search this blog

Sunday, June 7, 2015

101 ancient Eurasian genomes?


Update 10/06/2015: 101 ancient Eurasian genomes (Allentoft et al. 2015)

...

I'm hearing rumors that Nature is about to publish an important ancient DNA paper on the Late Neolithic to Bronze Age transition in West Eurasia. The paper is one of the main outcomes of The Rise, a Gothenburg University-based project focusing on the rise of Bronze Age societies in Northern Europe.

Below is a Google translation of a recent article from NovostiNK which is almost certainly about the imminent publication of this paper.

It's interesting to note the claim by one of the Armenian researchers that present-day Armenians are almost indistinguishable from the Bronze Age samples from what is now the Republic of Armenia.

This sounds very reasonable based on what we've learned to date from present-day Armenian variation (for instance, see here and here), but it's difficult to know what it means exactly without knowing what types of analyses were carried out and thresholds used. Hopefully we'll learn that in "a few days".

Today Armenians are descendants of people who lived in the territory of Armenia 5 thousand years ago, told reporters on Friday the head of the Laboratory of Molecular Biodiversity Institute of the National Academy of Sciences Levon Yepiskoposyan, referring to the international genetic study.

Genetic analysis of 101 DNA samples from different parts of Eurasia was made to clarify the genetic portrait of a man of the Bronze Age.

Eight DNA samples were taken from various archaeological sites throughout Armenia. They date back to the middle and late Bronze and Iron ages.

"The results of genetic studies have shown that DNA samples from the Bronze Age have been found on the territory of Armenia have a genetic portrait that is almost indistinguishable from the genetic portrait of people living today in our territory," said Yepiskoposyan.

He stressed that based on the results, we can conclude that modern people living in the territory of Armenia has enough deep roots.

"By participating in this study, we were able to solve not only the question of the genetic, historical and archaeological analysis, but also to some extent to respond to allegations that the Armenians living in the territory of Eastern Armenia only 200 years" said Yepiskoposyan.

He also noted that Armenia was the only country in the region that participated in the study. "Perhaps the reason is that we do not have the problems associated with our history and our neighbors apparently have such problems," said Yepiskoposyan.

"The study was conducted to study the genetic portrait of generations past and solve one of the most important questions - are we, modern people, direct descendants of the peoples living on the territory of a 5 thousand years ago," said Yepiskoposyan.

In turn, the director of the Institute of Archeology and Ethnography of the National Academy of Sciences Pavel Avetisyan added that the results of this study, conducted at Copenhagen University, will be published a few days later in a scientific paper, authored by 44 experts from 13 countries.

"Armenia for the first time take part in this kind of program, and the results we got were quite interesting," said Avetisyan.

Source: Современные армяне являются потомками людей, населявших территорию Армении 5 тыс. лет назад

218 comments:

1 – 200 of 218   Newer›   Newest»
Aram Palyan said...

I think the number of 5000 years is a extrapolation from MBA/LBA data.
Unfortunately they didn't mention aDNA from Early Bronze Age that could be quite interesting.
But it is the just beginning, so let's be patient and optimistic.

Davidski said...

I can believe that Armenians are almost indistinguishable from Armenian samples dating to 3500 years ago, or the Middle Bronze Age.

I don't believe that Armenians are almost indistinguihsble from people living in what is now Armenia 5,000 yeas ago. So I'd say that guy got a bit ahead of himself there with that quote.

Aram Palyan said...

I think it is also the result of some journalistic sensationalism. :)

Mike Thomas said...

Cannot wait for full pub.
But im detecting political undertones ....

Nirjhar007 said...

You mean the Reporting is misleading?.

Davidski said...

The comment that Armenians are direct descendants of the people who lived in Armenia 5,000 ago is misleading, because the oldest Armenian samples are only 3,500 years old.

Nirjhar007 said...

OK Where is the source that gives the Information that they are 3.5 KYO? if you can show its important i'm not kidding!.

Davidski said...

Because the article says that the oldest Armenian samples are from the Middle Bronze Age, which was 3,500 years ago.

Anyway, you'll find out soon that they're no older than 3,500 years, so there's no point debating this.

Nirjhar007 said...

Aram,
As Said ''Eight DNA samples were taken from various archaeological sites throughout Armenia. They date back to the middle and late Bronze and Iron ages.''
So IMO the Earliest samples of Middle bronze age can go back ( though i'm not sure as i don't have accurate info on Armenian Bronze Age) can go back as far ~ 2500 BC?.

Aram Palyan said...

Nirjhar

I think this study will not give a definitive answer on the question that interest all of us. I mean the transition from Neolithic to Bronze Age in all Eurasia and the question of IE identity.
We desperately need some BA aDNA from South Central Asia to understand better the origin of Indo-Iranian branch of IE.

Nirjhar007 said...

So Its around 1500 BC?:(

Aram Palyan said...

I think we just need to wait little bit.

For the question of Bronze Age in South Caucasus it is a matter of debates. Georgian archaeologists have calculated with radiocarbon method that the Bronze Age start at 3600 BC but some others think it is too early and they place the start at 3200BC.
The early Bronze Age is associated with Kur-Arax culture. It ends at 2700-2500 BC.
The late Bronze Age is between 1800-1200 BC. After 1200 BC starts the Iron age.

Nirjhar007 said...

ok So it can be before 2000 bc Period and close to 2500 bc! BTW on Indian aDNA i can only say we will have aDNA results soon!:).

Aram Palyan said...

The timeline of Marc Haber's study shows various admixture event between 3000-2000 BC.
We also see some important changes during this period on archaeological level.
And the most amazing thing is that the legendary founder of Armenians Hayk is dated to 2500 BC.

My theory is that is that R1b came during this period. But I could be wrong.

Davidski said...

The Armenian samples in this paper won't be older than 2,000 BC. You'll see.

Nirjhar007 said...

According to My Version Armenians are the people of the Original PIE Stock of S Caspian-Central Asian area and then they should show continuity from the Neolithic period!....

Davidski said...

Well your version is wrong. But in any case, this paper doesn't have any Neolithic samples from the region, so it won't tell us anything about that.

Nirjhar007 said...

To be precise the Continuity should be observed from the 4500-4000 BC Period!..

Aram Palyan said...

BTW
This is the website of the project "The Rise"
http://the-rise.se/

They have a collective multidisciplinary approach to understand the Bronze Age. I think it is a wise idea because the amount of scientific information is so huge that a single person will have hard time to keep in mind all this interrelations. I think Asian and Near Eastern scholars should also have such projects.

Aram Palyan said...

According to My Version Armenians are the people of the Original PIE Stock of S Caspian-Central Asian area and then they should show continuity from the Neolithic period!....///

Nirjhar
So in this case the R1b would come to Armenia from South Caspian-Central Asian. Right?
But how the continuity will be kept if they are a newcomers? Shouldn't they create an admixture event?

Nirjhar007 said...

David Unless the reporting was made by people who where drunk or Nationalists :D we should see samples before 2000 BC period also!
Aram Yes I agree maybe Money is a matter?.

Nirjhar007 said...

Aram,
''But how the continuity will be kept if they are a newcomers? Shouldn't they create an admixture event?''
Admixture event? yes in that (4500-4000 bc) period but after that there is the continuity...

Mike Thomas said...

The problem is terminology. It's the same in Balkans. Every country's archaeologists have a different definition of "Bronze Age". To Avoid confusion lets just work with the raw dates

Nirjhar007 said...

Of Course.

Fanty said...

" Every country's archaeologists have a different definition of "Bronze Age"."

That may be.

I realised that with the middle ages.
English (US and UK) and German (Germany, Switzerland, Austria) speaking sources dont agree on the dates for "Early", "High" and "Late" Middle age and connect those periods also to different indicators/definitions for those 3 ages, wich also apear somewhat nation based on both sides (medieval Britan plays a great role in the US/UK ideas about the 3 ages and the Empire for the German ideas)

Davidski said...

Here's a list of Armenian samples from The Rise project.

