search this blog

Tuesday, October 21, 2014

Ancient genomes from the Great Hungarian Plain


This open access paper on the genetic prehistory of the Great Hungarian Plain is full of surprises. Here are a few of my observations:

- Four of the genomes from a Neolithic farming context produced two Y-haplogroups previously identified in Mesolithic European hunter-gatherers (I2a and C6), and one of the samples (KO1) could probably pass for a Mesolithic hunter-gatherer overall, suggesting that males of hunter-gatherer origin played a major role in early European Neolithic societies. But what's happened to the C6 since then?

- The two Bronze Age genomes, BR1 and BR2, look very present-day French, and probably western French at that, in both the Principal Component and Admixture analyses. Indeed, they clearly show a northern influence relative to all of the Neolithic farmers and the Iron Age IR1. And yet, BR2 belongs to Y-haplogroup J2a1, which is generally seen as a Near Eastern marker.

- IR1 is described as a pre-Scythian genome with both East Eurasian and North Caucasian affinities (it's not clear in the paper whether it belongs to Y-haplogroup N and mtDNA G2a1, or vice versa, although either way works in this context). However, it also shows significant Northern European-like ancestry, and is even inferred to have fair hair, which makes me think that its eastern shift might be in large part due to Eastern Hunter-Gatherer (EHG) or Yamnaya-related admixture, which is now pervasive across Northern Europe (see here).

- Many people, including myself nowadays, see the Carpathian Basin as potentially a major staging point for the expansion of Y-chromosome haplogroup R1b into Central and Western Europe during the Bronze Age. And yet, it's again missing from the line-up.

- The T allele at SNP rs4988235, associated with lactase persistence into adulthood in Europeans, is only present among the two most recent genomes: BR2 and IR1. This suggests that selection for this allele, which now reaches frequencies of well over 50% in much of Europe, post dates not only the Neolithic but also the early Indo-European period, and was possibly most intense during the metal ages.

- Some of the Neolithic samples are clearly shifted towards the Bedouins (Bed) in Figure 2, relative to Oetzi the Iceman, a Copper Age genome from the Tyrolean Alps, which is generally considered to be typical of European Neolithic farmers (see below). So perhaps further sampling of Neolithic remains from southern Europe, in particular the southern Balkans, might reveal early farmers who actually cluster with Near Eastern populations, rather than Europeans?

- The authors found a sweetspot for extracting ancient DNA from humans: "the petrous portion of the temporal bone, the densest bone in the mammalian body". The amount of endogenous DNA salvaged from this part of the skull exceeds those from other bones by up to 183-fold. This is obviously great news, and probably means we can expect many more ancient genomes to be published in the near future.



Citation...

Gamba, C. et al. Genome flux and stasis in a five millennium transect of European prehistory. Nat. Commun. 5:5257 doi:10.1038/ncomms6257 (2014).

See also...

First I1-M253 from prehistoric Europe

Genetic continuity and shifts across the metal ages in the Carpathian Basin: analysis of ancient Hungarian genomes CO1, BR1 and IR1

308 comments:

1 – 200 of 308   Newer›   Newest»
Mike Thomas said...

The supp data clearly shows that Y-DNA of IR sample was N, and mtDNA was G2a...

Shaikorth said...

The main paper is confusing issues a bit, but the supplementary data (for example, table S13) pretty clearly says Iron Age individual is Y-DNA N-M231 (no deeper SNP's tested) and mtDNA G2a1.

Ponto said...

It seems to me from the PCA that most of the Bedouins are genetically located in Africa, not the Near East or Arabia.

If the Neolithics are shifted towards the Bedouins maybe the first Neolithic ancestors came from Africa not Anatolia or the Near East.

Shaikorth said...

Those Bedouins are noticeably more Eurasian than real North Africans like Mozabites or Egyptians.

A related interesting point is that it looks like some ancestral components cancel themselves out in mixture models. Those ANE-less Bedouins can be successfully modeled as Chechen/Yoruba mixes in ALDER (in Anthrogenica member Everest's test), with the Chechen component being dominant. Chechens have lots of ANE and to cancel it out the Yoruba proportion was considerably bigger than the African Bedouins demonstrate in ADMIXTURE.

Returning to the topic of Iron Age Hungarian, someone @ EUpedia is confused by Russo-Armenian marriages and PCA projections.

Davidski said...

Someone at Eupedia is always confused.

Grey said...

"suggesting that males of indigenous hunter-gatherer origin played a major role in early European Neolithic societies"

I think googling "aboriginal stockman" might give a clue as to how it worked with native HGs recruited to act as herders. In Australia the colonization was too fast for those individuals to spread that knowledge among the aboriginal population but in Europe the farmer spread was a lot slower so maybe it was different.

.

"But what has happened to the C6 since then?"

If a two tier society is created the ydna gene flow is often one way from the upper layer to the lower so say for the sake of argument the lower layers loses 1% of ydna percentage per century while the mdna remains the same percentage.

Then if the lower layer started at 80% it would take 8,000 years to disappear to trace levels whereas if it started at 20% it would only take 2000.

.

"The two Bronze Age genomes, BR1 and BR2, look very present-day French, and probably western French at that ... Y-haplogroup J2a1, which is generally seen as a Near Eastern marker."

I imagine WHG as coastal CHG with I and J as originally the western and eastern branches. That may or may not make sense however.

It's interesting though if you drew a line that was somewhere around 100 to 200 miles from the coast all around the med,atlantic, baltic etc you'd get a rough map of the mammoth steppe.


.

"Many people, including myself nowadays, see the Carpathian Basin as potentially a major staging point for the expansion of Y-chromosome haplogroup R1b into Central and Western Europe during the Bronze Age. And yet, it's again missing from the line-up."

What if it was already full?

Say you had a population of R1b west of the Black Sea who wanted to move away from a problem but the Carpathian basin was already densely settled (relatively speaking) maybe their only options were to go around it i.e. north of the Carpathians or by sea either to Anatolia/Caucasus or south into the Aegean and beyond and only expanding (beyond the size of refugee coppersmiths among a larger population) once they arrived somewhere that wasn't already densely settled with farmers.

ryukendo kendow said...

Can anyone who is familiar provide some information on the archaeological background of the cultures? Their origins, etc. to help us interpret the data.
Deeply appreciated.
Making things easier, I'll post them here:
KO1/KO2 Early Neol. Koros
NE1/NE2 Middle Neol. ALP
NE3/NE4 M. Neol Bukk culture
NE5 M. Neol. Late ALP
NE6 M. Neol LBK
NE7 Late Neol Lengyel
CO1 L.Copper, Baden
BR1 E. Bronze Mako
BR2 L. Bronze Kyjatice
IR1 Iron, Pre-scythian Mezocsat

Early observations:
1) For the first time, a Caucasus-like contribution is found in Neolithics, in Fig 4 ADMIXTURE. (In Supp figs using NgsAdmix this is not the case, but not sure what to make of this new software that's only 1 year old.) Also, as David said the neolithics show a pull eastwards to ME in pca, which is not the case in Ancient DNA from neolithic thus far. Both phenomena are slight but significant.
This throws a spanner in the works. It increases the possibility that the Basal contribution in PIE was Caucasus-like but came from a source in the balkans, though judging from affinity to the Caucasus throughout Europe today vs in ancient samples, a contribution from these hungarian neolithics is not nearly enough. Perhaps there is indeed an even more ME-like pop somewhere further south as he said.

2. The bronze age individuals are very 'Western', but are not Iberian-like, rather being shifted to northwest rel. to geographical position. This is very mysterious.
Chad, what do you think about this, w. regard to the whole BB resembling modern N-Central Europeans thing? Ppl with more archaological knowledge about the Mako and kijatice cultures, your input would be welcome. They are somewhat obscure, difficult to find info about.

3. IR2 literally plugs the 'black sea hole' between Caucasus and E Eur, which is interesting to say the least.

Tesmos said...

NE3's(X2b) gnome is now available aswell.

https://drive.google.com/folderview?id=0B49499_WVp2fclFjbUhHeWRKaEk&usp=sharing#list

Davidski said...

I'll run the following samples if Felix makes them available:

BR2
IR1
KO1
NE1

The rest are too low quality and/or not interesting enough. Even BR1 is probably not worrying about, but might be worth a look if BR2 is too big (and it is fucking huge).

jackson_montgomery_devoni said...

@Grey,

You said...''I imagine WHG as coastal CHG with I and J as originally the western and eastern branches. That may or may not make sense however''

What exactly do you mean by this?

Davidski said...

Actually, CO1 looks pretty good too; Baden Culture genome with some sort of North or East Euro admixture.

Grey said...

@jackson

Populations mapping to geography / climate / food source zones.

If you look at maps of the western end of the mammoth steppe it's mostly the zone that is beyond a certain distance from the sea.

At the same time one of the arguments for why the WHG could be dark in northern latitudes is a seafood diet providing the vitamin D - which seems plausible but made me wonder about a coastal spread and a possible coastal vs interior split rather than an east vs west or north vs south one.

So in HG terms instead of a west-east line dividing WHG and EHG (with SHG in Scandinavia) maybe WHG all along the coast and up the rivers from the med along the atlantic to the baltic with EHG as the deep interior population.

They mostly come to the same thing but not quite.

It's just a thought though in case it makes the numbers work better.

(If you mean the IJ part that's just me wondering aloud about cro-magnons.)

Chad Rohlfsen said...

David,
Late Baden was overrun with Steppe people from the SE, before it disappeared. I'm not sure if this is late enough, for maybe someone that is just local. They are pretty close to local farmers.

It might not hurt to get all of the snps you can, to make that EEF component. Do they have as many snps as the Hinxton samples?

I don't know if they will be more Basal than Stuttgart. Her ancestors came out of this region. Otzi looked more Central/East Euro than her, on your plots too. Just as he does, next to these. They may be projecting Otzi and the other hunters, outside of KO1. I think that KO1 has an mtdna line that is common in Armenians. That is interesting. It must've been picked up in SE Europe, on the way to the Near East.

Chad Rohlfsen said...

Excuse me, the Copper Age sample could just be a local, and not a person with Steppe ancestry. Although it is possible for the Western Steppes to be more like Hungarian farmers than Yamnaya.

Chad Rohlfsen said...

I wonder if IR1 symbolizes the potential of Uralic mixed ancestry of certainly Eastern Steppe groups, possibly even a few Western ones.

Chad Rohlfsen said...

I hate to be a buzzkill, but if Reich didn't sample anything West of Samara, we may not get a good picture of the people that entered Europe. It is possible that Corded Ware is full NW Yamnaya, and Bell Beaker full SW Yamnaya. Both could logically descend from Western cultures within Yamnaya. Remember, there are at least 9 separate culture groups, within Yamnaya. I would expect anyone West of Samara to be close to as much WHG as ANE, if not a hair more. Farmers may have pushed those Balkan hunters East, into the heavier ANE areas. WHG does jump from Bulgaria to Ukraine.

Chad Rohlfsen said...

Here me out, before anyone freaks out. Dnieper Donets were the Western 1/3 of pre-Yamnaya. They are likely just as related to Swifterbant as they are to Samara. They were likely more WHG than ANE. They could be similar to the Motala people, or slightly more "Karelian". A mix with those "Armenian" types, would create a modern Central European, basically.

