Friday, April 27, 2018

The mystery of the Sintashta people


During the Middle to Late Bronze Age, the steppes southeast of the Ural Mountains, in what is now Russia, were home to communities of metallurgists who buried their warriors with horses and the earliest examples of the spoked-wheel battle chariot.

We don't know what they called themselves, because they didn't leave any written texts, but their archaeological culture is commonly known as Sintashta. It was named after a river near one of their main settlements; an elaborate fortified town that has also been described as an ancient metallurgical industrial center. Another of their well known settlements, very similar to Sintashta, is Arkaim, pictured below courtesy of Wikipedia.


Sintashta is arguably one of the coolest ancient cultures ever discovered by archaeologists. It's also generally accepted to be the Proto-Indo-Iranian culture, and thus linguistically ancestral to a myriad of present-day peoples of Asia, including Indo-Aryans and Persians. No wonder then, that its origin, and that of its population, have been hotly debated issues.

The leading hypothesis based on archaeological data is that Sintashta is largely derived from the more westerly and warlike Abashevo culture, which occupied much of the forest steppe north of the Black and Caspian Seas. In turn, Abashevo is usually described as an eastern offshoot of the Late Neolithic Corded Ware Culture (CWC), which is generally seen as the first Indo-European archaeological culture in Northern Europe (see here).

Below is a Principal Component Analysis (PCA) featuring 38 Sintashta individuals from the recent Narasimhan et al. 2018 preprint. Note that the main Sintashta cluster overlaps almost perfectly with the main CWC cluster. The relevant datasheet is available here.


Moreover, many ancient and present-day South and Central Asians, particularly those identified with or speaking Indo-Iranian languages, appear to be strongly attracted to the main Sintashta cluster, forming an almost perfect cline between this cluster and the likely Indus Valley diaspora individuals who show no evidence of steppe ancestry.

This is in line with mixture models based on formal statistics showing significant Sintashta-related ancestry in Indo-Iranian-speakers (for instance, see here), and high frequencies of Y-haplogroup R1a-Z93 in both the Sintashta and many Indo-Iranian-speaking populations.

Some of the Sintashta samples are outliers from the main Sintashta cluster, and that's because they harbor elevated levels of ancestry related to the Mesolithic and Neolithic foragers of Eastern Europe and/or Western Siberia. This is especially true of a pair of individuals who belong to Y-haplogroup Q. However, this doesn't contradict archaeological data, which suggest that the Sintashta community may have been multi-cultural and multi-lingual. Indeed, it's generally accepted based on historical linguistics data that there were fairly intense contacts in North Eurasia between the speakers of Proto-Indo-Iranian, Proto-Uralic and Yeniseian languages.

Thus, it appears that there's not much left to debate because ancient DNA has seemingly backed up the most widely accepted hypotheses about the origin of Sintashta and its people, and their identification mainly as Proto-Indo-Iranian-speakers.

However, a sample from a Sredny Stog II culture burial on the North Pontic steppe, in what is now eastern Ukraine, has complicated matters somewhat. This individual, known as Ukraine_Eneolithic I6561, not only clusters very strongly with the most typical Sintashta samples, but also belongs to Y-haplogroup R1a-Z93. On the other hand, none of the CWC remains sequenced to date belong to this particular subclade of R1a (although, obviously, they do belong to a host of near and far related R1a subclades).

I've never seen anyone worth reading propose that Sintashta might derive from Sredny Stog II instead of Abashevo. And no wonder, because Sredny Stog II was long gone when Sintashta appeared in the archaeological record.

But if CWC remains continue to fail to produce R1a-Z93, while, at the same time, the steppes of eastern Ukraine and surrounds are shown to be a hotbed of R1a-Z93 from the Sredny Stog to the Sintashta periods, which I think is possible, then ancient DNA might well force a serious re-examination of how the awesome Sintashta culture and people came to be.

See also...


222 comments:

  1. Is there any other reason to assume that those Q1a outliers are Uralic speakers besides the fact that they are genetic outliers? Is it possible to divide all these Sintasha outliers into subgroups according to their different biases, fex towards WSHG/Dali_EBA type of groups, Okunevo type of more eastern groups and such?

    ReplyDelete
  2. @Huck Finn

    You can probably use the Global25 to do that. The same Sintashta samples as those in the above PCA made it into the Global25 datasheet.

    https://drive.google.com/file/d/1FSzKKknFGcOgfyA76q9PD7B-n-MJs7L8/view?usp=sharing

    But I reckon Matt might beat you to it when he sees this post. :)

    ReplyDelete
  3. @Davidski: something has already been done, from a Finnish forum and made by Vastandus:

    https://i.imgur.com/CHglSLS.png

    It seems to me that at least some of the outliers are biased towards kind of "local" WSHG, such as Sintasha_MLBA_o3. Q1a, on the other hand, does not sound too Uralic, even though Selkups do harbour that lineage, if I recall it right.

    How about placing Dali_EBA into the PCA, as a kind of benchmark?

    ReplyDelete
  4. @Huck Finn

    Dali_EBA is highlighted in the same PCA here...

    https://drive.google.com/file/d/1HKC3mKkCGA6lItug7qpuC20HsPwO7bQD/view?usp=sharing

    ReplyDelete
  5. @All

    OK, here we go again...

    Commenting on this, Rai said, “any model of migration of Indo-Europeans from South Asia simply cannot fit the data that is now available.”

    Indus Valley people did not have genetic contribution from the steppes: Head of Ancient DNA Lab testing Rakhigarhi samples

    Nirjhar and Sanuj need to get in touch with Rai ASAP to confirm that this isn't what he meant to say and that the Jagran article is really still current. If you don't know which awesome Jagran article I mean, see here...

    Indian smoke and mirrors

    ReplyDelete
  6. @Davidski: Thanks, it seems that one of the outliers practically overlaps Dali_EBA. But, Q1a? Maybe somewhat different N was residing on the European side of Ural mountains (those Sintasha related samples to the SW of Dali_EBA)? N1c can't have been too far, as it is found in Sargat kurgans somewhat later.

    ReplyDelete
  7. Huck Finn wrote,

    "It seems to me that at least some of the outliers are biased towards kind of "local" WSHG, such as Sintasha_MLBA_o3. Q1a, on the other hand, does not sound too Uralic, even though Selkups do harbour that lineage, if I recall it right."

    What is the precise classification of the Y-DNA of those two individuals? Q1a-F1096 does not appear to be associated with present-day speakers of Uralic languages in particular; branches of it have been found among Turkmens, Azeris, Chinese & Vietnamese, Chukotko-Kamchatkans, a Paleo-Eskimo (Saqqaq), a Peruvian, some parts of Europe (North Caucasus, Poland, Hungary, Northern Ireland), etc.

    On the other hand, Q1b1a-L54 is in fact associated with speakers of the Samoyedic branch of Uralic languages as well as with the Kets, the last remaining speakers of a Yeniseian language. Yun‑Zhi Huang et al. 2017 ("Dispersals of the Siberian Y‑chromosome haplogroup Q in Eurasia," Mol Genet Genomics, DOI 10.1007/s00438-017-1363-8) have reported finding Q-L54 in 16/19 Ket, 5/7 Selkup, and 2/4 Enets. The sample sizes are quite small, but Q-L54 clearly is a common Y-DNA haplogroup among extant Kets and Samoyeds.

    Q1a-F1096 and Q1b1a-L54 are not closely related; their TMRCA in Q1-L472 is estimated to be approximately 26,000 [95% CI 24,200 <-> 27,900] ybp according to YFull YTree v6.02. (The Q1 clade also subsumes all indigenous American members of Y-DNA haplogroup Q.)

    ReplyDelete
  8. @Ebizur and re: "On the other hand, Q1b1a-L54 is in fact associated with speakers of the Samoyedic branch of Uralic languages as well as with the Kets, the last remaining speakers of a Yeniseian language."

    Unluckily I'm not aware of a more precise classification of those individuals and Samoyedic speakers indeed do have Q. Samoyedic, on the other hand, is based on eastern expansion of Proto Uralic into Sayan area and only after that into more northern regions. So, I'd guess that paternal Q was originally not a Proto Uralic lineage? The absence of paternal Q in other Uralic groups is probably not coincidental.

    ReplyDelete
  9. @Huck Finn & Ebizur

    One of the two Sintashta samples that belong to Q1a was further classified as belonging to Q1a2 or Q-L56. So it's possible that they would both belong to Q1a2 if there was enough data.

    Various branches of Q1a2 are thinly spread around most of Northern and Central Eurasia today, and some appear to be relatively common (but still uncommon) in Scandinavia.

    So obviously this is not a typical marker of any present-day Uralic speaking population. However, that doesn't mean it wasn't a marker present amongst Uralic speakers who lived near and in Sintashta towns during the Bronze Age.

    But yes, I'd be a lot more confident that some of those Sintashta outliers were Uralic-speakers if they belonged to N1c.

    ReplyDelete
  10. Rai said...“There is no question of the model being flawed. It is a most solid piece of work—no new study will overturn it. Our own work which will be out very soon provides solid evidence for the model.”

    “any model of migration of Indo-Europeans from South Asia simply cannot fit the data that is now available.”

    In honor of our distant Indo-Iranians cousins, I'll simply say “Shah Met" (Checkmate in Persian).

    ReplyDelete
  11. Q1a are likely not the Uralians, but Yenisenians or Kelteminarisns if it's not the same.

    ReplyDelete
  12. @Davidski
    Can you add Sintashta_MLBA_o3 samples? They are R1b-M73+ and I think they are even more ANE-shifted.
    And about Uralic languages: I think for two main reasons haplogroups N-TAT and N-P43 are real markers of uralic speakers:
    1. Ket people and their neighbours selkups and khantys are autosomally very similar. Ket people almost entirely are Q1a+ and selkups and khantys have both Q1a and N. Kets speak Yeniseian and selkups and khantys speak Uralic.
    2. Some Siberian non-uralic languages are somewhat related to Uralic and their speakers have some haplogroups N-TAT and N-P43. Like Turkics, Yukagirs, Eskimo-Aleuts. But languages like Nivkh do not have similar relations to uralic and their people lack haplogroups N-TAT and N-P43.

    ReplyDelete
  13. @REZA

    These two samples are in the above PCA and the Global25.

    Sintashta_MLBA_o3:I0941
    Sintashta_MLBA_o3:I1028

    But I can't plot them right now. You can use this datasheet to do so.

    https://drive.google.com/file/d/1q9ygTvpM5U83Lq3-TDPdWR83Y_vUNjeb/view?usp=sharing

    ReplyDelete
  14. Proto Uralic was spread both to the east and west by Seima Turbino traders and they could have been bilingual in Info Iranian too. Check Parpola's latest paper on this

    ReplyDelete
  15. Q-L56 (formed 26000 ybp, TMRCA 20000 ybp) should be labeled Q1b according to ISOGG 2018. Q1a2 would be an older name for the same clade. If the ancient Sintashta samples belong to Q-L56, then it is much more plausible that they might have some relationship with Yeniseians, Ugrians, Samoyeds, and/or Turks.

    Q-L56/M346 subsumes both Q1b1a1a-M3 (formed 15300 ybp, TMRCA 13700 ybp), which is one of the typical Y-DNA haplogroups of American aborigines, and Q1b1a3-L330 (formed 16200 ybp, TMRCA 7900 ybp), which is a clade now found mainly in north-central Eurasia. Q1b1a3-L330 subsumes Q1b1a3a-Y20260, which seems to occur with low frequency among Turkic peoples (e.g. Uzbeks) and Hungarians (formed 7900 ybp, TMRCA 7300 ybp, or TMRCA 3300 ybp for the predominant Q-L332 subclade), and Q1b1a3b-B287, which occurs among extant Kets and Selkups as well as in Astana, Kazakhstan and Kemerovo Oblast, Russia (formed 7900 ybp, TMRCA 4100 ybp according to YFull). Members of Q-L330 also have been found in the southern Altai and northwestern Mongolia. It also appears that the Ob-Ugrians (Khanty and Mansi) share a great deal of recent common ancestry with the Selkups and the Kets.

    ReplyDelete
  16. Must it be one or the other?

    Instead of direct lines, multiple steppe groups could have interacted in waves, with the descendants of Stredny Strog mixing with CWC peoples to form Sintashta?

    ReplyDelete
  17. @Chetan

    In regards to Parpola's latest, you mean this?

    Finnish vatsa ~ Sanskrit vatsá - and the formation of Indo-Iranian and Uralic languages

    Interesting read, but he keeps pushing for Tripolye culture as Late PIE. I don't know if that's ever going to work, but I guess we need more aDNA from Tripolye remains to be sure.

    ReplyDelete
  18. Parpola's own reasoning is always wrong.

    ReplyDelete
  19. I'll simply say “Shah Met" (Checkmate in Persian).

    Indeed Richard :
    https://twitter.com/vagheesh/status/989831821704814592

    ReplyDelete
  20. Narasimhan is often mistaken, he has a bunch of errors in the haplogroup table. He puts haplogroup A, where there is no haplogroup. He has a huge number of disputed definitions and definitions by one SNP(sic!), for example, he defined for sample I1949 from Ganj Dareh(8241-7962 calBCE) R1 the fact that there is a failure, because he is R2a where only positive.
    Etc... Etc... Etc...

    ReplyDelete
  21. @Nirjhar

    Indeed Richard:
    https://twitter.com/vagheesh/status/989831821704814592

    You mean that something CHG like entered the steppe from the Caucasus, mixed with steppe ENF and expanded in the form of Yamnaya? Is it 2015 again? Either way, thanks stating the obvious and then using the obvious to refute something that has nothing to do with Rai's comments.

    ReplyDelete
  22. @All

    We now have several pre-Bronze Age populations sampled from what is now Iran, and none of them look like plausible matches for the southern ancestry in Yamnaya and other Bronze Age steppe populations based on uniparental markers (Y-DNA and mtDNA).

    Yes, one of the Chalcolithic samples belongs to Y-hg R1b-Z2103, which is common in Yamnaya, but this sample is not C14 dated and actually shows some sort of steppe ancestry.

    Of course, everyone is entitled to interpret the data in their own way and believe what they like, but no it doesn't much look like there was any gene flow worth talking about from what is now Iran into the Eneolithic/Bronze Age steppe.

    ReplyDelete
  23. Uniparental markers alone don't provide a complete picture, in fact the evidence from using them alone can be misleading. The time transect from Samara clearly shows ancestry shift from EHG to Khvalynsk which has additional ancestry related to farmers from the Neolithic in Iran, and even more of the same input into Yamnaya. It could be that it is female mediated. But it is clear that the formation of the Yamnaya involved ancestry from the South related to that of Neolithic Iranian farmers and that all the evidence points to that fact that this type of ancestry did not exist in Europe or the steppe in the Paleolithic.

    ReplyDelete
  24. David,

    Have you seen the Yamnaya sample from Ozera ? It has no ANF admixture, only the Iran_N. Where do you think it got that admixture from ?

    In the proximal qpAdm models, Sarazm_EN comes across as a good source of that Iran_N.

