search this blog

Monday, June 6, 2016

Comic relief from Hofmanova et al. at PNAS


PNAS has a new paper on the Neolithic transition in Europe. I don't know what the authors were puffing on when they computed the inferred mixture coefficients, but they look like crap, with Loschbour-related admixture (in other words, indigenous European ancestry) peaking near the Caspian Sea, including among North Caucasians and Kalmyks, who today live just northwest of the Caspian, but are recent migrants to Europe from Mongolia.

Moreover, western Turks appear to show fairly even ratios of Loschbour and early Aegean farmer admixture, which is also strange.

I pointed out this problem when the paper was posted for review at bioRxiv (see here), but my comments were ignored. The co-authors responsible for this analysis are Lucy van Dorp, Saioa Lopez and Garrett Hellenthal. The paper was edited by Eske Willerslev of the University of Copenhagen. Holy shit!


Zuzana Hofmanová et al., Early farmers from across Europe directly descended from Neolithic Aegeans, PNAS June 6, 2016, 2016, doi: 10.1073/pnas.1523951113

See also...

The discrepancy

191 comments:

Chad Rohlfsen said...

Some odd stuff, but Mesolithic Greeks are as I figured. EN-like.

Davidski said...

This paper can't be taken too seriously when it shows Kalmyks from Mongolia as almost 100% Loschbour.

Davidski said...

And where are you seeing the Mesolithic Greeks? They only report their mtDNA HGs.

Chad Rohlfsen said...

You're right. Still reading. Im ignoring the wild stuff and looking at the formal stats early in it.

Chad Rohlfsen said...

There are some distinctions between the Greeks and Anatolians vs Iberians.

Davidski said...

How do you know that they didn't fuck up the formal stats as well?

Samuel Andrews said...

This is getting annoying. The pattern in ancient DNA studies is multiple teams sequence DNA from the same ancient people and it takes years before anyone decides to move on to a differnt topic and sequence DNA from new ancient people. The genetic origins of Neolithic Europe are already figured out. Lets move on. Get genomes from Mesolithic Aegean or get genomes from Bronze and Iron age Aegean.

These researchers are wasting their money and skills.

Davidski said...

What skills?

Samuel Andrews said...

Dude it takes skill to write get DNA from bones and write this paper. They made mistakes but are still skillful people.

Davidski said...

Mistakes like this ruin the credibility of population genetics.

Hector said...

This is sublime comedy gold.

What the paper is saying is that among the listed groups(all of which are ancient except Yoruba) Loschbour has the closest affinity to Kalmyks overwhelmingly. Obviously it does not mean Kalmyks mostly descend from Loschbour.
The result will be similar if Han Chinese were analyzed this way. Han Chinese would be very distant from all of the listed groups but if you had to choose the closest group(to make up the whole pie as 100 percent) it would be Loschbour most likely.

If modern mongols are included in the listed group the Loschbour component will shrink by a good margin. If Russians are included the Loschbour component will almost completely disappear. However inclusion of Russians will mess up all other modern European populations analyzed on the map by the way.

So much huff and puff from misunderstanding.

Davidski said...

It's a supervised test, so the East Asian of the Kalmyks has to be described as one or more of the reference samples, and considering Basal Eurasian input in all of the reference samples except Loschbour, it's classified as Loschbour.

All they had to do was use an East Asian population in the reference panel.

Rob said...

What I find interesting is yet further evidence of a '2nd Neoithic wave' after ~5500 BC with higher CHG, having now included Kumtepe in their analysis, also seen in the later Greek Neolithic samples

Davidski said...

Maybe, but if they're classifying Kalmyks as 100% European hunter-gatherers, then that casts doubt on all of their results.

capra internetensis said...

"Sampled groups that contribute most to this mixture indicate a high degree of shared ancestry with the target group relative to other groups."

Kalmyks have a high degree of shared ancestry with Loschbour relative to Neolithic farmers or Yoruba. No shit.

What's the big deal?

Davidski said...

@capra internetensis

You can stick your relativism and semantics where the sun don't shine.

Kalmyks do not share any ancestry with Loschbour within any reasonable timeframe. So the results are misleading when they show Kalmyks, from Mongolia, as the samples with the strongest links to indigenous Europeans.

Matt said...

Questionable why they've left it in. The curious thing here is that they've actually added a lot into their SI Appendix* that's relevant - SI10 shows that using this method (which they go into more detail about, but I don't fully understand yet) moderns are much better matches for all populations.

They had time to improve their PCA as well, but not that.

*and if people want to talk about their D stat analysis of relatedness between ancient farmers, that could be interesting since looks like they have some results with significant Z. They really have added a lot in their SI and in references to their SI in the paper that would be good to discuss here, rather than let that figure overshadow the rest of the paper.

capra internetensis said...

The study doesn't have anything to do with freaking Kalmyks. It's not trying to identify East Asian admixture or different kinds of hunter-gatherer ancestry; it is trying to discern relatedness to different Neolithic populations. The only thing that graph is saying is that Kalmyks don't have Neolithic ancestry to speak of.

The section that refers to the figure is specifically referring to the Neolithic farmer ancestry. There is a different section discussing hunter-gatherer admixture which uses f4 statistics.

You seem to be deliberately reading it in a misleading way for some bizarre reason.

Davidski said...

Hang on, but you said this...

The section that refers to the figure is specifically referring to the Neolithic farmer ancestry.

Really, so how come Loschbour-related ancestry in many parts of Turkey is on a par, or almost on a par, with farmer ancestry?

That's just stupid, even if we take into account your comment above, and the fine print you posted.

You seem to be deliberately reading it in a misleading way for some bizarre reason.

But it's a mixture analysis, showing admixture proportions. Any reasonable observer who knows something about Eurasian population history will tell you that Kalmyks or Nogais don't have much, if any, Loschbour admixture. And Turks certainly don't fucking well have equal levels of Loschbour and Anatolian farmer admixtures.

So there you go, it's just wrong. Plain and simple. So now that I've explained this to you, and assuming you understood what I said, then why did you just accuse me of deliberately reading it in a misleading way for some bizarre reason?

Onur said...

@capra

Those supervised analyses suffer from the lack of an East Eurasian reference population. None of their results can be trusted for this reason as explained by Davidski.

Shaikorth said...

"All they had to do was use an East Asian population in the reference panel."

Presumably they did, in the panels that had modern populations like those used in figure S30, because there Kalmyks are 0% Loschbour.

In any case the only thing they write about Loschbour's ancestry is that it is closest to modern Eastern European populations. Unless the news outlets reporting about this paper are dumb enough to interpret the Kalmyks as Loschbour's closest relatives instead (so far has not happened, https://pnas.altmetric.com/details/8517493/news), including them in ancients-only modeling should be no more than a harmless oversight.

Alberto said...

Yes, not including an East Asian reference is a very obvious mistake. I guess we can all see that, and it does make that graph useless. But the paper is huge (in size and amount of details), so let's not just get stuck in that small part.

The D-stats seem a bit surprising too in that many results that you would expect to be rather insignificant turn out to be pretty significant. They don't show the number of markers, so that could be an issue in some/many of them.

The Kumtepe samples do seem to have some CHG admixture, though still only had time to skim through it.

Also from this quick look it seems that the late Neolithic sample (Klei10) seems to have acquired lighter pigmentation without showing any strong signs of gene flow from differentiated groups, so it might have been just by selection?

Davidski said...

Shaikorth,

What about the other "harmless oversights" in that analysis? Like the high levels of Loschbour ancestry in Anatolia relative to many parts of Europe?

Is that OK just because it's not reported in the press?

Shaikorth said...

I think including non-europeans in tests where everyone was modeled as a mix of ancient europeans was only an oversight and not worse, since they did not write conclusions trying to present the distorted results as products of actual high WHG ancestry in modern Turkey and so on.

Of course, since they don't explain what's actually going on with those results, that oversight can go from harmless to harmful as soon as press interprets things wrong.

Davidski said...

Adding East Asians to the reference panel and limiting the analysis to Europeans wouldn't solve all the problems, because the levels of Loschbour admixture are just too high in Southern Europe.

And for the icing on the cake, these jokers used Ashkenazi Jews to represent Poland.

truth said...

The Polish sample looks exactly as the Sicilian ones, so they probably used Jews.

Nirjhar007 said...

How the hell they gave importance to such low cov. sample (Kum6), Dave?.

Davidski said...

No idea. I don't care much for the rest of the paper after seeing that mixture analysis.

Saioa Lopez said...

As many of the commenters have noted, the main issue here is that you should NOT interpret Figure 3 as showing admixture among the sources at left leading to the groups on the map. It is only meant to show for each target group (circle) which of the surrogate groups they are most closely related to relative to all included surrogates. Some of these patterns could result from admixture, but in many cases it's likely attributable to older ancestry sharing (there is no sense of timescale here). These patterns will change if you include different surrogate groups, so you need to be very careful about interpretation. We did a version that included Han as a surrogate as well, which did soak up some of the Loschbour contributions in many groups just as David suggests would happen. As other commenters have noted, the Han did not really affect the groups we were interested in (the main interest was looking at matching to Neolithics). But apologies if this caused confusion. Also, all groups are placed on the map based on sampling location, to make things standardized. As has been pointed out as well, when interpreting signals you need to take into account that the sampling location may not be a great representation of where that group is "from" -- e.g. if we had included Jamaicans from London!"

Saioa, Lucy and Garrett

Iranocentrist said...

Davidski can you stop you're swearing please, it might be keeping away study researchers from participating here.

Davidski said...

@Saioa Lopez

As other commenters have noted, the Han did not really affect the groups we were interested in (the main interest was looking at matching to Neolithics).

But hang on, to me it looks like you failed to correctly explore the main interest, because your failure to include an East Asian reference resulted in skewed results for the western Turks. You know, the samples closest to the sampling locations of your ancient samples.

Take another look at the Turkish results; their levels of Loschbour-related ancestry are on a par with those of Scandinavians, and that's due to East Asian admix in the Turks. And I'm willing to bet that this has also resulted in them matching the wrong Neolithic samples.

So I'm sorry but to me it looks like a mistake, especially since you seem to have put a lot of emphasis on these regional Turkish groups in your paper.

Davidski said...

@Iranocentrist

Davidski can you stop you're swearing please.

OK, good point.

Shaikorth said...

According to p.99 of the supplements those western Turkish samples (Turkish and Turkish_Balikesir) actually are modeled with a significant Neolithic portion even when modern populations are included in the panel, they're the only group to do this besides Sardinians.

Davidski said...

Look, if minor East Asian ancestry isn't accounted for, then them, as well as any other samples with minor East Asian/Siberian, are likely to get skewed results, and this means that we don't get to see the correct patterns of Neolithic ancestry across West Eurasia, or even Europe.

I'm not moving from this position.

Roy King said...

@Davidski,
I do find your ad hominem argument and brazen language not convincing. I agree that your not seeing the forest through the trees in this article. I know some of the coauthors who do first rate work in genetics and Neolithic archaeology. To me, what is most interesting are the new data from the Greek Early Neolithic site and the two Greek Late Neolithic site, showing an increased CHG component in the Late Neolithic sites 9as well as in the Kumtepe sample). Please discuss the overall inferences of the study rather than focusing on petty disagreements on statistical methodology.

Shaikorth said...

The East Eurasian in Turks, Kalmyks etc. is accounted for in fig S30. which is what they referred to in that comment.

Arch Hades said...

Modern Western Turks are as WHG as they are ENF? LOL

capra internetensis said...

@Davidski

If someone runs TreeMix with a limited set of populations, and TreeMix infers some impossible migration edge, you don't just dismiss TreeMix as wrong and worthless and the authors as incompetent. If this is treating some of the Turkish samples weirdly there may be a reason for it. Not that they are literally WHG of course, but the paper makes no such claim.

Of course it could turn out to be wrong but I don't know how you know so much about the ancestral composition of Turks that you can say that.

Alberto said...

What is interesting is that Early Neolithic and Late Neolithic samples from Greece are just like Early Neolithic Anatolians. This is unlike the rest of Europe, where we find increasing level of WHG as farmers were mixing with Mesolithic populations.

What this means is that probably the Mesolithic population from Greece was not WHG, but just like Anatolians (as Chad said above already). This idea is reinforced by the two Mesolithic samples having mtDNA K1c, which is different from all other European Hunter-Gatherers.

So we could say that from a genetic point of view, Greece during the Mesolithic and Neolithic was "West Asian", and not "European". However, modern Greeks are European. So at some point between the Neolithic and today there must have been a migration from within Europe to Greece (or more than one).

This team said to have Bronze Age samples too, if I remember correctly (Chad reported about it in some conference last year). I don't know why these samples didn't make it into the study, but I hope we'll see them soon.

Chad Rohlfsen said...

The greater change they saw in Greek mtDNA was the late Neolithic, which we can get a hint at being CHG related. It's not a lot of CHG, but it's there. Not a surprise. I have a feeling that once these are analyzed, we could see that EN Greeks are distinct from Anatolians. I think there is minor CHG in Anatolians that is not in Greeks or Iberia EN. Danubians always have a CHG shift vs Iberians in PCA. I can get this on qpAdm too.

Nirjhar007 said...

Alberto,

Yes. I also can't wait to see the Mycenaean aDNA . One of the most important aDNA, regarding the IE question, that will get revealed.

Nirjhar007 said...

Chad,

The greater change they saw in Greek mtDNA was the late Neolithic, which we can get a hint at being CHG related. It's not a lot of CHG, but it's there. Not a surprise.

I am sorry if I am getting you wrong, but are you suggesting that, this supposed shift towards CHG related ancestry, happened due to females again?.

bharatiya said...

@Nirjhar
Recall you mentioning DNA from Swat and BMAC do you know any more details about these studies

Nirjhar007 said...

