search this blog

Wednesday, June 29, 2016

German Bell Beakers in the context of the prehistoric Near East


Fascinating stuff, and basically in line with the generally accepted archaeological model of Bell Beaker origins in Iberia.

Of course, these TreeMix results don't necessarily mean that German Beakers are a straight two-way mixture between Yamnaya pastoralists and Chalcolithic Iberians; they simply suggest that the ancestors of German Beakers experienced a significant pulse of admixture from an Chalcolithic Iberian-like population.







All of the samples used in this analysis are freely available for download at the Reich Lab website here.

Update 30/06/2016: Below is the same graph but with German Corded Ware added to the line up. The German Beakers still receive a significant admixture edge from the base of the Iberia Chalcolithic branch. However, the percentage of this edge as a proportion of their ancestry has fallen to 33%.


145 comments:

ryukendo kendow said...

Looks like there is indeed a preference for Iberia over Germany MN, and the distant connection nMonte finds is confirmed.

Could you upload the residuals?

Thanks!

Karl_K said...

This, of course, simplifies explanation of the R1a/R1b split within Europe.

Determining the exact route taken by steppe people into Iberia will be extremely interesting.

Richard Rocca said...

We are looking at a dominant male replacement in both Iberia and Italy (except Sardinia) that is due to Bell Beaker and R1b lineages....

Spain-Chalcolithic:
Haplogroup I = 6 of 9 = 67%
Haplogroup G = 2 of 9 = 22%
Haplogroup H = 1 of 9 = 11%

Italy-Chalcolithic:
Haplogroup I = 3 of 4 = 75%
Haplogroup G = 25%

We know that those two populations plot with modern Sardinians which is... surprise, surprise... the place where I2a and G2a survived best in SW Europe. So, while some R1b expansion "out of" Iberia may have occurred, those males seem to have expanded "into" Iberia a few hundred years before that. That Chalcolithic Portuguese mtDNA is overwhelmingly H is also a data point in that direction. This is the model that Jean Manco has proposed in her book and is the likeliest, albeit not only, explanation.

Davidski said...

What about a migration from Iberia to Central Europe at some stage during the Chalcolithic/Bronze Age?

In any case, here are the residuals plots.

https://drive.google.com/file/d/0B9o3EYTdM8lQRVpyRnI5Z0lFclU/view?usp=sharing

bellbeakerblogger said...

Fascinating. This is a breakthrough for several of the archaeological models. I hope this gets maximum visibility.

I think this illustrates that for over a hundred years archaeologists have been very close to the solution but did not have the resolution to understand the migration pattern. Now it is fairly clear that Chalcolithic Iberians (of unknown origin) were moving into the lower Rhine and mixing with steppe herders.

ryukendo kendow said...

Thanks!

Looking across the residuals from 2 to 4, there is indeed the expected pattern of high mbuti affinity to Iberia Chalcolithic. There's also a pattern of BB underfitting to Armenia or Anatolia Chl--cannot tell from the pic as the word is cut off.

Onur said...

Davidski, why did not you add Corded Ware to your TreeMix analysis?

Onur said...

@RK

Looking across the residuals from 2 to 4, there is indeed the expected pattern of high mbuti affinity to Iberia Chalcolithic.

Why expected?

ryukendo kendow said...

Read the comments section of the previous post.

Davidski said...

I'll test German Corded Ware using the same topology tomorrow.

Richard Rocca said...

Davidski said... "What about a migration from Iberia to Central Europe at some stage during the Chalcolithic/Bronze Age?"

The distribution of R1b-DF27 looks to be from such an expansion and in the archaeological front is attested to by Palmela points.

Onur said...

@Davidski

I'll test German Corded Ware using the same topology tomorrow.

Great. That will be a significant point of comparison to test the various theories.

Karl_K said...

@bellbeakerblogger

"Now it is fairly clear that Chalcolithic Iberians (of unknown origin) were moving into the lower Rhine and mixing with steppe herders."

I don't think that is what this is saying at all. It is saying that R1b carrying steppe people went straight to Iberia, and then moved east toward Germany after the admixture with Iberians.

Olympus Mons said...

Eheheheheheh

Olympus Mons said...

Ladies and gentleman...
Its called r1b _ from shulaveri shomu 2 bell beaker.

Or if one rather it... The birth of the Oestrimnis people that led to the Bell Beaker.

Olympus Mons said...

@ryukendo kendow,
Man...if ever in Lisbon let me offer lunch.

Olympus Mons said...

@ryukendo kendow,
Man...if ever in Lisbon let me offer lunch.

bellbeakerblogger said...

@Karl,

I don't think we're necessarily in disagreement. There's a trace leading to Iberia but I make no real supposition from there.

Olympus Mons said...

@Richard Rocca,
regarding archaeological traces, read me:
http://blogs.sapo.pt/cloud/file/eb6b52b82097d41dfa0e5797a2fa7945/olympusmons/2016/From%20Shulaveri%20to%20Bell%20beaker.pdf


Its not just Palmela points.
I have a 50 pages thesis.. But in the end, the proof is here. In the Sorraia Horses of Portugal. Check minute 2.30
The Wild Sorraia Horse: Was it the first horse before domestication?


Those are the Tarpan horses they arrive in Iberia, before expantion as Bell beaker...

Karl_K said...

@bellbeakerblogger

I would base it mostly on the R1b data. It is hard to imagine that Iberian farmers all of a sudden started moving East just in time to interact with a hidden first wave of R1b rich steppe people in Central Europe.

It seems more likely that some R1b rich pioneers made it early to Iberia, found a new synergistic niche, and started expanding out with a new culture, probably building on one that they picked up in Iberia.

Olympus Mons said...

bellbeakerblogger,

" a trace leading to Iberia.." well, you/we already know there is a trail leading OUT of Iberia, right!? - :)

Olympus Mons said...

@ Karl_K,

Have you read my Chapter III - Pumped out by the Sahara (http://shulaveri2bellbeaker.blogs.sapo.pt/chapter-iii-pumped-out-by-the-sahara-5842) and also Chapter IV - The Oestrimnis Civilization (http://shulaveri2bellbeaker.blogs.sapo.pt/chapter-iv-the-oestrimni-5393) ??



Karl_K said...

@Olympus Mons

I haven't read it, and I doubt I would agree. As I said, I think they went very quickly to Iberia. Maybe some side branches went down to Africa or Italy. They were looking for something, maybe easy metals.


Olympus Mons said...

@Karl_K,
In the end the truth will come out. It does not matter. We are just having fun :)

See, what I don't get is: Its written everywhere in archaeology, with signs, billboards, neon lights... Its people! and people have habits and tell tales that do not change easily.

The same "package", of man, cattle, dogs, spelt, tarpans horses (or domesticates), carnelian beads, etc, etc, with the same habits have left a tell tale mark in Tell tsaf, in shiqmin, in Merimde, in Alentejo portugal, as bell beaker, as Unitice, as celts ,etc...

And I, as wrong as I might be (and I dont care), can see it clearly. - Well, we will see.

Alberto said...

Before we had IBS from Bell Beakers where many Iberian populations showed surprisingly high on the list, given that Bell Beakers from Germany cluster with North Europeans. But that could be interpreted as input from BB_Germany to Iberia (though already it was strange, because BB input to NW Europe would still be clearly higher).

But adding to that a connection of BB to Iberia_Chalcolithic, plus the archaeology, kind of makes for a pretty good case. I can only confirm in the Dstats/nMonte that Iberia Chalcolithic does look really good:

Bell_Beaker_Germany
"Iberia_Chalcolithic" 29.7
"Yamnaya_Kalmykia" 24
"Yamnaya_Samara" 17.75
"Anatolia_Chalcolithic" 8.3
"Hungary_EN" 5.3
"Hungary_CA" 5.1
"Motala_HG" 4.9
"Hungary_HG" 2.75
"Satsurblia" 1
"Loschbour" 0.8
"Ami" 0.2
"Iberia_EN" 0.2
"Anatolia_Neolithic" 0
"Esan_Nigeria" 0
"Esperstedt_MN" 0
"Eastern_HG" 0
"Iran_Neolithic" 0
"Israel_Natufian" 0
"Levant_Neolithic" 0
"Iran_Chalcolithic" 0
"Armenia_Chalcolithic" 0
"Iberia_MN" 0
distance=0.002496

Now, a large migration of Iberian females to Central Europe at the same time as a large migration of steppe males to Central Europe, and them mixing there looks kind of strange. Or even if the migrations were of men and women, why did the steppe men end up with the Iberian women and the rest disappeared?

So we're left with a hypothetical migration of steppe-like guys (or mostly guys) to Iberia. But I agree this had to be pretty fast, without much mixing along the way. But how would that be? By sea?

Strange, but certainly intriguing.

Olympus, note that the Sub-Saharan was not brought by the R1b guys, it's rather contributed by the putative Iberian_Chalcolithic females.

BTW, adding Corded Ware does not change the fundamental:

Bell_Beaker_Germany
"Iberia_Chalcolithic" 26.5
"Yamnaya_Kalmykia" 24.7
"Yamnaya_Samara" 10.8
"Corded_Ware_Germany" 9.1
"Anatolia_Chalcolithic" 8.4
"Hungary_EN" 6
etc..

Krefter said...

This supports the idea of Iberian_Chal ancestry in Bell Beaker but more evidence is needed for the argument to be convincing, since German_MN and Iberian_Chal were so similar to each other.

Karl_K said...

"So we're left with a hypothetical migration of steppe-like guys (or mostly guys) to Iberia. But I agree this had to be pretty fast, without much mixing along the way. But how would that be? By sea?"

This could be a dumb thought (due to lack of evidence), but it could have been by horse. The early Iberian 'Bell Beaker' culture was eating horse. If someone arrived riding one, that could get interesting.

But of course, hard to find evidence for rare events.

Karl_K said...

And of course... the genetic evidence that Iberian wild horses were a source of later domesticated horses...

Grey said...

Ancient Iberia had a lot of soft metal.

Open Genomes said...

https://www.gedmatch.com/fcgi-bin/u_compare2.fcgi?kit1=M291439&kit2=M472767&chart=0&resolution=1000&threshold=100&shared=1&win_size=&bunch_limit=&xsubmit=Submit

Comparing Kit M291439 (I1706 LevantBA 2490-2300 BCE) and M472767 (I0232 Srubnaya 1850-1200 BCE R-Z93)

Minimum threshold size to be included in total = 100 SNPs
Mismatch-bunching Limit = 50 SNPs
Minimum segment cM to be included in total = 1.0 cM

Largest segment = 4.1 cM
Total of segments > 1 cM = 231.3 cM
140 matching segments

311638 SNPs used for this comparison.

How the hell is Srubnaya in the Middle-Late Bronze Age so closely related to the Levantine Early Bronze Age from 'Ain Ghazal in Jordan?

