search this blog

Monday, December 15, 2014

ANE is the primary cause of west to east genetic differentiation within West Eurasia


Here's a Principal Component Analysis (PCA) and an accompanying biplot based on output from an improved version of my ANE K7 ancestry test. Let's call it the West Eurasia K8. This one gives more accurate estimates of Western European Hunter-Gatherer (WHG) and Near Eastern admixture proportions, thanks to the use of new ancient samples.
When rotated accordingly (like here), the results are basically indistinguishable from those I get with genotype data (for instance, see here and here), which suggests that they're correct and based on ancestry proportions that are close to the truth. The Past3 data sheet used to create the PCA is available here. You can view a spreadsheet of the results with extra samples here.

Clearly, ANE is the main agent causing the west to east differentiation in dimension 2. Note that even a small rise in ANE, say, 4-5%, creates significant distance between samples on the PCA plot.

East and South Eurasian admixture has a similar effect, but must be more considerable to make an impact on a West Eurasian-specific PCA like this (and it does with the obvious Volga-Ural outliers, who come from Chuvashia and Tatarstan).

On the other hand, Near Eastern admixture without ANE creates almost the opposite effect. Note, for instance, that Neolithic genomes Stuttgart and NE1 show much higher levels of Near Eastern ancestry than most Europeans, and yet they're amongst the most western samples on the plot.

This suggests that the Near East, and in particular the Caucasus, experienced a significant rush of ANE admixture after early Neolithic farmers left the region for Europe. Alternatively, Caucasus populations may have carried even higher levels of ANE than they do today, before newcomers from the Near East mixed with them. But either way, a lot of ANE arrived in the Near East at some point.

It also suggests that, overall, the populations that moved west across northern Europe after the Neolithic, and shifted northern European genetic structure to the east, did not carry high ratios of Near Eastern ancestry. Instead, they harbored high ratios of ANE and WHG. What these ratios were exactly I haven't a clue, but ancient DNA should tell us that soon.

Below are the ancestry proportions for the five ancient genomes in this analysis, in chronological order. It's interesting to note (yet again) the rising and falling Near Eastern admixture, from the Mesolithic to Neolithic and then from the Neolithic to Bronze Age, respectively, as well as the steady rise of ANE from the Bronze Age to the Iron Age.

Loschbour (Mesolithic)

ANE 0
South_Eurasian 0
Near_Eastern 0
East_Eurasian 0
WHG 99.5
Oceanian 0.5
Pygmy 0
Sub-Saharan 0

Stuttgart (Neolithic)

ANE 0
South_Eurasian 0
Near_Eastern 72.19
East_Eurasian 0
WHG 27.8
Oceanian 0
Pygmy 0
Sub-Saharan 0

NE1 (Neolithic)

ANE 0
South_Eurasian 0
Near_Eastern 69.82
East_Eurasian 0
WHG 30.17
Oceanian 0
Pygmy 0
Sub-Saharan 0

BR2 (Bronze Age)

ANE 9.62
South_Eurasian 0.08
Near_Eastern 43.96
East_Eurasian 0
WHG 45.44
Oceanian 0.48
Pygmy 0.23
Sub-Saharan 0.19

Hinxton4 (Iron Age)

ANE 15.08
South_Eurasian 0.06
Near_Eastern 35.44
East_Eurasian 0.46
WHG 48.5
Oceanian 0
Pygmy 0
Sub-Saharan 0.46

See also...

The fateful triangle

Bell Beaker, Corded Ware, EHG and Yamnaya genomes in the fateful triangle

282 comments:

«Oldest   ‹Older   201 – 282 of 282
Pierre de Laclos said...

Just to be clear, I have no problem with men going off and doing their own thing.

However, this is obviously not a hobby for you guys. Some of this data will be published.

Much of this research, the sequencing of the ancient genomes, for instance, is paid for with public money.

So, I'd like to know why my tax dollars are funding research that is not inclusive of women.

Time to write a letter.

Fanty said...

"
East Europe is often described as part of North Europe.
"

Yeah. Its the same concept I use.

Northern Europe = Northwest and Northeast

Southern Europe = Soutwest and Southeast

Western Europe = Northwest and Southwest

Eastern Europe = Northeast and Southeast

I generally avoid the term "central Europe", as its hard to say where that is. I think there are at least 2 points in Europe that claim to be the center. Both base this on 2 different concepts of calculating the center. I think one was in Czechia.

Krefter said...

"Much of the data that you are sharing here will ultimately be published. Yet, there is clearly a pattern of excluding women on this information sharing forum."

It just happens most of the posters are male, women were never intentionally excluded. Cultural and instinctual trends, like women are better students in grade school and high school, isn't sexism.

What Davidski said is that copper age men saw women as a commodity, he never said he saw women in that way.

Davidski said...

rk,

There's no way CWC will cluster south of the Poles and Scandinavians.

Pierre de Laclos said...

@Krefter

"Cultural and instinctual trends, like women are better students in grade school and high school, isn't sexism."

Sure, it's all due to cultural and instinctual trends.

