I'm not too serious about validity of my Models, but it is certainly fun trying to figure possible combinations, that's why I'm doing it. I might be way off sometimes, sometimes probably right on. Few interesting things I learned doing them. One is that there is no way of recreating CW with very high Baloch without using 50% or more Yamnaya. Or recreating LBA/IA "German" without 50% of EEF, because of very high Mediterranean admixture. So there are things that can't be changed and are obvious, but one the other hand exact tuning might be way off.
There is always a question about the Samara Outlier guy. He is quite different than the rest of known Yamnayans. I wonder if he is more like new discovered Ukrainian HG, or he was perhaps from NW Yamnaya population. Well, technically he belongs to Yamnaya Culture, but he is very, very distinct genetically. I can use him to reconstruct Estonia CW and couple of CW Swedes, and receive 90% of Yamnaya admixture in them. When I use typical Yamnaya sample the Baloch and Caucasian shoots way too high and Med drops too low. The problem might be that Yamnaya in Estonia is probably not the same as Yamnaya in Germany, though both sources come from Yamnaya culture. This might be confusing, that's why I avoid calling Samara Outlier, Yamnaya. We might need to rename few things when we have genomes from all the Yamnaya.
I agree. Bronze Age in Northern Europe looks like a very crazy period. Whole populations were dying off, and new ones explode from small pockets. Never mind huge invasions to start with.
LeBrok on Eurogenes I found a really excellent posting about your brainteaser with the model. And it gives a very good insight about why the the tumulus culture looks crucial in the creation of the modern NW European. Especially in the North Sea region. Elp or Sögel is a tumulus/urnfield offspring from about 1800 BC. But I guess stil a good candidate for the spread of R1b U106/S.
As said in a previous posting my autosomal DNA comes close to: 1 Halberstadt_LBA In the posting
Halberstadt_LBA is crucial in his reasoning.
It's a long and excellent comment from G. Dekaen on Eurogenes (
25 feb 2017) about "Massive migration from the steppe is a source for Indo-European languages in Europe (Haak et al. 2015 preprint)" I doubted if i should publish it here. But for our quest it's so crucial.... I hope he agrees with it.....
"Judging from Figure 3, we can see that:
Early Neolithics: 90-100% farmer, 0-10% HG
Middle Neolithics: 80% farmer, 20% HG (Esperstedt with 40% HG is outlier)
LN Corded: 80% Yamna, 15% farmer, 5% HG
LN Bell Beaker: 45% Yamna, 40% farmer, 15% HG
LN Karsdorf: 75% Yamna, <5% farmer, 20% HG (outlier, likely Yamna+HG)
LN Benzingerode/Aberstedt: 40% Yamna, 60% farmer
EBA Unetice: 45% Yamna, 25% farmer, 30% HG
LBA Halberstadt: 55% Yamna, 45% farmer
Modern Czechs (closest we have to Germans) : 50% Yamna, 35% farmer, 15% HG
Modern Belarussian (to compare to Corded): 50% Yamna, 25% farmer, 25% HG
CA Hungary: 80% farmer, 20% HG
BA Hungary: 15% Yamna, 50% farmer, 35% HG (Note: K16 shows that the two individuals that make up BA Hungary are very different, one seemingly 0% Yamna, 50% farmer, 50% HG (!) and the second being 30% Yamna, 50% farmer, 20% HG which might show a significant HG bounceback here along with the possibility that mixing in Hungary was not complete yet by 1600BC)
Iron Age Hungarian (from K16): 60% Yamna, 40% farmer
Modern Hungarians: 45% Yamna, 40% farmer, 15% HG
Some ideas:
1. Corded and Bell Beaker are quite different, with the latter potentially being the by-product of a 50-50 cross of Corded and MNE.
2. Bell Beaker and Unetice are somewhat discontinuous as there seems to be a substantial reduction in farmer DNA with a concomitant rise in HG DNA with apparent stability in Yamna ancestry. At first, I thought this would lend credence to my idea that Unetice derives in large part from a source NE. of the Steppe in E. Corded/Fatyanovo-Balanovo as a result of interaction between this possibly non-IE culture and NW IE. Catacomb culture from the south. However, Unetice's Y-DNA is clearly WHG and NOT from NE. Europe, so it could represent a local resurgence of HGs in C. Europe.
3. Discontinuity between BA Unetice and LBA Halberstadt, which is unlike any non-southern European population in that it lacks HG. However, a 50-50 mix of Unetice descendants and Halberstadt would seem to perfectly match Czech's component distribution. I wonder what is the archaeological affiliation of Halberstadt? It's funny, Halberstadt looks like it's Early Hallstatt/Late Urnfield which I figured would be the descendants of, and hence quite similar to Unetice, yet Halberstadt is quite different from Unetice.
