search this blog

Wednesday, March 25, 2020

The origins of East Asians (Wang et al. 2020 preprint)


Over at bioRxiv at this LINK. Here's the abstract:

The deep population history of East Asia remains poorly understood due to a lack of ancient DNA data and sparse sampling of present-day people. We report genome-wide data from 191 individuals from Mongolia, northern China, Taiwan, the Amur River Basin and Japan dating to 6000 BCE - 1000 CE, many from contexts never previously analyzed with ancient DNA. We also report 383 present-day individuals from 46 groups mostly from the Tibetan Plateau and southern China. We document how 6000-3600 BCE people of Mongolia and the Amur River Basin were from populations that expanded over Northeast Asia, likely dispersing the ancestors of Mongolic and Tungusic languages. In a time transect of 89 Mongolians, we reveal how Yamnaya steppe pastoralist spread from the west by 3300-2900 BCE in association with the Afanasievo culture, although we also document a boy buried in an Afanasievo barrow with ancestry entirely from local Mongolian hunter-gatherers, representing a unique case of someone of entirely non-Yamnaya ancestry interred in this way. The second spread of Yamnaya-derived ancestry came via groups that harbored about a third of their ancestry from European farmers, which nearly completely displaced unmixed Yamnaya-related lineages in Mongolia in the second millennium BCE, but did not replace Afanasievo lineages in western China where Afanasievo ancestry persisted, plausibly acting as the source of the early-splitting Tocharian branch of Indo-European languages. Analyzing 20 Yellow River Basin farmers dating to ~3000 BCE, we document a population that was a plausible vector for the spread of Sino-Tibetan languages both to the Tibetan Plateau and to the central plain where they mixed with southern agriculturalists to form the ancestors of Han Chinese. We show that the individuals in a time transect of 52 ancient Taiwan individuals spanning at least 1400 BCE to 600 CE were consistent with being nearly direct descendants of Yangtze Valley first farmers who likely spread Austronesian, Tai-Kadai and Austroasiatic languages across Southeast and South Asia and mixing with the people they encountered, contributing to a four-fold reduction of genetic differentiation during the emergence of complex societies. We finally report data from Jomon hunter-gatherers from Japan who harbored one of the earliest splitting branches of East Eurasian variation, and show an affinity among Jomon, Amur River Basin, ancient Taiwan, and Austronesian-speakers, as expected for ancestry if they all had contributions from a Late Pleistocene coastal route migration to East Asia.

Also this part is interesting, but surprisingly naive:

The findings of the original study that reported evidence that the Afanasievo spread was the source of Steppe ancestry in the Iron Age Shirenzigou have been questioned with the proposal of alternative models that use ancient Kazakh Steppe Herders from the site of Botai, Wusun, Saka and ancient Tibetans from the site of Mebrak 15 in present-day Nepal as major sources for Steppe and East Asian-related ancestry [28]. However, when we fit these models with Russia_Afanasievo and Mongolian_East_N added to the outgroups, the proposed models are rejected (P-values between 10 -7 and 10 -2), except in a model involving a single low coverage Saka individual from Kazakhstan as a source (P=0.17, likely reflecting the limited power to reject models with this low coverage). Repeating the modeling using other ancient Nepalese with very similar genetic ancestry to that in Mebrak results in uniformly poor fits (Online Table 5). Thus, ancestry typical of the Afanasievo culture and Mongolian Neolithic contributed to the Shirenzigou individuals, supporting the theory that the Tocharian languages of the Tarim Basin—from the second-oldest-known branch of the Indo-European language family—spread eastward through the migration of Yamnaya steppe pastoralists to the Altai Mountains and Mongolia in the guise of the Afansievo culture, from where they spread further to Xinjiang [5,7,8,27,29,30]. These results are significant for theories of Indo-European language diversification, as they increase the evidence in favor of the hypothesis the branch time of the second-oldest branch in the Indo-European language tree occurred at the end of the fourth millennium BCE [27,29,30].

I'd say the authors are putting too much faith in their qpAdm mixture models. They ought to know that qpAdm has some serious limitations, especially in regards to fine scale ancestry. I would urge them to become better acquainted with the uniparental markers of the Iron Age Shirenzigou samples instead of forcing the ideas that these individuals harbor Afanasievo-derived ancestry and lack Tibetan-related ancestry.

See also...

They mixed up Huns with Tocharians

A surprising twist to the Shirenzigou nomads story

Afanasievo people may well have been proto-Tocharian speakers (Ning et al. 2019)

346 comments:

1 – 200 of 346   Newer›   Newest»
Samuel Andrews said...

This paper refers to all European, Mid Eastern admix in ancient Mongolia/Western China as "West Eurasian." This genetic concept of West Eurasian need a serious update.

First of all, because we know why West Eurasian exists, why Europeans and Southwest Asians are related.It is because, Europeans are 50%+ Middle Eastern (Stone age). This is the whole reason why West Eurasian became a genetic category. Mystery solved.

Second of all, because we can easily break down what West Eurasian admix in Asia is European & what is Middle Eastern. We don't have to call it all mysterious "West Eurasian-related" admixture anymore.

Admix in an Asian pop which arrived after 5,000 BC and with significant Stone age European ancestry, is European not West Eurasian. Afanasievo, Andronovo in Central Asia that's European admixture.

BMAC admix in Central Asia, that's Middle Eastern. BMAC-types is from where all the Middle Eastern ancestry in Central Asia comes from.

West Eurasian is too ambiguous. Together you can vauglly put BMAC & Andronovo under the West Eurasian category, but you can also break them down into two separate categories: One European, one Middle Eastern. But, they don't do that. No ancient DNA experts do.

Samuel Andrews said...

@Davidski, What's your opinion on this? FST stat is what they always site to say the four ancient West Eurasian pops were completely unrelated?

"In West Eurasia, there were at least four divergent populations each as genetically differentiated from each other as Europeans and East Asians today (average FST=0.10),"

Is FST-distance a reliable tool for understanding deep relationships between populations?

Davidski said...

@Samuel

You're right, FST isn't a reliable measure of that sort of thing.

Some of the FST distances between Amerindian popuations are as high as between continental groups, but that's just due to recent genetic drift.

So yeah, it's misleading.

Ryan said...

So there were 3 Yangtze River farmer language families? That seems like a lot.

Samuel Andrews said...

@Rob,
"They spend too much time talking about Afanasievo, and completely miss the important point about a Neolithic - Bronze Age shift in Mongolia,. They completely ignore the main source population- putative Altai-Yenesei basin foragers."


You are referring to ANE-rich admix which moved east deeper in Asia, like the kind in Okenevo?

Rob said...

@ Sam- yeah, and into Baikalia

Samuel Andrews said...

Yes, I guess that migration does go unnoticed when compared to Afanasievo.

Synome said...

The WSHG influx may be related to the ancestors of the Xiongnu and Hunnic peoples, who show ancestry of this type.

I admit I was disappointed the the evidence given for the Afanasievo-Tocharian connection isn't stronger. Still no smoking gun there.

Davidski said...

Their insistence that the Shirenzigou nomads were an Afanasievo related group, and Tocharian speaking, is pretty funny.

All they did was mess around with the right pops in qpAdm, and voila, they got what they wanted.

Obviously, the Shirenzigou nomads are Hun/Saka related with the wrong type of R1b.

Samuel Andrews said...

@Davidski, I'm not agreeing with them. But, Andronovo can't explain R1b Z2103 in Iron age Central Asia. It would make sense, the R1b Z2103 seen in Huns and Shirenzigou is ultimatly from Afanasiveo.

Davidski said...

@Samuel

There's no Z2103 in any Huns or the Shirenzigou.

They belong to Central Asian subclades of R1b and one of the Shirenzigou probably belongs to R1a-Z93.

Anonymous said...


They clearly got something wrong with haplogroups. They use the ISOGG 2016, while they have R1b1a1a2a1a in the Afanasievo that gives the non-Afanasievo's L11/PF6539/S127, and Chemurchek has the R1b1a1a2a2 that gives the Afanasievo's Z2103.

Anonymous said...

Another ones of the candidates for the Tocharians.

383-535 calCE (1625±20 BP, UCIAMS-226554, PSUG-5469) Mongolia_IA_Xianbei Yes Khovd, Bulgan sum, Uliastai deed denj II kurgan 1 burial 1 (secondary) Mongolia 45,39 90,798 1240K capture 1 5,534433 848539 M R1b1c

890-982 calCE (1115±20 BP, UCIAMS-226555, PSUG-5470) SOME DOUBT - The genetic data and direct 14C date are from the same exported petrous bone. However, this individual is the only representative of their genetic cluster and the archaeological context is not fully clear, so we treat this archaeological and chronological context assignment as tentative. Mongolia_Medieval Yes Khovd, Bulgan sum, Hudzhirtyn gol II, kurgan 1 burial 3 (secondary) Mongolia 45,35 90,85 1240K capture 1 0,79158 546668 M R1b1c

https://www.yfull.com/tree/R-PH155/

Rob said...

But Kudos for the sampling. Pretty cool stuff

Davidski said...

I wouldn't trust their Y-haplogroup assignments, not at that level anyway.

Samuel Andrews said...

Yes interesting to see first major ancient DNA study in East Asia.

Something interesting, is this confirms the East Asian look, fair skin & slanted eyes, long predates 6000 BC because it exists in East Asian lineage which were separated by at least 5000 BC.

And the East Asians are more diverse, than the Europeans. Russians more related to Spanish than Chinese are to Ulchi.

Ryan said...

Isn't Z2103 the only R1b in pretty much every branch of Indo-European today?

CrM said...

There's a J1 in an Afanasievo from Mongolia, would J-L283 also be found among IE-related groups?

Davidski said...

There's also J1 in EHG and Khvalynsk, so it was around in Eastern Europe at low levels since the Mesolithic, but it didn't become a major lineage of the Bronze Age pastoralist groups expanding from the steppe.

I'm not aware of any L283 in Afanasievo, Yamnaya, Corded Ware etc.

Samuel Andrews said...

@CrM, Nice call.

@All, So, I guess Y DNA NO and C2 are the most original East Asian y-haplogroups. The Y DNA D1 in Joman can be explained as being from Onge-related groups. Y DNA Q1a is from ANE. So, all that's left is N0 and C2.

Samuel Andrews said...

Mongolia_BA mostly has same clades of Q1a as Okenuvo. Rob, is right migrations from Central Asia into East were important.

Davidski said...

@Arza

If you're reading this, does this look like R1a to you?

https://www.ebi.ac.uk/ena/data/view/SAMEA5613663

And which R1b subclades are you seeing in these samples?

M012

https://www.ebi.ac.uk/ena/data/view/SAMEA5613665

M15-1

https://www.ebi.ac.uk/ena/data/view/SAMEA5613668

Rob said...

These Shirenzigou samples are all over the place- they are quite diverse. Just look at a PCA which includes them. Some of them even need Turan admixture
So constraining a model to ''The only parsimonious model for this group that fits according to our criteria is a 3-way mixture of groups related to Mongolia_N_East, Russia_Afanasievo, WSHG.'' seems problematic. And where's the holistic understanding ? What really catalysed the formation of nomadic groups in Inner Asia ?

Davidski said...

They're obviously fishing for their desired outcome: Shirenzigou > Tocharians.

But this is very unfair to others, like historical linguistics, who will use these sorts of papers to guide them.

For the next ten years we'll be reading how the Afanasievo people and Shirenzigou nomads were Tocharians, even though that's not supported by any solid, direct evidence.

FrankN said...

Rob: "These Shirenzigou samples are all over the place.."
Even more all over the place are the Wuzhuhangguoliang samples. Some could pass for modern Mongols, other for Xiognu, and overall they display quite a drift towards ancient Nepalese. Still, the paper treats them as a one homogeneous population, while in contrast breaking up the far more homogeneous ancient Mongolians into a dozen seperate clusters.

Based on this non-Analysis, they than conclude that Wuzhuhangguoliang is ancestral to modern Han (with some Taiwanese and occasional "West Eurasian" admix), and "that Tibetans harbor ancestry from a population closely related to Wuzhuangguoliang that also contributed more to Qiang and Han than to other present-day East Asian Groups". Yeah, Tibetans and Hans have been cousins for millennia..

Apparently, getting aDNA from China comes with a price. I can understand that no senior staff member from Harvard or MPI Jena (except Wang, of course) wanted to see his name in the list of authors.

Dave (and others) - once the aDNA is out, there is a lot of work ahead in trying to better understand that Wuzhuangguoliang amalgam.

Samuel Andrews said...

@FrankN, yeah and they barelly talk about the Wuzhuhangguoliang samples. I don't see them in any stats anyalsis.

gamerz_J said...

@Davidski

"I would urge them to become better acquainted with the uniparental markers of the Iron Age Shirenzigou samples instead of forcing the ideas that these individuals harbor Afanasievo-derived ancestry and lack Tibetan-related ancestry."
Do the Shirenzigou samples have direct Tibetan ancestry in your opinion?
I recently read a paper on Himalayan populations and Tibetans oddly seem to have some sort of West Eurasian ancestry but not from South Asia in the author's opinion which makes me wonder if something is going on in the Qinghai plateau.