The IBD is far higher than Kura-Araxes or Yamnaya. There's no IBD between the Levantine Bronze Age and the Iranian Chalcolithic at all. There is however some relationship (about 80 cM) with Yamnaya, and even Afansievo:

Comparing Kit M291439 (I1706 LevantBA 2490-2300 BCE) and M828815 (RISE552_Yamnaya 2849-2143 calBCE)

Minimum threshold size to be included in total = 100 SNPs
Mismatch-bunching Limit = 50 SNPs

Largest segment = 3.0 cM
Total of segments > 1 cM = 46.9 cM
30 matching segments

What's this all about?

Karl_K said...

@Open Genomes

Any idea what the coverage of the genomic data are? Any diploid data? IBD is very sensitive in this regard.

Open Genomes said...

IBD between Levantine Early Bronze Age (2490-2300 BCE, coverage: 3.749 total SNPs: 718277) and Srubnaya (1850-1200 BCE R-Z93, coverage: 2.839 total SNPs: 971,022) even above 3 cM (100 SNPs):

Comparing Kit M291439 (I1706 LevantBA) and M472767 (I0232 Srubnaya)

Minimum threshold size to be included in total = 100 SNPs
Mismatch-bunching Limit = 50 SNPs
Minimum segment cM to be included in total = 3.0 cM

Chr Start Location End Location Centimorgans (cM) SNPs
2 71,610,940 74,190,547 3.4 259
3 3,079,951 3,814,711 3.1 143
11 20,115,145 21,312,360 3.2 220
17 74,836,719 75,760,213 4.1 154
22 16,427,231 17,352,450 3.9 136

Largest segment = 4.1 cM
Total of segments > 3 cM = 17.7 cM
5 matching segments

Colin Welling said...

@Karl

"It is saying that R1b carrying steppe people went straight to Iberia, and then moved east toward Germany after the admixture with Iberians."

That makes about as much sense as saying that Iberians ferried their way to Ukraine.

Southwestern Europe has very little steppe compared to Northwest Europe and bell beakers. It is obvious that most Steppe ancestry in Eastern Bell beakers, had to come directly from the steppe. That is also where their r1b came from.

I can't believe people are still sticking to the idea that r1b came through Iberia. Steppe came directly from the east and r1b was in the steppe...

"This, of course, simplifies explanation of the R1a/R1b split within Europe."

What split? R1b was dominant in the yamnaya on the steppe. This isn't 2012

Open Genomes said...

Possible reasons for high IBD between the Levantine Early Bronze Age (2490-2300 BCE) and Srubnaya (1850-1200 BCE R-Z93):

1. The Uruk Expansion (northward)
2. The Anatolians (southward)

Given that there is by far higher IBD between the Levantine Early Bronze Age and Srubnaya than either Yamnaya or Afansievo, *and* that the Levantine Early Bronze Age is at least 450 years earlier, it the Uruk expansion seems more likely to be the cause.

The Uruk Culture is clearly Sumerian. (There is continuity with known Sumerian sites in southern Iraq.) The Uruk Expansion reached the North Caucasus.

https://en.wikipedia.org/wiki/Uruk_period#Uruk_expansion

Did the widespread Uruk Expansion leave any traces in the Y-DNA of the North Caucasus?

Too early (listed east to west):
J1-Z1842:
https://yfull.com/tree/J-Z1842/
J2a-CTS900:
https://yfull.com/tree/J-CTS900/
G2a-Z6653 (old G-P16):
https://yfull.com/tree/G-Z6653/
G2a-L1266:
https://yfull.com/tree/G-Z6653/

Too late:
L-M317:
https://yfull.com/tree/L-M317/

What is it then?

Karl_K said...

@Colin

"What split? R1b was dominant in the yamnaya on the steppe. This isn't 2012"

OK. Now I am seeing all the R1b in Corded Ware in Eastern Europe!

Thanks for pointing it out to me!

Colin Welling said...

@karl,

The split is between bell beakers and corded ware, both of whom have roots in eastern europe. The ydna split was already evident in eastern europe!!! there is no need to come up with fantasy about a migration from the steppe to iberia to central europe, to purify r1b and carry r1b to central europe. R1b was already concentrated in eastern europe (yamnaya) and it moved directly to central europe with the majority of steppe ancestors in the bell beakers

Open Genomes said...

Even higher IBD between Levantine Early Bronze Age (2490-2300 BCE) and BR2 (Late Bronze Age, Kyjatice Culture, Ludas-Varjú-dűlő, Hungary, J2a-Y17946*) than with Srubnaya (1850-1200 BCE R-Z93):

Comparing Kit M291439 (I1706 LevantBA) and F999933 (BR2, Hungary, 3.2ky)

Minimum threshold size to be included in total = 100 SNPs
Mismatch-bunching Limit = 50 SNPs
Minimum segment cM to be included in total = 3.0 cM


Chr Start Location End Location Centimorgans (cM) SNPs
2 72,017,801 74,931,709 3.8 290
5 91,139 1,368,660 3.3 154
8 11,357,562 12,668,042 3.1 130
9 38,458,570 71,189,984 4.4 181
9 138,101,543 140,058,004 3.8 190
12 6,349,553 7,690,103 3.2 169
17 14,225,791 14,865,931 3.6 162
18 4,286,997 5,150,529 3.0 106
Largest segment = 4.4 cM
Total of segments > 3 cM = 28.2 cM
8 matching segments

324072 SNPs used for this comparison.

Olympus Mons said...

@Colin Welling
Can't believe people are still stuck in the r1b from steppe and yamnaya. Pathetic. This os not 2014 anymore. Dude this is 2016. And yamnaya and steppe died out way back. :)

Open Genomes said...

BR2 BR2 (Late Bronze Age, Kyjatice Culture, Ludas-Varjú-dűlő, Hungary 1270-1110 BCE) is in a Y-DNA J-M67/CTS900 clade with Chechens, Aegean Greeks, Northeast Italians, Sardinians, Andalusians (Puerto Ricans), and Dutch.

It seems that BR2's ancestors went west across the Mediterranean in search of metals, perhaps as part of the Minoan expansion. There are in fact J-CTS900 people in the Levant today, who are not directly descended from Northeast Caucasians.

This IBD bypasses Unetice (and presumably the earlier Bell Beakers).

https://yfull.com/tree/J-Y17946/

Open Genomes said...

For comparison, the IBD between BR2 and Srubnaya:

Comparing Kit M472767 (I0232 Srubnaya) and F999933 (BR2, Hungary, 3.2ky)

Minimum threshold size to be included in total = 100 SNPs
Mismatch-bunching Limit = 50 SNPs
Minimum segment cM to be included in total = 3.0 cM


Chr Start Location End Location Centimorgans (cM) SNPs
3 61,248,582 63,649,383 4.3 849
5 73,590,100 75,317,053 3.0 357
6 2,965,688 4,050,735 3.2 339
9 2,314,646 3,706,970 3.7 489
9 38,743,508 71,042,580 3.7 235
12 6,346,984 7,762,379 3.4 326
Largest segment = 4.3 cM
Total of segments > 3 cM = 21.3 cM
6 matching segments

Olympus Mons said...

Did any of you read anything about iberia chalcolithic?
Well start by reading the last chapters of my thesis. Is all there.

Stop behaving like archeology does not exist!!! - get real you adna freaks ;)

Olympus Mons said...

Alberto
5.9 kiloyear event was under way and loads of north africa population were crossing over to iberia.
Lovely intriguing story. I have read the papers...so just read my thesis. All references are in there....

Open Genomes said...

There is an IBD segment between
M291439 (I1706 Levantine Early Bronze Age, 'Ain Ghazal, Jordan, 2490-2300 BCE) and F999933 (BR2, Late Bronze Age Kyjatice Culture, Ludas-Varjú-dűlő, Hungary, 1270-1110 BCE)
and between:
M472767 (I0232 Srubnaya Novoselki, Northern Forest, Samara Russia, 1850-1200 BCE) and F999933 (BR2, Late Bronze Age Kyjatice Culture, Ludas-Varjú-dűlő, Hungary, 1270-1110 BCE)

but the same segment is *not* shared between
M291439 (I1706 Levantine Early Bronze Age, 'Ain Ghazal, Jordan, 2490-2300 BCE) and M472767 (I0232 Srubnaya Novoselki, Northern Forest, Samara Russia, 1850-1200 BCE).

The two IBD segments are on opposite phases, two alternate alleles on chromosome 9.

Chr Start Location End Location Centimorgans (cM) SNPs


Srubnaya and BR2:
9 38,743,508 71,042,580 3.7 235


I1706 (Levant EBA) and BR2:
9 38,458,570 71,189,984 4.4 181

My guess then is that there is a common connection between the Bronze Age Levant and BR2 (Mediterranean > Central Europe, perhaps via Spain) and the Near East and Srubnaya, originating from a common source in the Northern Near East. Perhaps this has something to do with the Kura-Araxes culture and it's Levantine extension, the Khirbet Kerak culture?

Perhaps the Kura-Araxes culture went from Anatolia to Crete in the Early Bronze Age, then spread out from there?

Kura-Araxes culture from Chechnya to Syria and Israel

Khirbet Kerak in the Early Bronze Age

Karl_K said...

@Colin

"The split is between bell beakers and corded ware, both of whom have roots in eastern europe. The ydna split was already evident in eastern europe!!!"

What split? Who are you arguing with, yourself?

I am talking about the fact that the R1a and R1b steppe people were split from the start, and did not migrate into Central and Western Europe all mixed up together.

The R1b people went all the way West immediately, and then went back East from there until they met up with R1a step people.

How can you disagree with that?

Grey said...

@Colin Welling

It might not make sense for large tribes but metal workers island hopping to Iberia is easy enough to imagine.

Grey said...

Just a thought

any evidence of horses used more in ancient mining than oxen?

(size?)

Rob said...

"don't think that is what this is saying at all. It is saying that R1b carrying steppe people went straight to Iberia, and then moved east toward Germany after the admixture with Iberians."

I'd say that has about 1% chance of panning out
;)

Let's note here that the out of Iberia movement precedes any sort of "steppe" admixture into Iberia, and what more, the "steppe" admixture was indirect, probably not from the actual steppe, & via a secondary stay in Central Europe

Olympus Mons said...

