Autosomal analysis of Yamna, Corded Ware and Bell Beaker samples

[Angela]

Fascinating stuff. I hope someone helps me to get this fixed.

Similar as lack of WHG in some Asian IE populations put Yamna origin of PIE under question, also lack of Gedrosia in Balts/East Slavs is a problem when Yamna had 30% of it. So going same style I would say that PIE originated in Baltics and then moved into Yamna :)
Because those German Corded samples (and Yamna samples) do not look like ancestors of Balts. Too much of goodies they have that we dont. But we can easily be partial ancestors to Yamna, Corded...

I am obviously missing something, because Yamna - Corded - Balts cultural archeological chain is mainstream and logical.
OK, those were German Corded samples, would be interesting to look at Middle Dnieper Culture (also Corded, according Gimbutas one of centers for Corded migrations in different directions) admixtures.

I don't think the "story", in so far as we're talking about the genetics and the language is as clear as it appeared it might be when the Haak et al paper first came out, do you? I've just reread it all, including the supplement, and I think the paper itself is pretty cautious about all of this; it's we hobbyists who have made some pretty unsupported (so far ) comments.

As you say, I think the archaeological, "cultural" progression is pretty clear...Yamnaya, Corded, Balts. But, also as you say, the Corded samples that we have are not a very good fit for Balts. The fact that they score "high" in terms of correlation with Yamnaya samples is, in my opinion, a function of their high EHG scores, which "pull" them toward Yamnaya, not that they are necessarily "sons" of Yamnaya. Indeed, I don't know that I would think that Corded itself is a "son of" Yamnaya...perhaps a brother, however.

I do think that it is important to get samples from the Middle Dnieper culture, but I have a feeling that they might be higher in EHG as well, higher perhaps than the "Yamnaya" samples.

I suppose what I'm badly expressing is that I think it all depends on how you define the "Indo-Europeans". If you define them as this specific group of people, with a specific genetic makeup, whose culture coalesced on the Pontic Caspian steppe between 4,000 and 3,000 BC, then perhaps these other groups aren't these specific "Indo-Europeans", although they are definitely part of the same phenomenon?

Also, I think it's informative to take a look at the stats in Haak et al that model the migrations not in terms of Yamnaya, but in terms of EHG. I think this has to be considered in light of the fact that Haak et al are talking about a massive migration of genes from eastern Europe coming into the rest of Europe. These might not all be specifically "Indo-European" genes or Yamnaya genes, in my opinion. My only quarrel with the paper might be that they don't make these subtleties sufficiently clear, which leads to a lot of confusion.

Looking at it from the linguistic angle, they are similarly careful to say that this movement of "people" from the area of the steppe may be responsible for the spread of some Indo-European languages, not all of them.

As to your point about the lack of WHG in some " Asian Indo-Europeans" ( a la the leak from Patterson) we'll have to wait and see what their next paper shows, I think. They obviously have samples they haven't yet published, perhaps from the Caucasus or south of it. Indeed, they have Yamnaya samples they haven't yet published, apparently. So, time will tell. If indeed that turns out to be the case when the ancient samples are examined, then perhaps the source and trajectories of these expansions may have to be altered to accommodate the additional complexities?

What I do know is that it seems to me that the Reich group researchers are correct to limit themselves to models based on ancient samples, and to adjust those models as new samples become available. I think we've seen what incorrect conclusions can be drawn when models do not only include ancient samples.

[bicicleur]

there is only one definition for Indo-Europeans : people who speak an Indo-European langauge
Reich has said nothing about IE, anciant DNA doesn't say what language these people spoke
he talks about a massive migration during corded ware, replacing 75 % of the indogenous people, mainly by R1a people
he also shows at the same time a lot of BB R1b people arrived in Western Europe
those R1a corded ware and R1b BB are not the same as the EHG or neolithic R1b who arrived earlier and lacked the 'Armenian-like' admixture
I repeat, Reich didn't say anything about IE
but he proved a lot of theories about the spread of IE - most notable the Anatolian origin theory - to be wrong

[MOESAN]

the cradle of first individualized PIEan language (and the population speaking it, mixed or not) is NOT the same thing as the path taken by one or more of the subsequent sets of IEan speakers who spred the language later and farther - if we focalize too much on a single launching, perhaps we never shall find the answer...
the genetic (and phenotypic) influence of a Southern population upon the Steppes tribes is sufficiently proved even if the too localized 'armenian' hypothesis and the two heavy weight given to it are out of play - "southerners" influenced Steppes tribes through West (Cucuteni, Balkans-Carpathians), through East (BAMC) and surely through Caucasus (Maykop) - physycally, the southern element would be rather a kind of 'cappadocian-indo-iranian' type with high enough level of autosomes 'west-asian' ('gedrosia' could have been in Steppes before that) - ANE is to largely spred to give us a clue here - what we have as bones remnants don't disprove this analysis at all, it's the opposite - same for DNA - it's true all that points more in the direction of S-E Caspian than to Anatolia... (the types in Cucuteni were more "western" or "south-anatolia" at first, 'danubian' type, but at Metal Ages I don't know, too unprecise abstracts, and people travel) -

now, the language...? a well developped culture had more chances to pass its language bit it is hard to be sure, it depends on a lot of conditions -

[arvistro]

I suppose what I'm badly expressing is that I think it all depends on how you define the "Indo-Europeans". If you define them as this specific group of people, with a specific genetic makeup, whose culture coalesced on the Pontic Caspian steppe between 4,000 and 3,000 BC, then perhaps these other groups aren't these specific "Indo-Europeans", although they are definitely part of the same phenomenon?

