went over the Argar culture paper, and I dug into the supplementary files. Not a single R1b-M269 sample directly dates (C14) to a range above 2300 BCE. The oldest L51 sample on record is the Afanasievo sample I6222 (R1b-P310*). I’ve seen the bam files, and there are no signs of contamination on the Y-chromosome. All the calls were consistent for both derived and ancestral SNPs, and there were no downstream calls past P310*. Downstream calls would have been a good sign of contamination. That and it’s practically impossible for a modern male to carry P310*. Furthermore, his autosomal profile aligns with the other Afanasievo samples. Corded Ware (including SGC) has V1636 like Yamnaya_Caucasus, Khvalynsk, and Progress (and possibly Srubnaya). CW has Z2103, L51, and Q1b-Z5902 like Afanasievo. The denial of these facts is insane. The links are completely obvious, and I haven’t even mentioned the archeological, linguistic, mtDNA, and autosomal DNA evidence.

vAsiSTha said...

@copperaxe

Congrats on the blog. There are many things wrong with your article.

1. pg2001 is charcoal dated to 4300bce not 4900bce.

2. The dates reference model doesn't simply use 5 EHG samples and 7 iranN samples, but rather a pooled sampleset with many related samples as well.

So in ehg_pooled, khvalynsk, ukr_meso and few others are included. In iran_pooled- ganj_dareh, areni_c, geoksyur, sarazm, tepe hissar etc are included.

They ensure that they get a working qpAdm model with ehg_pooled and iran_pooled with acceptable p-value, and then with same samples they use dates.

The advantage being that more the samples, better the dates results.

Andrzejewski said...

@Vladimir “ Maybe what they found 4400-4000 BC is the moment of mixing of the steppe Eneolite CHG/EHG with the Ukrainian Neolithic/Mesolithic WHG/EHG? If they write that they catch only the last mixing, then EHG is mixed twice, first with CHG, and then as part of WHG/EHG.”

What would be interesting is to solve the puzzle of what population was the first speaker of anything ancestral or resembling a PIE speech: was it the Progress/Piedmont tribe, Khvalynsk, Samara HG, or was the PIE formed only after mixing with Ukraine Mesolithic & Neolithic HG and some EEF.

Gaska said...

@Matt

The Narva signal in Bohemia is the key to understanding this situation, remember that some of the most steppe (and ancient) samples are women (Papac, 2.021)-After reading that paper, direct emigration from Yamnaya (not just any forest-steppe culture) to Bohemia seems clear to me. And that was around 2750 BC. There are some cases of R1a in the CWC older than that date (Poland ?) but in no other European region has R1b-L51 appeared except in Bohemia (the others are very late samples). If the migration from the steppes was joint, why have they not found L151 in Poland, Moravia, the Baltic, etc. Doesn't it seem strange to you? There is an obvious connection between the late neolithic cultures of central Europe and the CWC and perhaps (just perhaps) groups of northern WHGs reached Bohemia where they mixed with women from the steppes (90% of the CWC markers in Bohemia are typical from Yamnaya-Afanasievo and the rest have been documented in the Ukrainian Neolithic)- Or maybe those L51 coming from the forest steppe or Russia reached Bohemia 300-400 years before the Yamnaya.

Do you think that U106 has its origin in Ukraine? I think it has its origin in southern Scandinavia, perhaps Jutland or even Holland. A move to Bohemia cannot be ruled out and if finally L51 appears in Ukraine we will celebrate it anyway

Fair enough, I was surprised by Razib's insult

Rob said...

One could find CHG as early as Epipaleolithic, there's archaeological clues to the possibility. But the main diffusion & acceleration of its % in the steppelands was late Neo- early Eneolithic

Genos Historia said...

Hey, Rob here is my response to the critic of my video you made last week.

So, in my video please understand I didn’t say LBK totally replaced HGs. I said HG populations almost definitely survived segregated from farmers in different niches, to an extent but not as much as you believe.

I also made it clear that I believe LBK forcefully took land from many HGs and as a result “replaced” many of them. I stand by this claim.

The argument I will give you, is the same argument I gave in the last part of my video for why all the migrations are examples of invasion and conquest.
The argument is simple. I’ll apply it to this specific case of LBK.
The fact is HGs did live across central Europe before LBK. LBK did not enter a virgin land. There’s no way to explain how LBK occupied their land without LBk coming in forcefully. There’s no way to explain not just how LBK occupied their land, but how they did so in such a huge way without coming in forcefully.
LBk had huge population size, they built villages everywhere, their expansion was fast. There’s no way to explain this without LBK forcefully kicking HGs off land.

“What you (and some geneticists) are doing is looking at LBK - who are the immigrants / descendants of Balkan farmers and claiming they replaced WHG (the natives) without sampling the latter”

I disagree, the reason we only have LBK DNA from a certain period, is the LBK culture is all there is in the archaeological record after 5500 BC. They are the only skeletons we can get ancient DNA from. The HG presence pretty much disappears. Once again, the best answer for this, is many HGs had been replaced.

If HGs thrived in different niches, as you suggest, where is the archaeological documentation of this?

I mentioned in my video. How, the sudden disappearance of old cultures from the archaeological record supports the idea of population replacement. It supports the idea the old villages had been destroyed and replaced by new ones. Which is what archaeologists call cultural change.

Genos Historia said...

“Secondly, TRB is not from western Europe, but a heterogeneous pan-regional horizon. Don;t worry about your 'G25 models' simply look at the cacophony of uniparentals in Czech TRB and their absence in western Europe.”

I’ll make a video down the road, to finally set the record straight on western vs eastern farmers and how TRB is of western European origin.

I don’t have time to get into the evidence right now. Besides, I’ve presented the evidence to you before but you ignored it.

Also, heterogenous pan-regional horizons didn’t really exist in prehistory. Populations and population movements did. Pots=people. This is the best way to explain the new western European farmer ancestry in TRB.