@Rob, what do you mean by
"and what more, the "steppe" admixture was indirect, probably not from the actual steppe, & via a secondary stay" ?

Rob said...

@ OM

Oh I just mean that steppe admixture in German BB was mediated via steppe groups in the Carpatho - Danube region, as well as possibly earlier pre-CWC cultures (like GAC) of north-central Europe.

In turn, at some point after 2200 BC, this Danubianized steppe component then reached Iberia. So, I doubt there was any direct steppe movement to Iberia in 2800 BC, indeed there is virtually 0 evidence for this

la señora bibiloni said...

Probably my question is redundant, but here it goes:
We find both Iberian_Chalc and Yamnaya in Bell Beakers. Fine, guy from the East & Girl from the West (or the opposite) tie the knot and there we are.
But how much Yamnaya is there in Iberian_Chalc?
Do we find "Yamnaya" in Iberian_Chalc? Or just South-of-the-Caucasus stuff with a bit of EHG?
I am not arguing, just asking, because all this thing about how to get faster to Iberia is getting confusing...

Gioiello said...

The point of departure is of course where R-L51 and the first subclades will be found in aDNA. No doubt that it is in Italy (I invited Lazaridis to test Tyrrhenian Italy). The unique doubt I have is about R-L11*, which seems expanded from German people around the Baltic to Crimea to Burgundia Iberia and the Isles during the Medieval migrations and I have no proof as to the other R-L11* basal haplotype if it is of old Italian origin or come with Germans it too. Italy has all the oldest subclades from R-L389* except just R-L11*.

Rob said...

Senora

The Iberian Chalcolithic dates from 3300 - 2200 BC, thus the latter half of this period coincides with the BB phenomenon. Of all examined Spanish Chalcolithic samples to date, they essentially look like a continuity from the preceding Spanish Neolithic (which are Anatolian farmers + 20% WHG). Nothing too extravagant, nothing too exotic (no steppe, no Trans-Caucasus mixture). .

But none of the Spanish samples thus far are actually BB culturally. I.e. they come from non-BB contexts.

Colin Welling said...

@grey

It might not make sense for large tribes but metal workers island hopping to Iberia is easy enough to imagine.

But it was large tribes migrating from the steppe that caused central european bell beakers to be about 1/2 to 1/3 steppe derived. They brought r1b with them.

@karl

I am talking about the fact that the R1a and R1b steppe people were split from the start, and did not migrate into Central and Western Europe all mixed up together.

Yes, they did not come to central europe, as a single culture, mixed together. In no way does that suggest r1b took some weird route into central europe. You don't need to create some complicated method of arrival for r1b or r1a into central europe to explain the separation of the two haplogroups along cultural lines in central europe.

In no way does stating that r1b came from the steppe, to spain, to central europe help explain why r1b and r1a stayed divided along cultural lines in central europe.

And on top of postulating some random migration path for r1b into central europe, you have disassociated the introduction of r1b into central europe from the majority of steppe heritage in bell beakers. That is because southwest europe cannot be the source of the majority of steppe ancestry eastern bell beakers.





Colin Welling said...

@rob

Oh I just mean that steppe admixture in German BB was mediated via steppe groups in the Carpatho - Danube region, as well as possibly earlier pre-CWC cultures (like GAC) of north-central Europe.

In turn, at some point after 2200 BC, this Danubianized steppe component then reached Iberia. So, I doubt there was any direct steppe movement to Iberia in 2800 BC, indeed there is 0 evidence for such a caricatural theory.


Totally agree. There is also the r1b that was found in the upper balkans dated to around 2,800 bc

Colin Welling said...

@olympus

You do a lot of self promotion, lol.

I can't wait till we test dna from the western steppe and eastern balkans

Gioiello said...

YFull has no doubt that the basal haplotype is in Italy
R-L51 L51/M412/S167/PF6536 * PF6535 * CTS10373/PF6537/FGC39+2 SNPs formed 6300 ybp, TMRCA 5800 ybp info
id:ERS257000ITA [IT-CA]
but I not always agree with YFull.

Gioiello said...

Anyway R-L11* had a bottleneck of other 800 years as to R-L51
R-L151 P310/S129/PF6546 * YSC0000191/PF6543/S1159 * L52/PF6541+9 SNPs formed 5800 ybp, TMRCA 5000 ybp info
R-L151*id:YF03902
but my R-L23-Z2110* is the oldest line: 7200 years as to my Full Genome and so far all the close subclades are in Western Europe.

Ariele Iacopo Maggi said...

Interesting results for Anatolia Chalcolithic with Eurogenes k13 GEdmatch.

1 South_Italian @ 14.004732
2 Cyprian @ 15.955604
3 Sephardic_Jewish @ 17.035538
4 Central_Greek @ 17.244877
5 Algerian_Jewish @ 17.439276
6 Italian_Jewish @ 17.526861
7 East_Sicilian @ 17.649364
8 Italian_Abruzzo @ 19.468262
9 Tunisian_Jewish @ 19.643234
10 West_Sicilian @ 19.847778

la señora bibiloni said...

@Rob

Thanks for the explanation.
If there was no steppe dna in Iberian_Chalc, why are people speculating how the steppe guys got to Iberia to make up the Bell Beakers? Couldn't the Bell Beaker expansion start as an Iberian_Chalc DNA tribe and go picking other DNA's along the way?

Rob said...

I think the view that BB was a steppic phenomenon from the outset, with steppe people somehow getting to Iberia already by 2800 BC, is probably a minority one at present; but one which cannot be ruled out until we get actual BB samples from Iberia too (shall soon). But maybe those people who think it was can present evidence for steppe colonization of Iberia in 3000 BC.....

Rather, I think a more reasonable hypothesis is along the lines of what Ryu pointed out in the previous thread with admixture components ( as tree mixes might be a little broad brushed), and one supported by virtually every archaeologist. And that is, BB was a fusion of several elements culminating along the Rhine - upper Danube region, but apparently dominated by R1b from Eastern Europe from a male Perspective.

I suspect early Iberian BB which just look like other Iberian Chalcolithic samples

Rob said...

Ryu
Can you do your 'sequential dropping' approach also for Rathin, and add the Remedello specimens from Italy in evaluating both Rathlin & German / Czech BB ?

Davidski said...

I've added Corded Ware Germany to the analysis. Check out the update above.

Poise n Pen said...

If r1b came out of iberia or italy then really steppe is nothing to do with anything. And it's been basically impossible this wasn't the case for some time due to the way H mtdna spread around.

It's just some crazy circular logic that the people in yamnaya have steppe ancestry and come from the steppe therefore europeans came from the steppe. For some reason it's impossible that this 'steppe' dna comes from anywhere but the steppe even though those steppe people are much more like udmurts than NW europeans.

And today there is one r1b group in the extreme east and the main group sitting in western europe with all the r1a people in between so obviously there is already r1b in NW europe by the time of yamnaya but the powers that be have no interest at all in finding this. They would much rather make up a fairy tale that modern whites are either newcomers to their lands and don't deserve them, or some kind of recent mutations.

Davidski said...

You're not making any sense.

Bell Beakers and Corded Ware have a lot of steppe ancestry, and this is why they're in this cluster here with Yamnaya and Poltavka, and it's also why they're rich in Y-HG R1.

https://drive.google.com/file/d/0B9o3EYTdM8lQa3k5cTJjME9yUVk/view?usp=sharing

They're also the ancestors of modern Northern Europeans. If you don't get this by now, then get another, less demanding hobby.

Grey said...

Rob

"maybe those people who think it was can present evidence for steppe colonization of Iberia in 3000 BC"

Copper workers, originally of steppe origin island hop to Iberia following metal deposits - merge with Atlantic Megalith culture - expand out of Iberia.

PF said...

@ Ariele

Wow, that's beautiful. I knew this cluster was real!

Ariele Iacopo Maggi said...

Off topic I know...
"I1584-AnatolianChL" with MDLP K13
Using 1 population approximation:
1 Italian_South @ 4.199142 2 Greek_Coriell @ 5.807376 3 Greek_WGA @ 7.880704 4 Maltese @ 10.142474 5 Cretan @ 10.439176 6 Italian_EastSicilian @ 10.601636
7 Greek-Islands @ 11.053112 8 Italian_WestSicilian @ 11.503212 9 Turk_Jew @ 11.648053 10 Sicilian @ 12.027288

Using 2 populations approximation:
1 50% Greek_Coriell +50% Italian_South @ 3.854614

No doubt in my mind that there was a second wave migration from anatolia to SE Europe that predates the iron age and was very CHG and east med. That Cyprus component (the less exotic levantine) is probably from anatolia chalcolithic.

Rob said...

Grey

There was no "island hopping", as fun as it sounds.

Karl_K said...

@Rob

"I think the view that BB was a steppic phenomenon from the outset, with steppe people somehow getting to Iberia already by 2800 BC, is probably a minority one at present"

Definitely. I don't know who is saying that BB started when steppe people arrived. But by the time they got to Germany, they had a ton of steppe DNA with some Iberian Copper Age admixture.

Also, there are people in Iberia today with lots of R1b, yet are lower in steppe ancestry and higher in local copper age ancestry.

I am not saying this R1b rich steppe ancestry really came 'directly' from the steppe, only that it didn't come across Central Europe like the Corded Ware. Maybe they came through Hungary and then quickly made it to Iberia.

In any case, the BB cultural package was picked up by people heavily admixed with steppe ancestry. Then it spread out of Iberia.

So probably early Iberian BB will be locals, and later ones will have increasing more steppe ancestry.

ryukendo kendow said...

@ Davidski

If the overlap/quality is not too poor, could you include Remedello in the D-stats list as well?

@ Rob

I'm waiting for Chad for some more columns, so we can see if the cosmopolitanism remains if we exclude the damaged genomes. This plus the Remedello should give us some new insights.

A question though-what is the sociological model for the spread of BB? So far all the 'network'/'transhumance' models seem quite unrealistic, but it doesn't look like a mass directional spread of a cultural group either.

ryukendo kendow said...


Just to KIV that the nMonte analysis didn't just suggest ancestry from Iberia, pointing to ancestry from Anatolia or the Balkans as well, as it indicated a special and stable preference for Balkan genomes, so the final picture is likely to involve more than just Central and Southwest Europe.

Rob said...

Karl

Quite an interesting hypothesis, like a sub-Alpine route, or via Balkans then Italy ?

