If, like Iosif Lazaridis, you subscribe to the idea that the Yamnaya people carry early Anatolian farmer-related admixture that spread into Eastern Europe via the Caucasus, then I've got great news for you.
We now have a human sample from the Eneolithic site of Nalchik in the North Caucasus, labeled NL122, that packs well over a quarter of this type of ancestry (see
here). Below is a quick G25/Vahaduo model to illustrate the point (please note that Turkey_N = early Anatolian farmers).
Target: Nalchik_Eneolithic:NL122
Distance: 2.1934% / 0.02193447
60.8 Russia_Steppe_Eneolithic
26.2 Turkey_N
13.0 Georgia_Kotias
On the other hand, if, again like Iosif Lazaridis, you subscribe to the idea that the Indo-European language spread into Eastern Europe via the Caucasus in association with this early Anatolian farmer-related admixture, then I've got terrible news for you.
That's because NL122 is apparently dated to a whopping 5197-4850 BCE (see
here). This dating might be somewhat bloated, possibly due to what's known as the reservoir effect, because the Nalchik archeological site is generally carbon dated to 4840–4820 BCE.
However, even with the younger dating, this would still mean that early Anatolian farmer-related ancestry arrived in the North Caucasus, and thus in Eastern Europe, around 4,800 BCE at the latest. That's surprisingly early, and just too early to be relevant to any sort of Indo-European expansion from a necessarily even earlier Proto-Indo-Anatolian homeland somewhere south of the Caucasus.
This means that NL122 effectively debunks Iosif Lazaridis' Indo-Anatolian hypothesis. Unless, that is, Iosif can provide evidence for a more convoluted scenario, in which there are at least two early Anatolian farmer-related expansions into Eastern Europe via the Caucasus, and the expansion relevant to the arrival of Indo-European speech came well after 5,000 BCE.
I haven't done any detailed analyses of NL122 with formal stats and qpAdm. But my G25/Vahaduo runs suggest that it might be possible to model the ancestry of the Yamnaya people with around 10% admixture from a population similar to NL122.
Target: Russia_Samara_EBA_Yamnaya
Distance: 3.4123% / 0.03412328
72.6 Russia_Progress_Eneolithic
18.2 Ukraine_N
9.2 Nalchik_Eneolithic
However, I don't subscribe to the idea that the Yamnaya people carry early Anatolian farmer-related admixture that spread into Eastern Europe via the Caucasus (on top of what is already found in Progress Eneolithic). Based on basic logic and a wide range of my own analyses, I believe that they acquired this type of ancestry from early European farmers, probably associated with the Trypillia culture. For instance...
Target: Russia_Samara_EBA_Yamnaya
Distance: 3.2481% / 0.03248061
80.2 Russia_Progress_Eneolithic
13.6 Ukraine_Neolithic
6.2 Ukraine_VertebaCave_MLTrypillia
0.0 Nalchik_Eneolithic
Another way to show this is with a Principal Component Analysis (PCA) that highlights a Yamnaya cline made up of the Yamnaya, Steppe Eneolithic and Ukraine Neolithic samples. As you can see, dear reader, there's no special relationship between the Yamnaya cline and Nalchik_Eneolithic. The Yamnaya samples, which are sitting near the eastern end of the Yamnaya cline, instead seem to show a subtle shift towards the Trypillian farmers.
Indeed, I also don't exactly understand the recent infatuation among many academics, especially Iosif Lazaridis and his colleagues, with trying to put the Proto-Indo-Anatolian homeland somewhere south of the Caucasus. Considering all of the available multidisciplinary data, I'd say it still makes perfect sense to put it in the Sredny Stog culture of the North Pontic steppe, in what is now Ukraine.
Please note that all of the G25 coordinates used in my models and the PCA are available
HERE.
See also...
The Caucasus is a semipermeable barrier to gene flow
278 comments:
«Oldest ‹Older 201 – 278 of 278@Gabru
It's not an assumption.
Have a close look at Khvalynsk I0434 (~5,000 BCE). Steppe Neolithic is very similar.
This is the story for Golubaya_Krinitsa, but steppe neolithic is not the most ideal model for this, but so far the best thing that has happened in terms of distances for Golubaya_Krinitsa, even better than that of progress
https://postimg.cc/5Q3xWkyD
@All
Does anyone know of a paper or an article with a description of figurines found at sites with a Cernavoda I culture context?
@DragonHermit
The connection of the Besiktas assemblage is explicitly tied to the northeast Balkans, likely for good reasons. Considering that the Huriyet article showed stylized figurines I can see why. These are typical for the Balkans. What is odd though, is that figurines are more tied to cultures before the collapse. From what I have read only Cernavoda I sites sometimes come up with figurines, but far less abundant than before.
Maybe be a sign of some sort of ethnogenesis?