For instance:

1. telling women to "suck it up" when they disagree with you intellectual position,

2. ignoring their contributions, and then publishing them as your own,

3. accusing them of racism in writing online (smearing) and then ripping off their ideas and publishing them,

4. blocking their papers from publication (as the reviewer),

5. ignoring their emails (not even responding with a no) when they apply for positions in your research groups,

6. sending women secretly online sexual invitations when they show interest in your field (rather than take their academic potential and interests seriously),

7. secretly colluding online with other men to undermine the ideas and reputation of women colleagues,

8. talking over them,

9. telling them that they are distracting from your discussion, rather than taking their ideas seriously. (and then ripping off their ideas).

And these are only things that have happened personally to me or my immediate friends. I'm sure it's just the tip of the iceberg.

Yep, the absence of women on these forums must only be due to the fact that women aren't interested in population genetics, prehistory, archaeology or anthropology.

Krefter said...

Pierre, this is getting off subject so it'll be the last thing I say about it.

Your complaints can be explained as specific to women and simply human. I agree those things happen, but I don't see what it has to do with this blog. If you asked men, they could also give a long list of how women overlook things too. I'm not saying one is worse than to other or equal.

It's hard to stop persecution that isn't written in law. But all sorts of persecution will always exist, like bullying(anti-bullying campaigns, are ridiculous), and we can't expect to solve all of them. Miss treating each other is human nature, and a necessary evil.

"6. sending women secretly online sexual invitations when they show interest in your field (rather than take their academic potential and interests seriously),"

Honestly this is the only gender-specific one on your list I could find. All the others could simply be miss understandings.

There are very few men who will try to dis credit a woman's ideas because she's a woman. That doesn't make any sense. Gender competition is for little kids.

Chad Rohlfsen said...

Just ignore them. That simple.

Seinundzeit said...

Pierre de Laclos,

How is any of this related to the discussion occurring here? Anyway, your simply pulling things out of thin air.

David,

I've always wondered if there was some way to construct synthetic "ASI" samples from the non-West Eurasian portion of Paniya genomes, like you've done with the ANE cluster? The Paniya seem to be around 45% West Eurasian, and 55% of what Reich et al. labelled "ASI". Considering that they are close to being "simple" hybrids of "ANI" (ANE+Near Eastern ancestry, but with a lot of ANE in that mix) and "ASI" (some sort of ancestry basal to "Crown Eurasians", to steal RK's terminology, or perhaps just some sort of ENA ancestry. Hard to decide between the two options, although I think the former option is more likely), would this be feasible?

Your ANE cluster is very robust, so I'd assume you could easily "mask" those segments among the Paniya (for what it's worth, they seem to be around 25%-30% ANE).Take out their 45% ANE+Near Eastern, and you have perfect ASI samples. The Onge are distant relatives of "ASI", but synthetic samples created out of Paniya genomes would basically constitute "ASI", at least as understood by Reich and Moorjani et al.

Seinundzeit said...

Also, to clarify, when I said "pulling things out of thin air", I didn't mean that women aren't discriminated against in intellectual contexts. These are very serious issues, and they need to be dealt with on a sustained and serious basis.

Instead, what I meant is that "Pierre" is just bringing random unrelated stuff up, just for the sake of side tracking the discussion, and just for the sake of making this blog look bad (I have no clue how discrimination against women relates to anything that has occurred in this conversation we all are having).

Davidski said...

Sein,

Isolating ANE from South Amerindian genomes is fairly easy, because they appear to be just a two-way mix of ANE and ENA.

South Indians, even those with seemingly high levels of ASI, are a mixture of ANE, Basal Eurasian, probably something WHG-like, and a component, or components, native to South Asia.

I wouldn't know what I'm looking for exactly. It is frustrating, but I think it's a hopeless cause until we see a pre-Neolithic genome from India, or at least South Central Asia.

Grey said...

@Fanty

"I generally avoid the term "central Europe", as its hard to say where that is."

I'd agree with that most of the time but I think there can be a significant difference at certain points.

For example there is a quite clear northern limit to the spread of farmers. If there was an HG rebound then the northern 1/3 is where it's likely to have come from.

So in the farmer era it can be defined as the northern limit.

I'd also say there were two distinct overland channels into Europe: one north of the Carpathians and one along the Danube, which I think will be significant in terms of copper age movements.

So in that era it can be defined as the central channel as opposed to the northern one.

Balaji said...

Davidski,

You found the following D statistics.

D(Primate_Chimp Dai : Basque_French Spanish) -0.0029 -2.822

This indicates that the Dai are closer to the Basque than to the Spanish which is reasonable since the Spanish have more EEF than the Basque. But then how do we explain the following f3 statistics?

f3(Spanish;Dai,French_Basque) -0.000975007 0.000124464 -7.83365

The Spanish are not the only population with higher EEF than the Basque who have negative f3 stats involving Basque and Dai.

f3(Greek;Dai,French_Basque) -0.000365561 0.000151607 -2.41124
f3(East_Sicilian;Dai,French_Basque) -0.000569748 0.000184245 -3.09234
f3(Tuscan;Dai,French_Basque) -0.000432676 0.000151683 -2.8525
f3(Portuguese;Dai,French_Basque) -0.000915428 0.000242182 -3.77992
f3(Bulgarian;Dai,French_Basque -0.00118302) 0.000144746 -8.17306
f3(Romanian;Dai,French_Basque) -0.00125626 0.000141042 -8.90695
f3(North_Italian;Dai,French_Basque) -0.0002888 0.000159868 -1.80649
f3(West_Sicilian;Dai,French_Basque) -0.000558259 0.000196153 -2.84604

One explanation that comes to my mind is that the population that brought in ANE to Europe also brought along some ENA.