I think Indo-Europeans are well defined, any people who speak one of IE languages. Proto Indo Europeans are also defined I guess - people who spoke the scientifically reconstucted PIE.
For those folk who went all directions from Yamna it makes sense to find a new name. Because a) it is not known if they cover all IEs; b) it is not known if they spoke PIE. Yamnoids? :) or "Yamna Indo Europeans".

Btw - modern Samara folk (where Yamna's samples were taken) has <1% Gedrosia.

I guess next good question is when and how did they loose it?
Cultures in Samara:
Yamna, Catacomb... - and maybe there is no need to move any further. This is what Maciamo comments (bold mine):

As I explained above, the Balto-Slavic branch probably descends from the pure R1a people from around Belarus and central western Russia, who later founded the Catacomb culture. Since their mtDNA shows no West Asian lineages whatsoever (as opposed to 45% of N1a, K, T, W and X in Yamna), they cannot have had West Asian admixture, including Gedrosian.

So initial Gedrosia IE folk were pushed away from Yamna. Btw didn't they (pushed folk) all turn out Centum? And was replaced by Balto-Slavic Catacombs that lacked Gedrosia. Satem?

Could "Yamna Original" be already Centum?
Could Satem speaking Balts live in North before Yamna and therefore would not need any Yamna direct ancestry to speak IE?
Could Catacombs be part of development of Satem Indo-Iranians, originally from North?

[Angela]

[QUOTE]

I think Indo-Europeans are well defined, any people who speak one of IE languages.

Great, for a literal, but, forgive me, not all that helpful definition in terms of the holy grail of Indo-European studies, i.e. the location of the linguistic, cultural, and genetic Indo-European homeland. The assumption has been that there is total overlap of all those elements. I'm not so sure about that. I'm also not sure Haak and Lazaridis et al have proved that Yamnaya is it. What they have proved, in my opinion, is that Yamnaya type genetics have had a large influence in Europe, more, at least as it looks now, in northern than in southern Europe. Otherwise, how could Corded Ware be 73% Yamnaya or modern northern Europeans 50% Yamnaya? I know all this might seem pedantic, but my business is words. Words matter. Definitions matter. Imprecision of language is a result of fuzzy thinking and leads to misinterpretations.

For those folk who went all directions from Yamna it makes sense to find a new name. Because a) it is not known if they cover all IEs; b) it is not known if they spoke PIE. Yamnoids? :) or "Yamna Indo Europeans".

I agree with all your qualifiers, and I like that definition. :) From now on I will call them Yamnaya Indo-Europeans.

Btw - modern Samara folk (where Yamna's samples were taken) has <1% Gedrosia.

I guess next good question is when and how did they loose it?
Cultures in Samara:
Yamna, Catacomb... - and maybe there is no need to move any further. This is what Maciamo comments (bold mine):
Originally Posted by Maciamo
As I explained above, the Balto-Slavic branch probably descends from the pure R1a people from around Belarus and central western Russia, who later founded the Catacomb culture. Since their mtDNA shows no West Asian lineages whatsoever (as opposed to 45% of N1a, K, T, W and X in Yamna), they cannot have had West Asian admixture, including Gedrosian.

I don't know about their not having had West Asian, but as I have been saying, I think the Baltic area's "Yamnaya Indo-European" scores may be inflated to a degree because of excess EHG/SHG, especially in the case of Finland, for example. That doesn't mean they don't have influence from Corded Ware and thus Yamnaya to some degree. Corded Ware moved into Latvia and Lithuania rather early didn't it? At any rate, the culture definitely came from Corded Ware. The language I don't know. I'm sure you know a great deal more about this than I do. Could it have been a more Uralic type language? That's one of the supports for the location of the "homeland" on the Pontic Caspian steppe isn't it...i.e. that this would explain the Kartvelian and Uralic influences on Indo-European?

In terms of genetics, the formal stats in Haak and Lazaridis et al are pretty unambiguous in showing that the northern populations cannot be modeled solely as prior Europeans plus a big influx of EHG. That "West Asian" component is a necessary ingredient.

