search this blog

Wednesday, June 10, 2015

101 ancient Eurasian genomes (Allentoft et al. 2015)


It'll take me a while to digest all of the information in this massive new Allentoft et al. paper. But I've already noticed that, just like in Haak et al. 2015, the Yamnaya samples are again from the eastern half of the Yamnaya horizon. This time, however, not all of the Yamnaya individuals carry Y-haplogroup R1b; one of the five samples belongs to Y-haplogroup I2a (see here).

So I'm wondering what more westerly Yamnaya sites will reveal in the future, considering the predominance of Y-haplogroup R1a among the Corded Ware individuals sampled to date, and the close genome-wide relationship between the Yamnaya and Corded Ware?

Abstract: The Bronze Age of Eurasia (around 3000–1000 BC) was a period of major cultural changes. However, there is debate about whether these changes resulted from the circulation of ideas or from human migrations, potentially also facilitating the spread of languages and certain phenotypic traits. We investigated this by using new, improved methods to sequence low-coverage genomes from 101 ancient humans from across Eurasia. We show that the Bronze Age was a highly dynamic period involving large-scale population migrations and replacements, responsible for shaping major parts of present-day demographic structure in both Europe and Asia. Our findings are consistent with the hypothesized spread of Indo-European languages during the Early Bronze Age. We also demonstrate that light skin pigmentation in Europeans was already present at high frequency in the Bronze Age, but not lactose tolerance, indicating a more recent onset of positive selection on lactose tolerance than previously thought.

Allentoft et al., Population genomics of Bronze Age Eurasia, Nature 522, 167–172 (11 June 2015) doi:10.1038/nature14507

See also...

R1a-M417 from Eneolithic Ukraine!!!11

728 comments:

«Oldest   ‹Older   601 – 728 of 728
Davidski said...

Colin,

Your argument, as far as I can see, is that present-day Northwestern Europeans received their high frequencies of R1b and high level of Yamnaya-related genome-wide ancestry from a mass Yamnaya migration into Western Europe via the Carpathian Basin.

But the evidence we have contradicts this, because no one sampled to date from the ancient Carpathian Basin even shows as much Yamnaya-related ancestry as present-day Northern Europeans, let alone more.

So based on the evidence we have to date it seems that something very different to what you're arguing took place. Here's what I think the data shows:

1) Present-day Northern Europeans received most of their Yamnaya-related ancestry from Corded Ware.

2) A large part, if not most, of the R1b in Northern Europe comes from Corded Ware.

3) Most of the R1b in southern and western Europe comes from the Yamnaya bands that moved into the Balkans and Hungary.

4) The Yamnaya bands that moved into the Balkans and Hunagry were somewhat different from the eastern Yamnaya sampled to date; with more EEF/WHG and less EHG/ANE.

5) The high frequencies of R1b in much of Europe are due to massive found effects well after the Yamnaya migrations, and thus need not be accompanied by high levels of Yamnaya genome-wide ancestry, and usually aren't.

Let's see what ancient genomes from obvious Yamnaya offshoot cultures from Hungary and the Balkans show us in the near future. They might offer something new, but I suspect they'll back up the scenario I outlined above.

pnuadha said...

You have correctly summarized what i think is most likely. But even if hungarian yamnaya was not the main source for genetic changes in northwest europe, I would suspect a migration, north of the caspian and separate from CW, to have carried lots of r1b, autosomal dna that was more likely to end up in Northwestern Europe than Central Europe (as a way of balancing the the greater autosomal influence of corded ware in Central Europe), and Italo-Celtic.

I just can't mesh corded ware with the 3, simultaneous, events that took place in northwest europe. Those events were a dominance of r1b, an autosomal shift towards yamnaya being no less profound than central europe, and italo-celtic taking over as a fresh from the steppe migration.

Alberto said...

@Seinundzeit

"I take this evidence to be determinative, mainly because you don't see such good fits everyday. We are dealing with "perfection" here, from a stas-based perspective.

From a purely pragmatic perspective, nothing shown via ADMIXTURE or PCA can beat this."

Again, I agree. I'm absolutely interested in these results. I'm just trying to understand why they happen while at the same time they break other models completely.

"Also, I think it's time to scrap the ANE-WHG-ENF model."

Yes, probably. But cross check the qpAdm results with the ANE-WHG-ENF model. What happens? A population with high levels of WHG goes to South Asia and from the mix: There is a sharp increase in ANE and a very sharp decrease in WHG (that almost disappears). So I'll say it again: what if ANE is just a composite of WHG+South Asian? Wouldn't that explain why we're not finding WHG in places that according to the steppe hypothesis (and to these qpAdm results) should have a good amount of it?

Now let's see about the Near East. EEF had WHG levels starting at 25-30% and no ANE. Ancient Anatolians (and probably Transcaucasia, Syria, North Levant, North Mesopotamia) could not be far from those levels. And then when ANE appears, WHG disappears.

Why Moroccans (who lack ANE) do instead keep some good 15-20% WHG?

And why do Basques, who are south Europeans, have North European levels of WHG, when coincidentally they have very low ANE? Similarly Spaniards have more WHG and less ANE than other south Europeans.

Haplogroup R has a clear South Asian origin. And when R appears in North Eurasia (WHG territory), ANE appears at high levels and WHG decreases sharply.

Why EHG would never show as admixed between WHG + MA1 (who believed to be pure ANE)?

So I think this hypothesis is worth a look, because if we discover that ANE is WHG+SE, it could be a breakthrough to understand prehistory.

Arch Hades said...

David, wouldn't the British(who are Northern Europeans), French, and Spanish get most their Yamnaya ancestry and R1b from the Bell Beakers?

Unknown said...

Alberto

"because if we discover that ANE is WHG+SE, it could be a breakthrough to understand prehistory."


By SE: do you mean South Asian?

Of your theory is correct, it could be hard to tease out- at least for eastern europe. Id expect that it arrived in several waves, the first one as early as 18 kya

Gioiello said...

@ Nirjhar007
I am seeing that YFull came back with its date of the separation between R1a and R1b: 22200 ya. Anyway English ancestry Jones line of descent is 18900 ya as R1a-M459* (pretty as R1b1*):
R1a PF6233/F3570 * FGC32014/Y215 * CTS8008/M726... 108 SNPs formed 22200 ybp, TMRCA 18900 ybp
R1a*
R-YP4131 YP4148 * YP4134 * YP4188... 98 SNPs
id:YF03626 new
R-M459 PF6194/F2215/M701 * V1412/PF7514/M623 * CTS11411/M798... 37 SNPs formed 18900 ybp, TMRCA 14300 ybp
-------------------------------------------
I think that it is very likely that at that time they were in the Siberian/Northern European corridor. R1b1-L389- and R1b1-L389+ separated later (16700 ya) and very likely they were Southwards, and I think that the line of separation was the Caucasus: R-V88 and R1b1-L389+ Westwards and R1b1-L389- Eastwards. Of course will know more only from aDNA.

Alberto said...

@Mike

Yes, South Eurasian (or simply South Asian).

It surely arrived in waves. We have already MA1 in Siberia 24 ya. who was first believed to be pure ANE (in David's current model he's about 85% ANE + 15% WHG, I believe). And he's R (and in pre-ANE models he showed a lot of South Asian, though it was attributed to being archaic).

All modern Europeans (but not WHG or EEF) have shown a clear f3 admixture signal with Dai, who have no ANE, so that's another data point that would support this.

Worth investigating, IMO.

Unknown said...

Alberto
Surely that Dai affinity, long noted, has something to do with Haplgroup K being of SEA origin. Again, perhaps linked with haplgroup R, but perhaps very basally so.?

Alberto said...

Yes, probably it is all related to that tree (K -> P -> R, or however it goes).

Simon_W said...

I had a hard time spotting the first of the BA Armenians in David's PCA! Only after having read Matt's comment I knew where I had to search. It's a very interesting result. RISE397 dates to roughly 950 BC, which is rather late, we may expect this to have been an ethnic Armenian. Looks like in the middle of Balkars and North Ossetians.

RISE407 is only little earlier, about 1000 BC, and very similar to the former, she sits among Lezgins and Chechens.

RISE423 is a bit from an earlier era, about 1300 BC, looks again like in the overlap area of North Ossetians and Balkars.

RISE408 looks interesting and slightly different from the others. He looks like slightly shifted towards Balkan populations. And he's the one that is labelled „Iron Age Armenian“ in Supplementary Table 9. Although according to Supplementary Table 1 he's dated to about 1100 BC, 1009 BC at the latest. Interestingly he's the one with haplogroup J2b, which is common in parts of the Balkans. A connection?

Simon_W said...

Perhaps in the Bronze Age the central North Caucasus and Armenia were genetically very similar. It's just strange that the Georgians inbetween are different, they are more southern shifted, towards Iranians and Kurds, in the PCA.

Unknown said...

Simon
Definitely. I think bronze age Balkans and northern Near East will be linked by J2b & R1b-Z2103

Simon_W said...

I'd say the bridge between Armenia and the central North Caucasus may have been a highway on which Near Eastern loan words were passed to the steppe. Kartvelians may have been somewhere nearby, near the Black Sea, which would explain the shared loans.

Grey said...

Kristiina

"so I do not get how Pashtuns can be modeled without input from India."

Maybe the input from India came from the "wrong" direction i.e. a boomerang. HGs from India expanding to north Eurasia, picking up some genes useful/necessary for survival in the Eurasian interior to form EHG and then at some much later point expanding back down south again.

.

Colin Welling

"I just can't mesh corded ware with the 3, simultaneous, events that took place in northwest europe."

Do they have to be simultaneous? The distributions of the west euro R1b clades look like dramatic founder effects. They might not have happened straight away but as a result of people trying to adapt to the acid soils along the Atlantic coast with the first few groups to figure it out having a population boom.

.

Alberto

"Why Moroccans (who lack ANE) do instead keep some good 15-20% WHG? And why do Basques, who are south Europeans, have North European levels of WHG, when coincidentally they have very low ANE? Similarly Spaniards have more WHG and less ANE than other south Europeans."

Perhaps WHG/EHG are actually more coastal/interior. This would still mean they were more west/east on average but may explain the quirks better.

If so then originally it may have been WHG in NW, SW and SE Europe and ANE in NE Europe simply because of the L shaped coastline.

.

Also more generally although there are three east-west routes into Europe created by the physical geography: 1) north of the Carpathians to the Baltic/North Sea, 2) along the Danube and 3) southern coast, the Atlantic running north-south could potentially act to combine all three so distinctions between northern Europe vs southern Europe might get blurred.

Krefter said...

Alberto,
"And why do Basques, who are south Europeans, have North European levels of WHG, when coincidentally they have very low ANE? Similarly Spaniards have more WHG and less ANE than other south Europeans"

The high WHG in Spain compared to SouthEast Europe is very easily explained: Middle Neolithic Farmer ancestry. Cardiel immigrants became about 25% La Brana-1 after 2,000 years of living there.