By the way, Suvorovo is 4.4ky BC to 4ky BC.
Visualisation of I0434's position compared to other Steppe_Eneo related samples: https://imgur.com/a/XWOP5Oh
Hello everyone. If you are interested, here is a text file with all the ancient coordinates of G25 sorted by date.
https://www.dropbox.com/scl/fi/0q39lrsynq7prjc7mm8gq/G25-Ancients.txt?rlkey=33i5tycf3nd6glv1w7z6dleco&e=2&dl=0
Target: RUS_Khvalynsk_En:I0434
Distance: 2.2769% / 0.02276859
61.0 RUS_Progress_En
25.8 RUS_Samara_HG
13.2 RUS_Sosonivoy_HG
@Matt,
Harvard is full of woke/DEI hire idiotes now, just look at their recent scandal with their president. Sarazm 4290 does not have 15% AHG, and it has too high CHG ancestry to be ancestral to any Ip groupe, who do not have this .
sample: Sarazm En:I4290
distance: 4.9606
Ganj_Dareh_N: 53
AASI: 1
RUS_AfontovaGora3: 17.5
Marmara_Barcin_N: 0
GEO_CHG: 28.5
@Arsen
'of course, this is a stupid statement, but it turns out that the Neolithic steppe was similar to modern Tajiks and Pamiris 😅'
Aren't Rushan and Shughnani Tajiks the South Asians with the highest Sintashta ancestry? If you google them they are the ones who sometimes have a European like appearance, and even sometimes a native american like appearance in addition to the persian look. Perhaps the cocktail of Steppe_MLBA and Iran Neolithic Sarazm-ey DNA makes them autosomally more similar to Steppe_EMBA. I am just a layman so I don't know if there is a big or negligible difference between things like similar ancestry and direct descent ancestry, would there be a large difference between a population with direct ancestry that plots closely to Steppe Eneolithic on PCA versus a population with different ancestries that happen to make it close to Steppe Eneolithic on PCA? A half-French half-Polish person might resemble a North German on a PCA but they probably are less likely to have blond hair for example
There is some literature suggesting the earliest Kurgans in Bulgaria show some continuity with local traditions, compared to the slightly later ones.
We know that by 2800BC nearly fully Yamnaya-like individuals can be found in Northern Thrace (Boyanovo).
https://www.academia.edu/35796532/Pit_Graves_Yamnaya_and_Kurgans_along_the_Lower_Danube_Disentangling_IV_th_and_III_rd_Millennium_BC_Burial_Customs_Equipment_and_Chronology
https://www.researchgate.net/publication/285673711_Pit_graves_in_Bulgaria_and_the_Yamnaya_Culture
https://www.academia.edu/106869532/The_old_man_and_the_three_babi_an_exceptional_burial_from_the_Pamukli_bair_barrow_near_Malomirovo_southeast_Bulgaria
I don’t think that the resulting steppe n is ideal for the population that lived at that time in the foothills, it lacks something...
there seems to be too much ehg in it. I can’t get the Nalchik sample out of it at normal distances
Yes, you're right.
@Scott G
plus, we should not forget that these peoples have a bit of Tarim origin associated with ANE populations
@Gio. i wondered which Latin words you had in mind. And what period.
Remember, a fair scatter of the Latin lexicon as we have it — the Romans had adopted from Etruscan.
And we know the Etruscans already (whatever the nature of their baseline language) had adopted chunks of Punic (Semitic) lexicon and culture (eg religious and trade) and that some of that also filtered to them through Greek contact. The Phoenicians, after all, were out and about while Rome was still relatively obscure and Latin localised.
So anyway, there are at least three other contributions to Latin lexicon when we first see it, one or possibly two of those non-IE.
Just a point of logic.
@Gio. i wondered which Latin words you had in mind. And what period.
Remember, a fair scatter of the Latin lexicon as we have it — the Romans had from Etruscan.
The Etruscans already had themselves adopted chunks of Punic (Semitic) lexicon and culture (eg religious and trade).
Some of that also filtered to them through Greek contact.
The Phoenicians after all (and varieties of Greek) were out and about while Rome was still relatively obscure and Latin localised.
So anyway, three other contributions to Latin lexicon when we first see it, one or possibly two of those non-IE.