Hatsarat Bronze Age ca. 2000 BCE 15 (RISE417)
Nerkin Getashen Bronze Age ca. 1200-1300 BCE 13 (RISE415)
Noraduz Bronze Age ca. 1200-1300 BCE 12 (RISE414)
Noraduz Iron Age ca. 700 BCE 18 (RISE420)

Source:

http://www.nature.com/srep/2014/141127/srep07104/full/srep07104.html

Colin Welling said...

@Mike

Cannot wait for full pub.
But im detecting political undertones ....


A huge amount of it which is probably why he jumps the gun as david pointed out. I won't be trusting his opinion but thankfully this is a cooperation.

So how do you think the arminian heritage and language came together? They have a lot of r1b and ANE but little WHG like dna as far as i know. I would guess that the Armenians got their language and r1b largely from the same migration out of the steppe but I'm wondering why they wouldn't have more WHG. Actually this is an open question to anyone, especially david. The lack of WHG in eastern IE peoples, with steppe heritage, makes sense since their steppe heritage probably comes from the eastern steppe hight in ANE but low in WHG like stuff. But the armenians are somewhat more perplexing seeing how they might have gone by a western route, round the black see, to get to armenia. Perhaps the very southern yamnaya had much less WHG and was more of a EEF/ANE mix with r1b mostly from the ANE side. Or perhaps there was some combination of founder effect in armenians with IE r1b along with diffusion of EEF/ANE from neighbors.

Davidski said...

I'll wait before commenting any further, because I have to see the results or even run the Armenian genomes myself to be able to say anything really useful.

Alberto said...

Pity, 2000 BC is a bit too late to be too informative. I hoped for an Early Bronze Age sample.

BTW, I think that the Anatolian (Kumtepe) sample from that other abstract will be from the EBA (c. 3000 BC), quite uninformative for the Neolithization of Europe (contrary to the abstract's claims) but maybe more informative for the Indo-Europeanization of Europe. According to this article, the human remains found at Kumtepe are from the EBA (from Troy's proto-settlement) and not from the earlier Late Neolithic period (4800-4000 BC):

http://www.tayproject.org/TAYages.fm$Retrieve?CagNo=2011&html=ages_detail_e.html&layout=web

We'll see...

Colin Welling said...

BTW, mike, did you see the news on a neollithic northwest anatolian who had something besides EEF, WHG, and ANE? I think this other component should be fairly well represented in neolithic anatolia. And its is safe to say that not much anatolian entered the european neolithic. We don't know what the levant was at that time but I am still not convinced that Levant was the source of european EEF minus some WHG.

This anatolian guy may support the idea that the southern balkans was already heavily EEF before the neolithic and this anatolian guy may have been too related to the levant for the levant to be the source of the european neolithic. I guess we will see.

Davidski said...

The late Neolithic sample from Anatolia shows that the ancestors of early European farmers migrated to Europe from Anatolia before a genetic shift happened in Anatolia. That's actually what the Omrak et al. abstract says:

"The scene presented by Kumtepe is compatible with geneflow into Europe from or through the neolithic core area in Anatolia. And it is likely that this occurred early, perhaps just after the neolithic core area had been established in southeastern Anatolia."

This is compatible with other ancient data from Neolithic Anatolia, which will probably be made public fairly soon.

Aram Palyan said...

Colin
You can trust him. He has a paper on Balkanic markers an idea that is quite unpopular in Armenia. Well what You see is a text composed by journalists, based on journalist's unscientific questions, who have little clue in genetics. I am pretty sure most of the journalists didn't understand what he tried to explain them. :)
In any case You are right this is a great international cooperation.

Mike Thomas said...

@ Colin W

Hey. Long time..

*"So how do you think the arminian heritage and language came together? "

On the one hand, one cannot look past massive migration = language change. I accept that, and anything contrary would be rare/ unusual. But the converse isn;t necessarily true. Genetic/ population continuity does not equal language continuity, esp when we're talking about Bronze Age to ~ Late Antiquity. Language changes can occur through multiple avenues...

* "makes sense since their steppe heritage probably comes from the eastern steppe high in ANE but low in WHG like stuff. But the armenians are somewhat more perplexing seeing how they might have gone by a western route, round the black see, to get to armenia"

Or perhaps, it came directly from the east, where there is no WHG.

* "Perhaps the very southern yamnaya had much less WHG and was more of a EEF/ANE mix with r1b mostly from the ANE side'

If by 'very southern Yamnaya' you mean that right on the Black Sea shore line - Id expect it to have rather generous amounts of WHG (?)


* "BTW, mike, did you see the news on a neolithic northwest anatolian who had something besides EEF, WHG, and ANE? I think this other component should be fairly well represented in neolithic anatolia. And its is safe to say that not much anatolian entered the european neolithic. We don't know what the levant was at that time but I am still not convinced that Levant was the source of european EEF minus some WHG.

This anatolian guy may support the idea that the southern balkans was already heavily EEF before the neolithic and this anatolian guy may have been too related to the levant for the levant to be the source of the european neolithic. I guess we will see.'

Yes i saw that. Im not sure about it, exactly what it;ll be, but as David said, i don think it'll be anything too mind blowing.
Demographically, it has been long assumed that the ENF component in EEF (to use genetic terms) came straight from the Levant, as if by sea, with little input over the Anatolian-Bulgarian landbridge. This might by partly correct - at least for the early neolithic. But research in Turkey has vamped up recently (eg see Mehmet Ozdogan's works - on Academia). There was a healthy forager, pre-Neolithic population in Anatolia. The same is *not* true for the southern Balkans. Virtually nothing in Bulgaria, and a mere handful of sites in Greece, like Franchthi etc - where nevertheless - there is stratigraphic hiatus between final Mesolithic an earliest neolithic settlements. If a handful of GReek foragers did admix, it'll hardly be genetically significant. The first European foragers (those 'UHG"), I hypothesize, to significantly admix with incoming farmers can only come from the northern Balkans - Hungary region - as we indeed caught K01 'in the act'.

Davidski said...

Aram,

Colin is confusing the Armenian scientist quoted in this article with an Armenian guy who's been posting somewhat confused "scoops" about the new study on an Armenian Facebook site.

They're two different people. Although I'm still wondering why anyone would say based on data from the MIDDLE Bronze Age that Armenians are direct descendents of the people who lived in Armenia 5,000 years ago. But, as you say, it might have been an error made by the journalist, who maybe confused 3500 YBP with 3500 BC.

Roy King said...

Clearly, according to Near Eastern archaeological nomenclature, 1500 BCE is LBA and 2000 BCE is MBA. In either case, aDNA from these dates precedes the IE Armenian language--The area was speaking an Uratian/Hurrian language at the time. What will be fascinating also will be the results from:
West Caucasus, Marchenkova Gora Bronze Age ca. 3000-2000 BCE
Bulanovo ca. 3000-1500 BCE

both of which are RISE samples and may come out with the 101 aDNA paper.

Roy King said...

@Alberto
"According to this article, the human remains found at Kumtepe are from the EBA (from Troy's proto-settlement) and not from the earlier Late Neolithic period (4800-4000 BC)"

I don't know why you insist that the Kumtepe sample is EBA. Anatolian archaeology is clear about the Late Neolithic Kumtepe being ca 4500 BCE. the term proto-Troy is used on account of Kumtepe's geographic relationship to EBA Troy, not a direct temporal relationship.

Alberto said...

@Roy

I gave a link, you can read there. There are no human remains from the Late Neolithic time. The ones found where dated to the EBA.

But who knows, maybe they did find other ones for this study and they are indeed from 4500 BC. Who knows. I'm used to misleading abstracts already, that's why I think this might be another misleading one.

Nirjhar007 said...

Disaster 2000 BC will not be much of a help.....

Aram Palyan said...

Roy
You mean that modern Armenians are the direct descendents of Hurrians?

Nirjhar007 said...

Aram,
Urartian Do Show some levels of IE lexicon http://books.google.it/books?id=M2aqp2n2mKkC&printsec=frontcover&hl=it#v=onepage&q=urartean&f=false
E.g., Ur atu 'to eat', PIE *ad/ed, Ur aš 'to sit, PIE *as/es, Ur burgana 'fortress', PIE *bhrgh- (German Burg, Greek pyrgos).

Aram Palyan said...