I think it is possible
But the only problem is that these areas are going to be too EEF to account for the level of steppe admixture in BB Germany. Indeed, we kind already know that western Hungary and northern Italy were pretty much still 'farmer' like until 2500 BC, and possibly later, which is all too late for BB. Of course, they could have 'jetted' through with no intermixture at all.
However, the "archaeological trail" for this scenario isn't there, either

The other aspect to consider is if the Iberian Chalc admixed with steppic component in Germany, as two unrelated but fusing waves

Karl_K said...

@Rob

I don't think them mixing in Germany makes it any simpler. The R1b people would have had low EEF admixture when they met up with the Iberians. And it makes it hard to explain why R1b is most prevalent on the Atlantic coast.

Do the R1b Irish Bronze Age genomes also show Iberian Chalcolithic admixture?

Rob said...

@ Ryu

I couldnt give a concise answer on that. Specialists have been debating that for years, but I hope aDNA will add another level of evidence - learning about the social from the biological.

But here are a good few papers to begin with (should be avail via academia.edu, otherwise email me ill send them)

* Families, Prestige Goods, Warriors & Complex Societies:Beaker Groups of the 3rd Millennium cal BC Along the Upper & Middle Danube. V Heyd and When the West Meets the East. The Eastern Periphery of the Bell Beaker phenomenon & its relation to Aegean Early bronze Age

* What linked the Bell Beakers in third millennium BC Europe? Vander Linden

* The earlier Bell Beaker migrations: Migrations to Britain & Scotland. A Fitzpatrick

* Origin of the Bell Beaker phenomenon. The Moroccan connection,. J Turek


I get the sense that scholars see it as successful hosehold lineages which circulated females, and became the core of successful settlements with time. At the same time, it seems like an expansive ideology from the west; regardless of what one makes of the issues of reservoir effects with Iberian RC dates, looks like an expansive western European phenomenon which succesfully moved east after 2500 BC, often at the expense of CWC - after which it follows in central Europe.

So the question is how to reconcile is large steppe component with its western origins: the theory which resonates with me is that the eastern half of bell Beaker (ie our German samples) formed in Germany, as a fusions of the elements you spoke of yesterday, then expanded back east, and also west/ south toward southern France & northern Italy, replacing the local variety of BB.

The other aspect is, as you point out, BB is not pastoral. It is a mixed economy, but more farming than pastoral. In fact, when BB appears in ECE, it brings with it farming again, which appears lacking/ scant in CWC.
In western mainland Europe, BB appears to have just developed or grafted itself onto pre-existing changes, perhaps accelerating them. In Britain, it appears it was more of a "BB package" arriving en bloc.

But I haven't had the time to synthesize all views into anything coherent yet. There are still outstanding issues with dating, etc, also

Rob said...

@ Karl

"I don't think them mixing in Germany makes it any simpler. The R1b people would have had low EEF admixture when they met up with the Iberians. And it makes it hard to explain why R1b is most prevalent on the Atlantic coast."

Doesn't the phylogeny of L51 look like it split into several directions from Germany or northern France ?
So again, your scenario would require the prospective BB people to have moved all the way to Iberia, from the steppe, via whichever specific route , without minimal or no admixture. Possible, but at the moment I fall back to the central European route, and the 'Rhine fusion corridor' (eg as stated by Turek & Fitspatrick). Frank N also pointed out some distinctly proto-Beaker looking CWC assemblages in Malopolska.
But, hey, im happy to wait a few more months on this, whatever i confirms.

"Do the R1b Irish Bronze Age genomes also show Iberian Chalcolithic admixture?"

I asked Ryu to check this also.
But i believe the Rathlin paper suggested they didn;t (?)

Kristiina said...

Correction!
The IE reconstruction I refererred to above is for 'drag a wagon' and not for the word wagon itself, but the word 'wagon' is in most cases derived from this same word.

http://starling.rinet.ru/cgi-bin/response.cgi?root=config&morpho=0&basename=%5Cdata%5Cie%5Cpiet&first=1&off=&text_proto=&method_proto=substring&ic_proto=on&text_meaning=&method_meaning=substring&ic_meaning=on&text_hitt=&method_hitt=substring&ic_hitt=on&text_tokh=&method_tokh=substring&ic_tokh=on&text_ind=&method_ind=substring&ic_ind=on&text_avest=&method_avest=substring&ic_avest=on&text_iran=&method_iran=substring&ic_iran=on&text_arm=&method_arm=substring&ic_arm=on&text_greek=&method_greek=substring&ic_greek=on&text_slav=&method_slav=substring&ic_slav=on&text_balt=&method_balt=substring&ic_balt=on&text_germ=&method_germ=substring&ic_germ=on&text_lat=&method_lat=substring&ic_lat=on&text_ital=&method_ital=substring&ic_ital=on&text_celt=&method_celt=substring&ic_celt=on&text_alb=&method_alb=substring&ic_alb=on&text_rusmean=&method_rusmean=substring&ic_rusmean=on&text_refer=&method_refer=substring&ic_refer=on&text_comment=&method_comment=substring&ic_comment=on&text_any=*weg%27he-&method_any=substring&sort=proto&ic_any=on

There is also a word for axle in proto-IE: *ak's- (attested in Old Indian, Old Greek, Slavic, Baltic, Germanic, Latin and Celtic)

The other interesting proto IE reconstruction is 'yoke'

http://starling.rinet.ru/cgi-bin/response.cgi?root=config&morpho=0&basename=%5Cdata%5Cie%5Cpiet&first=1&off=&text_proto=&method_proto=substring&ic_proto=on&text_meaning=&method_meaning=substring&ic_meaning=on&text_hitt=&method_hitt=substring&ic_hitt=on&text_tokh=&method_tokh=substring&ic_tokh=on&text_ind=&method_ind=substring&ic_ind=on&text_avest=&method_avest=substring&ic_avest=on&text_iran=&method_iran=substring&ic_iran=on&text_arm=&method_arm=substring&ic_arm=on&text_greek=&method_greek=substring&ic_greek=on&text_slav=&method_slav=substring&ic_slav=on&text_balt=&method_balt=substring&ic_balt=on&text_germ=&method_germ=substring&ic_germ=on&text_lat=&method_lat=substring&ic_lat=on&text_ital=&method_ital=substring&ic_ital=on&text_celt=&method_celt=substring&ic_celt=on&text_alb=&method_alb=substring&ic_alb=on&text_rusmean=&method_rusmean=substring&ic_rusmean=on&text_refer=&method_refer=substring&ic_refer=on&text_comment=&method_comment=substring&ic_comment=on&text_any=yoke&method_any=substring&sort=proto&ww_any=on&ic_any=on

There are also more words related to cattle: Proto-IE: *uks- bull, ox

The IE word for horse is '*ek'w', but it is strangely absent in Slavic, Germanic and Baltic languages (or found in a marginal meaning). However, horses were known in the Globular Amphora Culture and Yamna (e.g. Pit-Grave site of Mikhailovska). Do not know if there are horse remains on Corded Ware sites.

huijbregts said...

Davids TreeMix chart have generated a lot of attention and rightly so. They seem the revive the theory of an Iberian ancestry of Bell Beakers.
Alhough I wonder whether you don't get a very similar TreeMix if Baalberge like pops have immigrated into Chalcolithic Spain.

But the nMonte analyses which show a lot of Iberia_Chalcolithic in Bell_Beaker are NOT VALID.
nMOnte estimates admixture percentages by repeatedly selecting mixtures with smaller distances to the target pop.
The internal calculations presuppose that the columns of the datasheet are orthogonal (independent).
I expect that the algorithm is reasonably robust against small deviations from orthogonality.
Anyhow this is not a problem when nMonte is applied to calculator sheets, because the authors have the ambition to create independent columns.
But Dstat sheets are exceptional in having an extremely high correlations between the columns. In the correlation matrix you will hardly find a correlation smaller than 0.9
This really wrecks the nMonte algorithm.
There is a way out. With a PCA it is easy to transform the Dstats into orthogonal scores.
Here a are two nMonte runs which I did on the D-stats3b sheet.
The first is the flawed analysis on the raw Dstats, the second is the valid analysis on the (first 3 columns of) the PCA-scores.

Bell_Beaker_Germany
"Yamnaya_Samara" 24.75
"Iberia_Chalcolithic" 17.3
"Baalberge_MN" 12.6
"Corded_Ware_Germany" 12.2
"Yamnaya_Kalmykia" 8.55
"Hungary_EN" 7.75
"Motala_HG" 4.9
"Satsurblia" 3.65
"Iberia_MN" 3.55
"Tunisian" 1.7
"Loschbour" 0.95
"Hungary_HG" 0.8
"Israel_Natufian" 0.55
"Levant_Neolithic" 0.5
"Poltavka_outlier" 0.2
"Moroccan" 0.05
"Anatolia_Neolithic" 0
"Esperstedt_MN" 0
"Iberia_EN" 0
"LBK_EN" 0
"Villabruna" 0
distance=0.00225

Bell_Beaker_Germany
"Yamnaya_Kalmykia" 11.6
"Yamnaya_Samara" 11
"Satsurblia" 10.6
"Corded_Ware_Germany" 9.8
"Motala_HG" 7.5
"Iberia_MN" 7.1
"Poltavka_outlier" 6.6
"LBK_EN" 6.35
"Baalberge_MN" 4.4
"Anatolia_Neolithic" 4.05
"Iberia_EN" 3.6
"Hungary_EN" 3.5
"Hungary_HG" 2.8
"Loschbour" 2.4
"Iberia_Chalcolithic" 2.2
"Esperstedt_MN" 2.1
"Levant_Neolithic" 1.8
"Israel_Natufian" 1.5
"Villabruna" 0.6
"Tunisian" 0.3
"Moroccan" 0.2

It is not too hard to reproduce these caluculations (don't scale the PCA, use 3 columns of the scores).
But I will add the calculation to nMonte. Give me a few days.

Grey said...

Rob

"There was no "island hopping""

maybe not but anyway

map of ancient ore deposits

http://3.bp.blogspot.com/-_6pmQGgTT-M/UA2lbrxZ5KI/AAAAAAAAFEw/Lm-DGjmEu6U/s1600/ores.jpg

if there was a maritime or partly maritime route i'd suggest either

1) direct from steppe via those islands north of crete to sardinia to iberia

or

2) from Carpathian Basin to Italy to Sardinia to Iberia

(if the latter i've always thought there might two lots of BB-like populations - maritime one to Iberia and riverine one from central europe to the rhine mouth)

a dramatic population explosion from very small numbers to very large would be more plausible if the Atlantic coast - or at least the areas where the copper was - was relatively under populated by the Atlantic Megalith culture because of acid soils (caused by heavy rainfall) and if the steppe people brought a form of sustenance that worked better on those acid soils.