And you are seriously asking that question here?.

Alberto said...

@Chad

Yes, I remember they mentioned that they saw continuity from Late Neolithic to Bronze Age. Which makes it even more mysterious (though I think the samples were only early Bronze Age and none after 2000 BC?). Also it was only mtDNA, so we'll have to wait for the rest of the genetic material to really know.

Kurd Dgk said...

The PCA (fig 3) is unacceptable and should be re-done, as it does cast doubt on their methodology and the rest of the analysis

Chad Rohlfsen said...

Nirjhar, what are you talking about? If you're trying to troll, I'd keep it moving along.

Davidski said...

capra internetensis & Roy King,

There are no semantics that you can offer to rescue this paper.

The authors clearly focused on modern Turks because they saw these samples as particularly relevant to their study.

And the results they obtained for most of these Turks defy any sorts of expectations or interpretations. Scandinavian-like western Turks, eh?

Classifying the Kalmyks and Nogais as ~100% Loschbour is also ridiculous, but OK, let's put that down to me being confused about the intentions of the authors, assuming of course, that their intention was not to offer any sort of logical picture of the genetic structure of Europe.

Clearly it wasn't, was it?

Roy King said...

@Davidski
Could you discuss the f3 and f4 results showing increased CHG in Kumtepe6 and Late Neolithic Greek samples or are you fixated on modern Turk results?

Davidski said...

I have now mostly recovered from the shock, horror and dismay of viewing their mixture analysis. I'm now going over the formal stats.

One thing that jumped out at me was their use of the Khomani as one of the outgroups.

The problem is that the Khomani have minor Anatolia Neolithic-related ancestry, so unless I'm missing the fine print in the paper saying that they removed the admixed Khomani individuals from their runs, then I'd urge caution in the interpretation of all of the formal stats with the Khomani as the outgroup.

Kurd Dgk said...

I posted some IBS results for Kumtepe 6 when I first acquired it about 6 months ago. They showed a higher similarity to CHG than almost all of the other ancients. Curiously, Iceman looked most similar to Kumtepe. I posted the results on AG forum at http://www.anthrogenica.com/showthread.php?6218-A-first-look-at-Kumtepe-6&highlight=Kumtepe

The caviot with Kumtepe was, it had very few overlapping calls with my other genomes. The other caviot was some of the genomes in the run were pseudo diploids, whereas some were true diploids

Kurd Dgk said...

I don't believe I did f4s using Kumtepe

Gioiello said...

@ Davidski

In the world there are people who lie knowing to lie and other that may be wrong, but they believe in what think, thus, when they have the first suspect that things may be in a different way, they try to settle the shot.
I think you belong to this second kind. The explication of the paper is clear: "Significance: One of the most enduring and widely debated questions in prehistoric archaeology concerns the origins of Europe’s earliest farmers: Were they the descendants of local hunter-gatherers, or did they migrate from southwestern Asia, where farming began? We recover genome-wide DNA sequences from early farmers on both the European and Asian sides of the Aegean to reveal an unbroken chain of ancestry leading from central and southwestern Europe back to Greece and northwestern Anatolia. Our study provides the coup de grâce to the notion that farming spread into and across Europe via the dissemination of ideas but without, or with only a limited, migration of people".
I have explained in these 10 years who is behind these positions, who funds the researches and why. Not by chance here the usual exponent of the levantinists and of Stanford, Mr Roy King, blethers.

Matt said...

Their comment is

"The Khomani San were selected as an outgroup as they are considered to be the most diverged extant human population, having diverged from all other modern humans at least 100kya, and are highly unlikely to have experienced substantial Eurasian admixture.

However, we also tested the Mbuti and Yoruba in the place of the 6=Khomani San and we note that our results were robust to the choice of sub-Saharan African population utilized.

The results are also consistent for all the levels of filtering and for the dataset without transitions."


The first part is questionable though, presumably they didn't get statistically significant differences with the Mbuti or Yoruba groups.

Although I'm not sure how any admixture effect would shift many of their statistics like f4(Early farmer, Iceman, Aegean, Khomani) in any case. (Possibly if the Khomani have Early Farmer related ancestry that less related to Iceman and Aegean farmers, I guess).

Their main comment is:

We calculated f4 statistics to examine whether the Aegean Neolithic farmers shared drift with genomes from the Spanish Epicardial site Els Trocs in the Pyrenees (3, 12) that is distinct from that shared with Early Neolithic genomes from Germany and Hungary.

In a test of the form f4 (Germany/Hungary EN, Spain EN, Aegean, ‡Khomani), we infer significant unique drift among Neolithic Aegeans (not significantly in Bar8) and Early Neolithic Spain to the exclusion of Hungarian and German Neolithic genomes (SI Appendix, Table S21). The best explanation for this observation is that migration to southwestern Europe started in the Aegean but was independent from the movement to Germany via Hungary.

Kristiina said...

Alberto "What this means is that probably the Mesolithic population from Greece was not WHG, but just like Anatolians (as Chad said above already). This idea is reinforced by the two Mesolithic samples having mtDNA K1c, which is different from all other European Hunter-Gatherers."

I agree with that. You should check this map of habitable areas 20 000 years ago (the map should be from The Times Atlas of World History (Times Books (4th ed) 1993). (https://chiefio.files.wordpress.com/2013/05/livable-ice-age-20kybpasiagl.gif)

During the Ice Age, the Alps surely acted as a barrier between eastern Mediterranean and Western Europe, but you can see that Greece and Turkey look like being almost one land.

Capra Internetiensis "If someone runs TreeMix with a limited set of populations, and TreeMix infers some impossible migration edge, you don't just dismiss TreeMix as wrong and worthless and the authors as incompetent. If this is treating some of the Turkish samples weirdly there may be a reason for it. Not that they are literally WHG of course, but the paper makes no such claim."

It has several times been repeted that there has been an important genetic turnover in the Near East after the Neolithic. On the basis of language families, you can easily argue that there were movements of people from North-Africa and Eurasian Steppe (and Europe). All steppe people, whatever they were, should autosomally be close to WHG/EHG, so I do not think that it is a surprise that there is something similar to WHG and EHG north of Caucasus. According to your treemix trees what is the relation between WHG and EHG? Are they on the same branch? Treemix trees often make the split between Mediterranean/Near Eastern branch and the Eurasian branch. Have I also seen argued that Loschbour is closer to ENA than most other Mesolithic European genomes.

Davidski said...

Capra the goat still doesn't understand the problem..

If someone runs TreeMix with a limited set of populations, and TreeMix infers some impossible migration edge, you don't just dismiss TreeMix as wrong and worthless and the authors as incompetent. If this is treating some of the Turkish samples weirdly there may be a reason for it. Not that they are literally WHG of course, but the paper makes no such claim.

The reason why the Turks show inflated, Scandinavian-level WHG admix is because their East Asian admixture is inflating it. Not just replacing it, but inflating it.

I guess if you're semantically inclined, like he obviously is, then the result isn't wrong, it's just different.

In my, perhaps less philosophical opinion, however, it's wrong, misleading and useless.

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

Thanks Matt.

Again, for the philosophically and semantically inclined, "substantial" here could mean over 10% or even 20%.

The Khomani San were selected as an outgroup as they are considered to be the most diverged extant human population, having diverged from all other modern humans at least 100kya, and are highly unlikely to have experienced substantial Eurasian admixture.

But some of the Khomani do have very clear Anatolia Neolithic admixture. Not over 10%, but easily discernible. So yeah, there's that.

Kristiina said...

Rob, and Villabruna is said to be WHG, and it is argued that WHG came from Eastern Mediterranean. I must say that after Villabruna and paleolithic Europeans the relation between WHG and ENF is not so clear for me any more.

Davidski said...

She probably means this part from Fu Q et al.

One possible explanation for the sudden drawing together of
the ancestry of Europe and the Near East at this time is long-distance
migrations from the Near East into Europe. However, a plausible alter-
native is population structure, whereby Upper Palaeolithic Europe har-
boured multiple groups that differed in their relationship to the Near
East, with the balance shifting among groups as a result of demographic
changes after the Glacial Maximum.

Rob said...

I thought as such.
Seems unlikely. Nothing notable before the Neolithic.

Gioiello said...

Davidski has left a new comment on the post "Comic relief from Hofmanova et al. at PNAS":
She probably means this part from Fu Q et al.
"One possible explanation for the sudden drawing together of the ancestry of Europe and the Near East at this time is long-distance migrations from the Near East into Europe. However, a plausible alternative is population structure, whereby Upper Palaeolithic Europe har-
boured multiple groups that differed in their relationship to the Near East, with the balance shifting among groups as a result of demographic changes after the Glacial Maximum".

And if R1b won't be found in Middle Easterner aDNA, it will be true the second hypothesis.

Kristiina said...

Yeh, I agree that I am aware of those statements. In any case, In Posht et al, it has been argued that there was a repopulation of Europe:
"Our demographic modeling reveals a dynamic history of hunter-gatherers, including a previously unknown major population shift during the Late Glacial interstadial (the Bølling-Allerød, ∼14.5 ka)." (Posht et al., 2016)

In Fu et al. they distinguished this Villabruna cluster with Near Eastern affinities: Figure 4, Drawing together of European and Near Eastern populations ~14,000 years ago. "We observe an affinity to Near Easterners beginning with the Villabruna Cluster, and another to east Asians that affects a subset of the Villabruna Cluster."

So it is argued that Villabruna had Near Eastern affinities compared to European Palaeolithic genomes but no ANE. This means that we have Villabruna with Near Eastern affinities before the Neolithic and the Neolithic farmer genomes with Near Eastern affinities.

On top of that, Villabruna cluster was defined (close to) WHG, and we know that after the Ice Age WHG spread northwards.

That is why I automatically in my mind linked Villabruna cluster with the eastern Mediterranean/Italy, but maybe it is the other way round, and the flow went from Europe to the Near East.

Rob said...
This comment has been removed by the author.
mickeydodds1 said...

'Fu Q'

Well that pretty well sums up the tenor of some of this threads comments.

Kristiina said...

I had a closer look at Fig. 3 of this new paper. Please correct me if I am wrong but below the figure it is explained that "large pies, enclosed by borders matching key at left" which means that the dark circle around Loschbour only identifies it but the "WHG" component is not "black" but "green" and "brown" which means that Kalmyks and Loschbour do not share anything. In that figure there are two "Near Eastern" components that are wide spread in Europe: dark green which is at 35-40% around the Mediterranean and which peaks in Pal7 (Greece) and Barcin genomes; and brown which is absent in Greek and Anatolian Neolithic genomes but present in LBK Germany and Loschbour. Today this brown component peaks in people who live on the west coast of Turkey. As Loschbour is WHG and Villabruna cluster is close to WHG with Near Eastern affinities, one could conclude that the pre-Neolithic Near Eastern component is the brown component. Now the question is where it comes from?

Shaikorth said...

Fig 3's Loschbour pie is Loschbour as a mix of other ancients in their panel, which aren't WHG, so the result doesn't represent it well. It's a misleading result like the Kalmyk one in its own way.

Loschbour is better represented as a mix of modern Northeast Europeans than those ancients. Shown in figure S24.

Samuel Andrews said...

@Kristina,
"As Loschbour is WHG and Villabruna cluster is close to WHG with Near Eastern affinities, one could conclude that the pre-Neolithic Near Eastern component is the brown component. Now the question is where it comes from?'

There's no difference between Loschbour and Villabruna. They have the same amount of Near Eastern affinity. The recent paper that sequenced Villabruna's genome ignored the possibility more Near Eastern affinity in WHG than in earlier Europeans could be the result of European WHG migration into the Near East instead of Near Eastern migration into Europe.

WHG can't be modeled as part Near Eastern because it lacks Basal Eurasian. However Near Easterners can be modeled as part WHG.

The first real Near eastern, with Basal Eurasian, migration to Europe came in 6000 BC with EEF.

Samuel Andrews said...

Continued....

So the brown score in the pie charts of this study don't represent any type of pren-Neolithic Near Eastern ancestry, since Near Eastern ancestry didn't exist in Europe till the Neolithic except it was probably in Greece before then.

Alberto said...

Kristiina, yes, that seems to be the correct way to read the chart. So rather than a mistake it just looks like a slightly confusing presentation (using the same colours to identify the ancients as used for each of the components inside the pie charts).

Re: the D-stats, the complete set is available in SI (S02), with most of them repeated using different outgroups (Mbuti, Yoruba, Chimp, Khomani), and with all sites and transversion sites only.

The ones that I'm still trying to understand are those of Sheet 8, in the form F4(Ancient, Aegean : CHG, Outgroup), where all of them are very significantly negative, even with CHG-rich populations like CW or BB. Foe example:

Bell_Beaker_LN Bar8 : KK1 Khomani -0.0257 -5.223
Corded_Ware_LN Bar31 : SATP Khomani -0.0662 -9.433
Corded_Ware_LN Klei10 : KK1 Khomani -0.07 -10.842

Like if all these Aegeans were CHG. Have we seen this before with Anatolia_Neolithic?

Gioiello said...

Samuel Andrews says: "The recent paper that sequenced Villabruna's genome ignored the possibility more Near Eastern affinity in WHG than in earlier Europeans could be the result of European WHG migration into the Near East instead of Near Eastern migration into Europe.
WHG can't be modeled as part Near Eastern because it lacks Basal Eurasian. However Near Easterners can be modeled as part WHG.
The first real Near eastern, with Basal Eurasian, migration to Europe came in 6000 BC with EEF".

"So the brown score in the pie charts of this study don't represent any type of pren-Neolithic Near Eastern ancestry, since Near Eastern ancestry didn't exist in Europe till the Neolithic except it was probably in Greece before then.

Alberto says: "Like if all these Aegeans were CHG".

I reason about the Y (and the mt), i.e. what may be the proof of aDNA.