    ReplyDelete
  25. Also,

    The PCA clearly shows that in comparison to the Hunter gatherer stage, the later steppe groups are significantly closer to South Asians. Unless we have good no of atleast Bronze Age genomes from South Asia no such conclusion can be derived for South Asia.

    Infact, since the EHG/WSHG lies behind the Steppe_mlba on the PCA, I can just as well say that the South Asians are on a cline towards WSHG. PCA alone does not prove anything.

    ReplyDelete
  26. @music lover

    Uniparental markers alone don't provide a complete picture, in fact the evidence from using them alone can be misleading.

    All of the ancient and modern Iranian sample sets show uniparental markers characteristic of the South Caspian region.

    In other words, just by looking at their Y-DNA and mtDNA profiles it's already clear that they show affinity to what is now Iran.

    Bronze Age steppe samples don't show this. It's rather unlikely that their uniparental profiles have been affected in any significant way by gene flow from the South Caspian.

    Mixture models based on genome-wide data are informative, but they can also be misleading if the same or similar ancient components are involved, so they often can't provide evidence of direct links between populations like hi-res uniparental markers can.

    What we need is to build a complete picture based on all of the data, not just statistical fits from genome-wide data.

    ReplyDelete
  27. @Davidski Yes it's that one. His theory regarding the spread of Uralic languages into their present territories is very interesting.

    But if we accept the framework of expanding Yamna culture = the primary vector of expansion of IE languages, then I think that automatically disqualifies Tripolye and many other cultures.

    ReplyDelete
  28. « The Vara Settlement

    The sudden cooling and the onset of severe winters required the construction of a new kind of settlement and dwellings called a vara (Avestan Vendidad, a book of the Zoroastrian scriptures, chapter 2.25 - part of Zoroastrian scriptures). Vara is both the name of a settlement and the dwellings that made up the settlement (from vara, enclosure).

    The concept of the vara enabled sustainable living for a people and their live-stock in a mountainous region beset with harsh winters. Surviving severe winters without migrating to warmer regions must have been an incredible challenge and a profound development for the people of those days.

    If we put the mythological aspects of the legend aside, the description of the vara in the Vendidad indicates the start of settlement / urban planning in Aryan history. The Jamshidi concept was for the vara to be a self-contained, self-sustaining communal dwelling area built according to a set of uniform principles. There were to be separate areas for humans and animals, as well as for seed and hay storage. Fruit trees and crops were to be planted within the vara area. Water for the inhabitants and crops was to be brought to the vara via a channel and stored in a reservoir. Designated festivals also included a sharing of food resources. In addition, during the Jamshidi era, clay began to be used as a building and construction material for the first time. The houses of the vara were to be constructed using clay and wooden pillars.

    The vara settlement was to be of three sizes: a settlement of a thousand inhabitants with nine streets, six hundred inhabitants with six streets, and three hundred inhabitants with three streets. »

    - Pre-history of the Aryans, Zoroastrian Heritage Institute. http://www.heritageinstitute.com/zoroastrianism/aryans/prehistory.htm

    « Arkaim was a circular stronghold, umbegone by two concentric bastions made of adobe with timber frames, and covered with unburnt clay bricks. Within the circles, close to the bastions, were around sixty dwellings with hearth, cellars, wells and metallurgical furnaces. They opened towards an inner circular street paved with wood. The street was lined by a covered drainage gutter with pits for water collection. At the centre of the complex there was a rectangular open space. The complex had four ingangs, consisting of intricately constructed passages and oriented towards the cardinal points. Evidence suggests that the complex was built according to a plan, which indicates that the society had a developed structure of roles and had leaders with great authority. »

    Wikipedia: Arkaim https://en.m.wikipedia.org/wiki/Arkaim

    « Against the context of the rigveda it is important to note that the Sintashta sites share some cultural features described in this text. These are simple settlements fortified with ramparts and ditches, with a circular or rectangular fence or wall built from unfired clay and wooden frames (pur, RAU 1976). And there are remnants of horse sacrifices (aśvamedha) and primitive horse drawn chariots (ratha, raθa) with spoked wheels (ANTHONY and VINOGRADOV 1995). A real "tripura", Arkaim, was discovered in 1987 by G. ZDANOVICH. It has two circular walls and two circles of dwellings around a central square. The external wall was built from soil packed into timber frames before being faced with adobe bricks. »

    Michaelwitzel, the home of the Aryans https://sites.fas.harvard.edu/~witzel/AryanHome.pdf



    ReplyDelete
  29. @Chetan
    Chetan, what are your views on Parpola's 2015 magnum opus 'The Roots of Hinduism'? I think he overstretches himself in some places, but even if half of what he says is correct, it does put a lot of recent genetic findings in perspective.

    ReplyDelete
  30. @Davidski, Chetan

    From that paper it looks like the theory of "Late PIE Tripolye" involves Sredny Stog elites taking over originally non-IE Tripolye and inventing wheeled vehicles. The obvious alternative is that late Tripolye and PIE tribes invented those wagons independently of each other. If there's no Sredny Stog-looking ancestry in late Tripolye burials that was probably the case.

    ReplyDelete
  31. @velvetgunther, Shaikorth

    Parpola bases his entire Tripolye argument on the existence of native wheel-wagon terminology in PIE, so he reasons the PIE speakers had to be the inventors of the wagon-wheel technology otherwise they would have borrowed the vocabulary from a different language. Since Trypolye, at the time (I don't know if this has changed since then), had the oldest evidence of wheeled wagons, he argues that Late PIE expanded from Tripolye. So that argument rests on weak foundations to begin with.


    Regarding "The Roots of Hinduism", I think Parpola's knowledge counts when it comes to the Indo-Iranian area, especially Indo-Aryan, but he overstretches himself again in trying to come up with his model of PIE expansion.

    ReplyDelete
  32. He equates Catacomb = Iranian. Which I think is flawed along with the Srubna =Iranian theory since neither culture expanded to Central Asia, yet we have Iranian tribes appearing in Central Asia only a few centuries after the Indo-Aryans

    ReplyDelete
  33. @Davidski The only data we have from prior to the BA from the steppe zone, are a single Hunter Gatherer and 3 samples from Khvalynsk. This is uninformative to base any analysis on uniparental markers. However, it is a fact that ancestry related to Iranian farmers from Ganj Dareh and Tepe Abdul Hosein are present in significant proportions in the Yamnaya. From the extensive data published in Mathieson et al, we know that this ancestry is not present anywhere West of the Caspian, i.e. Europe in the Neolithic. From the EHG samples and the newly published samples in this paper from North of Central Kazakhstan, and from the ANE samples, and from the various outlier samples in Sintashta, and the lingering presence of West Siberian Hunter Gatherers that later admix into Steppe MLBA East without any additions of Iranian farmer related ancestry, the indications are that the steppe zone, does not harbor Iranian farmer related ancestry. The only possible solution left is that this ancestry arrives from the South, because there are clear increases of this ancestry from the Neolithic into the Eneolithic and then finally to the BA forming the Yamnaya.

    ReplyDelete
  34. @music lover

    The only data we have from prior to the BA from the steppe zone, are a single Hunter Gatherer and 3 samples from Khvalynsk.

    Not really, because we also have Ukraine_Eneolithic, but in any case, I'm not sure why that matters, because there's actually no unambiguous, direct evidence of gene flow from Iran into the following steppe populations, which translates now into a pretty big database of uniparental markers from the BA steppe...

    - Khvalynsk
    - Ukraine_Eneolithic
    - Yamnaya
    - Afanasievo
    - Sintashta
    - Petrovka
    - Andronovo
    - Srubnaya

    I mean, if there was any significant gene flow from the South Caspian region into the steppe that resulted in the formation of Khvalynsk and Yamnaya, then shouldn't we expect to see this reflected in these later steppe populations?

    And what about Corded Ware and Bell Beakers? Where are the South Caspian uniparental markers in these Yamnaya-related groups?

    The only steppe populations that actually show unambiguous markers from Iran and Turan are some Steppe_MLBA outliers from Central Asia, Sarmatians and Scythians, but that's not surprising and too late to prove anything about what happened on the steppe during the Eneolithic/Bronze Age.

    ReplyDelete
  35. "In other words, the preprint observes that the migration from the steppes to South Asia was the source of the Indo-European languages in the subcontinent. Commenting on this, Rai said, “any model of migration of Indo-Europeans from South Asia simply cannot fit the data that is now available.

    So, not a Yes.

    I'll simply say “Shah Met" (Checkmate in Persian).

    Indeed Richard :
    https://twitter.com/vagheesh/status/989831821704814592


    Checkmate indeed.

    ReplyDelete
  36. @Davidski sounding more and more like the OITers every day when it comes to this Iran-Yamnaya-Gene-Flow connection. Just plain denying the evidence.

    Nobody is a winner when it comes to the great DNA sweepstakes. Its something we all just have to get used to.

    ReplyDelete
  37. @Davidski 1) The movement of all later BA steppe populations into Europe is heavily male mediated. Since the ancestry flow from the south of the steppe was female mediated, you don't expect to see maternal markers from the South of the steppe in Corded Ware, Bell Beakers, or indeed Sintashta, Andronovo, Petrovka and Srubnaya or any other population descended from Yamanya 2) If anything, the difference in ancestry between Ukraine_Eneolithic compared to Ukraine Neolithic is a type of ancestry descending from populations from the South of Ukraine, as this type of ancestry is absent in Europe or the Steppe at any time period prior. 3) Uniparental markers are subject to genetic drift in a way overall ancestry proportion is not 4) Unlike the Y-chromosome, the maternal uniparental markers are way more diverse, so with the sample sizes we have we cannot compute statistics, though an Armenia_Chalcolithic sample has a U4a maternal haplogroup later found in Yamnaya. 5) Simply invoking uniparental markers alone, we know that Yamnaya_Samara don't have the right Y chromosomes seen in later Bell Beakers, so are you going to claim that Yamnaya related ancestry or ancestry from the steppe did not arrive in Europe? 5) I don't even want to get into Anatolia and the presence/absence of steppe ancestry there, and it's association with IE. It is possible to invoke fringe arguments there related to elites and cremation and so on. These points examined together, basically make it clear that there is ancestry in Yamanya, in significant proportions that is descended from populations containing a type of ancestry that is completely absent on the Steppe, Europe, or Siberia prior to the Neolithic and is seen in Iran and the South Caucuses at very early times.

    ReplyDelete
  38. There is gene flow from Iran/Caucasus into the steppe, during the expansion of agriculture into those regions

    See: https://www.academia.edu/9424525/Map_Expansion_of_farming_in_western_Eurasia_9600_-_4000_cal_BC_update_vers._2017.2_

    But it's too early even for the proposed model of Proto Anatolian from the Caucasus.

    ReplyDelete
  39. @music lover

    All of these steppe and closely related populations show a lot of southern mtDNA haplotypes that weren't present on the steppe prior to the Eneolithic.

    - Khvalynsk
    - Ukraine_Eneolithic
    - Yamnaya
    - Afanasievo
    - Sintashta
    - Petrovka
    - Andronovo
    - Srubnaya
    - Corded Ware
    - Bell Beaker

    That's a hell of a lot of southern mtDNA haplotypes, wouldn't you agree?

    However, none of these haplotypes shows an unambiguous link to what is now Iran, past or present. And I won't even get into the Y-DNA side of the argument.

    Coincidence? Drift? Nope.

    Simply, there were no migrations from what is now Iran into the steppe that gave rise to Khvalynsk and Yamnaya.

    These migrations, or maybe just gradual gene flow via female exogamy, came from somewhere else south of the steppe.

    I don't know from where exactly, but I already know it wasn't from what is now Iran. I'm betting on a part, or parts, of the Caucasus.

    ReplyDelete
  40. Would like to see some pigmentation data on them. Are they as depigmented as that 2009 study showed Andronovo to be I wonder. 38 samples is a sufficient size.

    ReplyDelete
  41. I'm not sure Rai should be drawing those conclusions from the ONE sample he has in his study. No one is going to accept a dating of the Rigveda from that.

    ReplyDelete
  42. R1a-Z93 will be found in eastern Corded Ware (Fatyanovo-Balonovo)-> Abashevo. We have no ancient dna from eastern Corded Ware yet, so the absence of R1a-Z93 does not disprove a Corded Ware origin. Corded Ware is probably themselves derived from Stedny Stog

    ReplyDelete
  43. @Davidski I reiterate, all movement of populations after the formation of Yamnaya, were extremely male mediated. So examining maternal markers in Bell Beakers or Sintashta or any other population descended from Yamnaya makes no sense whatsoever and these are non-data points. This is a well established fact from the Y chromosome data; 95% R1x, meaning that you won't find any markers from South of the steppe there. The only data we have that is even remotely conclusive is the autosomal data, and it is a fact that NW Iran works as a source for Yamnaya and Khvalysnk. I am forced to reiterate twice. If uniparental markers are the only source of information, then by that definition. Yamnaya_Samara will not be a source for any of the Bell Beakers, or indeed later Europe. Does that invalidate ancestry from the steppe into Europe? This is a yes or no question. Please answer this prior to any other arguments presented.

    ReplyDelete
  44. Speaking of relation of Cucuteni-Tripolie and steppe here are some significant quotes from the Manzura paper:



    Speaking of late copper age:

    "This new wave of ideological impact on steppe groups was apparently accompanied by influence in other cultural spheres including the domestic one. However, results of this inflhence manifested themselves at another stage of development. In general, it was just the beginning of active adaptation of steppe territories by Cucuteni-Tripolye communities. In the course of this process the steppes were literally ‘domesticated’by farming groups and included in theeconomic, social and ideological aspects of their culture, and the former cultural margin beganto be transformed into a cultural periphery, actively linked to Cucuteni-Tripolye groups in anorganic relationship."

    Then speaking of the bronze age:

    "At the same time, different Tripolye groups penetrated the steppes of the Azov-Dneprregion where the distinctive Dnepr-Bug group appeared. It contained Tripolye, North Caucasianand local steppe elements. Remarkable changes can be noted in earlier steppe cultures such asKvityana, Dereivka and Molyukhov Bugor. Here, the impact of the Tripolye culture can nowbe traced not only in the mortuary but in the domestic sphere as well. Owing to this influencea distinct metalworking centre emerged in the Kvityana culture (Ryndina 1998, 170–9). The traces of the influence can also be seen in the Tripolye vessels or their imitations, and in anthropomorphic and zoomorphic figurines found in both settlements and graves.It signifiesthat at this stage ( early bronze age) the influence of the Tripolye culture embraced all aspects of the steppe communities including social relations, economy, ideology, symbolic system and so on.The fully acculturated steppe groups finally became a proper periphery of the Tripolye culture."

    I ask every people with a decent IQ how can we rule out the chance that the language of the R1a guys in eastern ukraine was a "farmer" related language with such a tremendous acculturation process coming from the west: I requote in capital

    "THE INFLUENCE OF THE TRIPOLYE CULTURE EMBRACED ALL ASPECTS (ALL!) OF THE STEPPE COMMUNITIES INCLUDING SOCIAL RELATIONS, ECONOMY, IDEOLOGY (!!!), SYMBOLIC SYSTEM AND SO ON. THE FULLY ACCULTURETED STEPPE GROUPS FINALLY BECAME A PROPER PERIPHERY OF THE TRIPOLYE CULTURE."