Nirjhar

I know that, they are other words also from PIE in Urartean. But in the Steppe theory this would be some later borrowings from IE into Urartean. It is assumed that Hurro-Urartean languages are Neolithic languages before IE.

Nirjhar007 said...

Urartu AFAIK appears quite late, it is possible that Urartians conquered their historical kingdom, suppose from the Caucasus since their language has been connected with Northeast Caucasian languages. The area of Van maybe was already occupied by Armenians or other IE Folks!!. So the substratum can be rather an adstratum or superstratum, an influence due to domination and historical contact.

Davidski said...

Weren't the Hurrians ruled by an Indo-Aryan Mittani elite?

That seems to be the consensus.

Nirjhar007 said...

They weren't Indic but an Iranian Branch close to the religion of Indic!

Aram Palyan said...

F. Konig thinks that Urartean was a small elite language that was not spoken by the common people. I think the situation was mix. In some states Hurrian ruled IE people , in others IE people ruled Hurrians. There was a multitude of 'Hurrian' states during Bronze Age.
There is a state called Armanu-Subartu ( Arme-Shubria 1300 BC ) it is thought to be a mix of Armenians and Hurrians

Nirjhar007 said...

Ah that's a practical scenario really.

Roy King said...

@Davidski,
No, the earliest Hurrians, such as those from Urkesh (circa 2200 BCE), did not have an Indo-Aryan elite. later, in the LBA (1500-1300 BCE) do you find the Mitanni whose elite had Indo-Aryan names. Clearly Indo-Aryan enters the region in the LBA.

Nirjhar007 said...

The weren't Indo-Aryans for truths sake!:)

Nirjhar007 said...

Guys,
Jackson has informed in the previous thread that-
//I was told by someone that is working on the ancient Armenian DNA project that all of the ancient samples that they have tested so far are female except for one.//
any comments?.

Davidski said...

Right, so 100 of the samples are females and one is male?

I doubt it.

Karl_K said...
This comment has been removed by the author.
Nirjhar007 said...

He meant among the 8 samples of Armenia?:).

Davidski said...

I can believe that seven out of the eight ancient Armenians are female, but not that 100 of the 101 samples in this study are female.

That would be the dumbest sampling strategy in the history of science.

Nirjhar007 said...

LOL:D

jackson_montgomery_devoni said...

7 out of the 8 ancient Armenian samples are females apparently. At least that is what I have been told. The samples from elsewhere I am sure are a healthy mix of both genders.

Krefter said...

My guess on the Bronze age Armenian Y DNA is J2a.

Mike Thomas said...

All we need is 5-8 good coverage genomes from Central Asia ; neolithic to BA, and this entire debate will end (?!)

Romulus said...

It will be interesting to see when R1a came to replace R1b on the Steppe

Davidski said...

R1a didn't replace R1b on the Samara pre-Ural steppe.

Still plenty of R1b there.

Krefter said...

"It will be interesting to see when R1a came to replace R1b on the Steppe"

Slavic(R1a-Z283), Germanic(R1b-L11, I1-M253, I2a2a-M223), and Finno-Urgic(N1c) have erased most of the older R1b-ht35(Z2105 and L23) in Russia. mtDNA from Yamnaya-types though may have survived very well there, and invading groups from mainland Europe had a bunch of it too.

Aram Palyan said...

I guess we need to wait to 11 June at least
http://www.nature.com/nature/archive/index.html
The last issue is 4th June the next will be 7 days later.

Nirjhar007 said...

Mike,
//All we need is 5-8 good coverage genomes from Central Asia ; neolithic to BA, and this entire debate will end (?!)//
Yes areas like of Tajikistan,Kazakhstan,Turkmenistan Northern Iran and of Course Armenia:)...

Davidski said...

Aram,

I was betting on a Sunday night/Monday morning, because Haak et al. and a couple of other big papers came out at that time, but you might be right about Thursday.

Lazaridis et al. was published on a Thursday.

Davidski said...

Anyway, this is the page to check.

http://www.nature.com/nature/research/biological-sciences.html

Mike Thomas said...

Apart from Armenia, do we know where else they sampled ?

Davidski said...

As far as I know:

- Late Neolithic, Bronze and Iron Age remains from Germany, Poland and/or Scandinavia
- Copper, Bronze and Iron Age remains from Bulgaria
- Bronze and/or Iron Age remains from Hungary
- Sintashta from Kazakhstan
- Maikop from Russia
- Yamnaya from Russia
- Afanasevo from Russia

I haven't been able to confirm any of this, but I'm pretty sure most of that list is correct.

Krefter said...

We've seen everything in that list except Bulgaria and Maikop. The Bulgarian ones might reveal details about modern Balkan and Italian origins.

Mike Thomas said...

Wow !

Davidski said...

Krefter I'd say you're confusing Afanasevo and Sintashta with the South Siberian Andronovo/Scythians. We've seen nothing to date from Afanasevo or Sintashta.

Also, this will be the first time we get a good look at Bronze Age samples from Scandinavia, and maybe also Poland.

Colin Welling said...

@Romulus

It will be interesting to see when R1a came to replace R1b on the Steppe

Im guessing it may have happened with the fall of CT culture, and their subsequent movement eastwards, a migration which mallory proposed. Another possibility is late CW derived migrations moving east. Those are the only scenarios that fit the time frame and supply the demographics as far as i can tell. R1a needed to be prominent on the steppe by the time of the androvono 2000 bc.

Based on the r1b findings in the samara in mesolithic sample, all 7 of the yamnaya samples, and somewhat varied nature of r1b types id say that R1b dominated the eastern yamnaya. It is pretty obvious that bell beakers, and hence west europeans, got their r1b from the western yamnaya. Given the dominance of r1b in the bell beakers and western europeans we should expect r1b to dominate the western yamnaya too. I think the dominance of r1b in the west (bell beaker and modern west euros) is more statistically relevant as far as inferring high rates of r1b in a sect of the yamnaya. Given what is likely a very high prevalence of r1b in the western yamnaya, to the near exclusion of r1a, i think its unlikely that we will find much r1a in the yamnaya across their horizon. Sure the yamnaya horizon was big but it was also highly mobile and should have exhibited a significant degree of continuity across its vast span, where not just one but both ends probably lacked r1a. In short, I agree that r1b was largely replaced on the steppe.

To further argue the point of the a majority replacement of r1b on the steppe, as I noted a while back, yamnaya-like r1b seems to have settled predominantly in the areas where the first IE languages settled. That is Turkey (anatolian), Altia region (Tocharian), and west europe (Italo-Celtic). It also seems that certain areas were better at preserving earlier steppe dna such as the Southern Urals and the Altai. Both areas are mountainous (different habitat from the surrounding area) and both areas harbor more yamnaya-like r1b. Also, both the southern Urals and the Altai have better preserved earlier steppe-like mtdna compared to the surrounding area.

The story I suggest is that yamnaya mostly left the steppe towards turkey, altai, and west europe. Europeans just northwest of the steppe, high in r1a and fueled by populations related to the CW or CT, expanded into the steppe thereafter and replaced most of the r1b that remained. The expectional areas that did preserve some of the yamnaya-like r1b from the eastward movements of r1a were the more isolated mountain regions of the southern urals and the altai.

Colin Welling said...

As far as I know...

Still no CT? But wow, overall that a great spectrum of samples.

Aram Palyan said...

Sintashta is older than Andronovo.
Also Sintashta was related to copper mining people. If Grey's theory is right then we could see some R1b there. These R1b later could remain among Bashkirians.

Also Sintashta is important from the point of testing Proto-Indo-Iranian theory.


Davidski said...

CT cremated their dead, so there aren't many remains to choose from apparently.

But CT was in large part derived from Starcevo, and we have one sample from this culture, as you probably know.

Corded Ware were obviously one of the earliest populations rich in R1a-M417, so I'd say this map teaser from Allentoft et al. is very revealing about what we'll see in the paper, hopefully this week....

https://drive.google.com/file/d/0B9o3EYTdM8lQQlB1N3hZbFhfUVU/view?usp=sharing

Mike Thomas said...

Ok so if, as the graphic suggests, Western yamnaya was R1a; to which it spread to CWC and Afansievo; where did west european R1b come from ?