I also came across this, it might be of some interest. https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1008552#pgen-1008552-g006

gamerz_J said...

@Samuel Andrews

Do you know how much interaction there was between the "West Eurasian"-related groups in Mongolia/Western China and BMAC?
Was not BMAC also partly South and East Asian?

"And the East Asians are more diverse, than the Europeans. Russians more related to Spanish than Chinese are to Ulchi."

Btw, agree that the "East Asian Mongoloid" phenotype is much older than 6000 years but would you happen to have an Fst chart or citation for the last sentence? Interesting if true.

Rob said...

@ FrankN

'These Shirenzigou samples are all over the place.."
Even more all over the place ''

To clarify, that do make sense and form a cline, but not as characterised in the paper. Shirenzigou are a product of the trade route on which they sit on. So, I dont think that particular oversimplification much to do with the paper being from China

Arza said...

@ Davidski

M4 is Z93.

I will have M012 and M15-1 in ~3 hours.

@ Archi

They clearly got something wrong with haplogroups. They use the ISOGG 2016, while they have R1b1a1a2a1a in the Afanasievo that gives the non-Afanasievo's L11/PF6539/S127, and Chemurchek has the R1b1a1a2a2 that gives the Afanasievo's Z2103.

That R1b1a1a2a1a is L52. But the sample is "QUESTIONABLE (mtmatchmax=0.861)".
Haplogroup assignments are fine.

Arza said...

Re: new Italian samples (9309 supposedly is https://www.yfull.com/tree/R-S1161/ )

Chimp Yamnaya_Samara 9323_Oss 9309_Oss -0.0026 -0.532 22584 22703 424969
Chimp Yamnaya_Samara 9323_Co 9309_Co -0.0004 -0.092 27305 27330 517879
Chimp Yamnaya_Samara 9323_Oss 9309_Co -0.0001 -0.022 27761 27766 524360
Chimp Yamnaya_Samara 9323_Co 9309_Oss -0.0021 -0.417 22259 22352 420138

Chimp Yamnaya_Samara Sicily_MN 9309_Oss 0.0037 0.955 27502 27299 512361
Chimp Yamnaya_Samara Sicily_MN 9309_Co 0.0037 0.996 33899 33648 632308
Chimp Yamnaya_Samara Sicily_MN 9323_Oss 0.0056 1.551 35605 35206 664128
Chimp Yamnaya_Samara Sicily_MN 9323_Co 0.0035 0.952 34904 34662 653181

Chimp Yamnaya_Samara Bulgaria_Dzhulyunitsa_N 9309_Oss 0.0024 0.329 8065 8026 149695
Chimp Yamnaya_Samara Bulgaria_Dzhulyunitsa_N 9309_Co 0.0030 0.456 9754 9695 181750
Chimp Yamnaya_Samara Bulgaria_Dzhulyunitsa_N 9323_Oss 0.0011 0.164 10058 10036 188194
Chimp Yamnaya_Samara Bulgaria_Dzhulyunitsa_N 9323_Co -0.0063 -0.960 9854 9978 186293

Chimp Yamnaya_Samara Bulgaria_MalakPreslavets_N 9309_Oss -0.0043 -0.948 20158 20332 375498
Chimp Yamnaya_Samara Bulgaria_MalakPreslavets_N 9309_Co 0.0007 0.171 24802 24765 460363
Chimp Yamnaya_Samara Bulgaria_MalakPreslavets_N 9323_Oss 0.0020 0.479 25696 25595 475891
Chimp Yamnaya_Samara Bulgaria_MalakPreslavets_N 9323_Co 0.0012 0.288 25259 25197 471163

Chimp Yamnaya_Samara Hungary_LateC_Baden 9309_Oss -0.0015 -0.395 27297 27380 510731
Chimp Yamnaya_Samara Hungary_LateC_Baden 9309_Co -0.0007 -0.192 33594 33641 630421
Chimp Yamnaya_Samara Hungary_LateC_Baden 9323_Oss -0.0002 -0.061 35270 35285 661758
Chimp Yamnaya_Samara Hungary_LateC_Baden 9323_Co 0.0002 0.054 34698 34685 651198

Chimp Yamnaya_Samara Sardinia_Neolithic 9309_Oss 0.0007 0.147 18280 18253 341256
Chimp Yamnaya_Samara Sardinia_Neolithic 9309_Co 0.0022 0.456 22133 22034 415808
Chimp Yamnaya_Samara Sardinia_Neolithic 9323_Oss 0.0022 0.493 23016 22913 431348
Chimp Yamnaya_Samara Sardinia_Neolithic 9323_Co 0.0045 0.967 22773 22570 427050

Chimp Yamnaya_Samara Italy_N.SG 9309_Oss 0.0088 2.276 27738 27251 511421
Chimp Yamnaya_Samara Italy_N.SG 9309_Co 0.0090 2.453 34078 33470 630973
Chimp Yamnaya_Samara Italy_N.SG 9323_Oss 0.0104 3.022 35715 34981 662961
Chimp Yamnaya_Samara Italy_N.SG 9323_Co 0.0097 2.767 35172 34493 651975

Rob said...

ANother paper, simialr samples ?

https://www.biorxiv.org/content/10.1101/2020.03.25.008078v1.full.pdf

Some very different conclusions

Anonymous said...

@Arza

I think they got it wrong with ISOGG version. it's very likely that they wrote nonsense, it's the Chinese. I don't believe them yet. Errors with the definition of haplogroups in such papers are not the first time.

Arza said...

Kovalev A.A. The Great Migration of the Chemurchek People from France to the Altai in the Early 3rd Millenium BCE

That's why I6222/SHT001 is:

SHT001 AT-26 Tooth - SHT001.B Early Bronze Age Afanasievo
SHT001.B0101; SHT001.B0102 M N1a1a1a1 R1b1a2a1a (R-P311)

Anonymous said...

@Arza "That's why I6222/SHT001 is: SHT001 AT-26 Tooth SHT001.B0101; SHT001.B0102 M N1a1a1a1 R1b1a2a1a (R-P311)"

I6222 AT_26, Grave #2, MONU_26, Grave #2 Pinhasi, Ron; Cheronet, Olivia; Sirak, Kendra; Dashtseveg, Tumen petrous (CBD) 5067 3316-2918 calBCE (4415±31 BP, OxA-36222) Confident - The genetic analysis was done on a different individuals from the sample used to produce the direct 14C date (published in Wilkins et al. Nature Ecology and Evolution 2020). However, the 14C date matches to the Afanasievo period, and this individual clusters genetically with previously reported individuals from the Afansievo culture from the Altai region of Russia, providing confidence int he cultural assignment. Mongolia_Chalcolithic_1_Afanasievo Yes Bayankhongor aimag, Erdenetsogt sum, Shatar chuluu kurgan 2 Mongolia 46,4 100,82 1240K capture 1 0,136 146406 M R1b1a1a2a1a 25 R

Do you have sources?

Vladimir said...

This job is even more interesting: "A dynamic 6,000-year genetic history of Eurasia’s Eastern Steppe", Choongwon Jeong, Ke Wang, Shevan Wilkin ... https://www.biorxiv.org/content/10.1101/2020.03.25.008078v1.full.pdf

Anonymous said...

@Vladimir "Jeong"

This paper has very bad tables, they are hard to look at, they are completely uninformative.

Anonymous said...


In general, I don't believe these two papers in the least. The fact is that Chemurchek in their Y-haplogroup turns out to be typical Afanasievo/Yamnaya, but autosomal they have nothing in common with them at all! And Vice versa, the Mongolian Afanasievo they have autosomal exactly the same as the Russian Afanasievo, but Y-haplogroups have nothing to do with them!

I have a feeling that April 1 came early.

Arza said...

Genotypes of 191 ancient and 383 modern East Asians reported in Wang et al. bioRxiv 2020

Anonymous said...

@Arza

Genetics does not confirm the hypothesis of Kovalev, Chemurchek people have nothing in common with France, if they still had R-P311, then this option could be.

Arza said...

I6222 Y-DNA calls: https://pastebin.com/mAcLwZaQ

Anonymous said...

@Arza

Do you have source of I6222 or SHT001? Or did someone just type it?

Arza said...

@ Archi

Reich Lab has published genotyped data: https://reich.hms.harvard.edu/datasets

Anonymous said...

@Arza

Is it possible to get Y-calls out of there?

Jatt_Scythian said...

Can someone break this down for me?

1. WHat is the native range of EHG, ANE and WSHG? WHich areas are West Eurasians native to?

2. WHat was the grographical distribution of Afanasievo? And did they have any other ydnas besides R1b-Z2103?

3. What is the most likely Tocharian haplogroup?

4. Did the ancient Nepalese harboer any ASI or West Eurasian ancestry?
I'm kind of dissapointed if they didn't. It just weird that Tibetans exist on both sides of the Himalayas. I was hoping the south side of the Himalayas be more SOuth Asian. Oh well.

Jatt_Scythian said...

The second paper says ANA was the local population of the altai? I always figured they would be WSHG? AM I reading this wrong?

Jatt_Scythian said...

WHere is the WSHG and ANE continum? Seems like it doesnt exist given these results. I mean just look at the location of Mal'ta boy.

AWood said...

@Jatt_Scythian

As I understand it, based on Yana, northern Eurasia was ANE at the earliest, some 31,000 years ago, at some point before the Native Americans crossed the bridge to modern Alaska, a large chunk of northern Eurasia was replaced with ANA. Considerably later, western populations moved back east to a now ANA dominant population

Arza said...

@ Archi
Yes, it's possible, although with some caveats (no number of reads, consensus call in case of differences between reads etc.).

Do you want to see Chemurchek?

Anonymous said...

@ Gioiello

Kovalev's hypothesis is that the Chemurchekians came to the Altai from Neolithic (Megalithic) France in the middle of the 3rd millennium BC, but the Chemurchekians have nothing common with Neolithic France, according to the Y-haplogroup they are not distinct from the Afanasievans, but absolutely unlike anyone autosomally, supposedly have something to do with the Botaians and Turkmenistan (BMAC). That's nonsense.

Thus age ostensibly Afanasievo's R1b1a1a2a1a-L52 3316-2918 calBCE, that is the oldest sample L52! Long before everyone else. With all the rest this Afanasievan is no different from the rest of Afanasievans.

1 April)))

Jatt_Scythian said...

What is Yana?

WHat would you estimate the range of ANE was?

I thought I read about Y K2/P in the Russian Far East and mt U in the new siberian islands region?

Arza said...

@ Davidski

M4 is R-Z93

M012 is R-PH155 (xM335)

M15-1 is hmm... BY14575. It seems that it's R-PH200 level. Additionally in one of the two files it's derived for Y92825 but ancestral in both files to Y104457 which is on the same level (R1b2b2) according to ISOGG. But since Y104457 is not present in the YFull tree he would land here: https://www.yfull.com/tree/R-Y81807/

Samuel Andrews said...

@Arza, "I6222 Y-DNA calls: https://pastebin.com/mAcLwZaQ"

Thanks for confirming this Afanasievo is R1b L52. It isn't shocking. Now we have a Corded Ware & Afanasievo R1b L52.

Anonymous said...

@Samuel Andrews

This is not from the source, but from the interpreted file. What is in the source code is still unknown, so it is too early to talk about confirmation.

AWood said...

Hmmmm... food for thought. So rather than it being Yamnaya -> Bell Beaker, maybe Afanasievo -> Bell Beaker? That's a hell of a trip. (50% tongue in cheek here)

Arza said...

@ Samuel Andrews

"Thanks for confirming this Afanasievo is R1b L52. It isn't shocking."

It shouldn't be. After all Afanasievo has the ancestry from the West.

Vladimir said...

Theoretically, of course, someone from R1b-L51 could go with Z2103 to the East. In this case, L51 should be somewhere in the Eastern part of Ymnaya, probably in the Volga region. This fits well with the concept of m269 being located in Volosovo, which is closely connected to the Volga region.

Anonymous said...


Once again, the fact that there is something unclean with the interpretation of Y-haplogroups show Chemurchekians. They somehow turn out to be Yamnians in the Y-haplogroup and have absolutely nothing to do with Yamnians in autosomes. This absurdity clearly shows that something is wrong, well, there can be no such thing.


Arza said...

@ AWood

There is no M269 in Khvalynsk.
There is no M269 in Piedmont.

There is M269 in I5884 with barely any "steppe" ancestry.
There is M269 in Olmo di Nogara practically without any "steppe" ancestry (see Dstats posted in this thread).

Yamnaya/Afanasievo is more or less Khvalynsk + Piedmont with a Western shift.

Khvalynsk + Piedmont has no M269.
There are "Western" samples with M269.

https://i.postimg.cc/k7CN0tyC/M269.jpg

Conclusion?

gamerz_J said...

@Jatt_Skythian

Yana is an 30k year old population/lineage that together with Mal'ta boy forms the ANE clade. ANE were widespread in much of Siberia, though the Yana samples are from North-East Siberia north of Lake Baikal IIRC.