    ReplyDelete
  45. @music lover

    You appear to be claiming now that mtDNA data are irrelevant in establishing the source of southern ancestry on the Bronze Age steppe, because gene flow from Yamnaya to later steppe groups was significantly male-mediated?

    What if I tell you that Yamnaya shares a lot of its mtDNA gene pool with Srubnaya, Sintashta, Andronovo, Corded Ware, Bell Beakers, etc., including many of its southern haplotyes? I can prove this too. You can have a look for yourself.

    So let me reiterate again, there is no direct evidence of any gene flow from what is now Iran into the steppe during the Eneolithic/Bronze Age.

    ReplyDelete
  46. @music lover

    If uniparental markers are the only source of information, then by that definition. Yamnaya_Samara will not be a source for any of the Bell Beakers, or indeed later Europe. Does that invalidate ancestry from the steppe into Europe? This is a yes or no question. Please answer this prior to any other arguments presented.

    And by the way, Bell Beakers and indeed present-day Europeans share southern mtDNA haplotypes with Yamnaya that can be traced back to the Caucasus (not to Iran).

    ReplyDelete
  47. @coldmountains "R1a-Z93 will be found in eastern Corded Ware (Fatyanovo-Balonovo)-> Abashevo. We have no ancient dna from eastern Corded Ware yet, so the absence of R1a-Z93 does not disprove a Corded Ware origin. Corded Ware is probably themselves derived from Stedny Stog"

    Hmm agreed. But it's the absence of L657 in Sintashta-Andronovo that is more concerning. The day L657 is found in Sintashta/Andronovo and L51 is found in West Yamna, I would consider the steppe model to be sealed.

    ReplyDelete
  48. @Davidski Europe_MN has overlapping overall autosomal ancestry related to Anatolian farmers and also mtDNA with Yamnaya so it is hardly a surprise, you can have 100% male mediated ancestry and still see this sharing. I am saying that using uniparental markers to establish flows of ancestry is fundamentally flawed. And I re-asking/iterating my yes or no question. I am just trying to understand your assumptions. If you claim that uniparental markers are unambiguous then you are implying that ancestry from populations on the steppe in the Bronze Age did not make ancestry impact into Europe, because there are no Y chromosomes from Yamnaya that work. We can discuss, any further comments after you unambiguously answer this question, because we need to define a single methodology before discussing data or its interpretation.

    ReplyDelete
  49. @music lover

    I can't give you a yes or no answer to your question for two important reasons:

    1) Present-day Europeans share two relatively young Y-haplogroups with Eneolithic/Early Bronze Age steppe populations, which are R1a-M417 and R1b-M269, while Bronze Age steppe populations don't appear to show such an unambiguous Y-haplogroup link to pre-Bronze Age Iranian populations

    2) Despite claims of male-biased gene flow from the steppe into Europe, present-day Europeans share many mtDNA haplotypes with Bronze Age steppe populations that they don't share with Neolithic farmers or hunter-gatherers west of the steppe, while Bronze Age steppe populations don't appear to show such an unambiguous link to pre-Bronze Age Iranian populations

    Hence, we can infer that Bronze Age steppe populations migrated into Europe just from uniparental data, but we can't infer that any pre-Bronze Age Iranian population migrated into the steppe from such data.

    Unless of course you can explain how else, say, R1a-M417 and R1b-M269 suddenly showed up in Europe west of the steppe after the Neolithic? But I can't.

    At the same time, there are no such dilemmas when talking about the supposed gene flow from Iran into the Eneolithic/Bronze Age steppe. Unless you can give me an example of something?

    ReplyDelete
  50. Manzura was not credible, he distorts and overrates. His method of discussion with his opponents (like proponents of the Kurgan hypothesis) is unscientific.

    ReplyDelete
  51. @ music lovers

    Are you implying something happened like which happened with Neanderthal Genes where the Genes are present without their specific Haplogroups surviving ? So basically Genes from Iran without their Haplogroups surviving ?

    ReplyDelete
  52. But the southern mtDNA haplogroups that accompanied the southern genome-wide ancestry into the steppe did survive, and did very well there.

    It's just that they weren't from Iran.

    ReplyDelete
  53. @Davidski It is a yes or no question. We are talking about the scientific method here. It doesn't matter what data you use, but the methodology needs to be sound. It is a fact, that the Yamnaya_Samara Y chromosomes are not the the Bell Beakers of Europe or later Western European populations, including into the modern day. So if uniparental markers unambiguously explain shifts in ancestry, your inference has to be that we do not have a plausible source population from the steppe bringing ancestry into Europe in the Bronze Age. However, if you choose to base your scientific method on the fact that we haven't sampled the right Y chromosomes, because these societies are patrilocal and that uniparental markers have biases in sampling, drift as well as massive variance in the way in which they correlate with the actual mixture proportions in the resulting admixed populations in a way autosomal ancestry doesn't, then it is clear that ancestry related to the Yamnaya, which have the right autosomal mix arrive in Europe. Similarly, the Y chromosomes may not match Iran, but the early populations of North West Iran have the right autosomal ancestry as a source for the Bronze Age populations of the steppe. Any claims contrary to this is bias, not science.

    ReplyDelete
  54. @Ric Hern I am just stating in no ambiguous terms that Davidski's scientific approach of analyzing uniparental markers and claiming that there was or there wasn't population movement is fundamentally flawed, and worse, treating the situation where the currently sequenced Yamnaya as a source for Bell Beakers in Germany and NW Iranians as a source for the Yamnaya, as different is not only flawed in a technical genetic sense but also from a scientific philosophy sense.

    ReplyDelete
  55. @music lover

    "Does that invalidate ancestry from the steppe into Europe? This is a yes or no question. Please answer this prior to any other arguments presented."

    No. Off course it doesn't. Based on *only* uniparental markers the steppe is by far the most parsimonious source.

    ReplyDelete
  56. @epoch2013 If you look closely at the Yamnaya_Samara Y chromosome calls, they don't match the Y chromosomes in Western Europe. They are related but not the same. I just want to apply the same rational scientific basis in every situation.

    ReplyDelete
  57. Yamna people must have come from somewhere, right?

    What's so special about the Pontic Caspian steppe that it was able to harbour a population distinct from others surrounding it, from the Neolithic? Certainly not geographic topology.





    ReplyDelete
  58. @music lover

    "They are related but not the same."

    And if you look closely at the available Iranian Y chromosome calls you'll won't find any that match *any* R1b apart from one Z2103. Z2103 is pretty evolved. Now tell me, do you really think that an evolved R1b subclade could have originated from an area were there aren't *any* related, but not the same, clades of R1b, where there doesn't seem to be one single ancient R1b apart form that? Or do you believe that it is most likely that said mutations originated in a place where we can also find related but not the same samples, i.e. it's brothers and cousins and (grand-)parents?

    Mind you, while this is nt a yes/no question it *is* a Boolean: Either you pick the former *or* the latter.

    ReplyDelete
  59. @mzp1

    Why didn't farmers settle there? The area is covered with thick Löss soils, the best available soils for cereals.

    If it wasn't special in some kind of way, why didn't CT colonize it?

    ReplyDelete
  60. @epoch213 the Yamnaya Samara are derived for a particular haplogroup that is not found in Western Europe or in the German Bell Beakers. This is exactly the same situation with NW Iranian farmers and the Yamnaya

    ReplyDelete
  61. In the previous thread Rob asked a very simple question and no one answered.

    @Rob

    "Based on this when do you envisage the IA migration/ colonisation/ invasion occurred & how ?"

    I'm not seeing any explanations just people saying: "it happened, deal with it". I'll just list up the most well known scenarios and why Andronovo can't be the Indo-Iranian homeland.

    1. The IAMC Indo-Aryans scenario.
    This one as you know is the Swat one with some contacts between the Indus and Andronovo. Yet this scenario cannot explain the Mittani and the Indo-Aryan Kassite names.

    2. Late BMAC Indo-Aryans
    This was the thing back in the day but modern day archaeology have moved away from it. There seems to be less BMAC-Indus contacts during the Late BMAC period. Also, too late for the Mitanni and while we can't discuss the Rigveda anymore it is also too late for that like 1. above. Mind you Witzel denies the existence of horses in Early BMAC, even though we have proof that they were riding horses before the Andronovans did based on cylinder seals and cheekpieces, but even he doesn't deny the Indo-Iranian influence in the Zagros he just downplays it.

    3. Early BMAC ninjas
    Like 2. a steppe group adopts the entire BMAC culture but keeps the Indo-Iranian languages. This one can explain Indo-Aryans in the near east but like 2. it is the definition of special special pleading. How come the Indo-Iranians adopt the entire culture of BMAC but keep the language? Why couldn't they adopt the language as well? We know from history that the Indo-Iranians elites rarely imposed their languages. The Mitanni elite did not impose their language on the Hurrians. The Persians ruled over 44% of the world's populations and still did not impose the language on their subjects, other than the absorbed Elamites who were utterly decimated by the Assyrians by that time.

    Basically, most theories are entirely dependent on Indo-Iranian Andronovans adopting the entire culture of BMAC. It is less convoluted if we just consider BMAC to be the culture of some Indo-Iranian group.

    ReplyDelete
  62. @David and all:

    DA45 is an incredibly high coverage genome, with 26 Gibabases of reads.

    He's extremely important, because he's our first high-coverage true ancient East Asian genome.

    DA45 Y: O-F871*, primarly South Chinese
    mtDNA: D4b2b2b, found in one example from Russia
    Gedmatch: Z031856

    1240k SNPs in 23andMe format:
    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA45/genome_DA45-1240k.zip

    This individual is not really the same as any Han Chinese from Beijing or Han Chinese from the south. However, he has strong connections to minority people from South China, as well as some steppe peoples under Chinese influence.

    Based on his K15 Orace4 results, this leaves open two possibilities, and his exact ethic origin would depend on the degree of Han admixture among these two groups, as well as the connections to South Chinese minorites.

    A more general group that DA45 could belong to would be the Qiang nomads of Western China, who are thought to be of Tibeto-Burman origin, and ancestral to the Naxi, Tibetans, and Tanguts.

    I think there's a possibility that he's part of the Tanguts of the Western Xia. The Tanguts were a Southern Chinese Tibeto-Burman-speaking people who migrated northward from eastern Tibet and establshed the Western Xia Empire in Northwest China.

    There were other groups near ancient China that were in contact with the proto-Han Chinese Neolithic people of the Yellow River Valley civilization. For example the Tu (Monguor) were a proto-Mongolic people who lived to the north and east of ancient China. The problem with DA45 is that he clearly has a strong South Chinese Tibeto-Burman component, and that this would be lacking in the nomadic people north of China.

    Here are DA45's K15 Oracle4 results:

    Notice his close affinity to the Yizu (Yi) of Yunnan, and the Naxi of Yunnan and Sichuan.

    Using 1 population approximation:
    1 Yizu @ 4.303969
    2 Naxi @ 4.581689
    3 Japanese @ 9.458829
    4 Tu @ 10.700116
    5 Tujia @ 12.893726
    6 Miaozu @ 15.433413
    7 She @ 17.395081
    8 Lahu @ 18.746147
    9 Xibo @ 21.739904
    10 Tibeto-Burman_Burmese @ 23.256525
    11 Hezhen @ 23.266956
    12 Vietnamese @ 25.274090
    13 Cambodian @ 28.311504
    14 Malay @ 29.679453
    15 Dai @ 30.302412
    16 Mongolian @ 47.340858
    17 Kirgiz @ 50.739670
    18 Uygur @ 52.440399
    19 Kazakh @ 52.818272
    20 Buryat @ 56.257896

    Using 2 populations approximation:
    1 50% Tu +50% Tujia @ 3.188284

    Using 3 populations approximation:
    1 50% Miaozu +25% Tu +25% Xibo @ 2.880796

    Using 4 populations approximation:
    1 Miaozu + Miaozu + Tu + Xibo @ 2.880796
    2 Hezhen + Miaozu + She + Tu @ 2.883514
    3 Miaozu + She + Tu + Xibo @ 2.906702
    4 Hezhen + Miaozu + Miaozu + Tu @ 2.920462
    5 Hezhen + She + She + Tu @ 2.960979
    6 She + Tu + Tujia + Xibo @ 2.975740
    7 Japanese + Lahu + Tujia + Xibo @ 2.994522
    8 Dai + Tujia + Xibo + Xibo @ 3.008526
    9 Dai + Hezhen + Miaozu + Xibo @ 3.019304
    10 Japanese + She + Tu + Yizu @ 3.021522
    11 Miaozu + Tu + Tujia + Xibo @ 3.026930
    12 She + She + Tu + Xibo @ 3.030369
    13 Dai + Miaozu + Xibo + Xibo @ 3.057559
    14 Hezhen + She + Tu + Tujia @ 3.059136
    15 Dai + Hezhen + She + Xibo @ 3.059402
    16 Dai + Hezhen + Tujia + Xibo @ 3.060216
    17 Japanese + Lahu + Miaozu + Xibo @ 3.076950
    18 Hezhen + Japanese + Lahu + Miaozu @ 3.102531
    19 Hezhen + Lahu + She + Tu @ 3.106411
    20 She + Vietnamese + Xibo + Xibo @ 3.107765

    ReplyDelete
  63. @music lover

    But what of the two scenario's is parsimonious? It was a boolean question.

    ReplyDelete
  64. @Vara Some early Andronovo groups appeared in Turkmenistan and Uzbekistan by 1700 BCE.

    https://books.google.co.in/books?id=OZ0gAAAAQBAJ&pg=PA357&lpg=PA357&dq=alakul+andronovo+indo-aryans&source=bl&ots=SUIqW1Oh0M&sig=KDTx-BPK9Fwr5FkqETZmefx3BhM&hl=en&sa=X&ved=0ahUKEwizvtTvjNraAhUHRY8KHRY8B2AQ6AEIcjAN

    We can know infer that these groups took the IAMC path from the Siberia/Altai region. The origin of both branches of Indo-Aryans would be from these groups.

    ReplyDelete
  65. Chetan,

    IAMC -> Indus Colonisation?? Come one buddy. IAMC didnt have close to the population density to get full dominance in South Asia + Mittanni.

    It is just a backwater.

    ReplyDelete
  66. “We can know infer that these groups took the IAMC path from the Siberia/Altai region. The origin of both branches of Indo-Aryans would be from these groups.”

    That would require travelling via Afansievo culture
    Perhaps some Afansievo were R1a?

    ReplyDelete

  67. @epoch213 Of course the latter, but this has no bearing on our interpretation of Iranian Neolithic related ancestry in Yamnaya. Now why don’t you answer my yes or no question.

    ReplyDelete
  68. @Mzp and Rob Perhaps there were many interactions involved. But we are not yet in a position to obtain a full answer.

    And mzp, I didn't say the IAMC colonized Indus. Andronovo groups from Central Asia migrated into the Indus. I doubt there was any colonization ala the 18th century British conquest of some Indian regions, if you have such an event in mind. Much more likely these groups emerged successful in the region after the time of transition (end of IVC). Did you read the reference?