Gaska said...


@SS

Lying?, you are obsessed with the P310 argument in Afanasievo (perhaps because it is the only argument you have or maybe Mr Rocca has given you some idea to defend that this sample is not contaminated). You can believe what you want, but Mongolia is not the explanation for the origin of that lineage. Remember that Afanasievo is 100% Yamnaya and Papac and colleagues have killed the possibility that Yamnaya has anything to do with L51 ergo Afanasievo is not the explanation either. And of course the sample is contaminated, everyone knows it.

Regarding the paper on El Argar, I think you are also wrong. All the samples from Cueva de la Lechuza are dated to 2800 BC and only one is R1b-L51. The grave goods are typical of the Mediterranean Neolithic and there were no intrusions from other cultures. Why deny that L51 can appear sporadically in other regions of Europe ???. Some very rare samples of R1b-M269 and U106 etc appear in Patterson's paper on GB (some poorly dated and others dubious)

Matt said...

@Genos; do you really believe that the genetics support that when new pottery styles emerge in the European archaeological record, we see genetic turnovers? Did Food Vessel Pottery in the British Isles mean a genetic turnover, etc? Did the plain pottery of El Argar indicate a genetic turnover? Etc.

If that's not your opinion though, what do you mean by "pots=people in prehistory"?

Tigran said...

@Does ANE's western ancestry come from Europe too as opposed to Iran-Central Asia? This would make the roots of Ymanaya's founding population even more rooted in Europe.

What about the IUP/East Asian ancestry? Wasn't the younger Bacho Kiro sample some sort of y P?

Copper Axe said...

Also...

"Thus, we combined all early Steppe pastoralist individuals in one group to obtain a more precise estimate for the genetic formation of proto-Yamnaya of ~4,400 to 4,000 BCE (Figure 2). These dates are noteworthy as they pre-date the archeological evidence by more than a millennium"

Its been a few days and I'm still confused about this statement. Archaeological evidence for what exactly? The maturity of steppe pastoralist lifestyle fueled by the introduction of the wagon?

Has anyone ever suggested that the Yamnaya ethnogenesis was one of tribes from different environmental niches (Caucasus mountains, steppes, forest-zone) all coming together, thus forming the Yamnaya culture? Prior to ADNA at least.

They quoted David W. Anthony there but in his book he pretty much suggested that the Yamnaya was a homogenization of various eneolithic steppe pastoralist cultures. In general the position has always been that these steppe pastoralist cultures were derived from the foragers of the eastern european steppes, which has been confirmed. Yamnaya genrally positooned to have developed out of Sredny Stog, Khvalynsk or both.

So even in the event that they are correct here, I dont see how this is noteworthy as it predates the archaeology of the Yamnaya. The Yamnaya were never a 'starting point' to begin with.

It is noteworthy however, because you dont have any evidence of such migrations to begin with. Not in 3000 BC, not in 4000 BC, not in 5000 BC.

Simon Stevin said...

@Gaska

1) not an argument. You haven’t provided any evidence as to why the I6222 calls should be disregarded. That contamination only effected the X-chromosome, not the Y-chromosome, not the Y-DNA SNP calls, or the autosomal DNA. Contamination is measured via the X-chromosome. I6222 clusters autosomally with the other Afanasievo samples. No modern male contaminated it with P310*. That’s next to impossible, nor did a modern male alter I6222’s autosomal profile, an ancient profile which doesn’t exist anymore. There were no downstream calls, and all the derived and ancestral SNPs were consistent.

2) CW and Afanasievo share L51, Z2103, and Q1b2a (Z5902). CW shares V1636 clades with Khvalynsk, Progress, Yamnaya, and potentially Srubnaya. They all have a common origin in Eastern Europe, on the Pontic-Caspian and East European forest steppes, located in European Russia and Ukraine. The oldest U106 comes from early Bohemia CW as well. Corded Ware and Yamnaya descend from the same CHG + EHG population. This profile only existed in Eastern Europe prior to the LNEBA. R comes from the ANE side of EHG, and R1b/R1a are from Eastern Europe. CW and Yamnaya are roughly contemporaneous (neither derives from the other) but they share a common origin, and that means common lineages. Hence the shared links between CW, Yamnaya, and Afanasievo. L51 comes from the PIEs/WSHs of Eastern Europe. L51 has a common origin with Z2103 at roughly around 4100 BCE; that’s right around the time of Sredny Stog and Progress.

3) The Narva HGs all belong to Y13200/M73 (the ones that could be derived past P297 that is). None are V88 or M269, while I4550 is Q1b-Y2700. These lineages obviously come from their EHG/ANE side (R and Q that is), not their WHG side. Furthermore, the Bohemia CW paper stated that the additional HG in CW, could have come from any WHG/EHG admixed, surrounding group, like the Ukraine_N pops (that points to Sredny Stog), not just Latvia_MN/Narva: “We find that when either one of Latvia_MN, Ukraine_Neolithic, or PittedWare is added as a source, almost all (280 of 285) model fits (P values) improve and most of them by several orders of magnitude (table S17). While all (n = 95) two-way proximal models lack strong support (P < 0.05; table S17), the addition of either Latvia_MN (57 of 95 supported models), Ukraine_Neolithic (53 of 95 supported models), or PittedWare (32 of 95 supported models) to the sources drastically increases the number of supported models (table S17). These results show the presence of excess Latvia_MN/Ukraine_Neolithic/PittedWare-like ancestry in Bohemia_CW_Early relative to all known Yamnaya and Central European Neolithic groups.” In general, CW has more EEF/WHG than Yamnaya (20-35% vs 10-20%).

3) You are completely wrong, and you didn’t read the supplements. Read the paper and the supplementary files. The Copper Age samples range from 3300-2300 BCE archaeologically/indirectly, but not directly, as in direct carbon dating (C14). That’s where you probably got your bogus 2800 BCE figure, by choosing the middle of the indirect range. None of the C14, directly dated samples with R1b-M269, date past 2300 BCE.