Davidski said...

Those results make sense, because only f-stats of less than -0.001 are significant.

ryukendo kendow said...

@ Davidski
The point that I'm making is that Poles and Balts likely have their WHG driven to higher levels than CWC by later events.

Shaikorth said...

Not necessarily later events, if the higher WHG was present in N-C Europe before CW arrived. And that's assuming CW actually had less WHG than Poles.

Balaji said...

The significance or otherwise of the f stats should be a function of their z value. Values of 3 or greater occur among the following.

f3(Spanish;Dai,French_Basque) -0.000975007 0.000124464 -7.83365
f3(East_Sicilian;Dai,French_Basque) -0.000569748 0.000184245 -3.09234
f3(Portuguese;Dai,French_Basque) -0.000915428 0.000242182 -3.77992
f3(Bulgarian;Dai,French_Basque -0.00118302) 0.000144746 -8.17306
f3(Romanian;Dai,French_Basque) -0.00125626 0.000141042 -8.90695

The Bulgarians, Romanisns, Portuguese and Spanish also all have f3 of around -0.001.

Davidski said...

rk,

Balts and northern Slavs derive all of their genetic structure from the former CWC and Yamnaya horizons. It's not a coincidence that R1a reaches very high frequencies in this part of Europe.

Chad Rohlfsen said...

I think that I understand what he's trying to say. MNE groups are well SW of the LNE/EBA groups. Those groups are well off of the line to Yamnaya, so there has to be an extra mixture with groups with higher WHG and possibly some ANE, between MNE and EBA. Possibly movements in by Narva or another group that was between Yamnaya and Central Europe.

Davidski said...

Balto-Slavs expanded from within the core zone of the CWC horizon. This should be obvious to anyone.

There were still pagans who built kurgans and sacrificed horses in what is now northeastern Poland and Lithuania during the Middle Ages, until they were done away with by crusaders from western Europe, who were invited over for this purpose by the Masovian duke.

How much WHG, EHG and ANE do you think these horse people had? Was it more or less than what Central Europeans have?

Chad Rohlfsen said...

Those Middle Age folks probably had the same as people in the region do today.

Corded may be mixed with those hunter/farmer mixed folks, on the Eastern edge of Central Europe, or they expanded from a Western region of Yamnaya, that was more like modern Central Euros. I think that Beaker vs Corded will be similar to Irish vs Swedish.

Davidski said...

Why would Swedes be more similar to Corded Ware than Northeast Europeans, who still live in the areas from which Corded Ware expanded to the west, carry a lot of R1a, and also show more Corded Ware-like mtDNA than Scandinavians?

What is the logic behind this?

Chad Rohlfsen said...

The logic being, that is where they plotted prior to mixing more with locals, then pulling South, and then more South and East with Late Bronze and Iron Age migrations, like IR1.

Krefter said...

Davidski, the Yamna Laz described has to little WHG and to much Near eastern to fit in east Europe today. CWC on a PCA is only slightly more northwest than Yamna.

If your guess on Yamna's placement is accurate, Laz's description of Yamna is accurate, and that PCA shows accurate distances, then there's no way CWC fits in NE Europe.

NE Europeans have a to low near eastern ancestry to be just like Yamna or CWC. CWC would have had to mix with HGs if they were like NE Euros.



Davidski said...

Post-Corded Ware population movements had less of an impact on Northeastern Europe than on other parts of Europe. This is why today Northeastern Europeans show such high levels of hunter-gatherer ancestry.

And the idea that Balto-Slavs originated from a population that came from the middle of the Corded Ware horizon, but wasn't directly related to Corded Ware people, is just kooky, because of the uniparental marker evidence of very strong ties between Balto-Slavs and Corded Ware people.

In Brandt et al. the Corded Ware mtDNA shows the lowest fst with the Ukrainian sample.

Davidski said...

Krefter,

What makes you think that the place on the PCA where CWC is sitting is not where the Belorussians, Ukrainians and Lithuanians are sitting?

How did you work out that this is not possible? It appears to be very possible as far as I can tell.

Krefter said...

"Post-Corded Ware population movements had less of an impact on Northeastern Europe than on other parts of Europe. This is why today Northeastern Europeans show such high levels of hunter-gatherer ancestry."

That could be partly the case. Spanish for example could have more near eastern ancestry than Basque because of recent near eastern admixture.

I don't think that's the whole story though. Yamna and BR1+2 were less HG than NE Europeans.

"What makes you think that the place on the PCA where CWC is sitting is not where the Belorussians, Ukrainians and Lithuanians are sitting?"

CWC might fit around Ukraine, but they look to southern to be by Lithuanians, Poles, and Belorussians. I would expect a bigger northern jump from Yamna, if they were just like modern NE Europeans.

NW Euros, central Euros, and Basque look like they have around the same WHG as copper age central Euros, but there's a big WHG rise with Ne Euros and Scandinavians.

I don't know though I'm just making guesses based on PCAs other people create.

Chad Rohlfsen said...

I'm not denying the links, but going by Reich's words. He said that Corded plotted with North Central Euros(which includes Scandinavians) not Eastern Europe, or North East Europeans.

I wouldn't be surprised to see them fall along Northern Germany, Northern Poland, and Sweden.

Chad Rohlfsen said...