    ReplyDelete
  69. So you think the immigrants had already crossed the BMAC and were in South Asia before 1400BC, the cut-off date for the data in the Reich study?

    ReplyDelete
  70. @ Supernord

    “Manzura was not credible, he distorts and overrates.”

    Haha . Easy on there Prof
    You have Kotova, Potekhina, Ivanova, Morgunova, Rassamakin, Parzinger all seeing deep ideological transformations

    @‘Old Europe
    The issue ImO is that the genetic impact did not come from C-T itself

    ReplyDelete
  71. @Mzp The paper doesn't have any cut-off date like that, as far as I'm aware. It only says steppe ancestry started appearing in South Central Asia from 2000 BCE or so.

    ReplyDelete
  72. "Indeed, it's generally accepted based on historical linguistics data that there were fairly intense contacts in North Eurasia between the speakers of Proto-Indo-Iranian, Proto-Uralic and Yeniseian languages."

    Is that why there are iir loanwords in FU but no FU loanwords in Iir? Lol

    It is not because of close contacts between the 2, a bunch of IIr people lived with FU people and never came back. It was a 1 way transmission.

    ReplyDelete
  73. The interesting question for me is whether Sintashta was a CWC offshoot OR a merger of Westerly CWC elements with local Poltavka-like groups, and, if so, to what extent?

    If the proportions were significant(say, 30% Poltavka went into the mix), then the aggregate Western input was quite EEF shifted compared to the G25 CWC averages. So much so, that it might've been essentially like contemporary NorthEast Europeans in terms of deep ancestry/2D clustering.

    ReplyDelete
  74. @music lover
    "the Yamnaya Samara are derived for a particular haplogroup that is not found in Western Europe or in the German Bell Beakers"

    You're referring to the same conclusions of the balanovsky paper?

    Rob

    The revised steppe theory of gimbutas if I'm not wrong put the protoIE in Sredni Stog that had if I recall correctly a 30% farmer input. The same proportion the steppist say was enough to IEize western and central europe.

    ReplyDelete
  75. @Musiclover,
    "the Yamnaya Samara are derived for a particular haplogroup that is not found in Western Europe or in the German Bell Beakers. This is exactly the same situation with NW Iranian farmers and the Yamnaya"

    The mismatch between Iran & Yamnaya is mtDNA, the mismatch between Yamnaya and Bell beaker is Y DNA. Those are different kinds of miss matches because Y DNa & mtDNA behave differently. Y DNA is prone to founder effects. Yamnaya & Bell Beaker actually have a close Y DNA relationship. They belong to two different clades of R1b L23. R1b L23 is only like 8,000 years old. Most mtDNA haplogroups are much older than that.

    mtDNA is a better way to measure population relationship because it doesn't experience founder effects as often as Y DNA. This is why Sintashta, Yamnaya, Bell beaker share mtDNA. So, the lack of a clear mtDNA relationship between Yamnaya & Iranian farmers (which we now have lots of mTDNA data from) is good evidence that Yamnaya didn't have much ancient iranian ancestry. It isn't comparable to the Y DNA mismatch between Yamnaya & Bell beaker.

    ReplyDelete
  76. @Rob

    HAHAHA to you. Tripolie was generally nothing to do it. It, except for imports, onto the Steppe is almost no effect. Do not confuse it with the Balkan-Carpathian metallurgical province common influence. This Steppe has made expansion with the formation of the Usatovo culture type. Read Kotova etc...

    ReplyDelete
  77. @Chetan

    Unavailable for viewing. As far as I know there are no Andronovo artifacts in Iran. Witzel explains the Indo-Iranian influence in the Near East by BMAC.

    "It is probable that this move was preceded by successive spearheading forays of (non-IIr. speaking) mountain peoples into Mesopotamia, such as the Guti, Lullubi, and Kassites3 (c. 2250-1750BCE), who were as yet only marginally influenced by IIr. languages and customs. Some of them are perhaps represented by the sudden expansion of BMAC materials into Susa, Shahdad, Tepe Yahya, Hissar, the Gulf, Baluchistan, the S. Indus area"

    PS. I'm starting to agree with Aniasi on the Indo-Aryan substratum in West Iranic.

    ReplyDelete
  78. Why is it so important that BMAC cultural contributions be matched by BMAC genetic contributions? The Yamnaya were critically affected by cultural developments in the Balkans, despite having at most 7% ancestry from them (even less if the West Asian source of Southern ancestry in Yamnaya also had EEF). If BMAC represents the IIr-IAr population we are faced with the additional problem that South Asians don't have ancestry from them either.

    The Turks, for example, carried cultural elements from Chinese (e.g. twelve year animal cycle in their cosmology, with 'dragon years' replaced by 'snail years', and certain bureaucratic administrative practices in their Khanates, that were quite remarkably out of place in the history of Steppe confederations) as far West as Bulgaria, despite having almost no genetic contribution from Han Chinese proper.

    ReplyDelete
  79. @music lover

    It is a fact, that the Yamnaya_Samara Y chromosomes are not the the Bell Beakers of Europe or later Western European populations, including into the modern day.

    WTF?

    Yamnaya and Western Europeans share R1b-M269. Moreover, there is plenty of R1b-M269 (Z2103+) west of the steppe in Eastern Europe.

    And Western Europeans share mtDNA haplotypes with Yamanya that obviously came from the steppe.

    So even if you don't accept the obvious R1b-M269 link between Western Europeans and Yamnaya, you still have to concede that there is uniparental marker evidence of population movements related to Yamnaya into Europe.

    Now please apply the same logic to the relationship between Bronze Age steppe populations and Neolithic/Chalcolithic Iran.

    ReplyDelete
  80. Open Genomes wrote,

    "DA45 is an incredibly high coverage genome, with 26 Gibabases of reads.

    He's extremely important, because he's our first high-coverage true ancient East Asian genome.

    DA45 Y: O-F871*, primarly South Chinese
    mtDNA: D4b2b2b, found in one example from Russia
    Gedmatch: Z031856

    1240k SNPs in 23andMe format:
    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA45/genome_DA45-1240k.zip

    This individual is not really the same as any Han Chinese from Beijing or Han Chinese from the south. However, he has strong connections to minority people from South China, as well as some steppe peoples under Chinese influence."

    O2a2b2a-F871, and especially O2a2b2a-F871(xO2a2b2a1-N7/F2472), is found more frequently among some Austronesian-speaking populations (Amis, Filipinos, Polynesians, probably also the Batak Toba of Sumatra) than among any population in even southern China. It is in fact the second-most frequent Y-DNA haplogroup among Polynesians after Wallacean/Papuan/Melanesian C1b2a-M38.

    ReplyDelete
  81. @ Supernord

    "HAHAHA to you. Tripolie was generally nothing to do it. It, except for imports, onto the Steppe is almost no effect. Do not confuse it with the Balkan-Carpathian metallurgical province common influence. This Steppe has made expansion with the formation of the Usatovo culture type."

    I just said it's not C-T that was important. You're deranged & confused.

    " Read Kotova etc..."

    Okay ! Let us : -
    "The Hamangia influence in the burial rites of the steppe population was very important and caused to use stone in graves and above them, pits with alcove, new adornments of burial clothes. The strongest impact we have fixed for the population in northern area of the
    Sea of Azov, where the radical changes in the burial rite and the formation of a new Sredniy Stog culture took place. It was connected with the adoption of new religious elements connected with the formation of the centre of steppe metal working."

    Thanks for coming mate.

    ReplyDelete
  82. @ RK

    "he Yamnaya were critically affected by cultural developments in the Balkans, despite having at most 7% ancestry from them (even less if the West Asian source of Southern ancestry in Yamnaya also had EEF)."

    That's a mischaracterisation of the situation.
    Balkan BA, Yamnaya Bulgaria, Sredni Stog - all I2a2a1b like ALPc.
    The autosomal ancestry shifted grossly toward EHG/CHG after c. 3500 BC, but the male lineages stayed.

    ReplyDelete
  83. @music lover

    Did you catch Samuel's comment? Just in case you didn't, here it is again. Emphasis is mine...

    The mismatch between Iran & Yamnaya is mtDNA, the mismatch between Yamnaya and Bell beaker is Y DNA. Those are different kinds of miss matches because Y DNa & mtDNA behave differently. Y DNA is prone to founder effects. Yamnaya & Bell Beaker actually have a close Y DNA relationship. They belong to two different clades of R1b L23. R1b L23 is only like 8,000 years old. Most mtDNA haplogroups are much older than that.

    mtDNA is a better way to measure population relationship because it doesn't experience founder effects as often as Y DNA. This is why Sintashta, Yamnaya, Bell beaker share mtDNA. So, the lack of a clear mtDNA relationship between Yamnaya & Iranian farmers (which we now have lots of mTDNA data from) is good evidence that Yamnaya didn't have much ancient iranian ancestry. It isn't comparable to the Y DNA mismatch between Yamnaya & Bell beaker.

    ReplyDelete
  84. "Why is it so important that BMAC cultural contributions be matched by BMAC genetic contributions?"

    Nobody is arguing that. BMAC cultural contributions are simply too late.

    "If BMAC represents the IIr-IAr population"

    The root of Indo-Iranians can be explained with Jiroft-Helmand River complex as Sarianidi once did. The metallurgy and religion (the soma, fire worship and vara) of BMAC and BA Hissar has it's roots there after all.

    ReplyDelete
  85. The Dardic languages are right in between Indo-Aryan and Iranian, and are sometimes considered a third branch of Indo-Iranian.

    How do you get them there if Indo-Iranian split in the steppe?

    It only works if Indo-Iranian splits in South Central Asia.

    ReplyDelete
  86. Rob said...
    " You're deranged & confused.

    " Read Kotova etc..."

    Okay ! Let us : -
    "The Hamangia influence "

    You were disgraced. Hamangia is not Tripolye! This is a completely different culture, Middle Neolithic along BEFORE Sredniy Stog. Kotova describes not the influence of Tripolye culture, because it is not and she proves it, but the common influence of Balkan-Carpathian metallurgical province in the metallurgy.

    ReplyDelete
  87. DA85, a reasonably high coverage genome, is a mystery:

    Y: L-Y31213, seems to be Near Eastern.
    mtDNA: U4

    Gedmatch: Z710382

    1240k SNPs in 23andMe format:
    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA85/genome_DA85-1240k.zip
    SNPs read: 903693 percent covered: 75.41%

    The mystery is that he appears to be a combination of Uralic and "Tajik".
    What could that be? He's not close to the Tatars or Turkic people. The Uralic component is quite strong, however, he's not close to the Chuvash, who have Uralic ancestry. His Y L-Y31213 isn't the kind found in South Asia, and that wouldn't be something we would expect to see on the steppes.

    What is this about? Which Uralic-admixed steppe people had some Central Asian ancestry? He doesn't match the Hungarians either.

    K15 Oracle4 for DA85:

    Using 1 population approximation:
    1 Tatar @ 18.311073
    2 Chuvash @ 25.314087
    3 Nogay @ 25.758036
    4 Tadjik @ 26.459642
    5 Afghan_Turkmen @ 27.472700
    6 Mari @ 28.496529
    7 Moldavian @ 28.746031
    8 Croatian @ 30.691805
    9 Ukrainian_Lviv @ 30.730938
    10 Hungarian @ 30.779825
    11 Uzbeki @ 30.980347
    12 Erzya @ 31.049278
    13 Kargopol_Russian @ 31.099756
    14 Tabassaran @ 31.103912
    15 Afghan_Tadjik @ 31.205748
    16 Ukrainian_Belgorod @ 31.346106
    17 South_Polish @ 32.007481
    18 Serbian @ 32.055676
    19 Southwest_Russian @ 32.204113
    20 Ukrainian @ 32.270454

    Using 2 populations approximation:
    1 50% East_Finnish +50% Tadjik @ 10.547200

    Using 3 populations approximation:
    1 50% Chuvash +25% Makrani +25% Norwegian @ 9.917451

    Using 4 populations approximation:
    1 Mari + Southwest_Finnish + Tabassaran + Tadjik @ 8.085356
    2 Mari + North_Swedish + Tabassaran + Tadjik @ 8.150370
    3 Finnish + Mari + Tabassaran + Tadjik @ 8.165405
    4 Afghan_Pashtun + Mari + Southwest_Finnish + Tabassaran @ 8.298489
    5 Kalash + Mari + North_Swedish + Tabassaran @ 8.307493
    6 Afghan_Pashtun + Mari + North_Swedish + Tabassaran @ 8.357798
    7 Kalash + Mari + Southwest_Finnish + Tabassaran @ 8.363769
    8 Afghan_Pashtun + Finnish + Mari + Tabassaran @ 8.395761
    9 Burusho + Mari + North_Swedish + Tabassaran @ 8.432208
    10 East_Finnish + Mari + Tabassaran + Tadjik @ 8.436080
    11 Brahui + Chuvash + Mari + Norwegian @ 8.457118
    12 Brahui + Chuvash + Irish + Mari @ 8.490207
    13 Balochi + Chuvash + Mari + Norwegian @ 8.514278
    14 Balochi + Chuvash + Irish + Mari @ 8.518743
    15 Finnish + Kalash + Mari + Tabassaran @ 8.569997
    16 Burusho + Mari + Southwest_Finnish + Tabassaran @ 8.589872
    17 Brahui + Chuvash + Mari + West_Scottish @ 8.613857
    18 Balochi + Chuvash + Mari + North_German @ 8.639964
    19 Balochi + Chuvash + Mari + West_Scottish @ 8.644436
    20 Brahui + Chuvash + Chuvash + Norwegian @ 8.644952

    ReplyDelete
  88. @Open Genomes

    Thanks, I'll run DA85 in the Global25 later today.

    ReplyDelete
  89. No one knows where the Aryan languages have disintegrated.
    Soma is from West Siberia.
    Cult of fire from Europe, cremation from Europe.
    In the early Andronovo culture, cremation prevails in the Southern Urals, and such burials stretch from Western Siberia to the Yenisei river and then go to Eastern Kazakhstan.
    The prototype of the Varas was Sintashta.

    ReplyDelete
  90. @Davidski The point is that the Yamnaya and Western Europe, including the Bell Beakers who bring steppe ancestry into Western Europe do not share the same exact Y haplogroup. Yamnaya are Z2103, while Western Europeans are mostly L51. The timing isn't of consequence because at some level everyone shares uniparental markers with each other, just deeper along the tree so Iranians having haplogrup U. The fact that Bell Beakers and modern Western Europeans sharing maternal uniparental markers with Yamnaya is also not a valid argument because this could also be because both Yamnaya and Western Europeans share Anatolian farmer related ancestry prior to the Yamnaya migrations westward. In any event if you want to make a link between Bell Beakers and the Yamnaya as a maternal migration, be my guest.

    ReplyDelete
  91. @music lover

    Not only are the "Bell Beaker" R1b-L51 and "Yamnaya" R1b-Z2103 closely related sister clades, but there are now Bell Beakers in the ancient DNA record with R1b-Z2103.

    These are obvious smoking guns in uniparental data of Yamnaya-related migrations deep into Europe.

    But even if you want to ignore this for whatever odd reasons you might have, then there are strong links between Western Europeans and the Bronze Age steppe peoples in terms of mtDNA haplotypes that have nothing to do with Anatolian farmers, because they're Eastern European Hunter-Gatherer (EHG) derived haplotypes.