Just a point of logic.
target left weight se z
1 Russia_Khvalynsk_Eneolithic_I0434 Russia_HG_Karelia 0.242 0.129 1.88
2 Russia_Khvalynsk_Eneolithic_I0434 Russia_HG_Tyumen 0.373 0.123 3.03
3 Russia_Khvalynsk_Eneolithic_I0434 Russia_Caucasus_Eneolithic 0.384 0.0558 6.89
pat wt dof chisq p f4rank Russia_HG_Karelia Russia_HG_Tyumen Russia_Caucasus_Eneolithic feasible
1 000 0 9 3.53 9.40e- 1 2 0.242 0.373 0.384 TRUE
2 001 1 10 50.6 2.04e- 7 1 -0.432 1.43 NA FALSE
3 010 1 10 20.9 2.19e- 2 1 0.588 NA 0.412 TRUE
4 100 1 10 9.42 4.93e- 1 1 NA 0.587 0.413 TRUE
5 011 2 11 82.6 4.68e-13 0 1 NA NA TRUE
6 101 2 11 63.4 2.11e- 9 0 NA 1 NA TRUE
7 110 2 11 104. 2.94e-17 0 NA NA 1 TRUE
@William Anderson
and who is the author of this work done? Many, many thanks to him for his respect, now I have in my hands almost all the published samples + dates for them.
@ epoch
Yes C-T figurines & motifs are mediated via Cernavoda, who married C-T women extensively.
-------------
Like I said 10 years ago, ths early IE split probably came from I2, EEF-rich end of the westernmost steppe. Both volga-tards & Irano-tards gave me shit, but Untys shouldn't poke the Gods.
@ Meshiko
Sure it has, but nice work
Is that Ilipinar bundled in as well or just Barcin C ?
@Meshiko
https://eurogenes.blogspot.com/2020/06/perhaps-hint-of-things-to-come_2.html
https://eurogenes.blogspot.com/2020/06/the-precursor-of-trojans.html
Yes, the Allen DB has them together, although IMO they should be split, as Barcin_C is somewhat distinctive
@Matt
Do you know if there's a list of samples that make up these outgroups in the Kerdoncuff preprint?
Ethiopia_4500BP.SG’, ‘WEHG’, ‘EEHG’, ‘ESHG’, ‘Dai.DG’, ‘Russia_Ust_Ishim_HG.DG’, ‘Iran_Mesolithic_BeltCave’ and ‘Israel_Natufian
@Gabru
Khvalynsk is older than Progress, so your model is back to front in terms of dates and genetics.
You'll see what I mean sooner or later.
@ Wee
I understand very well what you say, because I spent all my life in disproving who tried to demonstrate that Italians don't exist, they always came from elsewhere, paper about the Imperial Rome, etc etc, but after the paper on Erfurt 1250 AD about the Jews demonstrated that they were 70% Italian, 15% from eastern Europe and 15% from Northern Africa, Middle East, Iran but with no certainty that they were "Jews". About the Phoenicians and the Zalloua's papers I wrote tons of letters and also in this case I demonstratec that the other way around happened. That Etruscans did come from Anatolia none thinks now that after the aDNA tests which found above all R-U152 and even R-L2. Certainly Etruscan had some words from Greek. From Phoenician perhaps some local words, but the only one Phoenician word in Sardinian is "tsippiri". No other. Anyway the "Dienekes' Anthropology blog" is always in the web whereas others where I wrote about 20000 letters aren't, and you may search for my letters.
@ Wee
This is a discussion of pretty much ten years ago. You may see what is alive and what is dead in what I wrote.
http://dienekes.blogspot.com/2015/05/2500-year-old-etruscans.html
@ Wee
This is a discussion upon Euphratic in 2011. Probably I didn't write in the blog yet:
http://dienekes.blogspot.com/2011/05/case-for-euphratic.html
http://dienekes.blogspot.com/2013/07/population-history-of-middle-euphrates.html
This thread had only one comment... in andrew is wee...
http://dienekes.blogspot.com/2015/02/italic-eteocretan-sea-peoples.html
this is a probable Italic language in Creta among others
This post could be interesting to read:
Gioiello said...
Simon_W, what you say to Rokus and to me is plausible. I wrote that in replying you, but I didn't send that to Dienekes, just because the doubts you pose were also mine:
"I thank Rokus for his defence of a possible origin of this test from "Italic" with good arguments from a linguistic point of view. Of course I should write a book for answering Simon W. Anyway the book of Woudhuizen I quoted above (The Ethnicity of the Sea Peoples) should be an unsuppressable start point, and Woudhuizen didn't recognize as "Italic" peoples amongst them that the Shardana, the Siculs and the Teresh/Etruscans (who of course didn't speak an Italic language).
I'd want to say only one thing, that the possible explications of "itspha" from an Italic "eitiuva", in the meaning of "movable property" (from the IE root *i/ei-" = to go) seems having developed in contact with a Lycian language, with a palatalization of "t" due to the "i" and p/b from "w" like in Lydian "dbi" (if I remember well) from IE *dwi-".
I read again the pages of Woudhuizen in "The Ethnicity of the Sea Peoples" and the possible link between Etruscan and Lydian (thus the possible origin of the Etruscans from the Aegean Sea) actually merits to be deepen (I know this hypothesis from at least 45 years, when I read the paper of Georgev about that in the "Biblioteca Nazionale" of Florence), but one should write a paper or a book about that and not only a post.
