PDA

View Full Version : Genome diversity in the Neolithic Globular Amphorae culture and the spread o IE



Olympus Mons
13-11-17, 17:18
Supplementary material from "Genome diversity in the Neolithic Globular Amphorae culture and the spread of Indo-European languages"
Published on 13 Nov 2017 - 13:07

It is unclear whether Indo-European languages in Europe spread from the Pontic steppes in the late Neolithic, or from Anatolia in the early Neolithic. Under the former hypothesis, people of the Globular Amphorae Culture (GAC) would be descended from Eastern ancestors, likely representing the Yamnaya Culture. However, nuclear (six individuals typed for 597 573 SNPs) and mitochondrial (11 complete sequences) DNAs from the GAC appear closer to those of earlier Neolithic groups than to the DNAs of all other populations related with the Pontic steppe migration. Explicit comparisons of alternative demographic models via Approximate Bayesian Computation confirmed this pattern. These results are not in contrast with Late Neolithic gene flow from the Pontic steppes into Central Europe. However, they add nuance to this model, showing that the eastern affinities of the GAC in the archaeological record reflect cultural influences from other groups from the East, rather than the movement of people.

Olympus Mons
13-11-17, 17:19
Indeed.

a. ....GAC appear closer to those of earlier Neolithic groups than to the DNAs of all other populations related with the Pontic steppe migration.
b. ....GAC in the archaeological record reflect cultural influences from other groups from the East, rather than the movement of people

Promenade
13-11-17, 18:54
Yes if I remember correctly a past paper even went as far as to refer to the GAC as an impenetrable genetic wall in between western neolithic Europe and eastern steppe, for some reason there was cultural diffusion without genetic contact. It's silly of them to state that GAC not having steppe ancestry has anything to do with IE languages coming to Europe from Anatolia though, especially northern Europe at that, you'd think they'd know better than that by now.

Angela
13-11-17, 21:42
Their conclusions indeed make no sense, imo.

What it shows in terms of ancient dna, however, is how it can overturn previous ideas. People argued about this for years. Now we know: Gimbutas was wrong, and these people were farmers.

New Englander
13-11-17, 21:59
So we are back to the R1b out of Anatolia, and R1a out of the Stepp? Or is is that Rs were not Indo European speakers? Seems like it is saying that the Farmers were Indo European speakers, and that population adsorbed East Eurasian tribes?

Olympus Mons
22-11-17, 17:09
Paper Out
http://rspb.royalsocietypublishing.org/content/284/1867/20171540

Looks like a lot more precise than the abstract!

Angela
22-11-17, 19:51
http://rspb.royalsocietypublishing.org/content/284/1867/20171540

Genome diversity in the Neolithic Globular Amphorae culture and the spread of Indo-European languagesFrancesca Tassi, Stefania Vai, Silvia Ghirotto, Martina Lari, Alessandra Modi, Elena Pilli, Andrea Brunelli, Roberta Rosa Susca, Alicja Budnik, Damian Labuda, Federica Alberti, Carles Lalueza-Fox, David Reich, David Caramelli, Guido Barbujani


Beware of coming to conclusions based just on abstracts I said to myself. :) The kurgan hypothesis is alive and well. David Reich would hardly put his name to it were it otherwise.

" There is now genetic evidence of population movements from the Russian steppes into Central Europe in the Bronze Age [20 (http://rspb.royalsocietypublishing.org/content/284/1867/20171540#ref-20),21 (http://rspb.royalsocietypublishing.org/content/284/1867/20171540#ref-21)] and Iron Age [22 (http://rspb.royalsocietypublishing.org/content/284/1867/20171540#ref-22)]. These processes may or may not have had large-scale consequences at the demographic and linguistic level, but the later expansion would be consistent with a spread of languages associated with the Kurgan hypothesis [18 (http://rspb.royalsocietypublishing.org/content/284/1867/20171540#ref-18),21 (http://rspb.royalsocietypublishing.org/content/284/1867/20171540#ref-21)]."