    At this moment, there are no such direct and unambiguous uniparental links between Yamnaya and Iranian farmers.

    ReplyDelete
  92. @ Supernord
    Please go easy on the Vodka. Read my original comment - "@‘Old Europe
    The issue IMO is that the genetic impact did not come from C-T itself"
    So maybe Mazura focussed on C-T, slightly off. Still better than others who completely missed the formative aspect of the Kurgan culture being dependant on acculturation and migration from the agro-pastoral west.

    ReplyDelete
  93. @davidski Can you point to a single Bell Beaker sample that is R1b-Z2103? As far as I've looked they are all R1b-L51.

    ReplyDelete
  94. @music lover

    Here you go...

    Beaker Central Europe I7044 R1b1a1a2a2c1
    Beaker Central Europe I2787 R1b1a1a2a2

    There are now Bell Beaker samples from the same burial sites that belong to both R1b-Z2103 and R1b-L51. Obviously, they arrived from the steppe at the same time.

    ReplyDelete
  95. @ Vara

    Yes i think the issue is the absence of convincing explanatory model for the I-A'zation of S-CA. There is certainly a palpable steppe presence throughout, but it is neither a demic mass migration nor an elite conquest. So we need new models. One would have thought a paper featuring Vidale and Frachetti might have attempted an explanation.

    IMO, Indian IE -speakers can be modelled as a 2-way 5-20% steppe + IVC.
    Whilst Dravidians and Munda can be modelled as IVC + extra 'basal ASI' or whichever we want to refer to it.

    https://i.imgur.com/usfAIKm.png

    Also captured in the image is the more complex ancestry of Iranian speakers :
    Swat _IA, Dardic groups, Pashtun and Tajiks can be represented as a 3 way mix of Turan Eneolithic, Steppe MBA and IVC, with Tajiks differentiated by a large steppe pull. (In part Dali related)
    Balochi, Brahui rather are rather differentiated by an 'old Iran Neolithic ancestry + modest steppe (20%)
    Central Iranians by contrast do not lie on the IVC- Turan axis, but are West Asian pulled (eg Iran Chalcolithic, additional Levant admixture) with some 15% steppe.

    So the unifying aspect is steppe MBA admixture. Was it a series of opportunistic migrations into various nexi who gained currency due to their adaptability and mobility ? It would be helpful to demonstrate a proto-Brahmin pastoral zone in northwest India which acted as a core for IA-zation.

    For those who remained unconvinced and perhaps think 1200 BC Kaskarchi samples are too late for Vedic culture, then they could argue that Dardic, Munda etc derive from various native groups, whilst the differentiation of Indo-Iranian, previously operant in the northern arm of the 'Middle Asian interaction zone (4000-2000 BC) took place exactly in that post 2000 BC period, when BMAC and IVC were fragmenting, with incoming pastoralists acculturating into these new societies...again due to their adaptability, etc.

    ReplyDelete
  96. @David and all:

    Sintashta and Iranian origins: Here we have our first R-CTS6 in aDNA. R-CTS6 is a subclade of R-Z93 that is somewhat rare today, but is mostly found among Iranic-speakers with a single very recent Ashkenazi Jewish Levite clade.

    DA126: Y R-CTS6
    So far, so good ... but his mtDNA is haplogroup F2c1.
    F2c1 is found today in China and Japan.

    DA126 1240k SNPs in 23andMe format:
    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA126/genome_DA126-1240k.zip
    SNPs read: 182219 percent covered: 15.20%

    DA126 has rather poor coverage, so he may not be suitable for Global25, which is too bad.

    However, we can see his calculator results in Gedmatch:
    Z580575
    He's quite distant from everyone today, but he seems closest to the Iron Age Scythian I10247. In K13 Oracle4, he comes out as a rather even mix of Northern Europeans (Irish, West Scottish, and Swedes) and northern East Asians such as Mongolians and Xibo. He also has an element of Caucasus ancestry (Tabassaran of Dagestan).

    A mystery. His Y descendants ended up in Iran, but this unusual combination did not.

    Using 1 population approximation:
    1 Tatar @ 23.425938
    2 Nogay @ 25.984245
    3 Afghan_Turkmen @ 27.064018
    4 Uygur @ 28.345562
    5 Uzbeki @ 29.191977
    6 Aghan_Hazara @ 29.206676
    7 Hungarian @ 31.015442
    8 Moldavian @ 31.296228
    9 Serbian @ 31.970407
    10 Chuvash @ 31.972649
    11 Austrian @ 32.156010
    12 Hazara @ 32.474373
    13 Tadjik @ 32.692543
    14 Romanian @ 32.865952
    15 Croatian @ 33.015057
    16 East_German @ 33.029274
    17 West_German @ 33.128757
    18 Kazakh @ 33.419693
    19 North_German @ 33.766376
    20 South_Dutch @ 34.148926

    Using 2 populations approximation:
    1 50% Hazara +50% Swedish @ 12.068843

    Using 3 populations approximation:
    1 50% Swedish +25% Tabassaran +25% Xibo @ 9.964124

    Using 4 populations approximation:
    1 Irish + La_Brana-1 + Mongolian + Tabassaran @ 9.809051
    2 La_Brana-1 + Mongolian + Tabassaran + West_Scottish @ 9.867119
    3 Hezhen + Swedish + Swedish + Tabassaran @ 9.886449
    4 Swedish + Swedish + Tabassaran + Xibo @ 9.964124
    5 Hezhen + North_Swedish + Swedish + Tabassaran @ 9.987449
    6 Hezhen + North_Swedish + Norwegian + Tabassaran @ 10.010129
    7 North_Swedish + Swedish + Tabassaran + Xibo @ 10.019039
    8 North_Swedish + Norwegian + Tabassaran + Xibo @ 10.053046
    9 Irish + La_Brana-1 + Lezgin + Mongolian @ 10.054248
    10 La_Brana-1 + Mongolian + Norwegian + Ossetian @ 10.058338
    11 La_Brana-1 + Mongolian + Norwegian + Tabassaran @ 10.062474
    12 Hezhen + Norwegian + Swedish + Tabassaran @ 10.072444
    13 La_Brana-1 + Mongolian + Orcadian + Tabassaran @ 10.101462
    14 La_Brana-1 + Lezgin + Mongolian + West_Scottish @ 10.104564
    15 Hezhen + Lezgin + Swedish + Swedish @ 10.106409
    16 Irish + La_Brana-1 + Mongolian + Ossetian @ 10.113779
    17 La_Brana-1 + Mongolian + North_Dutch + Tabassaran @ 10.122746
    18 Chechen + Irish + La_Brana-1 + Mongolian @ 10.137945
    19 Norwegian + Swedish + Tabassaran + Xibo @ 10.159996
    20 Hezhen + North_German + North_Swedish + Tabassaran @ 10.164265

    ReplyDelete
  97. DA85, Y: L-Y31213, mtDNA: U4 is a Hunnic elite sample from Hungary. It seems like Y-DNA L and mtDNA U4 have a love story(Armenia ChL had it too).

    MDLP K16

    # Primary Population (source) Secondary Population (source) Distance
    1 61.5% Udmurd (Udmurtia) + 38.5% Ishkasim (Gorno-Badakhshan) @ 7.86
    2 71.7% Udmurd (Udmurtia) + 28.3% Jatt (Haryana) @ 7.91
    3 75.1% Udmurd (Udmurtia) + 24.9% Burusho (Pakistan) @ 7.95
    4 60.6% Udmurd (Udmurtia) + 39.4% Tajik (Pomiri_Tajikistan) @ 8.02
    5 59.4% Udmurd (Udmurtia) + 40.6% Shugnan (Badachshan) @ 8.05


    Using 3 populations approximation:
    1 50% Shugnan_Badachshan +25% Tubalar_(Altai) +25% Vepsa_Russia @ 6.442919


    Punt k12

    1 Balochi + Estonian + Lezgin + Selkup @ 4.662281
    2 Balochi + Lezgin + Lithuanian + Selkup @ 4.662537
    3 Brahui + Estonian + Lezgin + Selkup @ 4.796335
    4 Brahui + Lezgin + Lithuanian + Selkup @ 4.799502
    5 Estonian + Lezgin + Makrani + Selkup @ 4.888319
    6 Balochi + Lithuanian + North_Ossetian + Selkup @ 4.915496

    ReplyDelete
  98. @Open Genomes and re: "Y: L-Y31213, seems to be Near Eastern." Is there any possibility that the result related to paternal lineage is wrong? Just by looking at Yfull it is very hard to see an apparent connection to Huns.

    ReplyDelete
  99. @ OG,

    I wonder if that may be one of those Y-L Huns we heard about a while back.

    @ RK

    There certainly is BMAC-like ancestry in South Asia. Gonur1_BA ancestry looks to be on par with Sintashta. I am looking at several models now, both with qpAdm and qpGraph. I have been pretty busy, but will post some models as soon as I possibly can, or just as I go.

    ReplyDelete
  100. @All

    You can find Global25 coordinates for the following ancient samples at the link below...

    DA28
    DA45
    Hunnic_Hungary:DA85
    Sidon_BA:ERS179073
    Sidon_BA:ERS1790729
    Sidon_BA:ERS1790730
    Sidon_BA:ERS1790732

    https://drive.google.com/file/d/1IuHoFKUvgQHK1OC6GIYPm9-NSVn0Lf0-/view?usp=sharing

    ReplyDelete
  101. O.K., it seems that L-Y31213 has been found among Kazakhs. Makes sense, then.

    ReplyDelete
  102. DA85 looks like one of those "Fenno-Scythians" in the flesh to me. Seems very similar to what the mediator of Central and East Siberian gene flow to Volga Uralics + Chuvash would look like.

    ReplyDelete
  103. Looking at West_Siberia_N, he is extremely ANE-rich and falls among Ket, Selkup, Mansi and other such peoples in Global25. This seems like an early Yeniseian, and the source of the "Central Siberian" substrate present in Nganasan, Ket, Selkup, Nenets etc more than anything else.

    Not quite Native American-like or Okunevo-like though.

    Would help to explain populations like Ket, which some studies say are even more ANE than Native Americans.

    ReplyDelete
  104. DA85 is very attracted to Sarmatian_Pokrovka. Western Huns = Eastern Huns + Scytho-Sarmatians? Global25 doesn't pick up Germanic or modern Eastern European ancestry for it.

    Hunnic_Hungary_scaled:DA85

    Sarmatian_Pokrovka 50.70
    Hunnic_Tien_Shan:DA101 23.05
    Altai_IA 17.95
    Udmurt 5.85
    Tajik 2.45
    Tabasaran 0.00
    Lithuanian 0.00
    Finnish 0.00
    Nordic_IA 0.00
    Slavic_Bohemia 0.00
    Germany_Medieval 0.00

    ReplyDelete
  105. @Shaikorth

    Care to do a fit for Hunnic Tien Shan?

    ReplyDelete
  106. Would be interesting to do fits w both Hunnics of the Tien Shan and Huns in Hungary... genetics may settle what historians havent.

    ReplyDelete
  107. @ Chad
    Yep same Hun which was tested by the small lab in Kazakhstan. They got it right with STRs .

    ReplyDelete
  108. @Ryu

    Looks like Scytho-Sarmatian+Altaian.

    Hunnic_Tien_Shan:DA101

    Sarmatian_Pokrovka 51.70
    Altai_IA 31.25
    Scythian_ZevakinoChilikta 11.80
    Scythian_AldyBel 5.25
    Scythian_Pazyryk 0.00
    Scythian_Samara 0.00
    Han 0.00
    Selkup 0.00
    Mongola 0.00
    Kanai_MBA 0.00
    Karasuk_o 0.00
    Zevakinskiy_LBA 0.00
    Kashkarchi_BA 0.00

    ReplyDelete
  109. Exploratory runs using the new East Asian samples.

    Wow, it looks like that Karluk sample is quite packed with BMAC-related ancestry. I wonder how representative he was of the average Turkomans? If he was typical, then it has strong implications for the eventual spread of BMAC ancestry north of the Oxus.


    Karluk_Medieval
    Sappali_Tepe_BA_o 40.15%
    DA28_scaled:DA28 34.95%
    DA45_scaled:DA45 10.6%
    Sintashta_MLBA_o3 10.25%
    Sarazm_Eneolithic 3.5%
    Sintashta_MLBA 0.55%

    Dali_EBA 0%
    Sintashta_MLBA_o1 0%
    Sintashta_MLBA_o2 0%
    Scythian_Samara 0%
    Kashkarchi_BA 0%
    Dashti_Kozy_BA 0%
    Sappali_Tepe_BA 0%
    Gonur1_BA 0%
    West_Siberia_N 0%

    Distance 1.818%

    When I remove all direct BMAC-related ancestry, the fit worsens drastically.

    Karluk_Medieval
    Scythian_Samara 57.4%
    DA28_scaled:DA28 31.2%
    DA45_scaled:DA45 11.4%

    Sintashta_MLBA 0%
    Dali_EBA 0%
    Sintashta_MLBA_o1 0%
    Sintashta_MLBA_o2 0%
    Sintashta_MLBA_o3 0%
    Kashkarchi_BA 0%

    Distance 5.2082%

    ReplyDelete
  110. @Chad

    Yeah, I'm getting substantial Gonur-like stuff in almost every single monte run for Brahmins and Kshatriya. Glad to see your formal models confirm this!

    The most BMAC-rich(but not the most steppic) Indian population are the Sindhis, intriguingly enough.

    ReplyDelete
  111. @ Anthro
    Disagree there
    Really not much apart from in Sindhi
    As a whole, IAs don’t have that much BMAC, as per the paper
    They can even be reduced to steppe and IVC

    ReplyDelete
  112. @Rob

    When I say "Indians", I don't refer to modern political boundaries: I apply the term to Indo-Aryan speakers living to the east and south of Pashtuns and Balochs lands as Indian. This is the pre-1948 'definition'.

    The dominant ethnic groups in Pakistan are Punjabi and Sindhi peoples, both of whom speak Indo-Aryan languages and whose ancestors practiced Dharmic faiths(Vedic heritage) back in the day.

    Unlike Sindhis, that Punjabi sample from Lahore(also Pakistan) gets considerably less Gonur-related ancestry despite what geography might suggest. Maybe muhajir ancestry from south india is diluting this, though.

    ReplyDelete
  113. Anthro
    Yes understand you
    So either Sindhi are atypical or all other IAs had their BMAC ancestry diluted

    ReplyDelete
  114. Rob, the thing is, there tends to be a correlation between high caste ancestry and Gonur shift. So, even if 8% doesn't seem like much overall, it's quite a bit when you consider it arriving in tandem with the 18% Sintashta-like. Same deal with Yamnaya and its single digit EEF(at least how I and OldEurope see it).

    ReplyDelete
  115. @anthro
    Could you post gonur %s for various modern castes plz.

    ReplyDelete
  116. @Rob "Please go easy on the Vodka." Do not project your state on others. I answer you on your messages me. Do you drink vodka before you write to me?

    ReplyDelete
  117. thanks stating the obvious and then using the obvious to refute something that has nothing to do with Rai's comments.

