search this blog

Friday, April 3, 2015

The teal people: did they actually exist, and if so, who were they?


The ADMIXTURE analysis in Haak et al. 2015 includes a series of intriguing teal colored components from K=16 to K=20 (see image here). The main reason I'm so intrigued by these components is because they generally make up over 40% of the genetic structure of the potentially Proto-Indo-European-speaking Yamnaya people.

But there's only so much one can learn by starring at a bar graph, so I thought I'd have a go at isolating the same signal with ADMIXTURE to study it in more detail. You can view the results of my experiment in the spreadsheet here.

I wasn't able to completely nail any one of the teal components from Haak et al., because I don't have access to all of the samples used in the paper (I'd have to sign a waiver to get them). Nevertheless, the signal looks basically the same.

Below is a bar graph based on the output featuring selected populations and ancient genomes from Europe and Asia. The Fst genetic distances between the nine components are available here.

Note that the teal component peaks in the Caucasus and the Hindu Kush, and generally shows a strong correlation with regions of relatively high MA1-related or Ancient North Eurasian (ANE) admixture. On the other hand, the orange component peaks among Early European Farmers (EEF), who basically lack ANE.

To learn about the structure of the three main West Eurasian components - blue, orange and teal - I made synthetic individuals from the P output to represent each of the components, and tested them with my K8 model. As expected, the teal component harbors a high level of ANE, while the orange component lacks it altogether. Refer to the spreadsheet here.

It's very likely that the teal and orange components from Haak et al. share these traits. I think this is more than obvious by looking at their frequencies across space and time in Eurasia.

I also analyzed the synthetic individuals with PCA based on their K8 ancestry proportions. The samples representing the orange component fall just south of the Stuttgart genome from Neolithic Germany, and this is basically where I expect Neolithic genomes from the Near East to cluster when they become available.

Interestingly, the samples representing the blue component are dead ringers for Scandinavian hunter-gatherers (SHG). However, I suspect this is something of a coincidence caused by the small number of Western European hunter-gatherer (WHG) and Eastern hunter-gatherer (EHG) genomes in the dataset. The algorithm probably doesn't have enough variation to latch onto to create both WHG and EHG components, and in the end settles for something in between, which just happens to resemble SHG.


But the fact that the orange and blue samples more or less pass for ancient populations leaves open the possibility that the same might be said for the teal samples.

So did the teal people actually exist, and if so, who were they?

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.

Were they perhaps the Proto-Indo-Europeans? Probably not. I'd say they were Neolithic farmers who eventually played a role in the formation of the Proto-Indo-Europeans. In any case, someone had to bring the Caucasian or Central Asian admixture to the steppe, and I have it on good authority that it was already present among the Khvalynsk population of the Eneolithic, albeit at a lower level than among the Yamnaya of the early Bronze Age.

Citation...

Haak et al., Massive migration from the steppe was a source for Indo-European languages in Europe, Nature, Advance online publication, doi:10.1038/nature14317

Update 16/11/2015: 'Fourth strand' of European ancestry originated with (Caucasus) hunter-gatherers isolated by Ice Age

640 comments:

«Oldest   ‹Older   401 – 600 of 640   Newer›   Newest»
Davidski said...

Maju,

R1a-M459/PF6235 is a sister branch of R1a-M417, not ancestral to it per se, since M459/PF6235 don't lead to M417.

And obviously the Corded Ware R1a-M417* came from Eastern Europe, because this where the Corded Ware population originated, and was clearly alien to Neolithic Central Europe.

Nirjhar007 said...

But David M459 is R1a1 and M417 is R1a1a1
http://www.isogg.org/tree/ISOGG_HapgrpR.html

Davidski said...

Helgenes50,

I think teal first entered northern and central Europe with early IE groups, but not all of the teal in Europe is IE inspired.

Nirjhar007 said...

David, Will you please clarify this suggestion?
''R1a-M459/PF6235 is a sister branch of R1a-M417, not ancestral to it per se, since M459/PF6235 don't lead to M417.''
How come R1a1 becomes sister of R1a1a1 rather than Ancestral and why it ''don't lead to M417''??

Karl_K said...

@Nirjhar007

"How come R1a1 becomes sister of R1a1a1 rather than Ancestral and why it ''don't lead to M417''??"

Have you never driven down a dead end road? Often they do not bother to give them unique names, but instead just call them by the last major street leading to it.

Davidski said...

Nirjhar,

R1a-M459 isn't R1a1, it's R1a1*, so it's a split from the R1a tree just before R1a-M417 or R1a1a1 branches off. In other words, people who belong to R1a-M417 don't belong to R1a-M459.

We can call it the uncle or auntie of R1a-M417, instead of sister. Whatever.

Nirjhar007 said...

Please be straight this is confusing,
'' it can be designated as belonging to haplogroup R1a1*(xR1a1a) and **it occupied a basal position to the vast majority of modern Eurasian R1a-related Y-chromosomes4**, although more basal (R1a-M420*) Y-chromosomes have been detected in Iran and eastern Turkey4.
So R1a1 is a ''dead end mutation'' and R1a1a M198 arrived from what?

Nirjhar007 said...

Davo,
//R1a-M459 isn't R1a1, it's R1a1*, so it's a split from the R1a tree just before R1a-M417 or R1a1a1 branches off. In other words, people who belong to R1a-M417 don't belong to R1a-M459.//
But didn't M198 derive from R1a1*?

Nirjhar007 said...

OK got it-
http://en.wikipedia.org/wiki/Paragroup

Unknown said...

@ Davidski

" came from Eastern Europe, because this where the Corded Ware population originated, and was clearly alien to Neolithic Central Europe."


Well, actually, the CWC was also "alien" in Eastern Europe; because there was no obvious 'proto-Corded ware culture' . It was a novel phenomenon throughout it's entire territorial extent

Alberto said...

About the origin of R1a, a key is what Krefter said:

"It isn't random almost every ANE-rich ancient Y-DNA result has come out R".

And ANE is not European. It's Asian. Whether North Asian or Central Asian or S-C Asian or Iranian we still don't know it exactly.

R is not related to European ancestry (for what we know). If we find some pure ANE population from around the Mesolithic times, it will be somewhere in Asia. EHG had some 45% Asian ancestry, so it's not too surprising they belonged to R1.

Unknown said...

I think that is evident also, Alberto.

But, such labels might be unwarranted when we 're talking about 10, 20 or 30 kya.

Anyone care to hazard a guess as to when R1 'arrived' in the east European plain ?

Unknown said...

and lets not forget the Late Neolithic, non-ANE, R1 (x 269) from Iberia. That might not bee too much of a red-herring.

lets remember that, without a doubt, Franco-Cantabria, was the densest LGM refuge in Europe. yet, we have a mere (?) 1 Y DNA sample from pre-Neolithic Iberia.

Davidski said...

R1 is a Mammoth steppe lineage.

The concept of Europe came later.

Krefter said...

Alberto,

No one disputes R was probably brought to Russia from somewhere in Asia. What some of us are saying though is that most modern R1 recently came out of Eastern Europe.

It's almost like if we were having a debate about the origin of modern Y DNA J, and someone said "Well, its ancestors came from Africa". Its distant geographic origin doesn't matter very much.

Unknown said...

The mammoth steppe extended all the way to France. So we should expect that, at least intermittently, ANE, R1 -derived peoples existed throughout Europe, and did not simply wait on the Dnieper-Don interfluvial for 20 thousand years ?

Davidski said...

It's hard to guess right now what the genetic partitions were like on the Mammoth steppe across space and time. We'll need a lot of ancient genomes to work that out.

But obviously what happened at some point was that Western European, Eastern European and Siberian Mammoth steppe hunters became fairly discrete population units. That doesn't mean all contact between them stopped, but clearly there was no regular gene flow until well into the Neolithic, when all sorts of other mixing happened as well.

That's when EHG/ANE entered Central and Western Europe, possibly for the first time ever. The R1b in Neolithic Spain is either a signal of intermittent contacts across the Mammoth steppe, or sporadic migrations from the Mammoth steppe to the Near East.

Unknown said...

https://drive.google.com/file/d/0B1vtTHobiXwVOTJWbVBfcU42aUk/view?usp=sharing

Davidski said...

Mike,

Your image is not accessible. You need to edit the permissions.

But anyway, here are a couple of maps that sum up the really important stuff.

https://drive.google.com/file/d/0B9o3EYTdM8lQdWI2cTZKdzRpSjA/view?usp=sharing

https://drive.google.com/file/d/0B9o3EYTdM8lQeDhsU24tSnVxMjA/view?usp=sharing

Unknown said...

Here is a simple, crude map aligning with (what I believe) the current consensus is on this blog for R1a. As they say - a picture says a thousand words.

Dashed lines represent possibly extinct movements

The numbers (eg 4 , 6, 16, etc) are years kya

https://drive.google.com/open?id=0B1vtTHobiXwVOTJWbVBfcU42aUk&authuser=0

Davidski said...

Mike,

MA1 belonged to R* (maybe something like R3) not R1.

Unknown said...

Right, but R1 might have come from central Asia sometime after 20 kya.

Krefter said...

The last thing I'll say about this is we need to realize is the sample set and sampling biased of ancient Y DNA(Almost all is from Europe so far) and that we need to pay attention to evidence of Rs history in modern DNA.

Davidski said...

Mike,

Paleolithic hunter-gatherers usually moved back and forth across the steppe, not up and down Asia. There were some very good reasons for this.

So I doubt R1 could have come from Central Asia after 20 KYA, when R was already on the Mammoth steppe by 24 KYA.

Krefter said...

This must be the best commercial DNA testing company. Everything about them is better, and it's at the same prive as 23andme and FTDNA.

https://www.tribecode.com/how-it-works/

Alberto said...

Yes, I agree with all the remarks, but still:

David, do you really suggest that ANE arrived in Asia (except Siberia) in the early Bronze Age?

Unknown said...

@ Davidski

"Paleolithic hunter-gatherers usually moved back and forth across the steppe, not up and down Asia. There were some very good reasons for this.

So I doubt R1 could have come from Central Asia after 20 KYA, when R was already on the Mammoth steppe by 24 KYA.'

Maybe true.
But something big called the glacial peak occurred c. 20 kya.

Davidski said...

Alberto,

People were a lot like animals until the Neolithic transition, and even well into it, in that they kept firmly to their habitats and niches.

If there was no reason for EHG/ANE populations to move west deep into peninsular Europe, then why would they move to the Near East?

My bet is that the first Neolithic Near Eastern genome we see, no matter where it comes from, will be orange like Stuttgart except minus the blue, or at least with very little of it. And it'll belong to something like Y-HG H or T.

Alberto said...

I'm not talking about the Near East. I refer to places like Central Asia, Iran, Afghanistan, Pakistan, North India... Do you really think that ANE arrived there in the Bronze age?

Davidski said...

I have no idea, but I have a feeling that what happened in most of Europe after the Neolithic also happened in the Near East and South Asia. So I think there's a chance that the earliest Neolithic Indus Valley sites will show something totally unexpected like even a southeast Asian population with no West Eurasian admixture. I'm keeping an open mind.

Karl_K said...

@Krefter

"This must be the best commercial DNA testing company. Everything about them is better, and it's at the same pri[c]e as 23andme and FTDNA."

You sound like an infomercial.

Clearly not "everything" about them is better. They do not have an enormous database of other users, which is the main interest of most users these days.

Any idea what kind of raw data they provide? And do they avoid health related regions? It's all a bit unclear on their site.

I myself will have to just pray that 23andme survives to do whole genome sequencing, because half of the elderly family members I got to be tested in the last few years (and have their DNA sample stored) have since passed away. What a waste if all those samples wouldn't be re-analyzed in the near future.

Grey said...

(h/t maciamo's cool maps)

This may be nothing as the autosomal data seems to come from different sources but if you roughly add the percentage of Atlantic autosomal component to the percentage of northwestern autosomal component from these two maps

http://cdn.eupedia.com/images/content/Atlantic-admixture.png

http://www.eupedia.com/images/content/Northwest-European-admixture.gif

the result seems to pretty much correlate with the LP percentage peaking where the two components most overlap.

http://rstb.royalsocietypublishing.org/content/royptb/366/1566/863/F1.large.jpg

Simon_W said...