Chad Rohlfsen said...

Felix said that he is doing BR1, before BR2. BR2 is 30 gig, compressed. He said it will take 3 weeks to complete.

Chad Rohlfsen said...

David,
Is that the only Beaker site they have so far? I think that one is the most "Iberian", for mtDNA. Might be the reason for the low ANE.

Chad Rohlfsen said...

Yeah, that's the one. "Pimp juicing" the local ladies, for 500 years. No wonder...

jackson_montgomery_devoni said...

@Grey,

Are you saying that both I and J are Cro-Magnon in Europe? I can see haplogroup I being Cro-Magnon in Europe but not J really. J is probably a rather new late comer to Europe. J probably started arriving during the late Neolithic or Bronze Age.

Grey said...

@Jackson

Not in Europe - near eastern equivalent - but i'm not really suggesting it. It was just a thought i had once that slipped in among the other comments.

Chad Rohlfsen said...

David,
You should look at NE7, too. Lengyel has a steppe influence, according to some. I have a feeling it's going to turn up that the Neolithic went pastoral, rather than a true steppe ANE.

Chad Rohlfsen said...

Wasn't there some LP in Germany and Iberia? I remember a paper on that.

Matt said...

http://www.nature.com/nature/journal/v514/n7523/full/nature13810.html – Ust Ishim's genome's out.

http://www.nature.com/nature/journal/v514/n7523/extref/nature13810-s1.pdf - Supplement

Michael said...

"The finding that the Ust’-Ishim individual is equally closely related to present-day Asians and to 8,000- to 24,000-year-old individuals from western Eurasia, but not to present-day Europeans, is compatible with the hypothesis that present-day Europeans derive some of their ancestry from a population that did not participate in the initial dispersals of modern humans into Europe and Asia."

This must be referring to the basal in EEF.

ryukendo kendow said...

@ Matt
It seems like Ust-Ishim was a basal member of the WHG-ANE-ENA clade, or a Basal crown Eurasian, in other words.

I won't be surprised if there were further radiations from this clade that did not survive, and we are only seeing the tips of the three branches that did.

Mike Thomas said...

Doesn't this finding somewhat put a spanner in David's theory of large scale migrations from the east, then(?) if so, the distance between the sample and modern Europeans should be less

Mike Thomas said...

Ie during the Bronze Age

Grey said...

Not sure I understand the Ust-Ishim results. Are they saying Basal is Neanderthal?

Davidski said...

Mike,

Clearly, the European Neolithic farmers and most Mesolithic hunter-gatherers are shifted west relative to most modern Europeans and Near Easterners, except the Sardinians and some of the Bedouins who lack ANE. This is very obvious evidence for massive population movements from the east from the late Neolithic and well into the metal ages, which spread a lot of ANE across West Eurasia.

See that's why BR1 and BR2 are positioned with the present-day French instead of with the Neolithic farmers or even the Chalcolithic Oetzi and CO1.

Mike Thomas said...

You're quite right- they're definitely shifted east. Although causation and exact source remains to be elucidated. .note the appearance of J2 from the Bronze Age , which could have contributed ...

barakobama said...

Geneticker did an analyses of Ko1. His results are nearlly perfectly identical to Geneticker's analyses of Loschbour. He belonged to Y DNA I2a1-P37, but not an I2a1b clade like Loschbour.

http://genetiker.wordpress.com/2014/10/23/analyses-of-an-early-neolithic-hungarian-genome/

We've caught a hunter gatherer in the act of joining of a farmer community( he could have had a farmer great great grandparent or whatever though). This confirms that Loschbour-like hunters lived as far southeast as Hungary. Also, there's a U5b2a5 from Mesolithic Croatia. Stuttgart's WHG ancestry could have come from as far southeast as Serbia or even further. ANE was probably pretty absent from SE Europe during the Mesolithic.

Most of the WHG as of far seems to be coming from WHG males. Gok2 could have had as much as 20% more WHG than Stuttgart, but all the Goks has farmer maternal lineages, and the only Y DNA sample is WHG I2-M438. As Neolithic-Mesolithic paternal lineages were largely replaced in the bronze age, the high amount of WHG ancestry in modern Europeans became hard to detract with mtDNA and Y DNA.

barakobama said...

I emailed Reich again, but this time about WHG-affinity of EHG, Yamna, and Corded ware.

I asked specifically asked if EHG has less WHG-affinity than northern Europeans, if Yamna has less than southern Europeans, and if Corded ware had more than Yamna and how are they compared to modern Europeans.

If he simply says EHG had less or more WHG-affinity than northern Europeans, that gives a hige clue as to how much ANE, how the Yamna were, and how Yamna-related ancestry modern Indo European-speakers have.

Davidski said...

Mike,

I'm guessing you must be a fan of Baron Munchausen, who pulled himself and the horse he was on out of a swamp by his own hair.

However, Caucasians and other Near Easterners aren't shifted east because they have admixture from themselves, but from a North Eurasian population that moved southeast after the Neolithic.

Chad Rohlfsen said...

So, Karelians are between 30-50% WHG, is my guess.

I think that BR1 and 2, are more likely getting ANE from a Beaker source, than an Anatolian one. They shift more towards Northern Europe than West Asia.

Chad Rohlfsen said...

Br1 at least, 2 is probably going to be a mix, having J.

ryukendo kendow said...

@ Barak
By the way, the only person here with light hair and blue eyes is a Neolithic individual.

barakobama said...

"@ Barak
By the way, the only person here with light hair and blue eyes is a Neolithic individual."

Yes, I noticed that. I didn't really look into it though. NE5 and BR2 had a single mutated allele in SNP Rs12821256. Having both alleles is suppose to give yellow hair(all-out), so I guess this means blonde hair existed in those populations.

I don't know what was unique about NE7. rs12821256 is the only blonde-associated SNP I've heard of and it only gets as high as 25% in N Europe, and so it doesn't represent all shades lighter than brown.

I'm not sure how accurate the Hirisplex system is in predicting anything in between black and blonde. I'm sure it is pretty accurate just I don't know what SNPs or patterns their looking for. NE6-IR1 had much higher light-hair probability than the older samples, maybe that means something I don't know.

OF the Sardinian-like samples; 7/8 had at least a G in rs12913832. 3/9 had G/G. BTW, Gok2 had G/G.

7/12 Sardinian-like ancient Europeans have turned out dark-eyed(58.3%) and 4/12 have turned out light eyed(33%). Today only 20% of Sardinians have a G allele and ~3.5% have G/G.

I don't think it's a coincidence so many ancient Sardinian-like Hungarians had a G allele and that Gok2 had G/G. It probably means at least some Sardinian-like ancient Europeans were much lighter-eyed than Sardinians. Also, now we know there was some diversity in hair color

It doesn't make sense that Sardinians probably have well over 20% Euro-WHG ancestry, but only ~3.5% have light eyes. Selection could be why so few do.

barakobama said...

I know why you mentioned NE7. And I want to make a point to the other bloggers. You think I'm pushing for Yamna to be the reason there's similar pigmentation in west-east Europe, because you think I'm ethnocentric and that it's a romantic origin I'm attracted to.

You want the origin to be more complex, diverse, unexpected, showing ancestry has almost nothing to do with physical appearance. That's actually a biased view point and an attempt to sugar-coat.

It's not a crime to mention pigmentation, and I'm tired of people treating it like it is. If anyone reads the description of this blog, they'll see it mentions physical anthropology.

People on these blogs intentionally ignore pigmentation and phenotype period because they know it leads to bad ideologies. If you got an isolated ancient person, he would notice the same physical differences we see today and would be interested in its history just like I am. They would recognize certain phenotypes as foreign or familiar, just like people today.

People outside of the very very small minority of those who enjoy researching human history, think of human genetics as very simplistic-straight forward and defined solely by physical appearance(watch family guy, you'll see). If I talk about ancient history to anyone all they ever want to hear about is what ancient people looked like and the origin of phenotype differences between modern people.

Isolated tribesmen today go nuts with interest when they see people from other areas of the world and who look very differnt for the first time.

I don't have an agenda when it comes to human phenotype, I'm simply interested. The reason I keep mentioning N-Euro-type pigmentation, is because that's the last one to develop in west Eurasia and probably the world and has the most mysterious origin.

ryukendo kendow said...

@ Barak
While phenotype might be highly correlated with the autosome, it is simply highly unlikely that pigmentation, which is controlled by fewer loci and is under strong selection, is.

Phenotype is a good indicator of origin. That Basal Eurasian and WHG have so similar phenotypes today, or that sinodonty+east Asian phenotype emerged independently in two places, is thus an extremely weird situation.

On the other hand that the lightest people are found in the area of the world where we get very high latitudes heated by the gulf stream, and thus the highest vitamin d shortage when transitioned to farming, is not surprising. Barcelona is the same latitude as New York, and Norway is as far north as Alaska. There are no other areas in the world as far north as norway where farming was even a thing.

barakobama said...

Ryuk,

There are no Basal Eurasians around today. West Asians are mostly Basal Eurasian, but none are 100%. I'm pretty sure Basal Eurasian is suppose to be a mix of WHG-related and sometype of unknown non-mainstream Eurasian ancestry, which is why even Bedouin cluster with other west Eurasians and WHG. The most popular view is that the shared physical features of west Asians, north Africans, Europeans, and south Asians descends from the UHG-ANE branch, but of course Basal is involved somehow.

"On the other hand that the lightest people are found in the area of the world where we get very high latitudes heated by the gulf stream, and thus the highest vitamin d shortage when transitioned to farming, is not surprising."

That's an interesting idea. It's obvious light-skin associated mutations gradually got more popular in Europe from the Neolithic-bronze age, and today are uniform all-over Europe. So i guess it could have developed independently in various populations.


" it is simply highly unlikely that pigmentation, which is controlled by fewer loci and is under strong selection, is."

I'm not arguing pigmentation is determined by ancestry from certain prehistoric populations. I actually believe the opposite(but of course sometimes it could be true). Ancient DNA has thrown alot of surprises at us, and proven this isn't true at all. Pigmentation-wise I could only pass as N Euro, even though genetically I'm the most southern a N Euro(~West German-N French) can be and my father and his brother are darker than most middle easterns(similar facial features though).

In another thread I suggested that shared pigmentation in east-west Europe(many different and distant ethnic groups) which did not exist in the the stone age, might one way or another be connected to the Yamna people or could have been introduced by them(obviously it existed in Neolithic Hungary though). Even if I'm wrong we can all probably agree it probably arose in the bronze age.

I suggested this because Yamna-like ancestry might range from 40-65% in N Europe, and the rest of their ancestry is mostly from native Stuttgarts and Loschbours. Since various Yamna-descended people that lived in central Asia and Siberia are the oldest example of a N-Euro-like pigmented population, that makes the idea even more convincing.

Because of the finds in Hungary, Gok2, and dark-Yamna, I'm considering native-Neolithic people to be the original source.

Davidski said...