Grey said...

the little islands north of Crete marked as sources of copper are Kythnos and Seriphos

http://3.bp.blogspot.com/-_6pmQGgTT-M/UA2lbrxZ5KI/AAAAAAAAFEw/Lm-DGjmEu6U/s1600/ores.jpg


apparently the ore on the islands ran out around the time of the Minoans


https://books.google.co.uk/books?id=UYpVBQAAQBAJ&pg=PA58&lpg=PA58&dq=copper+deposits+aegean+islands&source=bl&ots=MPbCI10G0I&sig=XCjuR7B-TX51mwuPXmACLsViJUc&hl=en&sa=X&ved=0ahUKEwjx17G0i8_NAhULKMAKHXZqDfYQ6AEIHDAA#v=onepage&q=copper%20deposits%20aegean%20islands&f=false


Karl_K said...

"a dramatic population explosion from very small numbers to very large would be more plausible if the Atlantic coast"

So you're thinking lactose tolerance? It does look like the Bell Beakers were more likely to be lactase persistent, and it matches the R1b distribution pretty well.

ryukendo kendow said...

@ Huijbregts

use 3 columns of the scores

Hmmm, why? Wouldn't this get rid of all the variation in the higher dimensions? Shouldn't we use all the dimensions?

From your results, it looks like using 3 columns results in only the big four/big six being matched properly, with some info lost. I might orthogonalise the sheets and try some fits myself.

Davidski said...

@huijbregts

Alhough I wonder whether you don't get a very similar TreeMix if Baalberge like pops have immigrated into Chalcolithic Spain.

They didn't. Iberia Chalcolithic is Spain EN (Early Neolithic) with extra local WHG admixture.

Karl_K said...

@Huijbregts

Bell_Beaker_Germany
"Iberia_MN" 7.1
"LBK_EN" 6.35
"Anatolia_Neolithic" 4.05
"Iberia_EN" 3.6
"Hungary_EN" 3.5
"Hungary_HG" 2.8
"Iberia_Chalcolithic" 2.2
"Levant_Neolithic" 1.8
"Israel_Natufian" 1.5
"Villabruna" 0.6
"Tunisian" 0.3
"Moroccan" 0.2

So this looks about the same to me. It is just splitting the Iberian Chalcolithic up into its components. Likely because the specific Iberian Chalcolithic sample is not from the specific population that had input into the Bell Beakers.

ryukendo kendow said...

@ huijbregts

Huij, here is an analysis, on an orthogonalized datasheet containing all 26 dimensions, utilising the same list of source populations that I threw at the Bell Beaker Germany before, no populations dropped:

[1] "distance%=0.1708 / distance=0.001708"


Bell_Beaker_Germany
"Unetice_EBA" 24.45
"Yamnaya_Samara" 17.7
"Iberia_Chalcolithic" 15.1 <-------
"Hungary_BA" 10.9 <-------
"Anatolia_Chalcolithic" 10.8 <-------
"Armenia_MLBA" 5.55 <-------
"Motala_HG" 3.55
"Iberia_EN" 3.1
"Samara_Eneolithic" 2.95
"Loschbour" 2.9
"Baalberge_MN" 1.35
"Moroccan" 0.8 <--------
"Yamnaya_Kalmykia" 0.45
"Poltavka_outlier" 0.2
"Satsurblia" 0.1
"Villabruna" 0.1

If Unetice EBA is dropped:

[1] "distance%=0.1737 / distance=0.001737"


Bell_Beaker_Germany
"Yamnaya_Samara" 24.9
"Iberia_Chalcolithic" 23.2 <---------
"Hungary_BA" 13.4 <---------
"Anatolia_Chalcolithic" 10.25 <---------
"Armenia_MLBA" 8.9 <---------
"Motala_HG" 4.85
"Baalberge_MN" 4.1
"Samara_Eneolithic" 3.55
"Poltavka_outlier" 1.85
"Loschbour" 1.6
"Andronovo" 1.2
"Yamnaya_Kalmykia" 0.6
"Iberia_EN" 0.5
"Satsurblia" 0.45
"Villabruna" 0.25
"Israel_Natufian" 0.15 <-------
"Corded_Ware_Germany" 0.1
"Masai_Kinyawa" 0.1 <-------


This is virtually the same situation as before, with strong indications of Iberian and Southeast European/Middle Eastern contributions, though the percentages have moved around slightly. The overall pattern is the same.

ryukendo kendow said...

I think the real changes may come when I receive the stats for the Bell Beaker with the RISE samples removed, from Chad, which would really help to clarify the "moroccan", "Natufian" and "Levantine" phenomena.

huijbregts said...

@ Ryu
I am glad you will look into this too.
I looked at the Scree plot (variance by dimension) and concluded that 3 dimension contain most of the variance. The 23 higher dimensions are mostly "noise", therefore I dropped them.
Your impression that only the big 4 or 6 are matched is interesting. Keeping only the first 3 columns is a way of aggregating. It is conceivable that only the deep ancestry is kept.
My main point is: if the columns are not orthogonal, the nMonte algorithm cannot work properly.
By the way, nMonte on scores takes a lot more time to converge. It takes at least 1000 cycles.

Grey said...

Karl_K

"So you're thinking lactose tolerance?"

That's one option although now I wonder if there was simply more under populated marginal cropland along the Atlantic coast because of the acid soil which was habitable with dairying - and the LT developed as a result of dairying becoming the staple diet.

looking at the ore map again

http://3.bp.blogspot.com/-_6pmQGgTT-M/UA2lbrxZ5KI/AAAAAAAAFEw/Lm-DGjmEu6U/s1600/ores.jpg

*if* something like this did happen then the hop from Iberia to Brittany and SW Britain/Ireland jumps out

It would be an exceptional fluke of course

ryukendo kendow said...

Hi Huij,

Honestly, it was a relief that the runs turned out the same haha. I had a suspicion that they would, but all the same it was nice to see this confirmed.

The way I thought of it was the following: the authors of qpAdm made sure to tell us that recent, i.e. post-admixture, gene flow must not occur between the outgroups and the sources if all the distances are to be measured properly, because only differences in deep ancestry, e.g. ANE, EHG, Natufian, Iran_N etc. should discriminate between the source populations as contributors. In such a method, the position of an admixed population is a simple euclidean matter of adding the spatial positions of the sources in proportion to the ancestry.

However, i would expect that in a method like nMonte the existence of post-admixture gene flow between, say, Bedouin and Armenia MLBA, or Bedouin and Cypriot, is information to be capitalised on, to zoom in on the identity of the actual source populations, instead of just 'noise'. As this gene flow 'distorts' the position of the source population in space, such that Armenia MLBA or Cypriot is 'too' close to Bedouin than what their deep ancestry will allow, if they really have recent gene flow with Bedouin, than, say, some mixture like Satsurblia+Anatolia_EN+Iran_N+Levant_Neolithic, i.e. a mixture exactly like Armenia MLBA or Cypriot in deep ancestry, but is missing the extra drift path shared between Armenia MLBA/Cypriot and Bedouin due to actual recent gene flow. So, if Bell Beaker really had Middle Eastern ancestry, it can be expected to share a very short drift path with the Cypriot and Bedouin columns caused by the existence of intermediary populations like Armenia MLBA, which the algorithm will attempt to account for by adding a Middle Eastern population that can mediate this drift path, instead of just a population or a mix of populations that matches in deep ancestry but does not create this 'distortion' towards the Cypriot and Bedouin columns in excess of what the deep ancestry of BB will allow.

I think these minor drift paths are very important because I get very different results for e.g. Moroccan if I allow Bedouin in both the rows and columns than if I allow Bedouin only in the rows and drop the Bedouin column, because Bedouin has a very long drift path with itself, and thus including Bedouin in the columns will cause the Bedouin contribution in Moroccan to decrease, as a smaller quantity of Bedouin ancestry than is expected from just deep ancestry already creates the 'right' distance from Bedouin in Moroccan. This causes the 'big six' estimation for Moroccans in mixes with Bedouin both in the rows and columns to be quite divergent from mixes where only ancients are allowed in the rows, or Bedouin is in the rows but not the columns. C.f. the 'distortion' we are talking about. The same for many other comparisons. This should be expected to influence other ancient populations as well; at least, those that contribute heavily to any subset of the modern populations in the columns to the exclusion of other columns.

I have a suspicion that this tendency to discriminate very minor drift paths makes your tool far more powerful than I thought originally, haha. Which explains its capacity to create nice ADMIXTURE like clines when applied on a range of moderns.

Davidski said...

@huijbregts

What about instead of keeping only three dimensions, remove the lesser dimensions defined by just one or a couple individuals?

ryukendo kendow said...

For anyone who wants to work on the orthogonalized data instead of the raw data, here is the sheet:

http://textuploader.com/533tq


Feel free to drop any dimensions, play with the stats, etc.

Davidski said...

New datasheets with only the UDG treated Bell Beakers (hence, they shouldn't show minor Sub-Saharan) and Remedello BA, not UDG treated.

https://drive.google.com/file/d/0B9o3EYTdM8lQWldwcHZSa2hhZEE/view?usp=sharing

ryukendo kendow said...

@ huijbregts

If I use your workflow, where I only orthogonalize the list of source populations only, I still get this:

[1] "distance%=0.176 / distance=0.00176"


Bell_Beaker_Germany
"Yamnaya_Samara" 25.35
"Iberia_Chalcolithic" 22.65 <-----
"Hungary_BA" 13.25 <-----
"Anatolia_Chalcolithic" 10.15 <-----
"Armenia_MLBA" 8.8 <-----
"Motala_HG" 4.85
"Baalberge_MN" 4.2
"Samara_Eneolithic" 3.45
"Poltavka_outlier" 2.7
"Loschbour" 1.6
"Andronovo" 0.8
"Iberia_EN" 0.65
"Satsurblia" 0.45
"Moroccan" 0.45 <-----
"Villabruna" 0.3
"Yamnaya_Kalmykia" 0.2
"Iberia_MN" 0.05
"Israel_Natufian" 0.05 <-----

All zero populations ommitted.

I think the difference between our two runs is more due to my run having 26 dimensions and yours including 3, which causes information to be lost. After all, 3 dimensions on a PCA cannot be expected to represent even the big six without loss of information, on purely mathematical grounds.