4. There seems to have been a significant change between Corded Ware and modern Belarussians (let alone Lithuanians or Estonians). There was at least a 40% demic displacement following Corded that culminated in the creation of Belarussians. The cumulative change of this/these population(s) was (40-90% displacement): 5-45% Yamna, 40-26.1% farmer, 55-27.2% HG. Given that the absolute minimum HG input was 27%, we need to look for a very strongly HG population which eliminates almost all of W. Europe (BB only had 15% HG, Norwegians only have 15% HG, Scots are the only ones close with 20-25% HG, Icelanders have 20% HG, followed by Orcadians who have about 17% HG). Intriguingly, Unetice basically seems to fit the bill at 45% Yamna, 25% farmer, 30% HG, and if we accept this, it would mean an 80-90% replacement by Unetice; the existence of Y-DNA I2 in Belarus today at 10-20% would seem to indicate an influx from the West at some point, but probably not an 80-90% replacement, so we can probably rule this out. There could have been a separate resurgence of EHG in Corded territory with Y-DNA R1a instead. On a similar note, I have a feeling this post-Corded change came from E. Corded/Fatyanovo-Balanovo along with the BA expansion of blue eyes; that population likely had a moderate amount of Yamna, moderate farmer and high HG to fit the requirements perfectly. In-fact, it seems that all modern N. Euro populations differ considerably from Corded in having much higher HG, so I wouldn't be surprised if there was a second major population expansion after Yamna that involved a group with much higher HG, from a population with roughly similar proportions as Unetice, but with possibly different Y-DNA (Unetice can't explain post-Corded change). In-fact, maybe we need to look for a dual expansion, one from resurgent C. Euro HGs/Unetice that re-expanded I2 and a separate E. Corded expansion of R1a, both of which increased blue eyes throughout Europe. This reminds me of this map from the ancient Bulgarian study that showed some sort of expansion out of C. Europe post-3000BC:
http://dienekes.blogspot.com/2014/05/ancient-dna-from-balkans-iron-age-thrace.html
5. Building on the above, considering that both EBA Unetice and BA Hungary show a sizable increase in HG in the EBA/MBA, the post-2500BC resurgence in HG can possibly be traced back to Central Europe/Pannonia and with an expansion of Y-DNA I specifically (Carpathians/Alps mountain refuge?). This seems to match all the Unetice samples being Y-DNA I2, along with IIRC the vast majority of Y-DNA from later Urnfield as well, which AFAIK is generally considered to have expanded from roughly the same area as Unetice, S. Germany/Czech/Austria etc. It seems that Y-DNA I2 is particularly associated with this phenomenon, so I'm not sure how Y-DNA I1 fits in and how it went from C. Europe to Scandinavia.
6. Between the CA and BA, Hungarians acquired significant HG mixture separate from the Yamna influx. I wonder what the cause was behind this simultaneous increase in HG in C. Europe 2200-1500BC?
7. Modern Hungarians can be modeled pretty successfully as a 50-50 mix of IA Hungarian with the more Yamna-like BA Hungarian. This would also seem to confirm that Yamna influence into the Balkans was much weaker (30%) compared to its expansion across N. Europe (German Beaker 45% Yamna).
or is it Basal Eurasian?
kept in Extended Data Table 2.
I forgot to mention for point 4 that, the study indicates that Corded was the result of a single, major expansion from Yamna and not a gradual process, which would seem to indicate that Corded was a pretty homogeneous culture (except perhaps its Eastern fringes?). This is the reason why I think we can directly compare these German Corded samples with Belarussians from farther East.
I was also just thinking that if there was a post-Yamna double expansion of HG from C. Europe/Unetice and E. Corded, this would help explain why fringe populations in NW. Europe like Scots and Norwegians have higher Yamna than Belarussians, Czechs, Hungarians - who were closer to Yamna and hence would be expected to have more Yamna - because these people were right in the heart of Europe, at the center of these two later, Yamna diluting expansions.