@ Maju

According to Underhill R1a and R1b diverged 25000 years BP. That's very old to me. The shallow divergence time you're refering to concerns primarily R1a1a1-M417. Although R1a1a and even R1a1 don't seem to be very old either (8000 y BP and 12000 y BP respectively, according to the R1a project on FTDNA).

The six R1a1-SRY10831.2*(xM417/Page7) found by Underhill in Iran and the Caucasus may be R1a1 or R1a1a, since apparently he didn't test for M17/M198/M512/M514. We know for a fact that R1a1 was present in Karelian EHG. And we know, thanks to Underhill, that R1a1* or R1a1a* is present in modern Iranians and Kabardins. So what does this tell us about the origin of R1a1? At the very least it tells that R1a1 may be from anywhere inbetween Karelia and Iran. But more realistically seen, the actual evidence from ancient DNA, evidence directly from the past, weighs more as far as the origin-question is concerned, than the evidence from present-day DNA, thousands of years later.

R1a1a1-M417 has been found in a Corded Ware individual from Germany. Germany is inbetween Norway and Turkey and the individual in question lived around 4400 years ago. Plenty of time to migrate to either place, or from anywhere in eastern Europe where the Corded Ware was also present.

R1a-Z93 doesn't have a West Asian centrality, see:

http://cdn.eupedia.com/images/content/Haplogroup-R1a-Z93-Eurasia.png

The 24 R1a-M420*(xSRY10831.2) chromosomes found by Underhill are really R1a*, a step older on the phylogenetic tree than the Karelian EHG. But, you see: present-day DNA suggested an Iranian or Caucasus origin for R1a1, whereas ancient DNA research found the same in Karelian EHG. Present-day DNA suggested an Anatolian or Norwegian origin of R1a1a1, whereas ancient DNA research found that haplogroup in the Corded Ware in Germany. There seems to be a pattern at work. Maybe present-day DNA isn't very reliable? There seem to be places where old paragroups are much better preserved than elsewhere.

So there's no need to assume that R1a1a1-M417 spread to Iran first and then to central Asia. Imho it was present at least in parts of the vast Yamnaya/Corded Ware area, and later in the Andronovo culture. The details have to be worked out by ancient DNA evidence.

Simon_W said...

@ Krefter

Are you sure there was I2a2 in the PWC? Afaik there's just Ajv58 who is positive for P37.2?

Also, to my knowledge the two Mesolithic Russians (one from Haak et al., the other from a Russian paper) were R1a1, but not anything further downstream.

Simon_W said...

@ Nirjhar007

Yes, presumably the earliest Indo-Iranians were not 100% R1a-Z93, they certainly also had other haplogroups.

Simon_W said...

@ David

That's a rather old tree you posted, the latest one I've seen looks thus:

http://jpst.it/y9xL

Gill said...

I think it's more likely that early neolithic indus valley will show a lot of ASE/ENF or perhaps ANE/ENF (the first Gedrosian?). Or maybe even all three (ANE/ASE/ENF). It should help us date when ANE came into the subcontinent and whether it came with ENF (as a part of this "Gedrosian") and/or whether ENF was already in South Asia before ANE.

I would expect a lot of J, L, H, G haplogroups. Mostly J and L.

Nirjhar007 said...

Just to be clear here there is absolutely NO way R originated in E Europe but Central Asia or SC Asia.
ANE is Native of Central Asia.
David,
// I think there's a chance that the earliest Neolithic Indus Valley sites will show something totally unexpected like even a southeast Asian population with no West Eurasian admixture. I'm keeping an open mind.//
Well 7000-6000 BC show ''SE Asian type'' Skeletons in Mehrgarh area but Interestingly in North India we have ''European'' Like Hunter gatherers as per Skeleton studies!! so its quite complex even in the Earliest neolithic Stages for the Subcontinent.

Nirjhar007 said...

Gill, I would certainly Expect R2a and R1a also!
I certainly Think ANE was there in the Subcontinent from Mesolithic period at least with ASE.
Then Came ENF.

Simon_W said...

Another thought regarding the Spain_EN-like admixture in Esperstedt_MN: Maybe it's from La Hoguette!

http://jpst.it/y9yn

Nirjhar007 said...

//My bet is that the first Neolithic Near Eastern genome we see, no matter where it comes from, will be orange like Stuttgart except minus the blue, or at least with very little of it. And it'll belong to something like Y-HG H or T.//
That may be applicable to Levant or Anatolia etc But certainly NOT with Iran-S Caspian areas.

Eireanach said...

My K15
Population  
North_Sea 34.43%
Atlantic 31.38%
Baltic 13.42%
Eastern_Euro 7.33%
West_Med 5.38%
West_Asian 5.47%
East_Med -    
Red_Sea -    
South_Asian 1.95%
Southeast_Asian -    
Siberian -    
Amerindian 0.44%
Oceanian -    
Northeast_African 0.20%
Sub-Saharan -    

1 Irish + Irish + Irish + North_German @ 4.162835

Hinxton 4
Population  
North_Sea 34.84%
Atlantic 31.86%
Baltic 13.89%
Eastern_Euro 5.72%
West_Med 6.34%
West_Asian 4.78%
East_Med -    
Red_Sea -    
South_Asian 1.93%
Southeast_Asian -    
Siberian 0.13%
Amerindian -    
Oceanian -    
Northeast_African -    
Sub-Saharan 0.51%

1 Irish + Irish + Irish + North_German @ 4.858814

Why do I have similar results to a man who died 2000 ybp? I am Irish

Maju said...

@Mike: "I clearly stated Karelia and Mesolithic. (Who would I make as ludicrous claim as arguing that Corded Ware is Epipalaeolithic ?"

In Europe Mesolithic (when used at all) = Epipaleolithic. Corded ware is well into Chalcolithic in fact, so...

Nirjhar007 said...

Maju, Europe had no Mesolithic development?

Unknown said...

Of course it did
Maju is confusing himself, and others

Nirjhar007 said...

Yeah it appears it did-
http://en.wikipedia.org/wiki/Mesolithic#Europe

Maju said...

"R1a-M459/PF6235 is a sister branch of R1a-M417, not ancestral to it per se, since M459/PF6235 don't lead to M417."

Per ISOGG:
→ R1a1-M459
→→ R1a1a-M512 (old good M17)
→→→ R1a1a1-M417

I agree that it's necessarily ancestral but it's clear that the Karelian HG belonged to a two-tier higher phylogenetic level that is now only represented by Iranians and some Kabardins.

... "obviously the Corded Ware R1a-M417* came from Eastern Europe"...

From Karelia? I mean because how much R1a1a1* exits today in Eastern Europe supporting that idea? We know only by archaeology that Corded Ware is related to Eastern European cultures such as Yamna but actually most of that relation is via Globular Amphorae to Baalberge, who are now claimed as genetically non-Kurgan (based on Baalberge aDNA), and the only other complementary source would be a very localized impact of Catacombs culture in Cuyavia (but these people are not from Cuyavia, nor carry around the cultural signatures of Catacombs, but of a modified Globular Amphorae c.)

You may think I'm dilly-dalling here but something we know about present day M417 is that in Europe it has two branches: a North German one (M417) and the mainline "Corded Ware" one (Z282). This suggests a more westerly origin of European R1a-417 than Samara-Khvalynsk (or Yamna/Maykop) somewhere not far from Poland in fact.

In addition to that, it must be reminded, there are two other branches that are Asian (one in Turkey and the other spawning from West Asia - read:
"surely Iran"), so I'm only talking above of the two European sub-lineages, i.e. assuming there was a short-lived (and so far undetected in terms genetic) subdivision of M417 between a European and West Asian branch of M417.

One possibility is that you're partly right and that European M417 comes from the Caucasus via Catacombs (Catacombs' origin is debated but some think it spawned from Maykop, while others instead think it's rather Yamna-like), acquiring its current characteristics with Corded Ware.

Another possibility is that you are 100% right and it's an offshoot of the previous expansion that affected Karelia (but the intermediate stage M17 is undocumented in Europe (barring the Kabardin singleton), so it's hard to make the case for it unless you simply deny all the evidence pointing to the Iranian homeland until the M17 stage, what I'm not willing to accept. For me therefore the Karelian occurrence of M459* and the CW one of M417* are not directly related. In fact M459 and M17 could have been around for some time (Underhill gives no chronologies for them, only considering M417 and below).

Finally it's even possible that R1a1a1-M417 arrived somehow via the Balcans (the Hamburg - Kiev - Ankara partial triangle can suggest that) and therefore is not related to Kurgan expansion until picked up by Corded Ware.

I'm remaining open to all three possibilities but, if I have to discard one, that would be the second one: that Corded M417 is somehow directly related to Karelian HG M459. Because it is the less parsimonious one of the three scenarios and seems mostly driven by the blinders of current aDNA knowledge, way too limited to allow us to draw conclusions when most of the relevant regions still remain to be sampled and the relevant ancient sequences are two singletons.

In any case the case for the European half of M417 being related to Corded Ware is strong. But before that stage the matter is very open.

truth said...

I don't like much the Orange (EEF) component because in reality it's a mixture of ancient near-east and WHG. That's why I prefer the admixture in the study of Kostenski, where there is a "brown" near-east component and EEF's show about half of it, and I think original farmers from near-east will mostly have :

http://anthropogenesis.kinshipstudies.org/wp-content/uploads/2014/11/Kostenki-ADMIXTURE.jpg

Karl_K said...

@Maju

"it is the less parsimonious one of the three scenarios and seems mostly driven by the blinders of current aDNA knowledge"

I love it! That darn aDNA, blinding everyone from the truth that is obvious from modern DNA studies.

So... what do people normally say here... I think R1a and R1b originally came from the Americas. That is probably where the highest diversity is seen today.

Nirjhar007 said...

Karl What Maju says is for the sake of possibilities and you very well know that Forget Asia Even within Europe there is a large area of aDNA which is yet to be explored.

Maju said...

@David: ISOGG (which I have to consult once and again in this debate) lists an intermediate phase between M459 and M417, which is M17. However it is correct that Underhill did not detect any M17(xM417) individual so I guess that you're somehow right in your "sister clade" claim. I'm conceding here because ISOGG structure may well be overpopulated with evolutionary stages that may not actually exist in terms of actual carriers.

On the other hand we have no clear references on how old may be M459, while M417 should belong to a well-defined bracket around the early Holocene.

Maju said...

Alberto is right when he emphasizes that ANE is Asian, in fact he should say Siberian. Instead Y-DNA R1 is unmistakably West Asian. So I personally think that they are unrelated, however some R1 may be related to the "teal" component, which is indeed West Asian. Here I mean R1a1a1 particularly, and maybe also the Yamna branch of R1b (unrelated to Western European R1b). Thankfully we have at least one Early Neolithic Iberian R1b that is not carrier of ANE not "teal", and that should be enough evidence to discard the generic and whishy-washy alleged association of all R1 or even R with ANE (or also with "teal").

"Anyone care to hazard a guess as to when R1 'arrived' in the east European plain ?"

Well, Karelia HG gives the latest possible date for arrival R1 in Eastern Europe but we don't know if it arrived by the steppe or the taiga, i.e. via Caucasus or via Central Asia. Or even it might have arrived via Balcans/Central Europe, although I do not particularly favor this scenario.

But the question is not really that one but rather: when did R1 that is actually ancestral to modern European R1, and not an erratic without legacy, arrived to anywhere in Europe? So far this question has just two very partial answers:

1. R1a-M417 (Euro subgroup) seems to have been in Central Europe since Corded Ware times.
2. Some very specific R1b (Yamna subclade) seems to have been in the Volga area at least since Yamna times.

Both seem to have arrived via the Caucasus and be associated to Highlander West Asian genetic affinities ("teal") of these two populations (rather than to ANE or EHG).

We still don't have any direct reference on how the major R1b (Western European subclade) arrived to the region where it is now dominant. But plausibly it must have been with Neolithic at the latest (because otherwise Basques would not have it in such over-dominating role).

Maju said...

@Davidski:

"R1 is a Mammoth steppe lineage.

The concept of Europe came later".

That's not correct at all and we see it very clearly in ancient DNA: all European HGs form a clade vs ANE and vs West Asia, so there was a de facto concept of Europe in terms ethnic and genetic in the Paleolithic. And this is in spite of Gravettian being associated to Mal'ta and very possibly of West Asian origin: the European exclusive identity of WHG, UHG, SHG and EHG cannot be denied.