Certainly R1b didn't come from Middle East if it won't be found in the aDNA there as has been found in Italy, Iberia and Samara (R-L23 subclades).
Very likely hg. J (J1 and J2) came from Caucasus and not Middle East.
Everything is explained.

Chad Rohlfsen said...

WHG can be modeled as a mix of Anatolian, GoyetQ, and AG3. GoyetQ is almost significantly closer to Ust-Ishim than WHG. This isn't reaolved just yet.

Onur said...

Kristiina, yes, that seems to be the correct way to read the chart. So rather than a mistake it just looks like a slightly confusing presentation (using the same colours to identify the ancients as used for each of the components inside the pie charts).

No, Kristiina is wrong. The black component is Loschbour, i.e., WHG. In that analysis every modern population (shown in small pies) is represented as a mixture of all the individuals or populations in the reference panel (i.e., the ancient individuals and Yoruba) whereas every ancient individual of the reference panel (shown in large pies enclosed by borders in their own color) is represented as a mixture of all the individuals or populations in the reference panel except himself/herself, that is the sole reason why Loschbour's pie chart lacks the Loschbour, i.e., WHG (black) component. So Kalmyks and Nogais are indeed represented as totally or almost totally as Loschbour, i.e., WHG, which is the result of the fact that that analysis lacks a reference individual or population with East Eurasian ancestry and that Loschbour is the only Eurasian individual in the reference panel who lacks Basal Eurasian admixture, making Loschbour the only reference individual to represent East Eurasian ancestry in the absence of a reference individual or population with East Eurasian ancestry. So no matter how you slice it, you cannot cover up the mistakes of the analysis, which are huge and inexcusable.

epoch2013 said...

"Needless to say, this "3rd source" should be looked for in the Balkans and Black Sea region. For this we do have evidence for- not the least- the Epigravettian-oid features in Satsurblia cave itself."

Here's my idea.

Fu paper states WHG is a sister clade of Gravettian. It also must have lived not that far from Iberia during LGM as the Red Lady of el Miron is modeled as 30% WHG. It can't have lived in Greece during the mesolithic as two mesolithic Greeks have K1c, not found among any WHG. It can't have lived in the northern Balkans during the early Gravettian as both Muierii2 and Ciaclovina1 show no kinship apart from the general K14 affinity.

We have a tiny bit of WHG in Vestonice16. You can see a slight affinity of Paglicci133 and Ostuni2, both of the Italian clade of the Vestonice group, to Villabruna.

We have Paglicci71, 18.000-19.000 year old having U5b2. Paglicci cave has horse paintings, horse being one of the two key species of game the Magdalenians hunted for.

Must be around the Adriatic, I would say. Although we still can imagine the Greek K1c's coming in later.

Alberto said...

@Onur

Ok, yes, that's probably the whole correct way of reading it, thanks.

Not that anyone's trying to "cover up" any mistake. I said above it was an obvious mistake and after Kristiina explained it I thought I should just correct my assertion. Now that you explained it further, I think again it's a mistake not to include an East Asian reference.

But really, the paper has a lot more than that chart.

Onur said...

But really, the paper has a lot more than that chart.

Surely I was not referring to the whole paper.

Gaspar said...

In this paper and in other papers , I see the main issue of arguments in what people refer to as middle-east. Too many have different views.
also, what is near east? is it Levant.

I think many of you people need to come to a consensus of what is a middle-east area.
for me:
Anatolia is never middle-east , it was never semitic in tongue.
south caucasus is part of caucasus it is not middle-east.
Persia/iran is not middle east.

In regards to this paper, the terms of locations are the confusing part.

In regards to Villabruna ( a sample that many detest because it has thrown a spanner in their theories ). His middle-eastern means what ?.............more likely his line travelled to the middle-east ..............people did move from west to east and not just east to west

Poise n Pen said...

There's nothing AT ALL wrong with it. Those are the best matches, and those are the things that they care about looking at. It's informative because it shows that the relation spreads out from a certain area.


It does point out how much bullshit these things are though. The problem is obvious from the words you use. You keep saying it's not lochsabur ancestry. WELL NO FUCKING SHIT. You can't tell the direct ancestry for ANYONE this way accurately. The fact people use it that way and that others believe it's valid just shows how much fapping in the wind all of this kind of analysis is.

It's not worthless really, but it is easy to make some analysis that supports any scenario you want (like steppe theory for example). And it's virtually ALWAYS used in some misleading way that fits an agenda.

batman said...

@ Kristiina

https://chiefio.files.wordpress.com/2013/05/livable-ice-age-20kybpasiagl.gif

This map is definitely outdated.

The extent of the ice and the finer lines of the inhabittable climate-zones during Late Paleolithic have been revised a number of times since 1993 - as both paleo-geology and paleo-botany have assembled a number of drill-cores across the Eurasian continent, as well as the Atlantic facade...

Here's one of them - clearifying that the Gulfstream avtually passed through the North Sea all through the late Weichsel, LGM and YD included:

http://geology.geoscienceworld.org/content/42/8/663.full?ijkey=h3AuZZV5q9nwk&keytype=ref&siteid=gsgeology

Please note that this test-material clearly states that YD represents the very lowest temperature of the Atlantic surface, from which the major part of (western) Eurasia's air-temperatures are defined.

Kristiina said...

Batman, it is probably a very valid paper but I cannot find any map of the Mediterranean basin there. In the abstract they say that "The results suggest that warm Atlantic water never ceased to flow into the Nordic seas during the glacial period; inflow at the surface during the Holocene and warm interstadials changed to subsurface and intermediate inflow during cold stadials. Our results suggest that it is the vertical shifts in the position of the warm Atlantic water that cause the abrupt surface warmings.".I would be glad if you could find me a better map of the Ice Age sea level in the eastern Mediterranean.

Onur, thanks for the explanation that is missing in the figure. According to Eurogenes K12_10K SNPs..., Kalmyks have only 1-4% Bedouin, which correspond to the small amount of dark green in the circle, and the rest is something else, and the rest cannot obviously be anything else than Loschbour. We usually have the same lack of resolution when talking about ENA. However, Loscbour's circle still shows Loschbour's affinity to farmer genomes. It may not be entirely correct, but it is still in line with this detected Near Eastern affinity of the Villabruna cluster. On the basis of your comments, it looks like Villabruna gene flow went also from Europe to Anatolia. Gioiello, you should be happy that Loschbour which is similar to Villabruna mostly consists of green colour that is found in Sardinia and Greece and is in line with your Italian/(Balcan) Ice Age refuge theory. WHG may have survived better in Estonia than in Italy for the lack/paucity of Basal in Estonia, but after the Ice Age the situation was surely different in both places.

Davidski said...

I hope the authors of the paper caught these comments from Poise n Pen.

It's not worthless really, but it is easy to make some analysis that supports any scenario you want (like steppe theory for example). And it's virtually ALWAYS used in some misleading way that fits an agenda.

This is the kind of nonsense you (the authors of the paper) are helping to propagate by using badly designed and highly misleading models like in your mixture analysis.

capra internetensis said...

@Davidski

You keep *saying* it's wrong, but you aren't actually giving us any evidence of that. I had a look at your CHG K10, for instance, which suggests that these Turkish populations are about 25% Anatolian Neolithic, 25% HG + East Eurasian, the other half CHG and SW Asian. How do you know that breaking up this 50% into closer-to-Barcin and closer-to-Loschbour in the proportions shown is an artifact of the East Asian admixture inflating the Loschbour-like proportion?

I'm perfectly willing to believe it's wrong, but you aren't telling us how you know this very technical thing.

Matt said...

So their ADMIXTURE analysis using only the ancient populations has a pretty clear distinction between the Aegeans and the Central European EEF by around K4, in the form of an element that is not 100% in the CHG, but is shared between CHG and Aegean farmers.

Note Kumtepe4 also has some alignment with the Indo-European / steppe populations, even at higher K. However as they note, D-statistics also show no greater link with WHG compared to EEF, so this is quite odd in that respect.
These may have looked different with all the Anatolia_Neolithic samples.
...
Looking through the table that Alberto has alerted us to, and their F4(Early Farmer, HG, Aegean, Khomani) transversions only stats, there are some tendencies to low significance with Pal7 (Greek Final Neolithic, 2000BC) as Aegean for some populations of Central European Early Farmers:

Esperstedt_MN HungaryGamba_HG : Pal7 Khomani 0.0144 0.846
HungaryGamba_CA Motala_HG : Pal7 Khomani 0.0140 1.138
HungaryGamba_CA LaBrana1 : Pal7 Khomani 0.0145 0.983
HungaryGamba_CA Loschbour : Pal7 Khomani 0.0087 0.631
HungaryGamba_CA HungaryGamba_HG : Pal7 Khomani -0.0084 -0.457
LBKT_EN LaBrana1 : Pal7 Khomani 0.0411 1.212
LBKT_EN HungaryGamba_HG : Pal7 Khomani 0.0429 0.974
Stuttgart HungaryGamba_HG : Pal7 Khomani 0.0316 1.605
SwedenSkoglund_MN Loschbour : Pal7 Khomani 0.0168 1.062
SwedenSkoglund_MN HungaryGamba_HG : Pal7 Khomani 0.0146 0.922
(the Z is kind of out of whack with the D for LBKT_EN - LBK in Transdanubia - however)

But definitely not so for the Iberian farmers (and also LBK_EN):
Spain_EN HungaryGamba_HG : Pal7 Khomani 0.0814 6.718
Spain_MN HungaryGamba_HG : Pal7 Khomani 0.0834 6.950
LBK_EN HungaryGamba_HG : Pal7 Khomani 0.0396 3.270
Starcevo_EN HungaryGamba_HG : Pal7 Khomani 0.0701 2.333
Baalberge_MN HungaryGamba_HG : Pal7 Khomani 0.0423 2.142

Kinda picking Pal7 a bit at random here, as it got the lowest stat (the stat where HungaryGambaHG appears to share more drift with Pal7 than HungaryGamba_CA) but true for other populations.

It's not that general a trend though as some EN Central Europeans perform like the Spain_EN with some Aegean_EN.

They do reach significance when considering non-transversions so perhaps nothing.

Davidski said...

@Capra

I've got all of these Turkish samples, and I know how they behave in a lot of different tests. There's no way they should ever be characterized as having as much Loschbour-related ancestry as Scandinavians.

Also, it's easy to see that East Asian ancestry is inflating the Loschbour-related ancestry in this mixture model.

Nogais are classified as almost 100% Loschbour here, but they're just North Caucasians with relatively high East Asian admixture. You can have a look at their genetic structure in the Haak et al. Admixture analysis.

Have a look at that carefully, and then come back here and tell us how accurate their results were in this mixture model in the context of Neolithic vs non-Neolithic ancestry.

OK?

Davidski said...

@Matt

I suspect that there are some issues with the formal stats in this paper. I'm checking that now and might have something to say about it later today.

Onur said...

@Kristiina

However, Loscbour's circle still shows Loschbour's affinity to farmer genomes. It may not be entirely correct, but it is still in line with this detected Near Eastern affinity of the Villabruna cluster.

In that weird supervised analysis Loschbour is being forced to be represented by one or more of the ancient farmer individuals in the reference panel and/or Yoruba, so in the absence of another choice he is of course represented as a mixture of various farmer individuals. So it does not indicate any farmer-like ancestry in Loschbour.

On the basis of your comments, it looks like Villabruna gene flow went also from Europe to Anatolia.

Villabruna, like all other WHG individuals, lack Basal Eurasian ancestry, so obviously WHG individuals do not have any farmer-like or CHG-like ancestry, if anything it is the other way round, it is farmers and CHG who have European hunter-gatherer-like ancestry.

Matt said...

Davidski, OK, thanks.

Btw, all, for the ADMIXTURE analysis, I took some screenshots and rotated them, because I found them hard to read:

http://i.imgur.com/haaSJb0.png

Rob said...

What's to be made of the high steppe like component in Kum 4 (or was it 6)?

Kristiina said...

@Onur ”So it does not indicate any farmer-like ancestry in Loschbour.” Well, if Loschbour was not like pre-Ice Age Palaeolithic European hunters and not like Neolithic Greek farmers, do we really have anything else than Gioiello’s Italian Ice Age refuge theory for WHG/Loschbour?

In any case, according to Eurogenes K12_10K SNPs, Kalmyks and Nogais are clearly different.
Kalmyk32: Naxi 44%, Nganasan 32%, Afanasievo 12%, European HG 4%, Atayal 2%, Bedouin 2%, Karitiana 1,7% Chukchi 1%
Nogay42: Naxi 10%, Nganasan 6%, Afanasievo 40%, European HG 16%, Atayal 2%, Bedouin 2%, Chukchi 2%
Kalmyk yDNA: C 71%, P 12%, R1a 6%, K 4.4%, N 3%, R1b 2.9%
Source: https://en.wikipedia.org/wiki/Y-DNA_haplogroups_in_Central_and_North_Asian_populations
C. 20% of Kalmyks’ mtDNA is Western Eurasian, and their most frequent mtDNAs are Altaian/Mongolian C and D. On the basis of their autosomal components and haplogroups, we all agree that they came from Altai/Mongolia. In this paper, do they really estimat the amount of true WHG, excluding ENA, CHG and EHG, in Kalmyks? I still cannot see why this Kalmyk issue should be so important here.

Davidski said...

I still cannot see why this Kalmyk issue should be so important here.

Because it's misleading.

Loschbour-related ancestry peaks in far Eastern Europe not because it's really found there at the highest levels, but because of recent East Asian/Siberian admixture.

Also, the paper actually claims that Loschbour-related ancestry peaks in Eastern Europe based on their results, and supported by Lazaridis et al.

However, their results are wrong, while Lazaridis et al. correctly found that Lsochbour-related ancestry peaks in the East Baltic region (not far Eastern Europe).