" we collected and typed samples of 17 individuals from the Megalithic barrow of Kierzkowo (Poland), which is archaeologically assigned to the Globular Amphorae culture (GAC). The GAC is documented in Central and Eastern Europe, from the Elbe to the middle Dnieper, around 5400–4800 BP. It plays a crucial role in this debate because it has been argued to be associated with the first Indo-European migrations based on its burial rituals, including burial of livestock, usage of domestic horse, and presence of amber sun-disks [1 (http://rspb.royalsocietypublishing.org/content/284/1867/20171540#ref-1)]. Gimbutas [23 (http://rspb.royalsocietypublishing.org/content/284/1867/20171540#ref-23)] argued that when the Kurgan culture expanded from its homeland in the steppe and forest-steppe of Ukraine and South Russia, it did so in three waves, thus leading to the diffusion of the IE languages (see electronic supplementary material, figure S1). The GAC people are regarded by Gimbutas as part of the first wave, associated with the spread of the Yamna culture from the Pontic region to the Danube basin and the Balkans, between 5100 and 4900 BP [1 (http://rspb.royalsocietypublishing.org/content/284/1867/20171540#ref-1)]. If Gimbutas' theory is correct, the people of the GAC should have Yamna related admixture, as well as genetic affinity to the populations associated with the later, Bell Beaker culture, documented in many areas of Europe 4800 to 3800 BP."

"The emerging picture is thus one in which migrations from the Pontic steppes into Central Europe left a trace in the genomes of the Corded Ware culture, but not in those of the GAC."

The supplementary data can be accessed here, and includes the mtdna results not only for Globular Amphora but for all the ancient samples they used for comparisons.
http://rspb.royalsocietypublishing.org/content/royprsb/suppl/2017/11/21/rspb.2017.1540.DC1/rspb20171540supp1.pdf

Angela
22-11-17, 20:16
This is the Admixture graph. I like it because it contrasts AN with Yamnaya.
9457


This is the PCA. Do Globular Amphora land on top of Tuscans or am I misreading it? It seems that some Iberian Chalcolitic also lands there, which if true validates the point that was made two years ago or more that modern Tuscans are similar to Iberia Chalcolithic, and which I endorsed, and which was ridiculed by the usual suspect.

Click to enlarge.
9458

Olympus Mons
22-11-17, 20:18
What I read is things like...

""and we ran some preliminary analyses on it. In the neighbour joining (NJ) tree inferred from the ϕST pairwise distances estimated for this subset, the Early Bronze Age people, represented by the Srubnaya culture, appear connected with the eastern Corded Ware peoples, and also close to the Yamna. The GAC samples are clearly separated from those populations, and show instead a closer relationship with the western, Late Neolithic, Bell Beaker population (electronic supplementary material, figure S12)."

Angela
22-11-17, 20:27
^^Obviously, since Bell Beaker has more MN than Corded Ware. It doesn't change the conclusions of the paper.

Olympus Mons
22-11-17, 20:35
The conclusion of the paper is that Yamnaya ans steppe is becoming smaller every day!
Time will tell how small.

Some traits in culture seemed to indicate PIE for a very long time... but suddenly when aDna does not show Steppe in them, the story becomes complicated.

Truth is a couple samples from forsaken caves in Iberia, not even from bell beaker proper, or even from possible related cultures (like VNSP) is enough for "them" to declare the death of bell beaker origin in Iberia.... but this paper already has more "but" and "If" then a child's play.

Angela
22-11-17, 21:56
You're exaggerating. The original Haak paper made a point of saying that it was talking mainly about Central and Northern Europe. The impact of the steppe genes on those populations is clear and irrefutable.

Later academic papers have shown that this caution was wise as there is much less steppe in Greece even after the Slavic Migrations. Spain also has less, and not just because of any possible subsequent gene flow in early Medieval times. We'll see about Italy.

It's only one trick pony internet "experts" who have promulgated the "Indo-Europeans" for dummies version, not the academics.

bicicleur
22-11-17, 21:58
The conclusion of the paper is that Yamnaya ans steppe is becoming smaller every day!
Time will tell how small.

Some traits in culture seemed to indicate PIE for a very long time... but suddenly when aDna does not show Steppe in them, the story becomes complicated.

Truth is a couple samples from forsaken caves in Iberia, not even from bell beaker proper, or even from possible related cultures (like VNSP) is enough for "them" to declare the death of bell beaker origin in Iberia.... but this paper already has more "but" and "If" then a child's play.

I think that this paper confirms what the Mathieson 2017 study already told about GAC.
But I don't understand how you can extrapolate this to other cultures and populations.

bicicleur
22-11-17, 22:08
You're exaggerating. The original Haak paper made a point of saying that it was talking mainly about Central and Northern Europe. The impact of the steppe genes on those populations is clear and irrefutable.

Later academic papers have shown that this caution was wise as there is much less steppe in Greece even after the Slavic Migrations. Spain also has less, and not just because of any possible subsequent gene flow in early Medieval times. We'll see about Italy.