North Central Europe is generally Germany, Poland, Denmark, Norway, and Sweden.

Chad Rohlfsen said...

Krefter,
CO1 doesn't have a lot of WHG. She is between Sardinians and Gok2. Gok2 is just under Basque levels, and a good 10-15% under NW and NC Euros.

Davidski said...

North-Central Europe on Lazaridis' PCA is where Czechs, Lithuanians and Belorussians cluster (right between the Northwest Europeans and North Russians). It looks like that's where the CWC samples are too, and the Yamnaya genomes are just southeast of there.

Also, keep in mind that the Corded Ware genomes have very high levels of Eastern European hunter-gatherer ancestry (around 36% low bound). Apparently, this type of ancestry peaks among Czechs, Lithuanians and Belorussians.

I really can't see any strong links between the Corded Ware data I've seen to date and Swedes. The only reason Swedes might end up close to the Corded Ware samples is because they usually cluster close to Lithuanians and Belorussians on West Eurasian PCA.

Nirjhar007 said...

Publication date of Reich et al. Someone please!!!!

Krefter said...

"Publication date of Reich et al. Someone please!!!!"

Reich said he'd have the paper ready to be sent to be reviewed by now.

Now we just have to wait for whoever the people were that has sent the paper to, to review and publish it. I think that's how it works.

Anyways I think we should expect it online in the next few months.

Nirjhar007 said...

Thanks Krefter! but i think we can yield some data from Emails then?
Good Day.

Krefter said...

"CO1 doesn't have a lot of WHG. She is between Sardinians and Gok2. Gok2 is just under Basque levels, and a good 10-15% under NW and NC Euros."

I never mentioned C01.

Basque have around the same WHG as NW Euros. In Laz they scored 33% WHG, and if you take the WHG out of their EEF, it's very close to English. Gok2 also has around the same WHG as NW Europeans.

People from the Balkans also don't have much less wHG than NW Euros. Croatians scored 29% WHG in Laz. If you look at Eurogenes PCAs Bosnians, Serbians, and Croatians ofnly cluster fairly close to Utah_USA, Germans, and even British.

WHG is very high(compared to Stuttgart and near easterns) in all of Europe except Italy and southeast Europe(Albany, Greece, etc.). It's next lowest in Spain and Portugal.

ANE makes NW Euros overall more related to WHGs like La Brana-1, than Basque.

Krefter said...

"Thanks Krefter! but i think we can yield some data from Emails then?
Good Day."

I tried last month and got no response. Someone else should try. I think they're fed up with inpatient people emailing them though.

I can't think of a good reason why Reich, Patterson, and Laz are so distant with people who ask them questions. I think they have a false sense of superiority. They can simply tell us what their guess is on when their paper will be online.

ryukendo kendow said...

@ Davidski

Balts and slavs might have originated geographically from the Corded Ware region, but Reich already said that Corded Ware genomes do not cluster with them today but with North-central Europeans, which means there must be some kind of genetic discontinuity.

Balts especially are simply far too autochthonous in their autosome to be the result or the source of any foreign intrusion into Europe. If there's one ADMIXTURE component that is extremely old and doesn't track any latter-day movements, I think it'd be Baltic.

I think Balts probably underwent a process with the same mechanics as the one that R1b-ized the Basques, leaving much of their autosome unchanged. We should remember that Yamnaya is modeled as 50% Armenian + 50% EHG, and while EHG might peak among finns/estonians for example that doesn't really mean much; we have to look for the Armenian affinity as well if it is IE influence we're looking for. Razib already said that 'Yamnaya had near east and Caucasus' from the talk at ASHG, and if CW inherited this from the large fraction of Yamnaya it had and CW was the IE-izing culture par excellence in Europe, then it is extremely difficult to reconcile a hypothesis that CW is most similar with Lithuanians with the fact that this Caucasus affinity is typically lowest in Lithuanians and Estonians in all Europeans.

ryukendo kendow said...

Actually, the alternative and more likely hypothesis is that the PIE protopop on the paternal side was overwhelmingly dominated by EHG ancestors represented by R1a, with the 'Armenian' like input represented by small slices of G2 and J2. This might mean that certain HG groups in Europe, probably more towards the east, were dominated by R1a even before IE-ization. This would also explain why R1a in all other areas is a subset of R1a in E.Europe, but the autosome of the Yamnaya is rather dissimilar to E.Eruopeans--the R1a EHG pool contributed to both E.Euros and the PIE protopop.

Krefter said...

Ryu, no one ever said CWC clustered with North-central euros opposed to east Euros. When people say east Europe they usually mean the northern half. Laz mentioned Yamna ancestry is highest in Lithuanians. How much of that similar ancestry is Yamna or from similar ancient people is unknown.

"but the autosome of the Yamnaya is rather dissimilar to E.Eruopeans--the R1a EHG pool contributed to both E.Euros and the PIE protopop."

I don't understand because Yamna brought R1a to east Europeans.

EHG didn't necessarily have mostly R1a. Who knows what type of founder effect could have occured after groups of near easterns and EHG mixed. It would make sense that had R, Q, I, or C.


Chad Rohlfsen said...

David,
Sorry, I was not aware that Reich considers Belarus as North Central Euro. Are you able to plot BR1, so that we can kind of get a gauge for what her breakdown might be if you are unable to pluck out any numbers? Thanks!