Let's take a look at the Dodecad K7b for another vantage point on the genetics, although of course there isn't going to be a complete correspondence between this type of blunt tool and formal stats.
This is Corded Ware

• 66.02% Atlantic_Baltic
• 27.10% West_Asian
• 3.73% South_Asian
• 3.13% Southern
• 0.02% Siberian
• 0.00% African
• 0.00% East_Asian

These are the scores for "West_Asian" for modern European populations. (this term means different things in different runs, of course)

Belorussians: 11.7
Germans: 11.9
Lithuanians 10.4
Finns 6.9

We're seeing the same phenomenon as with "Gedrosia". What happened to the "West Asian"? Clearly, the further north you go with modern populations, the more that West Asian drops. It also drops as you go west. Perhaps it's partly because at the margins of Corded Ware there was more mixing with pre-existing populations. Also, as Haak and Lazaridis specifically state in the paper, there was a resurgence of prior existing genes after the Bronze Age invasions, just as there was a resurgence of WHG in Europe between the early and the late Neolithic. Whether it was a resurgence in either case or just a question of the types of burials that survived (i.e.elite or dominant burials) is, I think, an open question.

Then, Admixture is going to "aggregate" gene alleles slightly differently depending on what is present and the modal population chosen.

(Armenians are about 53% West Asian in this calculator by the way. About half would be about 27%? Has anyone run Yamnaya through K=7b?)

Anyway, for what it's worth, this is what Dienekes discovered about the relationship between the components in his runs.

K7b as mixtures of K12b.JPG

K12b as mixtures of K7b.JPG
Interrelationships of K12b and World 9.JPG


As to this interrelationship between K12b and World this is what he had to say. Clearly, some more recent discoveries put some of these into question, but some still seem to be pretty accurate. Also, these are all more recent "clusters" formed of more ancient layers of more "Basal" groups.

• Gedrosia appears to be Caucasus + a slice of Siberian
• Both Siberian and Southeast Asian appear to be wholly East Asian
• East Asian on the other hand, appears to be mostly Southeast Asian + minority Siberian
• Northwest African appears to be Caucasus + a minority Sub Saharan
• Atlantic Med appears to be Caucasus + a slice of North European
• North European appears to be Atlantic Med + Gedrosia with a slice of Siberian
• South Asian appears to be Caucasus + East Asian
• East African appears to be Sub Saharan + minority Caucasus
• Southwest Asian appears to be Caucasus
• Sub Saharan appears to be East African
• Caucasus appears Atlantic Med + Gedrosia + slices of Northwest African and Southwest Asian

Anyway, I'm not at all saying that we should be relying on these or any other calculators, not when we have formal stats based on ancient samples. It's just that if we're going to discuss them we should try to understand how they relate to each other.

[Aberdeen]

Good post, Angela. But some of your comments make a non-science person like myself all the more skeptical of some of these terms. If Gedrosia is Caucasian + a slice of Siberian and Atlantic Med is Caucasian + a slice of North European, how can North European be a slice of Atlantic Med + Gederosia with a slice of Siberian? I feel as if I'm watching a dog chasing its tail. And why are these ancient remains being classified as part Armenian-like if Armenians are more recent? Why aren't we saying that Armenian is Yamnaya-like?

Just on the basis of the genetic data available, I'd be inclined to question whether the Yamnaya were just a Corded Ware derived group, rather than CW being Yamnaya derived. It's actually the archeological evidence and the subclades of R1a and R1b currently found in both India and Europe that have always made me a believer in the steppe hypothesis.

[Maciamo]

Could "Yamna Original" be already Centum?
Could Satem speaking Balts live in North before Yamna and therefore would not need any Yamna direct ancestry to speak IE?
Could Catacombs be part of development of Satem Indo-Iranians, originally from North?

That is exactly how I see it.

That's also why I always refer to Yamna people as PIE speakers, like many serious linguists.

[Maciamo]

We're seeing the same phenomenon as with "Gedrosia". What happened to the "West Asian"? Clearly, the further north you go with modern populations, the more that West Asian drops. It also drops as you go west. Perhaps it's partly because at the margins of Corded Ware there was more mixing with pre-existing populations. Also, as Haak and Lazaridis specifically state in the paper, there was a resurgence of prior existing genes after the Bronze Age invasions, just as there was a resurgence of WHG in Europe between the early and the late Neolithic. Whether it was a resurgence in either case or just a question of the types of burials that survived (i.e.elite or dominant burials) is, I think, an open question.

I think I can explain why Northern Europeans have low frequencies of West Asian despite their high Yamna ancestry. I think it mostly has to do with the way various populations mixed in Scandinavia between 4000 and 1000 BCE.

Around 4000 BCE, two distinct groups lived side by side in Scandinavia. Mesolithic HG (similar to modern Saami) and Neolithic farmers who were quite Sardinian- or Basque-like (mix of South European HG and Near Eastern farmers).

What we can tell from modern mtDNA and Y-DNA frequencies is that modern Scandinavians inherited genes from both, but their pre-IE Y-DNA is ultimately of Mesolithic origin (I1), while their mtDNA is overwhelmingly from the West European Neolithic blend (H, J, K, T, U5b, V) and only in minority from Mesolithic HG (U4, U5a).

Notice that Neolithic Swedes (Gökhem) and modern Sardinian and Basque people are just as low in West Asian admixture as modern Scandinavians. Yet modern Scandinavians also have considerable Mediterranean (15%) and EEF (35%) admixtures - much higher than Baltic or Uralic NE Europeans.