ANE later on, expanded both towards Europe and Central Asia and some ANE populations mixed with basal East Asians and crossed the Bering strait into the Americas.
Most Siberian populations have some ANE today with some papers suggesting a very high ANE percentage in groups such as Mansi and Nganasan.

Samuel Andrews said...

@Arza, R1b M269 isn't indigenous to Italy. The Bronze age Italy sample got his R1b M269 ultimatly from Bell beaker. Also, he did have Steppe ancestry, 10%, which equals 30% Bell beaker North Italy.

Samuel Andrews said...

IMO, R1b M269 isn't from Volosovo. Volosovo had it, but so did the Hg ancestor of Kurgan people. There's very few samples from Eneolithic Steppe outside of Khvalnsky. R1b M269 should pop up in Sredny Stog.

AWood said...

@Arza

I don't really follow your point. Certainly by 2600 BC, the central European Bell Beaker cultures were well accustomed to taking local wives with roots in the Mediterranean or western Turkey boosting their EEF, just as these East Asian genomes sometimes pop up with a M269+ male line despite having significant ANA ancestry. Nobody lives in a silo, the point is that it would be interesting if the founding L11+ branch derived from a lone Afanasievo male. Yamnaya looks nearly indistinguishable from Afanasievo, and in earlier works Yamnaya was often used, and maybe still is used as a proxy for a large portion of ancestry for modern Euros. Afanasievo had previously been purely Z2103, yet here is L52+ (assuming correct) as well as a mysterious J1. There is a lot we still do not know.

Anonymous said...

JuanRivera said...
There's a Q1a2 in Afanasievo too.

This is the Okunevian invasion. This is at the very end of the Afanasievo culture and the beginning of the Okunevo culture.

Rob said...

The Jeong (MPI) paper concludes differently, or the samples are truly differemt

''The succeeding EBA Chemurchek culture (2750-1900 BCE), a ruminant dairying society
(Wilkin et al., 2019) whose mortuary features include stone slabs and anthropomorphic stelae, has also been purportedly linked to WSH migrations (Kovalev and Erdenebaatar, 2009). Chemurchek graves are found throughout the Altai and in the Dzungar Basin in Xinjiang, China 203 (Jia and Betts, 2010). We analyzed two Chemurchek individuals from the southern Altai site of Yagshiin Huduu and one from Khundii Gobi (KUM001) in the northern Altai. Compared to Afanasievo_Mongolia, the Yagshiin Huduu individuals also show a high degree of Western ancestry but are displaced in PCA (Fig. 2), having also a strong genetic affinity with ANE207 related ancient individuals such as AfontovaGora3 (AG3), West_Siberia_N, and Botai (Fig. 208 3a; Fig. S9, S11). We find that these Chemurchek individuals (Chemurchek_Altai) are genetically similar to Dali_EBA (Fig. 3a)''

Jatt_Scythian said...

What are the ydnas of the Chemurchek culture?

Rob said...

@ Jatt

The 2 from Ning are R12b-Z2103 (but cont. slightly elevated). Those from Jeong, not avail yet

Rob said...

Based on these & previous papers, it seems there was a cline/cluster/continuum of fisher-foragers from west Eurasian to east (not realy surprising). But these show fixations of some rather divergent lineages. E.g. Amur basin (Devils Gate-type) and Mongolian plateau are Y-DNA C2 (with D in Japan and isolated areas of Nepal).
the Baikal was old clades of N.
To the west of that (Altai-Yenesi) was Q

Q is from the ANE populations
C & D must be very old Upper Paleolithic relics
N & O must have expanded during Holocene, intruding & overriding much of C & D, and perhaps AFG-3 types from Baikalia too

Davidski said...

@All

I've got the genotypes. I'll run them later today.

Anonymous said...


There is another interesting sample:

Late Xiongnu_sarmatian NAI002 / NAI003 NAI002.A0101 NAI002.A0101; NAI003.A0101 M C4a2c1 R1a1a1b1 (R-Z283)


Anonymous said...

@Samuel Andrews

West Eurasian is an ancestry branch that goes back to Kostenk1 14, much prior than neolithic or even the appearance of the Villabruna cluster.

Also, many ameríndians have slanted eyes, so this characteristic is in fact quite old.

Anonymous said...

@ gamerz_J

Yana it's actually Ancient North Siberian. Ancient North Eurasians are modeled as 88% ANS 12% "ancestral caucasian" by Martin Sikora.

Anonymous said...

mary said...
"West Eurasian is an ancestry branch that goes back to Kostenk1 14"

No one goes back to Kostenki 14 at all.

Samuel Andrews said...

@mary, Yeah there definitly is shared ancestry between Mesolithic Europe & Middle East and Kostinki is distantly apart of this shared lineage. Even, ANE is partially apart of this shared lineage but even more distantly. Its best attested by mitochondrial haplogrup U. They all share a Upper Paleolithic ancestor, who lived sometimes between 50,000-30,000 years ago.

But, the closeness between Europeans & Southwest Asians today, is mainly because Europeans are 50%+ Mesolithic Middle Eastern. Modern Europeans are a lot closer to modern Middle Easterners, than Mesolithic Europeans were to Mesolithic Middle Easterners. Hence, the main reason looking at modern DNA, people put Europeans and Middle Easterners under the same category is because Europeans are part Middle Eastern.

Vladimir said...

Another article: "Human Genomic Diversity Where the Mediterranean Joins the Atlantic", Candela L Hernández ... https://academic.oup.com/mbe/article/37/4/1041/5670533

Vladimir said...

Another article: "Population History and Gene Divergence in Native Mexicans Inferred from 76 Human Exomes", María C Ávila-Arcos ... https://academic.oup.com/mbe/article/37/4/994/5652085

gamerz_J said...

@mary

Right, I think ANE are a bit more West-Eurasian shifted than Yana.
But that is one of various different models.

zardos said...

@Samuel: That's not entirely true, wait for more data, especially from Dzudzuana as a root branch. The relationship of the West Eurasian branches is older and was closer even before the great panmixture of during the Neolithic and Bronze Age expansions. However, Basal Eurasian was more a group apart most likely, but we still haven't enough of these.

Anonymous said...


As expected, the Slab graves people were not related to the Deerstone people and to anyone else, he is an alien and has no descendants in the form of nations, he was dissolved in the population of Northern China.

Davidski said...

OK, here are the Global25 coords for the relatively high coverage individuals from this preprint...

Scaled

https://drive.google.com/open?id=1ixQrFPZII-LL4_z-16TQEwrDUqDB-sNC

Original

https://drive.google.com/open?id=1GgvTverqeKYdV_zzU4pWj762Pbq_CnnU

I'll write something up about these samples this weekend.

Luuk said...

There seems to be a lot of individuals from the Jeong et al study from which the haplogroups are not called, you can see the individuals marked in red in table 3. Does anyone know if it is possible to manually analyze and call the haplogroups of these individuals?

Leper said...

@Davidski. I can't see the Salar and Yugur samples from the article. You didn't add them?

Mr. J said...

@Unknown,

The samples marked in red from the Jeong paper are very likely contaminated.

Luuk said...

@Mr. J
Table 4 mentions there are 9 samples which have Human DNA contamination (>5% mt or X, or uncertain sex).
The rest of the excluded samples marked in red, were excluded because of Insufficient human DNA on shotgun screen (<0.1% human) or Insufficient human DNA after 1240K capture (<10K SNPs). So with different software it could be at least possible to define the base haplogroup.

Mr. J said...

@Unknown,

Ahhh ok thanks. Indeed it would be interesting to see the base haplogroups of those samples marked in red if possible then.

I can't get on my desktop computer right now to check but could you provide the list of the 9 samples that are contaminated here please?

Luuk said...

The ones unde Human DNA contamination (>5% mt or X, or uncertain sex): ARS004,BAA001,BRL003,CHN013,EME003,GUN003,KHO003,SKT011,TAV007

And of the 225 new individuals analyzed, 18 failed to yield sufficient human DNA (>0.1%) on shotgun screening (Table S4) and a further 6 individuals failed to yield at least 10,000 SNPs after DNA capture (Table S4). Both were excluded from downstream population genetic analysis.

Also in the list of Datasets passing quality filters there are some haplogroup results like CT.

Mr. J said...

@Unknown,

Ok great thanks.

andrew said...

"So there were 3 Yangtze River farmer language families?" Definitely. Long accepted in linguistic circles. But, at least a couple may be branches of a larger family.

Anonymous said...


The "Chemurchek" people are some kind of blown minds, in mito- and Y-haplogroups they are Afanasievo, but in autosomes they are more like Botaians.

capra internetensis said...

@Archi

AT-590B from Yashiin hodoo Chemurchek in both papers: IAG001 in Jeong et al, I12978 in Wang et al. But AT-628 from Kumdii Govi is listed as Chemurchek KUM001 in one, but Mongolia_EBA_1_Ulgii I13180 in the other. Then Wang et al have Chemurchek I12957, and Jeong et al have Chemurchek YAG001 (both from Yashiin hodoo as well), neither of which seems to be in the other paper.

In the Jeong et al paper Chemurchek are modelled as about 60% Botai + 40% Turan, while in the Wang et al paper they come out as about 50% Afanasievo + 40% WSHG + 10% Mongolia N. I don't know if this is a difference in methods or in which samples are pooled together. Jeong et al only tried Afanasievo + Botai and Afanasievo + Okunevo for models with Afanasievo.

I tried replicating these models in scaled G25:
I12957 - 62% Afanasievo, 28% Tyumen, 9% Mongolia N - dist 5.2366%
I12957 - 63% Botai, 37% Bustan BA - dist 5.2481%
I12978 - 51% Afanasievo, 35% Tyumen, 13% Mongolia N - dist 2.8479%
I12978 - 72% Botai, 28% Bustan BA - 5.1823%
I13180 - 73% Mongolia N, 18% Tyumen, 9% Afanasievo - 2.5616%
I13180 - 95% Botai, 5% Bustan BA - dist 32.1549%

The Wang et al Chemurchek model seems okay for I12978, the rest suck.

Anonymous said...

@ capra internetensis

The main thing there is not in the models, but in the complete contradiction between haplogroups and autosomes. The presence of Afanasievians does not say anything, you see, Afanasievan's modeling necessarily has Yamnaya, because they are close relatives. But the Afanasievians also absorbed the substrates, so the presence of something in common with the Chemurchek people who lived there does not say anything. But in the Chemurchekans there are not Yamnaya.

Models can be obtained much more accurate.

Target: MNG_Chemurchek_EBA_2:I12957
Distance: 1.1664% / 0.01166367
40.8 RUS_Afanasievo
38.2 KAZ_Botai
21.0 TKM_Gonur1_BA

Target: MNG_Chemurchek_EBA_2:I12978
Distance: 1.2380% / 0.01237992
49.8 RUS_Afanasievo
41.0 KAZ_Botai
8.4 TKM_Gonur3_BA
0.8 TKM_Gonur1_BA

Target: MNG_Afanasievo_1:I6222
Distance: 1.0968% / 0.01096830
46.8 RUS_Afanasievo
36.6 Yamnaya_RUS_Samara
13.2 Yamnaya_UKR_Ozera_o
3.4 TKM_Gonur1_BA

Target: MNG_Afanasievo_1:I6221
Distance: 1.0030% / 0.01003046
57.0 Yamnaya_RUS_Samara
24.8 RUS_Afanasievo
4.6 Yamnaya_UKR_Ozera_o
4.4 TKM_Gonur1_BA
4.2 KAZ_Botai
3.6 Yamnaya_UKR
1.4 TKM_Gonur2_BA

Adding Kazakhstan Yamnayans doesn't improve the case except this one.

Target: MNG_Chemurchek_EBA_2:I12978
Distance: 1.1976% / 0.01197567
41.0 KAZ_Botai
25.8 Yamnaya_KAZ_Karagash
23.8 RUS_Afanasievo
6.8 TKM_Gonur3_BA
2.0 TKM_Gonur1_BA
0.6 TKM_Gonur2_BA

Davidski said...

@Leper

The Salar and Yugur are modern samples.

I'll eventually update all of the Global25 datasheets with these new samples. But it might take a couple of days.

Leper said...

@Davidski

I know, I realized that you only added ancient ones for now. Thanks.

Is the dataset of the new Eurasia article accessible? Will you add them aswell?

Link: https://www.biorxiv.org/content/10.1101/2020.03.25.008078v1.supplementary-material

Davidski said...

@Leper

Please post the link if it's available and I'll add them.

Tetris said...

Chemurchek was said to be closer to Dali_EBA.

Target: MNG_Chemurchek_EBA_2:I12957
Distance: 1.6453% / 0.01645257
34.6 KAZ_Dali_EBA
27.2 RUS_Afanasievo
17.6 TKM_Gonur1_BA
17.0 KAZ_Botai
2.4 RUS_Shamanka_N
1.2 IRN_Shahr_I_Sokhta_BA2
0.0 KAZ_Kumsay_EBA
0.0 RUS_Sosonivoy_HG

Target: MNG_Chemurchek_EBA_2:I12978
Distance: 1.6380% / 0.01637982
31.6 RUS_Afanasievo
17.4 KAZ_Botai
16.2 RUS_Sosonivoy_HG
14.8 KAZ_Dali_EBA
9.6 RUS_Shamanka_N
7.8 TKM_Gonur1_BA
2.6 IRN_Shahr_I_Sokhta_BA2
0.0 KAZ_Kumsay_EBA

Anonymous said...