Mike Thomas said...

Colin,
CT stuff is in process.
See the abstracts previously posted

Davidski said...

R1b started its expansion in Central Europe with the Bell Beakers, who appeared at the tail end of the Corded Ware period.

So take what you will from that. But I think it's better to wait for this paper before getting too deep into the topic.

Krefter said...

@Mike,
"Ok so if, as the graphic suggests, Western yamnaya was R1a; to which it spread to CWC and Afansievo; where did west european R1b come from ?"

Maybe. It's probably just based on autosomal DNA though. We only have Yamnaya Y DNA from Samara. R1a and R1b could have been in every part. It doesn't have to be as simple as, west was R1a and east was R1b. If one of the Yamnaya's turn out R1b-L51, there isn't much more debating where West European R1b comes from.

BTW, the greatest diversity of R1b-L11 is in Central Europe. R1b-L51(xL11) is mostly found in East Europe. There's a very clear route that was taken. I'm tired of the controversy over West European R1b lingering on, it's pretty clear where it comes from.

Mike Thomas said...

Yes, of course

Aram Palyan said...

Mike
I think it is pretty clear that L51 came from steppe. But when it came is a question.
If it was already evolving in West Europe during Yamna period then what was their language?
So it must appear after Yamna reached Europe to be able to spread the IE.

Alberto said...

But it's still quite a mystery why R1a stayed only in Eastern Europe and R1b (and only R1b) moved to South and West Europe, no? If both where intermixed, why this separation? Especially when R1b seemed to be a minority among Steppe people (judging by CW and by modern distribution).

Also, is Armenian R1b derived from Yamnaya R1b (and if so, again, why R1b and not R1a came there)?

Davidski said...

There were groups on the steppe that were exclusively R1a or R1b going back thousands of years because of their strongly patriarchal culture.

That's why basically all of the Tarim Basin and Andronovo remains are R1a, while the Samara Yamnaya are all R1b.

By pure chance a small R1b steppe clan formed into the Bell Beakers and eventually took over Western Europe.

Colin Welling said...

R1b started its expansion in Central Europe with the Bell Beakers, who appeared at the tail end of the Corded Ware period.

R1b expanded up the danube into hungary as a western extension of the yamnaya. Then their descendants became a part of the eastern BB who moved west. The route and ynda are separate from CW.

Davidski said...

Yeah, so what? They were a tiny group of people during the Late Neolithic. We have no idea at the moment which river valley they crawled out of.

Mike Thomas said...

Well this study looks to more or less solve lingering questions for europe. Apart from the lack of Ukrainian samples; we have more russian yamnaya; Central European stuff; and bulgaria. Looks very promising !!

Davidski said...

I never said it didn't have any Ukrainian samples. I actually did see a comment online a while back that some of the samples are from north of the Black Sea, so they might be from Ukraine.

Colin Welling said...

Yeah, so what? They were a tiny group of people during the Late Neolithic.

A tiny group that changed the genetic landscape of western europe, or more something more akin to a mass migration.

Alberto said...

Yes, very promising. It can be a big step to answer many questions if we really get all those samples.

But wasn't Maikop DNA taken by the Lazaridis, Reich, Patterson... team? Do these teams share samples or maybe this other team also took independently DNA from Maikop?

Davidski said...

No Colin, the Corded Ware expansion was a mass migration from the east.

The Bell Beaker expansion in Central Europe that came later was an in-situ expansion from a couple of local hot spots, like Bohemia.

Davidski said...

Alberto,

I don't know for sure if this Allentoft et al. paper will have Maikop samples, but it's one of the rumors I picked up from a pretty solid source.

Anyway, yes, Maikop and other Caucasus samples are being tested by other research groups. I have no idea if they share their samples.

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

I didn't say that a small clan took over Western Europe by pure chance.

I said that a small R1b steppe clan formed into the Bell Beakers by pure chance.

It was also by pure chance that an R1a steppe (Repin Culture?) clan ended up moving all the way to the Tarim Basin.

What if their patriarchs decided to settle in different parts of the steppe a few thousand years earlier?

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

There was probably quite a bit of chance involved, especially during the hunter-gatherer days on the steppe. Some guy managed to spear an aurochs, while another got trampled by it, and things went from there.

Anywho, if you guys in Europe have a bit of time and cash to spare, you can do this course at Copenhagen Uni. Lecturers include Morten Allentoft and David Anthony.

http://rootsofeurope.ku.dk/roe_sommerskole/

Bronze Age population genomics of Eurasia - Morten Allentoft

Krefter said...

Insights into human history from genetics will be in history books and taught in schools in the future. No doubt about it. In my Geography book it says Slavs came from Northern Asia.

Nirjhar007 said...

Krefter,
Uh From where exactly?:D

Krefter said...

Just North Asia in general. It's one sentence and is stated as a consensus theory. The book is used to learn about the modern world and geography of the world, so origins don't matter. My point was that when it comes to pre-history, genetics help a lot. Also, before recently no one had any idea who the ancestors were of anyone before several hundred or a few thousand years ago. Therefore it's only a matter of time before insights learned from genetics will be in school books.

Mike Thomas said...

Krefter
Lol what kind of history books do they use in USA ?

rozenfag said...

> In my Geography book it says Slavs came from Northern Asia.

That's something strange. Most of the Russian historians agree that proto-Slavic language existed at least till 6-7th century and at that time proto-Slavs lived somewhere to the north of Carpathian mountains, though they disagree about exact location.

Karl_K said...
This comment has been removed by the author.
Karl_K said...
This comment has been removed by the author.
Mike Thomas said...

Karl

There were likely different subclades of R1b present and R1a. Perhaps only few came to dominate due to chance; as related groups also had similar structures. But the overall mechanism, "driving force", wasn't chance; but a very real phenomenon.

Colin Welling said...

The Bell Beaker expansion in Central Europe that came later was an in-situ expansion from a couple of local hot spots, like Bohemia.

First off, we have to admit that the quick change in the autosomal composition of west europeans by the addition of a large chunk of yamnaya-like dna was the result of a mass migration. There is no way around this. It is not feasible that a small group of yamnaya changed the autosomal dna of the west in such dramatic fashion. So this mass migration of r1b people came from a significant population dominated by r1b on the steppe.

The primary path they used was along the danube, temporarily settling in hungary, the only place in the BB world with a massive yamnaya settlement. CW does not fit the bill as the main source of the yamnaya autosomal component in west europeans because CW were r1a and CW territories like belarus and czech republic do not have more of the yamnaya component than northwest europe (the clear descendants of bell beakers).

The yamnaya migration to hungary, a territory within the eastern BB range, was not a dead end...

epoch1970 said...

@Collin

I think we hardly know the genetic composition of Western Europe. We have very little neolithic (0), mesolithic (1) or Bronze Age (0) aDNA of the area comprised of France, Benelux, Northwest Germany and Britain.

There is a theory that has (part of) the rise of BB situated at the lower Rhine. The megalithic culture had its core there. We need that aDNA.

Simon_W said...

That in ADMIXTURE analyses, North European-centered and European HG components are very weak in modern Armenians, has been known for a while. However, some people (like David) think that this is because modern West Asian components include quite a bit of steppe ancestry. Whether that's true could in theory be tested with aDNA. It would already be nice to see what was in Armenia before the modern West Asian genomic structure was established there.

The very weak WHG admixture seems to rule out an Armenian origin in the Balkans. Unless what arrived from the Balkans was a linguistically Indo-Europeanized Sardinian-like population, which seems unlikely.

The lowest f3 stats for Armenians involve Sardinians + Central and South Asians. Haber et al. dated this admixture event to roughly 2500 – 3000 BC. As was explained on p. 46 of Haak et al. 2015, the (very similar) f3 stat with LBK_EN + Sindhi is more negative than with Yamnaya + BedouinB. This means the most important admixture event in the formation of the Armenian gene pool was on a west-east axis, not on a steppe-Near East axis.

Simon_W said...

Probably a south central Asian population carrying some "teal" migrated westwards, perhaps also some forms of R1b, and mixed with rather EEF-like people, perhaps around 3000 BC.