Alberto said...

@Krefter

Yes, the case of Spain is not too relevant for the hypothesis, since the difference in ANE and WHG are quite small anyway.

But it's the overall picture of ANE appearing with the mixture of WHG and SE, with the extreme example in Pathans (well, of MA1, but that was too long ago). If Pathans 60-70% Sintashta and mixed with South Asians that had no ANE, why from the mix ANE goes up and WHG disappears?

Why in the Near East WHG disappeared with ANE, but why it stayed in Moroccans, even though they got some 20% SSA admixture?

I see no way of explaining ANE peaking in the Hindu Kush unless:

- It's native from that area (by native I mean it's there since much before the Bronze Age), probably Mesolithic.
- It's a composite of 2 elements that mixed there, in this case WHG and SE.

The first case goes against Pathans descending from Sintashta. The second backs it up.

What is your opinion? Worth investigating it?

Simon_W said...

@ Kurti
Yes, BA Armenians resembled modern Northeast Caucasians, and even more so modern central North Caucasians. But that they spoke a Northeast Caucasian language is dubious. Because they are very close to the population that admixed with EHG on the steppe to produce Yamnaya people. And Northeast Caucasian doesn't share a lot of Near Eastern loanwords with PIE, secondly direct Northeast Caucasian influence on PIE is scant, and last but not least Northeast Caucasian is dominated by J1 and J2, which apparently were rare on the steppe, in contrast to R1b which is shared between BA Armenians and the steppe. And if linguists say that Northeast Caucasian came from the south, which would be in line with them being dominated by J1 and J2, it may have superimposed an R1b dominated population of unknown, possibly PIE language that was identical to the population that admixed with EHG on the steppe. And actually the fact that BA Armenians are closer to Northern Caucasians than to Southern Caucasians makes it unlikely that they came from the south like the Northeast Caucasian speakers. Of course it's possible that they were genetically minimally affected and just linguistically assimilated to Northeast Caucasian speakers by the Bronze Age.

Simon_W said...

Nirjhar,
Yes, but the PIE-Uralic relationship shouldn't be ignored either; it's too much to be coincidental. And it's about morphological/grammatical stuff which doesn't get borrowed quickly and easily, quite in contrast to cultural loan vocabulary. I think if PIE connections with SemItic and Sumerian are to prove a PIE homeland in West Asia, then we need to see at least PIE influence in these clearly native West Asian languages, too. According to Gamkrelidse and Ivanov such influence does exist, and if they're correct, that would be an important point. Maybe what David called a possibility was what really happened: That EHG people moved across the Caucasus to West Asia and later they expanded back to the steppe. The first movement may have been PPIE, the second PIE.

Nirjhar007 said...

Gioiello ,
Thank you for your observations!.
Simon,
Of Course there are Morphological/Grammatical resemblances with PIE and Semitic-Hamitic family etc ,read the book i named buddy:).

Aram said...

Gioiello
They are saying Franco-Cantabrian theory not likely.
Neverthless this issue of R1b-L51 is quite intriguing that we still didn't found it in East Europe.

===============
European Journal of Human Genetics advance online publication 17 June 2015; doi: 10.1038/ejhg.2015.114

New clues to the evolutionary history of the main European paternal lineage M269: dissection of the Y-SNP S116 in Atlantic Europe and Iberia

Laura Valverde et al.

Abstract

The dissection of S116 in more than 1500 individuals from Atlantic Europe and the Iberian Peninsula has provided important clues about the controversial evolutionary history of M269. First, the results do not point to an origin of M269 in the Franco–Cantabrian refuge, owing to the lack of sublineage diversity within M269, which supports the new theories proposing its origin in Eastern Europe. Second, S116 shows frequency peaks and spatial distribution that differ from those previously proposed, indicating an origin farther west, and it also shows a high frequency in the Atlantic coastline. Third, an outstanding frequency of the DF27 sublineage has been found in Iberia, with a restricted distribution pattern inside this peninsula and a frequency maximum in the area of the Franco–Cantabrian refuge. This entire panorama indicates an old arrival of M269 into Western Europe, because it has generated at least two episodes of expansion in the Franco–Cantabrian area. This study demonstrates the importance of continuing the dissection of the M269 lineage in different European populations because the discovery and study of new sublineages can adjust or even completely revise the theories about European peopling, as has been the case for the place of origin of M269.

Aram said...

Ok further in the abstract they say another thing that is quite confusing.
So where is the origin of R1b-L11?
///
Third, an outstanding frequency of the DF27 sublineage has been found in Iberia, with a restricted distribution pattern inside this peninsula and a frequency maximum in the area of the Franco–Cantabrian refuge. This entire panorama indicates an old arrival of M269 into Western Europe, because it has generated at least two episodes of expansion in the Franco–Cantabrian area. ///

Krefter said...

@Aram,

They should have tested SNPs not just STRs. They can't learn anything about the distant past of R1b-M269 with STRs. What do more knowledgeable people here think?

Gioiello said...

I have often broken many peer reviewd papers in pieces, and many said that I am temerarious, but these scholars don't understand anythging of out matter. Unfortunately the paper of Valverde et al. isn't for free, but I uploaded the supplements and saw an image: R-M269 to North like L11 to Iberia like P312 and expanded to the Isles as L21 and to Italy as U152. I wrote in a blog of some friend of mine on FB: An effort yet: more Westwards, meaning that the Refugium wasn't in the Balkans but in Italy. But some say that they are thinking to the Franco-Cantabrian Refugium yet... thus they are out of head. I never said similar nonsenses. You should know which is my thought, and I'm waiting that the aDNA solves any question. I have only said that whether the YFull dates remain these, the expansion happened from Iberia, but I continue to think that they are underestimated for 1000 or 1500 years, and Bell Beakers were 4900 years ago not only in Iberia but also in Tyrrhenian Italy and southern France. About what happened before I have expressed many times which is my thought.

Simon_W said...

@ Colin

You claim Central Europeans have more Corded Ware DNA than Northwest Europeans, while Northwest Europeans have more Yamnaya DNA than Central Europeans. But what's the difference between Corded Ware DNA and Yamnaya DNA? (On a side note: we're speaking here of differences between Central European Corded Ware and Samara Yamnaya, there was probably more variation in the vast Corded Ware complex and in Yamnaya.) Basically Corded Ware was Yamnaya + MN Farmers + some extra EHG. And Yamnaya was Yamnaya. So, if we neglect for once the finer detail of extra EHG in Corded Ware, they're simply Yamnaya + MN Farmers. But MN Farmers existed everywhere, also in Northwestern Europe. So in the end, it boils down to different levels of Yamnaya admixture. I suppose it would be very hard to accurately distinguish between Corded Ware ancestry and Yamnaya ancestry in modern Europeans, and quantify these separately. Haak et al. calculated either with EHG or with Yamnaya, but never with both in the same calculation. Hence the EHG vs. Yamnaya proportions cannot be used to infer Corded Ware vs. Yamnaya ancestry.

Therefore the next question is: Do Northwest Europeans have more Yamnaya ancestry than central Europeans? That's a tricky question, Haak's calculations have been called into question by some here; at least the Scots seem to have relatively strong Yamnaya ancestry, other Britons have less. Central European Slavs have been affected by Medieval Slavic migrations. (But ironically you seem to believe this had reduced the Yamnaya affinity). German speaking central Europeans were probably affected by Bronze Age migrations from Southeastern Europe and the Roman Empire. Norwegians are not Northwest Europeans, they are North Europeans, and Icelanders have considerable Norwegian ancestry.

But, as I also told you, some BRONZE AGE Central Europeans (like Halberstadt_LBA, which was even from the LATE Bronze Age) had plenty of Yamnaya ancestry, even more than the German Bell Beakers had. So they had more than enough Yamnaya ancestry to explain the highest Northwest European levels of Yamnaya ancestry. And of course, for that reason I agree that some yet unsampled Central European Bell Beaker groups may also have had such high levels. But all Hungarian Bronze Age samples had clearly less than the German Bell Beakers.
(Previous attempts to reconcile this finding with the hypothesis of a wave of Italo-Celtic R1b people expanding from the Yamnaya in the Carpathian Basin have explained the discrepancy to the Samara Yamnaya as resulting from the different origin of the local Yamnaya at the very western end of the steppe, where there may have been more WHG admixture. But if this explanation was correct, it would follow that calculations dealing with the levels of Samara Yamnaya ancestry in Northwestern Europe are overly simplistic and useless.)

Simon_W said...

continued, @ Colin

I'm not sure that the predominance of R1b in German Bell Beakers cannot have originated via drift from an R1b minority in the Corded Ware. A man can in theory have many sons with many women, who in turn can have many sons, so proportions can change quickly within some generations, especially in the Chalcolithic, when population densities were lower than now. Also why should quick autosomal change be at odds with this? Obviously if the replaced R1a men were autosomally the same as the successful R1b males, the yDNA change would make no autosomal difference. And you forget that according to all evidence, also in Hungary R1b was at most a minority.

And what weighs even more: There was R1b-U106 in late Battle Axe Sweden. R1b-U106 is the brother clade of South-Central/West European R1b-S116, a close relative. So at least R1b-U106 which is very common in peoples of Germanic tongue came from this minority in the Corded Ware. Why should the same thing be impossible for R1b-S116?

A Yamnaya migration north of the Carpathians that was distinct from the Corded Ware migration is pure fantasy.

And your explanation of the lack of Yamnaya legacy in Bronze Age Hungary as resulting from complete emigration of the Yamnaya population is rather implausible. For sure we'd expect the BA Hungarians to be considerably Yamnaya admixed if at one point Yamnaya was a huge mass of people in Hungary.

Yes, it looks like Italo-Celtic branched from PIE or early IE steppe with R1b Yamnaya. But this doesn't mean all R1b in central-western Europe comes from this same branch.

There's also a problem that gets consistently ignored: How do you explain why Basques, Aquitanians and northern Iberians are more R1b than their IE neighbours? If your scenario holds true they started with 0% R1b and the Celtoid IEs with I don't know, 90, 95%? And after millennia have passed, the said non-IE groups (some of them linguistically assimilated by the Romans) have more than 80% and their IE neighbours have 70 – 80% or less. How did this unlikely thing happen? It doesn't suffice to point out that the Scottish Highlanders and the Irish have similarly high incidence of R1b, because that still doesn't explain what strange process took place in southwestern Europe, and secondly we don't know for sure when exactly they became Celtic speaking.

Helgenes50 said...

David,

4) The Yamnaya bands that moved into the Balkans and Hungary were somewhat different from the eastern Yamnaya sampled to date; with more EEF/WHG and less EHG/ANE.