@Tetris

Dali_EBA is in principle simultaneous with Chemurchekians so it's not clear who influenced who, but it's not much different from Botaians. The Botaians are exactly an ancestral population without influence Yamnayans.



Davidski said...

@Tetris

You're overfitting your models with too many similar populations, and breaking up the Dali_EBA-related ancestry into several different streams.

Davidski said...

Here are the distances...

Distance to: MNG_Chemurchek_EBA_2:I12957
0.05331791 RUS_Steppe_Maykop
0.05439808 KAZ_Mereke_MBA
0.05609068 RUS_Lola
0.05903345 KAZ_Kumsay_EBA
0.06710663 RUS_Potapovka_MLBA_o
0.07265112 RUS_Srubnaya_MLBA_o
0.07291195 KAZ_Oy_Dzhaylau_MLBA_o
0.07644157 KAZ_Kairan_MLBA_o
0.07673547 KAZ_Dali_EBA
0.07852790 KAZ_Maitan_MLBA_Alakul_o

Distance to: MNG_Chemurchek_EBA_2:I12978
0.05870263 KAZ_Dali_EBA
0.06381126 KAZ_Mereke_MBA
0.06573005 KAZ_Kairan_MLBA_o
0.06641366 KAZ_Kumsay_EBA
0.06816372 RUS_Steppe_Maykop
0.06892668 KAZ_Oy_Dzhaylau_MLBA_o
0.07814680 KAZ_Zevakinskiy_LBA
0.08420269 RUS_Lola
0.08454189 RUS_Sintashta_MLBA_o1
0.08885321 RUS_Srubnaya_MLBA_o

Leper said...

@Davidski

Data is here: https://edmond.mpdl.mpg.de/imeji/collection/cogJX_dvd94nNI33

Rob said...

@ Archie

''As expected, the Slab graves people were not related to the Deerstone people and to anyone else, he is an alien and has no descendants in the form of nations, he was dissolved in the population of Northern China.''

Hhm that's not really true. Slab graves share recent ancestry with DKSC, with the latter having elevated western-herder ancestry. Both complexes are dominate by Y-DNA Q1a

Davidski said...

@Leper

No items found!

I'm guessing the files will be up when the preprint is published.

Leper said...

@Davidski

Yep, just realized that the links they provided are empty.



DATA AND CODE AVAILABILITY


All newly reported sequencing data are available from the European Nucleotide Archive, accession number PRJEB35748. 1240K genotype data are available on the Edmond Max Planck Data Repository under the link:

https://edmond.mpdl.mpg.de/imeji/collection/2ZJSw35ZTTa18jEo."

Anonymous said...

@Rob "Hhm that's not really true. Slab graves share recent ancestry with DKSC, with the latter having elevated western-herder ancestry. Both complexes are dominate by Y-DNA Q1a"

As always, you write anti-scientific nonsense without knowing anything and without seeing anything. Archaeologists and anthropologists have long proved that they are absolutely not related. They lived in a very close neighborhood, mixing, but had completely different lifestyles and completely different origins.

You're lie, your goal is to write the untruth.. DSKC are previous Q1a2a (+neighborhood R1a...,N1c...), Slab graves (with Ulaanzuukh Slab graves) are new Q1a1(a1)! Autosomatically, they're completely different, too.


Rob said...

@ Archi
Yes, theyre different groups, but they still share ancestry which is as recent as EBA.
There's little point in explaining complex GW-data patterns to you (e.g. you use Ukraine Ozera in modelling Mongolia), so simply note the uniparentals.
Y-DNA Q is missing in Mongola Neolithic, Baikal Neolithic & Amur basin.
All Mongol BA groups, & subsequent, possess Q1a lineages

Rob said...

Moreover your claim (he is an alien and has no descendants in the form of nations, he was dissolved in the population of Northern China.''') is contradicted by the findings in Jeong et al write - ''During the EIA, the Slab Grave culture expanded northwards, often disrupting and uprooting former DSKC graves in their path (Honeychurch, 2015), and it ultimately reached as far north as the eastern Baikal region''

Anonymous said...

@Rob "but they still share ancestry which is as recent as EBA.
There's little point in explaining complex GW-data patterns to you (e.g. you use Ukraine Ozera in modelling Mongolia), so simply note the uniparentals.
Y-DNA Q is missing in Mongola Neolithic, Baikal Neolithic & Amur basin.
All Mongol BA groups, & subsequent, possess Q1a lineages"

The inability to distinguish Q1a1 from Q1a2 is already a clinic.
All people are A, so Mongols are not different from Swiss, they are the same people)))))) there is no difference between Hindus and Spaniards in general, they are R1!

There's no point in explaining anything to you at all, you'll never understand it anyway, for you thousands of years it's nothing, the lifetimes of Okunevo and Slabgraves are the same, and Chukchs, Han and Okunevo are the same.


"Moreover your claim (he is an alien and has no descendants in the form of nations, he was dissolved in the population of Northern China.''') is contradicted by the findings in Jeong et al write - ''During the EIA, the Slab Grave culture expanded northwards, often disrupting and uprooting former DSKC graves in their path (Honeychurch, 2015), and it ultimately reached as far north as the eastern Baikal region''"

You can't read! Jeong doesn't write anything about what happened to these people after this culture died, he only writes about the time of its existence, about its expansion. So there is no contradiction, it exists only in your head because you don't understand what it's about.



Anonymous said...


Separation time Q1a1 and Q1a2 is 26000 BP.

Rob said...

That's a genealogical split, not physical.
The fact remains that both Q1a1 and Q1a2 only seem to appear after the eneolithic. and its most likely they came from a common region which harboued both of them
You're really not very good at this.

Anonymous said...

@Rob "That's a genealogical split, not physical.
The fact remains that both Q1a1 and Q1a2 only seem to appear after the eneolithic. and its most likely they came from a common region which harboued both of them
You're really not very good at this."

It is a fact that this Anti-scientific nonsense contradicts all the data, complete untruth from a man who again wanted to deceive taking from the ceiling of his shameful delusions that he had just invented.

Neolithic Ust’-Belaya Russia Ust’-Belaya II, Angara river [I7336 / Mos111] 6300 – 6210 calBP M Q1a1b

Bronze Steppe Maykop Russia Sharakhalsun 6 [SA6004 / kurgan 2, grave 18, BZNK-003/4] 3336-3105 cal BC (4500±40BP ,GIN-12401) M Q1a2

Bronze Okunevo Russia Syda 5, Tumen, IPDN 6/H, Kurgan 3 [RISE718] 2573-2348 cal BC (3964 ± 31 BP, UBA-31600) Q1a2a1c-L330


It is a typical shameful Eurocentrism to consider everyone who lives in East Asia as a single population, who for tens of thousands of years had no history of their own, they say: it only changed something in Europe. Europe is also a common region.


Anonymous said...

Mesolithic Russia Duvanni Yar, Kolyma river, east Siberia [Kolyma1] 9668-9904 cal BP (8770 ± 27) M Q1a1a

Mesolithic Latvia Zvejnieki [I4550 / ZVEJ3] 6000-5100 BCE M Q1a2

LBA / IA Pakistan Swat Valley, Loebanr [I12134 / Grave 136, Individual 1 (B) in a double burial , 543 ] 1000-800 BCE M Q1a1b1

Mr. J said...

Unfortunately the authors of the Jeong study do not provide very detailed descriptions of the grave sites that the samples in that paper come from.

Unknown said...

So are y dans K2, P, Q1a, Q1b, R1b-M73 and R2 originally West Eurasian? If so how did Q1a and R1b-M73 become so big in East Eurasians?

Matt said...

FrankN: Apparently, getting aDNA from China comes with a price. I can understand that no senior staff member from Harvard or MPI Jena (except Wang, of course) wanted to see his name in the list of authors.

Dave (and others) - once the aDNA is out, there is a lot of work ahead in trying to better understand that Wuzhuangguoliang amalgam.


Don't be so sure that we can learn anything from this; it seems like the price of getting adna from China is that they do it themselves in a lab with poor coverage of variants and poor overlap. I'm not sure anyone will be able to do much with these data.

vAsiSTha said...

Is wang suggesting that tocharian entered tarim in late 4thmill bce from analyzing 400bce samples? hahahahaha.

anyway, this region is important and good that we have some samples. will merge the dataset and run some tests.

Simon_W said...

A handful of observations making use of the Olmo di Nogara samples:

Explaining Etruscans as a two-way mixture of Olmo di Nogara and the Latin-like Villanovan sample yields a very poor fit:

Target: ITA_Etruscan
Distance: 4.2346% / 0.04234587

74.6 ITA_Villanovan
25.4 ITA_Olmo_di_Nogara_MBA

Adding a lot from the HRV_IA-like Protovillanovan sample greatly improves the fit:

Target: ITA_Etruscan
Distance: 1.8647% / 0.01864681

55.4 ITA_Proto-Villanovan
28.6 ITA_Villanovan
16.0 ITA_Olmo_di_Nogara_MBA

However, when I'm trying to model the Latins as a mix of Proto-Villanovan, Olmo di Nogara and central Italian Chalcolithic samples, I again obtain a very poor fit:

Target: ITA_Rome_Latini_IA
Distance: 4.6348% / 0.04634833

62.8 ITA_Proto-Villanovan
37.2 ITA_Monte_San_Biagio_CA

I've tried various additional populations to improve the fit, that make some sense archaeologically: DEU_Lech_BBC, DEU_Lech_EBA, DEU_Lech_MBA, HUN_LBA and Bell_Beaker_ITA. DEU_Lech_EBA worked best, by far. But tell me if you find something that works even better.

Target: ITA_Rome_Latini_IA
Distance: 2.9947% / 0.02994696

51.4 DEU_Lech_EBA
22.8 ITA_Grotta_Continenza_CA
14.4 ITA_Olmo_di_Nogara_MBA
11.4 ITA_Monte_San_Biagio_CA

It looks like the Terramare culture was neither the source of the Etruscans nor of the Italics. It looks like a dead end, at least for now. And surprisingly, it seems likely that the Etruscans and the Italics arrived in central Italy at the same time, the Final Bronze Age, with the Proto-Villanovan culture. The Italics coming more from a northern, post-Bell Beaker South German source and the Etruscans from an eastern HRV_IA-like one.

Anonymous said...

@Simon_W
You combine the samples too much, each of them is unique and has a different fate, but the Etruscans do not go back to (P-)Villanova in any way.

Target: ITA_Etruscan:RMPR474b J2b2a-M12 H
Distance: 0.6830% / 0.00682953
42.8 Bell_Beaker_CZE
25.4 Levant_ISR_Ashkelon_IA1
14.4 Bell_Beaker_ITA
7.6 GRC_Mycenaean
7.0 ITA_Remedello_BA
2.8 Levant_PPNB
0.0 ITA_Proto-Villanovan
0.0 ITA_Villanovan
0.0 Bell_Beaker_ITA_o
0.0 GRC_Minoan_Lassithi
0.0 ITA_Sardinia_C
0.0 ITA_Sardinia_C_o
0.0 ITA_Sardinia_EBA
0.0 ITA_Sardinia_ECA
0.0 Levant_Canaanite_MBA
0.0 Levant_ISR_Ashkelon_IA2
0.0 Levant_ISR_Ashkelon_LBA
0.0 Levant_ISR_C
0.0 Levant_ISR_MLBA
0.0 Levant_JOR_EBA
0.0 Levant_Natufian
0.0 Levant_PPNC
0.0 MAR_Taforalt

Target: ITA_Etruscan_o:RMPR475b T2b32
Distance: 0.6210% / 0.00620998
29.6 Bell_Beaker_CZE
23.0 GRC_Minoan_Lassithi
13.6 ITA_Sardinia_C
8.2 MAR_Taforalt
7.8 Levant_PPNB
5.2 ITA_Sardinia_C_o
5.0 ITA_Sardinia_EBA
4.0 GRC_Mycenaean
3.2 ITA_Remedello_BA
0.4 Levant_ISR_Ashkelon_IA1
0.0 Bell_Beaker_ITA
0.0 Bell_Beaker_ITA_o
0.0 ITA_Proto-Villanovan
0.0 ITA_Villanovan
0.0 ITA_Sardinia_ECA
0.0 Levant_Canaanite_MBA
0.0 Levant_ISR_Ashkelon_IA2
0.0 Levant_ISR_Ashkelon_LBA
0.0 Levant_ISR_C
0.0 Levant_ISR_MLBA
0.0 Levant_JOR_EBA
0.0 Levant_Natufian
0.0 Levant_PPNC

Target: ITA_Etruscan:RMPR473
Distance: 0.9143% / 0.00914338
31.8 Bell_Beaker_CZE
26.6 GRC_Mycenaean
17.6 ITA_Sardinia_EBA
13.8 ITA_Remedello_BA
3.4 ITA_Proto-Villanovan
2.2 Levant_Natufian
1.8 Bell_Beaker_ITA
1.6 GRC_Minoan_Lassithi
1.2 Levant_ISR_C
0.0 ITA_Villanovan
0.0 Bell_Beaker_ITA_o
0.0 ITA_Sardinia_C
0.0 ITA_Sardinia_C_o
0.0 ITA_Sardinia_ECA
0.0 Levant_Canaanite_MBA
0.0 Levant_ISR_Ashkelon_IA1
0.0 Levant_ISR_Ashkelon_IA2
0.0 Levant_ISR_Ashkelon_LBA
0.0 Levant_ISR_MLBA
0.0 Levant_JOR_EBA
0.0 Levant_PPNB
0.0 Levant_PPNC
0.0 MAR_Taforalt

Anonymous said...