LivoniaG said...

Genos Historia said...
"There’s no way to explain this without LBK forcefully kicking HGs off land."

Genos Historia - the Dutch digs have shown that the LBK settlement stuck with loess soil, TRB sites were on sandy soil. The HG genetic component is in TRB is large.

Hunter-gathering doesn't use land the same way agriculture/food production does. Guys like Lewis Binford showed that mesolithic life-styles were very tied to routes and concentration of resources rather than land. (see Binford's Willow Smoke and Dogs’ Tails: Hunter-Gatherer Settlement System)

Hunter- Gatherers strategies in temperate climates mean people are constantly moving. So moving to non-farmed land would be quite natural. Especially because loess soil can be a stagnant environment for hunting and fishing, HG life-style groups might not have been all tat interested in the land. Though the attractiveness of food production apparently did inspire selective breeding of food plants more adapted to the coastal sand of NW Europe.

epoch said...

@Genos

"I disagree, the reason we only have LBK DNA from a certain period, is the LBK culture is all there is in the archaeological record after 5500 BC"

Have you forgotten about the hunter-gatherer from Körös?

Matt said...

@epoch, there are of course a few HG related samples which exist broadly in Central Europe up to at least the 4th Millennium, with 50-80% HG ancestry (Czech Republic and Germany here: https://imgur.com/a/QIMHtgU ). So I assume that there is some other question at work in Rob and Genos's discussion than whether HG people simply survived in Europe at all post the arrival of the farming horizon. (Because obviously they did).

@CopperAxe, earlier in the paragraph they referenced "The archaeological record documents that the early Steppe pastoralists cultures of Yamnaya and Afanasievo, with characteristic burial styles and pottery, appeared around ~3,300 to 2,600 BCE (35)", as the specific archaeological evidence they're talking about, so that is presumably what they are referencing (not wagons).

I agree though it doesn't seem particularly surprising that a genetic integration of separate groups might form over 1,000 years before a particular characteristic burial or pottery form might appear (or more, if there are reasons from a complex scenario that the method is lowballing things and this actually began earlier). So they could use more language in the paper to explain why they think it's noteworthy. It's not noted as noteworthy in other groups, so why here?

I think they might be saying this as a contrast with how they think there's a close correlation between admixture time and emergence of earliest Corded Ware cultures in North and Central Europe. In that case the "pots" (shorthand for all material culture, including pottery and burial traditions) began to emerge at more or less the same time as the admixture is inferred, and the "people" emerged. Indicating that perhaps the emergence of the material culture can be linked to the phenomena of admixture. Which is not the case for Yamnaya; the emergence of the distinctive Yamnaya material culture (the "pots") is not happening when the admixture is happening. So is there some other distinctive material culture that did emerge at the inferred date *or* not? Again though, this is not noted considering other groups they infer, e.g. SHG? (Although I think Torsten Gunther and others have suggested that the material culture of SHG does look like the product of combining waves?).

Arza said...

@Davidski
I was about to post a comment under a new preprint at bioRxiv, but the comment section isn't there anymore. Hopefully, this is just a temporary glitch.

You need to click on the comment counter (above the "Abstract" tab).

Rob said...

@ Genos

I can understand the intention of your video - it is a jab at the new school western archaeologists who overplayed the belief that migrations did not occur, or that past populations did not have a collective identity which can be reflected in the archaeological record (like 'pots'). But you have misunderstood the presmise and over shot the mark
Such views remained, in extreme, a minority view, and in your rush to negate it, your video has become a caricature, and that's fine if the intention is to make your fanboys laugh, but don;t tell me that you're right when that's not quite the case.
For ex;


''I also made it clear that I believe LBK forcefully took land from many HGs and as a result “replaced” many of them. I stand by this claim.
I disagree, the reason we only have LBK DNA from a certain period, is the LBK culture is all there is in the archaeological record''

This claim is not correct. It's one thing to not know, it's another to continually ignore lessons. To repeat, LBk colonized virgin lands which were not populated by HGs. I even linked a nice simple image for you, but it seems you also failed to understand that . Here is another map by same author (Marek Nowak), a polish archaeologist. The Green areas are LBK formed enclaves in fertile soil lands, whilst the sea of red around them were HGs areas (more vast, but lwoer desnity than nucleated LBK villages).
Of course, I this is not to suggest there was no occasional conflict, but as a whole your theory fails misrepresents the LBK colonization.

What many people, incl. 'genetic authorities' often struggle to understand is the difference between ''population replacement'' and initial avoidance amongst distinct cultures/peoples.




'' How, the sudden disappearance of old cultures from the archaeological record supports the idea of population replacement. It supports the idea the old villages had been destroyed and replaced by new ones. Which is what archaeologists call cultural change.''

This is incorrect. You haven't even understood the basic cultural progression.
LBK 'dissolved'' ~ 4900 BC. TRB did not appear for another 800 years. Your claim that TRB 'replaced' LBK is therefore nonsensical

LBK was followed by a series of 'post-LBK' cultures like SBG, Gatarsleben, Rossen, Lengyel. They are cultural successors of LBK villages, although some like Rossen now show high amounts of Y-hg I2a1 but still very EEF at a GW level
TRB then emerged c. 4000 BC, which then ''collapsed'' c. 3200 BC. Northern Europe was then left sparsely populated again, with some mobile GAC groups roaming around, in turn followed by intrusion of CWC.



Rob said...


''I don’t have time to get into the evidence right now. Besides, I’ve presented the evidence to you before but you ignored it.''


In order for you to present evidence, you would first have to understand the culture-historical basics. Apart from the Dutch 'TRB west' group, TRB is situated entirely east of the Rhine. How can it be 'west European' ?