The Indo-Europeans first moved into Scandinavia with the Corded Ware culture (R1a), but only had a moderate impact based on the 15% of R1a in Denmark and Sweden (a founder effect in Iron Age or medieval nobility could account for the higher R1a in Norway). When R1a got there, Yamna genes had already been diluted through intermingling with locals in western Ukraine, Poland and Germany, then got progressively diluted further at each generation it took for R1a to advance from Denmark to Sweden, then to Norway.

The same thing happened with R1b, except that R1b arrived much later in Scandinavia, probably from 1700 BCE when the culture changed to the Nordic Bronze Age, when more Yamna-like and Celtic-like elements start to appear, like burial mounds with chariots and bronze shields. These R1b people would be descended from the Unetice culture, which came to replace the Bell Beaker culture in Germany, Bohemia and West Poland.

So when Unetice R1b Proto-Germanics moved into Scandinavia, the locals only had a little Gedrosia/West Asian from heavily diluted Corded Ware R1a immigrants, and the R1b newcomers brought even more diluted Yamna genes. Since the Mesolithic-Neolithic blend remained dominant (over 50%) in Scandinavia, it's only logical that ancient Germanics and modern Scandinavians should have so little West Asian admixture. And indeed the Hinxton genomes showed that ancient Anglo-Saxons had extremely low levels of West Asian (about 1%).

It should be expected that Germanic migrations lowered the overall West Asian and Gedrosian admixture in the northern half of Europe.

It isn't so surprising that the ANE and Gedrosian admixtures should be higher today in the British Isles than in Germany, since Germany received far more Scandinavian immigrants in the Iron Age than the Britain. Within Britain, both the ANE and Gedrosian admixtures peak in Scotland (probably more in the Highlands). Unfortunately we don't have much data from Wales, and especially remote places like Anglesey and Pembrokeshire. That ought to be interesting.

[Alan]

[QUOTE=Angela;451730]

• Gedrosia appears to be Caucasus + a slice of Siberian
• Both Siberian and Southeast Asian appear to be wholly East Asian
• East Asian on the other hand, appears to be mostly Southeast Asian + minority Siberian
• Northwest African appears to be Caucasus + a minority Sub Saharan
• Atlantic Med appears to be Caucasus + a slice of North European
• North European appears to be Atlantic Med + Gedrosia with a slice of Siberian
• South Asian appears to be Caucasus + East Asian
• East African appears to be Sub Saharan + minority Caucasus
• Southwest Asian appears to be Caucasus
• Sub Saharan appears to be East African
• Caucasus appears Atlantic Med + Gedrosia + slices of Northwest African and Southwest Asian

Anyway, I'm not at all saying that we should be relying on these or any other calculators, not when we have formal stats based on ancient samples. It's just that if we're going to discuss them we should try to understand how they relate to each other.

So North European might have absorbed Gedrosia type DNA? Could this in support why East Europeans don't have any Gedrosia, because Northeast Europeans merged with it?

If not, than I honestly doubt modern Balto_Slavs are descend of PIE. More likely they are "Indo Europinized" farmers+H&G. Because let's be serious now. 2/3 of Indo Europeans have it. Even R1a rich South_Central Asians have it. Thracian samples had allot of it (they also had Caucasus and North European), also populations of the Urals and North Caucasus have it.

[giuseppe rossi]

[QUOTE=Alan;451795]

So North European might have absorbed Gedrosia type DNA? Could this in support why East Europeans don't have any Gedrosia, because Northeast Europeans merged with it?

If, than I honestly doubt modern Balto_Slavs are descend of PIE. More likely they are "Indo Europinized" farmers+H&G. Because let's be serious now. 2/3 of Indo Europeans have it. Even R1a rich South_Central Asians have it. Thracian samples had allot of it (they also had Caucasus and North European), also populations of the Urals and North Caucasus have it.

Gedrosia and North Euro have ANE affinity, while Atlantic Med don't. That's why Sardinians have zero Gedrosian and North European admixture on Dodecad k12b.

[Angela]

[QUOTE=Alan;451795]

So North European might have absorbed Gedrosia type DNA? Could this in support why East Europeans don't have any Gedrosia, because Northeast Europeans merged with it?

If not, than I honestly doubt modern Balto_Slavs are descend of PIE. More likely they are "Indo Europinized" farmers+H&G. Because let's be serious now. 2/3 of Indo Europeans have it. Even R1a rich South_Central Asians have it. Thracian samples had allot of it (they also had Caucasus and North European), also populations of the Urals and North Caucasus have it.

Isn't some of this just a function of how ADMIXTURE works? The "North Euro" cluster is the set of alleles "modal" or most common in Northern Europe today. It "hides" within it alleles from past migrations and more "basal" groups. Wouldn't any "Gedrosia" or "West Asian" or any other cluster of alleles show up only if they were in addition to or excess over whatever is aggregated into the North Euro cluster? Take a look at the second graphic: K12b expressed as a mixture of K7b components. "North Euro" there looks to have about 10% "West Asian". Or look at K12b for a comparison in terms of the World 9 components. So far as I know, in these runs "North Euro" is actually Northeast Euro. So, doesn't it stand to reason that "Northwest Euro" would show an excess of some more southern or southeastern "components"? What is showing in these runs could just be the excess.