It's only one trick pony internet "experts" who have promulgated the "Indo-Europeans" for dummies version, not the academics.

most new studies somehow confirm older studies
the truth just becomes every time a little bit more complicated and less straightforward

A. Papadimitriou
22-11-17, 22:51
The titles of W Haak studies are misleading (not supported by the data or the content of the studies themselves).

Angela
22-11-17, 23:33
The titles of W Haak studies are misleading (not supported by the data or the content of the studies themselves).

Yes, I noted at the time on my comments on this site that it seemed as if the title and the body of the paper, and especially the supplement, were written by different people. We know who did the computation population genetics analysis and that's what counted, then as now.

As I also noted more than once, the authors may come to regret that title.

Olympus Mons
22-11-17, 23:41
Oralde and Martiniano simultaneously was not innocent. Making a 26 year old Portuguese "kill" Portuguese BB centrality was not by chance.

ROS
22-11-17, 23:43
The titles of W Haak studies are misleading (not supported by the data or the content of the studies themselves).

You could expand more on the W Haak titles are deceptive for amateurs like me.

This is increasingly similar to the story of the Lord of the Rings.

Let's see Angela this graph that from time to time you present us is from this same W Haak that you are questioning up to the titles?


And on the other hand something that I do not understand as Spain has 0% WHG ?, this really can not be true or only compressible with a very partial study and not correlated with reality.

Pax Augusta
22-11-17, 23:54
This is the Admixture graph. I like it because it contrasts AN with Yamnaya.
9457


This is the PCA. Do Globular Amphora land on top of Tuscans or am I misreading it? It seems that some Iberian Chalcolitic also lands there, which if true validates the point that was made two years ago or more that modern Tuscans are similar to Iberia Chalcolithic, and which I endorsed, and which was ridiculed by the usual suspect.

Click to enlarge.
9458


Usual suspect has been very active today.

Angela
23-11-17, 00:43
Usual suspect has been very active today.

Ah, I was talking about the usual suspect who has his own blog. He was adamant that modern Tuscans are nothing like Chalcolithic Iberians. It was a terrible study and PCA and should be forgotten :) Wrong again; what else is new when it comes to him and Southern Europe?

As for the mini-usual suspect, is he quoting (without attribution) and twisting my ideas as he's been doing since 23andme days? :)

@Ros,
I'm afraid you've gotten confused. I don't disagree, in general terms, with the computational population genetics analysis in the paper or the explanation of it, for that matter, which I would bet was done and written by Lazaridis. No one with any standing in academia has found any fault with its general parameters, although newer data has refined certain things around the edges, and more data might do the same. Even in 2015, though, the data analysis and the essays clearly show that the steppe input is much smaller in southern Europe than in Central and Northern Europe.

What I objected to then and now is the title, which implies there was "massive migration" from the steppe to all parts of Europe. It was an overstatement and it doesn't fit with the body of the paper itself, as others upthread have also pointed out, and I would bet it wasn't Lazaridis' idea.

Of course, anyone whose posts are worthy of being read knows that the body of the paper itself makes no such claim. As you immerse yourself in the data and the comments on it, you'll discover whom it is best to ignore.

It is certain internet bloggers and posters who have promoted a sort of "Indo-Europeans for dummies" version of the data since the beginning and continue to do so. You have to be careful whom you read. Even in the very early days of 2013 and before some of us were aware that certain archaeological papers were being ignored, and that certain dna data was being massaged.

Sarcasm and misplaced analogies are not going to change hard genetic data, I'm afraid.

The Spanish have WHG, as other analyses by Lazaridis have shown. On this particular graphic it is all hidden in the EN, which is based on Central European Neolithic farmers like Stuttgart, who are about 94% Anatolian farmer and 6% WHG. People like Spanish Basque have more. This was all discussed on the original thread for the HaaK 2015 paper.


Here are the actual percentages. Tuscans have zero on this too in this particular model. This isn't engraved in tablets of stone like the one Moses brought down from the mount. The percentages will change a little bit based on the particular ancient samples available. However, the parameters are as you see them. No new samples have changed the general picture.

