search this blog

Saturday, August 27, 2022

Dear Iosif...

Update 29/08/22: Dear Iosif #2

...


I'm skimming through the Lazaridis, Alpaslan-Roodenberg et al. paper that just came out at Science. And I feel like someone punched me in the face.

Nevertheless, I'll try to be diplomatic. Suffice to say, for now, that there's some rather strange stuff in this paper.

The main problem is that the authors are attempting to study fine scale ancestry with a somewhat rough distal model. As a result, they miss important details.

For instance, this quote is from the paper's supplementary PDF file, freely available here.

However, the complete lack of association of R-haplogroup descendants and EHG ancestry in either Armenia or Iran is consistent with either a massive dilution of EHG ancestry in these populations resulting in the dissociation of Y-chromosome lineages from autosomal ancestry over time, or with a scenario in which R-M269 was not associated with substantial EHG ancestry to begin with.

Obviously, EHG means Eastern European Hunter-Gatherer. But why focus on EHG? Surely, this makes little sense when looking at the genetic prehistory of West Asia, because no one ever argued that this region was settled by EHG populations. It was widely settled by Yamnaya-related groups, with already heavily diluted EHG ancestry, during the metal ages.

OK, so the authors are actually aware about the potential dilution of EHG ancestry, but they don't really do anything about it.

If we're looking at the origins of West Asian R1b-M269, and using its association with autosomal DNA components as a guide, then we should be focusing on Yamnaya-related ancestry.

For instance, here's a fine scale ancient ancestry model based on Principal Component Analysis (PCA) data. It shows the ancestry proportions of two relatively high coverage Iron Age males from two different sites in Iran from the Lazaridis, Alpaslan-Roodenberg et al. dataset. Both belong to R1b-M269 and both show significant Yamnaya-related ancestry.

Target: IRN_HajjiFiruz_IA:I2327_all
Distance: 2.2930% / 0.02292994
39.6 Kura-Araxes_ARM_Kaps
24.2 IRN_Ganj_Dareh_N
18.2 Levant_PPNB
12.4 Yamnaya_RUS_Samara
4.4 Anatolia_Tepecik_Ciftlik_N
1.2 Han

Target: IRN_Hasanlu_IA:I4232_all
Distance: 2.5179% / 0.02517895
26.0 IRN_Ganj_Dareh_N
25.6 Kura-Araxes_ARM_Kaps
24.4 Anatolia_Tepecik_Ciftlik_N
15.8 Yamnaya_RUS_Samara
7.6 Levant_PPNB
0.6 IRN_Shahr_I_Sokhta_BA2

As a control, here's an earlier, Chalcolithic sample bearing Y-haplogroup J2b from the same region. Not surprisingly, this individual totally lacks the Yamnaya-related signal.

Target: IRN_HajjiFiruz_ChL:I4241_all
Distance: 2.7938% / 0.02793782
32.6 Kura-Araxes_ARM_Kaps
25.6 IRN_Ganj_Dareh_N
23.6 Anatolia_Tepecik_Ciftlik_N
18.2 Levant_PPNB

Overall, these results make perfect sense. I could probably locate very minor signals of EHG ancestry in the Iron Age samples, but that would be more difficult and much less certain, so I won't bother.

Soon I'll be able to rerun these analyses with Bronze Age samples from Dagestan and surrounds. That should bump up the levels of Yamnaya-related ancestry and improve the statistical fits (wink, wink, nudge, nudge).

Disappointingly, Lazaridis, Alpaslan-Roodenberg et al. go so far as to suggest that R1b-M269 may have originated in West Asia.

However, considering the scores of ancient Eastern European populations rich in R1b-M269 and many near and far related subclades of R1b, this makes no sense whatsoever.

Indeed, contemplating nowadays that R1b-M269 might be native to West Asia, where R1b only starts showing up in the ancient DNA record during the Copper Age, is about as stupid as claiming that gravity doesn't exist.

Largely due to their distal model approach, Lazaridis, Alpaslan-Roodenberg et al. also argue that the Indo-Anatolian homeland was located in what is now Armenia and surrounds. I'm far from convinced that this solution will stand the test of time.

In terms of the more widely accepted theory that the Indo-Anatolian homeland was located on the Pontic-Caspian steppe in Eastern Europe, the most important samples in the paper are the three Bronze Age individuals from Yassitepe in western Anatolia. That's because they're from a region that is traditionally seen as the entry point of Indo-Anatolian speakers into Anatolia from the European steppe via the Balkans.

Interestingly, individual I5737, dated to 2035-1900 calBCE or the Middle Bronze Age, belongs to Y-chromosome haplogroup I2a-P78, which surely must be a signal of European ancestry. I see this as a significant result.

Here's how the trio from Yassitepe look in my fine scale ancient ancestry model. Minor Yamnaya-related ancestry does show up, although, admittedly, it might just be noise in individual I5735.

Target: TUR_Aegean_Izmir_Yassitepe_MBA:I5737
Distance: 2.7507% / 0.02750748
58.4 Anatolia_Barcin_N
20.4 Kura-Araxes_ARM_Kaps
9.2 Anatolia_Tepecik_Ciftlik_N
5.6 IRN_Ganj_Dareh_N
3.8 Yamnaya_RUS_Samara
2.6 Levant_PPNB

Target: TUR_Aegean_Izmir_Yassıtepe_EBA:I5733
Distance: 2.7969% / 0.02796887
52.0 Anatolia_Barcin_N
27.2 Kura-Araxes_ARM_Kaps
8.6 Levant_PPNB
6.2 Yamnaya_RUS_Samara
6.0 IRN_Ganj_Dareh_N

Target: TUR_Aegean_Izmir_Yassıtepe_EBA:I5735
Distance: 3.1270% / 0.03127009
36.0 Kura-Araxes_ARM_Kaps
32.4 Anatolia_Tepecik_Ciftlik_N
26.0 Anatolia_Barcin_N
2.8 IRN_Shahr_I_Sokhta_BA2
1.2 Yamnaya_RUS_Samara
1.0 Levant_PPNB
0.6 MAR_Taforalt

This isn't much, especially considering it's already late 2022, but it's better than nothing. Fortunately, more samples from Bronze Age western Anatolia are on the way (wink, wink, nudge, nudge).

However, I'm not done with the Lazaridis, Alpaslan-Roodenberg et al. dataset yet. I'm planning to spend much more time on this blog in the coming weeks and months and will be using their samples in a wide range of analyses.

Citation...

Iosif Lazaridis, Songül Alpaslan-Roodenberg et al., The genetic history of the Southern Arc:A bridge between West Asia and Europe, Science 377, eabm4247 (2022)

See also...

Dear Iosif #3

But Iosif, what about the Phrygians?

Dear Iosif, about that ~2%

Dear Iosif...Yamnaya

119 comments:

EthanL said...

If we follow the method of these authors, then the Phrygians TUR_C_Gordion_Anc
must also be recognized as autochthons of Asia Minor. This is an obvious absurdity. Then why didn't Reich become Baron Renfrew?

Davidski said...

Yeah, my jaw dropped when reading this paper.

I've basically given up now trying to help/influence these people, because they live in another reality.

Rob said...

Renfrew is a Lad

Kavkasi said...

@Davidski
Are we going to see steppe ancestry from samples in central Anatolia in upcoming papers?

Davidski said...

@Kavkasi

I haven't heard about such samples yet.

LGK said...

Surely the Griffin Warrior result (no EHG) compared with variable levels in other Mycenaeans makes it pretty clear that EHG could and did become diluted very quickly. Or, even that local lines (including elites) could be integrated into a dominant culturally and linguistically "Indo-European" society without necessarily ever becoming admixed. Seems like an obvious context for understanding the Anatolian results which is ignored

Matt said...

Quick set of PCA of these Yassitepe vs Minoans/Mycenaeans and TUR samples from this paper and previously published: https://imgur.com/a/VASibNb

The Yassitepe samples look on PCA no more northern shifted than the samples from the 4500 / 3800 BCE from the Sea of Marmara to their north, or persons from Isparta city in Western Turkey in mid BA, or the Minoans. Maybe this does conceal some Yamnaya related ancestry in some/all of these groups tho?