The physical borders of Europe were only slightly different than today:

(1) the Caucasus was about the same

(2) the Turkish straits were not yet a strait but rather an isthmus but still narrow enough to define two distinct regions (plus nearby areas were surely not much populated at either side).
(3) the main difference may have been in Western Siberia, which would have been originally more European than it is today, as the natural border was a huge glacial late at the Obi. But still West Siberia is not Central Siberia to where the original ANE component belongs.

There was no such thing as "the mammoth steppe", except maybe for the mammoths, not for humans.

Maju said...

@Mike: "The mammoth steppe extended all the way to France. So we should expect that, at least intermittently, ANE, R1 -derived peoples existed throughout Europe, and did not simply wait on the Dnieper-Don interfluvial for 20 thousand years ?"

That's another good counter-argument indeed. I'm not favorable to it because the patterns of spread of all relevant R1-derived sublineages clearly spawn from West Asia in radial style and not from Siberia or Eastern Europe but, within the Davidski hypothesis, it is a very good counter-argument no doubt.

Nirjhar007 said...

Maju, where R originated?

Maju said...

Mal'ta boy was pre-R1, not quite "R3" but still the diverging node upstream enough not to be yet R1. Then he's quite drifted from that pre-R1 node he hangs from, so almost certainly R1 had already coalesced millennia before Mal'ta boy lived but not in Central Siberia but in West Asia or rather Pakistan/NW India (or Uzbekistan maybe).

The phylogeny of Mal'ta boy's lineage allows for R and R1 coalescing in the earliest Upper Paleolithic, roughly simultaneously to Q and Q1 and around the same Iran-Pakistan-Uzbekistan core area. In fact I'd dare say that the divergence between the R1a and R1b branches is from well before the LGM (my estimate for R1b-root is c. 48 Ka BP but, even if I'm going too radical, it should not be more recent than Gravettian dates, c. 32 Ka BP).

So we should stop thinking in catch-all terms of R1 and R1b when discussing the Holocene: we must talk of the specific sublineages.

Nirjhar007 said...

Maju, How R1b can be 48ooo YBP when R is said to have originated around ~27000YBP?!!o_O

Maju said...

@Simon: "According to Underhill R1a and R1b diverged 25000 years BP".

But Underhill's choice of mutation rates is very much criticable, he should have picked the Méndez one, which is average among the modern studies he referenced to and is also the only one calibrated to an ancient specimen (Anzick). So it's necessarily older.

"That's very old to me".

Not old enough for me. I mean: Mal'ta boy is actually from that very same date (24 Ka BP) and his lineage was clearly drifted (in parallel) well under the R1 node, so R1 must be quite older than Ma1. So add at least 8-10 Ka more to Underhills mega-hyper-ultra-conservative estimate, possibly even more.

"The shallow divergence time you're refering to concerns primarily R1a1a1-M417".

Conceded and surely my bad (was a bit confused by the nomenclature). Still R1a1a1-M417 is like 99% of all modern R1a.

"But more realistically seen, the actual evidence from ancient DNA, evidence directly from the past, weighs more as far as the origin-question is concerned, than the evidence from present-day DNA, thousands of years later".

I cannot agree? Ancient DNA is just another data point, more so in absence of any aDNA from West Asia that could serve as counter-reference. Of course detailed analysis of the haplotypes could maybe lean the scales in either direction but we don't have that (at least not that I know). For all I know the Karelian data point could even be a random erratic or could be a once more common but still peculiar branch, just as happens with the Yamna R1b.

"Germany is inbetween Norway and Turkey"...

Again my bad if I initially pointed to Norway. The actual M417* European individual is from North Germany, the Norwegian group is downstream (within Z282). Again haplotype analysis is the only thing that could help further discernment but it's possible that the CW individual was (roughly) ancestral to the M417* of North Germany or to Z282 or to both or (less likely) to neither. If you're looking for a "Gengis Khan effect", there you have a potential candidate but it needs to be more deeply analyzed in relation to its potential descendants.

Personally I do not expect him to be ancestral neither to the Anatolian M417* nor much less to the West-South-Central Asian Z93 branch. The only way this could be true would be if the actual origin of Indo-Iranian is not in the steppe Andronovo culture but in some West-South Asian group that would have to be derived directly from Corded Ware. I don't see that plausible, sorry.

"R1a-Z93 doesn't have a West Asian centrality"...

That map only shows frequency. You have to look at the haplotype structure in Underhill's supp. fig. 2 or at least read carefully my synthesizing analysis (too long to quote here): http://forwhattheywereweare.blogspot.com/2014/03/y-dna-r1a-spread-from-iran.html

Z93 is unmistakably West Asian by origin, Iranian again I would say, although maybe more easterner within this country than previous stages. IDK, maybe you could imagine something like BMAC but BMAC does not have the kind of expansion into Central Asia that would be needed.

"So there's no need to assume that R1a1a1-M417 spread to Iran first and then to central Asia".

There is if you consider carefully the structure haplogroup Z93, as I did. There's no reasonable doubt that Z93 overall spread from SW to North. So that "fourth branch" of the expansion of M417 has to go via Iran, regardless of what you think about the overall origin of M417 (which is indeed a bit blurry).

Maju said...

@Nirjhar: "Maju, Europe had no Mesolithic development?"

No. India neither AFAIK.The term in its strict sense is mostly limited to the Fertile Crescent and the Nile (there seems to have been a Mesolithic of Sudan for example, with cereal gathering and related sedentarization). Not sure how it applies to East Asia or America (there must have been some Mesolithic in some regions because their Neolithic is independent from ours).

However, if you use it in the loose, sloppy sense of "anything between the end of the Ice Age and the arrival of agriculture", then it's different. But I'm talking senso stricto.

Maju said...

@Truth: that's a very interesting analysis, yes. Much more productive than the "orange" component indeed. It suggests that Stuttgart and Ötzi were like 50% aboriginal European and that Gökhem was around 60% (a bit more if the "African" thing was aboriginal Iberian).

Maju said...

@Karl: No. It's not about raw diversity but about hierarchically organized structure, what I call the "geostructure". Modern America simply does not make it because it can be demonstrated, as happens with Armenian R1b, that their branches are terminal and include a number of founder effects from the Old World.

So don't attempt to ridicule my argument that way because it's not what you say.

Maju said...

@Nirjhar: "is said"? Someone says. I say otherwise because "is said" is not evidence per se, one must look at the reasons behind the claim. People say many things that are only in their imagination, even atheists use religious expressions just out of cultural inertia, so I guess it's just natural that academics also fall in the trap of scholasticism.

Gaspar said...

@nirjhar

The H and T connection in India ( south asia )do not match in paths to Europe. while H was based in west India, it travelled to the Levant and Europe via the south of the zargos mountains, while L and T was in union ( as per mendez paper )with SNP P326, this was 49000yo and originated in the sind valley of north India. The logic is L and T settled first in eastern India, while another branch travelled via Afghanistan to the Caucasus, Levant and Europe via a path north of the zargos mountains .........age of these markers via studies confirm this route.

Krefter said...

@Eir,
"Why do I have similar results to a man who died 2000 ybp? I am Irish"

Because Britons are closely related to Irish.

In my opinion, in the last 2,000YBP not much has changed, and Ireland has been more isolated from foreign admixture than Britain. The Roman empire was mostly an "elitist" thing anyways, so the 2,000YBP Briton is very closely related to former Roman regions to.

Unknown said...

was Franco-Cantabria really a Refuge and the densest populated at LGM in Europe ? It is the most explored though.
how were H, H1 and H3 Iberian Franco-Cantabrian european Aboriginals 50% in Stuttgart and 60% in Gökhem?
If Basal Eurasian were rather N1a and N2'W mtDNA non 'european' Neolithic pioneers.
H like R0, R2'JT and U8b'K just were all later incorporated into Neolithic across the Aegean.

Alberto said...

Maju, when I said that ANE is Asian I was not just meaning Siberia. Siberia does have a lot of ANE, but in Eurasia it peaks in Pakistan, with a secondary peak in the Caucasus. It's also pretty high everywhere in between (Central Asia, Afghanistan, Iran...). Even in Anatolia it is higher than in any European country.

And it does have a strong affinity with Y-DNA R and Q (which doesn't mean that every single group in the world that has R or Q is high in ANE ancestry).

And in agreement with you, I fail to see why would Eastern Europe be the origin of modern R1 or ANE.

Davidski said...

Maju & Truth,

Neolithic farmers from the northern Near East will show mostly the orange component, maybe with a little of the blue. You'll see.

Chad said...

The orange component is not Balkan. It is that same component spread with farmers into Africa, and Europe. Take the teal and Sub-Saharan out of Levantines, and they're right under EEF. They're the same damn people!!! These guys all have the orange component. Once again, Maju, you will be dead wrong when early farmer genomes from West Asia come out.

result: Loschbour Stuttgart Somali Ju_hoan_North -0.0178 -5.481 14961 15504 345573
result: Loschbour Stuttgart Kikuyu Ju_hoan_North -0.0108 -3.222 14444 14761 345573
result: Loschbour Stuttgart Masai Ju_hoan_North -0.0106 -3.424 14654 14969 345573
result: Loschbour Stuttgart Datog Ju_hoan_North -0.0098 -2.682 14820 15114 345562
result: Loschbour Stuttgart Ethiopian_Jew Ju_hoan_North -0.0193 -5.417 15206 15805 345573
result: Loschbour Stuttgart Tunisian Ju_hoan_North -0.0218 -5.778 15833 16538 345573
result: Loschbour Stuttgart Algerian Ju_hoan_North -0.0181 -4.690 15869 16454 345573
result: Loschbour Stuttgart Saharawi Ju_hoan_North -0.0184 -4.702 15793 16384 345573
result: Loschbour Stuttgart Egyptian Ju_hoan_North -0.0283 -7.658 15814 16735 345573
result: Loschbour Stuttgart Palestinian Ju_hoan_North -0.0277 -7.632 15994 16906 345573
result: Loschbour Stuttgart BedouinB Ju_hoan_North -0.0275 -7.121 15975 16878 345573
result: Loschbour Stuttgart Lebanese Ju_hoan_North -0.0253 -6.462 16085 16921 345573
result: Loschbour Stuttgart Iraqi_Jew Ju_hoan_North -0.0318 -7.633 16063 17118 345573
result: Loschbour Stuttgart Armenian Ju_hoan_North -0.0275 -7.081 16157 17072 345573
result: Loschbour Stuttgart Georgian Ju_hoan_North -0.0251 -6.288 16221 17055 345573
result: Loschbour Stuttgart Turkish Ju_hoan_North -0.0173 -4.600 16329 16904 345573

Unknown said...

@ Alberto

"And in agreement with you, I fail to see why would Eastern Europe be the origin of modern R1 or ANE."

But where R* arose; and that Central Asia was a repository for ANE might have little to do with the expansion of R1a M417

Unknown said...

Chad

"
The orange component is not Balkan. It is that same component spread with farmers into Africa, and Europe. Take the teal and Sub-Saharan out of Levantines, and they're right under EEF. They're the same damn people!!! These guys all have the orange componen"

Do you think you're possibly looking at Orange , a low resolution component, is obscuring the details.
Ie within this "orange" are west med, east med, Atlantic, and Red Sea elements. I don't think they're all straight Levantines.

Davidski said...

The orange is the ancient Near East. The modern Near Eastern K15 components are what came after the orange, after lots of population expansions, mixing and drift.

Matt said...

I think the orange component in this K9 may prove to be as close to the populations of the ancient Near East as "European" is to actual WHG and EHG on the genotype PCA. Near in world terms, still perhaps a little distant in West Eurasian terms? Or it could be the real deal.

Re: the ADMIXTURE analysis in the Kostenki 14 paper, I would note that at the K9 where there is a brown Middle Eastern component at around 50% in EEF (i.e. Stuttgart, as Iceman and Gok2 are later in time), it is around 60% in Druze, who are also shown as 30% Central Asian "teal" (equivalent) and 10% WHG.