Not good enough.

Matt said...

@ Rob, Kumtepe4. Not sure what it's about. The sample is only mentioned in passing in the f4 stats (relevant results seem : less related to EN and MN than Aegean Early Neolithic, less related to WHG than Aegean Early Neolithic, more related to Kotias than some Aegean Early Neolithic) and they don't seem to test the relatedness in the form f4(Aegean, Kumtepe4; Steppe Influenced,Outgroup) where Steppe Influenced would be Yamnaya, Corded, etc. which could have been useful to see. That sample is actually from around 4846 - 4618 BCE though IRC, reading more carefully (also Pal7 is from 4000BC, so I got that wrong upthread). There could be something wrong with the sample?

Roy King said...

@Matt
Kumtepe4 is in the Current Bio Paper Omrak et al.and dates to 2800 BCE to 3500 BCE. As such, it is actually an Early Bronze Age sample. If it has affinities to Yamnaya, it could reflect the earliest Indo-European influence in Anatolia--Anatolian languages. Kumtepe6 is of course much older and dates to 4500-5000 BCE. The f3 and f4 stats in the supplement of this paper demonstrate that Kum6 and Kum4 are bear different relations to other ancient samples.

batman said...


From the paper;

"We ... observe striking genetic similarity both among Aegean early farmers and with those from across Europe. Our study demonstrates a direct genetic link between Mediterranean and Central European early farmers and those of Greece and Anatolia, extending the European Neolithic migratory chain all the way back to southwestern Asia."

http://www.pnas.org/content/early/2016/06/01/1523951113.full

Based on the genetic results (only) - HOW can the authors define the area of origin - and thus the d-i-r-e-c-t-i-o-n-s of these gene-flows?!

Rob said...

@ Roy

"@Matt
Kumtepe4 is in the Current Bio Paper Omrak et al.and dates to 2800 BCE to 3500 BCE. As such, it is actually an Early Bronze Age sample. If it has affinities to Yamnaya, it could reflect the earliest Indo-European influence in Anatolia--Anatolian languages. Kumtepe6 is of course much older and dates to 4500-5000 BCE. The f3 and f4 stats in the supplement of this paper demonstrate that Kum6 and Kum4 are bear different relations to other ancient samples"

Thanks for clarifying. That is major finding. People have been waffling on and wasting time about some pie-charts, and overlooking this detail, which I have personally waiting for.

in this light, it is interesting that this 'steppe ancestry' appears already in western Anatolia Chalcolithic, but is absent in the Hungarian sample in Baden, c 2800 BC.

Roy King said...

Figure S16 of the supplement shows, under admixture, a large Yamnaya component in Kumtepe4>Kumtepe6, suggesting that Kumtepe4 does indeed acquire some Yamnaya input at a relatively early data!

truth said...

@Roy KIng

I don't think it's because of Yamnaya, but because of CHG:

Rob said...

@ Truth -> Roy

I'd agree. I think Kumtepe 4 is too early for hypothetical Yamnaya movements into Anatolia. Rather, it more probably relates to broader processes which were impacting Anatolia, the Balkans & steppe from a highland West Asian source.

batman said...

Obviously, Fu & al may be closer to describe the possible directions of movement:

"One possible explanation for the sudden drawing together of
the ancestry of Europe and the Near East at this time is long-distance
migrations from the Near East into Europe."

What if it was the other way around - from paleolithic Europe into the epi-paleolithic/mesolithic (PPN) Near East?

May that explain the genomes - such as mt-dna U8/K1 - among mesolithic Aegans and Scandianvians, both?

The sculptures from Gobeli Tepe obviously bear witness to an arctical origin of the group that built this famous megaliths.

http://www.thelivingmoon.com/43ancients/04images/Turkey/Gobekli/061122statue.jpg

https://mathildasanthropologyblog.files.wordpress.com/2008/05/gobekli-tepe-man.jpg

Meanwhile, up north:
http://siberiantimes.com/other/others/news/n0376-shigir-idol-is-oldest-wooden-sculpture-monument-in-the-world-say-scientists/

The ice-age refugia in the Western Baltics is still the o-n-l-y area of paleolithic Eurasia known to have carried survivors of the human genome throughout the LGM as well as the Dryas. Thus the question have to be asked, for obvious reasons...

Ducking this question is nothing but ignoring well documented facts.

Rob said...

@ Batman

"The ice-age refugia in the Western Baltics is still the o-n-l-y area of paleolithic Eurasia known to have carried survivors of the human genome throughout the LGM as well as the Dryas."

Do you actually believe that or are you just kindly providing comic relief ?

Ariele Iacopo Maggi said...

Rob

"Why is it surprising the pre-LGM Europe is somewhat different to post-LGM Europe?" Because a population (WHG/Villabruna) seems to appear out of thin air... That's rather surprising, don't you find?

Rob said...

@ Ariele

"Because a population (WHG/Villabruna) seems to appear out of thin air... That's rather surprising, don't you find? "

The Villabruna population doesn't appear out of thin air. The Villabruna population is a composite, and, rather, one it's components is *yet unsampled*.

So what some people call "thin air" is really the remaining > 50 % of Europe yet to be sampled (the Balkans, more of Italy, Ukraine, and I'd include Anatolia as part of Palaeothic Europe for cultural & geographic reasons).

capra internetensis said...

@Davidski

Thank you. That makes sense. You'd expect them to have maybe less Neolithic than Balkars and Lumyks but not virtually none.

capra internetensis said...

@Rob

Mediterranean Iberia too. For that matter we have only one sample from one corner of the Franco-Cantabrian refugium.

Rob said...

@ Capra

"
Mediterranean Iberia too. For that matter we have only one sample from one corner of the Franco-Cantabrian refugium.'

Wholly agree. We're missing the business end of Palaeolithic Europe. No wonder people are talking about ghost populations and 'white walkers'

Davidski said...

Matt,

How would you interpret these sets of f4 and D statistics?

The f4 stats are from Hofmanova et al., while the D stats were run by me. The first set of D stats uses the highest quality Anatolia Neolithic sample from Barcin from Mathieson et al. and CHG genotypes from Fu Q et al., and the second uses the same Barcin sample plus CHG genotypes from Jones et al.

Also, keep in mind that, as far as I can tell, the Barcin genomes from Hofmanova et al. and Mathieson et al. date to the same period.

f4 Corded_Ware_LN Bar8 Satsurblia Khomani -0.0367 -8.145
f4 Corded_Ware_LN Bar8 Kotias Khomani -0.0193 -3.437

f4 Spain_MN Bar8 Satsurblia Khomani -0.0327 -5.385
f4 Spain_MN Bar8 Kotias Khomani -0.0182 -3.136

versus...

D Corded_Ware_Germany BAR20_I0709 Satsurblia Khomani 0.0215 4.299
D Corded_Ware_Germany BAR20_I0709 Kotias Khomani 0.0205 4.408

D Iberia_MN BAR20_I0709 Satsurblia Khomani -0.0003 -0.06
D Iberia_MN BAR20_I0709 Kotias Khomani -0.0017 -0.339

...

D Corded_Ware_Germany BAR20_I0709 Satsurblia2 Khomani 0.0224 4.38
D Corded_Ware_Germany BAR20_I0709 Kotias2 Khomani 0.0226 5.168

D Iberia_MN BAR20_I0709 Satsurblia2 Khomani 0.0068 1.222
D Iberia_MN BAR20_I0709 Kotias2 Khomani 0 0.004

Clearly, something's horribly wrong. If I made a mistake, my apologies. But I'm pretty sure I didn't make any mistakes.

I checked the datasets that I'm using for consistency with the f4 and D stats published in Mathieson et al. and Fu et al., so I can say with confidence that my D stats should not be much different from correctly run f4 stats using the same ancient samples.

FrankN said...

Rob, Capra: In fact, we have little more UP/Mesolithic aDNA than for the wider circum-Alpine region (Bichon, Swiss/ French Jura, Suebian Alps, Villabruna), the Ardennes (Loschbaur, Goyet), and parts of Iberia. No Western France, no British Isles, no CE Plain, no Scandinavia/Baltics. Putting it differently: On top of what you have listed before - no Ahrensburgian, Swiderian, Maglemosian (and also nothing at all from N. Africa).
While I am certainly happy that we have at least some UP/Mesolithic aDNA, we all need to be aware that most (certainly much more than 50%) of Europe hasn't been sampled so far.

batman said...

@ Rob

Arrogance is the brother of Ignorance.
You seem to be related to both of them. I just wonder if that's via the y-dna or the mt-dna. But then, again - it could be by both...

batman said...

8.000 years old boats, fit to sail the oceans, made in England:

http://digventures.com/2016/06/has-the-worlds-oldest-boat-building-site-been-discovered-near-the-isle-of-wight/

batman said...

Spreading wheat-production and lactose-persistance:

http://www.bbc.com/news/science-environment-31648990
http://www.nature.com/news/archaeology-the-milk-revolution-1.13471

Tobus said...

@batman:...fit to sail the oceans

What makes you think that? The link you provided doesn't say anything about sailing the oceans. Europeans didn't develop genuine ocean-faring vessels until very recently (last 600-700 years).

Davidski said...

Don't get him started about boats and sailing.

Kristiina said...

Davidski, I agree that figure is difficult to read and ambiguous.
However, there was an important error in my comment! According to that Eurogenes chart Nogais are 16-25% Bedouin. Is that other black circle in Caucasus Nogai? In any case, North Ossetes and Balkars are not different from Nogais. All of them are 20-30% Bedouin according to that Eurogenes chart. Only Mongolic speaking Kalmyks are different. Is it a problem that in the figure Nogai/(North Ossete/Balkar), if that is what they refer to, does not show more of that green or brown component? Do we know if Afanasievo contains any Barcin ancestry?

Shaikorth said...

Because those Nogais are not identical to Balkars and North Ossetians, the latter are much less eastern. None of the North Caucasus populations actually have high levels of WHG, even in the paper they don't claim this so there's no need to justify figure 3 with anything other than an unsuitable panel.

Davidski said...

Shaikorth, you still don't understand.

In this paper Nogais are classified as overwhelmingly Loschbour. In the main analysis, posted above, which doesn't include Late Bronze Age BR2, they're almost 100% Loschbour.

But yes they have significant Southwest Asian ancestry, and thus a lot of Neolithic ancestry.

So the analysis showing them to be ~100% European forager is off the wall.

Davidski said...

Shaikorth

What don't you understand about the concept of an analysis designed to pick up the correct levels and types of Neolithic ancestry not being able to pick up any Neolithic ancestry whatsoever?

Because that's the problem here. You don't seem to see it as a problem. Why?

Shaikorth said...

I don't think they should have had Nogais or Kalmyks or North Caucasians in fig 3 because it may cause someone to interpret high WHG in them.

But as I said I don't think it's anything more than an oversight because they've done the same analysis with a panel containing the HO set's moderns and there was no Loschbour in Kalmyks or North Caucasians then. They also modeled Loschbour as a mix of moderns and ancients, and there was no Nogai or Kalmyk in it. I don't think they interpret their results to mean high WHG in Nogais or Kalmyks and I don't think press will interpret it that way either, so I don't see the paper's issues as a massive deal.

Their interpretation about the kind of neolithic ancestry in moderns is this:

"Modern groups matching the Neolithics—mostly from the Mediterranean and North Africa—strikingly match more to Bar8 from northwestern Anatolia than to the LBK genome from Stuttgart in Germany, indicating that the LBK genome experienced processes such as drift and admixture that were independent from the Mediterranean expansion route, consistent with the dual expansion model. "

Which may turn out to be wrong but is not unreasonable to assume Mediterraneans should have more contribution from Mediterranean Neolithic than from Central European Neolithic. Lazaridis et al's linked Finestructure run also interpreted Stuttgart's clustering with North Italians as Stuttgart being closer to North Italian neolithic ancestry than to Sardinian, even though Sardinians have more neolithic in total.

There is also this:
"An exception to this are two Turkish groups (Turkish, Turkish Balikesir), which each receivea substantial contribution (30%) from the Neolithic Bar31
sampled in modern-day Turkey. Notably, out of all modern groups Sardinians receive by far the highest genetic contributions (>97%) from Neolithic genomes under analysis (II), suggesting a strong genetic affnity between
modern-day Sardinians and these ancient samples even relative to any other modern-day groups (Figure S30)."

Figure S30 used a panel containing the moderns of Human Origins set on top of the ancients so it did have East Eurasian references for those Turks. So there might be some kind of genealogical connection as with Stuttgart and North Italians. In total affinity Western Turks are of course more distant from Stuttgart than pretty much any Southern Europeans, but that's different.

Davidski said...

OK, just stop and read carefully what I'm going to ask you.

Nogais have Anatolian Neolithic and European Hunter-Gatherer ancestry. Why then, should they be left out of a test that purports to test the levels and types of Anatolian Neolithic ancestry?

Just answer this question in as few words as possible.

Shaikorth said...

Because figure 3 clearly is not their go-to test in measuring ancient ancestry in moderns, so they could have just as well left the modersn out. They refer to it when they talk about the neolithics' relation to each other, but "relation to modern populations" section refers to other tests (like the one in fig S30.) with better references for moderns.

The figures leave potential for confusion and misinterpretation, which is bad, but since they don't draw those potential misinterpretations I don't see how this paper is as bad as something like Basu et al. which referred to ADMIXTURE components as actual ancestral components.

Davidski said...

You didn't answer my question.

You need to answer my question before we can move on.

Matt said...

@ Roy, thanks. I assumed the same rough dating for all the Kumtepe samples, since there's no specific dating for Kum4. A steppe related component in ADMIXTURE analysis would fit better with a relatively later dating.