Do you think that is the case for this Hungarian ?
( RISE479 )


SteppeK10

28.80% Near_Eastern
0.00% East_Asian
0.00% Siberian
0.00% Oceanian
47.38% WHG-UHG
0.00% Sub-Saharan
0.00% Hindu_Kush
23.83% Steppe
0.00% Amerindian
0.00% Southeast_Asian

HarappA


S-Indian -
Baloch -
Caucasian 1.29%
NE-Euro 63.60%
SE-Asian -
Siberian -
NE-Asian -
Papuan -
American -
Beringian -
Mediterranean 34.03%
SW-Asian 0.97%
San -
E-African -
Pygmy -
W-African 0.10%




Simon_W said...

If RISE408 is really from the Iron Age, unlike all other Armenian samples, his slight Balkan shift in the PCA would favour the theory of a Balkan origin of Armenian.

a said...

K15's
F999943
# Population Percent
1 North_Sea 40.24
2 Atlantic 31.48
3 Baltic 16.78
4 Eastern_Euro 10.88

Using 4 populations approximation:
1 Swedish + Swedish + West_Scottish + West_Scottish @ 9.272451
------------------------------------------------------------------
F999946

# Population Percent
1 North_Sea 30.98
2 Eastern_Euro 28.80
3 West_Asian 21.88
4 Baltic 12.12
5 South_Asian 4.50
6 Amerindian 1.72

Using 4 populations approximation:
1 Chuvash + North_Swedish + Tabassaran + Tabassaran @ 18.696758
-------------------------------------------------------------------------
F999942

# Population Percent
1 Eastern_Euro 33.37
2 North_Sea 32.72
3 West_Asian 16.21
4 Baltic 13.39
5 Atlantic 3.36

Using 4 populations approximation:
1 Chuvash + Erzya + Tabassaran + West_Norwegian @ 19.084686
---------------------------------------------------------------
M951285

# Population Percent
1 Eastern_Euro 33.52
2 North_Sea 28.14
3 West_Asian 14.93
4 Baltic 11.27
5 Atlantic 4.42
6 South_Asian 4.06
7 Amerindian 3.66

Using 4 populations approximation:
1 Chuvash + Erzya + North_Swedish + Tabassaran @ 16.525211
----------------------------------------------------------------
F999944

# Population Percent
1 Baltic 32.41
2 Atlantic 25.80
3 North_Sea 18.85
4 West_Med 12.55
5 Eastern_Euro 10.40

Using 4 populations approximation:
1 French_Basque + Lithuanian + Lithuanian + Lithuanian @ 9.193623
--------------------------------------------------------------------
http://www.y-str.org/p/ancient-dna.html

capra internetensis said...

@Alberto

An interesting proposal, but I don't think it works. If I understand you correctly, you are suggesting that *independent* mixtures of WHG and South Asian element X are being detected as ANE. But in that case the mixes would not share drift, so they will not be close in the D stats.

If WHG came to India and mixed with X there to form ANE 1, and X went to Europe and mixed with WHG there to form ANE 2, then the distribution of WHG alleles and X alleles in their genomes will not be correlated. At any given locus it will be random whether ANE 1 and ANE 2 match or not. If, in contrast, they both descended from some ANE which was an ancient mix of WHG and X in a small population, drift would fix the same alleles at the same loci, so they would match at these loci, and D stats would pick up the relationship.

Unless I am missing something here, which is entirely possible.

Alberto said...

@capra

I haven't done any test (I don't have the means), so it's just based on the patterns I'm seeing from different tests. I couldn't say at this point how it would work exactly.

But the Sintashta-Pathan case seems to imply that ANE can indeed appear from independent WHG and South Asian X (as you called it), without coming from an ancient mixed population. In other words, ANE would just be a component which is something like 70% WHG - 30% X. Basically like West_Asian is something like 60% ENF - 40% ANE, or like the "blue" in the "Teal K9" component is about 82% WHG - 18% ANE.

I think that the Sintashta case and the fact that EHG never worked as admixed of WHG + MA1 suggest that there could be something in this (plus other evidence that I wrote above about Near Easterners, Dai admixture signals, etc...)

Should be easy to test and see if there something to it or not.

pnuadha said...

somehow i failed to post my response to grey. anyhow

@grey

Do they have to be simultaneous? The distributions of the west euro R1b clades look like dramatic founder effects. They might not have happened straight away but as a result of people trying to adapt to the acid soils along the Atlantic coast with the first few groups to figure it out having a population boom.


The population that dramatically changed the autosomal composition of northwest europe was large. This large population, bell beakers, were dominant in r1b. Therefore, a lot of r1b was brought to northwestern europe.

capra internetensis said...

@Alberto

What test would you suggest?

It seems to me that if the appearance of ANE in Europe in the Copper Age is due to the arrival of an element X that admixed with WHG there, then this "ANE" will be distinct from the "ANE" of Amerindians, and they should not share ANE-specific drift. But ANE-rich Northern Europeans do share drift with Amerindians, so the component is real and not a composite.

I would appreciate you (or anyone else) double checking my logic here.

Krefter said...

@Helgenes50 and a.

Looks like this guy had a lot of WHG. His ADMIXTURE scores that you guys posted suggest the same. I think ANE K7 may be an exaggeration though. He could be key to how WHG stayed high after Yamnya-like admixture.

RISE479, Bronze age Hungary, Vatya culture.

ANE K7.
WHG-UHG: 83.81
ANE: 10.8
ENF: 5.39

Grey said...

Simon_W

"There's also a problem that gets consistently ignored: How do you explain why Basques, Aquitanians and northern Iberians are more R1b than their IE neighbours?"

Low initial population density.

If you have a 100 people moving into a region which already has
- 900 people then they become 10%
- 300 people then they become 25%
- 10 people then they become 90%+

e.g. copper miners moving onto a lightly populated mountain, able to trade for food and thus able to support a larger population than the terrain previously allowed could easily become a local majority while a 100 of their cousins moving into a more densely populated valley below become a minority.

.

Colin Welling

"The population that dramatically changed the autosomal composition of northwest europe was large."

All the maps of LBK show a gap along the western edge that more or less lines up with the border of the Atlantic coast bio-region.

http://cdn.eupedia.com/images/content/LBK_culture.png

http://www.ceeweb.org/wp-content/uploads/2011/12/Natura2000EU28Bioregion.png

If the neolithic crops didn't grow well in that region - maybe due to acid soil - then the population in that region may not have been large and so any replacement population may not have needed to be large either.

pnuadha said...

@Simon_W

You claim Central Europeans have more Corded Ware DNA than Northwest Europeans, while Northwest Europeans have more Yamnaya DNA than Central Europeans. But what's the difference between Corded Ware DNA and Yamnaya DNA?

I mean to say that Central Europeans have more true Corded Ware ancestry as inferred by archeology. However, both Yamnaya estimates and EHG are lower in central europeans, which means that northwest europeans must have some, significant, non-CW source of yamnaya related heritage and EHG. Ydna also supports a migration, separate from CW, to northwest europe which was heavy in r1b and lacking in r1a. (I explained above why r1b in west europe was not due to a founder effect from CW. Whether you agree with that reasoning or not, you must admit that the ydna evidence strongly disfavors the CW as a source for r1b because of how dominant r1b is with the CW.)

It is pretty obvious to me that this alternative source of yamnaya like dna in northwest europe is yamnaya to bell beaker to northwest europe. This would better explain the high amount of r1b which clearly comes from yamnaya, who themselves appear to be r1b dominant.

Do Northwest Europeans have more Yamnaya ancestry than central Europeans?

According to Haak, they do have more. Yamnaya seems to be the ultimate source of yamnaya related ancestry and EHG in northwest europe. If yamnaya is indeed the primary source for both of these "components" in Northwest Europe, then Northwest Europe necessarily has more (samara like) yamnaya heritage.

But ironically you seem to believe this had reduced the Yamnaya affinity

I entertained other senarios that reduced the CW heritage in Central Europe while noting that the actual CW heritage in Central Europe would still be greater than the actual CW heritage in Northwest Europe. Hence the greater yamnaya heritage estimated for northwest europe requires a non CW source for yamnaya like dna. .

Norwegians are not Northwest Europeans, they are North Europeans, and Icelanders have considerable Norwegian ancestry.

uhhhhh, they are northwest european. And the point is that they have more r1b, less r1a, and more yamnaya related ancestry than Belarusians and Czechs. They further the point that northwest europe has a significant source of yamnaya related heritage apart from the CW.

A Yamnaya migration north of the Carpathians that was distinct from the Corded Ware migration is pure fantasy.

What is clear is that Northwest Europeans have less, actual, CW heritage but more yamnaya related heritage than Central Europeans. The yamnaya heritage came fast. Based on archeology and linguistics, the most likely chain is yamnaya to hungary, and hungarian bell beakers to northwest europe. But if said chain is wrong, ill still assume that there is another source of yamnaya like heritage in northwest europe which was not mediated by CW.

But all Hungarian Bronze Age samples had clearly less than the German Bell Beakers.

Its not that unexpected when you suspect a mass migration out of Hungary towards Northwest Europe. We have already seen a rather quick replacement of r1b on the steppe by r1a; something similar probably happened in the hungarian plane after eastern bell beakers moved west.

There's also a problem that gets consistently ignored: How do you explain why Basques, Aquitanians and northern Iberians are more R1b than their IE neighbors?

This has nothing to do with whether or not r1b came to northwestern europe by CW or another eastern migration. But ill just say that the bulk of r1b did in fact come to west europe by an eastern migration related to the yamnaya. If you disagree thats another discussion, but a very silly one at that.

pnuadha said...

i reposted above because of my atrocious grammar in the first post.

pnuadha said...

@grey

If the neolithic crops didn't grow well in that region - maybe due to acid soil - then the population in that region may not have been large and so any replacement population may not have needed to be large either.

large compared to what? you are just throwing out an idea with very little development.

Do you think the west bound yamnaya were so small in numbers that they're ydna was not statistically representative of their western yamnaya brothers? At the same time do you think the western yamnaya were so massive in number that a small group of yamnaya could settle central europe, and dramatically change the autosomal composition of metal age central europe? Do you think that the bell beakers were so small in number that they're ydna would not be statistically representative of CW even if they derived from CW? At the same time do you think the population of western/northwestern europe was so small compared to central europe that a small group of CW folk turned BB could dramatically transform the autosomal composition of northwest europe?

What senario are you actually arguing or are you just throwing disorganized would ifs and hoping it will make us listen to your metallurgy hypothesis. BTW, r1b yamnaya were not a minority of the yamnaya and they were not just miners. This study proved that!

Unknown said...

Gioiello

Nothing much happened in Italy during the Palaelithic and mesolithic. It didn't serve as a refugium for any part of europe apart from itself. It was a sparsely populated Region. In fact, it wasnt permanently settled by modern humans until 25 kya - some 20 thousand years after other areas like Bulgaria.

We're you aware of these simple facts?

pnuadha said...