Preliminary K15 for BR1

North_Sea 24.45
Atlantic 27.9
Baltic 25.19
Eastern_Euro 2.93
West_Med 19.53
West_Asian 0
East_Med 0
Red_Sea 0
South_Asian 0
Southeast_Asian 0
Siberian 0
Amerindian 0
Oceanian 0
Northeast_African 0
Sub-Saharan 0

Davidski said...

Here's a PCA. Look for BA_Hungary. Easy to spot though.

https://drive.google.com/file/d/0B9o3EYTdM8lQMFNXX2FFZzIxRDQ/view?usp=sharing

Looks part Basque.

barakobama said...

BR1's results are strange and it doesn't seem an admixture based on modern populations suits him. We probably can't get deep-insight on his regional-European origins with your admixtures, just his overall EEF-WHG-ANE ancestry. Can you post his K13 and or ANE K7 results so we can get an idea what his EEF-WHG-ANE ratio is?

BR2's pretty evenly spilled out between Atlantic, North sea, and Baltic like East-Germans, Austrians, and Hungarians but he has unusually low east-Euro and high west-Med, and completely lacks middle eastern-centered east Med, red sea, and west Asian. We know he has some Stuttgart-like ancestry because of his high West-Med, but where's the East-Med and red-sea? I don't think his lack of middle eastern-centered components should be taken to seriously, it may be because this test doesn't fit him.

Overall his score is most similar to East-German, but it's still very distant.

barakobama said...

I suspect he's around 5-10% ANE, 35-40-% WHG, and 50-60% EEF. His near eastern-affinity are similar to Yamna just the WHG-ANE are complete opposites.

Davidski said...

The file looks legit, because the genome is where it's supposed to be on the PCA, if we consider that the PCA in Gamba et al. suffers from some projection bias.

K13

North_Atlantic 40.2
Baltic 34.56
West_Med 25.23
West_Asian 0
East_Med 0.01
Red_Sea 0
South_Asian 0
East_Asian 0
Siberian 0
Amerindian 0
Oceanian 0
Northeast_African 0
Sub-Saharan 0

K7

ANE 7.74
ASE 0
WHG-UHG 84.27
East_Eurasian 0
West_African 0.66
East_African 0.03
ENF 7.3

barakobama said...

It's safe to say his extra WHG-affinity compared to Neolithic European-farmers came from a non-Yamna/Indo European source.

Chad Rohlfsen said...

I think they're post Beaker....

It's more like 50 37 13, if you ask me.

barakobama said...

Holy shit!!!!!!!!! The K7 score is more extreme WHG than I expected. This guy is very interesting, and obviously not very Indo European(~7% ANE). Currently I don't have access to excel can you post his K13 EEF-WHG-ANE score.

Chad Rohlfsen said...

47.24049873
38.22790513
14.53159614

On the calculator

Chad Rohlfsen said...

Yamnayans were not homogenous, Barak. Yamnaya had 9 different cultures. Reich only tested the Samara Valley, the furthest point from Europe.

Chad Rohlfsen said...

The non-extreme, East Europe

Chad Rohlfsen said...

Their ANE k7 score cant be correct. Looks like what happened to Lithuanians and Estonians. I think the calculator score is correct, with the components. If they were 84% hunter, they should be off the charts.

Chad Rohlfsen said...

I think it's Western Steppe, heavily Dnieper Don, Mikhailovka, type, with a decent amount of WHG, prior to mixing with CT. Probably the R1b route, and I'll bet similar to the Beaker scores in Germany.

Davidski said...

The EEF-WHG-ANE test results look much better.

The K7 doesn't do well with ancient samples.

barakobama said...

How can he be 14% ANE? He's about as east as Spanish and SW French in the PCA, and scored 7.74 in ANE K7(which is similar to Iberians).

"Yamnayans were not homogenous, Barak. Yamnaya had 9 different cultures. Reich only tested the Samara Valley, the furthest point from Europe."

Yes that could be true(we can't assume) but they have Corded ware genomes from east-Germany, the exact people who brought Yamna-like ancestry to east Europe and probably similar to the ones who brought it to west Europe.

Chad Rohlfsen said...

Barak,
Spaniards are 12% ANE, and the Basque are 11%. English are 15%. BR1 is right where it should be.

Chad Rohlfsen said...

It's probably R1b/Beaker related. Corded Ware pretty much avoided Hungary.

Davidski said...

The most accurate estimates from Laz are here...

https://drive.google.com/file/d/0B9o3EYTdM8lQVGgyUEJpMWhFaEk/view?usp=sharing

And BR1's results do match.

barakobama said...

In the K13 calculator Spanish score under 10% ANE. Spanish scored 11.5% ANE, Spanish north scored 9.6%. He'd pass as the most eastern Spanish, so 14% is still too high. All of his near eastern is going to west Med, which is very strange. Like I said before modern admixtures obviously aren't fit for him, like it isn't fit for Mesolithic Europeans.

barakobama said...

He's like Basque in that he doesn't score(or very little) in middle eastern-centered components, with more eastern-affiliated WHG, more WHG, and slightly more ANE. He's also like a north European with some ANE subtracted.

Davidski said...

It's actually a she, from the early Bronze Age when Bell Beakers first entered the Carpathian Basin.

We need IR1 for comparison. I reckon he'll be very different, with ridiculous East Euro and West Asian scores.

barakobama said...

IR1 looks like a Finno-Urgic or Turk from Russia.

Chad Rohlfsen said...

BR2 may be the same. I don't think a J2 scoring similar scores will be a terrible surprise. J2 is heavy around the Caucasus. If that is correct that agriculture and domestication came from the South Caucasus, then a J2 could look very much the same. Just how a G and an I farmer can look the same. Not all J2's will be the same, just as not all R1a, or R1b will be the same. As more Corded and Beaker sites are tested, we will see the difference. It probably depends on their mate preference, or what's available.

Chad Rohlfsen said...

IR1 will be very North Caucasus like, with some East Asian admixture. Probably originated closer to the Caucasus, so more Armenian like..

barakobama said...

Hungary has obviously changed since the bronze. The bronze age Hungarians though fit in the bulky modern European cluster that includes Hungarians. They were probably moved south of BR1 and given post-Neolithic west Asian ancestry by Balkaners.

IRI's position in the paper's PCA and having Y DNA N make sense, couldn't he be from the people who brought the Hungarian language to Hungary?

Davidski said...

IR1 clusters with a few Russian/Armenians in the Gamba paper, and that's probably due to projection bias anyway.

He'll end up near the Ukrainians and Erzya on my plot. Just below them IMO.

Chad Rohlfsen said...

This is 1600 years before Magyars. IE's probably weren't just R1a, R1b. I bet they had J2, I, G, and N. It just depends on the location you're testing.

Chad Rohlfsen said...

That's possible, David. That much of a shift, compared to BR1 and 2, can't be because of ANE. I think it's more because it's shifting from more of a Near Eastern source. I'm really interested to see how much East Eurasian he had. Any guesses?

Chad Rohlfsen said...

Unless he's like 10-15% East Eurasian, causing the pull.

Chad Rohlfsen said...

Well, I know Genetiker's results are usually a little bunk, but this guy is showing good amounts of West Asian, and near 10% East Eurasian, on his runs.

He might be 45/30/15/10

barakobama said...

WA IR1 and Indo Iranian?

Davidski said...

IR1 can actually be described culturally as a Cimmerian. I'm not kidding.

Matt said...

Hmmm. That position actually reinforces the PCA from the paper, along a line with Northern French, but with obviously less compression from projection bias. Resulting in a position which falls between Basque and Northwest Europe.

When I've been thinking of Basques, I've thought of them to be simply be Stuttgart like people who somehow picked up a little Loschbour/Motala related admixture. A slight curiosity for Europe, where the main picture is a Stuttgart-Lithuania cline.

This seems to indicate that Basques may have slightly more representative of what South Europe may have been like during the Bronze Age, than I'm giving credit for, and that Bronze Agers in South Europe already had significant shifts north without having significant shifts east...? Then were shifted southeast.

Draw a line between the Bronze Age Hungarian and modern Hungarians and you get something that is where we might perhaps expect the Yamnaya samples to cluster (its still unclear).

Draw a line between the Bronze Age Hungarian and modern Bulgarians, to allow from the likely northeasterly gene flow from the Slavic Expansion, you get something from the northern Middle East.

http://i.imgur.com/BDl1ULk.png

Interesting how one sample can raise so many questions that depart from a simple clinal idea...

Tesmos said...

Davidski,

What happened to these Cimmerians in Hungary? Do you think they had a significant genetic influence on some modern European populations like Hungarians?

Davidski said...

I might be wrong, but I don't think so. They were just another late wave from the steppe that imposed itself as the ruling class over the more numerous peasants in the Carpathian Basin, and then melted away without leaving much behind. The Magyars did leave their language behind, so that's at least something, but the Scythians, Sarmatians and Cimmerians just went the way of the dodo as far as I know.

It seems the only steppe people to really leave a major trace in modern European genomes were the Yamnaya and related groups. I suspect the reason for this is that they came when population densities across Europe were very low.

Chad Rohlfsen said...

"Celto-Scythians"?!? Ha

Chad Rohlfsen said...

KO2, is now ready, David.

Chad Rohlfsen said...

Cimmerians are kind of linked to starting Hallstatt. Some "Germanic" tribes, like the Cimbri, and some in Northern Denmark, were reportedly "Cimmerian".

Chad Rohlfsen said...

The Taurisci are mentioned in links to Cimmerians, too.

Chad Rohlfsen said...

I know it's Wikipedia, but interesting none the less. If this person is rather Caucasus and East Asian, compared to the Bronze Age samples, it could actually validate the migration view.

"Paul Reinecke in 1925 postulated a North-Thracian-Cimmerian cultural sphere (nordthrakisch-kimmerischer Kulturkreis) overlapping with the younger Hallstatt culture of the Eastern Alps. The term Thraco-Cimmerian (thrako-kimmerisch) was first introduced by I. Nestor in the 1930s. It reflects a "migrationist" tendency in the archaeology of the first half of the 20th century to equate material archaeology with historical ethnicities. Nestor did intend to suggest that there was a historical migration of Cimmerians into Eastern Europe from the area of the former Srubna culture, perhaps triggered by the Scythian expansion, at the beginning of the European Iron Age. This "migrationist" or "invasionist" view, assuming that the development of the mature Hallstatt culture (Hallstatt C) was triggered by a Cimmerian invasion, was the scholarly mainstream until the 1980s. In the 1980s and 1990s, more systematic studies of the artefacts revealed a more gradual development over the period covering the 9th to 7th centuries, so that today the "invasionist" scenario is considered untenable, and the term "Thraco-Cimmerian" is used by convention and does not necessarily imply a direct connection with either the Thracians or the Cimmerians.

Archaeologically, Thraco-Cimmerian artifacts consist of grave goods and hoards. The artifacts labelled Thraco-Cimmerian all belong to a category of upper class, luxury objects, like weapons, horse tacks and jewelry, and they are recovered only from a small percentage of graves of the period. They are metal (usually bronze) items, particularly parts of horse tacks, found in a late Urnfield context, but without local Urnfield predecessors for their type. They appear rather to spread from the Koban culture of the Caucasus and northern Georgia, which together with the Srubna culture, blends into the 9th to 7th centuries pre-Scythian Chernogorovka and Novocherkassk cultures. By the 7th century, Thraco-Cimmerian objects are spread further west over most of Eastern and Central Europe, locations of finds reaching to Denmark and eastern Prussia in the north and to Lake Zürich in the west. Together with these bronze artifacts, earliest Iron items appear, ushering in the European Iron Age, corresponding to the Proto-Celtic expansion from the Hallstatt culture."