@ Davidski, yeah I see what's going on there with the D stats giving a result we would expect from previous work - Anatolia Neolithic and Iberia_MN equally related to CHG, Corded Ware more to CHG - with the stats from this paper being different - Bar8 being more related to CHG than Iberia_MN, and Bar8 even more strongly related to CHG relative to Corded_Ware, also implying Iberia_MN more related to CHG than Corded Ware is. I don't know that there's anything about f4 vs D stats themselves that would explain that difference, and as you say, yours are consistent with the previously published.

This is really stuff that should have been in and the resolved in the early print. That's the whole point of the process!

Shaikorth said...

If you meant figure 3. I answered - moderns, such as Nogais, can be left out of a model that is better for ancients. If you meant any such test in general, Nogais shouldn't be left out out of principle and weren't in this study. There is figure S30, and various figures using a global set of modern populations that are better for Nogais than a WHG-EEF modeling.

Davidski said...

So what's the point of the mixture test in Figure 3, if it can't correctly estimate the levels and types of Neolithic ancestry in at least some of the test samples?

Shaikorth said...

They refer to it when they talk about measuring Neolithic genomes' contribution to each other, and for that particular purpose there may be nothing particulary wrong with it - even though it may not get the particulars down which they admit:

"However, it is not possible to infer a direction for dispersal within the Aegean with statistical confidence because both the Greek and Anatolian genomes copy from each other to a similar extent. We therefore see the origins of European farmers equally well represented by Early Neolithic Greek and northwestern Anatolian genomes. "

Davidski said...

Why can't you just admit that their mixture analysis is a pile of crap? It is what it is.

Also, what's your opinion about the formal stats comparison I just posted for Matt?

Shaikorth said...

Basically, fig 3 isn't necessarily wrong for Neolithic Aegeans even though it is for moderns of the same region so I don't think everything in it is useless. Nothing more to that.

The f4-stat thing is interesting. Their samples are public and they give a rundown on their methods on page 49, so someone can test on them directly to see if the results can be replicated. Did you do those D-stats with and without transitions, and could you run f4's too?

Alberto said...

PLEASE, could we just move on from that Figure 3? Really, it's a bad test, not useful, let's forget about it.

The paper had this very surprising Kumtepe4 sample that might be from the Bronze Age! (Thanks, Roy King!) That's a huge novelty.

David, would it be possible for you to find that sample so you can run some tests on it? I agree that some D-stats in the paper look strange so it would be great if you could find Kumtepe4 and see what you can make out of it.

Shaikorth said...

Nitpick: those are not D-stats in the paper, they're f4 stats. The difference in results shouldn't be sizable, but we'll see.

batman said...

@ Tobus

Phrases like "ocean-worthy" and "sailing oceans" referes to the phase when logboats were exchanged with sewn skins or planks - based on a keel that made it possible to steer the boat (ship) and thus "move across the ocean".

There's no evidence claiming that ocean-going vessels - or sails for that matter - were unknown to NW Europe before modern time. That's obviosuly a modern myth, using 'absence of evidence' as evidence of absence.

Moreover, during the last decades we've seen a number of discoveries claiming that both keel-based ships as well as sails did exist during European Bronze Age, at the latest...

Here's a Brtittish expert commenting an image of a sailship, found on bronze-age pot-sherd:

http://www.culture24.org.uk/history%20&%20heritage/archaeology/art397577

“The picture on the sherd is definitely a ship. (…) It is an incised image done PRE-Firing, on a late Bronze-Age / Early Iron Age large storage vessel. Pottery of this period was usually undecorated, so this makes the image even more unusual.

The vessel is probably double ended - unfortunately one end is missing, with a single mast, and a large square sail, the top of the rimsherd being the top of the sail.” (Thorpe, 2014)

Then one may add some contemporary images from Scandinavia, proving that your assumption is flawed and this question far from resolved:

http://www.fotevikensmuseum.se/skane/s12/s12.htm

http://www.visitdenmark.com/sites/default/files/styles/galleries_ratio/public/vdk_images/Attractions-Activities-interest-accommodation-people-geo/History-wikings-royal-denmark/helleristninger-ved-madseba.jpg

batman said...

Another source contradicting your assumption is fond in Caesars logg from the campaign in Gaul. Here he refers to his observations at the coast of Bretagne, where he found some "veneti" (wends) sailing the English channel and its neigbouiring seas, "with great skill":

"... the Veneti both have a very great number of ships, with which they have been accustomed to sail to Britain, and [thus] excel the rest in their knowledge and experience of nautical affairs"

http://classics.mit.edu/Caesar/gallic.3.3.html

Rob said...

batman,

"Moreover, during the last decades we've seen a number of discoveries claiming that both keel-based ships as well as sails did exist during European Bronze Age, at the latest.."

now that is interesting, & definitely makes sense. Even earlier - the BB in the Copper age is clearly a maritime phenomenon

Roy King said...

@Davidski,
To settle the question of whether Kumtepe4 has intrusions (beyond Kumtepe6), from Yamnaya-like steppe DNA or from Kotias-like CHG, could you do f3, f4, Treemix stats using data from the data from the paper (which are public)to statistically test these alternative models?

Davidski said...

The plans are in motion.

FrankN said...

Dave: I subscribe to Roy's request. From the paper, it seems that Kumtepe6 has acquired substantial additional CHG. If true, this would greatly aid in identifying the time and nature of the Anatolian genetic turnover It could furthermore help to understand the genesis of the "Cypriot" element, i.e. the extra CHG without CW-typical EHG that is present around the Mediterranean.

Rob, batman e.a.: When it comes to early shipbuilding, this may be of interest:
http://www.academia.edu/4149354/S._Klooss_u._H._L%C3%BCbke_2009_The_Terminal_Mesolithic_and_Early_Neolithic_log_boats_of_Stralsund-Mischwasserspeicher_Hansestadt_Stralsund_Fpl._225_._Evidence_of_early_waterborne_transport_on_the_German_Southern_Baltic_coast

It deals specifically with the Stralsund logboats - 2 Ertebolle (ca. 4800 BC, incomplete, 8 and 9 m length recovered), and one early FB (3900 BC, 12 m length, the longest European Neolithic boat recovered so far), but also discusses the general Ertebolle/ FB logboat evidence from Denmark and N. Germany.

The logpoat's rumps were hollowed out from single tree stems, thinned down to 1-3 cm width. They incorporated additional wooden elements, especially a transom at the stern. As such, the technology for the boards being heightened with additional planks should in principle have been available. No such construction has been found yet, however, it provides a plausible explaination for regular cavities at 5-10 cm spacing along the upper rims that are frequently found with Ertebolle/ FB boats.
No signs of sails yet (but a lot of Ertebolle/FB paddle finds). All three Stralsund logboats possessed keels of some 5 cm thickness. The boats are generally believed to have beeen seaworthy, at least for travel along the coasts, possibly also for open-water crossing of the Baltic Sea (see link below). For the 1 km crossing over the Strelasund from Stralsund to Rügen, they seem overdimensioned..

http://www.academia.edu/11563227/S._Kloo%C3%9F_2014_They_were_fishing_in_the_sea_and_coppicing_the_forest._Terminal_Mesolithic_and_Early_Neolithic_wooden_artefacts_of_coastal_settlements_on_the_south-western_Baltic_Sea

Davidski said...

Well, Hofmanova's f4 stats involving CHG don't match my D stats involving CHG, so let's wait and see.

batman said...

@ Rob,

The BBC was definitely closely connected to the old, maritime culture of western Europe.

Today there's no expert on Mesolithic Europe doubting the evident use of ocean-worthy vessels during the pioneering phase of the re-population of NW Europe - already.

The founding population surviving Younger Dryas (in the area of Scania) did indeed spread along the shores and to the islands of the Baltic Ocean, as well as the North Atlantic - already 11.500+ years ago.

The spread of the late-glacial Bromme-Lyngby-culture, to the mesolithic sites of the Baltics, Brittain and Scandianvia is indeed very clearifying - as soon as you get the archaeological updates in place. Thus we have a plain and simpel time-line, starting with the onset of the Holocene 11.800 yrs ago, when the first settlers reached the Oslo bay area (Darbu) as well as the west-coast of Norway (Galta 3) 11.700, the islands outside middle Norway at 11.500 (Hitra, Froya, Leka, Vega) - reaching the islands around North cape around 11.300 BP (Sarnes 4).

Populating larger islands outside the coast of Norway - reaching the Shetlands, the Orkneys and Scottland - they obviously developed highly ocean-worthy vessels during the later Mesolithic.

A maritime culture expanding during the mesolithic already - from North Cape to Gibraltar - is by today a pretty well documented fact. (Whatever you may "believe".)

batman said...

Rob et al:

A consequence of these archaeological discoveries is that the traditional knowledge, traded through the old Egyptian and Greek academia and duely noted by Plato - actually claim that the "first mariners" arrived in the Mes from the Atlantic facade - at 11.600 yrs BP (!). Hitting the scientifically correct point in time (with five digits in a row) is obviously beyond guessing and mere incident.

The old civilizations obviosuly carried some very old traditions (schools) of knowledge, that contained important facts and events from the history of their ancestors. As the chronicles of Plato is a widely recognized fact, as a genuine collection of Greek traditions, we may have no choice than to admit that the Greek traditions explaining the 'atlantic culture' as pioneers of tool-making, boat-building and seafaring - was based on historic facts.

Another old site disclosing a skilled maritime culture is the 9.500 yrs old site at the distant island of Stora Karlsö, disclosing a male with hg I2 and, thus, later relatives 'all over Europe'.

A final point:

Recent research proves that the spread of agriculture within the Mediterranean -during early neolithic - was helped by the maritime culture that arrived there a few generations earlier. Thus we see that the descendants of y-dna C/F - such as G, I and J - are operating in collaboration to spread new settlements and populate mesolithic Eurasia, with a variety of 'agricultures'.

Later we find the R1's spreading along the same routes, but with an entirely new sets of cereals, cabbages and cows - to effectively cultivate and populate the vast, open fields of Europe and Eurasia.

Following the traces of the cattle-herders we find their R1a and R1b all along the Rhine and the Rhone, the Wizla and the Donau, as the Neva and the Volga - down to Goa-dal-quivir and the Nile, as well as the Eufrat, Tigris, Indus, Ganges and Yangtse...

Obviously these new dynasties - and their consequent populations - grew to become the larger populations of Europe. Thus we may see a 'cultural evolution' during the late ('modern') neolithic, that continued througout the bronze-age and iron-age Eurasia. (Until the great wars arrievd - with 'foreign masters', illegitimate rulers and 'cultural revolutions' that de-costructed the civilized world of Eurasian antiquety. The result is well known as the cultural and academical 'darkness' of the Middle-Ages, when the Eurasian populations were forced into centuries of submission and superstition - resulting in an academic and historic amnesia.)

The old dynasties (family-lines) of G, H, I, J and N were obviosuly not harmed by the growth of the new, agricultural dynasties of their cousin R1, the founder of the lactase-persistent producers of diary. On the contrary, as the bovine agriculture became a new and steady source of industrious production - promoting trade and transport along lakes, rivers and coast-lines. This job was still donme by the old specialists of boat.building, travel and trade - known as 'the maritime culture' from mesolithic time - still carrying y-dna I, G, J and N.

A qualified guess is that this common origin may explain why the descendants of makrogroup F, as well as "Loschbour" and the IE languages - are to be found all over the place. Still today... ;-)

FrankN said...

Dave:Hofmanova e.a. note (SI, p. 49): "In order to make our results comparable to other recently published paleogenomic studies [40, 41, 81, 102, 104, 135], we limited our analyses to the ∼300K SNP positions utilized in Haak et al. [102].
In addition, to ensure our results were robust to errors in genotype calling and comparable to the reference dataset, the ancient Greek and Anatolian genomes were analyzed using pseudo-haploid calls (using positions with a minimum genotype quality of Q30)."


Could this explain the differences between their f4 and your D-stats?

Ariele Iacopo Maggi said...

On a side note.
If it's found (in the upcoming paper) that there is a zagros hg admixture of >20% in the Levant (Syria, Lebannon) and it's less then 5% in SE Europe then a second wave of farmers with some CHG admixture (possibly the J2/E1b1 folks) but without zagros hg it's necessary to explain the extra CHG in southern europe. If we found that Greece/Sicily are 20% Zagros then a bronze age or a more recent migration is very much confirmed. And no, in no way this all thing put in question the rock solid validity of the steppe hypothesis.

batman said...

@ Frank

Stora Forvar and Motala are definitely direct descendants of Bromme-Lyngby, who survived both the LGM AND the Younger Dryas, contrary to the Magdalenien, Azilien, Creswellian, Hamburgian and Pertunian cultures. Though - as all the others disappearred during the Dryas-periods - along with the mammoths and the red deer - the Bromme-Lyngby-connection is so far the only one with a known, unbroken track-record throughout the late Paleolithic.

The YD survivors of the W Baltics (Skateholm/Bromme 12.700 -> Hasselberga 12.500 -> Lyngby/Ahrensburg/Swidrien 11.900) became ancestral to the post-glacial cultures of Swidrien, Ahrensburg and Fosna-Hensbacka - from which the first boat- and sledge-driven re-population of arctic Eurasia was initiated.

Soon we'll have some more a-dna from the latter.

FrankN said...

Addition: It also seems that their Bar8 (6212-6030 BCE) is slightly younger than the BAR20-I0709 (6500-6200 BC) used by you. Both plot quite closely together in their PCA (Fig. 2), with Bar8 being a bit drawn towards Stuttgart (ultimately WHG), while BAR20 drifts slightly towards CHG. But those drifts are minor, and probably not sufficient to explain the difference between your and their stats.

Maju said...

The only comic relief on this paper is how you have mistreated it, David. This study is most interesting on so many issues that I can't just believe that you decided to abuse it this way. You should reconsider and apologize.