Mike, can you give me a simplified version of where the large numbers of people were during ice age europe and how they spread out at the end of the ice age. How would you derive mesolithic europeans, across europe, from these refugiums. And just for fun, can you speculate on how it relates to EHG, WHG, and possibly EEF.

Finally, what is the relative population estimate for the balkans. I remember you saying that during the mesolithic greece and bulgaria were barely populated and that contemporary turkey was well populated. What about the rest of the balkans? What about the adriattic coast of the balkans? I want to speculate the possibility of EEF being native to southeast europe, for entertainment.

Alberto said...

@Capra

I think you give too much credit to ADMIXTURE. I don't think it's too sensitive to drift unless you specifically design something to find it. At normal resolutions it won't matter much, I think.

Anyway, I think that if qpAdm worked good in the Sintashta-Pathan case, it could be used to try EHG as a mix of WHG + South Asian (Dai, Malay, Sakilli,...) and see what it says. Or just some f3 stats with those populations could do.

Unknown said...

Colin

Ill email you.

Grey said...

Colin Welling

I'm not talking about central Europe. I'm talking about the gap between the western range of LBK and the Atlantic. It seemed very strange to me the first time I saw a map of LBK and yet no one seems to think it needs explaining.

Unknown said...

Anyone seen the R1a/b paper at biorxiv?

Unknown said...

No; what's title ?

There also "New clues to the evolutionary history of the main European paternal lineage M269: dissection of the Y-SNP S116 in Atlantic Europe and Iberia".

Gioiello said...


@ Mike Thomas, I don't know who you are, if truly whom the person we cannot name says or not. Of course I don't believe her, because I am pretty sure who Davidski and Nirjhar007 are and not who she says.
1) First of all my Refugium isn't of 25000 years ago: I have always spoken of the Younger Dryas onwards, and specifically my theory concerns the people that from Italy migrated 7500 years ago from Italy to Iberia as Zilhao demonstrated, and I have always said that I am pretty sure about what happened after that, but I didn't exclude that those people may have come from the Balkans or Asia Minor. Anyway I think that they were autochthonous, but if it will be demonstrated from aDNA that they came from the Balkans I'd agree with that. We were in the Cardial period. From Cardial descended all the Western Mediterranean cultures till the Bell Beakers.
2) What you say isn't true. The first Homo Sapiens Sapiens in Europe has been found in Italy about 45000 years ago. A man of more than 40000 years ago has been tested and resulted mt R. We have Paglicci 24000 years ago. We have Mesolithic mt HV in Sicily etc etc thus I take your words as the same prejudice against Italy which doesn't know pretty anything about prehistory in Italy.
3) Anyway the truth is demonstrated from the proofs, and I simply ask that Rinaldone culture, Arene Candide etc are tested for the aDNA. I have written a lot about the likely presence of many mt and y in Italy from many thousands of years, comprised my Y (R1b1a2-Z2110) and mt (K1a1b1e).

Krefter said...

Another R1a1a1b2a2. This time from Karsuk. Also, another one of the BBC R1bs is confirmed to be P312.

https://genetiker.wordpress.com/2015/06/11/y-haplogroups-for-prehistoric-eurasian-genomes/

Unknown said...

Gioiello

Please re-read what I said:

" it wasnt permanently settled by modern humans until 25 kya - some 20 thousand years".
There was a populatin depletion between the earliest, abortive human colonisations , and 25 kya, from which point there was permanent settlement. And, even in the mesolithic, parts of italy had only sparse settlements.

Any these are not my opinions, but those of Italian archaeologists. But I guess your entitled to your theorem until the burden of evidence proves or disproves it.

And my only "bias" against Italy is admiration.

Gioiello said...

@ Mike Thomas, the problem isn't to have biases, but to search the truth. Who thinks that my theories are due to the fact that I am Italian is wrong. My theories may be right or wrong, but are based upon indictions considered with a scientific attitude... to make hypothesis for doing experiments. Now we have the aDNA, which anyway may always be interpreted. The enthusiasm of the kurganists for the first R1b found in Yamnaya may be ingenue if one doesn't know the phylogeny of the haplogroup and "R1b phylogeny" was a thread of mine on Anthrogenica, which smal is following, but with some shyness: why he doesn't publish a tree also for R1b1, about that I have written tons of letters in these years?
And I don't like Italy even though I am Italian. I said that I am a litterate before being a genetists, and many years ago I wrote
Vai in culo, paese di merda,
affonda in tutto il tuo marciume.
Who bans me dislikes my language, and in italian I am worst than in English, but who looks at the last vicissitudes in Italy may say that my words aren't true?

Aram said...

Gioiello

What is Your opinion about the origin of R1b-Z2103 branch.
As far as I understand You, it must come from Balkans?
And what about Yamna?

Unknown said...

Gioillo

Please present your case simply and clearly- which you seem to be avoiding. Don't talk about 7 years ago about some thread I don't even read.
It's easy to do: state your understanding of phylogeny and illustrate your corroborative archaeological evidence.

Gioiello said...

As I said, I based my theories about the Y and the mt, not about the autosome, because I think it is misleading (for many reason I cannot resume here).
About R1b.
1) Italy has the highest variance of R1b1-L389+ (Of course you haven't to consider the "R1b1 FTDNA Project": they support the wrong idea that R1b was born in Middle East for justifying the numerous Jewish R1b, very likely introgressed from Europe or the Caucasus. I said that only two subclades seems old Jewish: an R-L277 and an R-PF7562, but I have found a person of Greek descent who could be the ancestor of this recent unique Jewish cluster: he is the father of Anne Hart)
2)R-V88 seems oldest in Sardinia/Italy or also Iberia
3) All the oldest subclades of R-M269 are in Western Europe, only a tiny subclade of R-M269-PPF7563- has been found in Turkey, but we don't know its oldest origin
4)R-L23* doesn't exist of course, but the subclades R-Z2103/Z2105 are diffused from East to West, but in Yamnaya and the Caucasus we find only the oldest ones, till Z2109*. Z2110 (mine) is above all European, and its subclade CTS7556 is Western European and ancestor of all the Eastern European CTS9219, which came thus from Western Europe.
5) L51 is fundamental Italian in its oldest subclades and my hypothesis is that it is older than 6400 years (see the YFull tree) and was amongst the "Italian" agriculturalists who migrated to Iberia 7500 years ago.
6) The expansion of the subclades happened with Bell Beakers, very likely from Iberia, but also Tyrrhenian Italy and south France may be at the origin of some of them, perhaps France for L21 and Italy for U152.
7) U106 is older, linked to the German world and may have come from L11 in the Western European region, in fact its brother clade R-S1200 is diffused above all in the Isles.
Something happened in the Bell Beakers world which will be clarified from aDNA.

Gioiello said...

As I said, I based my theories about the Y and the mt, not about the autosome, because I think it is misleading (for many reason I cannot resume here).
About R1b.
1) Italy has the highest variance of R1b1-L389+ (Of course you haven't to consider the "R1b1 FTDNA Project": they support the wrong idea that R1b was born in Middle East for justifying the numerous Jewish R1b, very likely introgressed from Europe or the Caucasus. I said that only two subclades seems old Jewish: an R-L277 and an R-PF7562, but I have found a person of Greek descent who could be the ancestor of this recent unique Jewish cluster: he is the father of Anne Hart)
2)R-V88 seems oldest in Sardinia/Italy or also Iberia
3) All the oldest subclades of R-M269 are in Western Europe, only a tiny subclade of R-M269-PPF7563- has been found in Turkey, but we don't know its oldest origin
4)R-L23* doesn't exist of course, but the subclades R-Z2103/Z2105 are diffused from East to West, but in Yamnaya and the Caucasus we find only the oldest ones, till Z2109*. Z2110 (mine) is above all European, and its subclade CTS7556 is Western European and ancestor of all the Eastern European CTS9219, which came thus from Western Europe.
5) L51 is fundamental Italian in its oldest subclades and my hypothesis is that it is older than 6400 years (see the YFull tree) and was amongst the "Italian" agriculturalists who migrated to Iberia 7500 years ago.
6) The expansion of the subclades happened with Bell Beakers, very likely from Iberia, but also Tyrrhenian Italy and south France may be at the origin of some of them, perhaps France for L21 and Italy for U152.
7) U106 is older, linked to the German world and may have come from L11 in the Western European region, in fact its brother clade R-S1200 is diffused above all in the Isles.
Something happened in the Bell Beakers world which will be clarified from aDNA.

Srkz said...

Some IBD maps
https://www.dropbox.com/sh/36ldhba817klryo/AADvlILVMQ2FsGM7Qxe6wGV5a?dl=0

(Only RISE98 have enough diploid SNPs to create a good map)

Helgenes50 said...

@ Srkz

Some IBD maps
https://www.dropbox.com/sh/36ldhba817klryo/AADvlILVMQ2FsGM7Qxe6wGV5a?dl=0

Thanks,

On this map, in France, we can see differences
You certainly have regional data.

Srkz said...

No i'm using HGDP French only now. It's just a painting algorithm - Basque affects South-West, Cornish - North-West etc.

Unknown said...

It won't let me paste the link from my phone. L23 Mrca is 3700BCE. L51 can only be from Eastern Europe.

Helgenes50 said...

@Srkz

OK! I understand
A French dataset would be too beautiful

Or Cornish is maybe affected by NW France ?

Nirjhar007 said...

Chad,Krefter
you guys are kidding right?

Krefter said...

@Nirj,

I misread Mrca as being a location. My bad.

Gioiello said...

@ Krefter, they aren't kidding. They don't know the last YFull tree, which I believe underestimated for at least 1000/1500 years, and we'll se if I will right or wrong:

R-L23 L478/PF6403 * L23/S141/PF6534 formed 6400 ybp, TMRCA 6100 ybp
R-L23*
R-Z2110 S12460 * Z2110/CTS7822 * S17864 formed 6100 ybp, TMRCA 6100 ybp
id:YF02873 ITA [IT-FI]: Gioiello Tognoni
R-L51 L51/M412/S167/PF6536 * PF6414 * FGC39/CTS10373/PF6537... 1 SNP sformed 6100 ybp, TMRCA 5600 ybp
5600ybp is the MRCA of the samples found so far, but it was born 6100 ya, but also MJost calculated some thousands of years more.

capra internetensis said...

@Alberto

I don't really know how ADMIXTURE works. I was thinking of D statistics, which are a lot more transparent.

Dai is only marginally closer to ANE than to Loschbour:
Loschbour, MA1; Dai, Chimp -0.0074 -1.11

Unfortunately I don't have the other D stats for Dai, so here is Japanese, showing that WHG is much closer to ANE than ENA is:
Loschbour, MA1; Japanese, Chimp -0.0048 -0.715
Chimp, MA1; Japanese, Loschbour 0.0454 7.324

Of course Japanese is not Dai, but Japanese is actually closer to MA1 than Onge is, so I doubt the difference is meaningful:
Chimp, MA1; Japanese, Onge -0.005 -1.148

Some f4 stats -with Stuttgart as outgroup, all I could find:
Kharia, Chimp; Loschbour, Stuttgart 0.001154 2.564
Kharia, Chimp; MA1, Stuttgart 0.001447 2.563
Onge, Chimp; Loschbour, Stuttgart 0.00191 3.452
Onge, Chimp; MA1, Stuttgart 0.001842 2.987

Almost perfect symmetry of MA1 and Loschbour in their relation to South Asian tribals.