Krefter,
My apologies, I instantly thought of CO1 as a Central Euro Copper Age representative.

Chad Rohlfsen said...

Looking at the k8 PCA, I am wondering if EBA Bulgarians were very similar to modern French people. Or, between them and Bosnians.

Chad Rohlfsen said...

David,
What is the French Basque score, with the new test?

Davidski said...

rk,

Can you actually show me a quote where someone explicitly says that the Corded Ware genomes don't cluster with Balto-Slavs?

Davidski said...

Chad,

I posted a spreadsheet above with all the results.

Davidski said...

rk,

You're reading too much into some of the tweets from the conference, and ignoring that terms like Northern and North-Central European are also used to refer to Lithuanians, Poles and Czechs.

There's nothing that I've seen that prevents the CWC from showing the highest affinity to the Balto-Slavs in the Human Origins dataset, and in fact that's my impression from all the information I've seen to date. Lithuanians and Czechs, for instance, are supposed to have the highest ratios of the non-local eastern ancestry carried by the CWC.

You might find this useful.

"The genetic variation you see in eastern and central Europe today is representative of the Yamna people. They know because they have ancient samples from those regions."

http://www.unz.com/gnxp/r1a1a-what-is-best-in-life/

Chad Rohlfsen said...

Okay, thanks.

If you made your Near East cluster 2-3% ANE, what will happen to everyones' results?

Davidski said...

Then in theory everyone's estimates will be slightly lower. But I don't think this is an issue, because Near Eastern groups aren't showing lower than expected ANE estimates.

The groups showing lower than expected ANE estimates are those that can be modeled as 0% ANE with formal stats.

ryukendo kendow said...

@ Davidski
Now we are reduced to reading different parts of Razib's corpus. You a line there, I a line here.

Ok, so be it, lets assume EHG peaks in Lithuanians. The issue I am trying to raise here is that the difference between East Europeans today and Yamnaya is that East Europeans probably have more WHG. I dunno, as opposed to that sentence you raised, the facts Razib dropped in the tweets were highly specific. Just because East Europeans are distinguished from other Europeans by high WHG, and East+Central Europeans have highest Yamnaya, does not mean necessarily that Yamnaya genomes resembled East European ones in having high WHG.

@ Krefter
I am trying to make sense of the fact that Yamnaya, Central Asians and Indians do not resemble East Europeans in having high WHG, but did probably have very high R1a; while high WHG populations today have the highest diversity of R1a, implying that there must be some movement from the latter to all the former, but for some reason this boosted ANE without boosting WHG to the same degree in the receiving populations.

One way out is for high WHG populations in Eastern Europe to have a high level of EHG prior to Yamnaya intrusions, and have the diversity of R1a preserved in EHG in Eastern Euros, while Yamnaya is only a subset of that. So East Europeans retain the diversity, but Yamnaya is responsible for the spread of a subset of the diversity--though a very wide spread at that.

Razib said in his tweets that 'Corded Ware localized to North-Central Europe.' Though of course David might be correct in extending this to cover balts.

I would like to reiterate that the non-local ancestry is probably referring to the EHG in Yamnaya, and thus that EHG was already present on the other side of the finnic peninsula from the karelians, aka in the ancestors of the Balts, is thus not surprising. Its really a very short way away. The question still remains as to whether they are similar to Yamnaya in overall ancestry, in which case we have to look at the '50% Armenian' and the 'Caucasus and Near East' as well.

Davidski said...

rk,

EHG looks like a mix of WHG and ANE, so the WHG in Eastern Europe might be from EHG, and indeed mostly of Corded Ware/Yamnaya origin.

Also, your argument that the R1a in Eastern Europe might have two sources, Proto-Indo-European and local non-Indo-European, doesn't work, because the vast majority of the R1a subclades in Europe and Asia aren't older than the Copper Age.

In fact, the Balto-Slavic and Indo-Iranian subclades are sister clades derived from R1a-Z645, which is less than 6,000 years old.

So basically either all of the R1a in Eastern Europe is Indo-European in origin, or none of it is, and we can extend that premise to South Asia too, because the R1a there is closely related to Eastern European R1a.

But if so, then who spread all of this R1a across both Europe and Asia during the Copper Age?

The fact that South Asians lack WHG, and perhaps even EHG, isn't a problem, since the Indo-Iranians are thought to have expanded from Sintashta deep into Asia, and then the Indo-Aryans are believed to have formed in a contact zone between Andronovo and BMAC.

In other words, autosomal DNA from the steppe got flushed out in Siberia and Central Asia, but the paternal lineages remained.

Tesmos said...

David, do you think there is a significant CWC influence in The Netherlands and/or NW Germany? Because they were part of the Single grave culture that was part of the CWC horizon.

Davidski said...

I think that the people who lived in what is now the Netherlands were a lot like Gokhem2 until the Single Grave culture got there.

Now they're around 15% ANE, and most of that probably came from the Single Grave people. This suggests that there was a lot more than 15% admixture from the CWC horizon and Yamnaya into that part of Europe.

Tesmos said...

Thank you for the infromation.

I have one more question about the new calculator. It looks to me that the ANE scores of European populations are a bit lower than the ones of the ANEK7 calculator.

Are the ANE scores of this new calculator more accurate or is this calculator more WHG centric?

Helgenes50 said...

In Netherland the amount of ANE may be higher because of immigration came from Norway to the historical periods.