About the genetic point of view I could say that, even though I appreciate the YFull team work and I sent them also my Full Genome and use daily their site (but they didn't recognize a SNP to me I have: I am YF2873 under R-Z2110 in the YFull tree), I esteem their dates underestimated for 1000/1500 years as to L51. I wrote a lot about the hypothesis which links Tuscan L51/CTS6889- with Iberia and Ireland due to the migration of 7500 years ago, and the L51/CTS6889+ with an expansion from Italy to Central Europe.
Of course your hypothesis is also that of all the Anthrogenicians who banned me and I am waiting like them for the aDNA response. What I appreciate of you is that we may discuss and wait for the scientific proofs without banning anyone. I have written a lot also about my R/L23/Z2110 with 92 private SNPs and the link I found so far with a Basque and an Englishman, all in Western Europe, and also about the fact that my acquired cousin Fabrizio Federighi (R-M269/PF7566/PF7569) from Tuscany has a MRCA with me at least 9300 years ago. Of course it is possible that our ancestors came from elsewhere and our lines live at a few kms from hundreds or thousands of years.
Monday, March 09, 2015 7:57:00 am
@Gabru
here above, the guy threw a link to g25 coordinates for almost all published ones , plus dates and coverage
when composing equations, you can focus on them
https://www.dropbox.com/scl/fi/0q39lrsynq7prjc7mm8gq/G25-Ancients.txt?rlkey=33i5tycf3nd6glv1w7z6dleco&e=2&dl=0
The supposed Proto-Anatolian samples, such as Cernavoda and Usatovo, unlike the later Yamnaya-like samples, don't really require any of the Khvalynsk/Steppe_N shift, or even Golubaya Krinitsa. Instead, they can be neatly modeled as Nalchik + EEF + some Ukr_N.
https://i.imgur.com/ScZVPri.png
@Davidski, no that's not specified; the written supplement doesn't go beyond that level of detail, and the supplementary tables are all around their more biomedically relevant results / archaic ancestry results.
@ancestralwhispers.org
As I demonstrated here, Yamnaya doesn't have any direct Nalchik ancestry.
You're missing some key samples in your models, like the more Progress-like Usatovo and early Sredny Stog.
@Davidski
Yamnaya probably doesn't. The point was to demonstrate whether Cernavoda and Cernavoda-like Mayaki samples have any additional Steppe Neolithic ancestry. What do you think?
Target: RUS_Steppe_Maykop
P-Value: 0.793
57.2 RUS_Tyumen_HG
18.1 RUS_Samara_HG
17.2 GEO_Kotias_Klde_HG
7.5 TUR_Barcin_N
https://pastebin.com/3Utwejz5
Assuming the model is legitimate, we can draw the following conclusions:
Cernavoda and Usatovo are irrelevant to the spread of Proto-Anatolian.
Or:
If Cernavoda and Usatovo are relevant to the spread of Proto-Anatolian, then the Nalchik + Ukr_N signal could be the vector of PIE language spread. It might represent the genetic profile of Novodanilovka chiefs. If that is the case, then Yamnaya and Sredny Stog might have direct Nalchik ancestry. If they do not, then it could be plausible that they were subjects to an elite language spread by the Novodanilovka chiefs. Further examination Novodanilovka, especially Y-DNA, would provide valuable insights.
@CHG
All these calculations are missing the Neolithic of the Caucasian steppe (it is clearly not similar to what I threw above) and the Mesolithic of the North Caucasian hunters, they would resolve all disputes and misunderstandings
It turns out one of the Usatove samples described in Penske was buried with figurines. From the Sup Info:
"The male child (7-10 years old) laid in a crouched position on the left side with the skull to the northeast. His arms were bent with the hands in front of his face. Patches of rot were traced on the bones of the limbs. The inventory
included a copper awl, a vessel made of shell-tempered clay, two anthropomorphic figurines and four jet beads."
MtDNA: U5a1a1, Y-DNA: G2a2b2a, G-P303
It's fairly nonsensical to average every Usatovo and every Cernavoda sample if that's what you did. It's very clear some of them cluster differently due to caucasus ancestry (KTLb).
Cernavoda-Usatovo are almost all I2a apart from single cases of R1b-V1636, J2, and G2a. Autosomically they are diverse, exist in a long cline. Some individuals might have Majkop type ancestry, mostly female mediated. This echoes evidence of trade between Usatovo and Majkop cultural trajectories.
Anyone looking at Cernavoda -Majaki & Usatavo sites should separate the late samples (< 3100 BC)- theyre Yamnaya period,
@EthanR
If you examine the targets, Usatovo was divided into two clusters: one representing the Nalchik-rich cluster and the other the Yamnaya-like cluster. The latter likely comprises individuals who are just Yamnaya at Mayaki, considering the burial context.