Krefter said...

DF27 in German Bell Beaker according to Richard Rocca. So, U152 and DF27 were around pretty early. This supports the idea P312 spread very early on and before the Celts or maybe from proto-Italo Celts or whatever.

"Genetiker found that the RISE560 Bell Beaker sample is M12124+, which is below DF27 as per Underhill 2015.

Source: https://genetiker.wordpress.com/2015...asian-genomes/"

Chad said...

Here's the abstract..

http://biorxiv.org/content/early/2015/06/15/020933

"We establish accuracy by a comparison across a wide range of time, indeed this is only y-clock giving consistent results for 500-15,000 ybp. In particular we show that the dominant European y-haplotypes R1a1a & R1b1a2, expand from c3700BC, not reaching Anatolia before c3300BC. This contradicts current clocks dating R1b1a2 to either the Neolithic Near East or Paleo-Europe. However our dates match R1a1a & R1b1a2 found in Yamnaya cemetaries of c3300BC by Svante Paabo et al, together proving R1a1a & R1b1a2 originates in the Russian Steppes."

Alberto said...

@Capra

The problem with MA1 is that it's too old and no population is really close to it. With EHG that is only 38% ANE but much more modern:

Loschbour EHG Dai Chimp -0.0137 -2.506 341763
Loschbour EHG Thai Chimp -0.0137 -2.614 341763
Loschbour EHG Kinh Chimp -0.0146 -2.722 341763

Significant enough, though Dai are actually from China (and the others from SE Asia) and have a good amount of East Asian. I don't have the stats with Austro-Asiatic or Sakilli to see if they're more significant or not.

In any case, this is obviously a long shot. But I think there are some hints to at least look into it in some more detail.

With Admixture the runs I've checked have Amerindians in them, so they are useless to know the origins or possible admixture of EHGs.

Alberto said...

For completeness, and to check is has nothing to do with Loschbour being more "basal" or something:

Loschbour EHG Yoruba Chimp -0.0001 -0.013 341763
Loschbour EHG Papuan Chimp -0.0037 -0.65 341763
Loschbour EHG Australian Chimp -0.004 -0.645 341762

These are insignificant. So the previous populations do have some specific affinity to ANE.

Krefter said...

Anyone know what's up here. Was DNA just taken from Neolithic Near East?

"Optimal Ancient DNA Yields from the Inner Ear Part of the Human Petrous Bone"

http://journals.plos.org/plosone/art...0129102#sec001

Unknown said...

Did anyone pick up why this study didn't use EHGs from russia in their modelling ?

Unknown said...

Krefter,
There's probably lots of genomes coming.

Unknown said...

Mike, they hadn't been published yet.

pnuadha said...

@Chad

The paper you just quoted, used the following as a reference for the claim that r1a and r1b were found in yamnaya.

S. P ̈a ̈abo et al (2014),Ancient human genomes suggest three ancestral populations for
present-day Europeans, Nature 513, pp 409-413.

but that finding wasn't in Paabo's 2014 paper. The author either made a mistake or maybe he is aware of some unpublished findings from Paabo.

Davidski said...

Alberto,

This is probably just barely picking up ANE affinity among the South Indians.

WHG EHG Pulliyar Chimp -0.0135 -2.347
WHG EHG Sakilli Chimp -0.0127 -2.209

WHG EHG Pulliyar Mbuti -0.01 -2.337
WHG EHG Sakilli Mbuti -0.0092 -2.259

In any case, EHG can't be modeled as WHG/South Indian or WHG/Dai ( which is along with the Onge the closest pop we have to ASI).

WHG/Sakilli is infeasible.

EHG as 100% WHG gives a chisq of 186.573. EHG as 100% Sakilli gives a chisq of 244.052. These are very high and thus unlikely.

EHG as WHG/Dai gives ancestry coeffs of 0.939/0.061 but a chisq of 182.691. Again, too high.

Unknown said...

Colin

Yes I noted that too- about the Pabbo paper quote.
The papers claims are interesting, but it'll take some coming to terms with the mathematics.

Davidski said...

EHG as WHG/Dai using more markers.

Coeffs 0.961/0.039 chisq 209.611.

Highly unlikely.

pnuadha said...

@mike

Do you have a background in math, or decent practice with it by reading up on papers that apply math?

Does anybody know why gene ticker has been able to draw out so much more detail about ydna than the authors of the paper? Is it because there is too much uncertainty with the given data if they try to be that precise?

capra internetensis said...

@David

Thanks for running those stats.

@Alberto

EHG cultures also have archaeological connections with Northeast Asia, so minor ENA input is plausible for them, distinct from any original ENA affinity in ANE.

Nirjhar007 said...

David, will you do any hg determination or we can take Genetiker's listings as more or less accurate?>:)

Nirjhar007 said...

Krefter, the article say:
Ten petrous bones were selected from archaeological specimens, representing a wide range of geographical locations and climatic contexts (Table 1, for repository information see S1 File). The specimens were selected from Central Europe, Central Asia, Southeast Asia, the Levant, Anatolia, and North Africa. The specimens are from Holocene archaeological contexts dated to between 10,000–1,800 calibrated years before present (cal. BP). The samples from Nubia, Jordan and Turkmenistan are from hot and arid regions. The sample from Turkey is from the Eastern Mediterranean (northwestern Turkey); the samples from Hungary and Serbia are from the Carpathian Basin/Southeast Europe, while the two samples from Cambodia and Vietnam are from tropical/subtropical Southeast Asia. We also included a metatarsal bone for one Neolithic individual from Hungary (Polgár Ferenci hát, PF280-443) as a control to confirm the differences between petrous and non-petrous reported in the previous study...
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0129102#pone-0129102-t001
They also took Samples from Parkhai it will be great though if we find any aDNA data from Parkhai, some have suggested the customs of Catacomb Culture has origins from that culture in Turkmenistan...

Davidski said...

You can take this list as more or less accurate.

https://drive.google.com/file/d/0B9o3EYTdM8lQaU54RWc0eEV2WUU/view?usp=sharing

I don't follow Genetiker's work.

Nirjhar007 said...

David, With all due respect we need the sub-mutations of the clades and it appears except him no one for some reason is giving much importance!

Davidski said...

Sure we do, but why do you think Allentoft et al. gave so little info about the Y-chromosomes?

Probably because there's another study on the way focusing on this.

Nirjhar007 said...

Which Study?:).

Gioiello said...

@ Nirjhar0007

The study David was referring to is only supposed, but I'd want to do other explanations:
1) a deep test of those samples didn't agree with their agenda
2) I know many illustrious geneticists who don't know YFull work and are acting with Geno 2.0 yet when I had last year my Full Genome.

Chose 1 or 2 as you prefer.

Nirjhar007 said...

Gioiello, do me a favor would you can you please run the analysis of the Sintashta and Karasuk genome to get their Y-DNA more accurately as Genetiker is doing?:).

Alberto said...

@David

Thanks for looking at it. It certainly doesn't look like Dai or any modern South Asian population contributed any significant amount to EHGs.

Do you have any hypothesis already about why those great results with qpAdm for Pathans as 70% Sintashta might be happening and why do they disagree with the previous models we know? And basically for the lack of WHG ancestry in S-C Asia if indeed the population has European origin?

Gioiello said...

@Nirjhar007

Unfortunately I have only a little lap top, very slow now, and ever was able to upload the program for doing that. When I was on Anthrogenica I used smal's analyses, whom I semt my data to, and now I use the YFull browser which works very well also on my laptop. Through it I did my analysis on the Sardinian SNPs from Francalacci and I wrote a lot about hg R-V88 but also on hg.G on FB blogs of Asciak. I think that Genetiker has done a good analysis, as smal is doing well his. I know that I should get all these programs and a more powerful PC, and all the knowledge for that... but I cannot do everything. In a free communinity we are protected from the fact that if anyone is wrong he may be corrected from others.

Balaji said...

Some people seem to be embracing the modeling of Pathans as 71% Sintashta, 16% Georgian and 13% Dai from qpAdm. With this new qpAdm tool, they want to throw out “out-dated” concepts such as WHG, ANE and BEA. But this will be a mistake. There is actual physical material from three WHG individuals, Loschbour, La Brana and Hungary-Gamba_HG. Similarly, there is physical material from MA1 and AG1 to represent ANE. BEA is a reasonable inference from the f4 statistics involving EEF.

We can do a sanity check of the modeling of Pathans as something like 71% East European by looking at f3(Mbuti;Loschbour,Pop). These were calculated by Davidski about a year ago. Here they are.

https://drive.google.com/file/d/0B-aM8HL5oc9WSm1KcEItbjZCa0k/view?usp=sharing

This is a good measure of “Europeanness”. Lithuanians top the list and Europeans constitute the top 38 of the list with South-Italian being number 38. Next come Near Easterners, Central Asians and Northeast Asians. Pathans are number 71 after Georgians, Lebanese_Christians, Kets and Druze. Here is a picture of people who are more European than Pathans.

https://en.wikipedia.org/wiki/Ket_people#/media/File:Ket_women_and_children_1914.png

We must take this qpAdm model with a huge grain of salt. It is possible that the model fits well because instead of gene flow from Sintashta to South Asia, there was gene flow from South Asia to Sintashta.

Unknown said...

FRom this data it is probable that R1b is coming up from Iberia.

I may also exist in unsampled regions. But it is NOT coming in with Corded Ware who are overwhelingly R1a.

Davidski said...

For a long time I believed that R1b moved up from Iberia into Central Europe, but could never make it stick. Now I know why.

Most of the R1b in Northwestern Europe could well be from Corded Ware, even if Corded ware were mostly R1a, due to a massive founder effect.

Keep in mind also that we only have Corded Ware samples from Germany, Poland, Scandinavia and Estonia at this stage, and R1a is common in all of these places. Corded Ware Single Grave samples from Holland might be mostly R1b.

Unknown said...

Dave
Ex Netherlands is possible. I leant toward the upper danube region, but the low countries for disseminating BB styles has always featured. But recently, perhaps unjustly, it has been overshadowed by the of out of Iberia concensus.

Simon_W said...

@ Colin

Well, archeology might suggest that Central Europeans have more Corded Ware ancestry than Northwest Europeans. But this isn't very conclusive reasoning. With the same logic you could argue that South Russians and Ukrainians have more Yamnaya ancestry than anyone else.