Haak graphic EN WHG Yamnaya
Norway - 30, 16, 54
Lithuania - 18, 30, 52
Estonia - 12, 37, 51
Iceland - 32, 19, 49
Scotland - 28, 23, 49
Czech - 35, 16, 49
Belarus - 25, 28, 47
Hungary - 39, 16, 45
Ukraine - 28, 27, 44
England - 44, 14, 42
Orkney - 34, 25, 41
South French - 57, 4, 39
Croatia - 44, 17, 37
French - 51, 12, 37
North Spanish - 59, 10, 31
Bulgaria - 55, 14, 31
Tuscany - 72, 0, 28
Basque - 54, 19, 27
Bergamo - 63, 13, 24
Spain - 78, 0, 22
Greece - 66, 14, 20
Albania - 65, 18, 17
Sardinia - 88, 7, 5

Remember, at least 40% of most of those steppe numbers are Caucasus or "southern" or heavily "Basal Eurasian" in origin.


In the supplement there's extensive modeling which tries to get a better fit. You really should read it for yourself.

Angela
23-11-17, 01:58
Ah, I was talking about the usual suspect who has his own blog. He was adamant that modern Tuscans are nothing like Chalcolithic Iberians. It was a terrible study and PCA and should be forgotten :) Wrong again; what else is new when it comes to him and Southern Europe?

As for the mini-usual suspect, is he quoting (without attribution) and twisting my ideas as he's been doing since 23andme days? :)

@Ros,
I'm afraid you've gotten confused. I don't disagree, in general terms, with the computational population genetics analysis in the paper or the explanation of it, for that matter, which I would bet was done and written by Lazaridis. No one with any standing in academia has found any fault with its general parameters, although newer data has refined certain things around the edges, and more data might do the same. Even in 2015, though, the data analysis and the essays clearly show that the steppe input is much smaller in southern Europe than in Central and Northern Europe.

What I objected to then and now is the title, which implies there was "massive migration" from the steppe to all parts of Europe. It was an overstatement and it doesn't fit with the body of the paper itself, as others upthread have also pointed out, and I would bet it wasn't Lazaridis' idea.

Of course, anyone whose posts are worthy of being read knows that the body of the paper itself makes no such claim. As you immerse yourself in the data and the comments on it, you'll discover whom it is best to ignore.

It is certain internet bloggers and posters who have promoted a sort of "Indo-Europeans for dummies" version of the data since the beginning and continue to do so. You have to be careful whom you read. Even in the very early days of 2013 and before some of us were aware that certain archaeological papers were being ignored, and that certain dna data was being massaged.

Sarcasm and misplaced analogies are not going to change hard genetic data, I'm afraid.

The Spanish have WHG, as other analyses by Lazaridis have shown. On this particular graphic it is all hidden in the EN, which is based on Central European Neolithic farmers like Stuttgart, who are about 94% Anatolian farmer and 6% WHG. People like Spanish Basque have more. This was all discussed on the original thread for the HaaK 2015 paper.


Here are the actual percentages. Tuscans have zero on this too in this particular model. This isn't engraved in tablets of stone like the one Moses brought down from the mount. The percentages will change a little bit based on the particular ancient samples available. However, the parameters are as you see them. No new samples have changed the general picture.

Haak graphic EN WHG Yamnaya
Norway - 30, 16, 54
Lithuania - 18, 30, 52
Estonia - 12, 37, 51
Iceland - 32, 19, 49
Scotland - 28, 23, 49
Czech - 35, 16, 49
Belarus - 25, 28, 47
Hungary - 39, 16, 45
Ukraine - 28, 27, 44
England - 44, 14, 42
Orkney - 34, 25, 41
South French - 57, 4, 39
Croatia - 44, 17, 37
French - 51, 12, 37
North Spanish - 59, 10, 31
Bulgaria - 55, 14, 31
Tuscany - 72, 0, 28
Basque - 54, 19, 27
Bergamo - 63, 13, 24
Spain - 78, 0, 22
Greece - 66, 14, 20
Albania - 65, 18, 17
Sardinia - 88, 7, 5

Remember, at least 40% of most of those steppe numbers are Caucasus or "southern" or heavily "Basal Eurasian" in origin.


In the supplement there's extensive modeling which tries to get a better fit. You really should read it for yourself.

Here is the link to where we originally discussed the Haak 2015 paper. I went on record saying the rest of Europe might be very different. There are a lot of other threads where we discussed it. Just use the search engine.

https://www.eupedia.com/forum/threads/30878-Massive-migration-from-the-steppe-is-a-source-for-Indo-European-languages-in-Europe?highlight=Haak+2015



I want to qualify one thing about the analysis in Haak. You'll see that the Finns are not included. That's because they couldn't be modeled with only those three populations. They needed Siberian. There's something else about the very North Eastern European samples like Finland, Estonia, and maybe even beyond that I've speculated about, and that is that the "steppe" component in them may be more heavily EHG/SHG like than is the case for other populations. I've always held out the possibility that as the far northeastern areas were refugia for hunter-gatherers similar to the hunter-gatherer element in steppe peoples, that might be skewing the analysis a bit, and that therefore the number of invading IE might be smaller to some degree than the percentages might indicate.