    Rai's Comments?.

    ReplyDelete
  118. Vara,

    What is your opinion on sites like Bustan and the data we have from there?.

    ReplyDelete
  119. @Shaikorth and Ryu, running comparisons of that HTS model against real Hunnic_Tien_Shan:DA101 - https://imgur.com/a/LO3FtZL

    Closer than any real average, but note that the real is further away from European and East Asian than the model, and the real is closer to South Asian, Central Asian, Near East and Siberians. (That is, the combination of Sarmato-Scythian and Altai_IA have more ancestry relating to both East Asians and Europeans than the real Hunnic sample).

    In the rank order of distances as well, while the model is closer than any real average, (probably because it roughly gets the East Eurasian:West Eurasian fairly right and of the right sort), Molaly_MLBA, Alpamsa_MLBA_Alakul and Oy_Dzhaylau_MLBA_o2 are closer than any of the Scytho-Sarmatian+Altaian averages. As makes sense from them being earlier Central Asians than Hunnic_Tien_Shan:DA101. Fits might improve from including these samples too.

    ReplyDelete

  120. @all


    if ukraine was so full of R1a during the copper /early bronze age and since ukraine was the platform from which the invasion of europe took place why we don't have R1a in central-western europe?

    @supernord
    "Cult of fire from Europe, cremation from Europe"

    So a big amount of future SC asia religion and culture coming from europe and at the same time we are told that europeans were surrendering their language to steppe invaders.
    A very unlikely scenario isn't it?

    ReplyDelete
  121. old europe said...
    @all
    "if ukraine was so full of R1a during the copper /early bronze age and since ukraine was the platform from which the invasion of europe took place why we don't have R1a in central-western europe?"

    It can be explained better with more detail- if you tell us the branch of R1a you belong.

    ReplyDelete
  122. @Matt

    The new fit takes three populations you suggested and a trace of AASI construct from Sein, but the distance doesn't improve much (2.1103). Notably the old fit didn't take Zevakinskiy LBA, a more temporally proximate Central Asian than the MLBA samples.

    Hunnic_Tien_Shan:DA101

    Molaly_MLBA 43.85
    Sarmatian_Pokrovka 18.00
    Scythian_ZevakinoChilikta 14.00
    Oy_Dzhaylau_MLBA_o2 10.65
    Alpamsa_MLBA_Alakul 8.70
    Scythian_Pazyryk 2.45
    Mongola 1.60
    Selkup 0.65
    AASI 0.10
    Altai_IA 0.00
    Scythian_AldyBel 0.00
    Scythian_Samara 0.00
    Han 0.00
    Kanai_MBA 0.00
    Karasuk_o 0.00
    Zevakinskiy_LBA 0.00
    Kashkarchi_BA 0.00

    ReplyDelete
  123. @old europe

    You have strange logic. Arians into Asia came in the Bronze age, while the Indo-Europeans from Eastern Europe to Central Europe came in Eneolithic. The cult of fire origin from Eastern Europe, the cremation of South-Eastern Europe. IE genetics denote to Central Europe CWC, CWC genetics connect to Eastern Europe. The spread of the chariot form of domesticated horses from Slovakia to the Urals indicates the region. The origin of cities Sintashta indicates the Balkan-Carpathian region.
    Sredniy Stog culture it's from the Don river, corded Dereivka culture also.

    ReplyDelete
  124. White people can never dominate Northern South Asians. It's impossible. Punjab would be too hot anyway for light-skinned steppe invaders and there is no where to hide from the sun. Why do you think even Northern plains IAs with high-steppe admixture are quite dark, compared to mountain ones. Those light-skinned steppe guys would get serious sunburn before they could conquer anyone.

    ReplyDelete
  125. Comment of the year this...

    mzp1:White people can never dominate Northern South Asians. It's impossible. Punjab would be too hot anyway for light-skinned steppe invaders and there is no where to hide from the sun. Why do you think even Northern plains IAs with high-steppe admixture are quite dark, compared to mountain ones. Those light-skinned steppe guys would get serious sunburn before they could conquer anyone.

    If anyone here thinks he's joking, think again.

    ReplyDelete


  126. supernord

    I talked about cremation and fire cult: as for cremation you say from south-eastern europe......but which period are you referring to?

    ReplyDelete
  127. I think I've just about had enough of this discussion for now. I have seen nothing to overturn the dating of Rigveda to pre/early IVC. I dont have time to keep up-to-date with every single article and then debate against a-priori positions.

    I'll look at new DATA as and when it comes out but for now I am sticking to the consistent, complete and coherent picture I have for IA/South Asian history which has Vedic Aryans as pre-IVC.

    The alternative, the AMT, is just a big black whole of a mess with lots of problems, highly unlikely scenarios, just a big lump of incoherent, incomplete, inconsistent nonesense.

    ReplyDelete
  128. mzp1 said...
    "White people can never dominate Northern South Asians. It's impossible. Punjab would be too hot anyway for light-skinned steppe invaders and there is no where to hide from the sun. Why do you think even Northern plains IAs with high-steppe admixture are quite dark, compared to mountain ones. Those light-skinned steppe guys would get serious sunburn before they could conquer anyone. "

    The northern Paternal cultures, didn't have to invade or dominate. All they had to do, is create a safe environment for their families and the refugees seeking a better future - giving people shelter/food/some early type of welfare.

    ReplyDelete
  129. @ Anthro

    "Rob, the thing is, there tends to be a correlation between high caste ancestry and Gonur shift. So, even if 8% doesn't seem like much overall, it's quite a bit when you consider it arriving in tandem with the 18% Sintashta-like. Same deal with Yamnaya and its single digit EEF(at least how I and OldEurope see it)."

    I'm not really understanding the analogy, because I see no parallel. The build up to Yamnaya on the steppe saw several shift in autosomal sway which included a significant WHG / EEG shift in the Mariupol & Sredny Stog phase. Also your figure for 'Yamnaya' seems cherry picked or based on a massaged definition of Yamnaya (only steppe Yamnaya and not Hungarian & Bulkan Yamnaya). If you wanted to analogise, you perhaps should have used BB. Anyhow, moving onto the relevant issue - in the set up I use Bengali score 0 Gonur, Gujarat 0.9 (vs 12% Sintashta) and Brahmin 6/ 16.
    By contrast:Swat would be 32 vs 16, Iran Persian 62 vs 24. Inverse.
    Sindhi show the same inverse ration (32 v 16). Anyhow just a curiosity, maybe related to later events.

    ReplyDelete
  130. @mzp1 you're just talking nonsense and nothing other only waste words. Northern India is the latitude of Kuwait, Kuwait is home to black men? No one expects to convince you of anything. Inappropriate profane nothing to explain, it is impossible to convince, he only uses his ridiculous inventions. You're bored with your profanity. Scientific dating of the Vedas as pre-IVC time does not exist, it is simply pure unscientific fraud.

    ReplyDelete
  131. "I think I've just about had enough of this discussion for now."

    ReplyDelete
  132. @All!!!!!!!!!!!!!!!!!!!

    Does anybody think they can do a Y DNA analysis of the Narasimhan 2018 genomes?

    ReplyDelete
  133. @Samuel A lot of them are wrong calls for sure. I think it's better to wait for the corrected data

    ReplyDelete
  134. @ mzp 1

    A lot of light skinned people can tan. It doesn't take long to get used to heat. I worked in temperatures as high as 50 Degrees Celsius and my skin are fine. So your reasoning is flawed...

    ReplyDelete
  135. @David

    Thank you for all the Global25s. :)

    Where did you hear that DA85 is Hunnic from Hungary? Or is that based on his Global25 results?

    As far as Nick Patterson / the Reich Lab and the Sidon samples go, I think if they had them they would appear in their .geno tables that they've been using for their studies. I know Nick Patterson never asked for them, and if they had direct access to them, of course they could extract the 1240k SNPs. (The Broad Institute after all wrote GATK, even though I'm using samtools + bcftools.) Many of the Willeslev / Copenhagen RISE samples show up in their 1240k .geno files now. The Bronze Age Sidon Haber et al. (2017) study has been published, so there's no restrictions on their use according to the Ft. Lauderdale principles. As I said, the aligned BAMs or CRAMs should have been made available as a requirement for publication, since they did of course use the alignments in their study. I would think that directly providing the .geno files would just "make things easier" even for the current Narasimhan study. (Yes, Bronze Age Sidon would be important for the Swat Valley samples, whose E-M123* is really found only among Galilee Christians.)

    So it's just to make their study better. I also provided the direct links to the BAM files, in case they want to use their own methodology to call the SNPs.(Mine is one of the two standard ways of doing it, and it's also been independently verified by the calculator results.)

    ReplyDelete
  136. Question: Why are some of you people so obsessed with skin color, when you ignore the really important things like rs53576 (A;A) = lack of emotional empathy, and rs1800497 (T allele) = bad at avoidance of errors?

    These two SNPs are the only ones that directly correlate (over 95%) with the results of psychological tests.

    These are extremely important for the socio-cultural adaptations of the steppe peoples vs. the Neolithic farmers in Europe and South Asia. I at least checked rs53576 for the European Neolithic, and they were almost all (G;G), which includes "higher concern for the in-group / tribalism". This is not true of Europeans today.

    We have these SNPs in the 1240k files, even for the Indus Valley Periphery. It's simple to check them. Anyone want to try and tell us the results? Does rs53576 correlate with Steppe ancestry?

    @Roy King

    ReplyDelete
  137. rs53576 "emotional empathy" frequencies for various modern populations:

    ALFRED database frequencies for rs53576

    rs1800497 "doesn't learn from errors" frequencies for various modern populations:

    ALFRED database allele frequencies for rs1800497

    Well, what about the various ancient groups? You all have access to the 23andMe format 1240k files ...

    ReplyDelete
  138. @Chetan
    We can know infer that these groups took the IAMC path from the Siberia/Altai region. The origin of both branches of Indo-Aryans would be from these groups.
    This has been, and is still my position (with reservations)...

    @Rob
    That would require travelling via Afansievo culture
    Perhaps some Afansievo were R1a?

    ...which would require to find R1a. Without R1a, then yeah, falls apart.

    ReplyDelete
  139. @Atriðr

    Perhaps some Afansievo were R1a?...which would require to find R1a. Without R1a, then yeah, falls apart.

    It doesn't fall apart because the Kashkarchi_BA samples from the Ferghana Valley are identical to Sintashta and belong to R1a-Z645.

    On the doorstep of India

    ReplyDelete
  140. @David and all Sintashta fans:

    The 1240k SNPs of the most coolest people who ever lived on on Earth - ever! are here! The Sintashta descendants, the Scythians, in R-Z94 > R-Z2124 > R-Z2125 (and > R-S23592) - and they even lived near the coolest places on Earth, Sintashta and Arkaim! :D

    DA2: Y: R-Z2125 mtDNA: N1a1a1a1 Gedmatch: Z511676
    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA2/genome_DA2-1240k.zip
    SNPs read: 680153 percent covered: 56.75%

    DA100: Y: R-S23592 mtDNA: C4b1 Gedmatch: Z906895
    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA100/genome_DA100-1240k.zip
    SNPs read: 914767 percent covered: 76.33%

    DA206: Y: R-S23592 mtDNA:HV13b Gedmatch: Z547295
    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA206/genome_DA206-1240k.zip
    SNPs read: 919803 percent covered: 76.75%

    DA385: Y: R-S23592 mtDNA: H13a2a Gedmatch: Z610860
    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA385/genome_DA385-1240k.zip
    SNPs read: 953822 percent covered: 79.59%

    DA243: Y: R-S23592 mtDNA: W1c Gedmatch: Z143419
    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA243/genome_DA243-1240k.zip
    SNPs read: 1065055 percent covered: 88.87%

    And while you all wait for the Global25 result, you can watch this video about the cultural and spiritual significance of kurgans to the Proto-Indo-European and Proto-Indo-Iranian steppe societies and other extremely cool cultures:

    Death and Burial among the Proto-Indo-Europeans:
    An analysis of the cultural significance of steppe Kurgan death rites and their relevance to contemporary descendant cultures

    ReplyDelete
  141. Typical Sintashta Proto-Iranian Scythian? :D

    DA100: Y: R-S23592 mtDNA: C4b1 Gedmatch: Z906895

    K13 Oracle4:

    Using 1 population approximation:
    1 Tatar @ 18.577372
    2 South_Polish @ 20.031906 (gimme dat!)
    3 Ukrainian_Lviv @ 20.073330
    4 Ukrainian_Belgorod @ 20.436430

    5 Hungarian @ 21.079231
    6 Kargopol_Russian @ 21.199106
    7 Ukrainian @ 21.232910

    8 Southwest_Finnish @ 21.246845
    9 Polish @ 21.305059 (Holy Smokes!)
    10 Southwest_Russian @ 21.324619

    11 Finnish @ 21.498823
    12 Croatian @ 21.906588
    13 North_Swedish @ 22.096758
    14 Russian_Smolensk @ 22.238264
    15 Moldavian @ 22.369724
    16 East_German @ 22.385723
    17 Erzya @ 22.736567
    18 Austrian @ 22.890787
    19 East_Finnish @ 22.992086
    20 Belorussian @ 23.567703

    Using 2 populations approximation:
    1 50% Chuvash +50% North_German @ 13.250097

    Using 3 populations approximation:
    1 50% East_Finnish +25% Norwegian +25% Tabassaran @ 9.208694

    Using 4 populations approximation:
    1 Chuvash + Finnish + Norwegian + Tabassaran @ 8.354923
    2 Chuvash + Finnish + Swedish + Tabassaran @ 8.526457
    3 Chuvash + Estonian + Norwegian + Tabassaran @ 8.586221
    4 Chuvash + North_Swedish + North_Swedish + Tabassaran @ 8.607311
    5 Chuvash + Southwest_Finnish + Swedish + Tabassaran @ 8.615351
    6 Chuvash + Norwegian + Southwest_Finnish + Tabassaran @ 8.623148
    7 Chuvash + East_Finnish + Norwegian + Tabassaran @ 8.729832
    8 Finnish + Mari + Norwegian + Tabassaran @ 8.745060
    9 Chuvash + Finnish + Irish + Tabassaran @ 8.758250
    10 Chuvash + North_Swedish + Southwest_Finnish + Tabassaran @ 8.772821
    11 Chuvash + Estonian + Swedish + Tabassaran @ 8.811392
    12 Chuvash + Finnish + Tabassaran + West_Scottish @ 8.838740
    13 Estonian + Mari + Norwegian + Tabassaran @ 8.844707
    14 Chuvash + Finnish + North_Dutch + Tabassaran @ 8.873098
    15 Mari + Norwegian + Southwest_Finnish + Tabassaran @ 8.914038
    16 Chuvash + La_Brana-1 + Norwegian + Tabassaran @ 8.927127
    17 Chuvash + Estonian + Irish + Tabassaran @ 8.933358
    18 Chuvash + North_Swedish + Swedish + Tabassaran @ 8.936755
    19 Finnish + Mari + Swedish + Tabassaran @ 8.947768
    20 Mari + Southwest_Finnish + Swedish + Tabassaran @ 8.947787


    mtDNA: C4b1 T146C G8251A
    1. AF346991(Khirgiz) Ingman
    2. AY195772.2(Asian) Mishmar
    3. EU007879(Mansi) Ingman_gyll
    4. EU482304(Yukaghir) Volodko
    5. FJ951476(Buryat) Derenko
    6. FJ951477(Buryat) Derenko
    7. FJ951559(Severo-Evensk) Derenko
    8. FJ951568(Evenk) Derenko
    9. FJ951617(Yakut) Derenko
    10. FJ951618(Yakut) Derenko
    11. MF522849(Pamir) Peng
    12. MF522850(Pamir) Peng
    13. MF522861(Pamir) Peng
    14. MF522888(Pamir) Peng
    15. MF523094(Pamir) Peng
    16. KU682913(Uyghur) Zheng
    17. KU683198(Uyghur) Zheng
    18. KU683293(Uyghur) Zheng

    ReplyDelete
  142. From a Russian source, quote:

    Hello Everyone,

    There is more accurate (?) Information about the place of collection of samples. Before the publication of the article, it is difficult to say 100%.