There are knock on changes on the distributions of components in other populations as well, as you go from David's ADMIXTURE of Stuttgart as 20% "European", 80% EEF to the K14's papers Stuttgart as 50% Middle Eastern, 45% European, 5% Central Asian. As the different components are essentially trying to approximate exactly the same population relationships (the two ADMIXTURE are not going to be giving different overall relatedness by very much).

Either way, we won't actually know until there is a sample.

Davidski said...

Eireanach,

Are you really asking me why you, as an Irishman, show such a strong genetic similarity to an Iron Age Celtic Briton?

Is this really what you're asking?

Maju said...

@Krefter: "Like I said before Underhill 2014 is just one study".

Do you have any other interesting studies to share with us? Otherwise it is "the study" and not "just one".

Maju said...

@Sik Raci

"was Franco-Cantabria really a Refuge and the densest populated at LGM in Europe ?"

At least between the LGM and the end of the Ice Age yes: http://forwhattheywereweare.blogspot.com/2010/10/revisiting-bocquet-appel-2005.html

Afterwards the Rhine and North France may have been similarly or more populated because of improved climate but I don't have clear data for the Epipaleolithic densities.

"How were (...) Iberian Franco-Cantabrian (sic) european Aboriginals 50% in Stuttgart and 60% in Gökhem?

Surely the 50% HG that seems to exist in Stuttgart/Ötzi is not Franco-Cantabrian nor Iberian nor in general Western HG but most probably from "cousins" from the much less populated Balcans, which were incorporated upon the first European Neolithic in Thessaly and nearby areas. The extra 10-15% HG in Gökhem may well be Westerner.

Maju said...

@Alberto: when I said Siberian I mean "Paleo-Siberian", i.e. Ma1, which is the reference genome for "ANE". Additionally, the Siberian nature of this component is beyond doubt particularly because the most affected by it are Native Americans and their close relatives of NE Siberia, then Kets and only then other populations. It's definitely not something from Europe, West Asia or South Asia but from LGM Siberia. More complex is to understand how and when exactly the various populations in these regions got it but the origin is very clear.

Nirjhar007 said...

Maju,
// But I'm talking senso stricto.//
Ok then i ask as i often do to give at least 3 basal characteristics for a culture to be called''Mesolithic''.
// I guess it's just natural that academics also fall in the trap of scholasticism.//
Then for example if R1b is indeed 30000 YBP why we have Malta Boy as R* 24kybp? An example of Isolation or Drift? or what?.
Gaspar,
Give a reference for your suggestion.

Nirjhar007 said...

Maju,
//BMAC does not have the kind of expansion into Central Asia that would be needed. //
BMAC IMO effected Sinastha,Andronovo,Arkaim etc and Brought The IE Influence towards Uralic folks anyway certainly We can't Put Z-93 to be spreading with A Single culture as its history can be more complex though i also think BMAC spread Some Z-93 but it was already there in Central-asia prior to that with ''Iranian Nomads''...

Unknown said...

Nirj

The terminology can be quite confusing, and depends on what shool of thought one comes form.

Generally European, esp central -northern European scholars distinguishes "Mesolithic" from Epipalaeolithic or Final Upper Palaeolithic on the basis of period. Any HG in the Holocene (ie after 11 kya) is termed Mesolithic, whilst before that they are called Palaeolithic, Pleistocene , etc huntergatherers.

In parts of southern Europe, the Neasr East , etc, "Mesolithic" is reserved for groups of foragers which show steps toward becomeing producers (ie farmers) - either some aspects of cultivation, or sedentism in settlement.

Now, how does this affect the fact that R1a (XM417) was found in Karelia 6 kya ? It doesn't.

Nirjhar007 said...

Thanks Mike, So in North we have the Term on the basis of time and in South on the Basis of Transition towards Farming.

Unknown said...

Yes, I think.

To keep it simple, it might be just best to use economy "forager' vs 'farmer' and absolute period (eg 8 kya) rather than clunky and inflexible terms like "Tardiepigravettian"

Nirjhar007 said...

Uh Mike Since such Transition came late in North Compared to South why can't we just put that Mesolithic is also late in the North?.

Unknown said...

The Mesolithic transition was not later in the north of Europe - it can't have been because it's defined chronologically (the start of the Holocene).

In fact, the Holocene is very well represented in central and northern europe. By contrast, there is a paucity of Mesolithic sites in SEE, and when seen they show more of a graudal transition from the Late Palaeolithic lithic industries and faunal assemblages.

The opening up of post-glacial northern Europe, with abudnant rivers and lakes must have provided a plentiful economy for Mesolithic hunters. The Balkans has fewer rivers, and might have been heavily forested - not condusive for hunting. What sites were existent seem clustered in specific regions.

Im not sure about Spain, etc - Ask maju.

Nirjhar007 said...

// it can't have been because it's defined chronologically (the start of the Holocene).//
My Bad I mean by transition the advent of food production in North of course so basically its just the time frame which dictates the identity not the way of life?.

Eireanach said...

@Davidski thanks for the use of your you fine calculators at gedmatch.

"Are you really asking me why you, as an Irishman, show such a strong genetic similarity to an Iron Age Celtic Briton?"

The dna has changed little then in 2000 years.

Is it possible to create a calculator with only Atlantic and North Sea markers?

Unknown said...

Nirj

There were also social changes. Gravettian hunter -gahteres were diferrent to Mesolithic ones

Eg: at 23 kya people in Alpine Europe hunter mammoth

By 17kya they had switched to horse and reindeer.

By 9kya (the Mesolithic)- the exploited a wide variety of fauna and flora, adapted to their specific ecological niche

Attendant with these changes in economy were changes in tools (lithic assemblages) and habitat locations, etc.

There was also massive flux in populations, regional settlement. It might be one region was unsettled for 3 000 years, then re-settled again By whom ? Their descendent who had shifted elsewhere for a while ? Or a wholly new population ? We need aDNA for this.

Alberto said...

@Maju

I think the confusion comes precisely because of having only one genome found in Siberia that was defined as an example of ANE.

Actually, MA-1 looks more like a proto-EHG (or as you said, paleo-Siberian).

But MA-1 in Siberia doesn't mean that all ANE originated in Siberia and spread from there, though it doesn't really matter much for our purposes. We just need to wait for a genome from Mesolithic Central Asia to see how it looks like, because that would be relevant to understand gene flow and migrations from the Neolithic onwards.

I think that one interesting experiment that David made some time ago was a K9 admixture test where ANE is somehow split in 2 groups: EHG and Central Asian. The results are quite interesting:

https://docs.google.com/spreadsheets/d/19dLSYTTibNTOllihAaYaI0bBpGr2K0FMKQ4iSPci-Jc/edit#gid=1108642209

Alberto said...

@Mike

"But where R* arose; and that Central Asia was a repository for ANE might have little to do with the expansion of R1a M417."

Yes. But we don't know from where did M417 expanded, so why choose Eastern Europe? Is there any strong reason for doing so?

Unknown said...

Yes, Alberto

Who knows if Ma-1 was a permanent colonization, or a failed one, which then required repeated colonization of Siberia from central Asia.

Unknown said...

ALberto
"But we don't know from where did M417 expanded"

Sorry I don;t understand (?)

Davidski said...

Two Siberian genomes, MA1 and AG2, separated by almost 10,000 years belong to the same clade, and are closely related to Mesolithic Eastern European Hunter-Gatherers (EHG). Three out of the four of these samples belong to Y-haplogroup R.

What's Central Asia got to do with it?

Unknown said...

Im not arguing pro or con either locality, but I think David you perhaps shouldnt artificially separate central Asia from the mammoth steppe. Populations were fluctuant & waxing /waning. Furthermore, I wouldn't discount the importance of north -south movements in addition to east-west. In fact, they might have been even more important

Alberto said...

Yes, whether MA-1 succeeded in populating Siberia is not too important. In any case, EHG, Siberian and Native Americans descend from a MA-1 like population.

But the important point is that ANE cannot be restricted to Siberia from 24 Kya till 5-6 Kya. It had to be in Central Asia long before, since the Paleolithic (maybe it originated in Central Asia, maybe in Siberia, we don't know, but it's not important).

Once we find out if it indeed was in Central Asia (or Iran, or C-S Asia) during the Mesolithic we will have a better understanding of population movements.

Unknown said...

I agree Alberto

But what the present evidence suggests at least is that "ANE", R, R1, e ven R1a not as important as to the fact that the major subclades of M417 appear stacked in EE. Finding some ancient R1 group in Palaeolithic central Asia wont change it.

Krefter said...

In the K9 spreadsheet African American's West Eurasian scores(WHG, EHG, Near eastern, Central Asian) aren't consistent with British, or any modern pop. Oracles doesn't always work when someone has complex ancestry. This is also true for me, and I'm sure with most people with multi-continental ancestry.

Central Asians have trans-continental ancestry and I don't think we can get the exact ANE K8(or whatever) score of their West Eurasian ancestors by subtracting their Siberian side. If it comes out a little differnt from Samara, it's no big deal.

Alberto said...

Sorry, Mike, I'm mixing two different conversations.

One is about the origin and expansion of ANE. I myself very much doubt that it was a Siberian (or better to say modern day Russia, to include EHG) thing until the late Neolithic. So finding ANE in Iran or Central Asia from the Paleolithic would be relevant to know about migrations.

The other one is about all modern R1a originating in Eastern Europe. For this I don't know, we would indeed need more modern (Neolithic) samples from Asia. But I just don't see any strong reason for placing the origin of it in Eastern Europe. Modern distribution of clades ans subclades don't really tell us where is the origin. It can be anywhere.

Grey said...

I think particular HG populations will likely conform to the flora and fauna of particular bio-regions so my guess is the mammoth steppe would have been a distinct forming ground for a particular north Eurasian sub-population with other distinct sub-populations in north Eurasia forming around other distinct habitats e.g. around the lakes or along the edges of the mountain ranges etc.

Given the connection between ANE and the Americas I'd suggest the mammoth hunter population were originally quite mobile along the whole range of the mammoth steppe from west to east.

What happened to that population after the mammoth were all gone is another question.

Grey said...

@Alberto

"The other one is about all modern R1a originating in Eastern Europe ... But I just don't see any strong reason for placing the origin of it in Eastern Europe."

I don't have s strong view either way but all it would take for a HG clade to expand dramatically to the point where it outnumbered all the other surviving clades put together is for a particular group carrying that clade to develop a critical competitive advantage for a brief moment in time.

There could have been a long chain of R1 HG clades from wherever but one who hit the jackpot by luck.

The critical point imo is these events happened when population density was very low so it was easier for dramatic changes to occur.

In most cases these dramatic changes stemmed from the default farmer vs HG transition due to the farmer's numbers advantage but sometimes the HGs got lucky and had a bigger advantage.

So if the primary R1a clade did stem form Eastern Europe the strong reason was a fluke - horses.

(And if the primary clade stemmed from somewhere else it was probably for the same reason.)

Davidski said...

Alberto,

The main subclade of R1a in Asia is R-Z93. Europe, on the other hand, is dominated by R-Z283.

So it would seem it's all even, and, as per Maju, we need those mysterious R-M420* lineages from Iran, that probably split from the R-M417 branch over 15,000 years ago, to work out where R-M417 came from.

No, not really.

Apart from R-Z283, Europe is also, and exclusively, home to R-CTS4385, which is the sister clade of the lineage ancestral to both R-Z283 and R-Z93.

So, Europe is where the main splits right under R-M417 happened. In other words, this is where the first major expansions of R-M417 derived lineages took place.

This leaves some room to wonder if R-M417 migrated to Europe from the Near East or Central Asia before expanding back into Asia again. But that's just convoluted and silly, especially when we consider that Karelia HG - a Mesolithic hunter-gatherer far up north with no Near Eastern admixture - belonged to an R1a branch parallel to that of R-M417.

To cut a long story short, the phylogeography of modern and ancient R1a (which Maju hasn't bothered to learn properly yet) argue for a European or paleo-Siberian origin of R1 and/or R1a, and a major expansion of R1a, mostly but not exclusively R-M417, from Europe to Asia during the Eneolithic and Bronze Age.

Alberto said...

@Grey

I have to admit that I wouldn't have thought that Western European R1b could have arrived only in the Bronze Age. So if it gets confirmed (and now it looks like a solid bet), I think we'll need some time to understand how it happened.