Maybe doing a direct comparison of f4 stats between these Yassitepe samples and the two Marmara with decent coverage (I10545, I10542) would identify if they are significantly different in relation to any other group?

Re; models Barcin_N, Ganj_Dareh, Levant_PPNB still seems like basically doing distal modelling by the standards of earlier papers. It's less distal than they propose but...

The true proximal approach would be "Is qpWave rank=0 / rank=1 / rank=2 rejected?" when using solely Turkish and West Asian sample populations that are proximal to the time?

None of this seems to have happened here. Lack of samples?

I would note that Lazaridis really did detect that signal in sample 2000 BCE MBA I5737 even with their model insisting on the separation of the EHG and CHG though: "EHG ancestry is not present in Anatolia during the Bronze Age except in the aforementioned outlier from Hatay region (6.5±2.8%; Z=2.3)" (this is Alalakh outlier, ALA019, who may be more likely to have Central Asian ancestry) " and very tentatively at 2.9±2.6% (Z=1.1) at Yassitepe in the Middle Bronze Age (I5737; 2033-1920 calBCE) out of 71 individuals representing every region of Anatolia." (I know this is quoting the supplement but not sure how else to cite this).

Davidski said...

Yes, and the dilution of steppe ancestry was obviously a powerful factor in West Asia that needs to be understood properly and taken into account.

But instead of doing that, they went so far as to give themselves the option that it didn't happen, and that instead there were no migrations into West Asia.

Matt said...

Looking at Vahaduo models, I think if you consider the Marmara CHL samples as a base, it's possible for the three samples at Yassitepe to have some Yamnaya related ancestry...

A said...

The Griffin Warrior was buried with the Pylos Combat Agate, which appears to show a 'Minoan' killing/defeating 'Mycenaeans' ... was he even a 'Mycenaean'?

Davidski said...

@Matt

Looking at Vahaduo models, I think if you consider the Marmara CHL samples as a base, it's possible for the three samples at Yassitepe to have some Yamnaya related ancestry...

How much are you seeing?

Matt said...

It looks like it's possible for it to be average of 5% or something I think? The models I tried were comparing the set of source (TUR_Marmara_Ilpinr_Chl and TUR_Marmara_Barcin_Chl, RUS_Yamnaya_Samara, GRC_Peloponnese_N, BGR_EBA_steppe_outliers, BGR_CHL and BGR_EBA nosteppe) using all samples and not averages and it did seem like they were picking a little Yamnaya.

It got more complicated when I tried putting Maykop in as a source as well (not because I think Maykop is likely but as a competing CHG rich source) and the models weren't decisive as to what was preferred, but this is arguably not geographically likely.

Arvind Pandit said...

700+ new samples and yet not one of them belongs to the R2 haplogroup?! Does this make any sense?

Vara said...

This is nice for Near Eastern studies.

LBA Hasanlu is an eastern Hurrian site with Indo-Aryan elite based on the golden bowl with "BMAC features". Seeing L1a almost confirms Parpola's Mundigak's tin road Mitanni hypothesis. On the other hand, Iron Age Hasanlu is mostly Urartian with a few incoming West Iranians.

It's very clear that there was a Yamnaya movement to south of the Caucasus (K-A samples). However, it's most likely Yamnaya did not spread IE languages south of the Caucasus as there seems to be strong correlation with R1b and Hurrians. Also, it can't be a coincidence that the Hurrians are first attested after Yamnaya ancestry reach south of the Caucasus. Whether the Yamnaya chieftains brought Hurro-Urartian south of the Caucasus or in the most likely scenario adopted it, it's probable some of the success of the Hurrians and perhaps even their association with equids is due to the Yamnaya chieftans.

As for Anatolians the context of how they received steppe ancestry matters.

LGK said...

@A

The excavators considered that this guy potentially had personal ties to Crete based on some of the other grave goods. His genetic profile leaves this as very possible.

However, the scene depicted on the agate seems to be something of a standard because nearly identical arrangements are found on various rings and seals at Mycenae. The "victor" does look typically "Minoan", it is possible they were indeed made on Crete and just looted by Mycenaeans. But the Griffin warrior could have been a Minoan conscripted or recruited by the Mycenaeans of Pylos. In truth there are many possibilities

ancestralwhispers.org said...

The science has made its verdict, and the long-awaited answer to the Indo-European origin question is finally here. It was an honour gentlemen, and for all the disagreements that this community had and will have in the future, maybe the real Southern Arc was the friends we made along the way.

Now seeing that the real star of the show was Armenia, the cradle of the Indo-Europeans, I noticed some pretty interesting outliers from that region.


I16546 -850 ARM_Bragdzor_EIA_father.I16536.brother.I16553 Bragdzor cemetery Armenia 40,91666 44,5 15,51% J-Z504 K1a4c

This man was was a part of a rather rich burial. Looking at his auDNA, he is evidently a Colchian. His high CHG and absence of Steppe ancestry makes him stand out from other samples in the region. His Y-DNA is present in Georgia, Northwest Caucasus and also Armenia.

Target: I16546
Distance: 3.8665% / 0.03866549
60.2 Caucasus_Hunter-Gatherer_GEO_CHG
25.2 Levant_Neolithic_Farmer_Levant_PPNB
10.4 Anatolia_Neolithic_Farmer_TUR_Barcin_N
3.4 Iran_Neolithic_Farmer_IRN_Ganj_Dareh_N
0.8 Papuan-Melanesian_VUT_2300BP_all

Distance to: I16546
0.04018039 Georgian_Imer
0.04082827 Georgian_Ajar
0.04159593 Georgian_Megr
0.04573203 Georgian_Kakh
0.04852606 Abkhasian
0.05019923 Georgian_Kart
0.05119247 Georgian_Svan
0.05131672 Ahiska
0.05225656 Georgian_Laz
0.05741580 Armenian_Hemsheni

And If I understood it correctly, his son is I16536. The son's autosomal is something like halfway between the Colchian and a local Steppe-rich Armenian.

Target: I16536
Distance: 3.9875% / 0.03987506
37.8 Caucasus_Hunter-Gatherer_GEO_CHG
23.0 Anatolia_Neolithic_Farmer_TUR_Barcin_N
15.2 Pontic_Steppe_Yamnaya_Pastoralist_Yamnaya_RUS_Samara
12.4 Levant_Neolithic_Farmer_Levant_PPNB
11.6 Iran_Neolithic_Farmer_IRN_Ganj_Dareh_N

Target: I16536
Distance: 3.0496% / 0.03049607
59.0 ARM_MBA
41.0 I16546


Another outlier is the "Urartian" from Armenia. Urartian proper samples from Van differ from the Steppe-rich Armenians or Etiunians as some call them. But this particular outlier is not like anyone else in the set. His haplogroup is G2a1, his auDNA is Georgian-like. I wonder if he's an early Iberian or a Diaokhi.


I17184 -664 ARM_Keti_Urartian Keti Armenia 40,8716 43,8445 51,31% G-Z6653 U1a1a1

Target: 176:I17184
Distance: 3.5473% / 0.03547340
51.8 Caucasus_Hunter-Gatherer_GEO_CHG
24.0 Anatolia_Neolithic_Farmer_TUR_Barcin_N
18.0 Levant_Neolithic_Farmer_Levant_PPNB
6.2 Iran_Neolithic_Farmer_IRN_Ganj_Dareh_N

Distance to: I17184
0.02586090 Georgian_Imer
0.02800585 Georgian_Ajar
0.02911075 Georgian_Laz
0.03159155 Georgian_Kart
0.03171192 Georgian_Megr
0.03180919 Ahiska
0.03544306 Georgian_Kakh


Another interesting sample is Aknashen Neolithic.