Simon_W said...

I can't rule out that these were the PPIE. I simply don't know enough about the presumed mutual lexical borrowings between PIE and ancient Near Eastern languages. A conceivable scenario would be that PPIE moved westwards along the South Caspian, then Anatolian branched off, followed by Tocharian, then one branch entered the steppe, and those who stayed behind around Armenia evolved into the Albanian-Greek-Armenian branch, from where the Greeks and proto-Albanians (Illyrians? or whatever Balkan language) eventually moved to the Balkans. At least the latter event would fit the overall genomic structure of Greeks and Albanians better than a steppe origin (strong West Asian component, plenty of J2).

Alberto said...

Yes, we need more data to understand the R1b phenomenon in Europe.

I was looking at Haak et al.'s paper and found something slightly strange. When they test modern populations based on the WHG/EEF/EHG(or Yamyaya) model, certain populations need input from a different source to get a good score. For example, Mordovian, Finnish and Russian need Nganasan as a 4th population, while Maltese, Ahkenazi_Jew and Sicilian need BedouinB.

After using all 5 populations (Figure S9.27, page 124), the population with the worst score (higher resnorm) is Scottish, and to a lesser degree Tuscans (when using Yamnaya, in Figure S9.27 C). It is subtle, but noticeable.

Scottish don't take any BedouinB, so I was wondering which population might improve their score. Most likely it is a Georgian-like population (or Tajik, Lezgin, maybe even Balochi?). But if so, why would they need it, while Czechs get a fine score with just Yamnaya (and somehow strangely, also English)? If we do get some ancient Armenian or even better Maykop DNA these days it would be interesting to test and see.

epoch1970 said...

"At least the latter event would fit the overall genomic structure of Greeks and Albanians better than a steppe origin (strong West Asian component, plenty of J2)."

But PIE has ties with proto-Uralic. The similarities are beyond loan-words: me- for I, ne- for negation. There is also the possibility that PIE was "built" on a Northwest Caucasian language. At the very least there is a strong tie, also beyond simple loan words.

There is almost no other place than the steppe that would fit.

epoch1970 said...

"Insights into human history from genetics will be in history books and taught in schools in the future. No doubt about it. In my Geography book it says Slavs came from Northern Asia."

One of the surprising things coming out of paleogenetics is the complete reinstating of the old history theories of migrating peoples. After the war terms as Aryans obviously became tainted. But serious archeologists opinions became tainted as well. The whole idea of migrations became somewhat tainted.

https://westhunt.wordpress.com/2014/11/01/the-inexorable-progress-of-science-archaeology/

Shaikorth said...

"After using all 5 populations (Figure S9.27, page 124), the population with the worst score (higher resnorm) is Scottish, and to a lesser degree Tuscans (when using Yamnaya, in Figure S9.27 C). It is subtle, but noticeable.


Scottish don't take any BedouinB, so I was wondering which population might improve their score. "

Adding Nganasan improved the Scottish score a bit (it actually did that for almost all Europeans) so perhaps the fact that Bedouins did not implies the issue is with something unrelated to Basal Eurasian or African, be that ANE, ENA or whatever. In that case a Near Eastern + ANE mix like Georgians wouldn't help. The fit not being better than it is may lie with something as simple as not using Kets or Selkups instead of Nganasans, who are quite extreme as far as Siberians go, as the fourth population.

Colin Welling said...

I think we hardly know the genetic composition of Western Europe.

No, we pretty much know that western europe changed dramatically in the late neolithic.

We have an iron age sample from brittain which looks like a modern northwest european. But even without that i would know that the introduction of yamnaya genes in northwest europe was abrupt. If it were instead a gradual process then the northwest europeans would not have more yamnaya related genes than the central europeans. So, the yamnaya genes that ended up in northwest europe did not do a whole lot of mixing in central europe; so it was not a gradual process.

Alberto said...

@Shaikorth

"The fit not being better than it is may lie with something as simple as not using Kets or Selkups instead of Nganasans, who are quite extreme as far as Siberians go, as the fourth population."

Yes, it could be, of course. It's just speculating. But why wouldn't it work good for Scottish but good for Lithuanians or Czechs? And Kets and Selkups are (in simplistic terms) EHG + Nganasan, and with those populations the score is worse than when using Yamnaya + Nganasan. So I have the feeling that what is needed is more of the non-EHG Yamnaya, i.e., the Georgian-like population (because that one is provided by Yamnaya, but mixed in a specific proportion with EHG, and adding more Yamnaya adds excess of EHG). Bedouins don't help because they have Sub-saharan and too much basal Eurasian.

It just needs to be tested with something like qpAdm and find the best fit for Scotts (better with an ancient genome if we do get it these days than with modern Georgians).

epoch1970 said...

@Colin

"If it were instead a gradual process then the northwest europeans would not have more yamnaya related genes than the central europeans."

According to this graph the English and French have less than Central Europeans:

https://drive.google.com/file/d/0B9o3EYTdM8lQd19HNEdENTdtZTg/view

Also, there is archeological evidence - for what that's worth - indicating a slower transition to CWC in the most western part. It might indicate some more continuity. There is a hypothesis that suggests a non-IE substrate in Germanic languages.

Shaikorth said...

@Alberto
Yes, Selkups are more EHG than Nganasans but they'd be replacing Nganasans while Yamnaya stays in the fit. Anyone with the dataset and some time on his hands could try to replicate the Haak fits and check that out.

As for qpAdm, Mauri M posted some experiments a while ago
http://terheninenmaa.blogspot.com/2015/03/estimating-ancient-genes-among-present_24.html

Yamnaya + SHG + EHG + EN + East Asian gives Scots a very good fit (chisq 1.475), and the combination seems to work for all NW and NE Euros as for example Lithuanians and Kargopol Russians get about as good fits. The question is if the source populations can be reduced without making the fit any worse.

Alberto said...

@Shaikorth

Thanks for the link. Interesting tests, though a bit inconsistent to read too much into them.

Let's see if these days we really get more clues and samples about this all.

Colin Welling said...

According to this graph the English and French have less than Central Europeans

...? and spanish have even less than the french and the moroccans have less than the spanish... who cares?

Northwest europeans have more yamnaya then central europeans which means the yamnaya source for northwest europe did not homogenize with the central europeans. Hence it was a fast change. Hence r1b come from the steppes in mass.

I suppose you didn't bother to look up the iron age brit because thats the actual proof that northwest europe changed abruptly.

Balaji said...

The statistic D(WHG,EHG;Armenian,Chimp)=0.0034 (z=0.68). This implies that Armenians have about the same amount or slightly more of UHG/WHG than of ANE. They have perhaps 16% ANE, 18% UHG/WHG and 64% BEA.

Krefter said...

There's WHG-ancestry in West Asia no doubt. Some posters have forgotten than ENF is not pure Basal Eurasian.

Mike Thomas said...

The ENF in Europe and anatolia perhaps; but what about Levant.? I highly doubt there was any WHG there

Chad Rohlfsen said...

EEF and that ENF have WHG/UHG. Bedouins are probably 40-50% WHG/UHG, and Stuttgart in the 56-70% range.

Nirjhar007 said...

David, which y-dna will Afanasevo have?:)

Balaji said...

There are two kinds of Bedouins with the following D statistics.

D(WHG,EHG;BedouinB,Chimp)=0.0116 (z=2.28)
D(WHG,EHG,BedouinA,Chimp)=0.0058 (z=1.18)
D(WHG,EHG;Armenian,Chimp)=0.0034 (z=0,68)

BedouinA has more ANE than BedouinB which accounts for the lower D statistic. The ANE in BedouinA can only be at most 16% The UHG/WHG of BedouinA may be about 25%. A rough estimate of the composition of BedouinA is 50% BEA, 25% UHG/WHG, 15% ANE, 10% SSA. BedouinA may be 60% BEA, 30% UHG/WHG, 0% ANE and 10% SSA.

Chad Rohlfsen said...

You need to run them together to find significance, as your Y and Z.

Grey said...