So yes, anything is possible. Any strange combination of circumstances can make such things happen. Even at a point (time and place) where population densities were not tiny (Bronze Age Italy, France, Iberia).

But I still think it should be more an exception than a rule, so these kind of strange scenarios need to be proved before jumping onto them by a few hints.

In any case, it's also important to remember that paternal marker does not correlate to genome wide DNA, so a population might change completely their Y-DNA while having a minimum genetic impact overall.

Alberto said...

@David

Thanks for the clarification. I will look into it more closely when I have time try to understand the deep logic in it. I've heard to much nonsense about haplogroups many times, so I stopped caring about it. But what you said might be right.

One question, though: How accurate are current estimates for each clade's age? This has been a controversial subject and every new paper was revealing new estimates for mutation rates, etc... I think there is a basic flaw in thinking that mutation rates happen at a specific pace all the time, in every circumstance. So I guess that we go for averages. But is this reliable enough? Has Ancient DNA proved that we have accurate estimates for the age of each clade?

Unknown said...

@ Alberto , David,

I think there is something to the fact that Europe is dominated by R1a L664, Z 283 and R1b L51, whilst south of the Caspian is R1a -Z93 and R1b- Z2103.

R1b and R1a split 20 - 17 kya.

Looks like potentially more complex patterns.

Davidski said...

I don't know if it can be proven with modern data, even using full Y-chromosome sequences, when certain mutations arose and lineages expanded. This is not really my area of interest.

But R-M417 is missing from all ancient samples older than the Eneolithic, and then suddenly shows up in scores of Eneolithic, Bronze and Iron Age remains linked with the Kurgan expansions. So either it didn't exist before the Eneolithic, or it was sitting quietly somewhere until then.

Unknown said...

Dave

"But R-M417 is missing from all ancient samples older than the Eneolithic, and then suddenly shows up in scores of Eneolithic ...."

I'm not doubting the recentness of what happened, rather the complexity of it.

Grey said...

@Alberto

"But I still think it should be more an exception than a rule"

I agree with that - the default ought to be farmers walk over the local HGs simply due to their advantage in numbers.

If you take all the potential farmer land outside Africa and East Asia i.e. India, Iran, Middle and Near East, Central Asia, Europe etc the farmers did simply walk over the local HGs - unusual stuff seems to have mostly only happened right on the edges of the viable farmer range.

(which when you think about it is where you might expect it to happen i.e. where the farmers were at their weakest and so the balance of power between them and the local HGs was at its most equal)

Grey said...

" the farmers did simply walk over the local HGs"

should have been "in almost all of it"

Nirjhar007 said...

Mike,
I understand the Situation of Mesolithic now thank you.
//I think there is something to the fact that Europe is dominated by R1a L664, Z 283 and R1b L51, whilst south of the Caspian is R1a -Z93 and R1b- Z2103.//
Ok i have to ask this where does R2a fit in this relation?
And yes certainly ''i think'' R1a-M417 Is there in Central Asian SC Asian Iranian Copper Age Sites.

Nirjhar007 said...

We also see R1a Z-93 In Eastern Europe with patchy seldom distribution and i think its at least in some part a result of Invasions of Scythians,Sarmatians etc Iranian Nomads coming from Central Asia.
It is notable also that Slavic mythology is indeed effected by Iranian for example in -*bog denoting divinity etc which comes from Aryan -*bhag.
A general description is here-
http://en.wikipedia.org/wiki/Da%C5%BEbog

Davidski said...

There are R-Z93* lineages in Poland and Russia that don't exist anywhere else.

On the other hand, typically South Central Asian R-Z93 (L657+) has never been recorded in Eastern Europe.

Nirjhar007 said...

Because that mutation happened after such migrations? and i'm talking Central Asia Iran bro not India directly here...
About the rare Z93* it can be from peripheral effect anyway Ancient dna from Central Asia should suffice.

Maju said...

@Nirjhar: The central characteristic of Mesolithic senso stricto is that there is a proto-domestication tendency, typically gathering of cereals, which is more easily gauged than other practices. Another trait is a tendency to sedentarization. I can't think of a third trait right now unless it is sickles but, like sedentarism, these are related to cereal gathering: it's just one trait: being proto-agricultural, proto-Neolithic. A typical Mesolithic culture is Natufian.

By contrast Epipaleolithic is defined by the lack of this proto-Neolithic tendency or economy, a mere continuity of Paleolithic after the end of the Ice Age, which is the most widespread case.

Of course there is room for debate and if you want to use "Mesolithic" for all, I can't stop you. But please don't use Epipaleolithic wrongly.

"Then for example if R1b is indeed 30000 YBP why we have Malta Boy as R* 24kybp? An example of Isolation or Drift?"

Exactly. Just look at the tree in the original Mal'ta genome paper: it's very apparent that the boy's lineage was very drifted after the divergence from pre-R1.

"BMAC IMO effected Sinastha,Andronovo,Arkaim etc and Brought The IE Influence towards Uralic folks anyway certainly"...

I'd like to read a well structured explanation on how to fit into archaeology the expansion of M417. Just "influence" doesn't seem enough and anyhow we are still lacking the European connection of BMAC, necessary for the European origin hypothesis that several people spouse here.

"We can't Put Z-93 to be spreading with A Single culture as its history can be more complex"...

One of the issues is that M417 and its direct descendants, including Z93, seem to have expanded suddenly in a very narrow span of time, say 1000 years or so. That seems to demand a very marked archaeological explanation and not a blurry chain of effects in a long time, so, in this sense I do understand the ones who theorize the Kurgan origin. However the geostructure of Z93 particularly does not fit with this scenario, so there's something amiss and maybe a Neolithic expansion fits better, regardless that in Central Europe its descendants may have hitchhiked Corded Ware.

Nirjhar007 said...

Maju,I have to go now after returning in some minutes i will be able to read and understand your comment which will be very interesting, anyway what is your opinion on Eastern European Z93? can it be in some part a result of Iranian type advents or in some cases peripheral as E Europe is the Western Border of ''Z-93 Zone'' to the East?.

Maju said...

@Mike: the Holocene is not a socio-economic transition per se, but a climate phase. There is no transition to agriculture in Europe: it arrived with the full "Neolithic package" (except pottery) from West Asia. That's why we speak of Epi-Paleolithic.

"Im not sure about Spain, etc - Ask maju".

In SW Europe there's also a lot of Epipaleolithic sites, notably along the coasts and often associated with shell middens, who some think were more than just trash dumps but had a spiritual significance of some sort.

Initially, in the Franco-Cantabrian region, we see a continuation of Magdalenian (but without rock art, except for a few decorated pebbles) in Azilian, which already displays the microlithization tendency typical of the period (but not exclusive of it). It somehow affected the Iberian region too, where it's called Microlaminar Epipaleolithic (it lacks harpoons).

In the second phase there are clear influences from further North in France (Tardenoisian culture, also of Magdalenian derivation), which in Iberia is known as Geometric Epipaleolithic. In SE Iberia particularly (Murcia, East Andalusia) there's continuity of Microlaminar (~Azilian) however even into the early Neolithic (Andalusian sites). This apparent continuity of Epipaleolithic toolkits is common in Cardial Neolithic sites (not just in Iberia but also in Italy, etc.), although there are also some (the majority in a few areas) that show a more clear intrusive (settler) characteristic. In this sense it's a bit puzzling that EEFs appear to be so extremely homogeneous, from the archaeological viewpoint one would have expected greater HG admixture as the wave advanced, at least in the South.

Maju said...

@Alberto: "ANE" has no other definition than affinity to Ma1 (and maybe to the related Afontova Gora sequence).

"But MA-1 in Siberia doesn't mean that all ANE originated in Siberia and spread from there"...

It does: by contrast West Asianness and Paleo-Europeanness form a clade vs ANE, even if they are also different among them. This means that, as the early UP wave expanded westwards into the "the Neanderlands" from the Indian subcontinent, it first branched off to Central Asia/Siberia and then, after arrival to West Asia, it branched off to Europe.

Only later some of that paleo-Siberian genetic pool mixed with other populations, and that's what we call "ANE". This admixture can have happened at various times.

"We just need to wait for a genome from Mesolithic Central Asia to see how it looks like"...

While we wait for more data, I can point you to a curious "Central Asian" (and very particularly West Siberian) component that was spotted in a study some years ago: http://leherensuge.blogspot.com/2010/07/central-eurasian-genetic-specifity.html

It's not quite "ANE" (but something more mixed with East Asian, judging on Fst) but it peaks in the Khanty (neighbors of the Kets), so it could have some relationship. It's not Uralic either because it's low in the Komi and it can't be described as Turkic or Altaic as well (although it may have a weak relation).

Maju said...

@Alberto again (forgot this): "I think that one interesting experiment that David made some time ago was a K9 admixture test where ANE is somehow split in 2 groups: EHG and Central Asian. The results are quite interesting"...

It's not ANE but other two distinct components, which should include some ANE. Anyhow, I do agree that the results are very interesting because, among other things, the influence of EHG is way too low compared with Haak's results for Yamna-like influence. Even Ukrainians are in the 5% range for EHG, showing much stronger affinity with WHG (45%) instead. This is very perplexing.

Maju said...

Chewing on this last, is it possible that Karelian and Samara EHG are not representative of the bulk of Eastern European HGs? They are from the very fringes of the region. Could something like Motala be closer to the real EHG average?

Maju said...

@Davidski: "CTS4385" is not listed in ISOGG. Can you explain what are you talking about?

Davidski said...

Here you go Maju...

http://www.yfull.com/tree/R1a/

Maju said...

@Nirjhar: "anyway what is your opinion on Eastern European Z93?"

The only thing I know of EE Z93 is that it's present and dominant within its kind, as terminal branches of Central Asian derivation, in the Samara area, per Underhill. If you have other information, please share: I can't form an opinion on data I ignore.

Nirjhar007 said...

Maju,
//Exactly. Just look at the tree in the original Mal'ta genome paper: it's very apparent that the boy's lineage was very drifted after the divergence from pre-R1//
OK.
//I'd like to read a well structured explanation on how to fit into archaeology the expansion of M417. Just "influence" doesn't seem enough and anyhow we are still lacking the European connection of BMAC, necessary for the European origin hypothesis that several people spouse here. //
I was speaking in sense of Language than Genetics but i think if BMAC Spread towards Urals and had share on development of Arkaim,Sintastha (Structures of which show clear BMAC influence) etc then it may brought some Z-93 with Indo-Iranian language which effected rather than influenced the Uralic folks vocabulary and that migration of Southern people may be the vanguard of the nomadic intrusions later to Easter European areas i.e. Iranian Nomads.
Now, the question is why would BMAC folks migrate there what was the purpose well in simple terms i think they explored the Metal ore rich areas of there for metal industry which is typical identification of Arkaim,Sintastha and unusual of Steppe cultures, in that case we can assume they used probably the pre-Scythian type nomads(Like Andronovans?) for labor works who existed there.
In Case of European connection for BMAC well it doesn't exist and we merely find some pre-Scythian Pots (Andronovans) in BMAC area when it started to fade probably that fading is related to the aridification that happened due to the 4.2 kilo year old event which also destroyed the Sindhu Civilization south to it and brought the late harappan stage.

Nirjhar007 said...

Maju,
//One of the issues is that M417 and its direct descendants, including Z93, seem to have expanded suddenly in a very narrow span of time, say 1000 years or so. //
can you explain that expansion a bit?.
//The central characteristic of Mesolithic senso stricto is that there is a proto-domestication tendency, typically gathering of cereals, which is more easily gauged than other practices. Another trait is a tendency to sedentarization. I can't think of a third trait right now unless it is sickles but, like sedentarism, these are related to cereal gathering: it's just one trait: being proto-agricultural, proto-Neolithic. A typical Mesolithic culture is Natufian.//
But What about C-N Europe according to you? As Mike has described the different stages of social changes i.e the changes in Hunting patterns, lithic patterns,. like Gravettian hunter -gahteres being different to Mesolithic ones?.

Davidski said...

Look here boys...

https://drive.google.com/file/d/0B9o3EYTdM8lQblJmbjVLQ3F2UHM/view?usp=sharing

PJL is the first South Asian sample down the R-Z93 tree, a Punjabi Jat from Lahore.

Nirjhar007 said...