Karl_K said...

@RK

"After all, 3 dimensions on a PCA cannot be expected to represent even the big six without loss of information, on purely mathematical grounds."

Absolutely. This is a complex situation, and probably with very little directly ancestral data, and a mix of quality.

Finding a best result can not be the same as when all things are linear and complete.

ryukendo kendow said...

In the other thread, Davidski has given me stats for the UDG treated Bell Beakers, which should not display the Sub-Saharan shift. Its true that the shift declines in magnitude, but a shift towards Levantines/Natufians/Africans continues to exist.


[1] "distance%=0.1848 / distance=0.001848"


Bell_Beaker_Germany
"Iberia_Chalcolithic" 18.95
"Hungary_BA" 17.95
"Yamnaya_Samara" 16
"Anatolia_Chalcolithic" 11.9
"Yamnaya_Kalmykia" 10.1
"Poltavka_outlier" 5.95
"Samara_Eneolithic" 3.5
"Motala_HG" 3.4
"Baalberge_MN" 3.35
"Loschbour" 2.55
"Armenia_MLBA" 2.45 <--------------
"Iberia_EN" 2.15
"Iberia_MN" 1.25
"Ulchi" 0.3
"Andronovo" 0.15
"Satsurblia" 0.05

To see whats hiding under the Armenia MLBA, I dropped it:


[1] "distance%=0.1897 / distance=0.001897"


Bell_Beaker_Germany
"Hungary_BA" 17.6
"Iberia_Chalcolithic" 15
"Yamnaya_Samara" 15
"Anatolia_Chalcolithic" 12.4 <--------------
"Yamnaya_Kalmykia" 10.7
"Poltavka_outlier" 8.4
"Baalberge_MN" 5.2
"Samara_Eneolithic" 3.3
"Motala_HG" 3.25
"Iberia_MN" 3
"Loschbour" 2.45
"Iberia_EN" 1.95
"Basque_French" 0.55
"Andronovo" 0.4
"Satsurblia" 0.4
"Ulchi" 0.3
"Masai_Kinyawa" 0.1 <----------------


Dropping Basque French:


[1] "distance%=0.1868 / distance=0.001868"


Bell_Beaker_Germany
"Iberia_Chalcolithic" 18.35
"Hungary_BA" 17.7
"Yamnaya_Samara" 15.05
"Anatolia_Chalcolithic" 12.75
"Yamnaya_Kalmykia" 11.7
"Poltavka_outlier" 7.3
"Baalberge_MN" 4.65
"Motala_HG" 3.25
"Samara_Eneolithic" 3.2
"Loschbour" 2.55
"Iberia_EN" 2.1
"Iberia_MN" 0.5
"Ulchi" 0.35
"Satsurblia" 0.25
"Andronovo" 0.2
"Levant_Neolithic" 0.05 <-----------------
"BedouinB" 0.05 <-----------------


Dropping BedouinB

[1] "distance%=0.1877 / distance=0.001877"

Bell_Beaker_Germany
"Hungary_BA" 17.85
"Iberia_Chalcolithic" 17.75
"Yamnaya_Samara" 15.2
"Anatolia_Chalcolithic" 12.55
"Yamnaya_Kalmykia" 11.1
"Poltavka_outlier" 7.6
"Baalberge_MN" 4.95
"Motala_HG" 3.25
"Samara_Eneolithic" 3.25
"Loschbour" 2.45
"Iberia_EN" 1.95
"Iberia_MN" 0.95
"Satsurblia" 0.35
"Ulchi" 0.35
"Andronovo" 0.25
"Corded_Ware_Germany" 0.1
"Moroccan" 0.1 <-----------------


ryukendo kendow said...


Dropping Moroccan:

[1] "distance%=0.1896 / distance=0.001896"


Bell_Beaker_Germany
"Hungary_BA" 17.5
"Iberia_Chalcolithic" 14.25
"Yamnaya_Samara" 14.1
"Anatolia_Chalcolithic" 12.8
"Yamnaya_Kalmykia" 11.4
"Poltavka_outlier" 9.5
"Baalberge_MN" 5.45
"Motala_HG" 3.3
"Samara_Eneolithic" 3.05
"Iberia_MN" 2.75
"Loschbour" 2.6
"Iberia_EN" 2.3
"Ulchi" 0.35
"Satsurblia" 0.25
"Andronovo" 0.15
"Corded_Ware_Germany" 0.15
"Masai_Kinyawa" 0.1 <-----------------

So a trace of African shift persists, which is created by either ancestry from highly Natufian people, or African ancestry; most likely highly Natufian people, as Moroccan, BedouinB, and Maasai seem to share this in common, while Esan_Nigeria or Yoruba-like ancestry is less of a common factor between them. This could have come from either the Southwest, from North Africa, or the Southeast, from the Middle East; the algorithm isn't quite clear on this.

Olympus Mons said...

Gosh!
I love denial!
Let me see f I get your: Kalr_K, Collin, etc , story right…

So now we all see the Iberia in Bell beakers, even on those as far away as “into” eastern Europe as the German bellbeakers…. Just imagine the others!
But, no! It’s not enough…steppe, steppe… Even if we already find all the hallmarks of a bell beaker in burials in the Tagus river by their kick-off 2900BC. Even if associated with bell beaker everywhere you find the same architecture as they had in Iberia, the same elite behavior as they had in Iberia, the same burials, the same a whole lots of things.. replacing what existed prior to their arrival. No, those brave steppe warriors just got to western Europe and dropped they yamnaya ways and embrace the southwest way of life… oh, I see , now I get it.

So we also know that along side with BB, and R1b for that matter, notice that actually they travelled and propagate Mtdna H1, which wait, has highest diversity in North Africa and, wait for it, wait for it… as Maju showed:
Reinoso~3700 0 (H1) ; Portalón~3000 0(H1) ; Mirador~2500 20 (H1)

See that 20 in mirador by bell beaker kick off? So… according to Karl-K and Collin… bunch of Iberian coming out of north Africa H1 women, just formed a gang (gosh and it had to be a lot!) and roamed out of Iberia to the arms of incoming Warriors from the steppe to give them the Iberian chalcolithic genetic component . Brilliant analysis. One got love ignorance . 

Oh wait, we also know that bell beaker horses and later bronze age horses trough out western Europe were actually the Portuguese lusitano hores (and some garrano)…. But wait, the horses also flee to the arms of the brave steppe warriors… Jesus H christ!

Olympus Mons said...

Something you all need to learn fast - Do not ask a scientist if its raining. better ask an idiot. the idiot will look out the window and tell if it is raining. A scientist will look at pressure instruments and so, and probably tell you that its a sunshine day.

Well regarding this issue of bell beaker people origins, let me tell you ... look out the fu++ing window. Its raining!

Karl_K said...

@Olympus Mons


"So… according to Karl-K and Collin… bunch of Iberian coming out of north Africa H1 women, just formed a gang (gosh and it had to be a lot!) and roamed out of Iberia to the arms of incoming Warriors from the steppe to give them the Iberian chalcolithic genetic component"

I am pretty positive you didn't read a single word of what I said, which is that the Bell Beakers that expanded out of Iberia were already largely steppe (Yamnaya-like) mixed with local Iberian Neolithic.

I even had a discussion with Rob about this very fact (see above).

I think you are now reading and posting while drunk.

Olympus Mons said...

@ryukendo kendow,
About the SSA component, people like me are intrigued by the possibility of having acquire (Chalcolithic Iberia) that SSA via L3 women in Egypt (broadly southeast Africa). Is it possible to include anyone with mtdna L3a? Maybe using PPNB Syria Tell Halula [H8] 6800-6000 BC by Fernández, E. et al. (2014)

Dont even know if that is available...

Olympus Mons said...

@Karl_K,
Sorry dude, sorry.. I meant ROB and Colin... Man, not drunk, but busy. Sorry again.

Olympus Mons said...

@ryukendo kendow,

Maybe you can help... is there a way of seeing if SSA existed in Iberia_MN or if it arrived with Iberia_Chal?

thank you.

Olympus Mons said...

@ Rob.
a steppe back? :)

No, no. I am all for wait for more samples. However:
1 - Steppe obsession is becoming pathetic. Steppe for you guys is Yamnaya, and when Yamanaya arise was after Maykop and those already were a mix popping after ubaid spreading DNA all over the region. So Steppe DNA does not exist. Yamnaya is EHG (local and south) CHG and Levant from South caucaus brought likely from fleeing R1b population from south Caucasus with M269 and L23 (some). - So, yes sample Europe and you will find steppe... but its just a proxy. By 3500 BC for intance you will find L23 in several eastern europe... but never L51. ever!

2 - Iberia up until 3500bc, was Maria, Manuel and carlos, living in caves, shelters and making shell middens. --Iberia calcholithic is over 100.000 -200,000 people arriving (Arrows and cattle). 3200 BC porto torrão had over 300HA (well over 20,000 people), Perdigoes (70Ha) pijotilla (100ha), san blas, etc. and then by 3000 was Zambujal were bell beaker is from.

DO you really thing that, say 200,000 people traveled from eastern europe without a trace in between??!

Rob said...

OM

Instead of a singular desire to prove a particular theory, you're better off taking a steppe back & let more data flow in. Its good to state hypotheses, but when one rants the same thing continuously, alarm bells start ringing.

Rest assured, if R1b came via Iberia, I'd welcome the finding.

But early BB Iberia will be just Middle Neolithic west European, +/_ some steppe; not north African. It's already evident

Karl_K said...

@Olympus Mons

Could you maybe write out a simple timeline for your ideas?

I think that many people don't quite get what you are saying.

Who came from where and when? Where did they start making Bell Beakers? What were the genetic components of the people? How did they spread?

I personally am quite confused by your posts. (perhaps because I am so drunk?)

Rob said...

Om

You've clearly misunderstood what Im suggesting

You are entirely correct about Iberia in 3200 BC: impressive developments take place. But, in your fixation to prove your pet theory, you exclude other regions and later periods from your analysis; and it is thus one dimensional instead of 4.

To begin with, there is little if any evidence for a migration from the south Caucasus via north Africa to Iberia in 3500 BC.
If you cant see this then your blind. Look at the Iberian samples we already have from 3500 - 2000 BC. Its 80% EEF and 20% WHG. Period

So there is impressive continuity in Iberia, and even movement out of it north. Yep
But its local stuff, not a new colonization you invented.