Here's a clearer demonstration of the composition of these clusters:
https://i.imgur.com/IojAAwu.png
However, there is a strong argument for considering I2a KTL001 and KTL006 as outliers within that cluster, as they are the only ones who score the Steppe_N proxy that is Khvalynsk_I0434. Therefore, they should indeed be separated from the average.
Additionally, MAJ023 appears to be an intermediate between the two Mayaki clusters.
Regarding the main KTL cluster harboring Caucasian ancestry, if not from Nalchik, what could be the likely mediator of said ancestry? Nalchik seems a more logical source than seafarers from the Caucasus, as some have postulated here. Anatolia also seems an unlikely source, given its lack of Levant ancestry.
Usatovo had contacts with Maykop.
@Davidski
Usatovo, yes, but not Cernavoda. The non-Yamnaya Mayaki cluster appears to be a continuation of the main KTL cluster, possibly indicating the source of Caucasian affinity, rather than Maykop.
@Rob
are you talking about UKR_Usatove_En samples? judging by what I see, as a southern proxy, they would be more suited to a sample of something like progress, but with less EHG
and again this is not pure CHG but a slightly Iranian shift
@Rob
Scroll up.
https://i.imgur.com/icY8qlI.png
look at the dates
The Yamnaya-like samples from Mayaki are Yamnaya, not Usatovo. Despite being from the same site, they are hundreds of years younger
The Caucasus affinity in KTL can’t be from Nalchik, there’s also a large time gap (Nalchik being much older). It’s probably from Majkop
Guys you are putting too much stock in what might be a misdated sample: "In the Nalčik grave we were able to take one piece of wood which dates the kurgan to 4410±25 BP,
i.e.
3092-2943 calBC 1
σor 3262-2924 calBC 2
σ, that is at the very end of the 4th
millennium BC (Fig. 16).Radiocarbon dating thus confirms a date late in the Majkop sequence for the complex".
In this case, the 5197-4850 date is kya not BCE.
https://www.academia.edu/33559720/The_Great_Kurgan_from_Nal%C4%8Dik_A_Preliminary_Report_mit_Andrej_Belinskij_und_Sabine_Reinhold_
@ancestralwhispers.org
Why didn't you use a Chalcolithic or Neolithic Balkans sample as EEF?
@Rob
"Every new sample needs to be reinvestigated, I cant see Nalchik included in your old run you've posted here.
Moreover, I remember your model for Piedmont steppe previously used the crap quality Khvalynsk individual, which would give false positives with qpAdm. The key is figuring out Piedmont/Vonuchka"
I ran 286 rotating qpAdm models with a similar setup to Lazaridis et al., 2022. Optimal model (p = 0.152735) = 80.9±1.6% RUS_Eneol_Piedmont, 8.7±1.3% UKR_N and 10.5±1.2% UKR_VertebaCave_MLTrypillia.
Fixed outgroups: Mbuti.DG, ISR_Natufian_EpiP, Levant_PPN, MAR_Taforalt_EpiP, RUS_AfontovaGora3, SRB_Iron_Gates_HG, TUR_C_AşıklıHöyük_PPN, TUR_C_Boncuklu_PPN, TUR_C_Çatalhöyük_N, TUR_Marmara_Barcın_N, TUR_Pınarbaşı_EpiP, WHG
Rotating sources/outgroups: ARM_Aknashen_N, ARM_Masis_Blur_N, AZE_N, CHG, EHG, IRN_Ganj_Dareh_N, RUS_Eneol_Piedmont, UKR_N, UKR_VertebaCave_MLTrypillia, RUS_Nalchik_Eneol, RUS_Eneol_Mountains, TUR_SE_Batman_ChL
qpAdm parameters:
details: YES
allsnps: YES
inbreed: NO
Output table
@Davidski
Thoughts?
Usatove has exactly a quarter of her ancestors related to chg-iran, excluding ehg, again excluding her Georgian origin. This is for Rob
Yeah, I agree with the comments that with these samples from Eneolithic Western Ukraine, on a cline, you really should model the whole cline.
Modelling an average point is sufficient when we have good reason to believe that they form a population that has been panmictic and where variation might just represent projection noise, and in that case to denoise and simply present results.
I'm not sure we should do so if we have obviously clinal sets of samples.
KTL A doesn't have significant Caucasian ancestry under proper modeling. Only when you erroneously model them without progress do they want the former (because both Khvalynsk and middle don are relatively "CHG" deficient, and their "EHG" ancestry is exotic (former too WSHG, latter too UKR_N))
The Yamnaya majaki samples aren't descendant of KTL. They are Z2103, higher Steppe and have zero or close to zero percent south Caucasian.