To compare, Rutuli has many Proto-Villanova.

Target: ITA_Ardea_Latini_IA:RMPR851 R1b1a2a1a-P311 H2a
Distance: 0.6947% / 0.00694730
37.6 Bell_Beaker_CZE
19.2 ITA_Proto-Villanovan
16.2 ITA_Sardinia_C
13.2 GRC_Minoan_Lassithi
11.2 ITA_Sardinia_EBA
1.6 GRC_Mycenaean
1.0 ITA_Remedello_BA
0.0 Bell_Beaker_ITA
0.0 Bell_Beaker_ITA_o
0.0 ITA_Villanovan
0.0 ITA_Sardinia_C_o
0.0 ITA_Sardinia_ECA
0.0 Levant_Canaanite_MBA
0.0 Levant_ISR_Ashkelon_IA1
0.0 Levant_ISR_Ashkelon_IA2
0.0 Levant_ISR_Ashkelon_LBA
0.0 Levant_ISR_C
0.0 Levant_ISR_MLBA
0.0 Levant_JOR_EBA
0.0 Levant_Natufian
0.0 Levant_PPNB
0.0 Levant_PPNC
0.0 MAR_Taforalt

And Rutuli's Pelasgan.

Target: ITA_Ardea_Latini_IA_o:RMPR850 T1a1a L208 T2c1f
Distance: 1.4360% / 0.01436030
36.2 GRC_Mycenaean
24.0 Levant_ISR_Ashkelon_LBA
18.8 Bell_Beaker_CZE
14.4 Levant_ISR_Ashkelon_IA1
4.6 Levant_ISR_C
2.0 Levant_ISR_Ashkelon_IA2
0.0 Bell_Beaker_ITA
0.0 Bell_Beaker_ITA_o
0.0 GRC_Minoan_Lassithi
0.0 ITA_Proto-Villanovan
0.0 ITA_Remedello_BA
0.0 ITA_Sardinia_C
0.0 ITA_Sardinia_C_o
0.0 ITA_Sardinia_EBA
0.0 ITA_Sardinia_ECA
0.0 ITA_Villanovan
0.0 Levant_Canaanite_MBA
0.0 Levant_ISR_MLBA
0.0 Levant_JOR_EBA
0.0 Levant_Natufian
0.0 Levant_PPNB
0.0 Levant_PPNC
0.0 MAR_Taforalt

It's all done on the same set of sources to see the difference.

Samuel Andrews said...

@Unknown,
"So are y dans K2, P, Q1a, Q1b, R1b-M73 and R2 originally West Eurasian? If so how did Q1a and R1b-M73 become so big in East Eurasians?"

Y DNA Q, R1b M73 are found in Central Asia, Siberia. But not in East Asia (China, Japan, etc). So they were never seen seen exactly as East Eurasian (Asian) haplogroups.

They both come ultimately from ANE so aren't exactly West Eurasian. They're ancient Central Asian not West Eurasian.

Arza said...

@ Samuel Andrews
Also, he did have Steppe ancestry, 10%

But rather from the background, acquired indirectly.

got his R1b M269 ultimatly from Bell beaker

I wouldn't be so sure.

9309_Oss
numsnps used: 247766

Russia_EBA_Yamnaya_Samara / Italy_N.SG / Hungary_LateC_Protoboleraz / Armenia_C

fixed pat wt dof chisq tail prob
0000 0 9 3.389 0.946865 0.025 0.077 0.772 0.126
0001 1 10 5.219 0.876071 0.102 0.244 0.654 -0.000
0010 1 10 21.241 0.0194716 0.317 0.948 -0.000 -0.265 inf
0100 1 10 3.390 0.97071 -0.008 -0.000 0.834 0.174 inf
1000 1 10 3.558 0.965094 -0.000 0.025 0.816 0.159
0011 2 11 27.134 0.00438576 0.193 0.807 0.000 0.000
0101 2 11 6.961 0.802207 0.065 0.000 0.935 0.000
0110 2 11 181.998 0 -0.393 0.000 -0.000 1.393 inf
1001 2 11 8.873 0.633658 0.000 0.050 0.950 -0.000
1010 2 11 52.141 2.57073e-07 0.000 0.759 0.000 0.241
1100 2 11 3.328 0.985647 -0.000 0.000 0.843 0.157
0111 3 12 695.624 0 1.000 -0.000 -0.000 -0.000
1011 3 12 70.188 2.95359e-10 0.000 1.000 0.000 -0.000
1101 3 12 9.104 0.694043 0.000 -0.000 1.000 -0.000
1110 3 12 217.131 0 0.000 -0.000 0.000 1.000

vAsiSTha said...

Yes Rob, will need to spend time on these samples.

interesting bit from Jeong preprint.
Chemurchek seems similar to Dali_EBA, ie Botai + Geoksyur/Sarazm/Parkhai

"We find that these Chemurchek individuals (Chemurchek_Altai) are
genetically similar to Dali_EBA (Fig. 3a), a contemporaneous individual from eastern Kazakhstan (Narasimhan et al., 2019). The genetic profiles of both the Yagshiin Huduu and Dali_EBA individuals are well fitted by two-way admixture models with Botai (60-78%) and groups with ancient Iranian-related ancestry, such as Gonur1_BA from Gonur Tepe, a key EBA site of the Bactria-Margiana Archaeological Complex (BMAC) (22-40%; Table S15; Fig. 3a), and ancient individuals from post-BMAC Bronze Age sites in southeastern Uzbekistan (Table S15). "

Samuel Andrews said...

@Avisthata, Yeah, In David's G25 PCA, Chemurchek are basically the same as Dali_EBA.

vAsiSTha said...

@samuel
thanks, didnt know. what is the label name for chemurchek in G25?

Samuel Andrews said...

@vAsiATha,

Their ID is. All the Mongolia samples have MNG at the begining. Actually, Chemurcheck comes out 80% Dali_EBA 20% Sintashta. Either, way looks like a Central Asian pop similar to Dali_EBA who mixed with Andronovo then migrated into Mongolia.

MNG_Chemurchek_EBA

Carlos Aramayo said...

@Dear all

This is a recently published paper in Nature, even though it´s a study in South Asia with current DNA (not aDNA) claiming that:

"...In the allele frequency based analysis, the Kols exhibited a closer genetic affinity with the Indo-European scheduled castes and tribal populations, rather than with Austroasiatic or Dravidian populations..."

https://tinyurl.com/tnudh95

Any comments?

capra internetensis said...

@Sam

Why use Sintashta? Too late, and Afanasievo works better in most models AFAICT.

Chemurchek I12978 scaled:
70% Dali EBA, 23% Sintashta, 7% Mongolia N - dist 2.5610%
56% Dali EBA, 35% Afanasievo, 9% Mongolia N - dist 2.1584%
45% Botai, 39% Afanasievo, 9% Mongolia N, 8% Bustan En - dist 2.1270%

Chemurchek I12957 scaled:
69% Dali EBA, 16% Bustan En, 15% Sintashta - 2.4422%
64% Dali EBA, 21% Afanasievo, 15% Bustan En - dist 2.3379%
44% Botai, 34% Afanasievo, 20% Bustan En, 2% Mongolia N - 2.6420%

Ulgii EBA I13180 scaled:
70% Mongolia N, 30% Dali EBA - 2.1524%

Can get decent models with Steppe Maykop too, and can swap out various sources of Iran/Turan and ANE.

Samuel Andrews said...

You're right. I just did a quick model. I didn't test with Afanasievo and didn't check the date of the samples.

Rob said...

Chemurchek & Dali_EBA are contemperaneous. My preferred way to look at it is with a slightly more 'distal' set-up, using relatively differentiated Source Pops. (e.g. Afansievo, Sintashta, WSHG, BMAC/ Bustan, Baikal_N, etc), accepting a slightly worse fit, to understand the ancestry streams coalescing

Analysing Chemurchek, Dali_EBA, Zevakinski shows that all these groups harbour WSHG, Afansievo, Andronovo, Bustan, but in varying degrees.

Dali_EBA
WSHG ~ 70%
Afanasievo ~ 14%
BMAC ~ 16%
Andronovo ~ 0%
(d2.8227%)

Chemerchuk:I12957
WSHG 45%
Afanasievo 30%
BMAC 25%
Andronovo 0%

Zevakisnkij LBA:I3753
WSHG 25%
Afanasievo 25%
BMAC 5%
Andronovo 35%
Lokomotiv 10%

Arza said...

Smyadovo I2181 is M269. YFull dates M269 to "TMRCA 6400 ybp". I2181 is dated to 6453 years ago.

4085 I2181_published I2181 21 tooth (molar) 2018 MathiesonNature2018 Gaydarska, Bisserka 6453 4550-4455 calBCE (5680±30 BP, Beta-432803) Bulgaria_C_published Smyadovo Bulgaria 43.05777778 26.98361111 1240K 4 0.063 71542 M R .. HV15 .. .. .. .. .. .. .. half,half,half,half S2181.E1.L1,S2181.E1.L2,S2181.E1.L3,S2181.E1.L4 PASS (literature, mtcontam=0.968-0.985)

fastq + Yleaf =

Sample_name Hg Hg_marker Total_reads Valid_markers QC-score QC-1 QC-2 QC-3
ERR2216591 R1b1a1b R-PF6452*(xA561,A1243,L237,FGC15335,FGC5253,CTS7768,F1343,CTS11567,S259,L744,P66_3,Z16400,S1115,FGC5642,L1308,ZS5732^^,BY23092,BY9405,ZZ38,A8058,CTS1078,GG480) 142630 809 1.0 1.0 1.0 1.0

Mr. J said...

@All,

From the Jeong paper.

Information about the three Khitan individuals in this paper.

''Khitan (ca. 900-1100 CE). After the collapse of the Uyghur empire in Mongolia in 840 CE, the Khitans
of northeast China established the powerful Liao Dynasty in 916 CE. Although based in Manchuria, the
Khitans conquered and controlled the steppe of present-day Mongolia through a system of garrisons and
long walls, deporting people from other conquered regions, such as northern Korea, to Mongolia (Kradin
and Ivliev, 2008). The dissolution of the Khitan empire in 1125 CE lead to a power vacuum in Mongolia
until the rise of Chinggis Khan in the early 13th century CE. To date, very few Khitan era graves have
been found in Mongolia. The site of Ulaan Kherem II (ULA) has yielded one Khitan-era grave
(ULA001), and two Khitan-era unmarked graves of a man and woman were also discovered during the
excavation of a Xiongnu settlement at Zaan Khoshuu (ZAA) beneath an older collapsed building (Nei
Menggu zizhiqu wenwu kaogu yanjiusuo and International Institute for the Study of Nomadic
Civilizations, 2015; Ochir et al., 2016). The man, found in a pit within the pit-house, was buried in a
simple pit with a quiver and arrows. The woman, found nearby a pit-house, was buried in full dress and
placed in a supine position with her head to the northwest inside a wooden coffin, along with pottery of
the Khitan era (https://edmond.mpdl.mpg.de/imeji/col...JSw35ZTTa18jEo). These burials are significantly
different in form and structure from other Khitan burials in northern China, where the core of the empire
was located. At present, no monumental tombs of high Khitan elites have been found in Mongolia.''

The Khitan era sample ZAA005 that belongs to J2a-L212 must be the same as the man mentioned in that information above from the ZAA site that was buried in a simple pit within a pit-house with a quiver and arrows...Does that sound correct?

Samuel Andrews said...

@Arza, "Smyadovo I2181 is M269." That's awesome if true. Please explain how you came to this conclusion.

Anonymous said...

@Samuel Andrews
This is to be expected.

I2430 40 tooth (molar) 2018 MathiesonNature2018 Gaydarska, Bisserka 6448 4545-4450 calBCE (5670В±30 BP, Beta-432806) Bulgaria_C Smyadovo Bulgaria 43.05777778 26.98361111 1240K 4 1.133 617793 M R1b1a .. K1a26


Arza said...

@ Samuel Andrews
See published output from Yleaf. Searching for PF6452 on YFull may be helpful too.

Arza said...

@ Archi
Y-DNA of I2430 is reported incorrectly in this spreadsheet. It's R1b1b2.