Here is my K16 breakdown (I used a ruler for proportions, so it should be reasonably accurate, sample size in brackets):
W.C.S.N HGs (14): 0% NE, 100% HG, 0% C. Asian
EHG (2): 0% NE, 80% HG, 15% C. Asian, 5% S. American
Early Neolithics (27): 75-85% NE, 15-20% HG, mostly zero, but trace elements (10 and 5%) C. Asian in Stuttgart/EN Hungary
Middle Neolithics (9): 70% NE, 30% HG, 0% C. Asian
Yamnaya (9): 0% NE, 50% HG, 50% C. Asian
LN Corded (4): <5% NE, 60% HG, 35% C. Asian
LN Karsdorf (1): 0% NE, 60% HG, 40% C. Asian
LN Bell Beaker (6): 25% NE, 55% HG, 20% C. Asian
LN Benzingerode Bell Beaker (3): 15% NE, 60% HG, 25% C. Asian
LN Alberstedt Bell Beaker (1): 30% NE, 50% HG, 20% C. Asian
EBA Unetice (7): 20% NE, 55% HG, 25% C. Asian
LBA Halberstadt (1): 20% NE, 55% HG, 25% C. Asian
Modern Czechs (closest we have to Germans): 25% NE, 50% HG, 25% C. Asian
Modern Belarussians (to compare to Corded): 20% NE, 55% HG, 25% C. Asian
CA Hungary (CO1) (1): 70% NE, 30% HG, 0% C. Asian
BA Hungary (BR1) (1): 35% NE, 60% HG, 5% C. Asian
BA Hungary (BR2) (1): 35% NE, 45% HG, 20% C. Asian
IA Hungary (IR1) (1): 20% NE, 40% HG, 35% C. Asian, 5% Siberian
Modern Hungarians: 30% NE, 45% HG, 25% C. Asian
CA Iceman (1): 75% NE, 25% HG, 0% C. Asian
Modern Bergamo: 40% NE, 35% HG, 25% C. Asian
So, western, central, southern and northern HGs seem to be 100% HG whereas the NE. Euro HGs would seem to have minor traces of the Teal C. Asian component, slightly more in Samara than in Karelia. The existence of pretty basal R1a and R1b among these EHGs testifies to R1a/b being either native to EHGs in NE. Europe OR was brought into NE. Europe via the C. Asian component and was gradually diluted/"decoupled" from its original ancestral background. I think we can effectively eliminate an origin of R1a/b among the ancestors of Neolithics in the Mid. East/West Asia proper. As some have been correct to point out, many of the most basal lineages of R1, R1a and R1b are located in N. Iran (was it also E. Turkey?) which is more strongly associated with the C. Asian than NE component, the former being around 50% with the latter 30% in Iran. Combined with Malta being R* (and IIRC, a basal Bhutanese R1b) and the general scarcity/absence of any R among farmers in Europe, I think we can generally rule out a "West Asian/Mid. East" origin of R1a/b and opt instead for either a C. Asian or NE. European. The Spanish R1b may simply be an early/rare case of R1b decoupling from its C. Asian roots and gradually making its way west with the spread of agriculture; this might also explain how we find an early Y-DNA H2 in C. Europe 5500BC which maybe came from close to Iran and also traveled westward. Alternatively, the Spanish R1b could be decoupled from EHGs that traveled westwards and were assimilated by Neolithics. I'm not sure which is more likely.
I also find the K20 interesting in that it distinguishes between "western" and "eastern" HGs. The "eastern" HG seems to be a merger of parts of the old HG and steals 50-70% from the Teal C. Asian component. This could be further evidence linking Teal C. Asia and EHGs together with R1a/b as the boundaries between these seem to be smaller - and hence, prospective gene flow increases/is more likely - than between the EHGs and the NE component. La Brana, the SHGs and the Hungarian HG seem to be purely "western," whereas curiously, Loschbour is shown as a 50% "western" and 40% "eastern" mix (with 10% NE). This might mean that the Hungarian HG may have been a little more representative of SE. HGs rather than C. HGs and that C. HGs had influence from EHGs, which might be attested by the finding of mtDNA U2e and U4 among Mesolithic German HGs. Still, it's even more curious why Motala and other SHGs show up as 100% "western" when they SHOULD be far more "eastern" shifted than Loschbour since they have vast amounts of mtDNA U2a and U4. As for the EHGs, they turn out to be 60% "eastern," 35% "western" and 5% S. American. Whatever the minutia, the point I'm trying to make here is that if you look at the ancient and modern samples, you see some very interesting changes. Among Neolithic cultures, you see that about 100% of all assimilated HGs are of the western/southern variety. This changes entirely with Yamna and ALL following cultures showing a massive shift to "eastern" HGs until finally we get to modern populations where the Grey "western" HG component is entirely missing/extinct. Mysteriously, it seems to even disappear/not exist among pre-IE peoples like Sardinians, Basques or people with no IE connection like NW Africans. How did this happen and what does it suggest?