I3931 -5908 ARM_Aknashen_N Aknashen Armenia 40,10093333 44,29051667 22,05% J-PF5174 I1

Target: I3931
Distance: 4.4364% / 0.04436350
30.0 Iran_Neolithic_Farmer_IRN_Ganj_Dareh_N
29.4 Anatolia_Neolithic_Farmer_TUR_Barcin_N
28.2 Caucasus_Hunter-Gatherer_GEO_CHG
12.4 Levant_Neolithic_Farmer_Levant_PPNB

He is probably admixed with Upper Mesopotamian farmers, but I imagine his local Mesolithic precursors will be a CHG-rich population with a good amount of Iran_N ancestry and a bit of Anatolian, giving us an idea of what the Trialetian Mesolithic and Armenia Mesolithic populations were like.
Assuming this hypothetical population survived the Neolithic and had a continuity, then in theory it can be a good source of the majority of Kura-Araxes ancestry, but with all the Armenian samples available, we don't really see any signs of anyone fitting that might have given the rise to Kura-Araxes culture, thus reinforcing the theory that the CHG push to proto-KAC came from West Caucasus, perhaps with the Sioni culture.

It is also interesting to witness the downfall of I2c in the Caucasus, apparently it used to be much more frequent in MBA-LBA Armenia, and probably in contemporary East Georgia as well.

EthanL said...

It is curious that the authors decided to solve the Anatolian problem, but did not collect samples of the 2nd millennium ВС from central and western Anatolia. I found only Kilis - south-east, it is in the zone of the Hurrian language

Targamos the Based, son of Kavkasos son of CHG son of said...

I don't know why we should disagree with this scientific breakthrough. Sounds like something an anti-vaxxer would do. Trust the science people.

Davidski said...

What scientific breakthrough?

They're just pissing around with some basic models based on wrong assumptions and missing reality.

EthanL said...

Blogger ancestralwhispers.org said...
The science has made its verdict, and the long-awaited answer to the Indo-European origin question is finally here.

The current verdict of genetic science is that it is not yet able to distinguish the Hattians, Hittites, Greeks, Phrygians and other peoples of pre-Roman Anatolia from each other

Targamos the Based, son of Kavkasos son of CHG son of said...

Yet this is a study written by such esteemed and knowledgeable scientists like Lazaridis and Reich. Surely they are one of the main authorities on population genetics? Do you imply there is some conspiration theory, Davidski? I have a hard time doubting these two. Again, I thought you were open-minded based on the vaccination comments from a while ago.

Matt said...

Also another thing to consider around Yassitepe and the Marmara_Ilpinar_CHL samples (4500-3500 BCE), is looking quickly at the distance difference in Vahaduo (with free choice of all the population averages in my file):

https://imgur.com/a/d6g7OVO

The populations that "peak" being closer to Yassitepe relative to Marmara_Ilpinar_CHL are all Steppe_EMBA. That hints that the "direction" that Yassitepe is shifted relative to Marmara_CHL is Steppe_EMBA. This is something that Lazaridis and collaborators should've tested with formal stats, and also whether the Yassitepe samples form a clade?

There is a complexity that comes in when we compare Yassitepe to the Marmara_Barcin_CHL at 3800 BCE though, the old sample that they've resequenced to 963510 SNPs, in that the shift shown by Yassitepe is lesser and instead towards Kura_Araxes and Maykop: https://imgur.com/a/E1PJ2zE

But compared to the Marmama_CHL the direction looks like it's steppe.

Rob said...

Analysing Bronze Age anatolia is difficult. There’s a lot of overlapping ancestry, and you need to be sharply attune to ydna , & know everything about the archaeology, demography, climate of the region
Some vague distal modelling obviously falls well short of such a threshold

A said...

"We're doing a study on Mycenaean DNA"

'That's great, so you got samples from Mycenae?'

"No, why would we want to do that lol?"

Aram said...

Vara

Where You have seen R1b in Hurrians? LoL. There are tons of samples from Alalakh to Van where HU languages were spoken and R1b is nowhere to seen.

The only place is NW Iran where there could be hypotethic link between R1b and Hurrian. But that place has no own records. So whatever they spoke is lost. And they could be even IE unrelated to Armenian and Iranians. Isolated Mannaeans.

Also Hurrian terms are attested in Near East before any Yamna appeared in Near East.

It's now clear there was no Basque scenario in Armenia. Something one could easily predict if he was attentive to Y dna and language correlation in Caucasus.

vAsiSTha said...

Oh no, a massive 3.8% yamnaya ancestry in Anatolia in a 6 source G25 model. WHat will Iosif do now?

Davidski said...

Don't worry about the Yamnaya, worry about Balkans + far western steppe. And in many more Anatolian samples.

You know, where I2a-P78 came from.

Rob said...

There's the R1b* from Ganziatepe too, all the way in SE Anatolia. So we have I2a2 in west and R1b in SE. Putative IE speakers moving along the 'old Neolithic highway' of the west- south, whilst northcentral/ northeast Anatolia seems to have been a non-IE block only terminally entered by IE speakers.

The PIE question needs to be resolved by sharper scholars, so other equally interesting questions about Hattic and Hurrians can be probed, and their possible affinities with Minoan, NEC, etc looked into.

vAsiSTha said...

Lol, anatolia is in east already by 2400bce. This puny one off outliers in 6 source models in the west mean jackshit.

vAsiSTha said...

Anyway, I had correctly predicted yesterday.

"Anyway, so you're claiming that you know for sure of steppe presence in west anatolia prior to 2400bce? What is that - like 2% steppe?

And the whole region of central anatolia till 1500bce is devoid of steppe ancestry but suddenly anatolian names end up east?"

For what its worth, I agree with you that R1b-M269 came from north to south given current data. That of course doesnt necessarily mean it brought languages.

Rob said...

Here are some quick 3-way models : BA Bulgaria + Arslantepe BA + extra Levant


Yassitepe:I5737 39.4 54.2 6.4

Ilıpınar_ChL:I1054 25.6 74.4 0

Ilıpınar_ChL:I10547 29.2 68.2 2.6


That's huge. + corroborating I2a. Archaeology also supports it. Big movement from Thrace into Anatolia.

Davidski said...

Paternal European ancestry at the entry point of the proto-Anatolians to Anatolia. Seems pretty significant.

Minor Yamnaya Samara ancestry is just for starters anyway. There are better models to be had, and more samples on the way.

Rob said...

oh I almost forgort Kumtepe 4- just for vasistha


Kumtepe_4: 43.4 56.6 0

Vara said...

@Aram

"The sources at our disposal are of course few and often difficult to interpret, but some features emerge with reasonable clarity:
The polity of Itabalhum was well-established with a documented succession of at least three kings. It is interesting to note that the seal legends of these kings include mention of the first-born son – or chosen heir – an unusual feature in Mesopotamia, and one which points to a particular system of kingship – perhaps somewhat similar to that known from Elam in the southern Zagros mountains.
404 Although the geographical contours of the kingdom are uncertain, it must have spanned a considerable area, encompassing a number of intermontane valleys. Its core may well have been located in the Urmia Basin. Excavated sites here, like Hasanlu and Dinkha Tepe, show connections with Northern Mesopotamia.
405 The evidence suggests that different sectors of the kingdom were controlled by different royal or princely lines that may all have been connected to the same dynasty. Šušarra clearly did not belong to the core of the kingdom, and its population, through the council of elders, on at least one occasion, seems to have acted against the policy of the nuldānum of Itabalhum .
406 Although the ruling elite bore predominantly Hurrian names, the kingdom also incorporated other population elements407, notably the so-called Lulleans, who seem to represent inhabitants of the higher valleys, and may have had a different ethno-linguistic background.
408 The kingdom was allied to several other Zagros kingdoms, presumably all subsumed under the confederate label “Turukkum”409, which indicates a political construction of some historical depth, documented first in sources from the time of Yahdun-Lim of Mari.
410 The apparent local features of society were coupled with a marked Mesopotamian influence, as evidenced by the use of cuneiform, knowledge of Akkadian, use of cylinder seals, etc.


Constituent, Confederate, and Conquered Space: The Emergence of the Mittani State (Topoi – Berlin Studies of the Ancient World/Topoi – Berliner Studien der Alten Welt) . De Gruyter. Kindle Edition.

Where are these Hurrian terms before 2500BCE? The only attested IE group in NW Iran and the Zagros in the 3rd and 2nd mil were Indra worshipping Indo-Aryan elites (Tukrish, IA and Ir. Tugra). Both archaeological and genetic evidence now support this "BMAC" elites stuff in NW Iran.