The relationship between all these is very obvious on a PCA. Yamnaya, KTL And several usatovo individuals reside on a cline between Gumelnita/tripolye and the classical Yamnaya/Afanasievo profile. For the individuals who fall off this cline, it's very evident what is going on.
@Vara
There are different burials at Nalchik.
NL122 is from the Eneolithic cemetery. So nothing to do with Maykop.
https://www.ncbi.nlm.nih.gov/sra?LinkName=biosample_sra&from_uid=39854872
David Anthony stated in his paper on "The Eneolithic cemetery at Khvalynsk on the Volga River" that there were 121 Eneolithic graves at Nalchik. It would be very bad luck if the sample they got the tooth from was dated from a later era. But it's a shame there is nothing about the burial details for the sample that could help really confirm that it's not a false Eneolithic. The y-dna seems to indicate against the sample being derived from Yamnaya, if anything, though.
NL122's unique R1b lineage does back up the claim that this is an early Eneolithic sample.
@ Vara
''3092-2943 calBC 1
σor 3262-2924 calBC 2
σ, that is at the very end of the 4th
millennium BC (Fig. 16).Radiocarbon dating thus confirms a date late in the Majkop sequence for the complex".''
There is
A) the Nalchik cemetery early Eneolithic ~ 4700 bc
3) the Grand Nalchik Kurgan ~ 3100 BC - which is just a Maykop kurgan
You're talking about the Majkop era kurgan - a single great kurgan. Different things
@ Ancestral Whisper.
Yes I saw that shortly after soz., I agree about Mayaki.
The other thing I would say is Cernavoda-Kartal are post 4000 BC, even if odd outliers push ~ 4200 BC. In which case, when there is a range of dates, Bayesian modelling confirms true date range.
Given this, if Nalchik burial ground was out of use by then, we would need to look for other sources of additiona CHG-rich ancestry. I previously went & did qpAdm models for almost every individual, but currently not at my main computer.
@ Anveṣaṇam
Ok nice work.
I personally excluded I0434 as it is an outlier in PCA and coverage, but no matter.
@ Arsen
''Usatove has exactly a quarter of her ancestors related to chg-iran, excluding ehg, again excluding her Georgian origin. This is for Rob''
Yes I know Stalin was a bad guy, but your points don't make any sense.
@Davidski
"Once those samples are published, you'll see that Progress Eneolithic is roughly 80% Steppe Neolithic and 20% Nalchik-like.
So since Progress Eneolithic has some Nalchik admixture, then it stands to reason that Yamnaya has some as well. And the correct way to look at this is that the Anatolian-related ancestry in Yamnaya that doesn't come from its Progress-like ancestry is from Trypillia farmers"
If Yamnaya has Nalchik admixture then that does kind of mean it has Southern Arc ancestry does it not? Except that unlike Lazaridis's model, it does not have any additional input post Progress and any ANF they take afterwards appears to be Trypillia-related EEF.
So his qpAdm models were after all correct in the Southern Arc paper (2% Levant_Neo and the like) just not his conclusions and timing of that admixture.
@gamerz
yes, although the thrust of Lazaridis' position is that the south cacuasian ancestry in WSH corresponds with the classical dating of when the ancestors of Anatolian broke off from archaic PIE.
@gamerz_j : still don't know about Lazaridis's modelling, but it seems like your description is accurate there, presuming what Davidski commented. As per the calculation upthread, if Nalchik is 40% "Southern Arc", and 20% in Progress En, and Progress En is ~90% in Yamnaya, then "Southern Arc" is 7% in Yamnaya. (And approx 3-4% in modern day Europeans). This period around 5500-4000 BCE seems like when its all these influences mixing into this one network (people like the Middle Don HG, Khalynsk Cemetary), and though the Eastern Anatolian ancestry in Nalchik is not a big contributor, perhaps it is there?
Then its 4000 BCE - 3000 BCE more when the pool of ancestry in the steppe has mostly fully formed? (Or at least the branch of it that later gives rises to Yamnaya/Corded - some other groups are still experiencing their own admixture dynamics as is attested by the Ukrainian Eneolithic and Steppe Maykop).
By extension it also seems like the Steppe_Maykop core cluster must have some Eastern Anatolian ancestry ultimately from Nalchik, too?
@Matt
Thanks for the reply, yea I basically am trying to ascertain the extent to which Lazaridis's models did capture reality even if he (most likely imo) is wrong about the timing of these admixtures. Well, the model that Davidski posted for Nalchik makes it look more "generic" than it is for lack of a better term, as it has substantial Levantine ancestry on top of ANF
Target: NL122_scaled
Distance: 2.4673% / 0.02467349
59.8 RUS_Progress_En
15.6 TUR_Marmara_Barcin_N
14.2 RUS_Darkveti-Meshoko_En
10.4 AZE_Caucasus_lowlands_LN
0.0 ARM_Aknashen_N
Of course that makes one think whether it is actually Nalchik in Progress but it does seem to work.