Well, if we extract the Teal C. Asian component (based on K19) from "eastern" HGs to leave only the EHG component in Blue, we can see that EHGs have a frequency ratio of 35:55 "western" HG to "eastern" HG, Loschbour/C. Euro HGs are 50:40, and La Brana/Hungary/Swedish HGs are 100:0. Yamna is 5:20 meaning they absorbed EHGs and not WHGs (we obviously knew that already). Corded is 10:20 also suggesting EHGs (unsurprising because Corded is completely intrusive in Germany). Bell Beaker is 10:25 and Benzingerode Beaker is 15:25 again suggesting EHG influx rather than a contribution from local Loschbour-like HGs. Unetice is 15:35 indicating a continued increase in the EHG:WHG ratio over time until finally, we get to BA Hungarian and LBA Halberstadt which provide the strongest evidence for my hypothesis. I pointed out earlier that the BA Hungarian sample has two very different samples (BR1 and BR2); I believe the earlier 2200-2000BC BR1 to represent the first sample which has a ratio of 25:25 reflecting probably a C. + S. Euro HG mixture. It is only in the Late Bronze Age however that we begin to see samples that are truly "modern" and reflective of the final transformations of the European gene pool (as I have argued). Notice BR2, 1300-1100BC has zero "western" HG and 35% EHG as well as LBA Halberstadt, 1100-1000BC having zero "western" HG and 35% EHG. I think this is pretty clear evidence of a post-2000BC/MBA influx of EHGs from somewhere north of Yamna and E. of German Corded, very likely in the Fatyanovo-Balanovo zone that I have postulated. Cultures that followed Unetice probably played a big role in this, namely Tumulus and Urnfield, and maybe we can even incorporate the Seima-Turbino phenomenon in this since IIRC, it has some sort of connection with Tumulus and seems like a pretty good candidate for bringing an MBA, super-EHG influx into Europe considering it originated way far east.
The IA Hungarian (IR1) sample shows a 20:15 ratio. This individual is obviously heavily derived from the Steppe given he has the highest C. Asian of any ancient sample in K20 (and a decently high K16). Given how high his WHG is in K20 (20%), it's very likely that all of IR1's NE (25%) together with his WHG descends from the Balkans, perhaps, the last stragglers of BR1? The question I have is, is the 15% EHG a part of the "native" Balkanites (since BR1 also had 25% EHG), but that would leave no EHG to remain on the Steppe (and I consider it unlikely IR1 came without any EHG), or was the EHG brought with IR1 and the "native" Balkanites had primarily NE+WHG composition.
Overall, K20 shows us that the HG "resurgence" in BA Europe came not from native/"western" HGs, but rather from "eastern" HGs. Furthermore, the changes from the BA to now show that there was a continued influx of "eastern" HG post-BA that ended up completely replacing "western" HG. That would seem to lend credence to my idea that Fatyanovo-Balanovo was the source of further population expansions post-Yamna and possibly the springboard for at least NW IE languages from Catacomb culture.
Yamna completely lacks the Neolithic component found in every other EN and MN culture! Neolithic European cultures therefore did not penetrate into Yamna as far as the Volga. Conversely, given the complete absence of Teal/C. Asian among Euro EN and MN cultures, this component was not present in the Middle East responsible for the Neolithic migration wave into Europe until after 6000BC. Teal likely didn't come through the Caucasus into Yamna because all the Caucasus peoples have 20-30% Neolithic and 50-60% Teal components, so if Teal came through the Caucasus into Yamna, Yamna would also have NE but they don't, therefore the Teal in Yamna must've come through C. Asia. You can try to make the case that the Caucasus was 100% Teal at that time, but I don't find that believable, I'm almost certain they had NE. This also means that all the mtDNA H found in Yamna was either C. Asian or less likely, EHG and had little to do with Euro/Mid East Neolithics unless it became "decoupled." Furthermore, if NE didn't reach Volga Yamna, I think that strengthens my case that the later Fatyanovo-Balanovo even further north also lacked NE, probably had even more EHG than Yamna and was responsible for an MBA increase in EHG and very likely, also blue eyes and blonde hair throughout C./N. Europe. IIRC, the northernmost pigmentation samples from Yamna near the Volga-Kama had 33% blue eyes, the highest anywhere in the Yamna cultural horizon.