Meanwhile I'm going to write something on my own, because there is serious stuff here, beginning by the fact of minimizing the importance of NW Anatolian farmers on European Neolithic and particularly on their modern legacy (they still seem to have influenced Central European Neolithic, but this one clearly suffered a recession).

Also K in Greek Epipaleolithic: first pre-Neolithic K known in Europe.

Grey said...

Ariele Iacopo Maggi said...
"On a side note."

Depending on the timing it might enhance the steppe hypothesis i.e. if some of this movement was the result of CHG-like people being pushed south by pressure from the steppe.

Kristiina said...

Maju, I agree with you. It was a good point!

batman said...

@ Kristiina

So far the Greek K1 is only paralelled (in terms of periods) by some ancient U/K from mesolithic Siberia and three K1 from later 'mesolithic' sites in Sweden.

So what's the point about them, actually?

Alberto said...

The thing is that the F4 stats and ADMIXTURE seem to agree about the high affinity of these Aegean samples and CHG. It's the PCA that doesn't quite show this high affinity and place them like regular Anatolian Early farmers.

Let's see if the genomes are available and we can see more tests to find out more about this.

Maju said...

@Batman: oldest K1 (K1c specifically) known in Europe and it is pre-Neolithic. So far I was assuming (and probably most others) that all K and K1 in Europe was strictly Neolithic. Well it's not the case.

More stuff: European mainline Neolithic's origins seem now to be decided in favor of NW Anatolia and not Greece, what is unexpected, considering the archaeological data. However Greek Neolithic seems to have made a much greater impact in SW Europe, surely via Cardial culture, and very particularly among Basques, which are the only population that show up with clearly stronger Greek Neolithic apparent ancestry and much weaker Anatolian Neolithic one.

We also see that in the long run the influence of Greek Neolithic appearance of ancestry is bigger than than of Anatolian one, suggesting that maybe Cardial Neolithic offshoots impacted the continent via the Atlantic.

Davidski said...

@Frank

Could this explain the differences between their f4 and your D-stats?

It can't. There's a problem somewhere. Either their formal stats with CHG are wrong, or mine are.

They can't both be correct.

Davidski said...

Unless of course there's some awesome substructure within the early Neolithic Barcin population. But I doubt it.

Davidski said...

@Maju

You don't have a clue what you're talking about.

Before you can comment on these sorts of papers you need to learn how to analyze the data yourself, preferably at a level on a par or better than what is normally seen in them.

Maju said...

Waiting for your lesson, professor.

Sarcasm intended, of course, because you are saying NOTHING: just disqualifying with ad hominem attacks instead of reasoning. And well, that is YOUR big problem.

Davidski said...

What's your opinion on the discrepancy between the stats Professor Maju? See here...

http://eurogenes.blogspot.com.au/2016/06/comic-relief-from-hofmanova-et-al-at.html?showComment=1465437986601#c4050738047622613249

FrankN said...

@Dave: "Unless of course there's some awesome substructure within the early Neolithic Barcin population.
Your K10 has the Barcin samples varying between 0% (BAR11-I1099) and 13.2% CHG (BAR2-I0708). MenM4-I0727 is shown with 5% EHG and 5.2% WHG. SW Asian goes up to 5.9% (BAR17-I1580). NE Asian, usually at 0%, peaks at 4% with MenM2-I0724 (3% with MenM3-I0726, here in addition 2.7% Oceanian). BAR14-I1102 is shown as 1.9% Sub-Saharan, 1% EHG.
Not sure whether that qualifies as "awesome", but it looks that EN W. Anatolia, or its source population (Levante?), already attracted quite a few people from across the globe. So, a certain degree of variation in D/f4-stats results, depending on the specific sample selected, should be expected. Whether such substructure alone, however, can explain the differences between your and Hofmanova e.a.'s results, I am not sure. Anyway, the somewhat "multicultural" character of AnatNeol is intriguing and deserving a closer look.

@Ariele, in relation to your "side note": Trying to figure out the cultural genesis of CW, I did some research on the prehistory of cord-incised pottery. Its earliest evidence is from the Japanese Jomon Culture, ca. 11,500 BC, and it becomes particularly common in Middle Jomon from the 7th mBC onwards. At that time, it also appears in Eastern India (Assam and Gulf of Bengal), presumably as a Jomon import.
http://www.himalayanlanguages.org/files/hazarika/Cord%20impressed%20Pottery%20in%20Neolithic%20Chalcolithic%20Context%20of%20Eastern%20India%20by%20Manjil%20Hazarika.pdf

Intriguingly, around 6600-6400 BC, cord-incised pottery is making its entrance into the Levante, more specifically NW Syria. To my, admittedly untrained eye, Indian and Syrian cord-incised sherds look quite similar, raising the possibility of cultural (and genetic?) contact, either direct or via the Persian Gulf/ Zagros.
http://www.academia.edu/2557253/The_cord-impressed_pottery_from_the_Late_Neolithic_Northern_Levant_Case-study_Shir_Syria_

OT: Another early occurrence of cord-incised pottery is Mesolithic Goncharka ware from the Amur Basin in E. Siberia. The earliest European pottery, Elshan ware from the Samara region (ca. 7000 BC), however, lacks cord incisions/ imprints, but they appear, as one of several decorative elements, in the Upper Volga and Valdai cultures after ca. 6000 BC, and in the S.Finnish/ Karelian Sperrings Culture (5.600-5.200 BC).

While it is tempting to speculate about Trans-Eurasian innovation flows (either the Northern, or the Southern/ coastal route, or both), clear, unbroken technological and stylistic chains have not yet been established (see the following paper for further details and discussion):
http://www.academia.edu/2116502/Early_pottery_in_Afroeurasia_Origins_and_possible_routes_of_dispersal

Davidski said...

For those f4 and D stats to be both correct, the substructure within early Neolithic Barcin would have to be really something, ranging from early European farmer-like to mostly CHG.

I can't see this as being a plausible explanation.

Chad Rohlfsen said...

I've sent a request for the samples. Hopefully, we can resolve this.

Chad Rohlfsen said...

Frank,

That East Asian, Papuan mess is what shows up in the low coverage samples. This is a common occurrence and should be ignored. Only pay attention to the ones with the best coverage.

capra internetensis said...

@FrankN:

AIUI wrapping a wooden paddle with cord stops it sticking to the clay when you shape the pottery, so it is a pretty straightforward functional > ornamental path. Unless the ornamental motifs are clearly related I'd be cautious of drawing any conclusions about recent connections.

Rob said...

@ Batman

(Sorry off topic of thread)
So one can accept that seafaring in Europe is much older than appreciated.
But how does this, then, prove that European Mesolithic groups all descend from a West Baltic refuge, esp. when the Baltic was under 300 feet of Ice during the LGM, and was not repopulated until 14 - 12 kya, then appears to have again become depopulated with the Younger Dryas (at least most parts of northern and NW Europe), to be again re-populated.

Do you think the idea that post-LGM was repopulated from southern France, Ukraine & Balkans is a serious error on the part of almost every archaeologist dealing in the matter ?

Onur said...

@Davidski

At his own blog Luis Aldamiz (aka Maju) is insulting you with a statement such as this one:

"elsewhere someone was being a crybaby about the Polish sample (may well be an error) or the Kalmyk sample (who are obviously most related to East Asians, not used here) but those are minor issues."

Here is its screenshot:

https://drive.google.com/file/d/0B9n4j3PQ81RcVEpUdWRraS1ldlE/view?usp=sharing

FrankN said...

@Chad: MenM2-I0724, with 4% NEA, is indeed low coverage (56,787 SNPs), but MenM3-I0726 (3% NEA, 2.7% Oceanian as per Dave's K10) has a decent coverage of 236.641 SNPs.

@Dave: I read you, and I suppose you are already busy running stats and preparing a new entry that looks at Barcin and Kumtepe in more detail. I could imagine the Hofmanova team, some members of which obviously follow this blog, doing the same. I am curious about the outcome. Let's hope we don't end up with more bickering, but instead gain some more insight into the population dynamics of EN/MN/CA Anatolia, which seem crucial to understand Mediterranean demography as a whole.

@Maju: The idea of Cardial offshoots influencing the Continent via the Atlantic is intriguing, and I don't want to rule it out in any way. In fact, the spread of Megalithism, apparently out of Brittany but possibly with Iberian roots needs to be seriously considered here.
However, note that there are other more or less well documented Cardial paths into Central Europe, especially via the Rhone and her tributaries:
(1) Archeology suggests significant Cardial influence on the hunter-pastoralist La Hoguette Group. One of the best documented La Hoguette sites, Stuttgart-Bad Canstatt, is just 3 km away from Stuttgart-Viehäusener Hof, from where the Stuttgart LBK sample originates. Interaction between La Hoguette and Rhinish LBK is well documented, e.g. for Nieder-Mörlen, an exceptionally large LBK setlement (up to 1,000 inhabitants) N of Frankfurt, next to the Bad Nauheim salt springs.
(2) The generally poorly researched Rössen Culture originated west of the Rhine, but spread towards the Middle Elbe (Rössen, a suburb of Leuna, is located 30 km W of Leipzig). There is some indication of Rössen having had a more pastoralist focus than LBK, which presumably made it more resilient to climate fluctuations around 5,000 BC. While the geographic origin of Rössen people is unknown (French archeology apparently is completely ignoring it), Brotherton/ Haak 2013 have demonstrated their mtDNA pointing towards the southern Gulf of Biscay.
(3) The MN Michelsberg Culture (MC) is generally assumed to have formed in the Paris Basin as a fusion of LBK and Cardial-influenced pastoralist groups such as La Hoguette, with additional influence from Brittany. MC quickly spread to the Rhine, and subsequently across much of Germany up to the Western Harz and Western Bavaria. MC is credited with neolithicising the British Isles, the Netherlands (Swifterband), and Scandinavia (c.f. the Gökhem Swedish MN aDNA), and regarded as a catalyst of the Funnelbeaker phenomenon.

Grey said...

Gioiello

"Very likely hg. J (J1 and J2) came from Caucasus and not Middle East.
Everything is explained."

I've been wondering that for a while.

truth said...

The Polish sample is not an error, they simply placed the Ashkenazies (polish jews) on the map in Poland, but I understand that can be misleading, since they genetically have nothing to do with ethnic Poles.

Maju said...

@Davidski: I was commenting on the entry as such which is clearly just an angry rant based on bad pretexts and some sort of generic dislike for something in the study, which I don't understand but you probably do, as it is your own emotion. When I commented there were already almost 150 comments and I don't have time to read them all.

To your question: I don't know. I tend to think that f4 is more reliable, more professional maybe, than D-stats but I'll let it at that, because otherwise I'd have to work too much, revising all the info on those algorithms, only to satisfy your ego (and have no certainty I could come up with an explanation), Maybe if you did not offend the researchers so much, they might have come up with an explanation, as Saioa Sánchez (?) herself tried to do above on another issue. I am not even interested in how Satursbilia compares here: those Epipaleolithic Caucasian genomes are somewhat weirdo: related to the Caucasus-Baloch-partialIndoeuropean component but also with lots of "basal Eurasian", which those tend to be low in, so their relations are probably bound to be weird just because they are outliers with their own unique makeup.

I think there's a lot of other interesting stuff in that paper and much more central: the origins of European Neolithic peoples and, through them, of modern Europeans.

Davidski said...

I tend to think that f4 is more reliable, more professional maybe, than D-stats.

Maju, you really shouldn't comment on genetic studies until you get a lot better at this.

Maju said...

@FranN: Nowadays I'm more interested in Bell Beaker (also with SW roots, it seems) and a bit less on Megalithism but it may well be a combo of both. We won't know for sure until enough Western European samples from those periods are sequenced and compared. But already seems quite apparent that even in "Late Neolithic" (LN) samples (mostly from Germany, but also in Spain) are not anymore just the primary Neolithic settlers but rather something a large bit more mixed with the Paleoeuropeans (HG). Only Ötzi retains the EEF makeup so late, if I'm correct, so probably Italy was out of this secondary process. What is unclear is if that extra HG element comes from a single origin or a few selected ones (i.e. a "mestizo" population expanded from specific areas in the West) or is a more generalistic absorption of ex-foragers living at the margins of Neolithic society, or even both cases, depending on geography: for example local admixture in Iberia and single origin expansion in Germany with Funnelbeaker...

We will probably get to know in due time, as soon as due attention is paid by researchers to the Western or Atlantic European arch in those phases.

Re. Brittany, I'd think it has a more French than Iberian origin, although there is clear connection with Iberia via the sea to Portugal and also via the Basque-Aquitanian region. It's certainly (along with nearby West France areas like Upper Normandy, Pays-de-la-Loire) at the origin of Irish Sea neolithic (Cornwall, Wales, Scotland and probably Ireland too), while most of English Neolithic is instead from further North (Nord-Pas-de-Calais and maybe nearby areas in Picardy and Belgium), with clear Danubian (LBK) influence and non-dolmenic. I was recently watching a good documentary on Neolithic and Megalithic Britain, which explained that there were wars, probably between these two ethnicities in Western England, and I got the feeling from some of the details, that the Westerners (dolmen builders, megalithists) won, at least partly, leading to the development of Stonehenge.

But for example Funnelbeaker seems more centered in Denmark (or maybe Low Germany-Netherlands?), so there may be several focuses, linked to each other by trade and "religion" (dolmenic megalithism) and maybe also by a feeling of ethnic difference vs. purer EEF groups, which in any case expanded in the case of Germany and nearby areas but probably not or not so much in Iberia and France (?). For example Treilles were pretty much "pure EEF", yet fully integrated in dolmenic megalithism and Bell Beaker.

So it's probably quite a complex mosaic and not a simple generalistic rule.