Shaikorth said...

RK,

"If there's one ADMIXTURE component that is extremely old and doesn't track any latter-day movements, I think it'd be Baltic."

The K15 components aren't ancient per say, but products of recent drift of modern populations, in the case of Baltic that's Balto-Slavic drift, or Lithuanian drift, whatever you want to call it. Baltic, North Sea and Atlantic are all heavily present in La Braña and Loschbour - at more even rates than they tend to be in modern Europeans - because WHG population has highly contributed to all these components. Same is true of components that WHGs get in other calculators like Eurogenes K36, Harappaworld etc.

Davidski said...

Tesmos,

It's hard to say which results are more correct. They only differ by a couple per cent at most. It might never be possible to estimate these proportions to within a range of less than 1-2%, because there isn't continental-wide differentiation between these components.

Davidski said...

By the way, from now on anyone who acts mentally unstable here will have their posts deleted.

So think about the impression you make, and don't bother responding to crazy stuff, because it won't be up for long anyway.

Krefter said...

Davidski, once ANE K8 is on GEDmatch, will we be able to put our scores on a PCA west Eurasia?

Davidski said...

Maybe, we'll see.

VOX said...

"Isolating ANE from South Amerindian genomes is fairly easy, because they appear to be just a two-way mix of ANE and ENA.

South Indians, even those with seemingly high levels of ASI, are a mixture of ANE, Basal Eurasian, probably something WHG-like, and a component, or components, native to South Asia.

I wouldn't know what I'm looking for exactly. It is frustrating, but I think it's a hopeless cause until we see a pre-Neolithic genome from India, or at least South Central Asia."

South East Asians appear to be a mix of East Asian + South Eurasian. If you start with a South East Asian (like a Vietnamese) and remove pure East Asian segments, the result might be an element similar to ASI. Such a model would assume Koreans/Japanese are relatively pure East Asians while Jehai and Kensiu are close to pure South Eurasian. Other populations like the Dai should be intermediate between the two.

Shaikorth said...

North Han and Japanese still have significant South Eurasian in this test. East Eurasian doesn't peak there but in Hezhen and Oroqen, who also have very little or no South Eurasian.

Krefter said...

I wish there was a way to take out south Eurasian noise in this test.

ANE K7 and K8 doesn't work well for people who have something tiny like 5% non-west Eurasian ancestry. African and east Asian percentages aren't accurate, and I scored 2% ASE in ANE K7. I don't know where to put the ASE, African, and east Asian noise.

VOX said...

"North Han and Japanese still have significant South Eurasian in this test. East Eurasian doesn't peak there but in Hezhen and Oroqen, who also have very little or no South Eurasian"

But then there is the problem of these of East Eurasians having North Eurasian admixture. I wish there was a workaround but in the meanwhile there are some present/future papers that address East and South Eurasian genetic components. It would be great if someone can get the samples.

http://www.ashg.org/2014meeting/abstracts/fulltext/f140120917.htm

http://www.ncbi.nlm.nih.gov/pubmed/24916469

http://www.ashg.org/2014meeting/abstracts/fulltext/f140123567.htm

Grey said...

"And the idea that Balto-Slavs originated from a population that came from the middle of the Corded Ware horizon, but wasn't directly related to Corded Ware people, is just kooky, because of the uniparental marker evidence of very strong ties between Balto-Slavs and Corded Ware people."

I think the R1 + I1 thing north of the Carpathians implies Corded Ware *amalgamated with* some of the HG that were already there and they *expanded together* afterwards leading to the Aesir and Vanir split in later mythology.

DMXX said...

"The fact that South Asians lack WHG, and perhaps even EHG, isn't a problem, since the Indo-Iranians are thought to have expanded from Sintashta deep into Asia, and then the Indo-Aryans are believed to have formed in a contact zone between Andronovo and BMAC.

In other words, autosomal DNA from the steppe got flushed out in Siberia and Central Asia, but the paternal lineages remained."

Alternatively, the whole issue of WHG/EHG is a red herring and has little to do with the Indo-European expansion, with ANE and some sort of EEF-like signal best representing this expansion instead.

What happens on one side of Yamnaya doesn't necessarily mean the genetic landscape was uniform throughout (e.g. there may well have never been much, if any, WHG/EHG around the Urals).

Davidski said...

It does sound like there was some EHG or even WHG near the Urals, because the Yamnaya genomes that Reich has come from the Samara Valley, which is at the eastern end of the former Yamnaya zone.

But the Indo-Iranian expansion didn't take place directly from Yamnaya territory, but from east of the Urals. So if the early Indo-Iranians only had small amounts of EHG, then by the time their descendents the Indo-Aryans appeared in India there may have been nothing left of it.

DMXX said...

^ Ah. The Yamnaya horizon was (like Andronovo) typified by common cultural practices over time superimposed by numerous interactions and/or innovations in the peripheries. We'll need data from Sintashta (or any of the old Scythian remains for that matter) to determine whether WHG/EHG made its' way deep into Asia or not.

The Armenians and Indians may not have any WHG(?), but it does seem to crop up among Iranians and Kurds. Do you have any insights or thoughts regarding this, Davidski?

Davidski said...

I suspect they're having problems in this area because they don't yet have any ancient genomes from the Near East and South Central Asia to carry out direct comparisons with.