I believe PIE existed in the N. Caucasus mountains from 3500-3000BC. Yamna shows essentially zero genetic contact with the Caucasus meaning either IE didn't spread into Yamna OR it spread through elite dominance. Since it's much easier/simpler for a language to spread via the momentum of population expansion rather than a complex process of political domination, I think it unlikely PIE spread into Yamna via elite dominance (although given the wealth of Maykop, it IS still possible!). Instead, I think the evidence is stronger for IE languages spreading from the Caucasus into Catacomb culture. The existence of mtDNA R1 at 8.3% in Catacomb may hint at the start of a Caucasus-->Catacomb movement of IE speakers and so we should see the NE component start showing up in Catacomb; we know that at some point, NE must've showed up in the Volga/Urals pre-Slavic expansion because all the Volga populations have it to some degree, Catacomb would be a strong candidate for this. IF this did not happen, then we have two main possibilities: PIE was spoken in another mountainous area with yew trees, the plough, cart, and copper around 3000-3500BC which can only be the Carpathians/Crimea(?) OR PIE transferred their language to Yamna/Catacomb via elite dominance. PIE originating with C. Asian component in C. Asia is impossible since there are no yew trees there (nor carts etc.), the closest ones are in Tajikistan. Here is a distribution map of the yew tree groups:
http://www.worldbotanical.com/TAXNA_files/image002.gif
It's interesting to examine the sequence of K16 changes from culture to culture post-Yamna in C. Europe (Germany, Hungary, N. Italy). The CA Hungarian and MN Farmers sample show us a ratio of 2.5:1 NE:HG among C. Euro CA farmers just before the Yamna expansion. Using that, we can see that German Corded represented a minimum 95% replacement of CA/MN farmers (no farmer substrate). Also, given the 1:1 HG:C. Asian ratio in Yamna, only 35% out of the 60% of HG in Corded could be explained by Yamna influx, meaning that as the Yamnayans crossed through Belarus/Poland into Germany, they brought an additional 25% HG with them (significant EHG substrate). That could explain the LN Karsdorf sample as a straight Yamna-EHG hybrid (and in-fact, Karsdorf at K20 has a 10:30 WHG:EHG ratio, testifying to its eastern provenance and NOT being native to Germany). German Bell Beaker shows a significant bounce-back of NE and a drop in HG and C. Asian and can be pretty successfully modeled as a mixture of 35% MN/CA Farmer and 65% LN Corded. I doubt this farmer contribution was TRB (or anything more western) because the EHG:WHG ratio between Corded and Beaker doesn't change much (if anything, HGs become slightly MORE EHG during Bell Beaker), so I wouldn't be surprised if Bell Beaker might be some sort of Globular Amphora + Corded combo. Benzingerode Beakers are more Corded, being about 80% Corded and 20% MN/CA Farmer. Using Yamna instead of Corded is less successful in modeling Bell Beaker as a mix with MN/CA Farmers, but that could be due to the Yamna samples being so far east.
From Bell Beaker to Unetice, we see relative continuity with a steady increase in the EHG:WHG ratio indicating continued migrations from NE. Europe. The same applies from EBA Unetice to LBA Halberstadt, the K16 proportions remain identical, but WHG at K20 disappears completely due to continued migrations involving EHGs, making LBA Halberstadt direct ancestors of C. Euros like Czechs/Germans.
As for Belarussians and Corded Ware, post-Corded we see a minimum of 30% total demic displacement. Going from 30-90% replacement, the cumulative change is from (a) population(s) with 64.2-23.1% NE, 34.2-51.7% HG, 1.8-25.1% CAS. Considering that after 3000BC, there aren't really any cultures left in E./C. Europe without any Central Asian (CAS), we can increase the minimum to at least 40% replacement with the following cumulative requirements: 48.5-23.1% NE, 40.8-51.7% HG, 10.6-25.1% CAS. This doesn't really help to narrow down possibly migrations too much considering we're looking at a 5000 year timeframe, but we can essentially eliminate any migration from S. of the Alps/Carpathians or the Caucasus. At some point, Belarus must have gone from ultra-low NE during Corded to 20% present NE, maybe that was with Urnfield/Lusatian considering Urnfield seems to have been a major player and source of migrations during the MBA. Also, at some point, the remaining WHG in Belarus (which almost certainly existed since it was present in Yamna/Corded) must've disappeared from a migration from further east.