Maju,
//The only thing I know of EE Z93 is that it's present and dominant within its kind, as terminal branches of Central Asian derivation,//
What is ''dominant within its kind''?

Nirjhar007 said...

Oh not Jats again David they are of Scythian origin btw Is this Z-93* a uncle or Brother to Z-94?

Maju said...

@Davidski: we have been there before: new branches of trees based on commercial DNA testing of NW Europe that become irrelevant when West Asia is sampled in an even much less extensive way. There is a problem of relevance: we cannot properly judge when there is such a strong bias in the sampling strategy as happens with commercial DNA companies, which invest nothing in overall field research.

So the new branch probably does exist, but which is the distribution? And also is that English individual different from North German M417*? Probably not.

Maju said...

"But What about C-N Europe according to you? As Mike has described the different stages of social changes i.e the changes in Hunting patterns, lithic patterns,. like Gravettian hunter -gahteres being different to Mesolithic ones?".

In Central-North Europe or in SW Europe is the same: nothing at all indicates an evolution towards agriculture in either region. There are changes? Sure, there are always changes and more so after a dramatic climatic shift but nothing proto-agricultural and therefore nothing Mesolithic "senso stricto".

All this is basic prehistoric knowledge. I feel as if I'd be at a classroom full of semi-illiterate teens.

Nirjhar007 said...

Interesting Again We find the Archaic R1a branches in British People.

Alberto said...

@Maju: Yes, the results are interesting, and while it's an experiment and I think the EHG cluster might be getting a bit lower scores than the reality, it's still in agreement with the D-Stats I've been seeing lately, which show low affinity between Europeans and EHGs (compared to WHGs), while showing strong affinity between EHGs and Siberians.

But the thing is: Yamnaya and Corded Ware do show strong affinity with EHG, so their presence at that time was real. It's just that at some later point it seems to have vanished, and been replaced by Motala-like HG ancestry, much closer to WHG but carrying a good amount of ANE.

The Central Asian admixture (from the Caucasus-like population) does persist, however (in Europe and Central-South Asia).

How or why this happened I don't know. Is it real? I don't know, but it does seem so. More ancient DNA will tell us what happened.

Nirjhar007 said...

Maju,
// nothing at all indicates an evolution towards agriculture in either region//
But isn't that the point of C-N European Mesolithic?? that no such transition is required as argued by scholars?

Simon_W said...

Maju, that R1a-CTS4385 is just a step upstream from the ISOGG accepted R1a-L664. So you can deny its importance, but this wouldn't remove R1a-L664 as a brother clade of the common ancestor of R1a-Z283 and R1a-Z93.

Nirjhar007 said...

Mike,Maju
I think from simple process of thinking Mesolithic can be without transition towards Agriculture and also with Transition as written here:
//*Some authors use the term "Epipaleolithic" for those cultures that are late developments of hunter-gatherer traditions but not in transition toward agriculture, reserving the term "Mesolithic" for those cultures, like the Natufian culture, that are transitional between hunter-gatherer and agricultural practices.
*Other authors use the term Mesolithic for a variety of Late Paleolithic cultures subsequent to the end of the last glacial period whether they are transitional towards agriculture or not.//
https://en.wikipedia.org/wiki/Mesolithic#Europe

Simon_W said...

Maju, the genetic case for an eastern, non-local origin of the Corded Ware people is very strong. It's only debatable how exactly this came about. Perhaps the Corded Ware culture had a local origin, as many archeologists believe and it was really their Globular Amphorae ancestors who had migrated in. I don't have a strong opinion about this. But the DNA was from the east, this can't be refuted with archeological arguments.

Likewise it isn't just a claim that Baalberge people were MN farmers of local stock with no eastern ancestry, it has been proved with ancient DNA. There is sufficient autosomal DNA from three Baalberge people and they cluster closely with other MN farmers from Hungary, Germany, Sweden and Spain.

And as for the Catacomb people's origin, the paper here
http://eurogenes.blogspot.com/2015/03/population-genetics-of-copper-and.html
found that the Catacomb people's mtDNA differs a lot from the Yamnaya mtDNA. There was an increase in haplogroup U4 fromabout 5% to above 30%. And U4 clearly points to a northern origin.

The basic structure of R1a-M417 is evident in the tree I posted: There is an old branch centered on Northern Germany (CTS4385, or, per ISOGG, L664). And all the rest is its brother clade. This in turn splits into Z283 (northern, central and eastern Europe – erratics also in West Asia) and Z93 (mostly known for its presence in southern, central and western Asia, but also present in eastern Europe, Poland etc.).

Alberto said...

Re: R1a. So for what I understand, we have to sister clades (R-Z93 and R-Z282) of R1a, one European and one Asian. But the father (R-Z645?) is missing? Are there no living persons with this clade?

And then we have the grandfather (R-M417) which is present in both regions?

And also we have an uncle (R-CTS4385?) in Europe?

Ok, so if this is correct, what exactly makes Europe a better place for the origin of both big clades than anywhere else?

Simon_W said...

Of course it's possible that R1a-M417 is from Anatolia and migrated to Europe with the earliest farmers, and that it was just rare and therefore not found in all samples of EEF examined thus far, and that by coincidence it got all of a sudden very common at the time when eastern migrants who had other haplogroups, perhaps C, migrated in. But to be realistic, this isn't a very plausible scenario.

Simon_W said...

@ Alberto

You described it almost correctly, the sister clades you're referring to are Z93 and Z283. Z282 is downstream of Z283, but encompasses almost everything, so my correction is a bit splitting hairs.

spagetiMeatball said...

David, I'm just surprised you don't think the teal people had any R clades present today, despite being around 40% ANE.

Simon_W said...

Well, there is the logically plausible idea that more basal splits in the phylogeny occured closer to the origin than the more derived splits, and the most basal split involves a purely European branch splitting from the rest. This would suggest an origin in Europe

Another, independent category of evidence is the distribution of old paragroups, like M417*(xZ645, CTS4385). Personally I don't think this to be a very conclusive kind of evidence, as some places, like the steppe and central Asia have experienced much more population upheavals than secluded mountain valleys in the Caucasus or Iran.

And then there is ancient DNA, providing a direct window to the past - wherever a haplogroup is found in ancient samples, it is dead sure that this haplogroup was present there at this time, no matter where it is present now.

pnuadha said...

@mike

In fact, the Holocene is very well represented in central and northern europe. By contrast, there is a paucity of Mesolithic sites in SEE, and when seen they show more of a graudal transition from the Late Palaeolithic lithic industries and faunal assemblages.

We both agree that SEE Europe and Italy may have harbored a people who genetically cluster away from WHG and EHG. Could you expand more on the demographics of Europe during the neolithic and how it might relate to the potential of 2 or three genetic clusters.

For example, can WHG, which seems pretty widespread in europe and exclusive to europe, have formed exclusively in the Franco-Cantibrian refuge? Or is that unrealistic and we must suppose that WHG developed before the ice age and settled in different refugiums or WHG developed as a merger of the expanding refugiums after the ice age?

I think it is likely that SEE mesolithic people were genetically distinct from the Franco-Cantibrian types and that WHG types were largely derived from the Franco-Cantibrian refuge. My reasoning is that EHG is different from WHG and the tree mapping of Laz suggests that the differentiation developed in no more that 20k years. Even then there was surely gene flow between the clusters at later periods. As for the SEE mesolithic groups they had about 10k years of ice age conditions to differentiate themselves front he Franco-Cantabrian (WHG) by way of small populations and extremely limited gene flow. But if we go one step further and try to associate SEE mesolithic people with EEF, can it be argued that the SEE mesolithic has the capacity to fuel large scale migrations during the neolithic. If not, then we must presume that the bulk of Basel Eurasian, came from the middle east during the neolithic.

Just give me your thoughts.

Alberto said...

Thanks Simon.

But I still can't understand the logic in it.

If the grandfather is in both regions, and each grandson in one region, why the possibilities of both grandsons originating in Europe is higher?

David mentioned the uncle as a point that favours Europe, but I also don't see how that changes things.

What is it I'm missing?

Simon_W said...

The uncle and the father grew up in the same home, and since the uncle is only found in one place, it is thought that he didn't migrate.

Simon_W said...

Maju, the Andronovo culture probably had lots of R1a-M417. There is no logical reason why it should have had only the further derived variants. OK if I understand you correctly, M417* hasn't been found in modern-day central Asia. But why should it be present there now, thousands of years later? Is there any logical necessity to this? And anyway we're speaking about the presence in samples not in entire populations. The Anatolian M417* may be from central Asia via Indo-Iranians, or even a late arrival with the Turks. Or it may be from European relatives of the Corded Ware. Possibly there are even different origins for Anatolian M417*.

An origin of Z93 around the BMAC area would imply that it originated later in the Indo-Iranian expansion to southern and western Asia and then back-migrated northwards. Possible, though I have my doubts about the conclusiveness of the present-day absence of Z93* further north.

I would also like to point out that Yamnaya wasn't just a very specific R1b clade completely unrelated to western R1b. Predominantly the Haak sample was, yes, but there was also one Yamnaya male with R1b1a2a-L23, and this is not that far from the R1b1a2a1a2 considered typical for western Europe, which was found in a German Bell Beaker individual who autosomally resembled modern northern Germans. Also Basques are not as close to EEF as Sardinians are.

Alberto said...

@Simon

"The uncle and the father grew up in the same home, and since the uncle is only found in one place, it is thought that he didn't migrate."

Why would they grow in the same home? They have possible fathers in both regions. Do you really mean that the very same individual had two male sons, one with the uncle's mutation and another one with the father's mutation? This is one in a million chances to happen.

Unknown said...

Colin W

"Mesolithic Euopre, Glacial refugia..."
I think SEE itself was diverse – by the Mesolithic – there was several clusters of Mesolithic groups, often separated by hundreds of Km. So perhaps there was not a single "Balkan HG" group. Some must have been essentially WHG, some almost EHG (around eastern Romania), some "Mediterranean - Anatolian".
As to the rise of WHG itself – my guess would be as good as anyone’s. But almaost certainly , we cannot trace it back to a single point of origin – eg Iberian LGM refuge. Because, by the time of the currently sample Mesolithics – c. 6 – 7 kya, there must have been significant admixture – as we can see from Iberia to NW Europe to the Carpathians the WHG are almost identical, and SHG are only separated by the admixing of EHG.

“SEE Mesolithic fuelling Neolithic..””



At present, I suspect minor admixture. The Mesolithic in SEE is not dense, apart from certain regions. The large demographic rise, the advent of a “full Neolithic package”, differential settlement patterns very much reflects the traditional view of demographic influx into SEE. But from where ? Im not sure its all Levantine, as Maju would maintain. Naturally, we might aloso be missing the fact that Mesolithic SEE foragers lived by the Sea – in areas now under water.



Unknown said...

Maju

I think maybe you’re confusing people's commenets . What I was describing to Nirjahar was the evolution of hunter-gatherers. I never stated that northern – central Europe verged on agriculture in the Mesolithic, but they *did* become slightly more sedentary; and certainly had different economic strategies and industry compared to their Palaeolithic predecessors.

And I have already described that there is seminological complexity with how one defines “Mesolithic”.

So I'm not sure what your issue is here.

Davidski said...

I've added Alberstedt_LN, BR2 and Halberstadt_LBA to the bar graph above.

Maju said...

@Simon: "The basic structure of R1a-M417 is evident in the tree I posted: There is an old branch centered on Northern Germany (CTS4385, or, per ISOGG, L664). And all the rest is its brother clade. This in turn splits into Z283 (northern, central and eastern Europe – erratics also in West Asia) and Z93 (mostly known for its presence in southern, central and western Asia, but also present in eastern Europe, Poland etc.)".

I didn't understand that at first but now I see your point and that of Davidski.

How do you know it is centered in North Germany, how do you know it excludes the Turkish branch of M417* mentioned by Underhill?

In any case it does seem that Z93 and Z283 (precursor of Z282) are "brothers", form a clade to the exclusion of the North German M417* but without knowing much of West Asian M417* we can't jump to conclusions about the origin and spread of M417. Looks promising but needs wider sampling and scholarly formalization of some sort before I can be persuaded, really.

It still does not explain the south-to-north pattern in Z93.

Maju said...