That doesn't mean that there aren't some differences based on different migration patterns and the gradual dilution of the Yamnaya Indo-European autosomal signature as time went on, as Maciamo pointed out. There's also the explicit statement in the body of the Haak et al paper that there was a "resurgence" in pre-Yamnaya ancestry. Some of that "resurgence" might just be that the "autochonous" stock as they would be defined at that time weren't being buried in tombs that would have survived.

As to the Baltic area, Finland in particular, or even some of the northern forest steppe areas, I do think that their actual "Yamnaya Indo-European" descent genetically might not be as high as it appears in Figure 3 of the paper, because that's partly just reflecting their high levels of EHG. They seem to have gotten some effect from Yamnaya, however. Some of their 44% EN has to come via Yamnaya. I don't think the TRB got that far, and even if they did, I doubt they were numerous enough to have that kind of impact all the way into the Baltics and Finland. Plus, the mtDna in that part of the world is different from that in western Europe. It has a more specifically "West Asian" cast, as opposed to Cardial and LBK, as if it came due north from the Near East.

Culturally, of course, they adopted the "Yamnaya Indo-European" package of the groups to their south. I'm not sure about the language. Could it have been Uralic? This is wild speculation, but look at the EDAR results in SHG. If that is accurate, it's far too early to be attributed to recent gene flow from Siberia.

I'm not married to any of these ideas, of course. I'm just trying to use what I know of admixture to correlate those results with what I know is closer to the "truth"...i.e. the formal stats in Haak et al.

[Angela]

Good post, Angela. But some of your comments make a non-science person like myself all the more skeptical of some of these terms. If Gedrosia is Caucasian + a slice of Siberian and Atlantic Med is Caucasian + a slice of North European, how can North European be a slice of Atlantic Med + Gederosia with a slice of Siberian? I feel as if I'm watching a dog chasing its tail. And why are these ancient remains being classified as part Armenian-like if Armenians are more recent? Why aren't we saying that Armenian is Yamnaya-like?

Just on the basis of the genetic data available, I'd be inclined to question whether the Yamnaya were just a Corded Ware derived group, rather than CW being Yamnaya derived. It's actually the archeological evidence and the subclades of R1a and R1b currently found in both India and Europe that have always made me a believer in the steppe hypothesis.

All of this analysis arises out of attempts to verify the Pontic Caspian steppe lands as the uhrheimat of the "Indo-European" peoples. The oldest ancient samples in the Yamnaya area show hunter gatherers (EHG) on the steppe with a certain genetic profile. Then, later, around the time of the "genesis" of the Indo-European cultural and linguistic "package", they found ancient people who are only 50% similar to these EHG. The authors are clear that they haven't found a good ancient genome which would supply the other 50%, although they know there must be resemblances to Near Eastern populations. After exhaustive modeling, the formal stats spit out a few modern populations that would "statistically" fit. Modern Armenians and Iraqi Jews came out at the top of the list, so this ancient population, for which we don't yet have a sample, must be similar to these groups. Now, as these groups are highly EN, that ancient population must have been highly EN as well. The stats are complicated, but I think the theory is as simple as that.

If I made an error, I know people won't be shy about pointing it out.

How can Yamnaya be a Corded Ware derived group if the innovations appear first in Yamnaya? Unless I'm not following you here. I'm not 100% positive that Corded Ware is genetically derived from Yamnaya in that Corded Ware people are descended from Yamnaya people who literally moved from the steppe around Samara to Corded Ware areas.The yDna signatures would, for the present, seem to indicate that wasn't the case. Perhaps there were people to the west on the Pontic-Caspian steppe, or slightly northwest, however, who were R1a but had a similar autosomal signature. They had to have some "Near Eastern" ancestry, however, because they are 73% Yamnaya with the rest what they picked up in Corded areas. Also, although, as I said, they are blunt tools, look at the West Asian numbers or Gedrosia numbers in the admixture runs for Corded Ware compared to the modern people in the area. Remember also that Haak et al call them a "related" group. So, while they may or may not be a "son" group, they're at least a "brother" group, I think.

The archaeology and culture are clear however. As far as Europe is concerned, the innovations came from Yamnaya.

In your honor...

[LeBrok]

Isn't some of this just a function of how ADMIXTURE works? The "North Euro" cluster is the set of alleles "modal" or most common in Northern Europe today. It "hides" within it alleles from past migrations and more "basal" groups. Wouldn't any "Gedrosia" or "West Asian" or any other cluster of alleles show up only if they were in addition to or excess over whatever is aggregated into the North Euro cluster? Take a look at the second graphic: K12b expressed as a mixture of K7b components. "North Euro" there looks to have about 10% "West Asian". Or look at K12b for a comparison in terms of the World 9 components. So far as I know, in these runs "North Euro" is actually Northeast Euro. So, doesn't it stand to reason that "Northwest Euro" would show an excess of some more southern or southeastern "components"? What is showing in these runs could just be the excess.

I used to think that admixtures were more precisely defined. After spending some time on them, seeing how they blend and disappear I think it is exactly the way you eloquently have put it above.