The change from CA Hungary to BR1 is really an oddity. BR1 indicates at least a 50% replacement of previous CO1 inhabitants (max. of 50% CO1 substrate in BR1) with a doubling of HG compared to CO1.This HG had a WHG:EHG frequency ratio of 25:25, and this combined with its only 5% CAS makes it extremely unlikely that this overall HG increase came from the east. It seems to reflect a local (and short-lived), perhaps combined C. Euro + S. Euro HG resurgence. BR1 is likely a dead end, that was neither IE nor left too many descendants today. BR2 represents an entirely new population with zero WHG, attesting to some sort of NE. Euro migration and can be pretty successfully modeled as a mixture of 50% CO1 with 50% of a NE. Euro population north of Yamna with 10% more HG and 10% less CAS. BR2, like LBA Halberstadt seems like a perfect, direct ancestor of Hungarians, with only a minor change of -5% NE and +5% CAS, so we seem to have a relatively stabilized European gene pool by 1000BC and major changes taking place between 2000-1000BC. IR1 doesn't seem to have affected the Hungarian population after BR2 given it has 5% Siberian, N1c Y-DNA, G2a mtDNA, and a 20:5 WHG:EHG ratio, all of which are extremely uncharacteristic of modern Hungarians. Yet again, another dead end IMO.
Nevertheless, IR1 may be pretty useful in determining the genetic composition of the Steppe and at what point it ceased to play a major demographic role in Europe. If IR1 truly represents Cimmerians, given its date of 800-1000BC, we can associate it with at least the Novocherkassk/Chernogorovka culture and assert that this culture and the Steppe from then on (given the gradual increase in Asiatic DNA) had a minimal demographic impact on Europe. If IR1 had 5% Siberian and if we estimate that its 20% WHG and 25% NE were picked up in the Balkans, then we can estimate that Steppe cultures from at least 1000BC in Ukraine probably had around 10% Siberian and zero/little NE, and therefore couldn't be the ancestors of basically any Europeans today. Even populations that experienced Steppe invasions for 1000 years (Moldavians/Romanians, Bulgarians, Hungarians) display only the faintest traces of Siberian <1%. It should've been obvious to everyone judging from the very Asiatic Scythian mtDNA circa 500BC and their likely R1a-Z93 that Scythians are not the ancestors of basically anyone in Europe. This same analysis is relevant for later Turkic peoples. I have previously postulated that Srubna may also have been the source of some MBA migrations between 2000-1000BC as a result of Catacomb+Fatyanovo-Balanovo mixture and a back-migration onto the Steppe followed by an expansion into Europe increasing blue eyes/blonde hair in a similar manner as Andronovo. The question on my mind is, did Srubna - which ended 1200BC - also have this 10% Siberian or did it end via migrations from further east by people who brought the 10% Siberian DNA, i.e. Andronovo/Karasuk. Maybe it was the very last major migration from the Steppe before nomadism became dominant?
Following Otzi, 3300BC, N. Italy experienced a cumulative population change of at least 50% to the modern day. The cumulative genetic proportion of this change ranging from 50-90% replacement was: 12-40.4% NE, 40-32.9% HG, 60-26.7% CAS. This strongly points to the most significant migration post-3000BC coming from around the Steppes.
As for C. Asia itself, we can confirm HG (specifically EHG) throughout the entire C. Asia as far South as Afghanistan and into Gujaratis and Sindhis, but no further. It seems to be a pretty constant 5-10% (elevated 15% among Tajik Pamirs). However, Burusho also seem to have 10% HG, did this come from Indo-Iranians? I doubt it, Burusho have shown to be tied with Kalash for having the most ANE in the Old World and I find it extremely unlikely that Burusho got their 35% ANE entirely from the powerful Indo-Iranian people without also switching their language like just about everyone else in the region. Furthermore, Burusho have high R1a (25%), R1b* (10%), C (8%), have a strong Siberian and E. Asian autosomal component and alleged links with Yeniseian languages farther north which might hint at time spent on the Steppe. The question is when? It's possible they could've been on the Steppe as early as Andronovo since Andronovo will likely have some E. Asian component due to having Y-DNA C (although no E. Asian seems to have filtered through to the 1000BC IR1 sample). Alternatively, Burusho E. Asian/Siberian ancestry (both absent in all I-I peoples) could perhaps be via Turkic contacts since there are Turkic loanwords into Burusho. This pre-I-I HG substrate could explain why Tajiks have a bit more HG than the other C. Asians (who could be mostly pre-II substrate descended like Pathans, Gujarats, Sindhs or have had their HG diluted by Turkic migrations, Turkmen, Uzbek) because Tajiks have HG from both the pre-I-I substrate and the later I-Is themselves. So, maybe HG already existed in C. Asia at 5-10% with the Indo-Iranian people bringing an additional 5-10%. Either way, if the Yamnayan 1:1 ratio between HG and CAS holds, then roughly 10-30% of C. and S.C. Asian genes derive from Yamna Indo-Iranians.