Could there be an unattested IE group there? Sure, but seeing the strong contacts with Mesopotamia makes it more likely that the Urmia groups were either Hurrians or one of their attested non-IE allies.

Aram said...

Vara

You can google and find many links on the Sumerian word tabira/tibira . Coppersmith, sculptor , usually believed to have Hurrian origin.

""
The Hurrians had a reputation in metallurgy. It is proposed that the Sumerian term for "coppersmith" tabira/tibira was borrowed from Hurrian, which would imply an early presence of the Hurrians way before their first historical mention in Akkadian sources.[25][26]
""

You can check here that the word tibira is attested since 3000 bc

http://psd.museum.upenn.edu/nepsd-frame.html

Also Mesopotamians were using a term Subir/Subartu to denote Hurrian language. Which also is a very old term associated with North Mesopotamia.

Aram said...

Ancestralwhispers

KA homeland can be in the mountains were Turkey's Georgian and Armenian borders meet each other.

Chobatreti, Kaps. If I recall correctly Sagona placed it in NE Turkey. But I would wait Sioni aDNA.

As for Armenia Neolithic. Masis Blur and Aknashen seems to have different forager ratios.

Distance to: ARM_Masis_Blur_(5633)-(5532):I3930
0.04625065 Armenian_Aintab
0.04789019 Armenian_Urfa
0.05139696 Greek_Central_Anatolia
0.05332561 Lebanese_Christian
0.05339815 Cypriot
0.05392003 Druze
0.05422073 Turkish_Trabzon
0.05565105 Palestinian_Beit_Sahour
0.05566732 Greek_Cappadocia
0.05617164 Georgian_Meskheti

vAsiSTha said...

"oh I almost forgort Kumtepe 4- just for vasistha
Kumtepe_4: 43.4 56.6 0"

Only you could show this much confidence for a 4400bce sample with 12000 snps.

@Davidski, while youre here whining over 3% yamnaya ancestry, remember this

Target: RUS_Progress_En
Distance: 3.2597% / 0.03259678
59.0 RUS_Khvalynsk_En
23.0 TJK_Sarazm_En
16.8 GEO_CHG
1.2 IRN_Hajji_Firuz_BA
0.0 ARM_Areni_C
0.0 AZE_Caucasus_lowlands_LC
0.0 AZE_Caucasus_lowlands_LN
0.0 DEU_LBK_HBS
0.0 IRN_Hajji_Firuz_C
0.0 IRN_Hajji_Firuz_IA
0.0 RUS_Darkveti-Meshoko_En
0.0 RUS_Samara_HG
0.0 UKR_N
0.0 UKR_N_o

Target: RUS_Vonyuchka_En
Distance: 3.3783% / 0.03378306
50.8 RUS_Khvalynsk_En
25.6 TJK_Sarazm_En
20.8 GEO_CHG
2.8 IRN_Hajji_Firuz_BA
0.0 ARM_Areni_C
0.0 AZE_Caucasus_lowlands_LC
0.0 AZE_Caucasus_lowlands_LN
0.0 DEU_LBK_HBS
0.0 IRN_Hajji_Firuz_C
0.0 IRN_Hajji_Firuz_IA
0.0 RUS_Darkveti-Meshoko_En
0.0 RUS_Samara_HG
0.0 UKR_N
0.0 UKR_N_o


Gökhan said...

They are trying to resurrect the dead and buried Anatolian Hypothesis. When they can't get what they want from the Anatolian neoletic examples, they make excuses for what they want to see in the transcaucasus, on the western Asian plateau. Shame. Its simply manupulation.

Wise dragon said...

Lazaridis contradicts himself and didn't really challenge the origin of R-M269 in the Steppe nor the Kurgan Theory. Always check the supplementary folks, it's where the meat is.

Hence in the very same study (from its ‘supplementary material’) he had to note this:

“Where did the R-M269 founder live? The early presence of this lineage in steppe samples and its association with steppe ancestry in many of its descendants may suggest that the R-M269 founder belonged to a population with EHG ancestry.☝ However, the complete lack of association of R-haplogroup descendants and EHG ancestry in either Armenia or Iran is consistent with either a massive dilution of EHG ancestry in these populations resulting in the dissociation of Y-chromosome lineages from autosomal ancestry over time, 👆or with a scenario in which R-M269 was not associated with substantial EHG ancestry to begin with.

At present, we have no archaeogenetic information on where the R-M269 population originated. The TMRCA of R-M269 descendants is 6,400 ybp and of the immediately upstream node R-P297 a much earlier 13,300 ybp. R-P297(xM269) chromosomes are found in hunter-gatherers from the Baltic(3, 458) as well as in a hunter-gatherer from the Samara region of Russia.(8, 9) This would suggest an EHG-associated origin of this lineage, but the “long branch” of R-M269 reduces greatly any confidence in the proximity of the earliest R-M269 bearers to these eastern European relatives. Yet, the data are equally consistent with a scenario in which the R-M269 founder did not have EHG. It is a challenge for future archaeogenetic research to pinpoint the origin of the R-M269 lineage.”🤨🤔

kaksoipiste said...

@Davidski

I doubt I-L596 would have been introduced by the speakers of Indo-Anatolian. It seems to have already been present in Anatolia during the Neolithic (Menteşe sample from the Southern Arc paper). Moreover, its modern distribution in West Asia is largely confined to Northern Caucasus groups ~ up to 30% in some even. A scenario which entails Northern Iranian/Armenian/East Anatolian BA populations mixing with Indo-Anatolians, and then migrating to Northern Caucasus from the South after a paternal haplogroup turnover doesn't seem very plausible to me.

I'm using Indo-Anatolian here as a short hand for Anatolian speakers keep in mind. I don't believe CHG is responsible for the spread of IE.

Samuel Andrews said...

@ancestralwhispers,

Yeah it is all about Armenia and the Orient. Europe isn't the origin of anything cool.

This is why the vast majority of Bronze age, Iron age Armenian males belong to European lineages. R1b and I2a.

Samuel Andrews said...

I think the high frequency of European Y DNA in ancient Armenia is the coolest discovery in this paper imo.

It makes it so apparent that Armenians were a Indo-European isolate living in southwest Asia.

Samuel Andrews said...

The Armenian ancient Y DNA is also great new support for an East European homeland of Indo-European languages.

Think about it.

80% of IronBronze age Italians carried R1b M269. 50% of Iron/Bronze age Armenians carried R1b M269. They lived contemporary to each other.

They were sons of the same Eneolithic tribe from Eastern Europe.

They also both speak the same language (Indo-European).

How could you deny their common language came from their common fathers?

Samuel Andrews said...

In 1990, people were confused. They were like what could possibly be the historical and genetic connection between Italy & Armenia.

Now ancient Y DNA makes the connection so real we can touch it.

This is amazing. It is obvious who the mysterious Indo-Europeans were.

Davidski said...

@kaksoipiste

I2a-P78 is obviously European.

These samples do have European autosomal DNA too. The models with Yamnaya-related ancestry that I posted are very conservative.

And there are more ancient western Anatolians on the way.

DragonHermit said...

"considering the scores of ancient Eastern European populations rich in R1b-M269"

Um, what? I know these authors are trying to make a mountain of a molehill (if even that), but let's not pretend like R1b-M269 has ever been found in EHGs.

Considering how old R1b-M269 is, and the only EHG R1b clades so far only converge, what, like 15 thousand years back with M269, it's quite possible M269 originated in West (or Central) Asia. We've been out of Africa for 50 years. 15k years is 1/3rd of that time. R1b could have easily sipped into Central/West Asia from Siberia in ancient times.

I agree that this "paper" is just a data dump, but let's be rational here.

Davidski said...

I'm still in awe of this stupid shit...

However, the complete lack of association of R-haplogroup descendants and EHG ancestry in either Armenia or Iran...

They're actually claiming that there might not be any steppe ancestry in Armenia or Iran.

There are more than 50 authors on this paper, and not one of them questioned this.

Nezih Seven said...

@Davidski

"However, the complete lack of association of R-haplogroup descendants and EHG ancestry in either Armenia or Iran...