As to the Baltic area, Finland in particular, or even some of the northern forest steppe areas, I do think that their actual "Yamnaya Indo-European" descent genetically might not be as high as it appears in Figure 3 of the paper, because that's partly just reflecting their high levels of EHG. They seem to have gotten some effect from Yamnaya, however. Some of their 44% EN has to come via Yamnaya. I don't think the TRB got that far, and even if they did, I doubt they were numerous enough to have that kind of impact all the way into the Baltics and Finland. Plus, the mtDna in that part of the world is different from that in western Europe. It has a more specifically "West Asian" cast, as opposed to Cardial and LBK, as if it came due north from the Near East.

By quick glance of admixtures of Corded Ware, it looks like Yamnaya was responsible for 50% of population replacement in North East and Central Europe. However these samples might have come from the Invaders elite burials, who were not mixed very well with locals yet. The true numbers of Invaders might have been much lower. On other hand, the invasions from steppe usually happen during population decline in Europe, like at the end of Roman Empire times. In this case locals are not very numerous and therefore ratio of invaders is much higher, than it would be in "normal" times.

[Aberdeen]

All of this analysis arises out of attempts to verify the Pontic Caspian steppe lands as the uhrheimat of the "Indo-European" peoples. The oldest ancient samples in the Yamnaya area show hunter gatherers (EHG) on the steppe with a certain genetic profile. Then, later, around the time of the "genesis" of the Indo-European cultural and linguistic "package", they found ancient people who are only 50% similar to these EHG. The authors are clear that they haven't found a good ancient genome which would supply the other 50%, although they know there must be resemblances to Near Eastern populations. After exhaustive modeling, the formal stats spit out a few modern populations that would "statistically" fit. Modern Armenians and Iraqi Jews came out at the top of the list, so this ancient population, for which we don't yet have a sample, must be similar to these groups. Now, as these groups are highly EN, that ancient population must have been highly EN as well. The stats are complicated, but I think the theory is as simple as that.

If I made an error, I know people won't be shy about pointing it out.

How can Yamnaya be a Corded Ware derived group if the innovations appear first in Yamnaya? Unless I'm not following you here. I'm not 100% positive that Corded Ware is genetically derived from Yamnaya in that Corded Ware people are descended from Yamnaya people who literally moved from the steppe around Samara to Corded Ware areas.The yDna signatures would, for the present, seem to indicate that wasn't the case. Perhaps there were people to the west on the Pontic-Caspian steppe, or slightly northwest, however, who were R1a but had a similar autosomal signature. They had to have some "Near Eastern" ancestry, however, because they are 73% Yamnaya with the rest what they picked up in Corded areas. Also, although, as I said, they are blunt tools, look at the West Asian numbers or Gedrosia numbers in the admixture runs for Corded Ware compared to the modern people in the area. Remember also that Haak et al call them a "related" group. So, while they may or may not be a "son" group, they're at least a "brother" group, I think.

The archaeology and culture are clear however. As far as Europe is concerned, the innovations came from Yamnaya.

In your honor...

Thanks for the hilarious graphic. It's a pretty good visual of how my head feels when I try to understand the interplay between these various admixtures.

I was just joking about Yamnaya being possibly a CW derived group in the genetic sense, but I don't think it's any more off than describing CW as a Yamnaya derived group. I think the reality is, as you say, that they were more like brother groups. I think perhaps CW is just ENG mixed with European Neolithic while Yamnaya got its Middle Eastern component from elsewhere, either the Caucasus or Neolithic farmers from the Ukraine. More data will eventually clarify things. But it's clear that CW and Yamnaya were related but separate phenomena, and that Yamnaya was distinctly Bronze Age with far reaching influence in both Europe and Asia. I was mostly making the point that I don't think people should refer to CW as Yamnaya, despite the genetic connection.

[Alan]

I hope you are not referring to the CW was brought to Europe .........but created in central Europe. it is made my all haplogroups livng in germany at the time plus merging of the migrating Yamnya haplogroups

Of course Yamna was created in Europe. My comment might gave the impression I meant something different. What I meant ist that Yamna and CW might be related groups but CW not necessary descending from Yamna.

What I meant is there might have been one group of Pastoralist who played a role in forming both Yamna and CW.

[Alan]

Isn't some of this just a function of how ADMIXTURE works? The "North Euro" cluster is the set of alleles "modal" or most common in Northern Europe today. It "hides" within it alleles from past migrations and more "basal" groups. Wouldn't any "Gedrosia" or "West Asian" or any other cluster of alleles show up only if they were in addition to or excess over whatever is aggregated into the North Euro cluster? Take a look at the second graphic: K12b expressed as a mixture of K7b components. "North Euro" there looks to have about 10% "West Asian". Or look at K12b for a comparison in terms of the World 9 components. So far as I know, in these runs "North Euro" is actually Northeast Euro. So, doesn't it stand to reason that "Northwest Euro" would show an excess of some more southern or southeastern "components"? What is showing in these runs could just be the excess.