Colin Welling

"First off, we have to admit that the quick change in the autosomal composition of west europeans by the addition of a large chunk of yamnaya-like dna was the result of a mass migration. There is no way around this. It is not feasible that a small group of yamnaya changed the autosomal dna of the west in such dramatic fashion. So this mass migration of r1b people came from a significant population dominated by r1b on the steppe."

I'm sure there was a mass migration into Europe from the steppe but I don't think it follows there had to be a mass migration of R1b.

If R1b was a minority population on the steppe connected to copper mining and working (maybe through originally coming from a region that had a lot of easily accessible copper) which spread along the trade routes then it might have spread more or less everywhere but in low numbers.

However any originally small group of miners that ended up somewhere where the local population density was relatively low at the time, for example Ireland or up a mountain, then maybe those initially low numbers of people expanded to fill the whole niche.

Don't the recent age and star shape of the European R1b clades imply a sudden, dramatic expansion like that?

Kristiina said...

Shaikorth, I went to see Mauri’s experiment and I wondered what EN (in your formulation) and Scandinavian Neolithic HG (in Excel file) mean. Does EN mean Ajvide like hunter gatherers or Gökhem like farmers? I suppose they are farmers as Ajvide like hunter gatherer ancestry is said to have been replaced. The highest EN score is in Lithuanians: 0.375 and 0.381. The list continues as follows:
Estonian 0.332, 0.306,
West Finnish 0.305
East Finnish 0.296
Bell Beaker - LN 0.290
Basque 0.270
Belarussian 0.270
Icelandic 0.275, 0.222
Scottish 0.261, 0.210
Unetice-EBA 0.248
Orcadian 0.242
Norwegian 0.223
Corded-Ware-EN 0.080
It is interesting to compare this ”farmer” ancestry with Yamnaya ancestry:
Lithuanians 0.248, 0.211
Estonian 0.311, 0.178,
West Finnish 0.198
East Finnish 0.162
Bell Beaker - LN 0.216
Basque 0.000
Belarussian 0.299
Icelandic 0.269
Scottish 0.017, 0.289
Orcadian 0.296
Norwegian 0.027
Unetice-EBA 0.369
Corded-Ware-EN 0.361
To sum up, if EN were non-IE speaking farmers, in North Europe, their ancestry is best preserved in Estonians and Finns. If Yamnaya is IE, this IE ancestry is observed unevenly all over Europe ranging from 0.00 to 0.44 (the highest score is in Mordovians!). Yamnaya component is lacking in Basques, French-South 2, Iron-Age Briton, Maltese, Sardinian and Sicilian.

A third component must be added to above two components. Western and southern Europeans are mainly derived from Hungarian Neolithic Farmers, e.g. Sardinians 0.922 and Spanish 0.828.

Grey said...

just a little map to show how close Samara is to the Udmurts for example

https://www.google.co.uk/maps/place/Udmurt+Republic,+Russia/@55.5906803,51.8412362,6z/data=!4m2!3m1!1s0x43e3e2af724d4051:0x102a3a583f19530

and Bashkortostan

https://www.google.co.uk/maps/place/Republic+of+Bashkortostan,+Russia/@54.052373,54.3172899,6z/data=!4m2!3m1!1s0x43d9b405559447fd:0x703c8a5a54345434

i.e. where there's lots of R1b still today

.

http://en.wikipedia.org/wiki/Haplogroup_R1b_%28Y-DNA%29

"Trofimova et al. (2015) found a surprising high frequency of R1b-L23 (Z2105/2103) among the peoples of the Volga-Ural region. 21 out of 58 (36.2%) of Burzyan Bashkirs, 11 out of 52 (21.2%) of Udmurts, 4 out of 50 (8%) of Komi, 4 out of 59 (6.8%) of Mordvins, 2 out of 53 (3.8%) of Besermyan and 1 out of 43 (2.3%) of Chuvash were R1b-L23 (Z2105/2103),[34] the type of R1b found in the recently analyzed Yamna remains of the Samara Oblast and Orenburg Oblast.[24]"

.

Southern Urals have a lot of copper

http://en.wikipedia.org/wiki/Kargaly



Shaikorth said...

Kristiina, the populations were not in the sheet's order in my earlier post, EN is Hungarian Early Neolithic and SHG is Scandinavian Neolithic HG (which apparently is different from Mesolithic HG's according to most recent information, being both more WHG and farmer-like).

Use caution when interpreting those results, for example Basques may lack Yamnaya on that sheet but they have it in other qpAdm tests and they also have EHG which is mediated by Yamnaya. Same goes for Kargopol Russians and Icelandic people (who have fits with Yamnaya and no Yamnaya). Yamnaya can apparently be substituted with added EHG + Farmer in qpAdm. A fit simpler to interpret would be WHG+EN+EHG+E-Asian since that addresses the issues of SHG overlap with WHG and EHG (perhaps EN too) and Yamnaya overlap with EN and EHG.

Colin Welling said...

Wow, grey. Its one thing to appeal to an abrupt founder effect in terms of ydna but a whole other story to try the same with autosomal dna.

Its silly to think that a small group would travel to northwest europe barely being diluted, despite their small numbers, and just outbread the locals like crazy then later on mix with them. However, the fact that the iron age sample was well mixed disproves this very unlikely event. The abrupt autosomal change in NW europe occurred because of a relatively large migration. What leads you to think of this migration as small in the context of yamnaya and large in the context of northwestern europe (along with central europe to avoid dilution), i have no idea. The numbers involved in this migration were big and they pretty much lacked r1a.

Kristiina said...

Shaikorth, thank you for his specification! However, to me it looks like that Hungarian Neolithic Farmer ancestry is the prevailing ancestry in Europe on the basis of that run and it should be originally non-IE. Samara hunter gatherer was very close to Karelian hunter gatherer and that ancestry is not really very frequent in the west so I do not know if it has anything to do with IE languages:
Unetice-EBA -0.00
Bell Beaker - LN -0.00
Orcadian -0.00
Bergamo 0.00
Belarussian 0.000
Lithuanians -0.00, 0.021
Basque 0.027
French 0.00, 0.046
Estonian 0.00, 0.076,
West Finnish 0.082
Corded-Ware-EN 0.18
Icelandic 0.00, 0.155
Scottish 0.00, 0.153
Mordovian 0.00, 0.15
East Finnish 0.114
Norwegian 0.151
Kargopol Russian 0.23, 0.17
Chuvash 0.325

I would assume that the European IE protolanguage was spoken in Unetice and Corded Ware cultures, but those languages do not equal any of the modern language families as they have important substrates from earlier Neolithic cultures.

Grey said...

Colin Welling

I agree you'd need an unusual set of circumstances: mostly males moving to a region of low population density, marrying local females and then being able to expand. You'd need something like copper miners from a farming culture moving to an island that was still mostly HGs.

Alberto said...

@Kristina

I think those tests are interesting, but they are inconsistent too take them literally. However, they do show one important point: Once you add SHG into the equation, the Yamnaya ancestry across Europe (especially the north) gets dramatically reduced when compared to Haak et al's estimations. We'll see if something similar happens especially in Southern Europe when we get Maykop or Central Asian ancient genomes.

Regarding R1b and NW Europe: Scottish show the most positive score among North Europeans in D-Stats of the form: (Loschbour EHG Pop Chimp).

Loschbour EHG Scottish Chimp 0.0202 3.771

Compared to:
Loschbour EHG Corded_Ware_LN Chimp -0.0189 -3.32
Loschbour EHG Yamnaya Chimp -0.0354 -6.545

It looks quite significant. Whatever mass migration replaced the British Islands' population it doesn't look extremely Yamnaya/CW-derived. (Note that CW was about 41% WHG and Scotts are about 45% WHG, so not much difference in the amount of it, the difference must be in the origin of it).

Shaikorth said...

Alberto, he also did tests with SHG but no EHG included, and then Yamnaya estimates were much higher everywhere (comparable to Haak's).

http://terheninenmaa.blogspot.com/2015/03/estimating-ancient-genes-among-present_22.html

I'd say EHG and not SHG is the main cause for reduction of Yamnaya in qpAdm fits, because Yamnaya can be replaced with EHG+EN. How do Scots compare in WHG-CW or WHG-Yamnaya tests, f4 or D?

Mike Thomas said...