They're actually claiming that there might not be any steppe ancestry in Armenia or Iran."

I am pretty sure that this is not what they meant here. They are basically referring to their observation: There is no correlation between EHG-related ancestry proportions and R-haplogroups; i.e. the ones who do not bear R haplogroups have the same (or even slightly more in some cases) amounts of EHG-related ancestry.

Despite the Y-chromosome movement southward attested by our data, any association between R-haplogroup bearers and Eastern hunter-gatherer ancestry was lost south of the steppe because these had similar proportions of Eastern hunter-gatherer ancestry as I-Y16419 bearers (the second most prevalent lineage in Armenia).

Rob said...

@ kaksoipiste

This I2a is new to Anatolia. It’s different to Neolithic Anatolian I2c*

Do some basic research before commenting

Wise dragon said...

@Davidski

"There are more than 50 authors on this paper, and not one of them questioned this."

The reason for this is plain and simple - peer pressure on scientists.  

Davidski said...

@DragonHermit

Considering how old R1b-M269 is, and the only EHG R1b clades so far only converge, what, like 15 thousand years back with M269, it's quite possible M269 originated in West (or Central) Asia.

No, it's not possible.

The whole network around M269 in ancient DNA is rooted in Europe.

Ancient R1b has nothing to do with West Asia except as a late, intrusive lineage.

Davidski said...

@Nezih Seven

Thanks.

But that's not a very useful observation by the authors. And it creates confusion.

M269 is indeed associated with steppe ancestry in West Asia when compared to samples that pre-date steppe incursions into the area.

That's what I had in mind when I posted this comparison.

Target: IRN_HajjiFiruz_IA:I2327_all
Distance: 2.2930% / 0.02292994
39.6 Kura-Araxes_ARM_Kaps
24.2 IRN_Ganj_Dareh_N
18.2 Levant_PPNB
12.4 Yamnaya_RUS_Samara
4.4 Anatolia_Tepecik_Ciftlik_N
1.2 Han

Target: IRN_Hasanlu_IA:I4232_all
Distance: 2.5179% / 0.02517895
26.0 IRN_Ganj_Dareh_N
25.6 Kura-Araxes_ARM_Kaps
24.4 Anatolia_Tepecik_Ciftlik_N
15.8 Yamnaya_RUS_Samara
7.6 Levant_PPNB
0.6 IRN_Shahr_I_Sokhta_BA2

Target: IRN_HajjiFiruz_ChL:I4241_all
Distance: 2.7938% / 0.02793782
32.6 Kura-Araxes_ARM_Kaps
25.6 IRN_Ganj_Dareh_N
23.6 Anatolia_Tepecik_Ciftlik_N
18.2 Levant_PPNB

Wise dragon said...


@ancestralwhispers.org

"The science has made its verdict, and the long-awaited answer to the Indo-European origin question is finally here."

Indeed, science has spoken, but not in the way you think it did. On the contrary, the case for  a “Near Eastern” Indo-European original homeland made by Lazaradis and Reich, isn't really strong or conclusive. If you take a closer look at their supplementary materials, you'll realize that Reich and his team don't really dispute, let alone disprove, the Steppe Origin of PIE.

Wise dragon said...

@Samuel Andrews

"The Armenian ancient Y DNA is also great new support for an East European homeland of Indo-European languages."

Absolutely.


The very same data from the Southern Arch paper can be used to support the Pontic Steppe as the PIE Urheimat.

Davidski said...

@Matt

Do you have the raw sheet done in the same way?

https://pastebin.com/0bmW1JmA

Rob said...

@ vasistha


“Only you could show this much confidence for a 4400bce sample with 12000 snps.”


It actually dates to ~ 3000 BC; Harvard got the date wrong in their datasheet. Yeah it's low coverage but produces internally consistent results across qpADM, PCA, etc. So...




Rob said...

Here's another one

Barcin_C (3800 BCE)
Arslantepe_LC
Yamnaya_Bulgaria



best coefficients: 0.578 0.422
Jackknife mean: 0.581135728 0.418864272
std. errors: 0.090 0.090

error covariance (* 1,000,000)
8163 -8163
-8163 8163


summ: Barcin_C 2 0.622045 0.581 0.419 8163 -8163 8163

fixed pat wt dof chisq tail prob
00 0 17 14.631 0.622045 0.578 0.422
01 1 18 28.618 0.0532476 1.000 0.000
10 1 18 66.834 1.53946e-07 0.000 1.000





Of course, Yamnaya is 'younger' than Barcin_C, but it could be Cernavoda or some other group like Mikhailovka, etc


John Thomas said...

It's all political of course.

The anti European animus in academia runs deep.

Matt said...

@Davidski, here's the raw in the same way:

Just S_ARC samples, raw (present in their Supplementary spreadsheet S1, whether new or old but resequenced): https://pastebin.com/naWJy8uH

(Also in case anyone is curious about what the other samples in thier list are, the samples appear to be sorted by country code /population ID, and the following is matching against the entire datasheet that Davidski gave, including S_ARC and others:

Countries/Pops: A to J, raw : https://pastebin.com/qXJdRwvd
K to P: https://pastebin.com/dV3VRvFR
Q to Z: https://pastebin.com/D20yNQnQ

where I couldn't match them because the sample ID just has too many differences to the Reich Lab anno to build a quick code to match them, and it has to be done manually, I've put "No Automatch". But anything that's not in the S_ARC supplement is in the Reich lab anno as far as I can tell. The list of individual IDs being in country order makes it fairly obvious. Barring a few samples that it seems like are from S_ARC but as they were just redundant duplicates of the same individuals, they've left them in the data but removed mention of their ID from the supplement except as a note on the individual they kept. There's nothing new, so I'm not going to bother doing anything manually to try and get the last stragglers I couldn't automatch.)

On a side note the Southern Arc paper adopted Davidski's habit of using country codes rather than country names. Makes integrating the samples in a little less painful!

@Vasistha, re; Barcin_CHL I1584, she's got 963,510 SNPs in the new data for this paper, so it should be much more robust to check if the same result as Rob found comes out, and also whether this holds up with the Steppe_Eneolithic samples, potentially Sredny-Stog etc rather than Yamnaya etc.

ancestralwhispers.org said...

@Davidski, EthanL, Samuel Andrews, Wise dragon

The secondhand Indo-Europeans should tread carefully when an OG like Aram is around in these parts.

@Aram

That's true, here's another interesting sample,

Target: IRQ_Nemrik9_PPN:I6445___female___BC_8750___Coverage_25.67%
Distance: 3.5608% / 0.03560845
46.0 Levant_Neolithic_Farmer_Levant_PPNB
29.0 Iran_Neolithic_Farmer_IRN_Ganj_Dareh_N
23.8 Caucasus_Hunter-Gatherer_GEO_CHG
0.8 Anatolia_Neolithic_Farmer_TUR_Barcin_N
0.4 South_Africa_Hunter-Gatherer_ZAF_2100BP

I did not expect so much CHG in Northern Iraq, and once again the Iran_N/CHG proportions seem almost equal. There's no sign of Anatolian ancestry either, rather just Levant_PPNB. Some of the first G2a2 samples from Anatolia (Boncuklu) would similarly display pretty equal CHG/Iran_N proportions. If this represents the Trialetian Mesolithic population, then I wonder if they're responsible for the G2a2 founder effect in later Anatolians.
This population might also be the reason for why we're seeing signs of CHG in Jeitun and the like, and also sign of Iran_N in Eneolithic Steppe.

Matt said...

Btw, for the samples that have been re-sampled and enriched, the following lists what those would appear to be, and what the change in SNPs would appear to be based on the latest anno file (52.2):

List of Enriched: https://pastebin.com/RXcy9n9X

Scaled G25 (for these specifically): https://pastebin.com/XeLT29eQ

(The complication is that I'm not sure if the anno file already contains some updates to the published samples, and also the new version of the file has introduced 4 different columns for SNP count overlapping with different panels, so I picked 1240k).

vAsiSTha said...

@matt

"@Vasistha, re; Barcin_CHL I1584, she's got 963,510 SNPs in the new data for this paper, so it should be much more robust to check if the same result as Rob found comes out, and also whether this holds up with the Steppe_Eneolithic samples, potentially Sredny-Stog etc rather than Yamnaya etc."