Anonymous said...

@Arza

Correctly.

I2430 is L754 from Mathieson 2017
R1b1a:A702:10038192G->A; R1b1a:CTS8612:18109555C->A; R1b1a:FGC36:13822833G->T; R1b1a:FGC41:7900883C->A; R1b1a:L761:16773870A->G; R1b1a:PF6271:23984056G->A; R1b1:CTS2134:14193384G->A; R1b1:L278:18914441C->T; R1:CTS2565:14366723C->T;

vAsiSTha said...

Dali_EBA does not require Afanasievo in qpAdm, though upto 5% ancestry might be present

Kazakhstan_Dali_EBA

Botai: 80.5% +- 1.8%
Geoksyur_EN: 19.5% +- 1.8%
chisq: 14.016
tailprob: 0.081
result file https://pastebin.com/9kTUPtt9


For chemurchek, this is the only model that works so far. Have clubbed the 2 samples although the 2 are not too similar to each other.

Mongolia_EBA_2_Chemurchek

Botai: 41.7% +- 2.6%
Geoksyur_EN: 15.8% +- 2.6%
Russia_Afanasievo: 35.9% +- 4%
Mongolia_N_North: 6.7% +- 1.1%
chisq: 9.82
tailprob: 0.132
result file https://pastebin.com/jU3jPfMp

vAsiSTha said...

Boisman_MN is a great neolithic label to use as east asian right pop in qpAdm as it is 1240K captured. shamanka samples were shotgun captured.

Samuel Andrews said...

@Arzra, I've read your post a second time. "ERR2216591 R1b1a1b R-PF6452*"....That isn't R1b M269. It is PF6452.

vAsiSTha said...

Another qpAdm model for Chemurchek, with Okunevo as proximal east asian source rather than mongolia neolithic.

left pops:
Mongolia_EBA_2_Chemurchek

Botai: 32.2% +- 3.7
Geoksyur_EN: 15.1% +- 3.7
Russia_Afanasievo: 31.9% +- 2.6
Russia_BA_Okunevo.SG: 20.9% +- 3.5
chisq: 7.51
tailprob: 0.2755
result file https://pastebin.com/aX5tV4Hz

Arza said...

@ Archi

It seems that this "R1b1a" comes from some (very) old tree. Current one looks like this:

I2430, derived only
R1b:CTS2134
R1b:L278
R1b1:A702
R1b1:CTS8612
R1b1:FGC36
R1b1:FGC41
R1b1:L761
R1b1:PF6271
R1b1a1b1a1a2a5a~:F3867.2
R1b1b:PF6289
R1b1b:PF6292
R1b1b:PF6304
R1b1b:PF6312
R1b1b:PF6344
R1b1b:PF6376
R1b1b2:FGC21015

https://www.yfull.com/tree/R-Y7777/

Arza said...

@ JuanRivera
Post a list of quadruples with IDs from the 42.4 dataset and I'll run them.

Anonymous said...

@Arza

Mathieson 2017 => ISOGG 2016

Everybody will use Mathieson's data, because it's not clear where you got these results at all.

I2430 is L754 from Mathieson 2017:
R1b1a:A702:10038192G->A; R1b1a:CTS8612:18109555C->A; R1b1a:FGC36:13822833G->T; R1b1a:FGC41:7900883C->A; R1b1a:L761:16773870A->G; R1b1a:PF6271:23984056G->A; R1b1:CTS2134:14193384G->A; R1b1:L278:18914441C->T; R1:CTS2565:14366723C->T;

The rebuttal can only come from YFull.

Arza said...

I'm using tree from Yleaf and it's based on ISOGG 2019 tree.

Rob said...

@ vasistha
Can you do a run for shirenzigou & zevakinskij BA

@ mr J
Seems some Turkics have more Turan admixture and YDNA J
So it’s from before they became culturally iranicized

@ Sam / Arza
I was expecting smadjovo to be V3616 like khvalynsk

Mr. J said...

@Rob

Are you talking about modern Turkics or the ancient Turkic samples with Y-DNA J and Turan admixture?

vAsiSTha said...

for Zevakinskiy_ba I3770, this is the best i got so far. cant seem to improve that tailprob.

Sintashta_mlba: 85%
Okunevo: 15%
tailprob: 0.035

will work on china_ia later.

Arza said...

@ JuanRivera

I2181 Dstats
https://pastebin.com/1T02Gxxp

Arza said...

@ Rob
Why like Khvalynsk? It has something to do with Usatovo?

Davidski said...

What's with the "Unknown" nicks in this thread?

It's against the rules.

Jatt_Scythian said...

Sorry that last comment was mine. I used the wrong email address. Any chance you can put my comment back up?

Rob said...

@ MrJ
Yes I mean TUK001 & the Uyghurs

@ Arza
Because, at least in some older literature, it was assumed that it was groups from Khvalynsk which were arriving to the Balkans. But whatever the case, it seems at present that the early 'steppe eneolithic' signature involved R1b-V3616

Jatt_Scythian said...

Wasn't posting much when the Yana aDNA results came out but does this paper shine light on the origins of ANE/ANS/WSHG populations, their range and phenotypes? Also is R2 associated with anything in this? I believe it was absent in this paper. Does this mean TOcharians made it as far as Mongolia/Gansu?

Jatt_Scythian said...

Also really cool to see R1b-L51 and R1a-Z283 in Asia. Wonder why they weren't more successful.

Never would have thought the Xiongu were so western.

Mr. J said...

@Rob and Davidski

I already posted this but I have not received any answers or opinions on it. I am curious as to what you two think about this. I don't want to fill this comment section up with repeats though so I won't ask it again after this.


From the Jeong paper.

Information about the three Khitan individuals in this paper.

''Khitan (ca. 900-1100 CE). After the collapse of the Uyghur empire in Mongolia in 840 CE, the Khitans
of northeast China established the powerful Liao Dynasty in 916 CE. Although based in Manchuria, the
Khitans conquered and controlled the steppe of present-day Mongolia through a system of garrisons and
long walls, deporting people from other conquered regions, such as northern Korea, to Mongolia (Kradin
and Ivliev, 2008). The dissolution of the Khitan empire in 1125 CE lead to a power vacuum in Mongolia
until the rise of Chinggis Khan in the early 13th century CE. To date, very few Khitan era graves have
been found in Mongolia. The site of Ulaan Kherem II (ULA) has yielded one Khitan-era grave
(ULA001), and two Khitan-era unmarked graves of a man and woman were also discovered during the
excavation of a Xiongnu settlement at Zaan Khoshuu (ZAA) beneath an older collapsed building (Nei
Menggu zizhiqu wenwu kaogu yanjiusuo and International Institute for the Study of Nomadic
Civilizations, 2015; Ochir et al., 2016). The man, found in a pit within the pit-house, was buried in a
simple pit with a quiver and arrows. The woman, found nearby a pit-house, was buried in full dress and
placed in a supine position with her head to the northwest inside a wooden coffin, along with pottery of
the Khitan era (https://edmond.mpdl.mpg.de/imeji/col...JSw35ZTTa18jEo). These burials are significantly
different in form and structure from other Khitan burials in northern China, where the core of the empire
was located. At present, no monumental tombs of high Khitan elites have been found in Mongolia.''

The Khitan era sample ZAA005 that belongs to J2a-L212 must be the same as the man mentioned in that information above from the ZAA site that was buried in a simple pit within a pit-house with a quiver and arrows...Does that sound correct?

Jatt_Scythian said...

Also with all these samples why do we still not have Tarim Basin samples>

Samuel Andrews said...

@Jatt-Scythian, I agree. Current knowledge on Tarim Mummies is still stuck in 2006.

With the ancient DNA we have now, looking at their Tarim Basin mtDNA it looks Siberian/Lake Bailkal/Mongolia Neolithic. As we know the Y DNA is R1a M417, I think we can assume Z93 (I don't believe early reports which say it is negative for Z93).

I read a little bit about the archaeology of the Xiaohe cemetary which dates 1800-1500 BC contemporary to Andronovo.

But, no experts try to compare Xiaohe cemetary to Andronovo cemeteries which is frustrating. Unlike older Steppe cultures, Andronovo/Sintashta did make wooden building structures and religious buildings, which might be similar to Xiaohe cemetary.

Anonymous said...

@Samuel Andrews
"I think we can assume Z93 (I don't believe early reports which say it is negative for Z93)."

Z93- is official information from the authors of the study, there is no reason not to trust it. It's official. So you can't think like that. It's not Andronovo.

Generally for some reason, old samples are not redetermined by new methods, and the Chinese for some reason prefer to test Mongolia and Manchuria.

Arza said...

@ Archi
Generally for some reason, old samples are not redetermined by new methods

They are... but they are being kept in secret for years.

and the Chinese for some reason prefer to test Mongolia and Manchuria.

R1a(xZ93) aka "ancient Russians" being indigenous to Northern China is probably a big no-no.

Rob said...

@ Juan

''The Bulgarian M269 may have UKR_N and/or Piedmont. Did a G25 Maps run''

What's a G25 maps run ? I didnt see him on the G25 list

Iggy said...

@Davidski I was doing some research on the racial makeup of ANE, which brought me here. Your blog is fascinating.

I was wondering what your conclusion was regarding the racial makeup of ANE? I've seen it said that they were East Asian in appearance? I'm somewhat of a novice, so please excuse the potential ignorance of this question.

I'd greatly appreciate input from anyone else with knowledge on the subject, so this question really goes out to anyone and everyone reading this.

Samuel Andrews said...

Harvard geneticists have bowed down to Indian nationalists. Maybe the do bow down to Chinese government.

But Chinese government has no reason to worry about Tarim Mummies. Tarim Mummies are not the direct ancestors of Uyghur. So, Chinese government doesn't have to worry about Tarim Mummies encouraging Uyghur nationalism.

Samuel Andrews said...

Balkan_Chl:I2181 is in WE9 PCA. I don't know where a link to that PCa is. He scores slightly more UkraineN than Yamnaya. But scores both.

Samuel Andrews said...

@iggy, I've wondered the same thing. IMO, ANE for sure did not have the very specific traits unique to East Asians. Because......

When, Malta boy (the original ANE genome) was discovered in 2013, the highlight was he had no relation to East Asians but is very closely related to Native Americans and therefore Native Americans aren't 100% East Asian. I doubt after this discovery, it will turn out ANE was secretly East Asian even though his DNA doesn't express a strong relation to modern East Asians.

Most current models for ANE's deep ancestry, do model it as part East Asian-related. This may be true. But, these models make ANE part East Asian related not East Asian. By East Asian related they mean DISTANTLY related. Like, how the Onge who look African are distantly related to East Asians and are called "East Eurasian" by geneticists.

What everyone agrees on is ANE has no 'recent' common ancestry with East Asians, like within the last 40,000 years. Due to this alone, we can assume ANE didn't look East Asian.

In my opinion, Native Americans look different from East Asians due to their ANE ancestry. So, ANE in my opinion probably had brown/'red' skin color like Native Americans do.

Because, of this I think Central Asia and Western Siberia, as recently as 4,000 years ago was probably inhabited by people who did not look East Asian But, instead looked more like Native Americans. While, today they more or less look East Asian. Which makes this detail interesting.

If near pure ANE still lived today in Asia, like they did until only 3,500 years ago, it would have been much easier to determine Native Americans origins from living people's DNA. It would have been obvious they derived from two different Asian 'races.'

Arza said...

@ JuanRivera
as Volosovo has the haplogroup

Opinions are divided, so better be careful with such assumptions.

Davidski said...

@Arza

Opinions are divided, so better be careful with such assumptions.

You mean people who've seen the data say that there's definitely M269 in several Volosovo samples, while some of those who haven't seen it claim that there isn't?

That kind of divided?

Davidski said...

@Iggy

Apparently MA1 and AG3 show some sort of East Asian-related ancestry, and if so, then that might mean ANE is partly of East Asian origin.

But I don't have a clue what that suggests about the appearance of MA1 and/or AG3, if anything at all.

I'm not even sure if the modern East Asian look existed back then, and even if it did, then it may not have been associated with the East Asian-related ancestry in ANE.

It's also possible that all Eurasians looked more or less alike back then anyway.

Simon Stevens said...

@Samuel Andrews So to summarize, ANE might have a component related to East Eurasians, but that doesn’t make it East Asian, only sharing a much deeper, common ancestral relation and not deriving from East Asian proper. Would you say this is similar to how WHG and AHG share a deep ancestral relationship while not deriving from one another? The same could be said of the deep common ancestry between Europeans and Native Americans as well correct? This is also highlighted by the fact that neither AG3 nor MA1, display any affinities to East Asians, but instead—like Yana and Steppe Maykop—to Native Americans.

Copper Axe said...

Is there information out there regarding the phenotypes of these various WSHG populations? I'm curious what they looked like.

Their interactions with the various steppe herders populations are interesting it is like they were meant to be rivals!

What do you guys make of the supposed L151 in the Afanasievo?

Cheers everyone and thank you David for running this blog!

Samuel Andrews said...

@Simon_Stevin, Yes, that's perfect summary.