Secondly, you would need to understand the genetics, and you do not demonstrate that you have, either. Where are your formal models supporting your claim ? In fact, you do not even need that - look at the uniparentals. TRB-related Czech & Polish groups had lineages like R1bV88 & C1a. Baalberg, a German variant of TRB had H2, I2b, I2c. Swedish TRB had I2a1b.
They're actually all different mixture of post-LBK and HG ancestries, mixed in different ways. They even plot distinctively

TRB was definitlively a heteogeneous group. Yes, they have some 'west Euro farmer' admixture, but that's due to social contacts, not due to a west European origin of all TRB groups.



''Also, heterogenous pan-regional horizons didn’t really exist in prehistory'''

Not true. You're just being sensationalist without the 'goods' ;)

Matt said...

Some of the interpretation is a little strange I guess - they comment that "Using DATES, we inferred that the oldest date of Steppe pastoralists gene flow in Europe was ~3,200 BCE in Scandinavia in samples associated with Battle Axe Culture in Sweden and Single Grave Culture in Denmark that were both contemporary to CWC"

That may be robust... But even if so, it doesn't necessarily indicate that the admixture pulse is actually happening at this time *in Scandinavia*. It could (more likely does?) reflect a pulse into one of the branching early Corded Ware groups that actually happened further east (e.g. with Ukraine GAC?), and is just more present in the BAC samples.

Davidski said...

Yes, as I pointed out above, DATES just gives a mixture date and can't tell us where that admixture took place.

Matt said...

@Davidski, yes, true.

Another thing on the same theme I might be slightly critical of is that they say:

"Strikingly, we inferred the timing of admixture in central Europe (Germany and the Czech Republic) and eastern Europe (Estonia and Poland) to be remarkably similar. These dates fall within a narrow range of ~3,000–2,900 BCE across diverse regions, suggesting that the mixed population associated with the Corded Ware culture formed over a short time and spread across Europe rapidly with very little further mixture (Table SC)."

However, when we look at the actual samples within the CWC in Czech Republic, then it looks to me like the fall in Steppe ancestry in the CWC group really does continue across the whole phase of 2900-2500 BCE, and more crucially we do see the presence of female outliers with 0% steppe ancestry (or very little) in the CWC set as late as 2600 BCE and all sampled ones fall between 2870-2620 BCE (with peak density at 2750-2620 BCE), as well as one male late on around 2700 BCE with as little as 30% steppe ancestry.

Visually: https://imgur.com/a/ndzQCXP

It doesn't look to me like a process where you have a few generations with 10% female admixture per generation, and then by 2900 BCE there is just admixture within the culture which homogenizes things down.

Instead it looks like a process where women are being incorporated into the community somewhat continuously over the whole period of at least 2900-2600 BCE.

If it all happened in the one century then we'd probably see some females incorporated into the culture very early, but then not really see any female outliers with no steppe from post 2900 BCE, and instead a number of individuals with low steppe. Whereas what we see is a scarcity of any low sustained low steppe people, even at the beginning of the sequence, and a high steppe main group that drops over time with females with no steppe being brought into the group across the whole sequence. That seems to indicate a process where women with no steppe are being brought in more slowly over a longer time than they suggest.

It's a relatively minor thing but there is some distinction; its a different case if you have a scenario where it's like 1 in 10 marriages per generation for 3 generations, and then things homogenize, than if there's a community that is practicing 2.5 in 100 marriage for 12 generations.

It's also a rare thing where we can see the admixture history unfolding in the samples themselves, so it provides a test case for DATES and it would've been good if they could talk about it.

Gaska said...

@SS

The discussion on CLL007 may be interesting. If you read the paper carefully you will see what the archaeologists say-“The type of collective burial in a cave with the associated grave goods indirectly date the collective necropolis to the Late Neolithic/Chalcolithic of Southeastern Iberia (3300-2300 cal BCE)”-So yes 2.800 BCE. Do you think they are talking about a BB or Yamnaya burial???? Seven males have been analyzed in the deposit, I2a (4), G2a (1) H-P96 (1) and L51 (1). Do you still think it is a BB culture or Bronze Age site? The other lineages seem typical of the Western Neolithic farmers, don't they?

I have found a document about the archaeologist who excavated the site, as it is in Spanish I translate it for you. In his exploration he found under a central slab 4 skulls placed in a disorderly way and very close to each other. Their position was different and all of them were missing the mandible. They were found surrounded by other bones in disorder, and it was observed that some were burned. With these data, the same author believes that in the cavity there were second grade burials, later disturbed by the animals and that there must have been a special concern to protect the skulls. Grave goods-7 silex arrowheads, two polished stone axes, ornamental shells and pottery typical of the pre-BB Spanish Chalcolithic (3,200-2,800 BC). Obviously we have already asked the researchers if they have dated that particular sample, because it is buried with the other individuals and the probability that it is a later burial is very low. It is not common sense that a BB would be buried in the same place and with the same grave goods as a Neolithic farmer. When we know the exact dating I will let you know.

* The sample is also ~200,000 SNPs, so it is not one of the <60,000 SNPs "low" resolution samples-They remove the sample because it contains Steppe Ancestry (their sole reason given in supplementary), and other samples from Copper Age do not. They have not done a contamination test to exclude him-In other words, they exclude a sample without any reason to do so, but this is not a surprise because this has already happened with other samples considered doubtful that have been ignored or excluded when they do not fit the theory that scientists are trying to prove (Switzerland, Germany or more recently in England) Definitely M269>L51 has very bad luck.

Rafael said...

These guys know that CWC and BB are European cultures, and they probably know that about Yamnaya, too. The problem is just that science writers don't often write well.

Simon Stevin said...

@Davidski

So far are all the Sredny Stog samples WHG + EHG, since I6561 turned out to be Srubnaya related (2134-1950 calBCE)? I’ve seen some people claim there’s minor CHG in Sredny Stog. When do you think EEF arrived into the picture, 4000-3500 BCE?