@Nirjhar: you make too many insidious questions and sometimes you even answer them yourself (i.e. Mesolithic and Epipaleolithic usage). But not sure if you realize.

As for "of its kind" it means obviously "of R1a" (context!)

pnuadha said...

@mike

But from where ? Im not sure its all Levantine, as Maju would maintain.

What areas would you suggest? What is else is left as a possible base for the EEF expansion besides SEE and the Levant? Turkey? Mesopotamia?

Also, what was the demographics like in Italy during the mesolithic and how much gene flow could there have been with the other refugiums? Bacause we do not see evidence for a 2nd "southern" component Italy was realistically one of WHG or EEF-like during the mesolithic.

Unknown said...

Colin

Yes Turkey mihjt have certainly contributed; at least later .
It had a healthy mesolithic population around Maramara and Antalya, but the Neolithic in NW Turkey is poorly known at present

Italy is a blank spot (genetically) at the moment; but I don't think it contributed much
In fact I think they Cardial people there were a demographic expansion from the Balkans

Nirjhar007 said...

Maju,
//But not sure if you realize//
I do.
//As for "of its kind" it means obviously "of R1a"//
So E European Z-93 is what unique?

pnuadha said...

@mike,

Thanks. It seems kind of random to me that the population in southeast europe would be so lacking in populations compared to the franco-cantibrium refugium during the ice age and the levant/turkey during the neolithic.

At the onset of the ice age i would have expected central and east europeans to move into, and concentrate, in SEE. After the ice age i would have expected conditions in SEE europe to be decent enough to match turkey and central europe.

Why was it apparently so barren (i realize you already mentioned that the dense sites could be under water).

pnuadha said...

Actually, its better to cut to the chase. Could you tell me which regions of west eurasia likely had "large" mesolithic populations and which ones were more sparse?

I know things a speculative but I'm curious about the base populations before the neolithic which can increase the capacity in the right circumstances.

Unknown said...

Colin

Most of central, northern, eastern and western Europe had notable populations. Id need to read more carefully as to where specifically these were. Ie some micro regions more than others.

(See a good book "Mesolithic Europe" Ed. G Bailey 2008. Also ;Hunters, Fishers & Farmers in EE' - R Tringham)

But for SEE
* Danube Gorge - Serbia/ Romania
* a few in Carpathian basin - Danube bend, jazgsjag region, etc
* a few sites in Istria, Epirus, and around Franchthi and north Argolid.
* Marmara (either side of Constantinople),
* seems almost none in Bulgaria

Marnie said...

@Colin

Just spent the day today at the 2015 Paleoanthropology Annual Meeting.

Without naming names, I spoke with several very senior researchers regarding current archaeological programs that are looking at the UP-Mesolithic-Neolithic transition in the following places:

Iberia, the Balkans, Asia Minor, Armenia and the Eastern Mediterranean.

Currently, only Iberia and the Eastern Mediterranean have funding for the kind of multi site, fine scale, long time scale projects that could be used to construct a detailed picture for the population dynamics of the UP-Mesolithic-Early Neolithic.

Even in Iberia, where funding and available sites are adequate, it will be another five years, at minimum, before a reasonably clear picture can emerge for Mesolithic population dynamics in Iberia. (That's according to a very prominent, very well funded researcher from Germany who is in charge of this program.)

For West Asia, where funding is far less adequate than in Iberia, the question of Mesolithic population dynamics won't be answered in the immediate future. The funding simply isn't there (in the Balkans, Asia Minor, or Armenia).

The disconnect between those working on PIE and population genetics studies, and archaeologists and paleoanthropologists trying to do the work on the ground to answer questions about population densities and dynamics during the UP-Mesolithic-Neolithic transition, is gaping and quite apparent.

Please, anyone on this blog, show me actual publications that can demonstrate population densities for Europe or West Asia on fine scale (time and geography) in the Late UP or Mesolithic.

Where are these publications?

Nirjhar007 said...

Marnie,
//For West Asia, where funding is far less adequate than in Iberia, the question of Mesolithic population dynamics won't be answered in the immediate future. The funding simply isn't there (in the Balkans, Asia Minor, or Armenia).//
That is just ridiculous.

Davidski said...

Ancient DNA from the Near East is on the way. Here's a list of samples being sequenced by one of the teams.

https://sites.google.com/site/pinhasierc/home/samples

spagetiMeatball said...

David, you're in Australia and you're interested in genetics. Can I ask you a slightly off-topic question. Do you know if anybody is interested in using DNA testing to shed some new light on the "Tamam Shud" case?

Davidski said...

Sorry, I have no idea.

Unknown said...

Marnie

I don;t think such as study exists. Its pretty hard to do because

1) no one person has an absoulte commnad of knowledge over all of Mesolithic Europe
2) youre right, palaeolithic research often takes a back seat to 'sexier' topics like Bronze Age migrations
3) its hard to make simple inferences from settlement site numbers to actual population densities. There are sveral factors which affects it 'archaeological visibility' and attendant issues of post-depositional loss, settlmeent patterns and strcuture (dipersed vs concentrated, upland - lowland seasonal shifts), etc.

But for SEE, here are two recent studies by specialists from SEE

"‘Pre-Neolithic’ in Southeast Europe>
a Bulgarian perspective" Gurova / Bonsall

"Adaptations and Transformations of the Danube Gorges
Foragers (c. 13.000 – 5500 BC): An Overview" - D Boric

Both on Academia

Unknown said...

And for italy / Dalmatia: the mesolithic / neolithic transition in NE italy and the Adriatic basin- Paolo Boagi

Marnie said...

@Davidski

So, for this Pinhasi study, let's actually plot out the coverage for the Balkans and West Asia:

Turkey (a very big place): early Neolithic only

Georgia: UP, Mesolthic, Early Neolithic

Israel: Early Neolithic

Armenia: Early Neolithic

Croatia: Mesolithic, Early Neolithic

Southern Coastal Greece: Mesolithic

Greece: Neolithic

Hungary: Neolithic

Bulgaria: Early Neolithic

Romania: UP, Mesolthic, Early Neolthic

Ukraine: Early Neolithic, Neolithic

Iran: Mesolithic, Early Neolithic



Missing:
--------

Turkey: UP, Mesolithic

Israel: UP, Mesolithic

Armenia: UP, Mesolithic

Croatia: UP

Southern Coastal Greece:
UP, Neolithic

Greece: UP, Neolithic

Hungary: UP, Mesolithic

Bulgaria: UP, Mesolithic

Ukraine: UP, Mesolithic

Iran: UP


So, in summary, with the exception of Georgia and Romania, poor coverage for the UP and Mesolithic for most of the Balkans and West Asia.

According to the researchers who are doing the Iberia study, they are looking exhaustively on fine scale at 500 sites, starting in the UP 30,000 years ago, carefully in fine detail tracing through the climatic variation of the late UP and Mesolithic, and into the early Neolithic.

In terms of population variability, according to the researchers I spoke to, it's in the late UP/Mesolithic where you would expect to see the greatest population variability. With the exception of Romania and Georgia, Pinhasi's study (not an archaeology or climate study) does not have sufficient ancient DNA coverage for the late UP-Mesolithic transition in the Balkans and West Asia.

This study would have to be married with the kind of exhaustive archaeological study being done for Iberia to make valid deductions about population densities in West Asia and the Balkans. (At least if we adhere to the standard that was described to me today by this prominent German researcher.)

It is not.

I would also note that Pinhasi's study has very poor coverage for North West Europe (which at least according to Willerslev could have been a population refuge during the UP-Mesolithic.)

Pinhasi's study looks like a good start. However, certainly not enough coverage to make the kind of grand conclusions that I've seen from both Pinhasi and Haak in recent publications ("mass migrations" during the Bronze Age, "on the brink of extinction").

As a member of the interested public, do I feel manipulated?

Yes.

Is there a loss of trust in this type of research?

Yes.

Am I much more ambivalent and skeptical of the publications coming from these researchers?

Yes.

Do I want to fund this type of research in the future?

No, not unless this research is probably paired with an archaeological study to look at climate and population densities on fine scale. Not unless there is appropriate coverage for the UP-Mesolithic-Neolithic transition so that population dynamics and climate are properly captured.

Otherwise?

Lot's of headline grabbing titles, but insufficient data supporting the titles.

Unknown said...

Marnie

A lot of it has to do with budget . Proper aDNA labs cost hundreds of millions of dollars

I'm sure the whole world will be sampled in 20 years time

Nirjhar007 said...


Marnie
//Iran: Mesolithic, Early Neolithic//
But which area North or South?

Krefter said...

@Marnie,
"So, in summary, with the exception of Georgia and Romania, poor coverage for the UP and Mesolithic for most of the Balkans and West Asia."

We should be grateful for any ancient DNA. I'm more disappointed than anything because it takes out the fun of trying to figure it out with modern DNA and other evidence.

If like in Europe a handful of samples from each region gives a very good idea who the people were genetically, we'll be well off with this data set from Balkans-West Asia.

The big-picture story of the genetic origins of Europe has pretty much been discovered. We can see the admixture events that lead to the creation of modern Europeans in Haak 2015. Much has been learned about west Asia via ancient Euro studies, and studies with huge amounts of west Asian DNA seal the deal for west Asia to.

Nirjhar007 said...

Krefter,
//Much has been learned about west Asia via ancient Euro studies,//
That is a what i should say a very funny statement anyway what is your idea on what ''West Asia'' is and which specific areas falls under it?

Marnie said...

@Mike

"1) no one person has an absolute command of knowledge over all of Mesolithic Europe"

Colin's question was about population densities for West Asia during the Mesolithic. He wasn't asking about all of Mesolithic Europe. There are no studies being done for West Asia of the type being done for Iberia. If so, and it is deemed necessary to fund a fine scale study for Iberia, the same fine scale study should be required for West Asia.

"2) youre right, palaeolithic research often takes a back seat to 'sexier' topics like Bronze Age migrations"

Personally, I find the Paleolithic and Mesolithic a lot 'sexier' than the very boring, nothing really happened Bronze Age. Humans were fully modern before the Ice Age. The Bronze Age or Neolithic in Europe? It's shuffling the same cards of very similar populations that haven't changed much at all in tens of thousands of years. Who cares. I actually find it pathetic to read the genetic ancestry forums and watch even prominent researchers agonizing for literally hundreds of hours about whether their y-dna is R1a or R1b, or quibbling over a few alleles here or there. It's really shocking to see how much the genetic ancestry community has pulled the wool over people's eyes as to how little these minor genetic differences matter.

"3) its hard to make simple inferences from settlement site numbers to actual population densities."

Hard? Hey, research is usually hard. You're talking to a chip designer. I don't want to hear about the difficulties, expense and long hours required to do careful research.

For some parts of Western Europe, this difficult to do research is being funded and carried out. Not so for the Balkans and West Asia.

"There are several factors which affects it 'archaeological visibility' and attendant issues of post-depositional loss, settlmeent patterns and strcuture (dipersed vs concentrated, upland - lowland seasonal shifts), etc."

Mike, don't try to baffle gab me. I know this stuff. There are top researchers who are working on estimating population dynamics in Iberia from the UP onward. They're dealing with the above challenges, one way or another.

However, this is not being done for West Asia or the Balkans. Without these kinds of studies, grand statements about population dynamics like "mass migrations", "almost total extinctions" are just fairy tales and thinly veiled attempts at self promotion of not very well done research.

Krefter said...

@Marnie,
"However, certainly not enough coverage to make the kind of grand conclusions that I've seen from both Pinhasi and Haak in recent publications"

My only concern with the Basal Eurasian-WHG-ANE, Neolithic transition, and massive migration from the steppe, etc. theory is that somehow the tests people are doing with DNA are inaccurate.

Being able to give accurate proportions of ancestry from pre-historic humans sounds like a fairy tale. I trust the experts know what they're doing, but it's still hard to believe.

Assuming the methods used are accurate, which they probbaly are, there's little room for debate. A massive migration from the steppe did occur, unless a pop just like Samara Yamna were hiding in a magical candy forests throughout Europe and decided to come ut 4,000-5,000YBP.

Krefter said...

@Nir,
"That is a what i should say a very funny statement anyway what is your idea on what ''West Asia'' is and which specific areas falls under it?"