"Would you say this is similar to how WHG and AHG share a deep ancestral relationship while not deriving from one another?"

The basics are the same, both sets have distant common ancestors. But, I would bet AHG and WHG are much closer to each other than ANE and East Asians are.

CrM said...

@Copperaxetotheneck

I'm only aware of these,
The first one is Botai, which is mostly WSHG + a bit of additional East Asian related.
The second one is from the Neolithic Urals, don't know much about it.
The third is Kelteminar, the culture is long speculated to be WSHG-related, but it may turn out to be something different.

https://imgur.com/a/CYQaTEY

Samuel Andrews said...

@Simon_Stevin," The same could be said of the deep common ancestry between Europeans and Native Americans as well correct? "

That's a good analogy. It doesn't make sense to call them related, but there is a distant common ancestor.

Ric Hern said...

@ Iggy

If Mal'ta Buret people are the closest to ANE then it is basically easy to see how they looked like. Just look at the human figurines they carved. There are several....

Simon Stevens said...

@Samuel Andrews Thanks that makes sense. Saying Mal’ta Buret is East Asian seems to me to be the equivalent of saying Native Americans are white or partly European because of the ANE, or that we all have Bantu Sub-Saharan African ancestors. Those populations and ethnic groups didn’t even exist yet, it’s like saying we are descendants of Chimps, no we just have a common ancestor. ANE share a very deep and divergent common ancestry with East and South Eurasians, they don’t derive from one another. What’s your opinion on Steppe Maykop and WSHG? I found Narasimhan et al. 2019 quite confusing. It seems that there were Siberian Hunter Gatherers that had East Asian ancestry (ESHG), and then there were those that didn’t, those being the WSHG. I’m guessing the arrival of East Asian related ancestry in Siberia and Central Asia occurs around the time of Botai, or roughly around the late neolithic and early cooper age. Steppe Maykop like WSHG, also doesn’t have that ESHG/East Asian component. I would guess that the distant, deeply divergent, common ancestry of East Eurasians, South Eurasians, and ANE, can help explain the occurrence of mtDNA C in some European remains such as the C1f/C1g in mesolithic Karelia, or even the C1e in modern Iceland. C1f has even been found in Scotland and Italy. C4 and C5 have been found in Scandinavia, France, Italy, Greece, Germany, Slovakia, Poland, Scotland, Ireland, and England. To my knowledge, mtDNA haplogroup C4a was discovered back in the early 2010s or so, in samples from the Catacomb Culture. Potapovka posses some C as well, with C1 making an appearance. MtDNA C1 is also very common in Native American populations. There is also the issue of the EDAR genes in SHG and Steppe Maykop. May have something to do with this distant, ancient common ancestry.

Jatt_Scythian said...

Why would China object to releasing data on the Tarim mummies? If they are Z93- then that mean Uyghurs are not really even partially descended from them.

ANE and WSHG don't look East Asia. And I'm happy this discover of Y P, K2, R* proves that none of these are SE Asian in origin.

Samuel Andrews said...

@Simon Stevin, " I found Narasimhan et al. 2019 quite confusing. It seems that there were Siberian Hunter Gatherers that had East Asian ancestry (ESHG), and then there were those that didn’t, those being the WSHG"

All West Siberia_N have 11% East Asian. They also have a few percent WHG. Overall, 85% ANE. Steppe Maykop's Asian ancestor was just like West Siberia_N so also had about 10% East Asian. East Asian admix, akin to Mongolia Neolithic, admixed with ANE as far west as Western Siberia by 6000 BC.

"I would guess that the distant, deeply divergent, common ancestry of East Eurasians, South Eurasians, and ANE, can help explain the occurrence of mtDNA C in some European remains such as the C1f/C1g in mesolithic Karelia"

Yeah, that could be. mHG C is related to mHG M8 which is present in East Asia. Related in the last 50,000 years so distant. Maybe, it is from ANE's 'East Asian-related' ancestor.

Kristiina said...

@ Afanasievo, Shirenzigou and Tocharian

Shirenzigou uniparentals are completely different from Afanasievo, in particular yDNA:
R1b-PH155 (R1b2b2-Y92825), R1b-PH155 (R1b2-PH491/Y141794), Q1a1a-F4734/F750/Y693/M120, Q1a2a2b1-F4917/F4956/F5400 (under Q-M25), R1a1a1b-F3044/Z645 (above Z93), O-CTS7553/M1759,
while Afanasievo is mostly R1b1a1a2a-L23 (+ some J1a, Q-M25, C-M217, R1b-L151 (?)).

As for mtDNA,
T1a1b: not been found in Afanasievo, Turan or Caucasus
I1b: found earlier in Turan
A17: found in Qinghai Iron Age
D4j1b: found in Qinghai Neolithic Gonghe Basin ZongriM1 3765 C14 BP
G3b: no ancient samples, but seems typical for Tibet
U5a2: not been found in Afanasievo
U5b2c: not been found in Afanasievo, seems to be from EBA Europe so maybe migrated with Sintashta
U4’9, U4: difficult to determine as U4 is widespread
Only one mtDNA haplotype, H15b1, can be connected with Afanasievo, but it probably went via Okunevo as H15b passed to Okunevo and Mönkkhairkhan MLBA.

If we look at figure 3 in ”A dynamic 6,000-year genetic history of Eurasia's Eastern Steppe” (https://www.biorxiv.org/content/10.1101/2020.03.25.008078v1.full.pdf), we can see the development. In MLBA, Afanasievo component virtually disappears. Khövsgöl pastoralists appear with WSHG. On the eastern fringe, there is Ulaanzuukh crowd that is autosomally like the earliest Neolithic inhabitants. Both these groups are heavy with yDNA Q. Sintashta elements with yDNA R1a1-Z93 appear during MLBA. During the Early Iron Age there are two completely different populations: Uyuk which consists of Sintashta and Khövsgöl elements (R1a1 and Q) and Slab Grave which is like Devil’s Gate (mostly Q1a1a-M120). Scytho-Siberian Saglys carry similar yDNA as Uyuk. Late Iron Age Xiongnus are very diverse, and I doubt that anybody thinks that they had preserved the Afanasievo language.

To su up, I cannot identify any group as a possible candidate for descendants and linguistic heirs of Afanasievo in MLBA, Iron Age or Xiongnu period.

A very interesting point is that there is R1b1a1a2a2-Z2103 both in Afanasievo and Chemurchek, but they are autosomally completely different. Chemurchek harbours Baikal EBA and BMAC ancestry and their language could indeed be related to BMAC languages. R1b only reappers with Xiongnus, but it is probably R1b2-PH155.

Davidski said...

@Kristiina

Shirenzigou uniparentals are completely different from Afanasievo, in particular yDNA.

To sum up, I cannot identify any group as a possible candidate for descendants and linguistic heirs of Afanasievo in MLBA, Iron Age or Xiongnu period.


Yep, that's exactly my impression.

vAsiSTha said...

qpAdm for CHN_Shirenzigou_IA. This is the only passing model so far

China_IA.SG

Kazakhstan_CentralSaka.SG: 66.7 +- 4.9
Nepal_Chokhopani_2700BP.SG: 17 +- 1.9
Kyrgyzstan_BA_Aygirdjal: 16.3 +- 3.4
chisq: 7.6
tailprob: 0.368
result file https://pastebin.com/CcybsqUb

Anonymous said...

@Kristiina

"T1a1b: not been found in Afanasievo, Turan"
T1a1 found in Afanasievo (and Andronovo), what T1a1b did not find is probably a small number of samples fluctuation because
Bronze Estonia Iru, Harju; Iru 14, cist [X02 / AI 4810: L5] 1090–910 cal BC (2834±28 BP, SUERC-80017 (GU47828)) M R1a-M198 T1a1b

"Chemurchek harbours Baikal EBA"

No, Botai.

-----------------------
ANE of course aren't any East Asians.

vAsiSTha said...

Reposting chemurchek qpAdm. BMAC is post 2000bce, so turan or east iran (SiSokhta) will have to be used as source.

Mongolia_EBA_2_Chemurchek

Botai: 32.2 +- 3.7
Geoksyur_EN: 15.1 +- 3.7
Russia_Afanasievo: 31.9 +- 2.6
Russia_BA_Okunevo.SG: 20.9 +- 3.5
chisq: 7.51
tailprob: 0.2755
result file https://pastebin.com/aX5tV4Hz

Kristiina said...

@ Archi

Yes, a typo. Black component in Chemurchek is ANE. As reference, this element is present at 80% in Dali EBA (Byan Zherek Kazakhstan I3447 2850-2495 calBCE R3/R1b).

Anonymous said...


Again, Okunevo has nothing to do with Chemurchek, because Okunevo was after Chemurchek. You can't model by placing descendants as sources.

vAsiSTha said...

@archi

okunevo earliest sample is older than the 2 chemurchek samples.

chemurchek dates
4473, 2570-2476 calBCE [Rcombine of 2567-2468 calBCE (3983±17 BP, GrM-12984) and 2833-2487 calBCE (4050±25 BP, UCIAMS-226530, PSUG-5433); use instead of less precise and inconsistent non-AMS dates of 2461-2152 calBCE (3838±40 BP, Le-6932, Le-6933, Le-6939 R_Combine)]

4468, 2571-2464 calBCE (3980±25 BP, UCIAMS-226526, PSUG-5429)

Okunevo earliest 3 samples

4555 2859-2350 calBCE (4023±56 BP, UBA-31597)
4482 2831-2233 calBCE (3961±62 BP, UBA-31599)
4411 2573-2348 calBCE (3964±31 BP, UBA-31600)

Anonymous said...

@vAsiSTha

Nonsense.

Chemurchek - 2476,2464 calBCE > Oldest Okunevo 2350, 2233, 2348 calBCE; 3983, 3980 BP > 3961, 3964 BP


The oldest Okunevo specimens are not older than Cememuchek. Playing with units of the year is just idiotic.

In any case, Chemuchek was before to Okunevo, it's an archeological fact.

And you don't use only one oldest Okunevo sample, you use all the Okunevo samples.

Anonymous said...

Especially the oldest Ohynevo sample is Afanasievan
Bronze Okunevo Russia Uybat V, 7053-169, kurgan 1, grave 1, sk.1 [RISE675] 2859-2350 cal BC (4023 ± 56 BP, UBA-31597) M R1b1a1a2a2

gamerz_J said...

Perhaps silly question but I noticed that ALDER admixture percentages differ somewhat from the ones inferred via qpADM. Does anyone know which ones are more reliable?

vAsiSTha said...

@archi, this will be my last reply to you, getting the flu is better than arguing with you. Oldest sampled labeled Okunevo_BA.SG by Damgaard and Narsimhan is RISE675. 4555, 2859-2350 calBCE (4023±56 BP, UBA-31597). Both these things you already agree upon.

RISE675 sample is similar as all the other Okunevo samples. It is not special.


Target,Distance,KAZ_Botai,MNG_North_N,RUS_Afanasievo
RUS_Okunevo_BA:RISE677,0.04403317,61.6,35.6,2.8
RUS_Okunevo_BA:RISE684,0.02307791,57.6,28.4,14.0
RUS_Okunevo_BA:RISE683,0.02255819,54.2,30.4,15.4
RUS_Okunevo_BA:RISE670,0.02242491,53.4,31.4,15.2
RUS_Okunevo_BA:RISE719,0.02082500,53.2,28.6,18.2
RUS_Okunevo_BA:RISE515,0.02654752,53.0,29.0,18.0
RUS_Okunevo_BA:RISE672,0.02306612,52.2,28.0,19.8
RUS_Okunevo_BA:RISE671,0.03245586,52.0,30.2,17.8
RUS_Okunevo_BA:RISE667,0.03433548,50.0,35.6,14.4
RUS_Okunevo_BA:RISE674,0.01416065,49.0,30.8,20.2
RUS_Okunevo_BA:RISE685,0.02296678,49.0,28.6,22.4
RUS_Okunevo_BA:RISE680,0.02273388,48.4,31.4,20.2
RUS_Okunevo_BA:RISE675,0.02484224,45.6,29.2,25.2
RUS_Okunevo_BA:RISE681,0.02610524,43.6,22.0,34.4
Average,0.02572378,51.6,29.9,18.4

Simon_W said...

@Archi, re: Etruscans

1. You seem to believe that the Global25/nMonte method automatically finds the true solution if you feed it with all the samples as source data. I strongly disagree with this belief. It's a great tool, but not this simple and powerful. The results you obtain are dramatically overfitted. IMHO it's more sensible to try just a handful of samples as source data, carefully selected. Either relatively proximate potential sources that archaeologically or otherwise make sense. Or just the few maximally differentiated distal source populations that capture the basic variation in West Eurasia.

2. In case we have several individuals for an ancient population it's better to use the average to minimize chance variation. Outlying individuals that are really not typical for the whole population are listed separately in the Global25 sheet, marked with the suffix _o.