Fanty said...

Cimmerians eh?

One of these mystified kind of people who vannished in the fogs of time.

I somehow have it in the back in my head that "Conan the Barbarian" ("Who lived in the age of Hyperborea") is meant to be the last of the Cimmerians?

I fail to read the word "Cimmerian" without Arnold Schwarzenegger is whirling a sword in my mind, lunging for revenge for the genocide on his people. ;-P

Matt said...

Had a look at genetiker's analyses / calculator runs for the Iron Age Hungarian. Relatively relative to modern populations, but once all the non-West Eurasian population content (which may be noise) was stripped out, the Iron Age sample was closest to Bulgarians and Romanians on Dienekes globe10 and globe13 in its components.

Main difference is that in globe13 terms it has less Mediterranean instead more of the West Asian component in these calculator.

So prima facie pretty much in agreement with the Gamba PCA, once projection compression gets resolved by doing a PCA.

Same thing with the Bronze Age Hungarian, genetiker's component list, minus its non-West Eurasian noise seems about in the right place.
Quick PCA based on globe13 components run - only West Eurasians, based on each population's estimated distance to each of the globe13 components in turn based on their component membership:

http://i.imgur.com/8iTe7K6.jpg

put through rotation to better match Eurogenes normal plots -

http://i.imgur.com/KzVhcDF.jpg

(also gives an impression where the West Eurasian globe13 components would sit in light of the West Eurasian PCs we normally see).

Modern Hungarians look like a hybrid of Bronze and Iron Age with only a small Slavic tilt perhaps, here.

Probably some error between genetiker's processing and my raggedly arse reprocessing of what he has given though.

Chad Rohlfsen said...

Nice, Matt!!
Looks like about 60%EEF, 20%WHG, 15%EEF, 5%ENA. Or there about. It looks like we can model a lot of Europeans with this sample, in various amounts.

David,
Can you do IBD with other people. Maybe run an IBD with this one and the Iron Age Celts, plus the Saxons. I'm not sure if they're close enough in age, but it would be interesting to see. Can you get ahold of the Iron Age Bulgarian samples? I wonder if the Iron Age brought the same kind of people to all of Europe.

Chad Rohlfsen said...

Oops, that last 15%, should be ANE.

John Smith said...

It now makes since why Sardinians look like an exact match to Early European Farmers, and that is because both Sardinians and Early European Farmers are dominated by WHG ancestry on the male side but not the female side the Linearback and the Sardinains are probably both at least 20% WHG but they couldn't see it because they were assumed they were free of WHG.

Chad Rohlfsen said...

David,
Will you be doing shared drift and such, for BR1?

Davidski said...

Yes, I will, I'm just waiting for CO1 and IR1.

It'll be interesting to compare all the stats in the context of the Copper Age, Bronze Age and Iron Age.

ryukendo kendow said...


@Chad @Matt @Barak
On the Bronze Age Hungarians--
Barak: "It's safe to say his extra WHG-affinity compared to Neolithic European-farmers came from a non-Yamna/Indo European source."
Chad: "It's probably R1b/Beaker related. Corded Ware pretty much avoided Hungary."
Matt: "This seems to indicate... that Bronze Agers in South Europe already had significant shifts north without having significant shifts east...?"

It seems to me increasingly likely that there is another 'NorthWestern' pole of variation in Europe than the 'stuttgart-Lithuanian' poles we have generally considered. This pole maximises itself amongst the Atlantic coast, but is associated with very high WHG instead of high EEF, distributes itself into Atlantic+North Sea+Baltic, but not East Euro, components in ADMIXTURE, and might be responsible for raising WHG scores in Europe to present levels after the Neolithic and IE invasions, both of which were WHG-poor compared to populations today.

Other than the Bronze Age Hungarian, some of the Hinxton genomes might also fall into this category, of demonstrating NW-shift with high WHG, low East Euro and MidEast components, and such.

This raises a huge number of questions, and opens a whole new avenue of inquiry. If the Bronze Age Hungarian is indeed 'Weestern shifted' compared to everyone, then this immediately raises the specter of BB, but then they are not Iberian-shifted, which complicates the received archaeological interpretation of BB as well. Last of all, there is still the elephant in the room--R1b.

Grey said...

@ryu

"This pole maximises itself amongst the Atlantic coast, but is associated with very high WHG"

Many years ago when I spent a lot of time in the remoter parts of North Wales and noticed (or imagined i noticed) distinct physical differences the label i used in my head was "white Berbers."

Matt said...

@ Ryu, possible but as an alternative, broadly Iberian like Basques (who might tentatively represent Bell Beakers) travelling through Central Europe where more Corded Ware type people lived might have lived might then approximate BR1, without the original Beakers being like Bronze Age samples. The Bronze Age samples representing a signal of Northwest pole might then be an illusion.

It's very difficult to tell. It seems to me so far like the Bronze Age sample and Iron Age Hungarian sample basically far from one another on an axis at which modern Hungarians are more or less at the center. That seems to indicate that eventually they contributed about this much to the Hungarian population. But also these may have been elite samples so may have been more extreme even in their day.

@ Chad

Cheers. genetiker also ran calculators for the Early Neolithic sample that looks like the most extremely Near Easterns like farmer, so using the same method here's the first two dimensions of another PCA (http://imgur.com/NIL8QSq) and a 2 dimensional MDS (http://i.imgur.com/AgwYmcd.jpg).

On both of them the Early Neolithic Hungarian is as it should be from the components on a cline from Sardinian to Near East, but closer to Sardinian.

The PCA are a little less compressed in Europe than the MDS, mainly because the principal components analysis has a couple of much smaller higher dimensions, 3 and 4, where it can put certain post-Neolithic influences and that mainly increase the relative differentiation of Southwest Asian from Mediterranean (probably Northeast African / Arabian peninsula influences which the West Mediterranean lacks). And the same for the Turkic influenced Nogais who I left in.

The MDS is trying to talk about all the distances in 2 dimensions (probably with a little more loss of information). So I think this means there is less freedom for the first two dimensions to represent the earlier and more general patterns.

Davidski said...

BR1 is really interesting, and quite a mystery. But the shift in genetic structure represented by the change from CO1 to BR1 can be seen in the archeology of the region...

"As a result we may establish that there seems to have been a short,ca. 200-year-long phase between 2800 and 2600 BC in the Carpathian Basin, when “Late Copper Age” material culture (late Baden, perhaps Kostolac?) was used contemporaneously with “Early Bronze Age” type material culture (earliest Makó and late Vučedol/earliest Somogyvár-Vinkovci)."

https://www.academia.edu/5717048/Transition_to_the_Bronze_Age_Issues_of_Continuity_and_Discontinuity_in_the_First_Half_of_the_Third_Millennium_BC_in_the_Carpathian_Basin

Davidski said...

I mean rapid shift, which probably means a rapid migration into the region.

Chad Rohlfsen said...

The Bronze Age samples show us a little more than cultural change. If they are around 45-50% EEF, then we know the extra WHG affinity in Northern Europe wasn't brought in, but was already there. Britain, for instance, would have only been 45% Neolithic/EEF. I have mentioned the possibility several times. It looks like a fact now.

The supposed Iberian shift, is probably more about lacking West Asian / east med, and such. Obviously it's a steppe source, for the ANE. Nothing out of west Asia.

Davidski said...

The Copper Age sample, CO1, looks like Oetzi, so BR1's ancestors migrated in from somewhere that had very high WHG and a bit of ANE. The question is where.

Chad Rohlfsen said...

I'll bet it's Dnieper Don and Mikhailovka. Those are the ones that Anthony had going through CT, on their way to the Carpathian Basin. I believe the Hungarian Beakers came via Germany. If it's from Mittelbe-Saale, then the Atlantic affinities make sense. Although, that could be a result of all European farmers without additional migrations out of West Asia, being modeled as West Med/Atlantic, based on their WHG input.

Chad Rohlfsen said...

Matt,
Thank you, for the plots. Very interesting!

David,
You should run NE2. They look more Near Eastern, and almost like they have some East African.

Davidski said...

I'll run NE6, which is eastern shifted and has reasonable coverage, so the shift is probably real.

Chad Rohlfsen said...

David,
Felix has NE5, out. That's one of the C6 males.

Davidski said...

Indeed, GEDMatch kit# F999927.

What are the K15 scores? I can't log into GEDmatch at the moment.

Helgenes50 said...

F999927


Population
North_Sea -
Atlantic 24.49%
Baltic -
Eastern_Euro -
West_Med 54.73%
West_Asian -
East_Med 18.57%
Red_Sea 2.21%
South_Asian -
Southeast_Asian -
Siberian -
Amerindian -
Oceanian -
Northeast_African -
Sub-Saharan -

Helgenes50 said...

The results are similar to those of Stuttgart

Helgenes50 said...

Gedmatch results
F999927

K13


Population
North_Atlantic 17.48%
Baltic -
West_Med 57.68%
West_Asian -
East_Med 24.23%
Red_Sea 0.61%
South_Asian -
East_Asian -
Siberian -
Amerindian -
Oceanian -
Northeast_African -
Sub-Saharan -

Chad Rohlfsen said...

EEF 95.95779343
WHG 1.552149656
ANE 2.490056911

That's what the calculator comes up with.

Chad Rohlfsen said...

Oetzi showed 5%WHG, 3%ANE. I suppose that both are more about drift, correct? Or do you think it's possible that small amounts of ANE came in from West Asia? I don't think that people in the South Caucasus can be more Yamnaya than Northern Europeans. There had to be at least single digits, to low teens there already. If later farmers had some South Caucasus influence, it would make sense.

Chad Rohlfsen said...

Or, just the fact that it's an ancient sample and modern components can misrepresent it?

Davidski said...

That's close to 100% EEF. The small bits of ANE are noise.

ryukendo kendow said...

@ Davidski @ Chad
"BR1's ancestors migrated in from somewhere that had very high WHG and a bit of ANE. The question is where."

Agree strongly with this one. WHG in Europe, though reduced to the margins at one point, made a strong comeback. Chad, if what you said is true, then I'm guessing you must think the cultures you mentioned should have had the strongest impact in Northwest Europe. Their impact on say, scottish and orcadian must be massive, as these are so far away from a Stuttgart-Yamnaya cline. Is there evidence for this?

@ Davidski
The paper showed that the Hungarian Neolithic was somehow less 'West MEd' than the other neolithic samples we have, and a bit more MidEast-shifted, towards Levantines and Caucasus.

This could be an important result, as it might mean that the neolithic in each region is not the same. But somehow the calculators are not showing this.

Could you do some f3 stats?

Chad Rohlfsen said...