At some point steppe admixture arrived. Im not claiming it was a big conquest or whatever, but it did happen. There was significant transformation in Iberia c. 2200 BC, the end of enclosures, etc. This probably allowed pastoral -descended groups from central Europe to move in, even if opportunistically.

Its not rocket science.




Olympus Mons said...

@Karl_k,
Yes thank. bellow is a post just 2 days ago. see if it works
________

Not only do they had SSA, but they picked up with L3 women in Egypt (merimde and el-omari)

Recap "real" history from Shulaveri2BellBeaker.
1- 7th millennia in southern Caucasus as Shulaveri-Shomu, where M269 was born, apparently coming from Anatolia (because of cattle and goats DNA).
2- By 4.900 “they” were completely kicked out – by then they were a mix of EHG, CHG and Anatolia Neolithic. All their settlements were abandoned and some have a layer of ashes to the one that replace them (sioni going to Kura araxes) with different pottery, different architecture, etc.
3- By 4.800 BC they where in tell tsaf north Israel. So suppose the place they were kicked out to by Ubaid or L1a from Iran, was to west and that is why the 2 places with higher variance of r1b is the eastern Anatolia and …. The place in Armenia where that r1b was just found, near sevan lake.
4- By 4.700 BC they were in Nalchik north Caucasus, and so forth that is why Yamnaya is so close to bell beaker. They, the Shulaveri, diluted the EHG in them and gave them CHG and Levant DNA.
5- By 4.700 BC were settling heavily in Merimde and el-Omari in the Nile delta in Egypt, and having cattle binge parties in Fayum, near the lake. Was L51 born there?
6- 4.000 BC Again as the same as with Ubaid, the pre dynastic pharaonic Egypt with the crazy Badarian on south Egypt moving north, kicked them out.
7- By 3.700 BC were arriving to Iberia, kicked out by the 5.9 kiloyear climatic event that made the Sahara desert, along side with berber (E1b1) guys.
8- By 3.300 were amassing in large cities in Iberia, porto torrão in the lowlands of Iberia as big as Ur city.
9- By 3,000 where building the Zambujal city where the bell beaker actually arised.
10- By 2.700 BC had crossed the pyrenes. … And that is the bell beaker story.

Isn’t it what the DNA is telling? – At least they told every one. When Periplus, the 700BC greek mariner met them there, in Lisbon, They told him, who they were.
“We are the people had have been living here for a long time, but were kicked out of our homeland (southern caucasus) by an attack of serpents (Ubaid/uruk). … We are the Oestrimnis!

See chapter – Those o fled the serpents.
http://shulaveri2bellbeaker.blogs.sapo.pt/suppl-i-they-who-fled-the-serpents-5061

Karl_K said...

@Rob

"Im not claiming it was a big conquest or whatever, but it did happen."

No doubt. New people with a still mysterious new advantage.

What are the options here? It looks like R1b came out of Iberia tightly linked to lactase persistence. Also there are the metallurgy advances. Likely some animal husbandry advances.

What is the significance of the Beaker, so that someone would want to be buried with one? Beer? Milk?

Karl_K said...

@Olympus Mons

"Isn’t it what the DNA is telling?"

I will read it again more carefully, but no. You have only a few data points and are making really huge leaps based on your assumptions.

Olympus Mons said...

Karl_K,

"It looks like R1b came out of Iberia tightly linked to lactase persistence."

Well by the time my Merimde guys were in Delta nile --- They were making yogurt and drinking milk?

http://www.nature.com/news/pottery-shards-put-a-date-on-africa-s-dairying-1.10863

Olympus Mons said...

@Karl_K,
yes, lots of assumptions... but highly sustained in my thesis. Highly.
Anyways, assumptions is what we do here by the ton.

Rob said...

@ Karl

"What are the options here? It looks like R1b came out of Iberia tightly linked to lactase persistence"

Quite possibly (although I haven;t looked into that specifically as others have).
But the reason why steppe admixture reached Iberia is simple inevitability.
If it's in France (eg sites like Sioni), then it'll make its way to Iberia as part of reciprocal movement of individuals & groups. Thus, i don't think it rests of one special attribute or formula to success, although BB clearly were elites, I suspect this network might have consisted - at least initially by different groups, and perhaps culminated by eventual dominance of one (i.e P312).

But that's all we can say now. We need to discover the actual timing and rhythm of the growth (ie more aDNA) to zero in on the social mechanisms. We don;t know how rapidly Iberia changed from predominantly H2 and I2 to R1b, do we ? But I think the key is that 80% plus of R1b in Iberia, is DF 27, so it looks like a founder effect, IMO, late or post-Beaker.

Olympus Mons said...

@ROB,
"If you cant see this then your blind. Look at the Iberian samples we already have from 3500 - 2000 BC. Its 80% EEF and 20% WHG. Period"


Don't treat Iberia as a monolithic. Sites sampled so far are not Bell beakerish as far as I know and far way from the Bell beaker route.
this is small part from my thesis...

"Augmented population headcounts at these sites coinciding with the timeframe when, down south, was the arriving of the Shulaveri and Berbers (imo), means most likely one thing: Run north from the incoming intimidating others and local populations had no option but to move northbound around the Serra da Estrela mountain range, keep running until you bump into the other huge Iberia River. The Douro River. There, where you can’t cross is the Coa Valley where humans between 3500 BC and 2000 BC cluster (again *533). Works done in the last 50 years in the area clearly show this spurs of population increase at times. Whatever words one uses to characterize the events that englobes and defines Chalcolithic Portugal it cannot be applied there, even later periods of bell beaker, social complexity and so forth. Is not there. Ever. They were left to be, serenely has they were, because the bell beakers moved southbound the Serra da Estrela Mountain range, being that the reason why Madrid is south of Cuenca Alta del Manzanares and actually are so genetically closed to Central area Portuguese. It’s the Bell Beaker route and I imagine they didn´t detour much from it. Coa Valley will most likely be still a cornerstone in understanding the population transitions in Iberia Chalcolithic, precisely because they, the people which lived there for a millennia where actually not that chalcolithic anyway...

So,
to Find steppe...
Sample - elite burials in Porto torrao and Perdigoes (south Portugal), Zambujal near Tagus river (Portugal) and 3000 - 2000 BC south of Serra da Estrela Mountain and south of Cuenca Alta del Manzanares moutain range. - Outside this Bell beaker route... you will find lots of Neolithic (EEF & WHG) and also lots of North Africa (Prot-berber) E1b1.

Olympus Mons said...

@Davidski,
Regarding your update today (30 Jun), My first comment on this thread was to be:

"Give Davidksi time enough and he will torture the data enough to make it all go way"

:)

Gioiello said...

Genetiker says that the R1b from Kura-Araxes is an R1b1a1b positive for V1956 (Y: 8430895 G>A)
Sample ID HG 8430895
REFSEQ G

Also the sample from 1KGP from Puero Rico is positive for V1956 and all the subclades from R-P297 are neagtive
HG00640 R-L389* A
HG00640 20120522 PUR R-L389*
L388/PF6468 • L389/PF6531 B2b

The sample from Puerto Rico is very likely of the Iberian R-L389 haplotypes and all the other known R-L389 which are most diffused and with the highest variance in Italy, thus the Caucasian sample from Kura-Araxes, as I am saying from so long, belongs very likely to the Caucasian R-L389 with YCAII=23-23 which is presupposed derived through a RecLOH from the basal 18-22 and 18-23 diffused in Italy, which are a sister clade of the R-L389 ancestor of R-P297. But the Villabruna sample of 14000 years ago has two certain mutations at the P297 level, thus it is more likely of the family which brought to the R-P297* and subclades. Anyhow no doubt that my theory of an “Italian Refugium” adds another proof. Let's go on.

Fanty said...

Wasnt there those mtDNA comparations of Bellbeaker, Cordedware and UNetice, some years or so ago?

That had claimed German bellbeaker mtDNA matches best with modern Spanish mtDNA and German Corded Ware and Unetice mtDNA would match best with modern Russian mtDNA? (while it also claimed that modern Russian mtDNA does better match with modern Spanish than with Corded one)?

Karl_K said...

That seems familiar.

huijbregts said...

@ Ryukendo, Davidski
In a slightly modified sample the first 5 dimensions explained 99.4% of the variance.
Yet a model with only 5 dimensions returned a result which was clearly different from the full model.
I think I understand what is the problem. With a small sample we have more columns then rows.
So the PCA is overspecified and each row gets its own Principal Component.
That is not what we want a PCA to do.
I have to think about the consequences. My first idea is that we should limit the dimensions to 5 and not add spurious score-columns.
This is a problem which is relevant for the Bell Beaker task, for the most striking difference between 5 dimensions and full model is exactly in the row Iberia_Chalcolithic: 2.45% and 19.4%.

Krefter said...

@fanty,

mtDNA affinity between populations is a complex thing. There's no way to measure it in a meaningful way. mtDNA studies wrongly assume there is and take affinity results from their testing methods seriously. Our Bell Beaker samples have lots of H and I'm pretty sure that's the reason they associated them with Iberia, even though Irish or French or Germans have as much H as Iberia.

Poise n Pen said...

"You're not making any sense.

Bell Beakers and Corded Ware have a lot of steppe ancestry"

In fantasyland where correlation implies causation, sure. If we know anything at all about science then we don't make assumptions that this is proven, and we don't read a paper by some heavily biased idiots and think their explanation is an actual proven fact instead of just one possible explanation.


"They're also the ancestors of modern Northern Europeans. If you don't get this by now, then get another, less demanding hobby."

You seem to be incapable of understanding that a relationship doesn't imply ANCESTRY. No one should be using the term ancestry because none of this actually shows ancestry in any way.

My cousin and I are highly related, which one is the ancestor of the other? Such a joke.

It's even funnier because the r1a heavy corded ware shows more steppe ancestry, when they are west of the yamnaya people and later in time. And there is evidence that Maikop was founded by Funnel Beaker culture, and the earliest evidence of a wheel has been found there recently.

There's two possible realities. Modern europeans' ancestors were savages who didn't invent anything, but they were native to much of the area they are already in... or they basically invented everything somehow, and then conquered europe completely in prehistory but expanded only westward nowhere else.

But the politically correct story is a third and laughable option. They were savages who didn't invent anything, stole it from elsewhere, then migrated to some completely different area.

Ok, let's say that's true then. So they did not domesticate cows for example then - so why are their descendents the ones who are lactose tolerant?! Why did they expand only west?! Why, if they developed among a bunch of r1a people, are r1b and r1a not found together everywhere today?!

Amazingly nonsensical and impossible crap yet everyone swallows it whole.