On a similar note, I found it interesting that Chuvash almost completely lack the E. Asian component that generally exists at a ratio of 1:2 E. Asian:Siberian (K16) among Oghuz Turks like Altaians/Tubalars. It's not entirely absent though because we can see it at trace frequencies in K17, 18 and 20. Given that Chuvash have 20% Siberian (a large part of which may derive from their Mari substrate who are known to have around 25% Siberian), we should expect to see about 10% of the Yellow component. Either Chuvash is yet another case of elite dominance by a Steppe people and language shifting by the local substrate OR Oghur Turks were noticeably different from their Oghuz cousins in being much less E. Eurasian. I consider the elite dominance model the most likely since that seems to be a pattern among modern Turkic-speaking peoples (Turks, Azerbaijans, Kumyks, Gagauz etc. all are predominantly "natives" of non-Turkic ancestry). We know many/most of the "Turkic" loans into Mongol came from an unknown Oghuric language, so they interacted closely early on. That makes it unlikely that Oghuz became "Mongolified" with extra E. Asian while somehow Oghurs escaped that fate. Therefore, Oghurs were likely just as E. Asian as Oghuz Turks. What's also interesting is that Altai Turks (assumed to be the Turkic homeland) differ considerably from their alleged Altaic relatives, the Tungus and Mongols in having what seems to be genuinely Yamna ancestry (1:1 HG: CAS ratio) in the amount of 30% for Tubalars and about 20% for Altaians and Kyrgyz (with some extra CAS not Yamna-derived). We can compare this to Tajik Pamiris who likely have 35% Yamna ancestry. Mongols like the Daur and Oroqen have none, nor do Tungus like the Hezhen and Ulchi. It's tempting (and sensible) to say that this Yamna component in Oghuz Turks reflects the Indo-Iranian Steppe substrate that was assimilated by Turks. Alternatively, it's possible, this could partially represent Turks having slightly higher W. Eurasian, non-Yamna substrate from pre-IE HGs that existed in Siberia since we know Euro HGs existed far to the east.
Also of interest is that the K16 has seven Uralic samples and not one (not even the far eastern Mansi and Ket-admixed Selkups) have any of the yellow East Asian component. Nor do the Yukaghirs, which I think suggests (given the ubiquity of this yellow E. Asian component among all Altaics) that Uralo-Yukaghiric did not originate/spread from the Baikal (IIRC, Jaska's idea), but further west and was not associated with Altaic languages (IIRC Janhunen's idea), OR Altaic was associated with Uralo-Yukaghiric and was primarily Siberian, only later absorbing E. Asian OR that Uralo-Yukaghiric WAS in the Baikal before Altaic moved in. I think all in all, this evidence strengthens cases of Uralic homelands further north and west of the Baikal. Even the Mansi which are among the eastern-most of Uralic peoples (apart from small tribes of Samoyeds) have about 30% HG, 20% C. Asian, 40% Siberian, 5% Far East Siberian. Some of that HG-C. Asian might be Yamna derived as Ugrics allegedly have a pastoralist history (max 40% Yamna ancestry). Saami in comparison are 50% WHG, 20% C. Asian, 20% Siberian, 10% N. Eastern, 5% Far East Siberian. It would've been great if they tested those 1500BC HGs from Karelia as well to see how much Siberian and Far E. Siberian pre-Saamic peoples had, maybe Saami/F-U people lowered Siberian rather than raised it in Finland? What is Uralic's connection to Far E. Siberian? I wouldn't be surprised if in both Mansi/Saami cases, they acquired this from some pre-Uralic contact/substrate.
Some other observations:
-Karelia HG is 60-62% WHG, 38-40% ANE in comparison to Motala12 being 80% WHG, 20% ANE. If we know EHGs were 60% WHG, 40% ANE and we know that Armenians (15% ANE) are 5% HG and 50%, we can calculate how much ANE CAS has. So, 2% of Armenian ANE comes from HGs, and the remaining 13% comes from their 50% CAS meaning CAS is 26% ANE. Surprisingly, that's less than EHG. Yamna is 50% HG, 50% CAS, so they likely had about 33% ANE.
-As for R1a, maybe it was more popular to the N. of the Steppe and expanded after R1b? We should remember that all these R1b samples come from a small area and are probably not fully representative of all Yamna. I'm confident that R1a WILL be found in the Yamna culture because Samara and Karelia seem to be very close, so I would imagine R1a/b existing in a single population; furthermore, Corded Ware which is supposed to be 75-80% Yamna has 3/3 R1a samples it seems which almost certainly guarantees R1a existing in Yamna as well.