And actually the question if modern Central Europeans have lower Yamnaya/EHG estimates than Northwest Europeans is completely irrelevant. Because, for the third time: Fact is, some Bronze Age Central European groups had enough Yamnaya/EHG to explain even the highest Yamnaya/EHG in Northwestern Europe. And modern Northwest Europeans are not descended from modern Central Europeans, but from ancient ones.

I don't rule out that R1b-P312 is from the Yamnaya in Hungary. But this has yet to be verified by aDNA. Also it's noteworthy, like I elaborated a couple of posts ago, that it isn't typical for the ancient Italics in Italy, who probably branched off from the Danubian Yamnaya. So apparently it wasn't very common there and got frequent in central Europe because of a founder effect. The map of R1b-P312/S116 in Myres et al. 2011 indeed shows that R1b-P312 is very rare in southeastern Europe. So one way or the other we can't get around founder effects. And judging from the latest evidence, R1b-U106 is most probably from the Corded Ware.

Furthermore according to all autosomal evidence, the Yamnaya input in Hungary seems to have been quite strongly WHG and less EHG/Armenian than the Samara Yamnaya. This might be a useful explanation for the dilution of EHG ancestry in modern Europeans, but it doesn't make the Hungarian Yamnaya a good candidate for explaining strong Samara Yamnaya-like ancestry. Even if the Yamnaya-descended people emigrated completely, I would expect to see some EEF type people showing a minor Samara Yamnaya admixture signal. But in fact, what follows are not really EEF people with weak Samara Yamnaya signal, but people with rather strong WHG and lesser EHG, and this is best explained by a different kind of Yamnaya input.

Imho archeology suggests that the first wave of Yamnaya heritage reached Britain with Bell Beaker people from the Bell Beaker-Corded Ware fusion area on the lower Rhine. Later on other waves followed in the Bronze Age proper and in the Iron Age. (To us Central Europeans Bell Beakers, Corded Ware and Yamnaya are not Bronze Age, but that's just a terminological issue.) I'm not sure which one was the most important as regards autosomal input, it may well have been the first one.

OK, I agree that the Basque/Aquitanian/Iberian problem is unrelated with the questions just discussed, so I leave that for another post.

And by the way, the dude you call „gene ticker“ calls himself Genetiker, that's the German word for geneticist. He isn't really a geneticist, rather a crackpot, but I appreciate him for trying to analyze the y-chromosomes.

Simon_W said...

It's by the way great that Felix has returned and is back uploading the kits to GEDmatch!

Simon_W said...

RISE71 had been uploaded longer ago by Sergei, it's kit M671253. She lived around 2100 BC in Denmark. The Eurogenes K15 oracle suggests she's most similar to 67% West_Norwegian + 33% Orcadian, a rather unexpected result for someone from Denmark, but then again it's very similar to Hinxton2's results! I have no idea where Falshöj is, the place where she was dug out. Based on this result, I would guess in Jutland.

Simon_W said...

The link http://jpst.it/zEfR shows a comparison between the frequency of R1b-DF27 and the pre-Roman languages of the Iberian peninsula. It really jumps into the eye that R1b-DF27 is associated with non-IE languages. In the Celtiberian and Lusitanian areas it's clearly less common. The only deviation from the clear pattern being an area in Castilia-La Mancha, this might be because of a stronger non-IE substrate there. The most likely conclusion imho is that R1b-DF27 reached Iberia before IE languages did. According to Genetiker R1b-DF27 was found in German Bell Beakers. This suggests that it spread from there to Iberia, but lost the IE language somewhere on the way.

R1b-L21 might be a similar case. However, the Yamnaya-related autosomal admixture is much stronger where it predominates. So it's more likely to have been associated with an IE language, though this needn't have been Celtic. And it's at least curious that it's also quite common in Basques.

pnuadha said...

@David

For a long time I believed that R1b moved up from Iberia into Central Europe, but could never make it stick. Now I know why.

Most of the R1b in Northwestern Europe could well be from Corded Ware, even if Corded ware were mostly R1a, due to a massive founder effect.


You have retailored your original hypothesis just enough to not sound delusional but its mostly the same story you're putting foreword. That is, Bell beakers formed from a mix of neolithics and Corded Ware, in which the later donates language, steppe like autosomal dna, and NOW r1b.

But does it really stick when you say that r1b was eastern yamnaya but only represented a minority of western yamnaya and their descendants the CW folk, but that this minority r1b lineage flourished like crazy when it got to central europe. This hypothesis looks tailor made.

Seinundzeit said...

Balaji,

That isn't how Haak et al. modelling (as implemented in qpAdm) works.

Also, one can easily verify South Central Asian admixture in Sintashta. David could try to model Sintashta as mixtures of Pashtuns + something else, and see if the model is as stunningly good as the model with Pashtuns at 60%-70% Sintashta.

That should settle it.

pnuadha said...

@simon_w

Well, archeology might suggest that Central Europeans have more Corded Ware ancestry than Northwest Europeans. But this isn't very conclusive reasoning. With the same logic you could argue that South Russians and Ukrainians have more Yamnaya ancestry than anyone else.

The two are completely different. And when I said archeology I wasn't just saying that corded ware was located in central europe, I was actually using a range of archeology over time to estimate whether or not central europe would retain more true CW heritage. The steppe is highly unstable and the yamnaya probably left en masse before their land was taken by others. Thats not the same for the CW region. Also, for me to be wrong the people diluting true CW heritage in Belarus, so that they had less true CW heritage than NW Europe, would need to have had less EHG than modern northwest europeans.

Do you think northwest europeans have more true CW heritage and the non CW heritage in belarus has a smaller ratio of EHG than that in modern northwest europeans?

Also it's noteworthy, like I elaborated a couple of posts ago, that it isn't typical for the ancient Italics in Italy, who probably branched off from the Danubian Yamnaya.

not really. i don't know if the italians studies spoke IE or if they had much heritage from hungarian yamnaya.

Furthermore according to all autosomal evidence, the Yamnaya input in Hungary seems to have been quite strongly WHG and less EHG/Armenian than the Samara Yamnaya.

Ill look into that and see if there are any fitting problems for (samara) yamnaya in the bahungarians who, i agree, had some of the hungarian yamnaya in them. On the surface I can't see any problems because we don't know how much yamnaya the ba hungarians had. You can vary yamnaya admixture in yamnaya over space or you could vary WHG in middle neolithic people over time and space...

Imho archeology suggests that the first wave of Yamnaya heritage reached Britain with Bell Beaker people from the Bell Beaker-Corded Ware fusion area on the lower Rhine.

Maybe there were other migrations as you say but they need to have high yamnaya and probably high r1b. I don't think we can expect that senario to exist for very long in central europe.

That said, the population of bell beakers was massive. How could they all get their ydna from a very recent founder effect of CW people? The time between r1b being a minority in CW and r1b being a majority of the large number of Bell beakers is so small.

Unknown said...

Colin/ Dave/ Simon

To me, what appears to fall into place is the Neolithic drop which we're all aware of, and the shift toward a predominantly pastoralist economy in much of western Europe, incl Britain during the period 3000 - 1800 BC (roughly). Did this coincide with the rise in dominance of L23 in WE ? Possibly.

I remain open, at present, on the exact route and chronology of it arriving west of the upper Elbe.

Chad said...

2600-1800BCE, Beakers, sure. I wouldn't call it primarily pastoral though. They were rather sedentary with lots of livestock. Expansion is probably more due to large population numbers, rather than living out of wagons. I still think there will be a 600 year lag between Germany and Iberia for R1b.

Unknown said...

Chad
Pastoralism doesn't necessarily imply nomadizing. There are different types of pastoralism, and true "nomadic pastoralism" arose in late Bronze Age.

PF said...

@Krefter

Unfortunately, no. I'm beginning to lose hope.

"Five of the samples had maximum endogenous yields >35% (35.36%- 69.63%) for part C while the other five samples had average yields <1% for all parts. All the samples with low yields are those from tropical/subtropical (i.e. hot) regions. All five samples from temperate regions yielded high percentages of human DNA reads, at least for part C (Table 1)."

Though, the authors do end on a somewhat optimistic note.

"However, the fact that Man Bac and Ain Ghazal samples, which are both from regions with environmental conditions which are unfavorable to aDNA preservation, yielded endogenous human DNA (although in low percentages), is encouraging as it suggest that in the future, analyses of petrous bones from regions such as sub-Saharan Africa, the Middle East and Southeast Asia, may provide sufficient nuclear DNA data to carry out genomic-scale analyses, although currently still at high costs. Further experiments involving additional samples and deeper sequencing will be necessary to investigate the potential of petrous bone sampling from hot regions, but given the great interest in human ancient DNA from these regions, such studies are well-warranted."


Davidski said...

The methods used by the Pinhasi team are very different from the methods used by the Reich lab.

Capturing SNPs in batches of samples like the Reich people are doing appears to be a relatively cheap and effective method of getting enough data to analyze with ADMIXTOOLS, which is really what is needed. The fact that the individual sequences are usually worse doesn't matter in the scheme of things.

So what you're reading in that article is only relevant to what the Pinhasi team are doing. In fact, the second hand info I'm getting is that the Reich lab has a lot of new samples, including from the Near East. I don't know if they have any from Central Asia, but I wouldn't be surprised if they did.

Nirjhar007 said...

Good To Know:).

Grey said...

It's worth thinking about what happened in the American version of this.

There was a mass migration starting in the east that gradually rolled west but at every stage there were people running ahead of the process: fur tappers, traders, prospectors, mostly male many of whom ended up marrying local women.

Normally - as in America - you'd expect something like this would eventually be swallowed up by the advancing wave so the question is could there have been strong enough founder effects along the Atlantic coast for these steppic runaheads to survive the advancing wave?

(autosomally if not culturally)

(as hinted at by the distribution of *some* of the R1b clades which seem to be centered on the Atlantic coast)

Grey said...

"FRom this data it is probable that R1b is coming up from Iberia."

Maybe some clades by sea to Iberia and then northwards from there but some by other routes e.g. via Rhine to Holland then expanding from there north and south.

Gioiello said...

@ Grey
"Maybe some clades by sea to Iberia and then northwards from there but some by other routes e.g. via Rhine to Holland then expanding from there north and south":
I'd agree completely with you if the point of departure were Italy. Of course that has to be proved or disproved from other data (and I always left open the possibility that R1b may have come to Italy from elsewhere, even though the pathway from R1b1-L389* is all in Italy... but of course also the oldest subclades may have journeyed).

Simon_W said...

@ Mike, Chad

About three years ago there was a paper on Dienekes' blog which dated this pastoral stage in Britain more exactly. The decline in farming started already before 3000 BC, perhaps 3300 BC and from 3000 BC on it was already fully down. And this lasted till the Middle Bronze Age, which started perhaps 1500, 1600 BC. One thing is crystal clear: The start of this pastoral stage has nothing to do with Bell Beakers, it's way too early! In central Europe, there are no Bell Beakers before roughly 2500 BC, and that's where the Beaker = Yamnaya supporters place their origin, don't they? In fact, it's even earlier than the arrival of the Corded Ware in central Europe. It's more or less synchronous with the Globular Amphora migration. But this didn't reach Britain, a far as we know.

http://dienekes.blogspot.com/2012/10/the-winding-road-to-agriculture.html

I'd say the end of this phase is more likely to have been coupled with the arrival of new people from the continent.