I don't think that the group that came in was really high in WHG. I think that it took place, well before. Remember farming stalled for 1000 years, nearly 150-200km, short of the Channel, North Sea, and Baltic Sea. Plus, all of the farmer mtDNA in Yamnaya. The Gok farmers were already admixed, before getting to Northern Europe, plus we have several hunter sites, way behind the lines, in Germany (Blatterhohle, Ostorf, etc). I'm sure that we will find many more, over time. I think it is safe to assume that Central Germany became even more hunter than the Gok farmers, as more time passed. Pastoralism would be the main drive of mixing, as it would force more contact between groups.

One final thing to remember, is the sparseness of farming settlements. They were a few dozen to a few hundred people, in Northern Europe, with much distance between them. Before pastoralism, there wouldn't be any pressure on the hunters to seek domesticated animals. Population densities were much lower. Stats that I've seen on pastoralism state that it could support 20 people/sq km. That leaves plenty of room for people that mix their hunter lifestyle, with some animals.

The archaeology supports the hunters going pastoral, to a degree. If farmers from Germany, were like Gok2 and 7, that's already 25-40%WHG. Then, factor in the hunters gaining in numbers in Scandinavia, Baltic, and Isles. That is why I think that Mesolithic survival in those areas was already at 50%+.

Remember, as far away as Samara was an "Armenian/Karelian" mix. I highly doubt anything in between would not have a decent amount of EEF.

I think an incoming group at 40/30/30, is sufficient to explain a 45/40/15, group in the Bronze and Iron Age, If Northern Europe was already around a 50/50 split. I think that the Bronze Age, is what really set modern levels, for Northern Europe.

Chad Rohlfsen said...

If we try to make Neolithic Europe like Gok2. Then we need someone with almost zero EEF, and not enough ANE, to explain modern levels, without a near wipe-out of the pre-metal folks. We need the end of the Neolithic to have people that are closer to modern levels of EEF, well North of the Alps and Carpathian Basin.

Chad Rohlfsen said...

BTW, the Blatterhohle people, started to become more agricultural around 3500BCE, the Ostorf group, stayed forager until Corded/Beaker times. So, we know that there were people that were straight hunter/pastoral, during the time that Beakers and Corded Ware showed up. There was no wipe out, and people could be way more than 50% WHG, in a few places in Germany.

Davidski said...

Yes, I'll do all the usual tests when I have BR1, CO1, IR1, KO1, NE5, NE6 and NE7.

I'll run them at the same time to save time.

Davidski said...

Do you guys have decent bandwidth? If so, why don't you try processing these samples with Felix's tools?

We might be able to get BR2 and NE1 a lot faster that way.

ryukendo kendow said...

@ Chad
I find this very plausible. Davidski's expectation that it was a migration might coexist with your theory to drive up WHG indep of IE and Neolithic in Europe.
One question: what do you think about mesolithic survival in the British Isles? Is there aerchao evidence for this? This would need to be high also.

ryukendo kendow said...

@ Davidski
Very excited to see your work.

Is there some way to localise the intrusive 'western' component in the bronze age? I suppose doing f stats modelling the bronze age hungarian as a mixture from a local neolithic sample, with all other populations in our data (incl. ancient pops) would be interesting to look at.

I am going to bet that the highest scoring is with Hinxton, or Orcadian.

Davidski said...

I think running BR1 against the ancient and modern samples in Human Origins with different tools might solve the mystery.

But, I suspect that this 4A Oracle sheet has the answer, which is that BR1 is the result of a Sardinian/Oetzi-like Baden population mixing with almost pure hunter-gatherer descendants from somewhere nearby.

https://drive.google.com/file/d/0B9o3EYTdM8lQUGhOam02ckxUR3c/view?usp=sharing

ryukendo kendow said...

@ Davidski
Thanks! And this chart is nice!
"BR1 is the result of a Sardinian/Oetzi-like Baden population mixing with almost pure hunter-gatherer descendants from somewhere nearby."

"Hunter-gatherer descendants" is almost certainly true. It would be great if we could find out if "somewhere nearby" was also true, using any of our typical tools. Esp with J2a and Bronze and etc being so suggestive. And of course the first match--Hinxton4.

Is there a method of comparing WHG alone from different modern populations?

Davidski said...

Looking at IBD sharing between BR1 and KO1, La Brana, Loschbour and Motala12 might work.

Chad Rohlfsen said...

I think that we will find many groups that are well over 50% WHG, running around in the Northern half of Europe.

Since the Bronze and Iron Age samples aren't much different from modern populations in terms of EEF, we could logically conclude that the EEF didn't move much in the Northern half of Europe. It would be WHG dropping, and ANE rising.

England is around 50% EEF today, the same as the old samples. There could logically be the same scenario in Germany, with groups ranging from Otzi, to Gok2, to almost Loschbour/Motala like.

It could be that Northern Europe has about the same amount of EEF today, that it did in the beginning of the Bronze Age.

Chad Rohlfsen said...

If Central Germany has hunter groups running around after 3000BCE, who are probably more WHG than anything; I see no reason why it would not be more common in Britain, Scandinavia, and the Baltic region.

Davidski said...

But it'd be better to run the IBD test with BR2, which is of a higher quality than BR1, and probably has several times more SNPs.

barakobama said...

NE5's ANE K7 results from GEDmatch.

WHG: 54.43

ENF: 45.57

Stuttgart's ANE K7 results from GEDmatch.

WHG: 43.28

ENA: 0.21

ENF: 56.21

As could be seen by NE5's K13 and K15 scores he's more WHG than Stuttgart. Stuttgart so far is the most near eastern-like of early European farmer samples.

Chad, NE5 scores ANE in the K13-conversions because the most WHG-like components in Europe also have the most ANE.

The only way to determine how much Euro specific WHG-ancestry modern Europeans have is by getting a genomes from the first farmers in south-east Europe. I bet Stuttgart had at least 5%, probably 10% or more. This means Balts are over 50% WHG, then another 20% or so is ANE, making them over 70% the same thing as SHG and EHG.

barakobama said...

The WHG-comeback in Europe revealed by BR1 is personally-interesting to me because my central-European U5b2a2-215G might have been involved in that event, or at least Stuttgart-Loschbour encounters. For a while I thought most N European WHG was from east Europe, but now I guess it seems it mostly comes from non-IE Loschbour-like central-west Europeans.

Chad Rohlfsen said...

U5b2a2 was in Blatterhohle. Lots of U5 there.

Chad Rohlfsen said...

Fisher-gatherer Germany Blätterhöhle, Hagen [BLA 11] 3922 ± 60 BC U5b2b(2)*

Farmer Germany Blätterhöhle [BLA 27] 3869 ± 59 BC U5b2a2

Fisher-gatherer Germany Blätterhöhle [BLA 8] 3726 ± 38 BC U5

Fisher-gatherer Germany
Blätterhöhle [BLA 9] 3681 ± 19 BC U5b

Farmer Germany Blätterhöhle [BLA 17] 3681 ± 19 BC U5b2a2

Farmer Germany Blätterhöhle [BLA 7*] 3666 ± 20 BC H5*

Fisher-gatherer Germany Blätterhöhle [BLA 24] 3616 ± 56 BC U5

Fisher-gatherer Germany Blätterhöhle [BLA 14*] 3603 ± 49 BC U5b2b2*

Fisher-gatherer Germany Blätterhöhle [BLA 21] 3577 ± 43 BC U5

Fisher-gatherer Germany Blätterhöhle [BLA 15] 3571 ± 47 BC U5

Farmer Germany Blätterhöhle [BLA 13*] 3513 ± 102 BC H5*

Fisher-gatherer Germany Blätterhöhle [BLA 1] 3508 ± 102 BC U5b

Fisher-gatherer Germany Blätterhöhle [BLA 12] 3449 ± 52 BC U5b2a2

Farmer Germany Blätterhöhle [BLA 16] 3429 ± 60 BC H11a

Farmer Germany Blätterhöhle [BLA 25] 3421 ± 63 BC U5b2a5

Farmer Germany Blätterhöhle [BLA 10*] 3418 ± 63 BC H1c3*

Farmer Germany Blätterhöhle [BLA 5] 3335 ± 136 BC H5

Farmer Germany Blätterhöhle [BLA 28] 3196 ± 103 BC J

Chad Rohlfsen said...

Forager/FB Germany Ostorf [SK 28a] M 3200 BC K

Forager/FB Germany Ostorf [SK 8d] M 3200 BC U5

Forager/FB Germany Ostorf [SK 35] F 3100 BC U5

Forager/FB Germany Ostorf [SK 12a and 18] M 3000 BC T2e 2 samples

Forager/FB Germany Ostorf [SK45a] ? 3000 BC J

Forager/FB Germany Ostorf [SK19] M 2950 BC U5a

There are a few U5s in Treilles, Bernburg, Schoningen, Spain, and Portugal as well. There's a lot more out there.

barakobama said...

There's also a U5b2a5 from Mesolithic Croatia. I haven't done much research but I read a paper on U5 that said U5b2a2 is most diverse and popular in Poland and Germany, or simply central Europe. It looks like a native post-LGM lineage of central Europe that was continuously assimilated by new comers(early farmers, Indo Europeans, etc.) and doesn't mark migrations into or out of(for the most part) central Europe. A native maternal-lineage that was constantly assimilated must have alot of hot women or prostitutes in it that were selected by immigrating men.

They should sequence a genome of one of those Neolithic U5 "Fisher gatherers". If nearlly pure WHG populations existed in central Europe as recently as 5,500YBp, that's big news and key to how WHG made a comeback.

barakobama said...

Chad, you should organize all the ancient mtDNA you can in saved Files. Instead of having to constantly look at the ancestral-hourneys sit. There are very quick and easy techniques for predicting their haplogroup by looking at their mutations.

Chad Rohlfsen said...

I'm sure there were nearly pure WHG people running around until almost Corded and Beaker times. Definitely some running around at 75%, I'll bet. The major mixing was said to take place between 3.5-5kya.

barakobama said...

CO1 was Sardinian-like and is only 800 years older than BR1 who has 35-40% WHG and only around 50% EEF(basically what CO1 was).

The late-Neolithic/bronze age WHG-comeback everyone is suggesting, makes the EEF-WHG-ANE test created by Laz very appropriate. The WHG score for many Europeans might be mostly from this possible WHG-comeback the EEF might mostly be from late-Neolithic people like Otzi and the ANE mostly from Yamna Indo Europeans(who contributed more EEF and some WHG, but if anything lowered the overall amount of both).

Even if it isn't that simple, if Stuttgart-like people brought farming to most of Europe(which seems to be true), then WHG and most ANE scores quite literally do come from hunter gatherers they encountered. It's like having parents from two different ethnicties or races, we can associate with both groups.

Someone should make a documentary or fictional story about farmer-hunter encounters and relations.

ryukendo kendow said...

@ Barak
Laz already provides an estimate for Basal and WHG in Stuttgart in his paper, where Basal is 44+-10%. So EEF is approx half WHG already. This means that Balts are in fact even more than 50% WHG, by the definition of Laz.

" A native maternal-lineage that was constantly assimilated must have alot of hot women or prostitutes in it that were selected by immigrating men."

Hahaha!

barakobama said...

Ryu,

What I meant was they over 50% of their blood is from WHG people of south-east Europe and mainland Europe who early farmers encountered. If you count their near eastern WHG, then it's around 70%.