I have already check almost all samples from Anatolia prior to this paper in a systemic manner (unlike Rob). The output can be seen here.

He is of course using younger samples to model a 3800bce samples. With no mention of competing right pops.

Steppe eneolithic does not work (p value e-8). Only models which work are

BarcinN + armeniaC/KuraAraxesKaps/meshoko/caucasusmaykop.

John Smith said...

This is the complete list of all "enriched" samples. Can someone please label them?

https://pastebin.com/raw/Y22BRS3V

For those who want to play with the new Southern Arc samples, you can find all the coordinates + Y-DNA, date and coverage in this list.

https://pastebin.com/raw/vwbBtKeZ

Vara said...

Aram

The 3000bce date is not correct. Lu E 19 is Early Dynastic (2900-2350BCE) and mentions equestrians as well making it more likely to be written at the end of that period.

Subartu is mentioned during the reign of Eannatum ~2500BCE.

The real expansion of the Hurrians is a few centuries after that. So yes, my point still stands.

vAsiSTha said...

BarcinC
BarcinN: 52%
Caucasus_Eneolithic: 48%
P-val: 0.11
Result file

BarcinC
Arslan_LC: 82%
Bulgaria_EBA_Yamnaya_o: 18%
p-val: 0.0006 FAIL
Result file

I think that should settle the matter.

Davidski said...

Does anyone know whether this sample is still current? That is, if it's been redated or pulled for some reason?

UKR_N_o:I3719

Arza said...

AADR V52.2

42 I3719 I3719 Mos53 (Grave 102 (289)) | Mos53 (289, H298, grave 102) | Mos53B (289, H298, grave 102) tooth 2018 MathiesonNature2018 Direct: IntCal20 6831 42 4983-4795 calBCE (5995±25 BP, PSUAMS-2303) .. Ukraine_N_o2 Dereivka I Ukraine 48.91422 33.76493 Repulldown on 3.2M snpset 1240K 1 0.107423 0.182 295680 284731 188830 101497 M n/a (no relatives detected) I2a1b1 .. .. 249.313296 H1 [0.96,0.984] .. .. .. .. .. .. -0.041 0.033 Model_Misspecified [0,0.024] ds.half S3719.E1.L1 PASS ..

Steven said...

Does anyone have a complete list of the haplogroups in this study?

Aram said...

This phrase is strange

""
Despite the Y-chromosome movement southward attested by our data, any association between R-haplogroup bearers and Eastern hunter-gatherer ancestry was lost south of the steppe because these had similar proportions of Eastern hunter-gatherer ancestry as I-Y16419 bearers (the second most prevalent lineage in Armenia).
""

The Y16419 came with R1b Z2103 to Armenia. Why they should have different autosomes?
The age of I2-Y16419 is 4000 years. Which exactly match MBA expansion.

Btw it's remarkable that I2a2b was absent from North West Iran. Despite geographic proximity.

elliv said...

What is EHG ancestery doing in Armenia 4000 BCE? From the Southern Arc paper:Balkan hunter-gatherer ancestry is absent here and EHG ancestry is absent in the Neolithic
samples, appears first at the Areni1 samples(10) at ~12% in the 4300-4000 BCE time frame, and
persists in later periods. The amount of EHG ancestry at Areni1 is comparable to the ~15%
across a vast swath of Southeastern Europe previously discussed. Note, however, while in the
latter case the EHG ancestry is sporadic prior to 3000 BCE, in Armenia it is present at least
1,000 years earlier.

claravallensis said...

https://twitter.com/iosif_lazaridis/status/1563953730499878926
thoughts?

Rob said...

@ Vasistha

your models are crap. The pRight set is inadequate to distinguish West Asian streams of ancestry, so I dont care what they show

There is a 1,000 year hiatis between Barcin N and C., same at Kumtepe
The post-4000 BC populations which then appear are novel, and have dual ancestry from the East and the north/west. Near 100% turnover

Like I said, you need to understand the region & samples in question in order to analyse them. I dont have the patience to listen to the BS from some smelly OIT-tard

Wise dragon said...


On Twitter:

Lazaridis on why the "EHG male / CHG female" argument doesn't hold very well:

Tiger Mike said...

I think we are still shy on samples of R-M269 from the right time period. We need to know more about any samples of this lineage from 5000 to 3500 BC.
Later samples could mean anything.

Davidski said...

@MikeW

The entire R1b phylogeny prior to 4,000 BCE in ancient DNA is located in Europe.

And M269 entered Asia along with steppe ancestry.

So why would M269 be an outlier from its relatives?

Davidski said...

@elliv

What is EHG ancestery doing in Armenia 4000 BCE?

Steppe people started moving into West Asia at this time.

That's the Sredny Stog period.

Davidski said...

@Aram

The Y16419 came with R1b Z2103 to Armenia. Why they should have different autosomes?

Yep, this argument in the paper involving Y16419 confused the crap out of me.

It's just a red herring.

vAsiSTha said...

@Rob

"your models are crap. The pRight set is inadequate to distinguish West Asian streams of ancestry, so I dont care what they show"

I don't care, go convince any geneticist that you are right, you seem to be losing the battle badly and are losing your mind.

"Like I said, you need to understand the region & samples in question in order to analyse them. I dont have the patience to listen to the BS from some smelly OIT-tard"

I don't need to take lectures from a foolish racist PoS.

vAsiSTha said...

@Vara

"This is nice for Near Eastern studies.

LBA Hasanlu is an eastern Hurrian site with Indo-Aryan elite based on the golden bowl with "BMAC features". Seeing L1a almost confirms Parpola's Mundigak's tin road Mitanni hypothesis. On the other hand, Iron Age Hasanlu is mostly Urartian with a few incoming West Iranians."

+1

I have written about this on my blog

Carlos Aramayo said...

@ Davidski,

I found only one sample holding Y-hg R1b from Bronze Age period (Supplementary table S1). See Number. 639, individual I146149, dated to 2308-2129 cal BCE, TUR_SE_Gaziantep_BA. Before the paper was released, Wise dragon commented that R1b new samples in Hasanlu, Iran, held no EHG (or Steppe) ancestry. Is this the case for this individual from Turkey?

Davidski said...

@claravallensis

https://twitter.com/iosif_lazaridis/status/1563953730499878926
thoughts?


It's no longer relevant, because we now know that CHG-related ancestry was present in steppe hunter-gatherers.

https://eurogenes.blogspot.com/2022/05/population-genomics-of-stone-age.html

CHG was on the steppe >5,000 BCE. So there's no point in looking for any correlations in the Yamnaya X chromosome.

The important point is that Near Eastern Y-chromosomes didn't make any impact there. And M269 is not a Near Eastern Y-chromosome, no matter what some still claim.

Davidski said...

@Carlos Aramayo

I haven't looked at I146149.

But the Hasanlu samples do have EHG and steppe ancestry.

The reason why a lot of people, including me, thought that Lazaridis couldn't find any EHG or steppe ancestry in these samples is because of a totally pointless discussion in the paper about the lack of correlation between EHG and M269 in Armenia.

Tiger Mike said...

@Davidski

"The entire R1b phylogeny prior to 4,000 BCE in ancient DNA is located in Europe."

What ancient samples are you referring to from the R1b-M269 lineage? The phylogenetic block for R1b-M269 is up to 100 SNPs long so it is difficult to find prior to its descendants R1b-M269>L23 (where Z2103 and L51 lay) and R-M269>PF7562.

Davidski said...

@MikeW

I said R1b. Prior to 4,000 BCE, R1b is based in Europe and has nothing to do with West Asia.

R1b and M269 are clearly intrusive to West Asia and it's obvious that they arrived there from the north.

The ancient DNA is already clear on that.

Rob said...

@ vasistha

You have pontificated endlessly but now there’s R1b and I2a in Anatolia . It is 100% clear what that means, so your models aren’t needed, nor are Reich’s/ Lazarides. Harvard aren’t an authority. Like you they don’t have a grasp of basics, and linguists reject their results
Youre not qualified to discuss these matters, and your blatant twisting of facts to project your own fantasies is what is racism. Yes you reek

Tiger Mike said...