Simon_W said...

@ Colin

I think your whole argument is flawed. What the ancient DNA evidence has shown is this: German Corded Ware had lots of Yamnaya ancestry, much more than modern Northwest Europeans. And all other central European LN/BA cultures had much less Yamnaya ancestry than the Corded Ware. But from central European Bell Beaker you want to derive modern Northwest Europeans. Maybe you don't want to, but the Hungarian ones* cannot have jumped over central Europe. The German Bell Beakers from Haak would barely suffice, if there was near complete population replacement in Britain. So is there a snag in it? No, because on the one hand there may have been variation in Bell Beaker groups, with some having more Yamnaya ancestry, probably from Corded Ware wives, and on the other hand later continental Bronze Age groups may also have contributed something, and a group like Halberstadt_LBA would do very well. There's no point in taking modern central Europeans into account and estimates about degrees of true Corded Ware ancestry based on archeological links.

Regarding the WHG in BA Hungary, it's a really striking change that occured from Copper Age CO1, 2800 BC, to Bronze Age BR1, 2000 BC. See the Eurogenes K15 analysis here:
https://www.youtube.com/watch?v=w2CkLqm1dwM
And coincidentally BR1 lived in precisely the area where the Hungarian Yamnaya had been. And there is also some ANE in BR1, so a pure WHG refugium somewhere in the mountains wouldn't work, just like local HG KO1 doesn't work either, because his ANE is too low.

As for the Italics, like I said, the Italic tribes are strongly associated with the southeastern yDNA pole of PC1, as this map shows (Italics in yellow): http://jpst.it/zA5J
Also they are associated with being ardent inhumation adherers. After the Protovillanovan had introduced cremation to central and southern Italy, they soon switched back to inhumation: http://jpst.it/zA5m
The map on the left shows the archeological cultures of Italy at the beginning of the Iron Age. The dark and the grey areas are dominated by cremation burials. The white areas are almost exclusively with inhumation burials. The map on the right shows the incidence of R1b-U152/S28. There is a remarkable correspondence between cremation burials and high incidence of R1b-U152.

*Just to avoid confusions, Bell Beaker proper didn't originate in Hungary, it migrated to Hungary from the West. And moreover it had a rather scant presence there. Check the Bell Beaker map by Heyd: http://jpst.it/zFoY
There is some thin Bell Beaker presence along the Danube. And then most of all the Maros culture, which was late and syncretistic, only partly Bell Beaker influenced. And from Maros we've got a male with G2a.

pnuadha said...

@Simon_w

There's no point in taking modern central Europeans into account and estimates about degrees of true Corded Ware ancestry based on archeological links.

Care to explain this? I gave a logically sound argument. If my assumptions are correct then the conclusion follows. It almost sounds as if you are trying to magically ignore the potential truth of this conclusion (CW was not the primary contributor of Yamna like dna) in favor of the opposite. It would be like me saying that the person who just got shot in heart is going to die and you saying "no, his lungs are fine". For a person to be alive they need heart and lungs. You can't magically say the heart doesn't matter because the lungs matter.

Unknown said...

Simon W
About BB, pastoralism etc
Your comments are correct. I didn't have the exact dates on me, but we're referring to same paper.!

Tone said...

You can all mock me if you want, because admittedly I know little about this stuff:

But using Occam's Razor, isn't the simplest solution to the problem of the origins of R1B in NW Europe simply that R1B (L23) was always there? The reason Scotts (for example) show a high affinity with Yamna people is because they (the Scotts) share recent ancestry with the natives on the eastern forest/steppe zone. (EHG?).

NW Europe was repopulated after the Ice Age fairly late around 8000 bc, and instead of being populated from Iberia as thought, perhaps it was repopulated from the East by a mesolithic group that also was ancestral to the Yamna. If so, the descendant people would still be closely related since this was relatively recent event. Also, NW Europe was a last bastion of Hunter Gatherers holding out against Neolithic farmers, so there probably wasn't too much mixing or replacement by Neolithic farmer groups prior to the Bronze Age.

It's possible that when LBK communities in Central Europe collapsed around 3000 bc, groups with native ancestry invaded not only from the East, but also from the extreme North West of Europe. Cousin groups if you will manifesting mostly as Corded Ware and Bell Beaker zones.

But I'm just speculating.

capra internetensis said...

@Tone

Not going to mock you (come on, have you not seen the whackos who are into this stuff?), but no, that is not a parsimonious option when you take all the relevant information into account.

L23 is calculated to be a good deal younger than that. Even if the dating is wrong, in your scenario there ought to be long branches (many mutations) between L23, having spread way back in the Mesolithic, and the daughter nodes in the Yamnaya (Z2103) and Western Europe (L51). This is not the case at all - the branches are extremely short. In the recent study of Hallast et al (which had about 1 mutation every 390 years if you put K at 50 000 years ago based on Ust' Ishim man) the M269, L23, and L51 levels are not even distinguishable -R1b1a2b-PH1348, R1b1a2a2-Z2103, R1b1a2a1b-Z2118, and R1b1a2a1a-L151 all branch from the same node. On Y Full's tree Z2103 has the same TMRCA as L23 to the century, and L51 is only 500 years later.

In addition, we do have Mesolithic DNA from Scandinavia, and they all have I2, not R1b, despite having a modest amount of EHG ancestry. Also Loschbour, a Mesolithic forager from Luxembourg, carried I2 as well and didn't have EHG ancestry at all. It would be hard to get to the Isles from Russia without passing through either the Low Countries or Scandinavia.

And if Bell Beaker originated in NW Europe, it would have to be from the Low Countries anyway.

So even though our sample of ancient DNA from Northwest Europe is totally inadequate, we can be quite confident that L23 wasn't there in the Mesolithic.

Gioiello said...

1) To offend who thinks different is a sign of strength or weakness?
2) Won't be the aDNA to decide who is right and who is wrong?
3) From YFull it is clear that R-CTS7556 was born in Western Europe from a brother clade of mine (R-Z2110) and expanded Eastwards being the ancestor of all the Eastern R-CTS9219 etc etc
4) Warning that the muleteer becomes the mule... non illa feris incognita capris / gramina, cum tergo volucres haesere sagittae.

Simon_W said...

@ Colin

The argument is pointless, because the situation with post-Corded Ware LN/BA Central Europeans is complex enough. Some Bronze Age groups had more Yamnaya ancestry than the German Bell Beakers, others had less. Did those who had more have this from Corded Ware or were they remnants of your presumed Yamnaya-Bell Beakers who must have migrated through Central Europe? A priori both is possible. Also there is the problem that the R1b dominated German Bell Beakers themselves have barely enough Yamnaya ancestry to explain Northwestern Europeans, so no need to refer to modern Central Europeans, the problem is already there in the late Chalcolithic.

But then again there were post-Corded Ware Central European Bronze Age groups with enough Yamnaya ancestry to explain Northwestern Europeans, so the problem isn't acute. The only difficulty is that early Yamnaya admixed Northwest Europeans seem to have been strongly dominated by R1b, whereas there was more haplogroup diversity in Central Europe.

These were the reasons why I consider your argument unnecessary.

Some further thoughts:

The idea that Corded Ware and Yamnaya were somehow different races is obviously wrong. Autosomally they were rather similar, the biggest difference is in the y-haplogroup frequencies. But even there it has become apparent that Corded Ware wasn't exclusively R1a, since we've already found two with R1b.

The study I had linked to ( http://dienekes.blogspot.com/2015/05/more-y-chromosome-super-fathers.html ) shows very different patterns for R1a/R1b and Central/Northwestern Europeans on the one hand and J2 and Greeks/Anatolian Turks on the other hand. J2 has very old TMRCA and accordingly Greeks and Turks don't have the yDNA bottleneck. This means that J2 really had to be brought by a massive wave of J2 males. But the pattern for R1a and R1b is different. There surely was a massive wave of people, that's evident from the autosomes, but they needn't have been uniform on the y-chromosome, until the bottlenecks happened.

I would also point to Unetice, which so far has yielded 3/3 males with I2, although they were quite strongly Yamnaya admixed. Three individuals isn't a lot, but I wouldn't be surprised if the whole culture was predominantly I2, which again would have to have been caused by founder effects, because there was no third, I2-dominated Yamnaya-like population invading Central Europe.

And the BR1 genome from Hungary suggests that the Hungarian Yamnaya was somewhere intermediate on the EHG-WHG scale and with very low Armenian-like admixture. So it's unlikely that there was a particularly strongly Yamnaya-like wave out of Hungary. Are there any alternative migrations, perhaps North of the Carpathians? The only thing that comes to my mind is the Globular Amphora migration, which also has links to the steppe and Northern Caucasus. So we might even speculate that the R1b in the Polish Corded male was a Globular Amphora (GAC) remnant. But I don't consider this to be a likely scenario, because the links of GAC are with the western steppe, not particularly eastern. And chronologically the GAC was superimposed and superseded by the Corded Ware, so we'd again be left with a mixed R1a+R1b population.

Simon_W said...

Also, regarding your argument, it's logically sound, but I doubt the premisses. I don't find it obvious that Central Europeans in general have more Corded Ware ancestry than Northwestern Europeans. The Corded Ware didn't migrate to Britain, that's right, but later cultures from precisely the area where the Corded Ware had been, did.

Grey said...

Gioiello

"I'd agree completely with you if the point of departure were Italy."

Could be; I don't have a strong view where it originated other than I think it's likely it will be somehow connected to the spread of copper working - possibly a quite peaceful process in itself as small groups spread along the trade networks but possibly acting as a catalyst for a lot of tribal scale expansions whenever they arrived at a new source region.

Simon_W said...

If modern central Europeans have Corded Ware ancestry because they live where the Corded Ware had been, then Bronze Age Central Europeans from the same area are likely to have been more Corded Ware ancestry, because they are chronologically inbetween the Corded Ware and modern Central Europeans.

Grey said...

"But does it really stick when you say that r1b was eastern yamnaya but only represented a minority of western yamnaya and their descendants the CW folk, but that this minority r1b lineage flourished like crazy when it got to central europe."

It might stick if you incorporate the flukiness of physical geography e.g. the massive copper field around Kargaly.

Whether indigenous or caused by people coming from the outside because of the copper field a connection between the local population and copper working could have been established leading to R1b spreading as an artisan caste within Corded Ware.

Gioiello said...