For instance, using a European Neolithic genome to try and estimate European hunter-gatherer ancestry in the Near East by proxy is probably not a good idea, especially if the signals of this ancient admixture are relatively minor or muffled in some way. But they did actually attempt this in the last paper, and not surprisingly, concluded that there was no European hunter-gatherer ancestry in the Near East.

DMXX said...

All of this is correct, but I was referring to the, let's say, "informal" (unpublished) comparisons done online, where certain Near-Easterners (e.g. Assyrians) scored negative WHG in the ANE-WHG-EEF model, whereas others (e.g. Iranians, Kurds) actually had positive WHG.

The common interpretation of this was that something resembling WHG was indeed present in the latter, given it wasn't swallowed up by EEF (which we know is partially WHG-like in nature). I was just wondering whether you noticed anything akin to this in your experiments.

Chad Rohlfsen said...

Looking at WHG in Kazakhs, Uighurs, Kyrgyz, and others, it flushed out in their periphery.

Krefter said...

"^ Ah. The Yamnaya horizon was (like Andronovo) typified by common cultural practices over time superimposed by numerous interactions and/or innovations in the peripheries. We'll need data from Sintashta (or any of the old Scythian remains for that matter) to determine whether WHG/EHG made its' way deep into Asia or not. "

Andronovo mtDNA is indistinguishable from Yamna. Considering they(and other north Asian IEs) were pretty much 100% R1a, Andronovo in my opinion was probably close to 100% Yamna.

I bet the Samara Yamna genomes Reich has are very similar to Andronovo.

Andronovo and Sythians have clear EHG mtDNA(U4, U5a, U2e).

We already have R1a-Z93 from IE-like northern Asians. This confirms to me Indo Iranians have EHG.

Chad Rohlfsen said...

Andronovo will not be 100% Yamnaya. There were agricultural people living there already. Yamnaya brought pastoralism after 2500 BCE. Looking at Kazakhs, I'd say that their pre-East Eurasian population was around 60-70% or so Yamnaya. That figure is only if all of their WHG is from Yamnaya, which may not be so. I don't have a problem with seeing a trickle of WHG past the Urals. Even AG-2 looked like he could be 15-20% WHG, if we take his position literally. Kazakhs don't have enough WHG outside of their East Eurasian, to get close to 100%.

Krefter said...

Chad,

What evidence do you have that there were west Eurasian people differnt than Yamna in central Asia.

Uralics are similar to Indo Europeans genetically, because although their linguistic ancestors were not in the same cultural and ethnic group, but because they are a similar EHG+Near eastern mix.

My guess would be that anything non-east Asian in Russia during the copper age was similar to Yamna.

Chad Rohlfsen said...

Uhm.. Who do you think brought farming to Yamnaya and Central Asia?

Here's a couple of things to read.

http://www.primtech.net/Zhusahn/Report.html

http://www.primtech.net/Zhusahn/Report.html 400 years older than Sintashta.

"This evidence is interpreted by the excavators, an assertion supported by many other scholars, as indication of a distinct route of transmission of wheat and millet from the central Asian mountains and into the steppes by the late 3rd millennium BC" (Frachetti et al. 2010).

Farmers almost certainly entered Kazakhstan from the West and the South.


Chad Rohlfsen said...

Also..

The wheat consisted of 13 whole seeds of domesticated compact free-threshing wheat, either Triticum aestivum or T. turgidum. Frachetti et al. report that the wheat compares favorably to that from the Indus Valley region in Mehrgarh and other Harappan sites, ca. 2500-2000 cal BC and from Sarazm in western Tajikistan, ca. 2600-2000 BC.

A total of 61 carbonized broomcorn millet (Panicum miliaceum) seeds were recovered from various Phase 1a contexts, one of which was direct-dated to 2460-2190 cal BC. One barley grain and 26 cerealia (grains unidentified to species), were also recovered from the same contexts. Other seeds found within the soil samples are wild Chenopodium album, Hyoscyamus spp. (also known as nightshade), Galium spp. (bedstraw) and Stipa spp. (feathergrass or spear grass). See Frachetti et al. 2010 and Spengler et al. 2014 for additional details.

Grey said...

"Alternatively, the whole issue of WHG/EHG is a red herring and has little to do with the Indo-European expansion, with ANE and some sort of EEF-like signal best representing this expansion instead."

There were sedentary pottery using HGs along the west shore of the Black Sea. A population like that would be pre-adapted to take up farming imo - especially if the abundant food source that had allowed them to be sedentary was drying up (e.g. wetlands around the Black Sea disappearing with increased sea level).

So what about...

1) Initially WHG + NHG (northern HG). WHG in the southern and central European strips from Britain to the Black Sea and NHG in the northern strip stretching from Scandinavia to *and beyond* the Black Sea east (with Scandinavia/Baltic being a border zone so mixed WHG +NHG).

i.e. WHG ends at the Black Sea while NHG carries on eastwards beyond the Black Sea


2) NEF (near eastern farmer) move to Thessaly and under their influence the sedentary WHG west of the Black Sea take up farming thus creating EEF (or one version of EEF anyway).

3) The newly minted EEF (WHG+NEF) create a hybrid culture with NHG on the border of the farmer/steppe zone

4) Further east a more fully pastoralist version of that culture develops with mostly NHG.