"Archeology suggests significant Cardial influence on the hunter-pastoralist La Hoguette Group"

Yeah, La Hoguette are quite clearly Epicardial and we know nothing as of now about their genetic makeup. Yet they were almost as influential as the well researched LBK peoples, from the continental branch.

As for (2) and (3) I'm getting notes from what you say because I don't have the most clear picture right now. Rossen replaces early LBK but seems (if I recall correctly) continuity of it, but Michelsberg is a clear march to the south from the Middle Rhine or somewhere over there (I have to update my knowledge) that pushes against Epirössen and seems to drive them out totally.

It's a very complicated constantly evolving ethno-cultural landscape. Very far from simple one-size-fits-all answers. Fascinating yet daunting.

Chad Rohlfsen said...

Frank,

That 236K is out of like 2 million. In David's test it will likely be significantly less than that and in the lower end of coverage. Perhaps, he can give the SNP count in the test.

What I can tell you about my runs is that only I0707,8,9,45, and 46 are the only ones over 90% coverage. I0726 is at less than 25% coverage of my 122k, and I0723 is less than 10%. The two you cite aren't worth citing.

Chad Rohlfsen said...

Correction, I0724 is less than 10% coverage.

Chad Rohlfsen said...

What makes you guys think la Hoguette is Cardial and not Danubian related? Please provide evidence.

Maju said...

EPI-Cardial, Chad, i.e. less "settler" and more probably "mixed" than the original Cardial. I'm overloaded to search for papers or articles right now (DIY: I'm sure that there is some info in the Internet although probably not too much and some in French/German) but I'm quite certain that re. La Hoguette there are two complementary explanations none of which is Danubian at all: it is just one Epicardial culture or group more, as so many others, or is one of those frontier cultures like Limburg, which might have been done by HGs converted to farming (or both at the same time). Personally I treat it as one of the many Western secondary branches of Cardial with all the cautions about the genetic affiliation of their authors as any other original group.

FrankN said...

@Chad: You certainly know more about aDNA quality issues than I do, so let's put possible NEA/ Oceanian admix in AnatNeol aside for the time being. But high quality I0708, with 13.2% CHG admix acc. to Dave's K10, certainly deserves a closer look. The same applies to I0746, which Dave's K10 has at 2.6% CHG, 2.6% EHG, and 2.2% SW Asian.

On La Hoguette, this is probably the best source for readers that don't understand German or French (scroll also down to the excerpts below the main article:
https://www.researchgate.net/publication/262187474_Der_La-Hoguette-Fundhorizont_in_der_Wilhelma_von_Stuttgart_-_Bad_Cannstatt_Anthrakologische_archaopalynologische_bodenkundliche_malakozoologische_radiometrische_und_saugetierkundliche_Untersuchungen


@Maju: BB will be clearly interesting, especially as the eponymous Beaker obviously sets forth CE traditions (CW, FB), while the chronology points towards an Iberian origin (but have dates been checked for resevoir effects - it's fish-eating country, today as most likely 5 millenia ago?). Intriguing here is also that the "Beaker package" seems to have its closest precedence in Lesser Polish CW, with daggers and arrowheads as dominating grave goods in contrast to the otherwise typical battle axe (c.f. Rob's links towards the end of the CW women entry of late May). There is quite a bit of "Iberian" mtDNA already showing up in CW, e.g. K1b1a1 in Oblaczkowo [RISE1] and El Mirador [I1272/MIR 2], or H10e in Eulau 2800 BC, and Bom Santo (POR) 3735 BC.

Ötzi is actually actually among the quite interesting "minor" findings of Hofmanova e.a. They show him as more closely related to Aegean than CE MN. You shouldn't qualify Ötzi as Italian, at best as Tyrolean. But most likely, he originated from the Mitterberg area S of Salzburg, which supplied the copper for his axe, and also provides the best isotopic match. The quite early emergence of copper mining in the Eastern Alps (Mondsee Culture etc.) has long been enigmatic. One of the theories has been immigration from the Southern Balkans (Vinca, possibly even Varna Culture); Hofmanova's analysis lends further support in this respect. [As Sardinia has also been a major copper mining area, with copper smelting evidenced far earlier than on the Italian mainland, one might even speculate that Sardinia's "EEF-ness" is in fact not owed to neolithisation, but to MN copper miners]

Chad Rohlfsen said...

Okay, well I've looked at several things on la Hoguette over the last couple of years. A lot of the stuff I see these days (dated 2012-2016) considers them as a group related to LBK, due to being a minority among LBK finds along with tempering being the same and no relation to Cardial or a ceramic HG pop being shown. So, I was just curious about what sources you drew the conclusion from.

FrankN said...

Chad - one issue about La Hoguette is that they precede LBK. The Bad Cannstatt site is dated from 5480 BC onwards, LBK reached the Rhine earliest around 5300 BC. LHG animals, sheep and goats, are typically Mediterranean (and not native to Europe, but imported from E. Anatolia/Caucasus/Zagros), and rather uncommon withion LBK that focused more on cows and pigs.
Another point in case is poppy, which was only grown by Rhinish LBK (not Elbe-Saale, Bohemia, Hungary). As poppy originates from the Mediterranean, LHG is generally assumed to have transferred poppy cultivation to Rhinish LBK. [This has btw. given rise to the discussion how "mesolithic" WHGs in fact were - they had already domesticated poppy, regularly coppiced hazelnut bushes, duck bones are occuring a bit too frequently and widespread in WHG/SHG assemblages to just have stemmed from hunting, and the question of pig domestication in Mesolithic Europe is under intensive review...]

Chad Rohlfsen said...

Frank,

What are your sources on those dates? That seems off. Tempering is also an issue, when LHG is the same as LBK and not Cardial. Also, you have bad information about sheep and goats in LBK. Sheep and goats are out-numbered by pigs less than 30% of the time, with SE Germany having more sheep than cattle. Even as far away as Hungary, sheep are more numerous than pigs. Even in the EN Koros, sheep and goats make up 2/3 of all domestic animals. Pigs are almost non-existent in Koros remains (3%).

Gioiello said...

It is known what I think about the R-L23 found in aDNA at Samara, just that they came from the Italian Refugium, where we have so far the oldest ancestor R1b1a (Villabruna: 14000 YBP).
I have said that from those haplotypes some descendant may have been born independently from the Western European haplotypes and perhaps migrated to Western Europe with the migrations of people (we are waiting that also about my R-L23-Z2110* some light is done with the tests we are waiting for), but about R-L277 I wrote a lot also on Anthrogenica, saying that the highest variance is in Italy (and also amongst Armenians). After of course I was banned from that forum.
I still wrote that amongst the R-L277 there was a clade completely Jewish, that found from the R1a-Levites paper with the first deep SNP test, and that a Moroccan Jews would have the first DYS392=13 being 14 the modal, but that isn't true, because in the smal's tree at R1b basal subclades of FTDNA, we have that:
1) not only smal haven't resolved the question I posed him about the fact that Mattoli is negative for L277 (confirmed from a SNP test at FTDNA and from his Big Y), and for that I said that his haplotype was the oldest
2) not only the two subclades BY3294 (Italian Stacy/Stasi and Lebanese I Lakkis) and BY3716 (Mattoli haplotype largely found so far only in Italy) are unified from smal with the SNP in the Multicopy region Y:22444952 (G>C), and I am waiting that also YFull recognizes these SNPs which are certainly reliable
3) that the cluster Y4366 has not only Armenians and also a Tuscan from the 1KGP and Jews only in late subclades
4) that that which seemed a Jewish cluster is really oldest in one Italian and one German
_f2d. R1b-Z2103, Z2105 > Y4362, L277 > M12135
B23717 Antonino Scafidi Moreci b.~1786, Altavilla Milicia Italy R-M12135
12 24 14 10 11-16 12 12 13 13 14 28 17 9-10 11 11 25 16 19 28 15-15-16-17 12 12 19-23 16 16 16 18 36-37 13 12 11 9 15-16 8 10 10 8 10 11 12 22-23 16 10 12 12 17 8 12 22 20 13 12 11 13 10 11 12 12
200183 Amon Bumgarner (1700s) Germany->PA->NC Germany R-M12135
12 24 14 11 11-15 12 12 12 13 14 28 18 9-9 11 11 25 15 19 30 15-15-15-15 11 10 19-23 16 16 18 17 34-34 11 12
and the Jewish haplotypes come out from more recent bottlenecks.

Now it is a little believable that Italy always has the oldest haplotypes supposed always come from elsewhere, being "more a population sink than source" as a kind person said on "Eurogenes blog", and I'd like to know the "sink country" where he comes from, but it is forbidden asking that to Davidski.

Thus, meanwhile we are waiting that some light from the next tests is shed about the hypothesis that the most part of the Western European R-L23 may have come from Alans or other nice peoples, I show these few remarks to you all.

FrankN said...

@Chad: Do you have a comprehensive source at hand on LBK faunal assemblages, from which you derived your 30% figure? Pigs are of course tricky, because it is difficult to tell them apart from wild boar. Recent Romanian analysis has demonstrated that several samples that were morphologically determined as wild boar in fact had Anatolian, i.e. domestic pig mtDNA.
It is well known that LBK subsistence models differed considerably from region to region. In SW Germany, e.g., meat (including game) consumption was substantially higher than in the Elbe-Saale region - unsurprising when considering that the Elbe-Saale region has the best soils in all of Central and Western Europe, while SW Germany is more suitable for transhumating animal husbandry. Climate and terrain also affect the animal husbandry mix, hence Köros isn't representative for LBK as a whole.

Almost all German LBK sites that I am aware of show cattle as dominating domestic animal. This includes Hienheim on the Danube (upwards from Regensburg), clearly in SE Germany, with 44(+12) cattle, 13 (+5) sheep/goat, and 13(+13) pig bones (numbers in brackets relate to uncertain identifications, esp. concerning the domestication status) [1].
For Elbe-Saale LBK, the best documented site seems to be Karsdorf: 12 cattle, 8 pigs, 8 sheep/goat [2].

When we move to the LBK-LG contact zone, the picture changes and pigs become more dominant:
- Müddersheim, Rhineland (35 km W of Bonn): 133 cattle (72%), 32 pigs (17%), 19 sheep/goat (10%) [1]
- Vaihingen n. Stuttgart, Younger LBK: 44% cattle (66% in Older LBK), 37% pig, 19% sheep/goat [3]
- Herxheim, Palatinate: 107 cattle, 111 pig, 62 sheep/goat [4, Fig. 11]

Contrast this to LHG showing no evidence of cattle or pig husbandry.

Further sites are discussed in [5]. Pay particular attention to p.295ff, where the Alsace, and possible LHG-LBK relations there, is discussed. The Alsace shows two LBK subcultures, with the SW differing in burial habits, as well as by preferring sheep/goats over pigs.

ON LHG in general, chronology, and poppy see [6]. A more detailed look into early poppy cultivation in Impresso/ Cardial contexts, which predates cultivation in the Rhineland (La Marmotta IT, La Draga ESP) is provided by [7] and [8]. Maybe Italians and Spaniards reading this can provide more detail on the a/m sites and WestMed poppy cultivation in general.

[Sourcss will fullow in a separate post].

FrankN said...

Ass addendum to my previous post, here the sources:

[1] https://www.google.de/url?sa=t&rct=j&q=&esrc=s&source=web&cd=14&cad=rja&uact=8&ved=0ahUKEwjf3KWZlZ3NAhWlbZoKHY2zCyk4ChAWCCwwAw&url=https%3A%2F%2Fopenaccess.leidenuniv.nl%2Fbitstream%2Fhandle%2F1887%2F28108%2FAnalecta-praehistorica-leidensia-XI-1978_005.pdf%3Fsequence%3D1&usg=AFQjCNGSfff95ULmO10VmC5eghyQ7lunag&sig2=BBomcCxXqz86BudOHyJAlA&bvm=bv.124088155,d.bGg

[2] https://openaccess.leidenuniv.nl/bitstream/handle/1887/19045/05.pdf?sequence=13

[3] http://www.tandfonline.com/doi/full/10.1080/00438243.2013.820649

[4] https://www.researchgate.net/publication/285584342_The_LBK_settlement_with_pit-enclosure_at_Herxheim_near_Landau_Palatinate_First_results
[5] https://books.google.de/books?id=jjmgBwAAQBAJ&pg=PA36&lpg=PA36&dq=LBK+fauna&source=bl&ots=IhTASa4F0V&sig=4yCYjGkh_Oa04wmRQo0jFKc3CTo&hl=en&sa=X&ved=0ahUKEwjs1p-Xi53NAhUBXCwKHQK5CZgQ6AEINTAE#v=onepage&q=LBK%20fauna&f=false

[6] https://books.google.de/books?id=2PAkBwAAQBAJ&pg=PA101&lpg=PA101&dq=La+Hoguette+Group+cattle&source=bl&ots=HTbdQyACOx&sig=ohaU_WMZg3MBOUqDCSxVtuj-u2Q&hl=en&sa=X&ved=0ahUKEwj62vaaqp3NAhWEF5oKHd6qAV0Q6AEIHjAA#v=onepage&q=La%20Hoguette%20Group%20cattle&f=false

[7] http://www.aggsbach.de/2011/07/papaver-somniferum-during-the-european-neolithic/

[8] http://archaeology.about.com/od/Domesticated-Plants/fl/Opium-Poppy.htm

FrankN said...

Here is a bit more on the poppy finds in La Marmotta, NW Rome, dated to 5700 BC, and as such earlier than Rhinish LBK.
https://books.google.de/books?id=_ehBAQAAQBAJ&pg=PT13&lpg=PT13&dq=La+Marmotta+poppy&source=bl&ots=qRgAsskjnp&sig=PUE-zQW_3opXjMV577Dkn7wBodU&hl=en&sa=X&ved=0ahUKEwiN9samsJ3NAhUGkywKHaEtBM8Q6AEIODAG#v=onepage&q=La%20Marmotta%20poppy&f=false

The following paper proposes in addition to poppy also naked barley as a presumptive LBK import from the Western Mediterranean.
http://www.raco.cat/index.php/Rubricatum/article/viewFile/269300/356849

Maju said...