Alberto

How do you propose the Haak/Lazaridis model failed to adequately factor in SHG contribution in northern europe ?

Davidski said...

They didn't fail.

Their conclusion that SHG was not an important source of ANE in northern Europe was correct, because the ANE ratio of SHG is too low and WHG too high.

So nothing has changed, but I see that hopes remain high to the contrary.

Alberto said...

At this point I'm willing to bet that Motala-like HGs did play an important role in the spread of ANE across northern Europe (or all of Europe). If they were in Scandinavia, they probably were also in Eastern Europe (at least NE Europe, not sure about Ukraine).

I'm not sure why Haak/Lazaridis model estimated Yamnaya admixture across Europe based on ANE levels (i.e, assuming ALL of the ANE in Europe came from the Samara/Yamnaya types). In their paper there are references to SHGs that point to their contribution to modern Europeans.

Also for a West Asian source of ANE they just used Bedouins, which isn't exactly the best option except to provide Sub-saharan admixture in a few South European populations.

I think their last paper is the best effort published to date, and I also give them credit for the huge improvement in their model compared to their previous one. So I'm confident that their next paper with more samples will continue to make a lot of progress and will improve their current model a lot.

Davidski said...

Okay, quit what you're doing and look at this page.

http://www.ebi.ac.uk/ena/data/view/PRJEB9021

Who wants to get in touch with Sergey or Felix? Because I don't have the bandwidth.

Alberto said...

@David

Motala12 did fit quite well with many population in that 4mix program based on your K8. So it's not a matter of numbers. Numbers match.

Formal stats also agree with those numbers.

What is your best explanation for the D-stats above involving CW and Scottish?

Mike Thomas said...

David I didn't say they failed; I asked albertos opinion as to why he disagrees

Helgenes50 said...

Who wants to get in touch with Sergey or Felix? Because I don't have the bandwidth.

That means that the genomes are available ?

Alberto said...

"Okay, quit what you're doing and look at this page."

Wow, so the genomes are available? Then surely that's more important than theoretical discussions. Let's try to see those results and then we can continue debating.

Davidski said...

Drop everything and look here!!!

http://www.ebi.ac.uk/ena/data/view/PRJEB9021

Mike Thomas said...

Ok I'll email Felix

Krefter said...

Andronovo and Sintashta were part European? So, it went Steppe-Europe-Central Asia. I'll wait for the paper. Maybe from East Europe, which could explain R1a-Z93 being a brother to R1a-Z283.

"e show that around 3 ka BC, Central and Northern Europe and Central Asia receive genetic input through people related to the Yamnaya Culture from the Pontic-Caspian Steppe, resulting in the formation of the Corded Ware Culture in Europe and the Afanasievo Culture in Central Asia. A thousand years later, genetic input from North-Central Europe into Central Asia gives rise to the Sintashta and Andronovo Cultures."

Krefter said...

Are all 101 available?

Mike Thomas said...

Totally jumping the gun here
But if true, it'll explain the replacement of large chunks R1b dominant Eastern steppe/ Central Asia by R1a (apparently Euro derived)

Davidski said...

Yeah, they're all there, but we won't know what they are exactly until the paper comes out.

Alberto said...

I hope the paper will be published tomorrow... Exciting times.

@Shaikorth

What I rally meant by saying that SHG drop the Yamnaya ancestry was that they reduce either Yamnaya or EHG (I count EHG as Yamnaya ancestry).

The results with SHG and without EHG do show greatly reduced Yamnaya estimates. For example, in Haaks paper Czechs and Scots are some 45% Yamnaya, but here they are below 30%. Norwegians go from 50%+ to 30%. French from some 30% to 19%, etc...

"How do Scots compare in WHG-CW or WHG-Yamnaya tests, f4 or D?"

I haven't seen those stats, so no idea. But the D-stats I posted above are significant enough to think that the lower numbers of Yamnaya admixture are probably much closer to the reality.

rozenfag said...

>During the late BA and Iron Age, the European-derived populations in Asia are gradually replaced by multi-ethnic cultures, of which some relate to contemporary Asian groups, while others share recent ancestry with Native Americans .

What?

Srkz said...

>>Who wants to get in touch with Sergey or Felix? Because I don't have the bandwidth.

OK, i'm in

Chad Rohlfsen said...

It looks like andronovo has Corded like admixture. Looks like this settles the gene flow direction.

Alexandros said...

Thanks for the abstract! Looking forward for the paper and your updated calculators!!

Alberto said...

@Mike

"Totally jumping the gun here
But if true, it'll explain the replacement of large chunks R1b dominant Eastern steppe/ Central Asia by R1a (apparently Euro derived)"

Yes, could be. But also interesting to note that they say North-Central Europe as the source of this migration and not Eastern Europe. This could explain why WHG/SHG ancestry dominates Eastern Europe today (without needing to wait for the middle ages Slavic expansion or Swedish/Vikings intrusions). It could be that the Eastern European intrusion into North-Central Europe rebounded rapidly and ended up being a North-Central European intrusion into Eastern Europe (though carrying R1a by then).

But let's wait and see what these genomes really show before jumping to conclusions.

Krefter said...

It's safe to assume Afanasevo was mostly Yamnaya-like and Sintashta/Andronovo were mostly Yamnaya with extra WHG and or MN-type ancestry. We know Bronze age Armenians(from at least 3500YBP) were just like modern ones. So, the main unknowns are what were Maikop and Ancient Bulgarians like, if they sampled them.

Davidski said...

Thanks Sergey.

Yep, I'll start working on new stuff once I have all the genomes.

Shaikorth said...

Alberto, Yamnaya is a mixture that contains EHG ancestry, but it's not really the same due to high Near Eastern ancestry portion.

There's also a qpAdm test in there with a Hungarian WHG+EN+Yamnaya+E-Asian fit and that gives almost the same Yamnaya numbers as Haak paper's S9.26-27 (Scots get 46,2% for example). The Yamnaya portion goes down when SHG is added into the fit and further so when EHG is introduced. Farmer goes up as well in North Europeans as that happens, and I think it's because Yamnaya breaks down into EHG or SHG (which is chiefly an EHG-WHG mix) and EN farmer in models.

SHG has so much WHG that it alone can't fully break Yamnaya down, but EHG in the fit is enough to allow for 0% Yamnaya fits for modern euros (be they Scots or Kargopol Russians). The difficulty of distinguishing EHG/SHG survival from Yamnaya with qpAdm is thus an issue. Haak et al didn't try a fit with EHG and Yamnaya simultaneously included, but it's quite possible that their method would face similar problems.

Alberto said...

Shaikorth,

The reason why I consider EHG ancestry as Yamnaya ancestry in Europe is not because they are the same, obviously, but because there's no clue that pure EHGs migrated to Europe. They did it mixed with a Caucasus-like population. So using EHG as a source is "substituting" Yamnaya ancestry. If Scots get 0% Yamnaya but 15% EHG, for me it means more or less the same as if they have ~30% Yamnaya.

So the important point is to see Yamnaya ancestry (and/or EHG) together with SHG, because SHGs had a different origin and imply a different model.

With only WHG the results are similar to Haak's, since it's they used the same sources.

Nirjhar007 said...

Yaawn i will wait for the Y-DNA and mtdnas.....

Nirjhar007 said...

//Yamnaya-like and Sintashta/Andronovo were mostly Yamnaya with extra WHG and or MN-type ancestry//
I would like to see the Y-DNAs actually before concluding anything....

Shaikorth said...

EHG's also appeared in pure form in NE Europe, and in mixed form not just with Yamnaya, but also SHG in Scandinavia at least. SHG still overlaps with Yamnaya because of its high EHG (the rest of it overlaps with WHG/EN), just not as much as a full EHG would. So it's not really certain that a fit with Yamnaya + SHG + WHG + EN would be any more informative than Yamnaya + WHG + EHG + EN for instance.

Nirjhar007 said...

Although we still don't know about the hg's but Afanasevo from Yamnaya was
expected, there is no problem, is a sort of proto-Scythian people.
But the thing is coining Sintasta and Andrnovo with NC Europe....

Davidski said...

You should be sitting down when reading the paper.