That doesn't mean that there aren't some differences based on different migration patterns and the gradual dilution of the Yamnaya Indo-European autosomal signature as time went on, as Maciamo pointed out. There's also the explicit statement in the body of the Haak et al paper that there was a "resurgence" in pre-Yamnaya ancestry. Some of that "resurgence" might just be that the "autochonous" stock as they would be defined at that time weren't being buried in tombs that would have survived.

As to the Baltic area, Finland in particular, or even some of the northern forest steppe areas, I do think that their actual "Yamnaya Indo-European" descent genetically might not be as high as it appears in Figure 3 of the paper, because that's partly just reflecting their high levels of EHG. They seem to have gotten some effect from Yamnaya, however. Some of their 44% EN has to come via Yamnaya. I don't think the TRB got that far, and even if they did, I doubt they were numerous enough to have that kind of impact all the way into the Baltics and Finland. Plus, the mtDna in that part of the world is different from that in western Europe. It has a more specifically "West Asian" cast, as opposed to Cardial and LBK, as if it came due north from the Near East.

Culturally, of course, they adopted the "Yamnaya Indo-European" package of the groups to their south. I'm not sure about the language. Could it have been Uralic? This is wild speculation, but look at the EDAR results in SHG. If that is accurate, it's far too early to be attributed to recent gene flow from Siberia.

I'm not married to any of these ideas, of course. I'm just trying to use what I know of admixture to correlate those results with what I know is closer to the "truth"...i.e. the formal stats in Haak et al.

So it could simply be that Northeast Europeans are "Indo Europinized" local H&G and Farmers. That is a possibility. I don't think Uralic had yet evolved.

[Alan]

I feel as if I'm watching a dog chasing its tail. And why are these ancient remains being classified as part Armenian-like if Armenians are more recent? Why aren't we saying that Armenian is Yamnaya-like?

This is because, as already pointed out quite a few times, Even if you replace Armenian with any other Western Asian group the genetic closeness to Yamna still remains. If you actually replace it with Iraqi Jew it is higher. And there is no other group in the Steppes which could be the source for it. Therefore this must be real "West Asian" input into Yamna.

[Aberdeen]

This is because, as already pointed out quite a few times, Even if you replace Armenian with any other Western Asian group the genetic closeness to Yamna still remains. If you actually replace it with Iraqi Jew it is higher. And there is no other group in the Steppes which could be the source for it. Therefore this must be real "West Asian" input into Yamna.

I'm well aware that some west Asian groups seem to be a mixture of Middle Eastern early farmer and EHG. I'm just saying it seems strange to me to call this mixture Armenian or Armenian like when found among Yamnaya, considering that Yamnaya seems to have preceded the existence of Armenians. It's similar to the point I was making about Corded Ware - CW is genetically connected to Yamnaya but I think it's a mistake to call it Yamnaya like when it preceded Yamnaya and embodied a different cultural package

[Alan]

I'm well aware that some west Asian groups seem to be a mixture of Middle Eastern early farmer and EHG. I'm just saying it seems strange to me to call this mixture Armenian or Armenian like when found among Yamnaya, considering that Yamnaya seems to have preceded the existence of Armenians. It's similar to the point I was making about Corded Ware - CW is genetically connected to Yamnaya but I think it's a mistake to call it Yamnaya like when it preceded Yamnaya and embodied a different cultural package

You still didn't seem to have understand it. I agree that the notion "Armenian like" might be wrong. But it is right to call it Near Eastern.
Why don't you understand that if a genetic component which is found in ALL Near Easterners, but not in any other group beside Yamna, must have been from the Near East.
We cant call a component "Yamna" if we actually try to break up the Yamna genome. There is no other source which might have possibly given this admixture to Yamna.

Or are you saying we should assume that Bedouins in Arabia are Yamna admixed, because this genes are also found in same.

[Aberdeen]

You still didn't seem to have understand it. I agree that the notion "Armenian like" might be wrong. But it is right to call it Near Eastern.
Why don't you understand that if a genetic component which is found in ALL Near Easterners, but not in any other group beside Yamna, must have been from the Near East.
We cant call a component "Yamna" if we actually try to break up the Yamna genome. There is no other source which might have possibly given this admixture to Yamna.

Or are you saying we should assume that Bedouins in Arabia are Yamna admixed, because this genes are also found in same.

You seem to have a reading comprehension problem. I'm saying that Yamnaya was a mixture of Russian hunter gatherer types and a Middle Eastern population that no longer exists. The fact that the Armenians are one of the populations that are somewhat similar to that vanished population doesn't mean we should call it Armenian. And when I say that some west Asians seem to be a mixture of that Middle Eastern population and other populations so may be partly Yamnaya, I'm talking about groups like Iranians. I don't know why you're classifying Arabian Bedouins as west Asians.

[MOESAN]

never heard of gedrosia in russia and siberian ...gedrosia is on the coast of the india ocean in modern eastern iran

http://www.antiquaprintgallery.com/a...p-198924-p.asp

north of Gedrosia was ancient ARIANA ....................some say means home of Aryans

Makran is the "capital" of Gedrosian ..............some say origin of LT-P326 marker

http://www.britannica.com/EBchecked/topic/359318/Makran

genes are genes and poolings are poolings
so genes present in Siberia 40000 BC were come there from Baluchis of today after a quick travel back in time???
or are they not rather allover spred old genes we find today concentrated in Pakistan and NW Europe pooled with truly southern genes?