.

so in a clockwise sequence of locations

http://upload.wikimedia.org/wikipedia/commons/5/52/Black_Sea_map.png

1) Thessaly
- NEF

2) Bulgaria/Romania
- NEF + (sedentary) WHG = EEF

(one version of EEF anyway)

3) Moldova/West Ukraine
- EEF + NHG (aka ANE)

4) East Ukraine (or further east)
- NHG (aka ANE)

#

Then maybe

1) the fully pastoralist group (ANE) start expanding.

2) they expand rapidly to the east over the steppe where there are only low density HGs to stop them followed by a slower expansion south against higher density farmers which requires a raid-advance-raid-advance sequence all the way to India changing their autosomal DNA along the way through female captives.

3) blocked to the west by the hybrid culture (EEF) the raid-advance-raid-advance sequence starts from the beginning eventually pushing them west and south: west into WHG/NHG HG populations north of the Carpathians who they incorporated and south onto the Thessalian NEF pushing them south in turn and/or incorporating them.

.

So returning to the original question

"Alternatively, the whole issue of WHG/EHG is a red herring and has little to do with the Indo-European expansion, with ANE and some sort of EEF-like signal best representing this expansion instead."

1) The above would explain why PIE east of the Black Sea was mostly ANE (aka EHG aka NHG).

2) If a hybrid culture (EEF+ANE) originally from the Moldova/West Ukraine region was pushed south and west by IE and in advance of IE

and they in turn pushed the EEF (WHG+NEF) from around Romania/Bulgaria south and west

and they pushed the NEF around Thessaly south and west then you'd have one IE wave but made up of four components in a domino effect.

Grey said...

"blocked to the west by the hybrid culture (EEF)"

should be (EEF+ANE)

so
pastoralist (ANE) push hybrid (EEF+ANE)

hybrid (EEF+ANE) push EEF (WHG+NEF)

EEF (WHG+NEF) push NEF

Dagnė Eitutytė said...

@Pierre de Laclos
I am reading a lot this blog and I like it very much.
So don't think women are not here, even if not so vocal as others.
Merry Christmas!

Ted Kandell said...

David, can you add the Samaritans to this plot? As you know, the Samaritans are one of the populations that are closest in the various Dodecad oracles to Stuttgart and NE1 after the "Sardinian" component is accounted for. They are a remnant of a pre-Hellenistic Persian Era isolate in the Levant, without detectable outside admixture after about 444-338 BCE. On the other PCAs they appear somewhere in the vicinity of the Lebanese Christians and Druze. Also, the Phocaean Greeks appear to be in a unique position somewhere between the Cypriots and and the Sardinian-like Mediterraneans, rather different from other Greeks. Adding these two populations would help "bridge the gap" between the Levantines and the Early European Neolithic Farmers, and I think both will help clarify the source and extent of the any "UHG" that may have arrived in Europe with the Neolithic Farmers. In lieu of Near Eastern PPNB Neolithic data, this should help identify the precise nature and location of the source population for the Neolithic immigrants. Maybe some kind of ANE arrived in the Near East before the PIE's, but just after 6000 BCE when the Early Neolithic Farmers left?

Davidski said...

Ted,

To be honest, I haven't looked at the Dodecad oracles. I don't think they're very useful because of the calculator effect.

But I added one of the Samaritans to both of the sheets in the K8 spreadsheet, and he doesn't appear to be a good fit for Stuttgart or NE1, or in fact their Near Eastern ancestors. The main reason is that he has almost 5% of ANE, and even some Sub-Saharan admixture, which is low but probably not noise.

https://docs.google.com/spreadsheets/d/1x8pm8sVcHqceiNFJMO082kxaBF5ePr4__bAK05VQRFw/edit?usp=sharing

You can see the effect this has on the Samaritan's K8 PCA position (look for the red dot).

https://drive.google.com/file/d/0B9o3EYTdM8lQNWpmN1FaWndobDQ/view?usp=sharing

However, without that ANE, the Samaritan might well be a better fit for the Near Eastern ancestors of NE1 and Stuttgart than the BedouinB are.

Anne Hart said...

Great granny is pushing 80. No one has referred to me as a commodity in all those years. Anyway, I'm mostly Lebanese with some Greek and Italian. Please let us know when your new DNA test comes out the K8, as I'd like to see whether I match any of those ancient DNA people. Sounds like lots of fun for this genteel lady. Enjoy...This site is so uplifting.

Ali Monsta said...

Hello. I have a question. I just took the ANE K7 test & I'm getting that I am mostly east African but on all of my other test, I am getting that I'm sub saharan or West Africa. So now I am confused. I know you are the creator of this project so, if you could shoot me an email to help me out, that would be great. Thanks! princessali1785@gmail.com

kristian frimer linde said...

Will there ever be tested more then the 103 reference population to West Eurasia K8,that would hel us find out whether Latvians or Lithuanians are the most WHG in Europe i think the some latvians as well as other peoples such as, Danes, Portuguese, Irish, Czech republic, Slovakia, Yemen, Madagascar, Tamils, other Indian peoples, Balinese, Polynesians, Algerians, Botswanan people, different regions of France, and Germany, Austrians, Swiss people as well as more samples from Finland(West and East) as well as at least some Saamis people should be tested.

Davidski said...

@kristian frimer linde

Lots of new samples in the new test, including Latvians. See spreadsheet here.

The Basal-rich K7

«Oldest ‹Older   201 – 282 of 282   Newer› Newest»