Going by your overfitted models, Etruscans have Mycenaean- and Minoan-like ancestry and even Levantine admixture. However, if we just use the basic variation in West Eurasia as source populations, namely: Barcin, WHG, Steppe, Anatolia_EBA, Ganj_Dareh_N, Natufian, Taforalt, then the models clearly show that the Etruscans have neither Anatolia_EBA nor Natufian ancestry, and hence they cannot really have Mycenaean-like, Minoan-like or Levantine ancestry, it's logical.

Anonymous said...

@fAsiSTha getting the flu is better than arguing with you

You can not argue with anyone because you are always wrong and are not able to understand what they write to you.

"sample is similar as all the other Okunevo samples. It is not special." sic!(((

Distance to: RUS_Okunevo_BA:RISE675
0.04400341 RUS_Afanasievo:I10564
0.04855224 RUS_Afanasievo:I5273
0.04882049 RUS_Afanasievo:I3388
0.05067386 RUS_Afanasievo:I6713
0.05100108 RUS_Afanasievo:I10565

Distance to: RUS_Okunevo_BA:RISE677
0.06257268 RUS_Afanasievo:I5273
0.06361116 RUS_Afanasievo:I10564
0.06363875 RUS_Afanasievo:I10565
0.06427939 RUS_Afanasievo:I3388
0.06543653 RUS_Afanasievo:I5270

Distance to: RUS_Okunevo_BA:RISE719
0.05506523 RUS_Afanasievo:I10564
0.05615283 RUS_Afanasievo:I5273
0.05681188 RUS_Afanasievo:I10565
0.05740035 RUS_Afanasievo:I3388
0.05954074 RUS_Afanasievo:I6713

Distance to: RUS_Okunevo_BA:RISE684
0.05677209 RUS_Afanasievo:I5273
0.05716144 RUS_Afanasievo:I10564
0.05718680 RUS_Afanasievo:I3950
0.05801913 RUS_Afanasievo:I6715
0.05909949 RUS_Afanasievo:I11112

Distance to: RUS_Okunevo_BA:RISE683
0.05402148 RUS_Afanasievo:I5273
0.05415025 RUS_Afanasievo:I10565
0.05511334 RUS_Afanasievo:I10564
0.05676901 RUS_Afanasievo:I3388
0.05710429 RUS_Afanasievo:I3950

Distance to: RUS_Okunevo_BA:RISE672
0.05364970 RUS_Afanasievo:I5273
0.05370354 RUS_Afanasievo:I10564
0.05579050 RUS_Afanasievo:I10565
0.05736994 RUS_Afanasievo:I3388
0.05818015 RUS_Afanasievo:I6713

Distance to: RUS_Okunevo_BA:RISE671
0.05730663 RUS_Afanasievo:I10564
0.05821039 RUS_Afanasievo:I5273
0.06090755 RUS_Afanasievo:I3950
0.06161842 RUS_Afanasievo:I3388
0.06313414 RUS_Afanasievo:I6715

Distance to: RUS_Okunevo_BA:RISE670
0.05327298 RUS_Afanasievo:I10564
0.05658206 RUS_Afanasievo:I5273
0.05814766 RUS_Afanasievo:I3950
0.05877440 RUS_Afanasievo:I3388
0.05888599 RUS_Afanasievo:I10565

Distance to: RUS_Okunevo_BA:RISE667
0.05713274 RUS_Afanasievo:I5273
0.05926323 RUS_Afanasievo:I10564
0.06070758 RUS_Afanasievo:I3950
0.06167755 RUS_Afanasievo:I6715
0.06193359 RUS_Afanasievo:I3388

Simon_W said...

@Archi

You see, the Global25 coords of a population or of an individual are just one point in a multidimensional space. They can be approximated well in wholly different ways. Per se they don't tell us in which specific way they have to be approximated. nMonte just tries to minimize the difference with the available source populations, but smaller difference doesn't always mean truer model.

Leron said...

Simon: Then you would have to look at the population existing between Anatolia_EBA and Levant that contributed to both Minoans and Etruscans before the rest expanded. Chalcolithic Transcaucasus and Taurus Mts. groups are your best candidates.

Anonymous said...

@ Simon_W

1. That's not true, he's not finding an exact solution, this method is suboptimal. But it is good for comparing samples with the same sources to show all their differences and to find those sources that were for a given sample and those that did not participate in the sample. It's Comparison!

2. By no means should you average it, if only because you have not proved in the slightest that it is one population! In ancient samples, you can never know that these people made up a single population and did not arrive yesterday! What there was no mechanical mixing. So you never use any average populations on PCA and ADMIXTURE, this modeling is the same. Modeling is not a statistic calculation, it's where you use the average, but in modelles you have to model each sample separately. It is proven that these people were not a single population, but came from completely different sources.

"Going by your overfitted models, Etruscans have Mycenaean- and Minoan-like ancestry and even Levantine admixture. However, if we just use the basic variation in West Eurasia as source populations, namely: Barcin, WHG, Steppe, Anatolia_EBA, Ganj_Dareh_N, Natufian, Taforalt, then the models clearly show that the Etruscans have neither Anatolia_EBA nor Natufian ancestry, and hence they cannot really have Mycenaean-like, Minoan-like or Levantine ancestry, it's logical."

This is absolutely not true, this is just your fantasy. Actually, I included them in the models, but they gave all zeros, and I excluded them from the models so as not to clutter them. You haven't even noticed that Taforalt and Natufian is presented there!

" but smaller difference doesn't always mean truer model."
Is this an excuse that you have the wrong models? So they are worse, simply because they are historically incorrect.



Simon_W said...

My apologies for riding around on this off-topic subject in this thread about East Asians. Just one last, additional remark. There's the problem that so far we've got only one Proto-Villanovan and one Villanovan. Hence we can't be sure if they're typical ones. They could both be typical, or just one of them, or neither one.

But: If both are typical, then the Proto-Villanovans can neither be the source of the Etruscans, nor of the Italics. Because the Proto-Villanovan individual produces a bad fit with the Latins. And because the Villanovan individual is more like a Latin than like the Etruscans. I think this scenario (both are typical and neither Italics nor Etruscans derived from Proto-Villanova) is pretty unlikely.

If the Proto-Villanovan individual is typical and the Villanovan individual by chance not a typical one, then the Etruscans are derived from the Proto-Villanovans, but not the Italics. This was the assumption I initially tended to. But how likely is it that the Italics with their considerable Steppe admixture were already in central Italy during the MBA, if the Terramare culture in the Po valley was still Sardinian-like at that time? Possible, but not very likely!

But what if the Villanovan individual is typical and the Proto-Villanovan individual is atypical? After all she's from coastal Abruzzo - Dalmatia, where HRV_IA comes from is just on the other side of the Adriatic. Then it follows that the Italics are derived from the Proto-Villanovans, but not the Etruscans. The Etruscans would then have come at the end of the Villanovan culture from a HRV_IA-like Balkan population (but hardly from Greece with it's Anatolia_BA admixture).

And if the Villanovan individual is atypical and the Proto-Villanovan one is not representative of the entire variation of this culture, then both Etruscans and Italics arrived with the Proto-Villanovan culture, as I assumed above.

Simon_W said...

@Leron

I don't understand your argument. According to my models Etruscans have neither Anatolia_BA nor Levantine nor Minoan-like nor extra-CHG. They are just Barcin, Steppe and WHG, that's all. Hence no Taurus mountains or Transcaucasus at all.

Leron said...

Simon: I might have assumed incorrectly that you still found them to have Ganj_Dareh_N (just no Anatolia_EBA or Levant).

Anonymous said...

@Simon_W
"According to my models Etruscans have neither Anatolia_BA nor Levantine nor Minoan-like nor extra-CHG."

It just doesn't exist in your models, you haven't shown any of it, and you're making a gross methodical mistake by uniting everyone into one population without any right to do so.

"HRV_IA-like Balkan population"
You have no right to consider this specimen as a source because it was later than ProtoVillanova, Villanova and Rutuli and Etruscan. You have not done a good research, you can immediately see that you did not go through the options at all because Bell_Beaker_CZE fits better than HRV_IA.

So you're making the wrong assumptions.

Simon_W said...

@Archi

1. You can also compare samples using a smaller set of source populations. That's what I often do. It's more clear and avoids overfittet models. And after all, what does it mean that one Etruscan supposedly (according to your model) had some Ashkelon_IA1 and another one some Natufian? Or that one Etruscan has some Sardinia_EBA, while a Rutuli from Ardea also has some Sardinia_C? Would that be a crucial difference?

2. Sometimes it makes sense not using the average, as in the case with the two Hallstatt_Bylany samples who are very different from each other. Sure there is also some variation among the samples from Latium and among the Etruscans but I don't have the impression that it's that vast, if the outliers are excluded as outliers. And all these samples are rather small, just a few individuals. If one of them has a vastly different background than the other ones, then that would show up in the average. But feel free to model all individuals separately, I'm not completely against this.

"This is absolutely not true, this is just your fantasy. Actually, I included them in the models, but they gave all zeros, and I excluded them from the models so as not to clutter them. You haven't even noticed that Taforalt and Natufian is presented there!"

Lol, yes in your models the Etruscans have neither Barcin nor WHG or Steppe. But please, man, these populations are basic for Europe. They are in all later populations. So in your models these components are represented by the many other populations you use. And yes I noticed the Taforalt and Natufian in some of your models. This is partly real and partly just an artifact of the overfitting. Taforalt is real in the Etruscan outlier and Natufian is real in the Ardea outlier.

"Is this an excuse that you have the wrong models? So they are worse, simply because they are historically incorrect."

Historically they are a lot more correct than models claiming that Etruscans had PPNB ancestry or Sardinian admixture...

capra internetensis said...

@Vasistha

Scaled G25 matches your qpAdm precisely:
Chemurchek avg - 44% Botai, 36% Afanasievo, 15% Geoksyur, 6% Mongolia N - 2.0630%
Chemurchek avg - 34% Afanasievo, 33% Botai, 19% Okunevo, 14% Geoksyur - 2.0137%
Dali EBA - 82% Botai, 18% Geoksyur - 3.2034%

But not here:
Dali EBA - 76% Botai, 13% Afanasievo, 11% Geoksyur - 2.5742%
Your outgroups look fine though, so qpAdm is likely more accurate.

I recall earlier (was reading but too busy to contribute) you were successfully running models with Satsurblia on one side and Kotias on the other to pin down CHG contributions; I wondered, is it possible to do the same with Turan, for instance put Geoksyur En on one side and Bustan En on the other, or something like that?

Jatt_Scythian said...

So both Chemurchek and Dali EBA were predominantly West Eurasian?

Anonymous said...

@Simon_W

1. See, you don't understand anything about samples at all. You choose only that small set of samples which are convenient to you from your ideological views! So you're not looking for the truth, but only adjust the models to your personal preferences. Look for the history of the samples before modeling, and then you will realize that between Natufan and Ashkelon_IA1, these are completely different people who come from completely different places! No one there has Natufans, but
Ashkelon_IA1 is the Seas People migrants (not Natufian), which you can see clearly! And shown in purely objective conditions, as opposed to your purely subjective and biased!

2. On average, you can model if it is proved that this is one population, but here it is proved that they are people of different origin. On average it is always the average temperature in a coronavirus hospital. This is never proof of anything.

"Lol, yes in your models the Etruscans have neither Barcin nor WHG or Steppe. But please, man, these populations are basic for Europe. They are in all later populations. So in your models these components are represented by the many other populations you use."

LOL, WHG = 0, Steppe is part Bell Beaker, Anatolia_Barcin_C = 0, Anatolia_Barcin_N is part Remedello, Sardinia, etc.

"And yes I noticed the Taforalt and Natufian in some of your models. This is partly real and partly just an artifact of the overfitting. Taforalt is real in the Etruscan outlier and Natufian is real in the Ardea outlier."

You're looking completely wrong, Natufian = 0 in all calculations. Don't talk about relevance here, you put your mistaken views ahead of the facts.

"Historically they are a lot more correct than models claiming that Etruscans had PPNB ancestry or Sardinian admixture..."

You're wrong, your models have proved that you wrote only mistakes. I proved you wrong. You don't want to admit it, but you know you're wrong. I've clearly proved it, you've only proved that you've written mistakes.

Your model is completely wrong because it's wrong by all standards, your historical views are completely wrong because you don't know their History at all.

Kristiina said...

@ David

Yes, I agree that it is quite crazy to insist that Shirenzigou samples lack Tibetan-related ancestry when one of the male samples carry yDNA O and three out of ten mtDNAs are related to Tibetans. Moreover, geographically, Shirenzigou, Xinjiang, seems to be located close to Hami, and Xinjiang is north of Tibet.

MtDNA A17 is a small haplogroup distributed in the area between Tibet and Thailand, and it can hardly be a coincidence that the only two ancient A17 samples are from Qinghai which is an area bordering Xinjiang: C2019, Iron Age, Qinghai Hualongqunke Qunke1 2480 C14 BP A17, and C497, Qinghai Hupo, unidentified period A17.

D4j1b has even a more restricted distribution in the same area, and it has recently been detected in Qinghai Neolithic Gonghe Basin ZongriM1 3765 C14 BP!

«Oldest ‹Older   1 – 200 of 346   Newer› Newest»