Davidski said,
"I said R1b. Prior to 4,000 BCE, R1b is based in Europe and has nothing to do with West Asia."

The R1b that I am aware of in Europe prior to the Early Bronze Age is NOT R1b-M269. Given the 100 or so SNPs in the R1b-M269 phylogenetic block, the M269- (negative) elements of R1b are not very relevant.

R1b-M269 is the only subclade of R1b that seems closely associated with Indo-European languages.

Davidski said...

By the way, the oldest M269 currently is from 4600-4000 BCE Romania.

Davidski said...

What's the oldest M269 sample in West Asia right now?

Rob said...

After the ice age due to the bottlenecks regions became very homogeneous in their uniparentals. So one part of Eurasia was full of I2, another part was full ofR1, another part was characterised by J2; another by G2.

It’s concerning some people are unable to understand this relatively trivial pattern

baickoz said...

Lazaridis wrote:

"Sex bias in the formation of Yamnaya? An objection I have seen about our model of the early Indo-European language history is that (supposedly) the Yamnaya were formed by admixture of EHG men with women from the Caucasus and this would imply a female-mediated language spread.

As far as I can tell, the sum total of the evidence is that Yamnaya men are dominated by a particular lineage (R-Z2103) and patrilineages common in the Caucasus or West Asia are not found among them.

BUT, the Yamnaya were descendants of admixture that took place 4,555±297 years BCE (Fig. S5 of our paper), or ~4400-4000 BCE (Chintalapati et al. 2022; elifesciences.org/articles/77625)

R-Z2103 is inferred to have been formed 6100ybp with a TMRCA of 5400ybp (yfull.com/tree/R-Z2103/) slightly preceding the expansionary phase of the Yamnaya/Pit Grave population

In other words, the fact that a novel patriline became very successful during the 4th millennium before the late 4th millennium BCE emergence of the Yamnaya tells us nothing about the chrY composition of the late 5th millennium BCE population of the steppe.

A direct test of the sex bias hypothesis is to fit the same model on the autosomes and chrX; since the chrX spends 2/3 of its life in females, under the assumption that CHG ancestry is mediated by females we'd expect more CHG ancestry on chrX than autosomes

The simple CHG-EHG model gives a CHG estimate of:

51.9+/- 1.3% (autosomes)
34.2+/- 8.5% (chrX)

In other words, the evidence is (2.1 s.e.) in favor of male CHG bias and _not_ the opposite

The evidence for male CHG bias is not super strong so we did not dwell on this point in the Southern Arc paper. But, I thought it would be useful to report here as I want people to be aware that the data don't point to a male EHG:female CHG mix and if anything the opposite."

What do you think?

Arza said...

@ Rob

If you're thinking of TUR_SE_Gaziantep_BA:I14649 it's R1b (xR-P297) [Yfull] or R1b1 (xR-P297) [ISOGG].

Tiger Mike said...

Davidski said...
"By the way, the oldest M269 currently is from 4600-4000 BCE Romania."

That's a questionable call but would be important if accurate. Remember the very long string in R1b-M269's phylogenetic block, plus the large delete. Having a SNP marginally called out of this string doesn't mean much.


Davidski said...
"What's the oldest M269 sample in West Asia right now?"

There really isn't any prior to L23. That's my point. We don't know. I've tweeted to Iosif Izaridis the same. I don't think he has much of answer to this, just speculation.


Davidski said...

@baickoz

It's a pointless discussion, because significant CHG-related ancestry existed in >5,000 BCE Eastern European hunter-gatherers that contributed ancestry to Yamnaya.

So there won't be any clear CHG vs EHG bias on the Yamnaya X, even if some women from south of the Caucasus contributed to the Yamnaya population (and they may have).

Rob said...

@ Arza

Yeah might be R1b-V3616 like in Arslantepe & the Khvalnysk network. Probably an early movement from steppe. In this case, this individual, and groups like him, appear to have become rapidly assimilated into local culture & language. Might not even have been PIE to begin with

Still, people were obviously making their way into Anatolia from both sides of the steppe. They didnt halt at Lake Van or Ezero.

vAsiSTha said...

"It's a pointless discussion, because significant CHG-related ancestry existed in >5,000 BCE Eastern European hunter-gatherers that contributed ancestry to Yamnaya."

It's not pointless, it just has to be understood correctly. Laziridis' assertion of male mediated chg in steppe applies collectively to both waves.

If one wants, they can test the x chrom for khvalynsk (and middle Don samples when they come out) and figure out the final answer of which sex mediated what in each wave.

Davidski said...

Khvalynsk is irrelevant.

Yamnaya is an offshoot of Sredny Stog from Ukraine.

Sredny Stog has as much CHG as Yamnaya, and the ancestors of Sredny Stog may have had more.

But CHG may have arrived on the steppe with women in any case.

vAsiSTha said...

@davidski

I mean choose whatever when samples become available.
Just identify population with first wave of CHG/Iran, say Pop ABC. Check their X vs autosomal.

Then check Steppe X chrom vs autosomal for model ABC + southern ancestry

Rob said...

It’s irrelevant. Yamnaya Acquired CHG from the lower Volga region, not directly from the Caucasus or Iran. It’s therefore decoupled from the original Caucasian patrilines

Davidski said...

Foragers in the Middle Don already had plenty of CHG, so it's a mystery where Sredny Stog got its CHG.

But it must have been pretty early, even if in separate waves.

Rob said...

Like - West Asian farmer ancestry was brought on mass to Britain. But this came from France linked to “ west European” I2a

Matt said...

I don't think the X vs A is that informative because you could have layers on this thing.

Say you have a situation where CHG enters the steppe via male migration, involving EHG females.

Then you have a situation that's the opposite of this where EHG men take wives from the admixed group to form another new group.

This happens in a patrilocal environment, and then the second group (EHG men+admixed female) expands as we know the Yamnaya to have done, with some strong founder effects on a particular y-dna lineage.

I'm not sure you would find any signal in either direction on the X:A, and you'd find a steppe Y chromosome.

vAsiSTha said...

If they're 1 shot admixtures or close, then it will work.
But if they're long-drawn admixture processes, results might get wonky.

But as long as std. errors. are contained, it should be fine.

Copper Axe said...

@Matt

"Say you have a situation where CHG enters the steppe via male migration, involving EHG females.

Then you have a situation that's the opposite of this where EHG men take wives from the admixed group to form another new group."

I recall Ancestralwhispers mentioning that something like that must've occured because there is a lack of CHG-associated mtdna lineages. I cant comment because I know nothing about mtdna, lets blame it on the "slow mutation rate" not triggering my interests.

I mean with the small population sizes of forsgers all sorts of scenarios are possible.

To go on a hit of a tangent, archaeology definitely shows a movement of "lithics" from the caucasus towards the volga in the later mesolithic. I covered that in one of my blog entries. But what I also recently discovered is that there were movements of Zarzian lithics on the eastern side of the Caspian towards the trans-Urals, this movement could just be a technological transfer. The archaeologists suggested migration but to be honest soviet era russian archaeologists ascribe everything to migrations (funnily enough more often correct than not).

Anyways my point being is that I dont understand why geneticists keep "postponing" the movement of CHG type populations in Europe when we have plenty of signs that we should look towards older connections. If possible they should look into the post-LGM period with the caspian sea flooding and retracting again, it should be relevant for EHGs, WSHGs, as well as CHG or EHG/CHG mixed pops.

It is kind of weird how all the sampling of the meso/neolithic seems to circle around the southern russian areas between the black and caspian seas. We have EHGs from the northern russian forest zones, Ukrainian meso/neolithic samples, several CHG genomes, mesolithic Iranians from Hotu etc. But the area I'm talking about right now in terms of foragers is only represented by those later steppe_eneolithic samples, the upcoming middle Don forager samples from Allentoft as well as upcoming pottery neolithic samples from the Volga. Nothing from the mesolithic however. Does anyone know why? Do we lack remains? Archaeologically it is a bit understudied compared to other areas so maybe that is why? To be fair Central asia has nothing in terms of foragers either....

Sabmiester said...