Regarding Steppe_Maykop, the core cluster does seem a similar population to me, but I am not very well versed in the archaeology of the area in order to say.
Also, something regarding Progress that I meant to ask @Davidski but if either you or @Rob have an opinion here, it would be appreciated, is whether the EHG part of Steppe_Eneo is simply less WHG than Karelia or whether there is some additional admixture going on. There are so many odd models using Tutkaul and Sarazm going around and while archaeologically (and haplogroup-wise) both seem very unlikely, I do wonder if at least Tutkaul-like pops have some role to play. And whether they could also explain the shift away from CHG that is seen in Steppe_Eneo relative to the Kotias sample etc.
In G25 and qpAdm it works, but then also models can do without it.
@Meshiko
But the issue is that if Yamnaya has Nalchik ancestry, and in the amounts specified above then it does have some Levant_N ancestry since Nalchik seems to take it. So the models were correct but the inferences on movements not
Read this. The article is in English
https://hfrir.jvolsu.com/index.php/ru/component/attachments/download/2484
Target: UZB_Zaman_Baba_N_contam:I8508
P-Value: 0.032
59.7 ± 12.1 IRN_Ganj_Dareh_N
40.3 ± 12.1 RUS_Tyumen_HG
I8508
Zaman Baba, Uzbekistan
Y-DNA = ()
mtDNA = n/a
4406 BCE
@Rob
no Rob, again you think that I have hostility towards my southern neighbors, I’m just stating the facts that not a single steppe sample has a gene with a pure mixture of Mesolithic Russian and Ukrainian hunters with Georgian hunters. Almost all of these steppe samples require a slightly Iranian shift as the south pole. And this does not mean that there was an additional cross with Iran. Just as I said, the Caucasian hunters were divided into northern and southern
and Stalin's DNA was Ossetian
@I have thoughts in my head that Sarazm, the northern and southern CHG have a common ancestor, I have no idea where he lived, but certainly, like Iranian farmers, he is connected with Western Asia
@ GamerZ
“whether the EHG part of Steppe_Eneo is simply less WHG than Karelia or whether there is some additional admixture going on. There are so many odd models using Tutkaul and Sarazm going around and while archaeologically (and haplogroup-wise) both seem very unlikely, I do wonder if at least Tutkaul-like pops have some role to play. ”
I’ve not seen stats behind “less WHG EHG” theory.
IMo it’s clear that Piedmont steppe needs something genuinely Central Asia with formal stats, because they always have an odd attraction to Pops like Mongolia -N (this is something I proposed 10 years ago when I was still a stats noob.)
There is in fact archaeological evidence of links between the southeast caspian and northwest caspian. The problem with this was that this evidence was misused by some individuals to claim this is evidence of “Jeitun migrations”, which is not the case.
Btw, I never use Sarazm, its 1000 years too late, it doesn’t give me gratification :)
Finally, with several little streams of “southern arc” ancestry we actually get a collapse of that model, because they’re all different and none are overriding. The overriding factor is the WHG admixed groups from the Dnieper -Don region EHG groups
@ arsen
“I’m just stating the facts that not a single steppe sample has a gene with a pure mixture of Mesolithic Russian and Ukrainian hunters with Georgian hunters. Almost all of these steppe samples require a slightly Iranian shift as the south pole. ”
I was the first person to discover & state that Progress En has Iran ancestry via Meshoko, on this blog in 2022.
So you’re obviously very confused about something.
Data for "Reconstructing the genetic relationship between ancient and present-day Siberian populations"
This collection includes the EIGENSTRAT format genotype data of 205 ancient individuals for the 1240K SNP panel. The individuals were previously published elsewhere but only FASTQ or BAMs were made publicly available by the original studies. We deposit the 1240K genotype call of these individuals which we used in our study titled "Reconstructing the genetic relationship between ancient and present-day Siberian populations".
https://edmond.mpg.de/dataset.xhtml?persistentId=doi:10.17617/3.QZBM1X
@gamerz_j, yeah, that's an interesting point about the broad-strokes model from David possibly slightly "genericizing" the ancestry in Nalchik (which is fine for a broad-brush model).
Still if you have AZE_Caucasus_lowlands_LN having like 25% Levant_N (and I can't remember the real number so went for something that felt on the higher end of plausibility), and Nalchik has 10% from a source like this (or we're approximating like that anyway), that's 2.5% Levant_N in Nalchik, and then you get only a small >20% pass through of that to Yamnaya, then we're in <0.5% territory. My intuition I would doubt that Lazaridis's models are necessarily actually clearly being validated very much by this finding. It's possible they are though, but seems like it could also be that this distinction just becomes too noisy to detect, and his models were only indicating in a plausible direction by luck. Depends on if you want to give those methods the benefit of the doubt perhaps. Kind of a "hot take" on my part though. If Lazaridis's models did not find such a trace in the Progress and Vonyuchka samples, I would not give him the benefit of the doubt on this one.