-No Haplo I or N from the Steppe or E. Euro forest-zone/HGs prior to 3000BC. Y-DNA N's absence from the Steppe until 1000BC (Gamba IR1) could mean that it expanded across N. Eurasia 3000-1000BC through the Forest-Belt and gradually filtered onto the Steppes. N is fairly infrequent among Turkic groups AFAIK, so could this be an indication that early Steppe Nomads like Cimmerians and Scythians were Ugrics or had a very strong Ugric substrate (Scythian mtDNA is also super-eastern, mtDNA F1b, A4, D, that doesn't even fit Uralics, might even be Turkic!) Either way, I think we're seeing some pretty solid evidence for early Turkic and Ugric substrate/peoples all the way in the western Steppes as early as 1000-500BC.
-Starcevo had Y-DNA H2! I'm not too familiar with H, but could this mean either H was introduced into India via the Neolithic (possibly with Dravidian languages?) OR this H2 originates from India/S.C. Asia, and traveled westward with agriculture. Given Starcevo lacks the CAS component in K16, could this mean ANE was absent in/near India/South-Central Asia 5000-6000BC? Stuttgart has 10% CAS, we also see it in one EN Hungary sample at <5%, so maybe it just got diluted/decoupled and there WAS CAS/ANE in India? Interestingly, Stuttgart's HG admixture seems to be EHG-derived, not WHG, so maybe some of the CAS comes from the assimilated EHGs? Judging by the absence of CAS in Munda-speaking Kharia, I think we can say that prior to Dravidians, India probably didn't have CAS and low ANE (Lodhi are also supposed to be Munda speaking, but they look admixed which explains their CAS; MA1 is mostly HG and CAS, only about 10% S. Asian which is the dominant component among all S. Asian Aryan speakers, so S. Asian is not very strong in ANE). Kalash lack the NE component that exists in other Iranic peoples like Tajiks and Pashtuns, not sure why (maybe they have less non-Yamnayan/I-I substrate?). It also looks like there may be a second H2, I0405 from Spain_MN that can be either I2a1a1 or H2. If it is I2a1a1, it is of the type popular among Sardinians and SW. Euros/Basques.
-The absence of Y-DNA J and E3b once again points to these lineages likely being largely confined to SE. Europe until maybe the MBA or EBA.
-They confirmed what many (including myself) have speculated before, that mtDNA I, T1, W, U2e, U4, U5a (and probably R1 & C1g IMO) strongly represent the inflow of ANE into Europe. Both EHGs and Yamna have mtDNA C, yet lack the Siberian component which I think means we can consider at least C1g/C4a6 as ANE. IIRC, we also have an mtDNA C sample from PPNB in Syria. It's interesting that they list H (certain subclades) as also being representative of ANE. That might explain the H found in Karelian HGs earlier. It's also interesting that they don't associate mtDNA J with Early Neolithics, I wonder where that leaves it? In the Fernandez study on PPNB from Syria, J was noticeably absent while being present in Mesolithic Greece, so could it be a southern HG marker?
-P. 51 confirms the presence of mtDNA H among Southern HGs (Spanish?) at a remarkably high frequency of 38.5%; U5b is another 38.5%, N* 15.4% and U4 (!) 7.7% and NO U5a. C. Euro HGs (I guess Germany) are equal parts U5a, U5b 32.1% each, followed by U 14.3%, U2 10.7%, U4 7.1%, U8 3.6%. E. Euro HGs are equal parts U5a and U4, 26.7% each followed by 20% U2, 20% C, 6.7% H.
-I hope they manage to add pigmentation data for the final print. Given the large scale of this study in time and geographic scope, it could really help settle many questions, like the timing/expansion of blue eyes, light skin and blonde/red hair, especially given the odd results we've seen from Yamna/Catacomb compared to later Andronovo. There must be some serious discrepancy there. I'm interested to see what pigmentation these EHGs and Samaran Yamnayans had; it's one of the main reasons I'm skeptical of Yamna being the last major demic displacement in Europe and accounting for the spread of IE. My theory is that blue eyes and blonde hair was likely spread by EHGs, especially during the MBA. Western HGs are universally dark-skinned, dark-haired and blue-eyed, but we saw from some of the Swedish HGs (IIRC, Motala12 and StoraForvar11) that some were light-skinned - and I consider it no coincidence that SHGs who are partially EHG are also partially light-skinned - meaning EHGs could also have been light-skinned, in addition to Andronovans being very blonde and heavily EHG derived. I think the connection is strong. EHG also seems to correlate almost perfectly with the European peaks in blonde hair and blue eyes, NE. Europe. Lastly, the Volga Yamna sample from a previous pigmentation study found the highest amount of blue eyes there out of the entire Yamna, 33%."