Davidski said...

There's a lot of CHG in the new Sredny Stog samples. Most look like Yamnaya.

But they haven't been published yet.

Simon Stevin said...

@Gaska

That is interesting, however, without direct C14 dating, this just amounts to mere speculation. 3300-2300 BCE is a wide range. Samples can bleed into earlier archaeological layers, as seen with the Shulaveri-Shomu debacle. We need direct C14 dating (hopefully calibrated for the reservoir effect) of CLL007 before we make any judgments. The highest C14 range for an R1b-M269 sample in that paper is 2296-2060 BCE (sample ID: MDP001), which is almost 2300 BCE. Moreover, two Iberian CA samples—CDM001 and CDM002—have yielded C14 date ranges of 2550-2234 BCE and 2579-2346 BCE respectively. This is despite them being archaeologically/indirectly dated to 3300-2300 BCE. Just because a sample is archaeologically dated to a certain period, doesn’t mean its from that period, nor does it mean it occupies the median or high end of the archaeological age range. CLL007–like the aforementioned two samples—could give a C14 date that corresponds to the low end of the age range (2500-2200 calBCE), which would correlate with the initial dispersal of steppe ancestry into SW Europe. C14 dating is the only thing that can confirm it. So far, the hard data does not suggest R1b-M269 entered Iberia/SW Europe in 3300-2800 BCE.

Konstantin Demidov said...

Greetings! I watched this heated discussion. A few comments.
1. As far as I know (I may be wrong) - representatives of the yamnaya culture found branches of r1b, which are practically absent in Europe. That is, the Germans, French, Spaniards, etc. have other branches of r1b. This fact destroys the theory of an invasion from the east.
2. According to scientists, the skills of cattle breeding and agriculture were brought to Europe by Neolithic farmers from the Middle East. And then these key technologies were received from them by groups of hunters in Europe and adjacent lands. r1b did not come from the Middle East - accordingly, they received these technologies directly from European Neolithic farmers (as an option) - or from them, but through the Caucasus. Otherwise, we have to admit that they invented them themselves.

Davidski said...

@Konstantin Demidov

As far as I know (I may be wrong)...

Yep, you're wrong.

R1b-L51 obviously came from the steppe, or alternatively, the forest steppe, but that's not much of a difference.

Early Czech Corded Ware samples practically identical to Yamnaya belonged to R1b-L51.

https://eurogenes.blogspot.com/2021/08/r1a-vs-r1b-in-third-millennium-bce.html

Copper Axe said...

@Matt

"earlier in the paragraph they referenced "The archaeological record documents that the early Steppe pastoralists cultures of Yamnaya and Afanasievo, with characteristic burial styles and pottery, appeared around ~3,300 to 2,600 BCE (35)", as the specific archaeological evidence they're talking about, so that is presumably what they are referencing (not wagons)."


Thats kind of what I mean though, there is nothing really "new" bout the pottery or the burial traditions of Yamnaya and Afanasievo. Repin, Khvalynsk, Sredny Stog and the LDC all share pottery styles and/or burial rites which you later see reflected in the Yamnaya and Afanasievo culture. What sets Yammnaya (and Afanasievo) apart is the relative uniformity of traditions across a wide geographical area, as well as widescale usage of the wagon, which allowed for travels and expansions deeper into steppe territories.

"Which is not the case for Yamnaya; the emergence of the distinctive Yamnaya material culture (the "pots") is not happening when the admixture is happening. So is there some other distinctive material culture that did emerge at the inferred date *or* not?"

Sredny Stog develops in the later 5th millenium BC, Repin forms just after 4000 BC, both of these could have their origins in the Lower Don culture. Khvalynsk develops in the late 5th millenium BC, the steppe_en samples with mounds and ochre usage in burials date to 4400-4000 BC etc. So a whole bunch of distinctive material cultures arose on the steppes, all with some degree of commonalities.

Konstantin Demidov said...

I would like to clarify. Which branch of rb is characteristic of the pit culture??? Modern r1b carriers in western Europe (France, Germany, Spain, etc.) - are they related to THIS branch??

Davidski said...

Yes, Corded Ware R1b-L51 and Yamnaya R1b-Z2103 are closely related subclades.

But more importantly, early Corded Ware people very, very similar to Yamnaya in terms of overall ancestry.

Konstantin Demidov said...

It can be assumed that if two different patrilineal lines pass through the bottleneck in the same period, have a similar autosomal background, have similar technological and cultural characteristics, then most likely both lines just got it all from some third.

Davidski said...

Probably from Sredny Stog.

Upcoming Sredny Stog samples look like Yamnaya and Corded Ware.

Konstantin Demidov said...

There is some uncertainty with the dates in the Sredny Stog - in my opinion, this is not a reliable basis.

Copper Axe said...

@Konstantin Demidov

R1b-L51 and R1b-Z2103 have a common ancestor which likly dates to around 4100 BC, thus within the Proto-Indo-European period.

https://www.yfull.com/tree/r-l23/

However, a divergence of lineages is just a genetic feature and doesn't have to imply a linguistic or cultural divergence. R1a and R1b diverged at the start of the LGM yet you had R1b and R1a carriers in the same ethnolinguistic communities. The picture is a bit more complex also as Yamnaya had I2 as well, there is a Q clade shared between Afanasievo and Corded Ware samples, one Afanasievo sample probably had R1b-L51 and several Corded Ware/Beaker samples have shown R1b-Z2103. Haplogroup J has been seen in Khvalynsk and Afanasievo A Danish Corded Ware sample had R1b-V1636 etc.

Matt said...

@CopperAxe, thanks for your view. Whether there's a break or commonality between burial and pottery practices at this time, I guess can't assess (whether continuity or discontinuity should be stressed). Perhaps they should run that section by DA.

Konstantin Demidov said...