WHG-UHG is probably pre-LGM like ANE and may have originated in west Asia. EHG probably got their WHG from east Europe not west Europeans who expanded after the LGM, because their U5 was U5a not U5b, and they probably lacked hg I. WHG in west Asia is also certainly not from post-LGM west Europeans.

Matt said...

Chad: The archaeology supports the hunters going pastoral, to a degree. If farmers from Germany, were like Gok2 and 7, that's already 25-40%WHG. Then, factor in the hunters gaining in numbers in Scandinavia, Baltic, and Isles. That is why I think that Mesolithic survival in those areas was already at 50%+.

I think the thing here for me is that the farmer communities seem to remain fairly genetically constant, after what seems like the initial stages where mixture happened.

So you have Stuttgart which is fairly early Neolithic, Oetzi who is Chalcolithic, the Gokhem farmers who are late Neolithic and the change in them seems really pretty modest. Not much change seems apparent in the Hungarian farmers over the time scales from over like, 3000 years, then there's a fairly sudden lurch in a Motala12-like direction.

(Skoglund's paper seems to show a fairly large 16% increase in WHG ancestry in Gokhem, but then PCA don't seem very much like this).

So if it is the case that farmers mixed with hunters locally, they either did it really infrequently, like building up very small percentages over thousands of years, then very suddenly a lot. Or the farmers who did mix with the hunters left the farmer cultural contexts we have excavated and were elsewhere for a long time, yet at the same time, not mixing into the Mesolithic contexts we have, then appeared on the record quite suddenly.

I'd imagine that on the whole, farmers who may have intermixed with hunter gatherers (and vice versa) might have been facing tough times.

So might have then disappeared or been demographically marginal afterwards. If they didn't, they might have faced a tough time dealing with other farmers, as y'know, their farming group is suddenly obviously in a way an outgroup to everyone else.

But then if you had a situation where a culture went right to the brink, to the point where its numbers were low and it *had* to intermarry quite extensively with the hunter gatherers, then was able to bounce back really quite strongly because of some new technology, some new means of subsistence or some climatic luck, that could change the status quo quite heavily, if there was then a barrier that prevented them melting back into the farmer mix (like if the farmers didn't really exist any more because the Corded Ware culture had done a lot of them in and absorbed what was left, that might "help", as it would be hard to be absorbed back into them).

...

But, I suspect that this 4A Oracle sheet has the answer, which is that BR1 is the result of a Sardinian/Oetzi-like Baden population mixing with almost pure hunter-gatherer descendants from somewhere nearby.

https://drive.google.com/file/d/0B9o3EYTdM8lQUGhOam02ckxUR3c/view?usp=sharing


Surprising. Based on the PCA plot, I absolutely would not have thought we were talking about a population which was 75% WHG and 25% EEF.

How does a population which essentially looks on a cline from Basques to Britain / Germany get to look like that?

I mean, this sample should be way more distant from where La Brana or Loschbour would be based on prior PCAs. It should be way closer to EEF, which contradicts a majority input from WHG-ANE cline people.

What do oracles with modern populations look like?

Davidski said...

Best fit are eastern Germans, but the best solution appears to be WHG + EEF.

https://drive.google.com/file/d/0B9o3EYTdM8lQREtrV2hoYk1kRG8/view?usp=sharing

Davidski said...

Felix uploaded NE7 to GEDmatch.

kit# F999928

Chad Rohlfsen said...

Population
ANE -
ASE -
WHG-UHG 59.98%
East_Eurasian -
West_African -
East_African -
ENF 40.02

Chad Rohlfsen said...

I think as more Bronze Age samples are available, we'll see a variety of types. Some will look more British like, and some more Tuscan type, depending on if they took a hunter or farmer type as a partner. Remember, El Portalon? They're 200 years younger than BR1.

Eduardo Pinto said...

NE5, NE7 and Stuttgart results hint to what pre-bronze Iberians might have been.

Take for instance my results below... I'm PT1 BTW
If you add the ANE score to the ENF and East_African you'll get identical scores to the Hungarian folks.

Population
ANE 8.03%
ASE 0.22%
WHG-UHG 52.47%
East_Eurasian 2.44%
West_African -
East_African 3.12%
ENF 33.72%

Chad Rohlfsen said...

NE7 looks like they're about 85% EEF.

Chad Rohlfsen said...

David,
Is there any chance that you could try to get something off of KO2? They're the most Bedouin shifted and maybe a couple hundred years older than Stuttgart.

Davidski said...

I'll have a look.

Davidski said...

Here's some stuff for KO2.

K15

North_Sea 0.18
Atlantic 18.32
Baltic 0
Eastern_Euro 0
West_Med 56.98
West_Asian 0
East_Med 21.62
Red_Sea 2.9
South_Asian 0
Southeast_Asian 0
Siberian 0
Amerindian 0
Oceanian 0
Northeast_African 0
Sub-Saharan 0

https://drive.google.com/file/d/0B9o3EYTdM8lQQThHbjI4RWJtVVk/view?usp=sharing

K13

North_Atlantic 14.7
Baltic 0
West_Med 57.9
West_Asian 0
East_Med 25.96
Red_Sea 1.44
South_Asian 0
East_Asian 0
Siberian 0
Amerindian 0
Oceanian 0
Northeast_African 0
Sub-Saharan 0

K7

ANE 0
ASE 0
WHG-UHG 44.42
East_Eurasian 0
West_African 0.01
East_African 0
ENF 55.57

Chad Rohlfsen said...

Tough to tell. It appears more Near Eastern shifted than Otzi.

Here is what the calculator scores.
98.26297152 EEF
-0.306360483 WHG
2.043388962 ANE

The calculator has KO2 more extreme than Stuttgart.. although..

Oetzi had 3% WHG, and 2% ANE. Either Oetzi and the others have some WHG, or the calculator is wrong, and it appears Stuttgart and KO2 were the max upon arrival to Europe.

The only thing is that the Oracle appears to have them as less EEF than Stuttgart, as the best mix is between Stuttgart and Sardinians, not two Stuttgarts. Very interesting. Mixed signals.

Chad Rohlfsen said...

Can f3 stats figure who is more Near Eastern?

barakobama said...

" and it appears Stuttgart and KO2 were the max upon arrival to Europe."

They're both from the same migration, same region, and are within 500 years of each other., so it makes sense they're like twins. I bet the first farmers were less WHG than them. Most of their WHG was certainly from west Asia though, a throw in the dark is that their Euro-WHG was 10-15%.

Simon_W said...

I haven't yet had the time to read through the whole discussion here, but as I already commented on Dienekes' blog: Some time ago I used to speculate that either Lengyel or Baden or both were associated with a dramatic increase of the „West Asian“ component, and possibly the arrival of R1b or J2. For archeologically and craniometrically, a connection with Western Asia has been made by some experts. Now we see: Both Lengyel and Baden were no different from early European farmers. In fact quite to the contrary, in the PCA and in the ADMIXTURE analysis they are even slightly closer to Sardinians than the preceding cultures in Hungary, and the blue Druze centered component is slightly weaker. Also the y-haplogroup of the Lengyel male, I2a, is nothing new.

The discontinuity between Baden, 2806 BC, and the early Bronze Age, 2082 BC, however does look dramatic! Judging from the PCA, the population changed in that short time from Sardinian-like to French-like, with an impressive increase of the North European component. Combined with picked up hints on the upcoming paper on Yamnaya I tend to think this must be ascribed to Yamnaya influence.

In the late Bronze Age, 1181 BC, we then see for the first time a J2a individual. Autosomally he is very similar to the early Bronze Age female; the J2 people from Western Asia appear to have had little autosomal impact in this population. It's interesting though, that the J2a male does have a little bit of the blue Druze centered component, in contrast to the early Bronze Age female. In theory this might be a legacy of the local Neolithic cultures, but I think it's more likely that this came back together with J2a. Personally I think that this J2 influence may have brought the Lemnian-related Raetic language to central Europe. The timeframe, 1181 BC, and the geographic proximity to the eastern Alps, where Raetic has been attested, make this a serious possibility.

With regards to the Iron Age male: His mt-DNA is G2a and the y-DNA N. So we have an East Eurasian connection there. And I wonder if the PCA and the ADMIXTURE analysis can really accurately assess the autosomal make-up of this individual, since there are no East Eurasian populations included in the analyses.

Finally, on pigmentation: It's amazing! Apparently the early farmers gradually evolved towards a lighter pigmentation, while staying on the whole Sardinian-like. (That's a combination that nowadays no longer exists!) That's probably possible because not too many loci are involved in that change. This means that the later increase of the North European component can't be merely an environmental adaption. It has been claimed by some authors that the LBK farmers and related cultures were „Nordics“, or proto-Nordics. On the whole they are far from being similar to present-day North Europeans – but craniometrically there is some similarity, and now we know that even the pigmentation was at least partly similar.

Chad Rohlfsen said...

David,
Just for shits and giggles, I ran Stuttgart and KO2 through the Near East/ANE Calculator. It is very interesting.

Stuttgart
Near East 90.024
ANE 9.976

KO2
Near East 90.957
ANE 9.043

Apparently, it doesn't take much more Basal to make 100% Near Eastern, by today's standard. If there were later Basal intrusions, plus waves of Afro-Asiatic speakers, ancient Near Easterners may have been extremely close to Stuttgart and KO2.

Chad Rohlfsen said...

It's certainly within the realm of possible that they represent the typical Anatolian farmer, at the time of migration to Europe. Farmers in the Natufian Culture, could be a different story.

Although, there is a 4000 year period of what could be a homogenizing of the population. Therefore they all appear similar, within a percent or two in Basal ancestry.

Matt said...

Davidski : Best fit are eastern Germans, but the best solution appears to be WHG + EEF.

https://drive.google.com/file/d/0B9o3EYTdM8lQREtrV2hoYk1kRG8/view?usp=sharing


Thinking about this, whether Oracles with Loschbour makes sense all seems to depend on how well the ADMIXTURE run represents Loschbour in the first place though, as IIRC the Oracles are based on matching population means from various populations to approximate another individual's components.

Like, the last Loschbour on K15 I saw from you was 29.69 North_Sea, Atlantic 29.57, Baltic 31.49 and Eastern_Euro 8.31, and while that's probably the right direction, the magnitude seems to be too close to existing North Europe.

So if Loschbour's distance to Europe is underrepresented by ADMIXTURE, then its contribution to a population might be overrepresented, even though the direction might be quite accurate.

For one comparison cross check, you could maybe to merge Loschbour+Stuttgart, or Loschbour+(Loschbour+Stuttgart) to get 75% Loschbour 25% Stuttgart, see if it would land anywhere near BR1. Probably too few overlapping SNPs for this to work though?

Davidski said...

Felix uploaded IR1 to GEDmatch.

kit# F999929.

Chad Rohlfsen said...

Matt,
There is no way that Corded Ware would have enough WHG to cause the change, plus many other factors.