On your other question about the possibility of those other populations as minority(?) contributors of some flow into Steppe_Eneolithic, I guess I don't really have a good opinion on it. I would think that with groups like Tutkaul or theoretical populations that are slightly like Samara_HG with 10% Botai or something like this, which in some ways share more deep ancestry with the other parts of Steppe_Eneolithic, it's gonna be harder and harder to distinguish between geneflow from these and populations that just branched off slightly earlier in the tree from the references we have.
There seems to be a problem...
Target: TUR_Marmara_C
P-Value: 0.747
48.4% ± 5.1 TUR_Mardin_PPNA
41.1% ± 3.9 TUR_Barcin_N
10.5% ± 2.6 RUS_Samara_EBA
https://pastebin.com/APBtk62i
Target: TUR_Marmara_C
P-Value: 0.418
49.8% ± 3.1 TUR_Barcin_N
41.4% ± 4.9 ARM_Aknashen_N
8.8% ± 2.8 RUS_Progress_En
https://pastebin.com/kV6tBiEz
@Rob
this is a purely personal opinion of mine, do not take this photo for a hundred percent instance
https://postimg.cc/TK3YyprP
@ Arsen- interesting. Good luck with your research
@Mr.Davidski
Why does RISE546 look so different from the others, where does it "drift" relative to the overall Yamnaya population? What is so different about it that it models so poorly? Could there be some hidden population that was involved in the ethnogenesis?
RISE546 is a low quality sample with a lot of damage and no UDG treatment.
https://www.researchgate.net/figure/Location-of-Komishan-and-other-Paleolithic-sites-mentioned-in-the-text_fig6_309770522
Davidski February 13, 2024 at 2:59 PM : "Heggarty is an idiot, and his language graphs are garbage."
Heggarty only provided the data and edited this extremely higgledy-piggledy unreadable paper. The entire calculation is the responsibility of the Russel D. Gray team, who are now pleased to receive a date of -6'120 CC for the first split of their PIE, even older than the already nonsensical -5'579 date of the Bouckaert/Atkinson 2013 version.
archlingo
@The Archaeogeneticist
You can post a link here to the relevant Plink files and I'll have a look.
Someone sent me the genotypes earlier today and there weren't enough markers for the G25.
@David
When is the common parent of this Suvorovo culture with SS? Because Proto-Anatolian should SPLIT around 4500 BC to 4000 BC, ..."
< NOBODY knows that. Ringe's computation ist pure lexicostatistics, i.e. without any time scale. The proponent of that date is D. Anthony, in combination with his western route view.
archlingo
@Simon
You wrote:
" Yet why did this admixture only flow in one direction? There seems to be no ANE/West Eurasian in East Asia (China, Korea, or Japan) despite the wide spread presence of Y-DNA Q. "
Are you daft or something?
There were multiple waves of west-to-east geneflow from Bronze and Iron Age West Eurasians in to East Eurasians.
Jeong, et al 2018 and Rogers et al, 2020-2021 make that amply clear. The gene flow was mostly from West Eurasian males and came with an increase in the haplogroups R1a, R1b and J.
This was found for Iron Age Mongolians, Xiongnu, Tûrk elites from Mongolia ca. 400-500 A.D., and also a medieval Uyghur sample.
Not mention many such similar patterns in other studies of Scythians, the Kyrgyz people, modern Uyghurs, etc. East Asia has experienced way more migratiory male impact from West Eurasians than Europe has from East Eurasians.
Also there IS some West Eurasian aDNA in modern Chinese, Koreans and Japanese.
"Yet why did this admixture only flow in one direction? There seems to be no ANE/West Eurasian in East Asia (China, Korea, or Japan) despite the wide spread presence of Y-DNA Q. "
You're wrong.
There was way more male veneflow from West to East, than East to West. Huge numbers of ethnic groups in China have mostly West Eurasian paternal haplogroups (Kyrgyz, Uyghurs, etc).
There was a huge male-mediated genetic impact on Iron Age Mongolia, Xiongnu, 400 A.D. Türks, mddieval Uyghurs, etc. All accompanied by an increase in West Eurasian male haplogroups like R1b, R1a and J.
Not to mention Scythians and other groups.
Basically there's a huge West Eurasian contribution to East Asia but not the other way around. Also, there IS West Eurasian ancestry in ethnic Koreans, Japanese and especially Chinese in China.
Post a Comment