[Angela]

You seem to have a reading comprehension problem. I'm saying that Yamnaya was a mixture of Russian hunter gatherer types and a Middle Eastern population that no longer exists. The fact that the Armenians are one of the populations that are somewhat similar to that vanished population doesn't mean we should call it Armenian. And when I say that some west Asians seem to be a mixture of that Middle Eastern population and other populations so may be partly Yamnaya, I'm talking about groups like Iranians. I don't know why you're classifying Arabian Bedouins as west Asians.

I don't see the issue here. According to formal stats the mystery ancient population which admixed with the EHG to produce the Samara Yamnaya Indo-Europeans can best be described as a population similar to modern Iraqi Jews and Armenians. If we didn't have an EEF genome, I'm sure the Reich group would have been able to model the peopling of Europe as involving a population that is very " modern Sardinian like". We all know now that indeed the EEF are very "Sardinian like" or vice versa.

What's the problem, therefore, with describing the Near Eastern mystery population as "Armenian like"? Do the authors or hobbyists really have to say every single time that Yamnaya can best be modeled as a population that is approximately 50% "ancient Karelian like" and 50% an ancient Near Eastern population that is most like modern Armenians and Iraqi Jews?

Perhaps this is one of those points on which we have to agree to disagree.

[Aberdeen]

I don't see the issue here. According to formal stats the mystery ancient population which admixed with the EHG to produce the Samara Yamnaya Indo-Europeans can best be described as a population similar to modern Iraqi Jews and Armenians. If we didn't have an EEF genome, I'm sure the Reich group would have been able to model the peopling of Europe as involving a population that is very " modern Sardinian like". We all know now that indeed the EEF are very "Sardinian like" or vice versa.

What's the problem, therefore, with describing the Near Eastern mystery population as "Armenian like"? Do the authors or hobbyists really have to say every single time that Yamnaya can best be modeled as a population that is approximately 50% "ancient Karelian like" and 50% an ancient Near Eastern population that is most like modern Armenians and Iraqi Jews?

Perhaps this is one of those points on which we have to agree to disagree.

There is no problem when professional geneticists refer to an Armenian like population, after having explained what it means. There isn't too much of a problem when amateurs refer to an Armenian like population. There is a problem when some amateurs assume that the "Armenian" (rather than "Armenian like") component must have come from a particular location because it's close to where modern Armenians live. I realize some people think the "Armenian like" component probably came from the Caucasus simply because the Caucasus is fairly close to Samara. But after looking at the archeological evidence, I've slowly come to the conclusion that the Armenian like portion of Yamnaya may have entered the steppe from the west, and I don't think calling the Middle Eastern originating influence in Yamnaya "Armenian like" should blind us to that possibility.

[Angela]

There is no problem when professional geneticists refer to an Armenian like population, after having explained what it means. There isn't too much of a problem when amateurs refer to an Armenian like population. There is a problem when some amateurs assume that the "Armenian" (rather than "Armenian like") component must have come from a particular location because it's close to where modern Armenians live. I realize some people think the "Armenian like" component probably came from the Caucasus simply because the Caucasus is fairly close to Samara. But after looking at the archeological evidence, I've slowly come to the conclusion that the Armenian like portion of Yamnaya may have entered the steppe from the west, and I don't think calling the Middle Eastern originating influence in Yamnaya "Armenian like" should blind us to that possibility.

Now I understand you. We indeed don't yet know where it came from, and I too have considered whether it could have entered the steppe from the west. The ANE portion of that ancestry would have to be explained, however, or "West Asian", Gedrosian or whatever, which has not shown up in any "farmers" from Europe.

I know, it's tail chasing time, yes?

[Aberdeen]

Now I understand you. We indeed don't yet know where it came from, and I too have considered whether it could have entered the steppe from the west. The ANE portion of that ancestry would have to be explained, however, or "West Asian", Gedrosian or whatever, which has not shown up in any "farmers" from Europe.

I know, it's tail chasing time, yes?

Well, we know that Y haplotype R probably originated somewhere in Asia and had thousands of years to develop, so some parts of it could have migrated to just about anywhere. I suspect most of the R1a portion just drifted north and west to end up mostly in eastern Europe, without a lot of mixing, so remained high in ANE until some R1a mixed with Neolithic farmers in eastern Europe to become Corded Ware. The R1b could have moved west across the steppe to mix with farmers in the Ukraine and/or Balkans, then moved east back on to the steppe when a drought made pastoralism a better bet.

Yah, I'm making me dizzy. And that isn't the only possible scenario, but I think it is one possible scenario. I've decided that the big problem with the Mykop idea, beloved by many, is that although there's considerable evidence of cultural flow, there doesn't seem to be any evidence of significant population flow in archeological terms. And we don't yet have the DNA data needed to test that idea. But those mountains may well have been a significant barrier to gene movement.