@Steven

"Does anyone have a complete list of the haplogroups in this study?"

https://docs.google.com/spreadsheets/d/1SDNwXADx4lxAr5myxOdJkLHw3K_CgzFibIKyrz-jhBU/edit?fbclid=IwAR2jsiJgmMxuk6HSBGEORw1tccijcmHeSeJSVdalwz8RHN9drXp7-uKbCQY#gid=56320326

Matt said...

A quick comparison of Lazaridis f4admix method with the same model in Global25+Vahaduo.

Lazaridis put the proportions for the f4admix model in the Supplementary Table 5, so I found matching IDs in the Global_25 (including the new set uploaded by Davidski this week from these papers), which left 1524/1852 (82%) of the modelled samples.

Then ran a model to estimate with the same 5x sources (CHG, EHG, Levant_PPN, Iron_Gates, Barcin_N).

Leaving aside whether they should be modelling these populations with these 5 sources and not doing proximal modelling and comparisons (IMO they shouldn't!).

Comparison of the proportions here, with a linear regression between the two: https://imgur.com/a/Mazx0co

In general the regressions are quite high between components (r2 - CHG: 0.94, EHG: 0.93, Levant: 0.79, Iron_Gates: 0.98, Barcin 0.9). The shakiest is the Levant proportion.

However, the slope of each is different. These slopes imply: 1) a substantially higher level of CHG in samples in f4admix, 2) a lower level of EHG in f4admix (this is still a slope of 0.9 though, so it's not a very large difference), 3) a lower level of Levant_N in f4admix, 4) a similar level of Iron_Gates, 5) a lower level of Barcin.

Generally these are pretty linear with noise, exceptions being that 1) Khvalynsk looks to break the trend of having less EHG in the f4admix method (so does steppe_maykop but that's more of a lack of West Siberian?), 2) f4admix seems to estimate that continuity is lower in the Levant after the Neolithic.

Looking at barplots, the f4admix seems to be more random or noisy with regard to Levant/Barcin in populations, including in donation to Afanasievo and Yamnaya. Even as relatively sympathetic to models where Caucasus populations can provide a plausible proximal source differentiating Yamnaya from Steppe_Eneolithic (with additional EHG ancestry), I don't think the unstable proportions of Levant donation are greatest evidence.

Davidski said...

I just ran some tests and Yamnaya Samara is coming out with BGR_Dzhulyunitsa_N ancestry, rather than anything Armenian.

Catacomb and maybe late Yamnaya might have some Caucasus ancestry.

I'll try and check this out in much more detail.

epoch said...

Does that autosomal/X ratio that he tweeted about also appear in Eneolithic Steppe, in Khvalynsk, Dereivka and in early CWC? Because the theory that a founder effect erased the Caucasus-related Y-DNA haplogroups in favour of Z2103 has this downside: There are other haplogroups that did the same: R1a Z93, that Eneolithic clade. We have the rumour of J1 in Khvalynsk, IIRC.

Rob said...

@ baickoz - It's BS. mental gymnastics

Synome said...

I think the Yhg issue needs to be strongly informed by the fact that the Steppe societies were from the evidence strongly patrilineal clans.

Over time, competition between clans reduces Y diversity and can quickly lead to near fixation of a single Yhg throughout a population. This is Zeng 2018, which seems to be consistency ignored in new papers but which I think provides the most plausible explanation of Y bottlenecks in the Late Neolithic/EBA worldwide.

Thus, there didn't necessarily need to be differential mating. The action was in group competition and differential male reproduction after the period of admixture.

alex said...

"Samuel Andrews said...
@ancestralwhispers,

Yeah it is all about Armenia and the Orient. Europe isn't the origin of anything cool.

This is why the vast majority of Bronze age, Iron age Armenian males belong to European lineages. R1b and I2a.

August 27, 2022 at 12:01 PM
Samuel Andrews said...
I think the high frequency of European Y DNA in ancient Armenia is the coolest discovery in this paper imo.

It makes it so apparent that Armenians were a Indo-European isolate living in southwest Asia.

August 27, 2022 at 12:06 PM"

The distance between R-L23's likely origin in the southern Russian steppe is much, much closer to Armenia than Britain, where I assume your paternal ancestry is from. R-L23 is no more indigenous to western Europe than it is to Armenia.

Samuel Andrews said...

R1b1a, from which L23 derives, is totally indigenous to Central-Eastern Europe. As Davidski has said the phnology of R1b before 4000 BC is entirely located within Europe.

It was common at least as far west as Serbia during the Mesolithic. It was not only common in the 'far-eastern' part of Europe. So it would be wrong to say R1b's native range is as close to western Asia as to western Europe. It is closer to Western Europe.

I mean, the oldest sample of R1b actually comes from Western Europe. It comes from Italy.

So yeah, the movement of R1b south of the Caucasus into southern Asia represents the movement of a totally indigenous European lineage into southern Asia.

Samuel Andrews said...

@Alex,

Ancient Armenian males belong almost entirely to yHGs R1b and I2. Both of which are totally European.

I'll demonstrate

Nearly all Mesolithic hunter gatherers from Latvia, Serbia, Ukraine belong to R1b and I2.

The majority Bronze age Unetice culture in central Europe belong to R1b and I2.

All the males from Bronze age Tollense site in Germany belong to R1b and I2.

So, when you see a population in the Middle East with the same Y DNA genepool as people deep in Europe going back to the Mesolithic, it is best described as a weird isolate in southern Asia whose paternal lineages are of European origin.

Samuel Andrews said...

@Alex,

My point is y-haplogroup R1b1a as a whole is associated with much of the continent of Europe not just the eastern part near Asia.

It is certainly more indigenous to Europe as a whole than to Armenia.

Samuel Andrews said...

Or I should say.....

R1b1a is more indigenous to Europe as a whole than it is to Armenia.

It can accurately be described as a European lineage.

Agelmund said...

Why are they pushing so hard for an Anatolian origin? Does the thought of PIE ethnogenesis being an indigenous phenomenon of Eastern Europe separate from Near Eastern civilisation cause that much butthurt?

Agelmund said...

"The distance between R-L23's likely origin in the southern Russian steppe is much, much closer to Armenia than Britain, where I assume your paternal ancestry is from. R-L23 is no more indigenous to western Europe than it is to Armenia."

The people in the British Isles have like 40-50% autosomal DNA from the people who spread it while the average Armenian has less than 10%. R1b-M269, especially R1b-u106 alongside R1a is native to the steppes and forests of Eastern Europe which is located in Continental Europe. Armenia on the other hand is behind a colossal mountain wall yhat has genetically and culturally separated Europe from the Near East for millenia. . Funny how people used to claim West Euro R1b had nothing to do with Indo-Europeans and now they're stating that West Euro R1b has nothing to do with Europe. Is there an antonym for "Eurocentric" where people are infuriated when facts point towards a European source for the spread of culture and heritage? Because it seems to me that such an attitude infects quite a lot of people

Andrzejewski said...

@Samuel Andrews “ Nearly all Mesolithic hunter gatherers from Latvia, Serbia, Ukraine belong to R1b and I2.

The majority Bronze age Unetice culture in central Europe belong to R1b and I2.”

Exactly. Dnieper Donetsk + Bug Dniester male Haplogroups.

But then Sredny Stog and Corded Ware were R1a1.

Which makes me think that a R1a pop in Forrest Steppe with up to 20% EEF ratio started Pre-Proto-IE.

AWood said...

I thought there were a few samples of R-Z2103 in Neolithic Ukraine which had low or zero CHG, and were basically pure EHG. Am I misremembering? This kind of kills the M269 from the Middle East..

In terms of no EHG in the Middle Eastern samples. That's probably because it was a Yamnaya population that arrived, not an EHG one. I think modern Europeans prefer Yamnaya or CWC ancestry rather than pure EHG if I'm not mistaken.

Andrzejewski said...

@AWood “I thought there were a few samples of R-Z2103 in Neolithic Ukraine which had low or zero CHG, and were basically pure EHG. Am I misremembering? This kind of kills the M269 from the Middle East..“

Are you talking about Mesolithic Bug Dniester and Dnieper Donetsk?