My bet is that the Sintashta remains will carry the ancestral R1a-Z93 lineages to all Asian R1a-Z93.

Nirjhar007 said...

It should have Z-93 there is also no problem will Z-93 be also there from Europe? that's the question IMO Z-93 is Asian in Origin...

Alexandros said...

I really wonder whether (and where) we will see any J2 patrilineages among the West Asian samples. Pitty we will apparently have only 1 Armenian male..

Nirjhar007 said...

I actually suspect some us here already know the results:D....

Aram Palyan said...

Hyperboreans? :)

Karl_K said...

"I actually suspect some us here already know the results:D...."

Well... the sequences have been available for several hours. It's just sorting out what is what, and "bandwidth" issues.

Nirjhar007 said...

Do anyone have any data yet on hg's?? Dienekes say ''genome bloggers can get to it thanks to the early data release''...

Nirjhar007 said...

Aram, yeah :D anyway since authors conclude NC Europeans ''gave rise'' to the Sintasta and Andronovo then it either related to R1a or Mtdna, i just hope R1b don't come up there as it will make things weird:P....

Davidski said...

Sample accession > Attributes

It tells you what samples they are.

Nirjhar007 said...

Thx:).

Krefter said...

I'm making a spreadsheet with the year, country, sex, site, and culture of every sample.

https://docs.google.com/spreadsheets...it?usp=sharing

Nirjhar007 said...

Krefter, when we will know about the hg's?:)

Alberto said...

@Shaikorth

Yes, you are right, Yamnaya is not the 100% source in the 100% accurate proportion of EHG across 100% of Europe. I'm just being practical.

When we have a Caucasus-like genome (let's hope in this batch there is Maykop DNA and it's the one we're looking for), we might have all the pieces. Then we can use all the population sources and see what happens. I think that such sample will reduce the Yamnaya/EHG in Bulgarians, Greeks or Tuscans. We'll see what happens with Western Europeans.

truth said...

So, that Russian study from one year ago, about R1b in Afansievo samples might be true actually.

Srkz said...

>>I'm making a spreadsheet with the year, country, sex, site, and culture of every sample.

Me too

https://www.dropbox.com/s/0gtissrsp3nwjqq/Samples.xlsx?dl=0

Nirjhar007 said...

Afanasevo have R1b? well what about Sintashta?.

Chad Rohlfsen said...

I'd bet on R1b with R1a coming in as Corded like in Andronovo. Steppes to SC Asia, not the other way around.

Nirjhar007 said...

its premature to tell that we have to wait a bit chad.....

Nirjhar007 said...

To be a more specific we need aDNA from India, BMAC, Jeitun, South of Caspian-Kurdistan etc etc and when we get them we will have enough, at the moment its mostly the areas concerned with the Steppe theory and its a bit biased, that's why some aDNA from example India will do real nice now:)......

Karl_K said...

Maybe Nirjhar007,

but the abstract seems pretty confident. "Genetic input from North-Central Europe into Central Asia gives rise to the Sintashta and Andronovo Cultures."

Nirjhar007 said...

we need many more samples from different parts of eurasia covering different periods, yes they are very ''confident'' and we will love to see why......

Karl_K said...

I agree, ancient DNA from India will be incredibly interesting. There was just so much going on. But as for the Indo-European language situation, I don't think ancient Indian DNA will be necessary to sort most of it out.

Nirjhar007 said...

IE Language situation? well there is no IE Language situation, the IE-Uralic theory is an exaggerated one.......

Karl_K said...

I don't follow. The situation to which I was referring was the branching of related IE languages from a single ancestral language. The basal branches must have split outside of India. I don't think there is really another option at this point.

Krefter said...

Nirijhar,

I agree the sampling chose is a bit bias. I mean the only samples they took from Italy comes from the first culture in Italy believed to be Indo European. Every sample here, except maybe later ones from Russia(East Asian-influenced) follow the Kurgan hypothesis.

Nirjhar007 said...

Karl,
I also don't follow Out Of India for PIE because it doesn't work well for Europe just as the Steppe theory is a failure for Asia!:) my placement for PIE is in South Caspian area and i think the Pinhasi study will have some samples around there so i can't wait...

Simon_W said...

@ Epoch1970

As for the proto-Uralic ties of PIE, I agree, they are important. I might at best speculate about old contacts of PPIE in central Asia. We don't know what languages were spoken in central Asia in PPIE times, and north of central Asia, there's already the Urals.

Alberto said...

For what I've seen, the steppe is going to be well covered with all these samples. I'd like to see the oldest ones from Afanasievo (3330-2930 BC). I'm guessing they'll have a lot of ANE.

The north (mostly NE) of Europe will be better covered too. Denmark, Poland, Czech Republic, one from Estonia... It should help to understand better the CWC, Unetice, BB cultures (though we already have samples of all of these).

The Armenian samples: The oldest is from c. 1800 BC. Male, it seems.

A couple of LN/EBA from Northern Italy will be interesting too. And then more samples from Hungary.

Did someone find any Bulgarian or Maykop samples? I didn't.

Overall it will be very interesting, but I'm afraid it won't answer all the questions just yet.

Nirjhar007 said...

Krefter,
It is actually reasonable because their priority is to follow the map of the most popular theory ( why would they put money one other theories?) but it doesn't mean its also the most reliable one and it don't need to be.
Simon,
PIE and Uralic ties if we simply put some observations are not very strong and mostly one way OTOH if one proposes ties with Semetic (as Bernard Seargent has shown or proposed) or even Sumerian they look far more convincing and deeper...

rozenfag said...

No, it seems there are no Bulgarian or Maykop samples. From the other hand there are couple of Montenegrin ones.

Nirjhar007 said...

Alberto,
The Afanasevo one will be very interesting i think it may well show R1a-M417 or R1b.

Krefter said...

I'm done with my spreadsheet.

https://docs.google.com/spreadsheets/d/1HuNPykGuq2PbHkUOL5dCiwrveIy-OGO2qOklwfsayW8/edit#gid=1630708280

Motable samples.

No Maikop. A few male Middle Bronze age from Armenia, so there'll probably be Y DNA. Some samples from supposed Early IEs of Italy(will probably get Y DNA). A bunch from Bronze and Iron age Russia. A bunch from Corded Ware and Unetice. A bunch from Bronze age Hungary and Scandinavia. A few from Bronze age Montenegro.

Krefter said...

"The Afanasevo one will be very interesting i think it may well show R1a-M417 or R1b."

All Afanasevo are female.

Nirjhar007 said...

Goodness:) then they concluded on the basis of Mtdna and autosome structure.

Alberto said...

Thanks Krefter, very neat to have all the samples there easily accessible. That was fast too.

Simon_W said...

Nirjhar, I wouldn't call this weak evidence:
http://en.wikipedia.org/wiki/Indo-Uralic_languages
Importantly, it's also about the grammar not just the vocabulary. PIE ties with Semitic and Sumerian are afaik just lexical.

Nirjhar007 said...

Simon,
PIE is more a language with a semetic substratum and a good looking common vocabulary with Sumerian and grammar also ( Read Bernard Seargents Les Indo-Européens. Histoire, langues, mythes).

Colin Welling said...

I agree you'd need an unusual set of circumstances: mostly males moving to a region of low population density, marrying local females and then being able to expand. You'd need something like copper miners from a farming culture moving to an island that was still mostly HGs.


Its a silly argument. How can you call the migration minor the yamnaya, but large in the context of both western europe? The migration that dramatically changed the composition of western europeans and central european bell beakers is not "small" in the context of the yamnaya nor is it small in the statistical sense of lacking r1a representing a departure from an area lacking r1a.

Colin Welling said...

@mike Totally jumping the gun here
But if true, it'll explain the replacement of large chunks R1b dominant Eastern steppe/ Central Asia by R1a (apparently Euro derived)


Great news for me :) Ive long theorized that the r1b was derived from yamnaya and that r1a belongs to a set of cultures which derive from the region northwest of the steppe.

Grey said...

Colin Welling

Well I expect we'll see soon enough if "boomerang" is the right word for what happened to R1b in the West.

«Oldest ‹Older   1 – 200 of 218   Newer› Newest»