There has to be a bounce back, caused by mixing with hunter groups, like the ones in Germany. If Corded Ware is 73% Yamnaya, then it is 36.5% "Armenian". Plus, there is the 36.5% "Karelian", which is probably mostly ANE. Where is the WHG? It had to be Mesolithic survivors, whom had taken up pastoralism between 3800-3000BCE. Corded Ware is not mixing into a pure hunter group either. They are going to have some EEF, from Globular Amphora, Baden, Lengyel, etc.

The climate change, which wiped out farming, had to be the culprit. I imagine a Neolithic collapse, which was exploited by those who knew how to hunt, and had some domesticated animals. The paper which spoke of the hunter comeback, had it starting 2000 years prior to Steppe intrusions. The only way WHG could climb so high, is those hunters who made due between farms, making a comeback. It is certainly easier to find Neolithic farming sites, than hunting sites, with temporary housing. I think that the paper on Neolithic Germany, will fill in the gaps.

Chad Rohlfsen said...

David,
You're on your own, with IR1. It's still being tokenized, with no end in sight.

Davidski said...

I'm about to put it into my dataset. Once I do that I'll run it and write up a blog entry.

Chad Rohlfsen said...

Awesome. Will that take a couple of days?

Davidski said...

IR1

WTF?

K15

North_Sea 23.36
Atlantic 23.4
Baltic 11.44
Eastern_Euro 19.14
West_Med 6.29
West_Asian 15.76
East_Med 0.04
Red_Sea 0.1
South_Asian 0
Southeast_Asian 0
Siberian 0.15
Amerindian 0.31
Oceanian 0
Northeast_African 0
Sub-Saharan 0

K13

North_Atlantic 38.52
Baltic 27.26
West_Med 11.35
West_Asian 19.67
East_Med 0.67
Red_Sea 0.86
South_Asian 0
East_Asian 0
Siberian 0.81
Amerindian 0.87
Oceanian 0
Northeast_African 0
Sub-Saharan 0

K7

ANE 18.09
ASE 0
WHG-UHG 57.17
East_Eurasian 2.45
West_African 0
East_African 0
ENF 22.29

But the PCA result matches with the one in the study.

https://drive.google.com/file/d/0B9o3EYTdM8lQREhTVzlPNU9hZFE/view?usp=sharing

barakobama said...

His EEF-WHG-ANE is like Balkan people with extra ANE and ENA. It's strange he goes against the European trend. He has high ANE and ENF, but low WHG.

Culturally you said he was Cimmerian. Can you give any insight into the origins of the Cimmerians, and what his results might suggest about them?

I would expect an Indo European people coming out of the steppe in the Iron age to be like we'd expect Yamna to be, not a mix of north European and Caucasus, which this guy looks like.

Davidski said...

4A Oracle results for IR1.

https://drive.google.com/file/d/0B9o3EYTdM8lQRE0xRlY5VTRrU1k/view?usp=sharing

barakobama said...

It's strange his European-ancestry is most similar to NW Europeans, since his people migrated through east Europe to get to Hungary.

Davidski said...

I'll try and write this up by tomorrow. But I'll have to think for a bit about what these results might suggest.

Chad Rohlfsen said...

Hell yeah, extra WHG is Mesolithic survival.. Guarantee it! He's actually pretty similar to me, as far as EEF/WHG/ANE/ENA. The Calculator had him a little different. 50/34/16. If there was an ENA section, it would probably score him at 50/33/15/2. Again, pretty much exactly like me. I think we will see that North of the Alps, we have nearly a 50/50 split in EEF/WHG(closer to 60% in Southern Germany, 50% in England, 40% in Scotland and Scandinavia.

All the Steppe people did was lower WHG, and raise ANE. They dropped EEF closer to the Alps, and raised it further North.

ANE being as low as it is in Bronze and Iron Age individuals means that there was an ANE survival in Northern Europe, to reach the 17-19% total, if they are only inputting 33-50% of the genes. He has a good amount of Caucasus ancestry. I'd bet his ancestors originate closer to Samara, than BR1.

Davidski said...

He looks like a mix of BR1 and something from near the Volga.

So he wasn't of pure steppe origin, despite his eastern uniparental markers.

Chad Rohlfsen said...

Here, Barak.

"Archaeologically, Thraco-Cimmerian artifacts consist of grave goods and hoards. The artifacts labelled Thraco-Cimmerian all belong to a category of upper class, luxury objects, like weapons, horse tacks and jewelry, and they are recovered only from a small percentage of graves of the period. They are metal (usually bronze) items, particularly parts of horse tacks, found in a late Urnfield context, but without local Urnfield predecessors for their type. They appear rather to spread from the Koban culture of the Caucasus and northern Georgia, which together with the Srubna culture, blends into the 9th to 7th centuries pre-Scythian Chernogorovka and Novocherkassk cultures. By the 7th century, Thraco-Cimmerian objects are spread further west over most of Eastern and Central Europe, locations of finds reaching to Denmark and eastern Prussia in the north and to Lake Zürich in the west. Together with these bronze artifacts, earliest Iron items appear, ushering in the European Iron Age, corresponding to the Proto-Celtic expansion from the Hallstatt culture."

Eduardo Pinto said...

"It's strange his European-ancestry is most similar to NW Europeans, since his people migrated through east Europe to get to Hungary."

The PCA was screaming that.
If he were so ANE-rich as you all thought him to be he wouldn't be plotting where he is, but in the right upper corner in Gamba's PCA.

barakobama said...

Chad,

IR1 obviously has ancestry from the Caucasus, northern Europe, west Europe, and volga. He doesn't look very special(Yamna-like) but like a mix of modern populations.

"Proto-Celtic expansion from the Hallstatt culture"

Sorry, but I gotsta correct this. You don't need to be an expert to see people confuse Celts with Gauls. Gaul or another name should probably be given to mainland European Celts. All the Celts in Italy, Germany, France, Balkans, Spain, and Portugal knew they were of the same origin and I've read ancient writers say they all called themselves "Celts".

It wasn't until very recent times that linguistics learned the Iron age people of Britain and Ireland spoke a related language. Mainland Celts or Gauls could be from the Hallstatt culture's spread, but defiantly not Celts in Britain and Ireland, who come from an older migration of Celts. If Italic and other Iron-age languages from Italy-central come from the same source that's more evidence of this Indo European-sect was spreading in western Europe long before the Hallstatt culture.

Chad Rohlfsen said...

Barak,
IR1 is closer to Yamna like, according to the "Armenian-Karelian" thing, than anyone so far.

You should read up on the migrations of Iron Age individuals. There are plenty of Hallstatt, and La Tene related movements and items in Britain. Also, Lusatian, from Central Europe, closer to Poland. There were several migrations out of Hallstatt/La Tene influenced groups, from 500BCE, up until the Belgae migrations.

Chad Rohlfsen said...

I do agree that there probably was Celtic spoken in Britain, prior to Hallstatt, but influences and migrations are visible out of that area.

Davidski said...

Late Bronze Age Bulgarian V2

North_Sea 1.32
Atlantic 32.69
Baltic 0
Eastern_Euro 7.14
West_Med 14.99
West_Asian 43.45
East_Med 0
Red_Sea 0
South_Asian 0
Southeast_Asian 0
Siberian 0
Amerindian 0.41
Oceanian 0
Northeast_African 0
Sub-Saharan 0

"Sample V2 was found in a flat cemetery dating to the Late Bronze Age (1500–1100 BC) near the village of Vratitsa, Bulgaria."

http://www.sciencedirect.com/science/article/pii/S000292971300459X

Davidski said...

K13

North_Atlantic 20.53
Baltic 0.01
West_Med 33.77
West_Asian 45.7
East_Med 0
Red_Sea 0
South_Asian 0
East_Asian 0
Siberian 0
Amerindian 0
Oceanian 0
Northeast_African 0
Sub-Saharan 0

Davidski said...

K15 4A Oracle for V2

https://drive.google.com/file/d/0B9o3EYTdM8lQQUhWb0dFNnNSQjQ/view?usp=sharing

ryukendo kendow said...

... ...

What.

Things are just getting weirder and weirder.
"Icelandic + Tabassaran". "Armenian + Basque". Geographical paradoxes all over.

Perhaps IR2 is an endogamous hybrid elite remnant from WHG-BB spread + Kurganization from the East.

And V2 is an remnant of the first West Asian/Caucasus exp into the Med-like pop of Neolithic Europe from centuries earlier.

All extremely conjectural.

Matt said...

@ Davidski, thanks loads for your hard work on this. Really interesting complement to the study which actually tells us completely new information that they couldn't.

Comments on IR1:

Not sure what's up with the WTFs and general surprise. Iron Age sample seems exactly what we would've expected from position in Gamba.
No East Asian, no Siberian. Almost equal levels of Eastern_Euro and West_Asian, with less West_Asian, unlike any modern population.

I thought the idea of East Asian ancestry in IR1 would be wrong placed in terms of modern terms - Where would modern Caucasian people be on a PCA like theirs skews to represented majority Mexican, Finns and Europeans? Somewhere relatively to the centre compared to Europeans. Surprised Alexander Kim (aka Sarkoboros, formerly Ahnenkult) didn't comment on this in his comment on this paper.

(and this is true at both K15 and K13 resolution, countering any argument that Eastern_Euro is somehow absorbing East Asian ancestry into itself and this is "hiding" an East Asian influence).

Suggests that ENA ancestry is still a dubious prospect in Yamnaya, if this sample has a lot of Yamnaya.
Also obviously a projected PCA is going to be closer to reality for a sample like IR1 likely truly close to the centre of the graph than ones at the edge. Projection issues are stronger at the outer edges of the graph become insufficient to capture the variation of samples beyond the modern space, which IR1 is not.

Also makes sense in light of history, right? Even if we think about a steppe origin, for whatever IR1 is, no reason it couldn't ultimately pick up a lot of West Asian ancestry in West Asia (Anatolia, the Aegean) before hitting Hungary. The Bronze Age Bulgarian David put up might be whatever a steppe IR1 could mix with, while itself being a descendant of the steppe via a very different route.

Another comment - If Hungary can be modeled as a merge of this sample and Bronze Age 1, as the PCA results overwhelmingly suggest and which is likely unless we take very strong interpretations of elite apartheid, that also suggests that Hungarian WHG-UHG by the ANE calculator should be (84.27+57.17)/2 or 70%.

(Likewise, question if modern Hungarians would actually look like an 8 population combination of IR1 and BR1's 4 population origins).

Outlines for me that there are accuracy issues with Oracles using Mesolithic samples beyond the modern range of the ADMIXTURE calculators they are based on and the K7 test for ancients.

Helgenes50 said...

That's interesting, the last results of V2, in this case, show that the North Atlantic can be composed of Eastern European.

Chad Rohlfsen said...

V2

EEF 82.87487773
WHG 7.79063069
ANE 9.334491575


Chad Rohlfsen said...

I can't wait to see that Bronze Age Dane. That should be very interesting. Hopefully, he can pull the y-DNA.

Chad Rohlfsen said...

V2, might be another J2.

ryukendo kendow said...

@Matt
The question of where Basal came into Yamnaya and by proxy as an intrusive neolithic input to other IE pops has been touched on here earlier. Personally I thought Caucasian input is more likely than a European/Balkan one, because eef in Europe has too little basal and to much WHG to produce the ratio seen in Yamnaya, and the basal input in yam/IE is explicitly localized to the caucasus in the new paper and also in ADMIXTURE analysis. Simon raises the mtDNA findings of Haak 2012 and how this points to MtDNA input from caucasus to unetice and corded ware.