@Copper Axe, I think there was some kind of culture, language and technology donor there. From which hunters r1a and r1b received both language and technology. Having received the technology, they still preserved and, perhaps, strengthened their tribal worldviews. Hence the sharp increase in the population. And the presence of separation, despite the commonality of linguistic and technological continuums.

vAsiSTha said...

@matt

What the paper means is that the admixture happened 4200bce or so in ancestors of yamnaya, but yamnaya itself is dated post 3300bce.

Gaska said...


@SS

According to Patterson the steppe ancestry arrived in Mallorca in 3.200 BC, that coincides in dates with ATP3 (3.400 BC) and of course with CLL007 (which is certainly closer to 3,000 BC than 2,800 BC).The dates are about to come out and then Haak and colleagues will have to think about whether they take the sample into consideration or not, and then we will value their work and their explanations. So we are calm because more samples of R1b-M269>L51 will appear when the appropriate deposits are studied. We are not in a hurry either, we will surely see four or five more versions of the steppe theory in the next few years.

1-Regarding the other comment you made, I think you should also talk to Max Planck and Harvard who are the ones who have definitely decided to liquidate Yamnaya and therefore Afanasievo. Nobody in the international scientific community mentions that sample because it has no relevance, but if you think it is the definitive proof of the origin of that lineage in the steppes, I guess you will look for all kinds of arguments to justify its validity. In my opinion it is a garbage sample and you should read the review process of that paper to understand what I am saying. You have used the same arguments that my friend Rocca has been using for years, if I may give you some advice, I would look for other independent opinions to contrast.

2-"CWC and Afanasievo share Z2103, and Q1b2"-Yeah, you are right. Z2103 is also in the BBC and in many European cultures (Vucedol, Maros, Latins, etc) although in very small percentages. On the other hand Q1b2 is only in the CWC and neither of these two lineages ever reached Western Europe. In fact, there are hardly any cases today.

3-"CWC shares V1636 clades with Khvalynsk, Progress, Yamnaya"-You are right again,everyone has known this for years

4-"They all have a common origin in Eastern Europe, on the Pontic-Caspian and East European forest steppes, located in European Russia and Ukraine"- You are probably right about V1636, Z2103 and Q1b2, although I would recommend you to wait for more samples from mainland Europe, there could be some surprises regarding the first two because their dating in the steppes are not too old

5-"The oldest U106 comes from early Bohemia CW as well"- His archaeological ascription is controversial.Although I suppose that you have already been informed about it, I can recommend some good papers on type A battle axes and their origin. In any case Bohemia is in Central Europe, not in the steppes, so actually its origin is in Czechia.

6-"Corded Ware and Yamnaya descend from the same CHG + EHG population. This profile only existed in Eastern Europe prior to the LNEBA"- You have forgotten the EEF component in both Yamnaya and CWC, because they have different percentages and different origins. This is not a minor issue

7-"R comes from the ANE side of EHG"- Currently, this is what it looks like

8-"R1a are from Eastern Europe"- That's what I believe

9-"CW and Yamnaya are roughly contemporaneous (neither derives from the other) but they share a common origin, and that means common lineages. Hence the shared links between CW, Yamnaya, and Afanasievo"-Yamnaya and Repin are much older than the CW"-The CWC does not have its origin in Yamnaya??? good to know. You know what is the genetic origin of these cultures??? maybe you can help Harvard and Max Planck in their search. I suspect that all steppe cultures in different degrees have something to do with the genesis of Yamnaya and CWC (Khavlynsk, Sredni Stog, Repin, Progress, Maykop etc etc).

10-The Narva HGs all belong to Y13200/M73 (the ones that could be derived past P297 that is). None are V88 or M269, while I4550 is Q1b-Y2700- Yeah, you are right again







Gaska said...

And now comes the interesting part

*These lineages (R1b-Y13200) obviously come from their EHG/ANE side (R and Q that is), not their WHG side and *R1b are from Eastern Europe. I don`t care about Q but I suppose that

You know what the autosomal composition of the Baltic HGs is???,
you know that they are WHGs, right?
have you also checked their mitochondrial markers?,
you know that R1b-M297 is much older in the Baltic??
you know that it has never been found in the steppes??
you know that this marker has been found in Asia thousands of years later than in the Baltic???

If you know all that I guess you would understand that the migrations were not east-west but west-east. Only if you find an older sample in central Asia or in the steppes we would have to change our mind, in the meantime, using the word "obviously" regarding the origin of this lineage in the EHGs sounds comical and nonsensical.

1-The bifurcation of R1 into R1a and R1b occurred 25.100 years ago (95% CI: 21.300–29.000)
2-R1b-P297 is the son of L754 and you know where is the oldest L754 that has been found so far, don't you? I don't need to tell you his name, but he has been eating spaghetti for 14,000 years and belongs to one of the most important WHGS clusters.
3-R1b-P297 is the father of R1b-Y13200 and these two lineages have been found abundantly in the Baltic (Latvia and Estonia) between 7271 and 3270 BC, i.e. 4,000 years of genetic continuity. Of course all of them fundamentally WHGs
4-R1b-V88 is a marker with origin in the Iron Gates HGs, where it has been found with an antiquity of 8.885 BC (Serbia, Padina), I suppose you also know the autosomal composition of these gentlemen, right? Of course it is much older than the Ukrainian samples-Again migrations seem to go from west to east right?.

Summarizing, we have R1b-L754, R1b-P297, R1b-Y13200 and even R1b-V88 in all WHGs groups (Baltic, Germany, Norway, Balkans, Italy, France, Spain etc), with a much higher antiquity than in the steppes or in Ukraine-Russia-Siberia-Asia and your conclusion is that obviously R1b is a lineage with origin in the EHGs. Congratulations

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

@SS you said

*"CWC and Afanasievo share R1b-L51- L51 comes from the PIEs/WSHs of Eastern Europe"-

You will have to explain this personally to Papac and colleagues because they have dismissed the Yamnaya culture precisely because of the lack of continuity in the male lineages which is what we have been saying for many years. Have you read Bohemia's paper or have you simply ignored it or do you not agree with what it says?

and *"Furthermore, the Bohemia CW paper stated that the additional HG in CW, could have come from any WHG/EHG admixed, surrounding group, like the Ukraine_N pops (that points to Sredny Stog................