@ Grey
I thank you, but unfortunately of my 10000 letters written a few have been read and are at disposal of everyone. Began Rootsweb to delete the most part of my more than thousand written ones during 2007. After, after having banned me, also DNA-Forums banned itself. Now also eng.molgen is out (I don't know why) where there are more that 1500 letters of mine. Of course I printed all what I wrote, but...
I never agreed with who thinks that the few lines survived were a few. It would be unbelievable that a line has survived for thousands of years from father to son. I think instead that one (or a few lines) survived from much more linked ones, so we'll find in the aDNA many lines extinct, and even about the miners or the copper smiths I think that it is a wrong solution, otherwise why many varied R1b1-L389*, many R-V88*, many R-M269/PF7562*, etc. are present in Italy? For the same reason I bet on Western Europe and not on Italy for R1a-M420*.

Grey said...

Gioiello

"and even about the miners or the copper smiths I think that it is a wrong solution, otherwise why many varied R1b1-L389*, many R-V88*, many R-M269/PF7562*, etc. are present in Italy?"

Yes I think the story is one of tribal migrations combined with small groups running ahead of the main advance. If the idea is correct then the regions with lots of clades would more likely be regions where a large scale migration took place and regions with a very strong single clade founder effect would more likely be the places started by run-aheads.

.

Just to add a point on one specific near-Italian R1b clade. If heavy rainfall causes acid soil (which is apparently the reason for acid soil along the Atlantic coast) then if you look at a map of euro rainfall then as well as the Atlantic coast there is also a peak around the Alps which maps quite neatly onto S28

http://www.printfree.cn/sites/printfree.cn/files/Image/20090314151759166.jpg

http://cdn.eupedia.com/images/content/Haplogroup-R1b-S28.gif

so did copper age Alpine farmers switch from wheat to oats and milk?

Gioiello said...

@ Grey
I don't know who you are (here is plenty of nicknames and many multiple), but when I wrote on Anthrogenica I didn't like alan's novels, even though he is a good person, nothing to do with many others there. Unfortunately your hypotheses are unverifiable, and science is done with verifications. There are many links I spoke about too, for instance those between the Alps and the Caucasus (Italy and Armenia share the presence of R1b1-L389*, even though Armenia with only one haplotype and Italy many; they have the most varied R-L277*, etc.), but unfortunately we need aDNA, i.e labs and money, and not our phantasy, which is for free. That said with all the respect with you.

Simon_W said...

I'd say, if R1b-P312 is from Yamnaya but not from Corded Ware, then it must be from the Hungarian Yamnaya. I'd say that's still a possibility I can't rule out. But this would entail that the R1b-P312 people reduced the Samara Yamnaya affinity rather than increasing it. Which would mean that the Samara Yamnaya affinity in Northwest Europeans is at least to a considerable part from the Corded Ware. And there's no reason why this can't be the case, if R1b-P312 males took Corded wives. And it would also mean that they were not a large mass of people that replaced the Corded folks, but rather a minority that was rich in R1b-P312. And so it would be quite equivalent to the hypothesis that they were a Corded Ware minority from the beginning.

Some doubts remain, because why didn't the R1b-P312 people manage to overwhelm Hungary in much the same way as they are presumed to have overwhelmed Germany? And what influenced the eastern Bell Beaker culture according to Gimbutas was rather a Yamnaya-Vucedol amalgam than pure Yamnaya, so this would suggest some amalgamation prior to migration, which would even more suggest that R1b-P312 should have entered the Bronze Age Hungarian gene pool.

Grey said...

@Simon_W

"why didn't the R1b-P312 people manage to overwhelm Hungary in much the same way as they are presumed to have overwhelmed Germany"

Speaking just to this bit if the population density of the original population around the edges of LBK was lower than in the core then that might be a factor - easier to swamp a smaller population.

.

Gioiello

"Unfortunately your hypotheses are unverifiable"

The R1b clades along the Atlantic coast look like they resulted from dramatic founder effects. If that is what happened there's probably lots of possible explanations why and some parts of each theory are likely to be verifiable or disprovable e.g. whether the L21 in Ireland originally came from the Pyrenees or whether BB in Iberia, Ireland and Holland all came from the same original source (but admixed with different groups due to taking different routes from that source).

So 100% verifiable probably not but disprovable, yes.


PF said...

@Davidski

Thanks for the clarification/info. That's good news indeed.

Gioiello said...

@ Grey

I was referring to this hypothesis: "Just to add a point on one specific near-Italian R1b clade. If heavy rainfall causes acid soil (which is apparently the reason for acid soil along the Atlantic coast) then if you look at a map of euro rainfall then as well as the Atlantic coast there is also a peak around the Alps which maps quite neatly onto S28 so did copper age Alpine farmers switch from wheat to oats and milk?" That seemed to me an argumentum ad hoc, and, as to it, unverifiable. But the problem is purely scientific. I have no problem to accept the expansion of the Bell Beakers from Iberia (even though hg. R1b-L51 and subclades arrived before from Italy, but we know the migration of 7500 BCE but very likely there should have been others), above all if the dates remain those of YFull (the expansion is clearly with Bell Beakers), but it is difficult to understand why R-U152 should be so huge in Italy and so slim in Iberia, above all if we know that Bell Beakers were contemporaneously from 4900 ya in Iberia but also in Southern France and Tyrrhenian Italy. For that, if this theory will be demonstrated true and not others (kurganists, Renfrew etc.), only aDNA will be able to say more. But R-U152 is dated at 4600 ya, i.e. 300 years after P312 in Iberia... as it seems that there aren't mutations in between P312* and its subclades, if they descend from different persons, they should be lived within about three generations from their ancestor.
Thus to think to a clan, or better to a chiftain who ruled on three "kingdoms" in Iberia, Southern France and Tyrrhenian Italy could be a solution, even though we have no support from history. But the other solution is that that I said above: the lines survived are only a few amongst a wider population... but the problem of other private mutations remains, even though very likely it is wrong to presuppose a constant mutation rate. Sometimes mutations are very slow.

Gioiello said...

I'd want to add more: for being sure that there aren't mutations amongst the P312* subclades, it'd need that the samples were tested from YElite or Full Genome and not from Big Y. I have had about 300 no calls out of 72000 snps, whereas Big Y has usually 12000 no calls.

pnuadha said...

@simon_w

These were the reasons why I consider your argument unnecessary.

Simon, we cannot have any fruitful discussion if you continue to ignore basic logic.

If a population has a much lower level of EHG than modern NW Europeans then they cannot be the primary source for EHG in NW Europeans. AT THE SAME TIME If a population has ample EHG but contributes ancestry to NW Europeans in proportion to the EHG ratio between the two, then they cannot be the primary source for EHG in NW Europeans.

Stop talking about 'early' problems... They are both necessary conditions for any valid explanation. They are in fact the same problem using the same math. EHG doesn't just amplify out of nowhere.

If you continue these nonsensical comments I won't even bother responding.

Also there is the problem that the R1b dominated German Bell Beakers themselves have barely enough Yamnaya ancestry to explain Northwestern Europeans

They did have enough, especially the latest batch from this study. I expect to find groups of beakers with higher levels of EHG as we get more samples, amidst the general pattern of eastern migrants mixing with locals and diluting their EHG. We need more snapshots...

The idea that Corded Ware and Yamnaya were somehow different races is obviously wrong.

you might want to add that water is wet.

And the BR1 genome from Hungary suggests that the Hungarian Yamnaya was somewhere intermediate on the EHG-WHG scale and with very low Armenian-like admixture.

Thats an extremely weak argument. You're talking about an area 1000 years later with neighbors that may have had next to no EHG.

So it's unlikely that there was a particularly strongly Yamnaya-like wave out of Hungary

OUT of hungary. Just like it was r1b OUT of the pc steppe.

Simon_W said...

Colin, I already pointed out the main flaw in your argument: It rests on the assumption that modern central Europeans have more Corded Ware ancestry than modern Northwest Europeans. This is extremely far from obvious. And so, since this has not been established, your argument rests on shaky feet. You might perhaps argue that the higher incidence of R1a in modern central Europeans proves their stronger Corded Ware ancestry. But since my position is that Bell Beaker and Northwest European R1b sprang from Corded Ware via founder effects, bottle necks and drift, I dispute that the higher incidence of R1a in Central Europeans is a safe indication of stronger Corded Ware ancestry. That's where it gets circular: You start from the assumption that a high incidence of R1b cannot be from the Corded Ware to prove that the R1b isn't from the Corded Ware.

On a more general note, IMO it's more advisable to study the ancient DNA we have instead of construing shaky arguments based on modern DNA.

BR1 cannot result from a mix of MN farmers + roaming WHG + Samara Yamnaya, because she doesn't have any West Asian/Armenian-like admixture. If you leave away Samara Yamnaya, the addition doesn't work either because she did have ANE and some EHG admixture. Apart from that, the change from CO1 to BR1 is huge, and hordes of roaming WHG in Hungary, in the time span between 2500 BC (when Bell Beaker arises) and 2000 (when BR1 lived) isn't credible.

pnuadha said...

@simon_w

Colin, I already pointed out the main flaw in your argument: It rests on the assumption that modern central Europeans have more Corded Ware ancestry than modern Northwest Europeans.

Its one of the more reliable assumption about the demographic history we do have. It simply is obvious that belarusians and czechs have more CW heritage.

But since my position is that Bell Beaker and Northwest European R1b sprang from Corded Ware via founder effects, bottle necks and drift.

Founder effect is saying that they ydna changes a lot but the autosomal dna doesn't change much.

You start from the assumption that a high incidence of R1b cannot be from the Corded Ware to prove that the R1b isn't from the Corded Ware.

No, you just fail to understand a simple argument; two of them actually. If there is a bag full of m&m's and you take a handful out of the bag while blindfolded and you get back only blue, you would not expect the bag to contain mostly red and a very small ratio of blue. The other issue is that Bell Beaker r1b was not a recent founder effect from a small population. Founder effects take time. It takes generations and generations for a population as large and as r1b dominant as BB to come from a small minority of r1b people.

BR1 cannot result from a mix of MN farmers + roaming WHG + Samara Yamnaya, because she doesn't have any West Asian/Armenian-like admixture. If you leave away Samara Yamnaya, the addition doesn't work either because she did have ANE and some EHG admixture. Apart from that, the change from CO1 to BR1 is huge, and hordes of roaming WHG in Hungary, in the time span between 2500 BC (when Bell Beaker arises) and 2000 (when BR1 lived) isn't credible.

Thats some incredibly shoddy reasoning which involves a bunch of really bad assumptions such uniformity of MN farmers, homogeneity of hungarian populations, and population continuity over about 1000 years from a populations that was on its way out. You also say that ba-hungary is inconsistent with having any yamnaya like dna; as far as i can see you just pulled that out of your behind. In haak the Ba-hungarians are estimated to have yamnaya and they do have the teal, non otzi component too.

All these false and/or poor assumptions and you can't simply see that East Central Europeans have more true CW heritage than Northwest Europeans?

«Oldest ‹Older   601 – 728 of 728   Newer› Newest»