Davidski replied that archeological evidence suggest a Balkan contribution, and also that a more ME-like/caucasus like pop might have existed further south in the balkans, and I agreed. It seems like V2 might be a smoking gun for this, and that cauc-like input could have come either side up the black sea as of now.

Chad Rohlfsen said...

Ryu,
V2 has WHG and ANE. It's less Neolithic than the EEF. It has to be a Caucasus source for Yamnaya.

I believe the Neolithic is North of the Caspian by 5200BCE. There is no evidence for movement from the Balkans into the East, this early. The Neolithic was adopted slowly by the Dnieper Don, so I don't see how someone that much further East, could be that much more EEF, without an alternate source for the EEF.

ryukendo kendow said...

@ Chad
That's a good point.
Perhaps V2 is instead a somewhat Kurganized and mesolithized individual, and so we've got the cart before the horse as far as directionality is concerned.

What's interesting is that we've finally unearthed the elusive "West Asian"/"Caucasus" component in aDNA in Europe, which simply refused to show up formerly, and that it is post-Bronze and post-IE as others have predicted.

Once we have localized the Basal source for PIE, the picture is pretty complete as far as IE is concerned IMO.

Chad Rohlfsen said...

Here is how West Asian scores at 100%, in the European and Near East calculator

EEF 88.75184039
WHG -2.440100982
ANE 13.6882606

Near East 77.400
ANE 22.600

I still think that a Caucasus source is the only option. mtDNA in the Neolithic Europe is missing the kinds of H that we in the Bronze and Iron Age. Some only match with the Caucasus/Near East, and the Northern half of Europe only, for some.

This person looks like they could be a three way mix of something Neolithic/Mesolithic, with something Steppe, and Anatolian copper/bronze.



Davidski said...

The West Asian component right now looks to me like a post-IE and post-4.2 kilo year event intruder into Europe.

https://en.wikipedia.org/wiki/4.2_kiloyear_event

Chad Rohlfsen said...

David,
It looks like some of it could've come with Indo-Iranians, as well.

ryukendo kendow said...

@ Davidski
Well, I guess its a possibility.
While the bronze age collapse is real, it would be entirely unclear why the results of a NE-Med centered dispersal would cause high West Asian in Russians and none in Basques. The balance of the evidence now is against it IMO.

@ Chad
Those are very nice numbers!
I simply can't wait for the Yamnaya genomes to go through ADMIXTURE. I'm gonna bet, that it will score very high "West Asian", and high "East European" and "North Sea", and that the latter two tracks IE in Europe. I'm also betting that it will not score in, or at most a trace of, baltic.

Davidski said...

Maybe BR1's results are leading me up a garden path?

That should be an early Indo-European genome. Not Proto-Indo-European, certainly, but early Indo-European in Europe. And yet, it lacks any sort of signal that we could attribute to post-Neolithic migrations from the Near East.

Anyway, here are CO1's K15 results. This is the Baden Culture sample.

North_Sea 1.34
Atlantic 31.99
Baltic 2.1
Eastern_Euro 0
West_Med 46.87
West_Asian 0
East_Med 17.5
Red_Sea 0.22
South_Asian 0
Southeast_Asian 0
Siberian 0
Amerindian 0
Oceanian 0
Northeast_African 0
Sub-Saharan 0

Davidski said...

BR1 is best modeled as 3/4 Loschbour and 1/4 CO1.

1 CO1+Loschbour+Loschbour+Loschbour @ 4.707253

But I'm still waiting for KO1. That might be prove a better match than Loschbour.

ryukendo kendow said...

Br1 is too low in ANE.
Also, you just used pre-IE genome's Western position elsewhere to argue for Eastern gene flow into modern day Hungarians and Europeans, and BR2 is West of modern Hungarians and thus cannot be the source of such gene flow.
The only genome we have thus far from a pop whose IE affiliation is sure is IR2, and he has high ANE and high West Asian.

Davidski said...

My argument isn't that BR1's population is the source of the eastern shift in modern North/Central/East Europeans, but rather an eastern population that contributed to the formation of the BR1 population and also, in even greater extent, to the making of modern Europeans.

So the fact that BR1 clusters west of modern Hungarians, and likely has lower ANE than them, doesn't matter. What is important is that we can see a major shift towards Northeastern Europe in Hungary from the Copper Age to the early Bronze Age, and the arrival of ANE.

Chad Rohlfsen said...

All I know is that if we don't start seeing someone that is over 14-16%ANE, then I am wondering if the flooding of the Baltic and North Sea caused a massive flow of Motala and Karelian like people into continental Europe, prior to IE movements. A few people between 9-16%ANE, isn't going to add a hell of a lot if they aren't more than 1/3 to 1/2 of our genes. It's either that, or the amount R1 lineages increasing significantly to the West means that they had a few kids and moved on without them, to new families, or most of their male heirs moved on, until the coast was massed with them. Then what we see is mostly the mixed result. Chalcolithic Germany or Britain, would be great to see.

David, can you do a shared drift with BR1? I am wondering if the Bronze Age replaced much more of the population in NW Europe, and subsequent flows during the Iron Age left little impact.

Davidski said...

BR1 - Eurogenes dataset

https://docs.google.com/spreadsheets/d/1ZT5UbhNOlv0QmStdhvv5QcxpezcRxQ6JL0PvSJX29a4/edit?usp=sharing

IR1 - Eurogenes dataset

https://docs.google.com/spreadsheets/d/18k_5asVmMEiw67Yr1tVUyMeIfFVsKtix0RBnXGoEOq8/edit?usp=sharing

What would be the reason/mechanism for large-scale pre-IE gene flow from Eastern Europe to Central Europe? And why would this be pre-IE gene flow if it came during the Corded Ware period?

Chad Rohlfsen said...

I think that the melting of the Baltic, and flooding of the North Sea, would be the reason for movement into Northern Europe. This would've happened a little before Corded Ware. The flooding began around 6500-6200BCE.

According the this site I'm looking at, the ebbs and flows of temps would put that during a warm period, and IE movements near a cold period. There was also a major shift in climate from 1500-1100BCE, where the Earth was 2 degrees warmer than it is today. That could cause migrations that disrupt trade lines and caused the Bronze Age collapse.

It's just speculation, and not likely to be a major player, but could still be possible on a minor scale.

Either way, something sure as hell happened in Europe, prior to IE migration. IE people, being an Armenian/Karelian-type of mix, doesn't move people from being Oetzi or Gok2 like, to ranges of 30-45%WHG. Yamnaya people didn't have much WHG, if they only had about half as much as modern Northern Europeans. Hunters made a massive comeback. Were they more like Loschbour, or Motala, or a mix of both?

Chad Rohlfsen said...

That means that IE people lowered WHG in the Northern half of Europe... not an increase.

Davidski said...

There's a major problem with calling the Yamnaya 50/50 modern Armenian/ancient Karelian, which is that modern Armenians have quite a bit of ANE.

This means that the Armenian-like EEF were actually more EEF-like than some people here imagine, and the ancient Karelians more WHG-like.

Chad Rohlfsen said...

Armenians only have about 14% ANE, but more basal in their 86% near eastern, than Stuttgart, just by a bit. Reich said the karelians had WHG level above and below modern Europeans. That means half of about 35-45%. So, ie people didn't have much. Even less later, if no trail of WHG exists in the west and Central Asia.

Davidski said...

The reason there's no trail of WHG into west and Central Asia is probably because the Armenian-like EEF that were part of the PIE mix didn't have an excess of WHG over Stuttgart, which makes sense, since Stuttgart is a genome from Western Europe.

Keep in mind that in Laz et al., WHG is not actually WHG, but excess WHG over what Stuttgart has, because EEF are a mix of Basal Eurasian and WHG.

Chad Rohlfsen said...

But if the Karelians had WHG at levels of modern Northern Europeans, we still have another 20% of WHG that needs to disappear. That's about another 8% of Basal, over Stuttgart.

Either way, it's mathematically impossible for IE's to have a ton of WHG, yet still have all of that EEF and ANE.

I can show that math again. The only way to reconcile is either IE's in Western Yamnaya had a lot less ANE and EEF to make room for WHG with the Caucasus as the source of EEF, since the West didn't have it, or for the WHG to be from a local source, waiting in the wings.

Davidski said...

Maybe you're confusing Corded Ware with Yamnaya?

We already know there was a difference between Corded Ware and Yamnaya, with Corded Ware being the temperate interpretation of the Yamnaya pckage, with foraging as one of its main forms of subsistence, and genetically only 73% Yamnaya-like.

Keep in mind that the Yamnaya package was spread across most of Northern and even Eastern Europe by Corded Ware, not directly by Yamnaya.

barakobama said...

" Reich said the karelians had WHG level above and below modern Europeans."

Did Reich really say this? I just wanna make sure. Karelian-hg WHG was then something in between 20(or even less)-49.5%(in laz terms).

It's good to finally get a somewhat clear answer from Reich. Yamna was then under 25% WHG in Laz terms. The post-Neolithic WHG rise in Europe then must have been local not Indo European. Finally the mystery is solved.

This also means Yamna may have been as much as 40% ANE. Considering that all Hungarian genomes up to BR1(2,080BC) completely lack ANE, and BR1 was 5-15% ANE, it means he had at the most something like 37.5% Yamna-blood.

It's interesting there could have been populations throughout mainland Europe before the Indo Europeans who were mostly WHG. WHG slowly rises in the Sardinian-like Hungarian samples. C01 scored only 39% ENF in ANE K7. The WHG-comeback and native-Sardinian-like farmers must to be how I1 and I2a2(both found in Neolithic Hungary) became such major paternal lineages of Germans.

Davidski said...

You guys are still blissfully ignoring the fact that Yamnaya didn't move across Northern Europe - Corded Ware did.

Corded Ware had its roots in the forest steppe zone of Ukraine and Russia.

Think about what that means.

Chad Rohlfsen said...

Let's just say that the average German was like Gok2, 75/25. That would mean at a 33% input, there couldn't be any EEF in the population coming in, to get to 50%, today. It's just not possible.

A 50/15/35 mix makes a lot more sense, for the steppes, and a population that was around 50/50 already, makes more sense. Even at a 40% replacement, the EEF doesn't go up much. It only goes to like 10%. That doesn't match the impression given at the conference. I'm not trying to be a thorn in your side brother, but it has to be something similar to this, with WHG already rebounded, before and during the IE expansion.

Chad Rohlfsen said...

I'm going by what Barak was told by Reich. The Karelians, he said, had more and less WHG than modern Northern Europeans. That puts it in the 30-50% range, for their people. If Yamnaya is only half of that, then it means their people should fluctuate in the 15-25% range. If Corded Ware is 75% Yamnaya, it's WHG can't be a great deal more, since they are entering Neolithic cultures in Poland and Germany. It could be more WHG than EEF, but certainly not enough to cover Northern Europe's gain in WHG.

«Oldest ‹Older   1 – 200 of 308   Newer› Newest»