Any European geneticist would have interpreted the results in Bohemia as definitive confirmation of a direct migration from the Yamnaya culture to Bohemia (autosomally and in certain uniparental markers they are identical) and yet they have rejected this possibility because of that Baltic signal. Evidently they had to look for solutions and the Ukrainian Neolithic may be one.

BUT

1-Steppe-ancestry in CW is NOT EHG-shifted when qpAdm shows that the EHG component is near identical or even higher in Yamnaya

2-It is clear that these are not EEF and not extreme WHG or EHG

3-i.e. we have to look for a population or culture that can provide such a peculiar signal ergo we have to look to the north (Belarus and the Baltic)

The Ukrainian solution could be possible, I won't deny it, but then answer me this question.

If this PWC/Narva like admixture was present in the Forest Steppe before the migration then shouldn't we see it in Fatyanovo or Baltic CWC?

For me, this autosomal signal seems specific to this R1b-L151 group

When the results of Sredni Stog are published we will find the solution, maybe they have already found L51 there and that's why they have liquidated Yamnaya

Rob said...

@ Copper Axe

''Sredny Stog develops in the later 5th millenium BC, Repin forms just after 4000 BC, both of these could have their origins in the Lower Don culture. Khvalynsk develops in the late 5th millenium BC, the steppe_en samples with mounds and ochre usage in burials date to 4400-4000 BC ''

Sredni Stog develops as early 5th Mill mostly from Donetz neolithic group.
According to Kotova, LDC is only partly ancestral to Konstantinovka Eneolithic group.

Copper Axe said...

@Rob

Don't worry I dont get my archaeological knowledge from AG comment chains. Although I cant remember the last time I've seen anyone talk about eneolithic steppe archaeology though...

Ps why are you keeping up with AG opinions arent you banned there? :P

I've seen several authors propose that Sredny Stog (middle, late phase etc) came from the LDC which is why I said may. Also that the Azov-Dnieper group came by way of a LDC and Sura synthesis. Its not some fringe comment chain take such as Yamnaya developing from Maykop. In general I think the much of the archaeology during the neolithic/eneolithic needs more modern reassessments anyways, I dont put all too much stock in all these material cultures, especially if its in Ukraine. I find it more convenient to talk about geography and timeframes.

Copper Axe said...

@Rob

I would really prefer it if you didnt delete 80% of your comments one hour after posting them by the way.

Rob said...

@ Copper Axe
Firstly, I can do whatever I want. Secondly, I presumed that’s where you get your Info from because that’s where you had to recruit a statistician from :) .
But it’s the kind of simplistic theory that those sissies would come up with- it’s a Khvalynsk 2.0; which will also fail

Davidski said...

@Copper Axe

What is that thing in your avatar?

Nick Patterson (Broad) said...

Davidski: Lighten up!

<My impression is that the authors see the people associated <with
the Yamnaya culture as Asians who simply used "far" Eastern Europe
as a springboard to expand into other parts of Europe.

<Davidski

Well this may be your impression but it's dead wrong. It certainly
isn't my belief. And then you go on a rant..

I'm working hard on the 4th and 5th millennium Steppe and for the
genetic origins of the Yamnaya a clear story is emerging, More
(I hope) soon.

Also we have known for years that Afanasievo is genetically Yamnaya.
This (as far as I can see) has nothing to do with dating various
admixture events. This paper is only marginally about phylogeny,


A technical remark. DATES is a pretty good method of determining
admixture dates, and experience shows its usally best to use multiple
samples even if only rather remotely related to the actual addicting
populations. You want to get rough approximations to the allele
frequencies of the sources and small samples will kill you.

Good news: DATES is robust to exactly which sources you use, Bad
news: DATES us a very poor method of determining the true source
populations,

Rob said...

btw, nothing against Altvred at all, he seems like a smart, cool guy. I just dislike the admins and mods at AG.

Copper Axe said...

@Davidski

Its a distant relative of the thing in your avatar :) I initially wanted to use a Yuezhi bust with cranial deformation from Khalchayan but then I figured since your iconic profile picture was of a Saka bust from Khalchayan I probably shouldn't. What if we became enemies like those two?

My avatar is a half-mummified head of a man from the Oglakhty cemetery, Tashtyk culture from around 100-300 AD. It's a pretty cool archaeological site.

THe Tashtyk culture is a successor of the Tagar culture in the Minusinsk basin during the Xiongnu period and collapse, The Tagar is more or less a typical eastern Scythian culture but they differ in that they kept their bronze age houses rather than going full nomadic.

It often has been interpreted as the archaeological culture of the old Kyrgyz (Yenisei Kyrgyz, modern Kyrgyz are a different Turkic people) however there is an archaeological shift between Tashtyk and the Chataas culture after 400 ad, and the Chataas culture is the one which can be directly linked to the historical Kyrgyz, followed by its successor the Askiz culture. It may be that these people kept their original ethnolinguistic identity, which presumeably was Iranian. I think the second option is a bit more likely.

In any case, it is probable that these people were a rather significant substrate of the old Kyrgyz, as their lifestyles were pretty much the same and the early medieval descriptions of the Kyrgyz often point to "red" hair being quite common amongst them, as well as blue/green eyes and other 'western features'like light skin and height. Having black hair was considered unlucky amongst them. Tattoo practises were also common amongst the Kyrgyz. Now given that these descriptions from Persian, Chinese and Tibetan sources they probably were a little exaggerated but its certainly interesting how we have mummies with reddish hair and tattoos during antiquity right where the Kyrgyz lived in the middle ages.