@Chad:

In your favor:

→ https://www.academia.edu/7477413/La_Hoguette_Limburg_and_the_Mesolithic_Some_questions

Against:

→ http://arheologija.ff.uni-lj.si/documenta/pdf30/30gronenborn.pdf

This last one is particularly interesting because it points to "Mesolithic horticulture" in France, Switzerland and SW Germany. However it is a bit old (no date typed but from bibliography, it's probably of 2003).

In the first one, they actually criticize the latter but their archaeological evidence is from two LBK sites ONLY and some of the criticisms against "Mesolithic agriculture" seem burning nails, notably:

The studies lend too much importance to ruderalsas anthropogenic indicators, since they occur naturally in some small open biotopes.

Well, some rare occurrences are substantially different from common occurrences. So something was probably happening in Western Europe before the mainline Neolithic arrived. We know for instance that they "herded" snails and brought them to Ireland from the Pyrenees (either in the Mesolithic or the Neolithic but they did and looks like a Mesolithic thing to do).

I can't conclude anything but I would agree with the first paper in expecting, especially in Western Europe, a more complex Neolithic landscape than we usually imagine. Genetic evidence supports:

(1) intensive admixture with Western aboriginals upon arrival of mainline Neolithic (which explodes into many distinct local cultures/groups). This may have been aided by the peculiarities of Oceanic climate, not too favorable for Mediterranean agriculture (too wet).

(2) oldest known modern-like mtDNA pools in the area between Navarre (Paternabidea) and Burgundy (Gurgy), roughly most of modern France but also parts of Iberia.

So for these reasons and others (dolmenism, funnelbeaker, bell beaker) something must have happened in Western Europe that was not just the "one size fits all" explanation of mere expansion and colonization. Neither EEFs nor early Indoeuropean groups can explain all the novel stuff we see in the region, which must have experienced processes of its own affecting decisively the modern genetic landscape. These traits include surely things from LCT+ to R1b-western, passing by mtDNA H (Western subclades: H1, H3, H4 and H7 at least, which are also found in North Africa at high frequencies, where they cannot be attributed to Cardial roots). The how exactly this happened is still to be unraveled and will need extensive research of Western European archaeogenetics.

Maju said...

@FankN:

"You shouldn't qualify Ötzi as Italian, at best as Tyrolean".

That would be misleading: he clearly belonged to Remedello culture, which was a North Italian population, much more extended than just South Tirol. The Alpine areas had only recently began to been settled in that period AFAIK, so there was no Alpine specificity but demographic and cultural dependence on the more fertile regions north or south of them.

"But most likely, he originated from the Mitterberg area S of Salzburg, which supplied the copper for his axe, and also provides the best isotopic match".

Then how do you explain his Remedello cultural belonging? Was he an exile that had adopted his hosts' cultural elements? Or did his family migrate northwards to become miners in a sort of "copper rush" of the time?

Gioiello said...

Maju, I am glad that you sometime remember being a quarter Italian! I don't reply these people and their "sink countries". Let them translate "Villabruna" in their languages!

Maju said...

@Gioello: I try very hard for my roots not to get in the way of proper reasoning, neither my quarter Italian, nor my 60% Basque nor anything else. First of all I am a person and all the rest are just circumstances.

Gioiello said...

Of course we all may be wrong in part or in all, but we should be neither a "person" with all its passions, but the "reason" itself, because a scientist searches for the truth and he'll get it more and more if he "reasons".
It doesn't seem that many people on this blog and elsewhere (but above all in the papers and the universities and the funds) are so. Only one person/reason all over the world wasn't surprised of the R1b1a at Villabruna: me.
P.S. Please, Maju, write "Gioiello". The "i" isn't an optional in Italian.

Nirjhar007 said...

Gioiello,

Tell me the truth. No theories. When did IE's come to Italy?.

Gioiello said...

Nirjhar007, I have written a lot also about that, but of course for languages we won't be able to get a Villabruna sample, thus we may only get theories.
Amongst the many theories about the origin of IE languages there was also the great Italian linguist Giacomo Devoto who thought to an origin in Central Europe, and his was the westernmost one.
I have a paper of a Dutch amateur linguist (but we all are amateurs) who speaks of an expansion from Adriatic of the languages of the Italic/Celtic stock to Iberia in the third millennium BC, thus not only a migration of 7500 years ago from Itay to Iberia,, but also a later one, and that would explain many things also about the R-L51 subclades.
Mallory&Adams, not me, said that Italy has the most varied presence of IE languages. That Latin descends from the pile dwellers of Adriatic and Northern Italy is demonstrated from its words like porta/portus etc.
Thus I should write a book, but I have no time now, but to think that IE languages were the languages of the R1b (and very likely also R1a) hunter-gatherers from the Italian Refugium doesn't seem absurd to me. And that centum languages (the oldest ones) are linked above all to R1b and the satem ones to R1a is easy to think. If Samara derived from R-L23 from the Italian Refugium we may think that they brought there the future satem languages.
Of course I think that the genesis of the IE languages is older than it is thought and Alfredo Trombetti thought Rhaetian-Etruscan as intermediate between an IE and a Caucasian language.

Nirjhar007 said...

Thank you for writing it . But if you think you should write a book. You really should do it.

Chad Rohlfsen said...

Maju,

That is one that I have read. The arguments I've seen against that site noted by Frank is that it is a site with mixed layers. I think they even found Bell Beaker stuff going down into an early layer. I'll have to dig for that one. Combined with the fact that La Hoguette always appears in LBK sites and immediate surroundings. Never is it far away from LBK sites. The fact that La Hoguette creates pottery much more like LBK than the Cardial (simple design aside), I believe along with others, this is probably more due to EN contacts and or trade between Danubians and Mediterraneans near the Adriatic and Alps. I think it would take some special pleading to place a sea-based Cardial group in Western Germany at about the same time as they get to Spain. Even southern France doesn't really look to take off until 5300-5100 BCE.

The Mesolithic roots of La Hoguette have pretty much been thrown out since 2012 or so, but there certainly is WHG admixture into the MN. Their effects on the culture of the farmers really appears minimal.

Matt said...

Although I don't think they're really to be taken literally / seriously, actually took the values for the Table A.IV values for donations and used then to generate a clustering analysis: http://i.imgur.com/travRUi.png

Even though the values are not literally realistic at all, it does seem like they somehow capture the relative structure of how the modern population cluster together, surprisingly more than I would think, though some positions are badly wrong. So it is systematic in modelling similar populations similarly, at least.

(One of the weird things (in amongst it all being weird) is how the more "Southwest Asian" the Eurasian ancestry in a population is, the more LBK related their method assigns to it, systematically - Ethiopian_Jew - 50 LBK, Kikuyu - 18, Yemen - 32. Doesn't make a lot of sense to me as to why that should be.)

batman said...

@ Rob

"So one can accept that seafaring in Europe is much older than appreciated."

Sure. There's even archaeological repports claiming that seavoyages were performed during the middle Paleolithoic (MIS 5/Eem).


"But how does this, then, prove that European Mesolithic groups all descend from a West Baltic refuge"

Who did ever claim such kind of logic?!

The point I made was simply that the Baltic refugia of Younger Dryas is, by now, proven to have been an obvious descendant to this maritime culture. One may even add that they, AFAIK, are the only proven group from the late Paleolithic that had maintained and developed this maritime tradition - throughout the terminal phase of the last ice-age.

Add the genetic studies revealing a strong, demografic bottle-neck during the YD and we end up with only one culture known to have survived the last, worst and terminal phase of ice-time - and we end up with the Baltic refugia as the only known origin of the described bottle-neck, as well as the following mariners.

"...the Baltic was under 300 feet of Ice during the LGM..."

Any proof of that?
Sure you haven't been misinformed?

Obviously you've been missing the last decade of updates from the wqide array of geologists, glaciologists and paleo-biologists that works with these questions, professionally. Some of these repports, clearifying this misconception, I've been linking to already (above).


"was not repopulated until 14 - 12 kya, then appears to have again become depopulated with the Younger Dryas (at least most parts of northern and NW Europe), to be again re-populated."

That's not based on evidence, but on the old pseudo-scientific 'deduction', where "abscence of evindence" becomes "evidence of abscense".


batman said...

@ Rob

"Do you think the idea that post-LGM was repopulated from southern France, Ukraine & Balkans is a serious error on the part of almost every archaeologist dealing in the matter ?"

Unless the archaeologists you refer to are quarternary geologers too they don't, normally, have a clue. AFAIK there's hardly any archaeologist questioning the outdated picture of "ice-time" and "The Scandianvian Ice Sheet" that has been, incorrectly, constructed by the geologists of the 20th century, to cover the entire Fenno-Scandia as well as the entire Baltic Ocean and large areas of its southern shores. Later discovceries of ice-free areas, plants and (larger) animals existing DURING the entire LGM (17-25 kyr BP) within several areas of Fenno-Scandia - some as far north as Middle Norway...

Since then a number of geological and paleo-botanical results have proven that this ice-sheet-model is dead wrong. Today we know that the "giantic ice-sheet" drawn by Anderssen and Burns 1985 is out of sync with the laws of termodynamics, hydrology and meterology. Moeover, it's directly contradicted by new discoveries and dating-methods...

For obvious reasons we can't aspect the normal archeologers or geneticians to encounter and disaprove of stetements commming from the quarternary sciences, even when they are doubtful and/or still discussed amongst geologists. But we may aspect that they both are able to discriminate (geological) facts from mere hypos and 'common opinions' - however consentius.

The populations existing in Eurasia - from Spain to Siberia - before the LGM, is said to have been "very homogenious" population in the arctic part of the Paleolithic world. We also know that they survived the LGM in SEVERAL places, to reunite during the Billeroed/Alleroed.

The genetic bottle-neck forming the extant populations did NOT happed during the LGM, but during the Younger Dryas - when several arctic species and humans - except from a handfull - went extinct.

Thus the refugiants surviving the long (but not extremely cold) LGM are found along the Atlantic Facade, forming the Billeroed and Alleroed populations, from which a few, only, came to survive BOTH Dryas-periods - in the Western Baltic.

Do we have any evidence that similar populations survived the last and terminal phase of the late Paleolithic in France, Spain, Italy, Greece, Balkans and Ukraine?!

Are there any proof of migrations from any of these areas - into Europe - after the onset of the Holocene?

If Pinhasi et al is rigth about a genetic bottle-neck at the end of the LGT, we have to recon that A terminal refugia were located. The archaological evidence of a SW Baltic refugia is - by today - very clear and factual, connecting the late-paleolithic Hamburg-Bromme-culture to the early-mesolithic Lyngby-Ahrensburg-Swidrien-culture(s).

Consequently it's an error to use the long-but-not-so-cold LGM, rather than the short-but-deeply-cold Dryas, as the ancestors of the 'modern' version of the actically adapted branch of the AMH, also known as 'Caucasians'. The refugias surviving the LGM in Spain, France, Italy and (possibly) in the Ukraine did NOT seem top survive the final cold of the YD - which is what matters...

Investigating the possible scenarios of the human re-population of northern Eurasia - and their mix with the tropical tribes, leading to the traditional demography of Eurasia and America - any serious genetician will have to consider the 'baltic scenario' as a possible alternative. Ruling it out before investigating the matter isn't an option anymore - if archeological facts are to be taken seriously. Whatever one may tend to 'believe'.

Rob said...

@ Batman

"Add the genetic studies revealing a strong, demografic bottle-neck during the YD and we end up with only one culture known to have survived the last, worst and terminal phase of ice-time - and we end up with the Baltic refugia as the only known origin of the described bottle-neck, as well as the following mariners."

The studies thus far certainly don;t mention anything about a "baltic refugium". I think that's your 'colourful' interpolation.


* "but on the old pseudo-scientific 'deduction', where "abscence of evindence" becomes "evidence of abscense"."

Yes, the old absence of evidence doesn;t mean evidence for absence axiom is the special pleading and the lowest form of argumentation. You are trying to dismiss the obvious settlement continuity in more southern part of Europe, and elevate the non-existing evidence from the Baltic. In essence, you are claiming up is down, and black is white Now, I take your geological inferences with great interest, but if humans dewelt that far north during the LGM, then we should find evidence, esp in well researched countries like Scandinavia and Germany, unless you suggesting some great Ice Age conspiracy is going on.


* "The genetic bottle-neck forming the extant populations did NOT happed during the LGM, but during the Younger Dryas - when several arctic species and humans - except from a handfull - went extinct'

Actually, it began during (& going into) the LGM, was exacerbated by the Younger Dryas, and the final hit was during the early Holocene, as late as the first contacts with farmers. So it was a multi-phase phenomenon.

* "
Do we have any evidence that similar populations survived the last and terminal phase of the late Paleolithic in France, Spain, Italy, Greece, Balkans and Ukraine?!'

Is this a serious question ?


* " any serious genetician will have to consider the 'baltic scenario' as a possible alternative. Ruling it out before investigating the matter isn't an option anymore "

I agree. We should do genetic testing on Baltic remains too. Now, we just have to actually find those humans living in the Baltic at 20 kya. We might be waiting for a long time.

batman said...

@ Rob

You're obviously not here to discuss new discoveries and facts, but to argue - like a second-hand politician - to maintain your own bias. Even when they are built on nothing but presumtions, predjudice and outdated models.

Refusing new facts and insights with old beliefs is hardly taking any branch of the historical sciences forward.

Rob said...

Lol. What politics ?
It's the ice age dude