Basal Eurasian, UHG, and ANE. How were they discovered?

West Asia's genetic history is intertwined with Europe.

Nirjhar007 said...

But what is West Asia?

Krefter said...

@Marnie,
"There are no studies being done for West Asia of the type being done for Iberia."

We can't expect people to do any serious research in West Asia in the near future because the circumstances are differnt there. Where are they going to get the money, and how do we know if they won't get caught in the middle of a war.

Marnie said...

@Mike

"A lot of it has to do with budget . Proper aDNA labs cost hundreds of millions of dollars"

There is no medical value to this type of research. It's curiosity only. There is no medical imperative for this research to be done. No need for faked up headlines that falsely lead people to believe they are more different from each other than they really are. The only "value" to this research is so that the genetic ancestry companies and Spencer Wells can sell a phony baloney story to some gullible people who get off on thinking they're directly descended from Agamemnon, Leif Eriksson or Oisin.

"I'm sure the whole world will be sampled in 20 years time."

I really don't care one way or another. I just don't want to read any more faked up papers with gaping holes in supporting data.

Nirjhar007 said...

Krefter whatever Is you notion on Central Asia, West Asia or South Asia is use specific terms like names of countries or geographic regions as those terms are not very conclusive, For Example Iran is included in the West Asian area by many but even within the country's geography there are stark differences in proto-history.

Marnie said...

@Krefter

"We can't expect people to do any serious research in West Asia in the near future because the circumstances are differnt there. Where are they going to get the money, and how do we know if they won't get caught in the middle of a war."

I just spoke with several researchers who are doing archaeological research in Turkey and Armenia. According to them, there is no problem doing archaeology in the Balkans, Turkey, Armenia or Georgia . . . just a lack of funds.

And if so, then papers should confine their conclusions to the funded areas only and openly state that they cannot make conclusions about the unfunded areas, because their fund masters were not interested in sufficiently funding research in the important UP/Mesolithic areas of the Southern Balkans, Turkey, Armenia and Georgia (instead of underhandedly pretending to have sufficient data coverage to make statements about these areas.)

Davidski said...

Mike,

Ancient DNA labs don't cost hundreds of millions of dollars. More like a few million bucks, and that's for the top notch labs.

Also, the costs of sequencing ancient DNA are falling with each year. I don't know how much it costs to sequence a few samples using that new baiting method from Haak et al., but it can't be more than a few thousand dollars, otherwise it wouldn't be done.

Marnie,

Paleogenomics has a lot of relevance to medical genetics. For one, knowing the precise population history of different human groups is very useful in GWAS work. Also, the methods developed when studying the prehistory of humans, just for the sake of knowing what happened, can be applied to studying the genetic history of pathogens.

Grey said...

Colin Welling
"Could you tell me which regions of west eurasia likely had "large" mesolithic populations and which ones were more sparse?"

HGs have high density around wetlands so if we assume for the sake of argument that the drowned land since the LGM used to be wetlands (would need to ask an expert but seems plausible enough) then the highest population density areas are likely to be underwater.

Take western Europe for example and you can see the big chunk of the Cantabrian refuge now underwater.

http://cdn.zmescience.com/wp-content/uploads/2015/01/31836.jpg

Looking at Europe as a whole

http://dailygrail.com/sites/dailygrail.com/files/imagecache/BlogImage-Large/storyimages/anteluvian-europe.jpg

the main areas might be that bit between Italy and the Balkans, the Aegean itself maybe and of course the Black and Caspian seas.

Also looking at the global map you can see why SE Asia maybe had a bigger effective population than India for a time.

https://62e528761d0685343e1c-f3d1b99a743ffa4142d9d7f1978d9686.ssl.cf2.rackcdn.com/files/48094/area14mp/ygrpfn22-1399582502.jpg

So (maybe) lots around wetlands and very few elsewhere.

Grey said...

Simon_W
"I would also like to point out that Yamnaya wasn't just a very specific R1b clade completely unrelated to western R1b."

If the dramatic expansion of specific R1b clades in Western Europe was connected to LP then it would have been a fluke which individuals from which clades had it and which didn't.

Alberto said...

Great to see all those samples are coming. I hope that by Autumn/Winter we can start getting some new sequences.

@Nirjhar

The Iran samples are from the Belt Caves, South Caspian area. Mesolithic and Neolithic. I think those 6 samples can be a key to understand many of the things we've been discussing lately.

Also the 150 samples from "Central Asian and Russian and Ukrainian Neolithic" should be very interesting.

70 samples from Romanian (from UP Mes and EN).

1 Baltic Mesolithic

76 samples from Neolithic Portugal

50 from Neolithic Greece

And much, much more...

Krefter said...

@Marnie,
"funding research in the important UP/Mesolithic areas of the Southern Balkans, Turkey, Armenia and Georgia (instead of underhandedly pretending to have sufficient data coverage to make statements about these areas.)"

That's a good point. With ancient genomes though all we need are a few ranging in age and region from west Asia and they'll probably answer a lot of genetic questions.

It doesn't look like there has ever been a major genetic impact caused by people from outside of west Asia into West Asia since at least the Neolithic, because they have so much Basal Eurasian. So the changes that occurred in west Asia over the last 10,000 years or whatever came from within west Asia.

If the Teal people came out of central Asia, they were still majority ancient West Asian.

Karl_K said...

@Marnie

"Hard? Hey, research is usually hard. You're talking to a chip designer."

Yes! Marnie is back! And, BTW, she is a chip designer!

Davidski said...

Marnie was busy, writing letters and stuff...

http://linearpopulationmodel.blogspot.com.au/2015/04/anonymous-online-use-of-internet-major.html

Karl_K said...

@Davidski

"Marnie was busy, writing letters and stuff..."

So... I see that you and your pals at the Max Plank Institute have decided to just not answer these letters?

Davidski said...

It's all done subliminally nowadays at places like the Max Planck. You don't really know that you've been contacted, you just feel it as a vibe.

Nirjhar007 said...

Alberto,
//The Iran samples are from the Belt Caves, South Caspian area. Mesolithic and Neolithic. I think those 6 samples can be a key to understand many of the things we've been discussing lately//
WOW!!! Just what the doctor ordered!! Seriously this is huge really huge this means quite soon the real origins of Proto-Indo-Europeans will be answered in terms of genetics and i have news i can't leak the details but BMAC will be tested the project is beginning and i am keeping contacts with the makers and arguing to sample Jeitun as well!! Jolly good times Alberto Jolly good times.....

Unknown said...

Ha ha Nirj I love your enthusiasm

Nirjhar007 said...

Mike I think in 3 years max. The PIE question will be done and dusted:)....

Maju said...

@Nirjhar: http://en.wikipedia.org/wiki/Western_Asia

Maju said...

@Grey: that first map of sea levels you linked to is not accurate: the Iberian side of the Franco-Cantabrian refuge (and in general all the Iberian peninsula) has a very short and abrupt continental platform, so only a few kilometers were lost to rising sea levels. On the other hand the area of emerged land near Brittany was surely larger.

The second and third maps are instead correct (for the LGM, not for other less extreme periods) - except that the Caspian Sea, a salty lake, was bigger rather than smaller and that the "emerged lands" shown in much of Northern Europe, particularly the Baltic, were under a thick ice sheet, at least most of them.

A general quality bathymetric reference for the whole World: http://cmtt.tori.org.tw/data/App_map/maplist.htm

Maju said...

As for the issue of Turkey I've been reading some recent materials in the last 48 hrs. and there are some novelties. The West Anatolian Neolithic now seems older than thought before (I stand corrected) but it's not clear if:

1) It's old enough? The oldest plausible date, based on livestock, I've seen is 6800 BCE, what overlaps at best with the initial Thessalian Neolithic, other more purely material data point to more recent dates, 6500, 6250... On the other hand some of these authors seem inclined to revise the initial Thessalian (or generally Greek) Neolithic to more recent dates but I do not have clear if their dates stand.

2) More importantly maybe, is it the same culture, grosso modo, as the Thessalian Neolithic one? Here the answer seems to be NOT, even if there were indeed contacts and influences between coastal West Anatolian sites and Thessaly or other Greek early Neolithic sites.

But archaeologically speaking the scenario is a bit more open than I used to think and it makes some sense that the Thessaly and West Anatolian wave were, if not directly related, at least quasi-simultaneous.

Refs.
· M. Özdogan, "Archaeological Evidence on the Westward Expansion of Farming Communities from Eastern Anatolia to the Aegean and the Balkans". Current Anthropology 2011.
· M. Özdogan, "Anatolia and the Balkans: archaeology". The Encyclopedia of Global Human Migration, 2013.
· B.S. Arbuckle, "Data Sharing Reveals Complexity in the Westward Spread of Domestic Animals across Neolithic Turkey". PLoS ONE, 2014.

Notice that all these studies are somewhat pro-Anatolian in their conclusions about Thessaly and European Neolithic in general, but the evidence backing such claims is blurry at best. The matter needs to be looked at with great care.

Also let's not forget that after the initial Neolithic, there was, c. 5000 BCE, another wave from Anatolia (related to Tell Halaf, it seems) that conquered much of the Balcans (Dimini-Vinca) and surely changed their demography in very substantial ways. Ignoring this "detail" may cause confusion.

Maju said...

@Marnie: "Please, anyone on this blog, show me actual publications that can demonstrate population densities for Europe or West Asia on fine scale (time and geography) in the Late UP or Mesolithic".

The only one I'm familiar with (and I consider a fundamental reference) is Bocquet-Appel et al. 2005: http://www.evolhum.cnrs.fr/bocquet/jas2005.pdf

It does not consider Epipaleolithic however, only UP in four stages.

Maju said...

@Davidski: "Ancient DNA from the Near East is on the way. Here's a list of samples being sequenced by one of the teams".

I just love it! Those samples should help settle much of these debates.

As a curious note Taforalt is not in "France" but in Arif (Northern Morocco). Also another line reads "African Upper Paleolithic - Russia", what is hilarious. XD

Marnie said...

@Maju

"It does not consider Epipaleolithic however, only UP in four stages."

According to some of the researchers I spoke to yesterday, most of the population fluctuation for Europe happens in the Late UP and Mesolithic. A UP study isn't going to capture this.

Marnie said...

@Maju

"I just love it! Those samples should help settle much of these debates."

Regarding the Pinhasi study, I can already see that there are huge gaps.

I don't think you will get the kind of definitive answers you are hoping for from the Pinhasi study.

See my post above for what is missing.

Unknown said...

Maju,
There was a paper last year about farmers going from South Anatolia and going to West Anatolia and Greece, via island hopping. I'm pretty sure European farmers and those going to Hacilar will be very much identical.

Gill said...

PJL aren't all Punjabi Jats, they could be any kind of Punjabi living in Lahore. Admixture shows them coming from all castes/backgrounds and other parts of India.

Also we don't know for sure if Jats are actually Scythian descended yet.

Nirjhar007 said...

// we don't know for sure if Jats are actually Scythian descended yet.//
Most likely Gill most likely.

Maju said...

@Marnie: I see an obvious gap in Greece particularly (Romania may not be similar enough) but anyhow I'm very happy about the large expansion of the aDNA database that this effort implies. It also seems to cover key Atlantic areas (Portugal, SW France), and there are even some African samples, so it's quite close to what I asked to Santa these Christmas. :)

postneo said...

The teal component in Georgia is later than that of yamnaya and could be due to recent iranization

@nirjhar
Jats being of Scythian origin is a fable. It may have some partial truth only.


@davidski
Ancestors of L664 did not come from Siberia or east Europe but were earlier migrants from south Europe and west Asia.

Exclusive z93 of east Europe is bunk. It's not ancestral to anything.

Maju said...

@Nirjhar: I told almost exactly the same to Marnie in private communications but there's nothing that seemingly will get her out of her conviction, which I'm sure it's at least largely wrong. *Big shrug* and let's move on.

Grey said...

Maju
"A general quality bathymetric reference for the whole World:"

http://cmtt.tori.org.tw/data/App_map/maplist.htm

saved that link, ty

Grey said...

If there are potential medical benefits from tailored medicine then as it becomes more obvious then rich people and governments in other parts of the world will start funding research into their own ancestries. If so the coverage will snowball rapidly.

«Oldest ‹Older   401 – 600 of 640   Newer› Newest»