    DA57 (Saka, Mount Khan Tengri): J-Y13534
    DA58 (Saka, Mount Khan Tengri): J-Z7706
    DA59 (Saka, Mount Khan Tengri): J-Y13534

    DA203 (Karakhanids, Central Asia): J-Z7706
    DA204 (Karakhanids, Central Asia): J-PF4610

    DA222 (Karluki, Mount Khan Tengri): J-Z7706
    DA230 (Karluki, Mount Khan Tengri): J-Z7706

    Khan-Tengri is a mountain that enters the Tien-Shan mountain system. It is located at the junction of the borders of Kazakhstan, Kyrgyzstan and China.

    ReplyDelete
  143. My guess is these new Asian Scythian genomes will be much more Sintashta-derived than the current ones. And Scythian scores for Turks in nMonte scores will go way down.

    ReplyDelete
  144. @David

    DA95: Y: N-Y16220 mtDNA D4b1 Gedmatch Z423957

    Y-DNA N-Z16620 is found among Buryat Mongols, Turks, and Poles.

    1240k SNPs in 23andMe format:

    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA95/genome_DA95-1240k.zip
    SNPs read: 902819 percent covered: 75.33%

    K13 shows that he's basically a mix of Siberian and East Asian.

    He's clearly a Tuvan or a nearby Buryat, not a Uralic speaker, even though he's within Y N1026, which includes many Finns, Estonians, Scandinavians (Saami?), Russians, and Belarussians:

    K13:
    North_Atlantic -
    Baltic 9.96
    West_Med -
    West_Asian 5.69
    East_Med -
    Red_Sea -
    South_Asian -
    East_Asian 21.44
    Siberian 60.84
    Amerindian 2.08
    Oceanian -
    Northeast_African -
    Sub-Saharan -

    K13 Oracle4:

    Using 1 population approximation:
    1 Tuvinian @ 6.381505
    2 Buryat @ 9.279637
    3 Altaian @ 13.595392
    4 Oroqen @ 15.390157
    5 Yakut @ 16.802637
    6 Mongolian @ 17.218132
    7 Hakas @ 18.956720
    8 Dolgan @ 19.880583
    9 Koryak @ 21.080681
    10 Evenki @ 22.139490
    11 Ket @ 23.287222
    12 Shors @ 23.305866
    13 Evens @ 24.651129
    14 Kirgiz @ 25.549837
    15 Selkup @ 26.074865
    16 Kazakh @ 27.028515
    17 Chukchi @ 27.287500
    18 Hezhen @ 40.628674
    19 Xibo @ 41.691662
    20 Hazara @ 42.158020

    Using 2 populations approximation:
    1 50% Hakas +50% Oroqen @ 4.689852


    Using 3 populations approximation:
    1 50% Buryat +25% Buryat +25% Selkup @ 3.439749


    Using 4 populations approximation:
    +++++++++++++++++++++++++++++++++++++++++
    1 Buryat + Buryat + Buryat + Selkup @ 3.439749
    2 Buryat + Mongolian + Oroqen + Selkup @ 3.620940
    3 Buryat + Buryat + Oroqen + Selkup @ 3.633861
    4 Buryat + Oroqen + Selkup + Tuvinian @ 3.808012
    5 Buryat + Evenki + Hakas + Tuvinian @ 3.968436
    6 Buryat + Buryat + Evenki + Hakas @ 4.068716
    7 Altaian + Buryat + Oroqen + Selkup @ 4.073336
    8 Kirgiz + Oroqen + Oroqen + Selkup @ 4.079275
    9 Buryat + Buryat + Dolgan + Hakas @ 4.082941
    10 Mongolian + Oroqen + Selkup + Tuvinian @ 4.093056
    11 Buryat + Buryat + Selkup + Tuvinian @ 4.123865
    12 Buryat + Ket + Oroqen + Tuvinian @ 4.180398
    13 Buryat + Buryat + Buryat + Ket @ 4.184972
    14 Mongolian + Oroqen + Oroqen + Selkup @ 4.215206
    15 Buryat + Buryat + Ket + Oroqen @ 4.220430
    16 Altaian + Evenki + Hakas + Oroqen @ 4.241297
    17 Dolgan + Hakas + Mongolian + Oroqen @ 4.242123
    18 Buryat + Mongolian + Selkup + Yakut @ 4.251205
    19 Buryat + Evens + Hakas + Tuvinian @ 4.289215
    20 Altaian + Hakas + Oroqen + Yakut @ 4.330460

    ReplyDelete
  145. @David

    DA81 has an interesting Y and mtDNA combination:

    DA81: Y: R-PH200 mtDNA: A16 Gedmatch: Z855800

    1240k SNPs in 23andMe format:
    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA81/genome_DA81-1240k.zip
    SNPs read: 934047 percent covered: 77.94%

    The quite rare mtDNA A16 is found among Uyghurs, a single Turk, and in Siberia.
    However, we've seen R-PH200 in the "Gepid" from Serbia, who is almost certainly a Hun.

    His K15 Oracle4 results have an interesting combination of Turkic people, Central Asians, and in particular a Hunnic Chuvash mixed with Uralic Hungarians and other Uralic people. No trace of East Asian.

    He seems to be either a Hun or a pre-Hungarian steppe Magyar:

    K13 Oracle4:

    Using 1 population approximation:
    1 Tatar @ 20.448389
    2 Nogay @ 26.107395
    3 Tadjik @ 26.292097
    4 Moldavian @ 28.399504
    5 Afghan_Turkmen @ 28.532135
    6 Chuvash @ 28.803125
    7 Hungarian @ 29.209324
    8 Croatian @ 30.250839
    9 Serbian @ 30.661474
    10 Tabassaran @ 30.859512
    11 Austrian @ 31.006783
    12 Ukrainian_Lviv @ 31.031282
    13 Romanian @ 31.324175
    14 East_German @ 31.537865
    15 Afghan_Tadjik @ 31.558739
    16 South_Polish @ 31.916679
    17 Mari @ 32.007210
    18 Uzbeki @ 32.136360
    19 Ukrainian_Belgorod @ 32.536568
    20 Aghan_Hazara @ 32.628681

    Using 2 populations approximation:
    1 50% Southwest_Finnish +50% Tadjik @ 10.932587

    Using 3 populations approximation:
    1 50% Afghan_Pashtun +25% Mari +25% Norwegian @ 10.291666

    Using 4 populations approximation:
    1 Mari + Norwegian + Tabassaran + Tadjik @ 8.085715
    2 Mari + Tabassaran + Tadjik + West_Scottish @ 8.233762
    3 Irish + Mari + Tabassaran + Tadjik @ 8.237353
    4 Mari + Swedish + Tabassaran + Tadjik @ 8.286062
    5 Mari + Orcadian + Tabassaran + Tadjik @ 8.487821
    6 Mari + North_Dutch + Tabassaran + Tadjik @ 8.562583
    7 Afghan_Pashtun + Mari + Norwegian + Tabassaran @ 8.564947
    8 Danish + Mari + Tabassaran + Tadjik @ 8.584302
    9 Afghan_Pashtun + Mari + Tabassaran + West_Scottish @ 8.709785
    10 Mari + Norwegian + Tadjik + Tadjik @ 8.729266
    11 Afghan_Pashtun + Irish + Mari + Tabassaran @ 8.769035
    12 Mari + North_Swedish + Tabassaran + Tadjik @ 8.783600
    13 Afghan_Pashtun + Mari + Swedish + Tabassaran @ 8.800719
    14 Kalash + Mari + Norwegian + Tabassaran @ 8.809497
    15 Chuvash + Norwegian + Tadjik + Tadjik @ 8.810736
    16 Kalash + Mari + Tabassaran + West_Scottish @ 8.920215
    17 Afghan_Pashtun + Mari + Orcadian + Tabassaran @ 8.924979
    18 Chuvash + Norwegian + Tabassaran + Tadjik @ 8.933625
    19 Mari + North_German + Tabassaran + Tadjik @ 8.947791
    20 Mari + Southwest_English + Tabassaran + Tadjik @ 8.949721

    ReplyDelete
  146. @OpenGenomes,
    "The quite rare mtDNA A16 is found among Uyghurs, a single Turk, and in Siberia.
    However, we've seen R-PH200 in the "Gepid" from Serbia, who is almost certainly a Hun."

    For the Gepid from Serbia. I got a good for him being 25% Bavaria Medieval, 25% Balkan IA, 50% Asian (forget exactly what type it was, maybe my model had some Sintashta in it). So, I think he was of mixed Hunnic and European origin (Balkan and German) not simply a Hun. But time will tell.

    ReplyDelete
  147. @David
    It doesn't fall apart because the Kashkarchi_BA samples from the Ferghana Valley are identical to Sintashta and belong to R1a-Z645.

    I don't think you're realizing that I've always been arguing for R1a traveling down what's now being called the IAMC.

    Now, is the Ferghana Valley satisfactory enough? Yeah, maybe.

    ReplyDelete
  148. DA162 is another relatively high-coverage genome.

    DA162: Y: Q-YP4000 mtDNA H13a2 Gedmatch: Z285172

    Q-YP4000 is a basal Q1a clade found in the Caucasus.

    1240k SNPs in 23andMe format:
    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA162/genome_DA162-1240k.zip
    SNPs read: 929425 percent covered: 77.55%

    K13 Oracle shows he's clearly a North Caucasian, either a Chechen or an Abkhazian. The extinct Ubykhs were horsemen who lived on the steppes in what is today Krasnodar Province, Russia. North Caucasian horsemen were sometimes also referred to as "Scythians" by the ancient Greeks. However, he doesn't seem to have all that much in the way of steppe Yamnaya ancestry, in spite of living close to the Yamnaya region.

    ReplyDelete
  149. @Open Genomes and re:"

    "DA95: Y: N-Y16220 mtDNA D4b1 Gedmatch Z423957

    Y-DNA N-Z16620 is found among Buryat Mongols, Turks, and Poles.

    He's clearly a Tuvan or a nearby Buryat, not a Uralic speaker, even though he's within Y N1026, which includes many Finns, Estonians, Scandinavians (Saami?), Russians, and Belarussians."

    Turko-Mongol etc. N-Z16620 is actually a rather close brother clade of N-L1025, both lineages descend from N-Y6058.

    DA171, Y N-L1025 mtDNA H2a1

    1 Lithuanian + Lithuanian + Lithuanian + Lithuanian @ 11.114440
    2 East_Finnish + Lithuanian + Lithuanian + Lithuanian @ 11.315384
    3 Erzya + Lithuanian + Lithuanian + Lithuanian @ 11.508182
    4 Estonian + Lithuanian + Lithuanian + Lithuanian @ 11.761994
    5 Kargopol_Russian + Lithuanian + Lithuanian + Lithuanian @ 11.880618

    Quite a difference, both very much related N-lineages apparently still coming from the Bronze Age Ural area.

    Many thanks for your efforts, enjoyable reading.

    ReplyDelete
  150. @Open Genomes & Huck Finn

    Y-DNA N-Z16620 is found among Buryat Mongols, Turks, and Poles.

    N-Z16620 isn't really found in Poles, unless you mean one or two Poles out of millions, so I'm not sure what you guys are discussing?

    ReplyDelete
  151. @Davidski: There's N-Y16220 in Poland, at least according to Yfull. Maybe there's just one and he happened to take the test?

    ReplyDelete
  152. @Huck Finn

    There's N-Y16220 in Poland, at least according to Yfull. Maybe there's just one and he happened to take the test?

    He had a highly unusual Y-hg for a Pole and that's probably why he ended up doing a full Y-chromosome scan.

    But I don't really know his ethnic background. Do you?

    ReplyDelete
  153. @David

    Maybe the N-Y16220 in YFull is a Pole of Tatar descent?

    ReplyDelete
  154. @Open Genomes

    Maybe the N-Y16220 in YFull is a Pole of Tatar descent?

    That was sort of my point.

    It's possible, and probably likely, that this person or one of his recent paternal ancestors was a Lipka Tatar.

    ReplyDelete
  155. @D and re: "He had a highly unusual Y-hg for a Pole and that's probably why he ended up doing a full Y-chromosome scan."

    Usually people know their Y-hg only after the test, but it is of course possible that he descends from Lithuanian Tatars or something. Then again, maybe he's just a Pole. Even Polish kings did have paternal N too, if I recall it right. It happens.

    ReplyDelete
  156. @David

    DA198 is a very different genome from all the others in the study. He's somewhat low coverage, so he may not be suitable for Global25, but it's worth a try, as you'll see.

    DA198: Y: G-PF3378 mtDNA: H2a2a Gedmatch: Z255267

    He has a signature Anatolian and European Neolithic Y, and a signature Steppe mtDNA.

    1240k SNPs in 23andMe format:
    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA198/genome_DA198-1240k.zip
    SNPs read: 496911 percent covered: 41.46%

    He appears to be a straight 50-50 European Neolithic Farmer and Steppe Poltavka mix.
    Poltavka of course is a direct continuation of Yamnaya. He's different from all the other Neolthic-Steppe mixes because he has a Neolithic Y and Steppe mtDNA, the exact opposite of almost all the others.

    Where did this happen? At the end of Cucuteni-Trypillia? The Baden Culture, which seems to have been totally Neolithic, but was surrounded by Corded Ware?
    A bride exchange with a patrilineal and warlike society would indicate peaceful trade relations, not conquest. Perhaps this happened just before the full-scale Steppe incursion into Neolithic Europe, at the very edge of the steppe.

    ReplyDelete
  157. @East Pole

    Sorry for late reply. I think you didn't quite understand what I wrote about Indo-Iranian language in the steppe, so I'll try again.