Onur said...

Thanks for your update, Davidski. So it seems there is something real between Chalcolithic Iberians and Bell Beakers. The mystery will probably be further solved when we have sequences from early Bell Beakers from Iberia.

Davidski said...

Yeah, it's not necessarily from Iberia though. I need to see what happens to this tree when I add the Swedish Gokhem TRB farmers. Hopefully they have enough markers.

Rob said...

For interests sake, can you include Czech BB & Rathlin, if enough markers
To see if there are any discernible variations in the broader BB network ?

Davidski said...

I don't have Rathlin in this dataset, because it was genotyped in a different way and shows slightly skewed results in D-stats sometimes. And the Czech Beakers don't have enough markers for this tree.

Onur said...

@Davidski

Yeah, it's not necessarily from Iberia though.

Sure.

I need to see what happens to this tree when I add the Swedish Gokhem TRB farmers. Hopefully they have enough markers.

That is a good idea. You can add other Neolithic/Chalcolithic farmers as well.

Ryan said...

When the hell are we going to get Bell Beaker samples from Iberia? :/

Fanty said...

"Why, if they developed among a bunch of r1a people, are r1b and r1a not found together everywhere today?!"

Ever heard of "drift"?

It can be exremely quick in small populations. Like say, neolithic tribes?

Also, there was a "smoking gun" already. We have R1a and R1b eastern hunter gatherers for example.

Gioiello said...

@ Fanty

"We have R1a and R1b eastern hunter gatherers for example".

The oldest R1b hunter gatherer (WHG) is Villabruna, Belluno, Italy, 14000 YBP.
What will you say if also an R1a will be found there?

Grey said...

Karl_K

"New people with a still mysterious new advantage.

What are the options here? It looks like R1b came out of Iberia tightly linked to lactase persistence. Also there are the metallurgy advances. Likely some animal husbandry advances.

What is the significance of the Beaker, so that someone would want to be buried with one? Beer? Milk?"

I imagine copper weaponry might give an advantage over stone but not sure how big but I think the first people to get their hands on bronze swords would have had a massive advantage until everybody else caught up and the first people to get bronze swords would be the people who made them.

Then again *if* the same thing happened in Armenia/Anatolia it might show a metallurgical advantage may not have been enough on its own to completely transform an originally larger population - so i wonder if there were two advantages along the Atlantic coast: metallurgy and dairying.

#

la señora bibiloni said...

"But how much Yamnaya is there in Iberian_Chalc? Do we find "Yamnaya" in Iberian_Chalc? Or just South-of-the-Caucasus stuff with a bit of EHG?"

I think the key point is the R1b. *If* the Iberian R1b was originally from the steppe and the ydna is mismatched with adna then that just implies there was a movement of males with some kind of advantage.

If the Iberian R1b is not from the steppe then the explanation is something different.

Karl_K said...

@Davidski

"I don't have Rathlin in this dataset, because it was genotyped in a different way and shows slightly skewed results in D-stats sometimes. And the Czech Beakers don't have enough markers for this tree."

Have you ever tried using imputation to fill in the missing SNPs?

It seems like Bronze Age Europeans would most likely work fine for imputation from statistics based on modern Europeans. Much older samples may or may not get some wrong SNPs stuck in.

Davidski said...

Nah, imputing markers into ancient samples, any ancient samples, would give too much ammunition to the crackpots.

Karl_K said...
This comment has been removed by the author.
Davidski said...

Their ascension numbers at the ENA should be listed in the relevant papers.

Olympus Mons said...

@Davidski,
As perdicted by me...
"Yeah, it's not necessarily from Iberia though. I need to see what happens to this tree when I add the Swedish Gokhem TRB farmers. Hopefully they have enough markers."

I have said it. :)

Regarding your update today (30 Jun), My first comment on this thread was to be: "Give Davidksi time enough and he will torture the data enough to make it all go way"

A: What does the statistics say?
B: what do you want it to say?



Ariele Iacopo Maggi said...

Bell_Beaker_Germany
"Remedello_BA" 49.1
"Yamnaya_Samara" 43.2
"Loschbour" 7.2
"Han" 0.5
"Iberia_Chalcolithic" 0
"Yamnaya_Kalmykia" 0
"Armenia_Chalcolithic" 0
"Hungary_CA" 0
"Hungary_EN" 0
"Satsurblia" 0
"Ami" 0
"Iberia_EN" 0
"Armenia_MLBA" 0
"Cypriot" 0
Remedello is strong with this one... Is everything all right with the remedello stats?

Olympus Mons said...

@Ariele Iacopo Maggi,

Is this Remedello_BA meaning the guy from 2.100BC, meaning he is just replacing Iberia_chalcolithic because he is full of it, or is it the all 3 samples of Remedello including the one from 3000BC ... which actually can really also be full of same material as Iberia_chalc.

First lets see what is Remedello 1, 2, 3. What are each.
Does anyone has it?

Hey, Is alberto around?
:)

Olympus Mons said...

And lets be clear...
If what is being running against on this stats is a single Remedello guy (RISE486) from 2100 and showing that he is best match for a bell beaker from 2500BC not that further north. - Its a slight of hand to throw people off the Iberia connection. Pure and simple. And Ugly. Hope that is not the case.

Colin Welling said...

@davidski

Does the fact that the migration edge, leading from farmers and feeding into bell beakers, occurs so close to the base of spanish chalcolithic mean that the farmer contribution to bell beakers was not simply spanish chalcolithic? Could the migration edge have occurred further up the spanish chalcolithic line?

Ariele Iacopo Maggi said...

Olympus Mons

Yes, I think this is a late Remedello with direct BB or central european ancestry, that shouldn't contradict the Iberian hypothesis.
With the same pops:
----Iberia_Chalcolithic
"Anatolia_Neolithic" 77.5
"Loschbour" 18.7
"Villabruna" 2.85
"Han" 0.95
----Remedello_BA
"Anatolia_Neolithic" 60.7
"Yamnaya_Samara" 17.9
"Hungary_HG" 10.8
"Motala_HG" 5.55
"Loschbour" 4.25
"Karitiana" 0.8
----Iberia_EN
"Anatolia_Neolithic" 85
"Hungary_HG" 13.95
"Karitiana" 1.05
----Hungary_EN
"Anatolia_Neolithic" 89.85
"Motala_HG" 5.35
"Hungary_HG" 4.55
"Dai" 0.25
----Iberia_MN
"Anatolia_Neolithic" 69.3
"Hungary_HG" 22.25
"Loschbour" 4.8
"Yamnaya_Samara" 1.65
"Mansi" 1.05
"Karitiana" 0.95

Fanty said...

@Gio:

"The oldest R1b hunter gatherer (WHG) is Villabruna, Belluno, Italy, 14000 YBP.
What will you say if also an R1a will be found there?"

First we need to see how much R1b there are, in Mesolithic Italy.
After all there are "J" EHG too, but that will not be a major Y-DNA in EHG I believe.

Olympus Mons said...

@Ariele,
I still would like Davidski and or Ryu to get out of their White horses and tell me that to throw people of the Iberian source of bell beakers, they didn't, didn't actually picked up a almost german bell beaker (RISE486) from 2100 BC im Remedello and say: look how it actually is so close to the German bell beaker from 2500BC.

Obviously this blog (as all others) is not about the truth or something, but about people doing ingroup and circuling the wagon. Which is fine. Its all fun.
It just makes more comfortable, if that is the case, the next time they pretend to be arbinders of anything else them theirs fantasies I can sent them all to a place I know. :)

Olympus Mons said...

@Ariele,
Do you think is possible to see how the 3 remedello samples evolve over the milenia?

Ariele Iacopo Maggi said...

I think so. Hopefully sooner or later we will know.

Davidski said...

@Colin

Does the fact that the migration edge, leading from farmers and feeding into bell beakers, occurs so close to the base of spanish chalcolithic mean that the farmer contribution to bell beakers was not simply spanish chalcolithic? Could the migration edge have occurred further up the spanish chalcolithic line?

If I understand your question correctly, then yeah, it seems that when TreeMix produces an edge from the base of a branch, then the tip of the branch is not the ideal reference sample.

What this might mean is that Iberia Chalcolithic did not contribute admixture to German Bell Beakers, but rather a related population did. But this population was more closely related to Iberia Chalcolithic than to Germany MN, so it's unlikely that it was a local population.

It looks like that one way or another German Beakers experienced a pulse of admixture from the Atlantic region.

Oinkselot said...

Oh really? Well, the Mongols crashed far more refined civilizations of China and India because they "invented" one single thing – horse archers. Germanic tribes destroyed Roman civilization, much more advanced than their own, and then collectively moved to its former territories. Linguistics, archaeology and genetics converge and lead to the same point that you put so nicely and accurately: “They were savages who didn't invent anything, stole it from elsewhere, then migrated to some completely different area.” That happened so many times in history that it seems quite plausible. All your efforts to “squeeze” them among the culture creators, such as “out of Anatolis”, “out of India” and “out of anywhere else but where they actually came from” quasi scientific theories, are futile. You know, it’s a clear case of cosmic justice – that the descendants of those who once enslaved half of human kind with the justification of “civilizing” them, now have to deal with their own barbarian origins. Of course, you may always choose to stick your head into the sand and live in your own version of reality. Just deal with it already.

Simon_W said...

@ Ariele Iacopo M., re: Anatolia Chalcolithic's affinities

Interesting, thanks for posting. I think it's related to what Harrison and Heyd wrote in 2007, that first, after 3000 BC, the eastern Aegean was brought into a cultural relationship with the early urban societies of the Near East, that around 2600/2500 BC the Cyclades, mainland Greece and the southeastern Balkans were incorporated into this wider Near Eastern exchange network, and that finally, between 2600-2200 BC, the central Mediterranean with Apulia, Sicily and Malta was drawn into this world.

The huge admixture proportions with Near Eastern/Cypriot-like people inferred for Italians based on linkage disequilibrum patterns is probably most of all simply the mixture between ancient South Italians and ancient North Italians. It makes sense that this happened at a relatively late date, when all the distinct ethnic groups of ancient Italy were unified under Roman rule. And North and South Italians are very distinct from each other, almost like different populations, yet this doesn't mean that they didn't mix with each other. They obviously mixed, and central Italy is a transitional area. The ADMIXTURE analysis in Fiorito et al. 2015 showed nicely how the red southern component is also strongly present in the north and the green northern one in the south. I'm not saying there was no Levantine immigration in the Roman age, but this probably had just a minor, more subtle influence.