Copper Axe said...

@Rob

"According to Kotova, LDC is only partly ancestral to Konstantinovka Eneolithic group."

I dont think this is true. In her work Ëarly Eneolithic on the Pontic Steppe" Kotova points out the many connections and parallels both the western and eastern Sredny Stog cultures have with the Lower Don culture and attributes both (and the Azov-Dnieper) group to have originated from LDC migrants. She does not once state that LDC is only marginally ancestral to Konstantinovka.

Also why spend 1/3 of your comments here shit talking people on Anthrogenica? I get that your probably pissed at the mods for not wanting you there but I remember you trying join the forum and particpating in the discussions of "sissies" so dont act like its beneath you or you are too good for the place because you clearly want to participate LOL. If anyone should talk shit its Davidski when they temp banned him twice in a row for no good reason.

Rob said...

@ Copper Axe

You shouldn't talk about matters you don't have a clue about-. You weren't even around when I was on AG (you're still a noob in this field). How is it if they mistreat Dave, or ban Vasistha just for posting a qpgraph, that powertrip behaviour by a bunch of dweebs doesn't carry through ? yes Im brash, so what. They take themselves waaay to seriously over there, given their mediocrity. They really need to get a sense of humour.
But what is to learn there ? Most of the threads are boring ancestry posts, a lot of pseudoscience thrown in, and the rest of is 5 years behind the times. Yes there's a couple of big talkers who act like they have scoops on aDNA studies, but they really don't.

So, it's great that you're a prima donna over there, but yes, it is way below me. So stick in you lane & echo chamber. Atta boy

vAsiSTha said...

@all

At which town in russia would you consider europe to end and asia to start?

Davidski said...

Orenburg

https://www.tripadvisor.com.au/Attraction_Review-g1833098-d6694439-Reviews-Pedestrian_bridge_between_Europe_and_Asia-Orenburg_Orenburg_Oblast_Volga_Distric.html

gamerz_J said...

Beyond the CHG/Ira_N/Yamnaya interaction topic, did anyone notice how they modeled Anatolia_N as a mix of Anatolia_HG and Iran_N? Anatolia_HG itself being something between WHG and Natufians apparently?

Would anyone have any thoughts on that, I thought the Dudzuana paper had shown Anatolia_N to mostly form a clade with Dudzuana, and though Natufian ancestry is possible how come there's Iran_N. But then again Feldman (2019) showed Iran_N-related ancestry in Anatolian farmers.

@Rob
CHG being close to Natufians was hinted at a pre-print regarding some Dudzuana-like samples I think, can't find it now but essentially they talked about ancient Caucasian populations 26kya or abouts clustering close to modern-day Levantines.

Also, if I may ask, have you run any graphs with Anatolia_N/Anatolia-HG?

Rob said...

@ gamerz_J

I've only looked at it preliminarily, but AHG can be modelled as something 'old West Eurasian' (or 'basal' according to received models) + 'pre-Villabruna'
On the other hand, some formulations do show ANE/ Yana/ ENA admixture as far west as AHG ~ 10% .
ive not made it to Barcin/ Mentese or the Central Anatolian Neo's as yet

But at face value (looking at the host of additional mtDNA uniparentals and archaeology), I do expect a new population moving in with farming, probably from the Fertile Cresc, as classically described - even though the G-W models from Feldman argue for 90% contnuity

gamerz_J said...

@Rob

Thanks quite interesting.

I think it's possible AHG has some type of ENA admixture, if I am understanding you correctly it's the same that ANE has right?

Something related to Villabruna almost certainly exists in AHG, or possibly something from the Balkans that had an important role in the formation of both Villabruna and AHG.

Uniparentally, there is not much of AHG in Anatolia_N but if the Lazaridis model is accurate, all these populations are close enough to each other (because of a common Dudzuana-like/common West Eurasian ancestry) that models showing continuity work, even though it's likely not the case, they are just genetically similar. So I think what you are saying makes sense.

Konstantin Demidov said...

A brief history of defining the border between Europe and Asia is as follows:
1. Initially, everything east of the Carpathian Mountains was considered Asia.
2. Then the Dnieper River was considered the border.
3. Then the Don and Volga were considered the border.
4. And it was only in the middle of the 18th century that the Ural Ridge began to be considered. From this point of view, Vorkuta is the easternmost.

vAsiSTha said...

"did anyone notice how they modeled Anatolia_N as a mix of Anatolia_HG and Iran_N?"

I remember posting here about one of my qpgraphs showing 10% additional IranN and 90% Pinarbasi for Barcin.

Davidski said...

@Konstantin Demidov

Scythia was considered a part of Europe since Herodotus's time.

https://commons.wikimedia.org/wiki/File:Herodotus_world_map-en.svg

Interestingly, his map shows a close correlation with my PCA.

https://drive.google.com/file/d/0B9o3EYTdM8lQelZhUGYtN1g1VnM/view?resourcekey=0-9FP1ZaFNjCJrxyoFmrOm7g

Silvia said...

What about the possibility that Yamnaya derives from Afanasievo, rather than vice versa? The dates attributed to the respective archaeological sites are really pretty close, or a few hundred years apart or even in overlapping range. In genomic terms while they are extremely similar or obviously belonging to the same clade, Afanasievo actually seem to have the more extreme values -- for example see PC15 on G25 -- indicating perhaps Afanasievo are more basal, and in process of migrating westward they could have experienced minor admixture leading to less extreme values at that component among Yamnaya?

Mark B. said...

Is it possible that the steppe-as-Asia thing is no more than an anti-nazi holdover. Even though the Aryan-from-Germany thing died a full lifetime ago, there could be virtue signalling going on - at least among themselves. If the proto-indo-europeans started out in asia, then take that, you naugty eurocentric nazis.