As time goes on, we'll see if that's correct or not.

ROS
23-11-17, 02:11
http://img5.fotos-hochladen.net/uploads/blankmapeurope57nabwr3eh.png

Who do we believe?

Pax Augusta
23-11-17, 02:22
http://img5.fotos-hochladen.net/uploads/blankmapeurope57nabwr3eh.png

Who do we believe?

This is an amateur map and made before the analysis of Yamnaya.

Angela
23-11-17, 02:46
Only read ACADEMIC papers, and papers after the advent of ancient dna testing. That's papers from Max Planck, Johane Strause, Kurt Alt, Wolfgang Haak, David Reich, Patterson, Lazaridis, Mathiesen, and others.

I've provided this link at least three times before. It's a good start. If you want more I can provide them.
https://www.eupedia.com/forum/threads/34850-Important-papers-for-newbies-to-Population-Genetics?p=524330#post524330

As Pax says, that darn thing you posted isn't even based on ancient samples. It's just guesswork. Why would you turn to that?

Ed. It just goes to show that when I said it was junk even before ancient genomes were available I was right.

LeBrok
23-11-17, 05:16
Was there a paper about Globular Amphora about half a year ago, having GAC as all I2a farmers with about 30% of WHG?

Angela
23-11-17, 05:31
It was Mathiesen. See:
https://www.eupedia.com/forum/threads/34212-DNA-of-Globular-Amphora?p=521295&highlight=GAC#post521295
https://www.eupedia.com/forum/threads/34052-The-genomic-history-of-southeastern-Europe-Mathiesen-et-al?highlight=GAC

LeBrok
23-11-17, 07:02
Right on. So no surprises here, generally continuation of EEF, with traces of EHG/Yamnaya. Surely there was some contact, but no big migration yet from the Steppe. But obviously some cultural and material exchange happened. Though hardly one can make a conclusion about language here.
On other hand, looking at Matieson figure 1, this 30% WHG looks rather like a fresh admixture, not the one that happened to EEF farmers when they entered Europe for the first time, but something recent. Might be the WHG who were pushed by expanding Steppe into central Europe, from Belarus area. Might solve the puzzle, what happened to Latvia H-G, when they were replaced by Yamanya type. Might explain change in culture too.
Here is another thought, was the fresh WHG responsible for ubiquitous I2a2 in Globular Amphora? But that's another story. Would be nice to have a sample from 4k BC for comparison.

Olympus Mons
23-11-17, 11:54
Lebrok.
Where do you see Steppe in GAC? how do yo unknow if Iron gates HG in the region already were WHG with a percentage of EHG why is that Steppe in there?.

There are jumps of thousands of years that are unknown. who is to say that one of these days we will not find "neighbors" of all of these that were full of CHG, and EHG and SHG and Mix of...

ROS
23-11-17, 17:46
https://cache.eupedia.com/images/content/European_hunter-gatherer_admixture.png

LeBrok
23-11-17, 18:13
Lebrok.
Where do you see Steppe in GAC? how do yo unknow if Iron gates HG in the region already were WHG with a percentage of EHG why is that Steppe in there?.

There are jumps of thousands of years that are unknown. who is to say that one of these days we will not find "neighbors" of all of these that were full of CHG, and EHG and SHG and Mix of... Enlarge the Mathieson figures and this paper's and you will see. It is there, unless you don't want to see it. ;)

Angela
23-11-17, 19:21
https://cache.eupedia.com/images/content/European_hunter-gatherer_admixture.png

NOT BASED ON ANCIENT GENOMES...

I wonder if I can just program the computer to spit out certain sentences. :)

It's true, however, that in most analyses the Spanish have more WHG than Central and Southern Italians, which is why they plot where they do on PCAs of European variation.

davef
23-11-17, 19:24
NOT BASED ON ANCIENT GENOMES...

I wonder if I can just program the computer to spit out certain sentences. :)

It's true, however, that in most analyses the Spanish have more WHG than Central and Southern Italians, which is why they plot where they do on PCAs of European variation.

You should buy a few textbooks on NLP and basic programming ;)