    In Global 25, the closest samples to Western Scythians/Sarmatians are these:

    Taldysay_MLBA2
    Karasuk
    Kyzlbulak_MLBA2
    ...
    Tajik

    Taldysay_MLBA2 is from Central Kazakhstan, c. 1500 BCE:

    Taldysay_MLBA2
    Sintashta_MLBA 52.2%
    Dali_EBA 17.5%
    Geoksiur_Eneolithic 13.2%
    Hajji_Firuz_ChL 9.8%
    Naxi 5.5%
    Gonur1_BA:I2085 1.8%
    West_Siberia_N 0%
    Sarazm_Eneolithic:I4910 0%

    Distance 3.0921%

    Kyzlbulak_MLBA2 is from SE Kazakhstan, also c. 1500 BCE:

    Kyzlbulak_MLBA2
    Sintashta_MLBA 45.1%
    Dali_EBA 30.3%
    Gonur1_BA:I2085 19.6%
    Geoksiur_Eneolithic 5%
    West_Siberia_N 0%
    Sarazm_Eneolithic:I4910 0%
    Hajji_Firuz_ChL 0%
    Naxi 0%

    Distance 2.2008%

    And these are the Sarmantian samples:

    Sarmatian_Pokrovka
    Sintashta_MLBA 65.7%
    Dali_EBA 12.2%
    Gonur1_BA:I2085 11.3%
    Naxi 6.6%
    Geoksiur_Eneolithic 4.2%
    West_Siberia_N 0%
    Sarazm_Eneolithic:I4910 0%
    Hajji_Firuz_ChL 0%

    Distance 2.0308%

    So we have a clear idea of where did Western Scythians come from and we have samples from their ancestors (and closest modern descendants). We also know that they spoke a language from that area of origin (around Bactria-Sogdiana), for which we have scarce evidence, but it's usually classified as East Iranian. But we know for the split timing that this was a very early form of East Iranian, very likely pre-Avestan and very close to the split with IA, and therefor very close to Proto-Indo-Iranian. All this is corroborated by the evidence we have from the steppe regarding Indo-Iranian substrate (place names, rivers,...).

    And this early Indo-Iranian language was spoken on the steppe (all the way to Ukraine) for maybe 1000 years.

    So there is no mystery as to why Balto-Slavic shares may common features with Indo-Iranian, including early Indo-Aryan. And there's no mystery as to why Uralic borrowings come from this language.

    Instead, the classical Kurgan model cannot explain anymore those similarities, because it has already been disproved by ancient DNA. You can't have Balto-Slavic and Indo-Iranian splitting c.2800 BCE in the CWC area and then Celto-Germanic splitting later from Balto-Slavic (or any similar scenario). That would place Balto-Slavic as very close to Celtic and Indo-Iranian as a very divergent branch from PIE and from Balto-Slavic.

    Yamnaya could never have been pre-Indo-Iranian either, as argued. And there is no language continuum from 4500 to 1000 CE (and later) in the steppe.

    So it's better to go with an explanation that is based on known facts than to go for old speculative ones that are basically disproved. Even if you don't like it as much as the old ones.

    ReplyDelete
  158. Re: the non-EHG side of Yamnaya, has someone tried with the new samples? I was starting with the model that was the best in the qpAdm ones posted by Davisdki here.

    http://eurogenes.blogspot.com/2018/01/another-look-at-genetic-structure-of.html

    Which was this one:

    https://drive.google.com/file/d/1w6sMd5mh7UybEV6LirFG8PZWI7FJkUkD/view

    Global 25 does not reproduce it exactly, though:

    Yamnaya_Samara
    EHG 58.8%
    CHG 34.4%
    Koros_EN 6.8%
    Koros_HG 0%

    Distance 5.7039%

    https://imgur.com/a/MkjS4MM

    The only samples that seem to work are Ukraine_Mesolithic, Geoksiur_Eneolithic and some Dali_EBA. Not sure if anything of it can be reproduced with qpAdm (using the 4 samples with larger contribution: EHG, Ukraine_Mesolithic, CHG, Geoksiur_Eneolithic).

    ReplyDelete
  159. DA221 is a very high coverage genome, with a very interesting ancestral combination:

    DA221: Q-L332 mtDNA: J1b1a1 Gedmatch: Z697149

    Y Q-L332 under Q-L330 is just above mostly Native American Q-M1107. It seems likely that Q-L330 is of Yeniseian / Ket origin, which explains its proximity to the Native American Q clades. One individual in Q-L332 is a Hungarian of Jasz origin. The Jasz descend from the Iaxyges, the "Royal Scythians".

    mtDNA J1b1a1 is a typical European Neolithic mtDNA haplogroup.

    1240k SNPs in 23andMe format:
    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA221/genome_DA221-1240k.zip
    SNPs read: 1042001 percent covered: 86.95%

    DA221's K15 shows him to be 50% Uralic, 25% Paleo-Siberian, and 25% Baloch, Brahui, or Pashtun. (Indus Valley, or Iranian Neolithic?)

    A most unusual combination.

    It's clear that he's a Scythian, but a mostly Uralic and Siberian Scythian, with perhaps some BMAC ancestry. He's not that close to the Mansi ancestors of the Magyars but rather to the Finns. A Scythian who lived in the forest zone of Central and Eastern Siberia. Did he get his mtDNA J1b1a1 from Corded Ware admixture with the European Neolithic, before Sintashta moved eastward? The patrilineal Scythians in this case seem to have incorporated males from the Uralic peoples. Or did some Uralic with Yeniseian ancestry people besides the Magyars adopt the horse-nomadic lifestyle of the Scythians?

    K13 Oracle4:

    Using 1 population approximation:
    1 Tatar @ 14.583388
    2 Chuvash @ 21.061611
    3 Mari @ 22.842422
    4 Afghan_Turkmen @ 23.346781
    5 Nogay @ 26.150249
    6 Uzbeki @ 26.662830
    7 Aghan_Hazara @ 29.126190
    8 Moldavian @ 29.428646
    9 Tadjik @ 30.084274
    10 Uygur @ 30.553617
    11 Kargopol_Russian @ 30.629723
    12 Erzya @ 30.849771
    13 Hungarian @ 31.214993
    14 Ukrainian_Lviv @ 31.307230
    15 Croatian @ 31.355104
    16 Shors @ 31.658257
    17 Hazara @ 31.841177
    18 Serbian @ 32.484200
    19 Ukrainian_Belgorod @ 32.491692
    20 Ukrainian @ 32.657677

    Using 2 populations approximation:
    1 50% Mari +50% Tadjik @ 10.739988

    Using 3 populations approximation:
    1 50% Finnish +25% Koryak +25% Makrani @ 7.131532

    Using 4 populations approximation:
    1 Balochi + Finnish + Finnish + Koryak @ 5.623512
    2 Brahui + Finnish + Finnish + Koryak @ 5.826793
    3 Balochi + Chukchi + Finnish + Finnish @ 5.827025
    4 Balochi + Finnish + Koryak + Southwest_Finnish @ 5.837285
    5 Balochi + Estonian + Finnish + Koryak @ 5.870986
    6 Afghan_Pashtun + Mari + Shors + Swedish @ 5.972068
    7 Brahui + Chukchi + Finnish + Finnish @ 6.004530
    8 Brahui + Finnish + Koryak + Southwest_Finnish @ 6.015066
    9 Brahui + Estonian + Finnish + Koryak @ 6.028364
    10 Balochi + Estonian + Koryak + Southwest_Finnish @ 6.028923
    11 Afghan_Pashtun + Mari + Norwegian + Shors @ 6.060086
    12 Balochi + Finnish + Koryak + La_Brana-1 @ 6.091079
    13 Balochi + Chuvash + Ket + Norwegian @ 6.095906
    14 Balochi + Estonian + Koryak + La_Brana-1 @ 6.111097
    15 Afghan_Pashtun + Chuvash + North_Swedish + Shors @ 6.153909
    16 Brahui + Estonian + Koryak + Southwest_Finnish @ 6.157790
    17 Balochi + Chukchi + Finnish + Southwest_Finnish @ 6.159769
    18 Balochi + East_Finnish + Koryak + Southwest_Finnish @ 6.161624
    19 Afghan_Pashtun + Mari + North_Swedish + Shors @ 6.165341
    20 Balochi + Koryak + Lithuanian + North_Swedish @ 6.171128

    ReplyDelete
  160. @Alberto said "So it's better to go with an explanation that is based on known facts than to go for old speculative ones that are basically disproved. Even if you don't like it as much as the old ones."

    Can you explain your model of late PIE migrations?

    ReplyDelete
  161. And now the Armenian_ChL samples have become really interesting too.

    Armenia_ChL
    Hajji_Firuz_ChL 38.4%
    Balkans_ChL 32.3%
    Sarazm_Eneolithic:I4290 14.5%
    EHG 8.2%
    CHG 6.6%
    Ukraine_Mesolithic:I1763 0%
    Ukraine_N 0%
    Koros_HG 0%
    West_Siberia_N 0%
    Dali_EBA 0%
    Geoksiur_Eneolithic:S8510.E1.L1 0%
    Tepe_Anau_Eneolithic 0%
    Tepe_Hissar_ChL 0%
    Seh Gabi_ChL 0%
    Barcin_N 0%
    Koros_EN 0%

    Distance 2.6179%

    Like something that connects the Balkans, through the north Pontic region to the North Caucasus, with SC_Asia through Iran to the South Caucasus. The samples can be seen having ANE (AfontovaGora3 in addition to EHG when using distal sources) so it's normal that it picks Sarazam_Eneolithic (plus it's Y-DNA is L1a). But it clearly has a European side too. These a key connections when we thing about the date of the samples (4300-4000 BCE).

    ReplyDelete
  162. @Chetan

    Can you explain your model of late PIE migrations?

    I don't have any specific model. I just look at what is likely and unlikely depending on what the data shows.

    At this point is not clear at all how Late PIE spread (I guess by this you mean PIE excluding Anatolian and maybe Tocharian?). At least for the European IE languages, I can speculate that the most likely is that they spread from the Balkans, but the data is scarce at present, so I'm not going to argue strongly for it.

    ReplyDelete
  163. @Alberto Fair enough. As of now, the data supports the steppe model for Post-Anatolian IE spread, I think

    ReplyDelete
  164. @Chetan

    Or probably you were asking more specifically regarding Indo-Iranian? How did it get to SC Asia in the first place?

    Well, again, that's difficult to say. From a genetic point of view all the possibilities are open, since there is a native substrate and then admixture from the west, north and south. From north and south is more limited, but it's there.

    Looking at the cultural transmissions, it's also unclear. A mix of "native" (pre-Bronze age) traditions, with influence from the west, more subtle from the south and mostly absent from the north (the steppe people are the ones who borrowed the culture from BMAC, but I guess it's still possible that they brought the language, even if rather unlikely).

    So I'l leave to make their own conclusions about it till the data leaves less room for speculation.

    ReplyDelete
  165. @Alberto

    Yamnaya can't be modeled with formal stats as partly of ancient Central Asian origin. I've looked at this very closely.

    And no, the latest data hasn't debunked the classical Kurgan model. But of course I don't expect you to be reasonable about this and admit it.

    ReplyDelete
  166. @Alberto Archaeologically, Andronovo is the only culture that links the northers steppes with South-Central Asia in the stipulated time-period of Proto- Indo Iranian. We have to take that into account too.

    But I agree the absence of L657 is a little concerning. Here, we have close to 50 new samples from the steppe and not a single one is L657! It is advisable to wait for the corrected version of Narasimhan's data sheets. I am waiting for at least a couple of R1as in it to turn out L657.

    ReplyDelete
  167. @Davidski

    Thanks for confirming that. So still no better CHG-like than Kotias...

    @Chetan

    Even if L657 turns up (which is perfectly possible), it will still represent 1% fo the Y-DNA of the Andronovo people. Which would mean that its high incidence in India is unrelated to any mass migration from the steppe.

    ReplyDelete
  168. "Even if L657 turns up (which is perfectly possible), it will still represent 1% for the Y-DNA of the Andronovo people. "

    There could have been an early L657 rich group in Turkmenistan/Uzbekistan. There were such groups (I gave a ref above), but they were not in the list of samples for this study.

    The dominance of L657 in India today is because of Iron Age founder effects probably.

    ReplyDelete
  169. Here's a total shocker for one of the "coolest" pre-Sintashta Balto-Slavic-Indo-Iranians ... if you think the Manchus are "totally cool!"

    DA39: Y: R-L645 mtDNA: N9a2'4'5'11 Gedmatch: Z780597

    1240k SNPs in 23andMe format:
    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA39/genome_DA39_1240k.zip
    SNPs read: 927452 percent covered: 77.39%

    Yes, R-L645 is the immediate parent of R-Z93. DA39 has no reads for the two R-Z93 SNPs, bue's not either R-Z283 or R-Z94, and is ancestral for most of the R-Z93 subclades.

    His mtDNA is N9a2'4'5'11, which is found in Japan and China.

    What's really unbelievable is his K15 Oracle4 results:

    He's a Jurchen, from the ancestors of the Manchus!

    The Jurchen were Tungusic speakers who lived in Eastern Manchuria. Or even more, something between a Jurchen and a Sibe (Xibo), a Tungusic people who live right next to North Korea. The Jurchen themselves were ruled for centuries (698-926) by the Korean Kingdom of Balhae

    Y'all do know that there's plenty of "R1a1*" all over Northern China, among Han Chinese, right?
    Zhong et al. (2010) R1a-M17 in China

    Basically, if these were Proto Balto-Slavic Indo-Iranians descended from the Corded Ware, they just got on their horses and rode east till they hit the Pacific Ocean! :D

    If this sounds unbelievable, take a look at these K15 Oracle4 results, and see for yourself. Aren't the samurai very cool? :D

    Using 1 population approximation:
    1 Hezhen @ 7.253376
    2 Xibo @ 8.164757
    3 Tu @ 19.758152
    4 Japanese @ 20.678120
    5 Mongolian @ 21.221468
    6 Naxi @ 27.574614
    7 Buryat @ 28.152800
    8 Yizu @ 29.415333
    9 Kirgiz @ 30.413364
    10 Tuvinian @ 30.716511
    11 Altaian @ 31.630821
    12 Kazakh @ 33.500027
    13 Oroqen @ 34.569004
    14 Tibeto-Burman_Burmese @ 37.497547
    15 Hakas @ 38.107147
    16 Tujia @ 41.443615
    17 Shors @ 42.083508
    18 Uygur @ 42.207119
    19 Hazara @ 43.875549
    20 Miaozu @ 44.124619

    Using 2 populations approximation:
    1 50% She +50% Yakut @ 4.282229

    Using 3 populations approximation:
    1 50% Japanese +25% Xibo +25% Yakut @ 4.011470

    Using 4 populations approximation:
    1 Oroqen + Tu + Xibo + Xibo @ 3.701355
    2 Hezhen + Oroqen + Tu + Xibo @ 3.869050
    3 Hezhen + Mongolian + Oroqen + She @ 3.884272
    4 Evens + Tujia + Xibo + Xibo @ 3.911974
    5 Mongolian + Oroqen + She + Xibo @ 3.954192
    6 Hezhen + Miaozu + Mongolian + Oroqen @ 3.977690
    7 Mongolian + Oroqen + Tujia + Xibo @ 3.991471
    8 Japanese + Tu + Xibo + Yakut @ 4.002586
    9 Hezhen + Mongolian + Oroqen + Tujia @ 4.011113
    10 Japanese + Japanese + Xibo + Yakut @ 4.011470
    11 Evens + Miaozu + Xibo + Xibo @ 4.014785
    12 Miaozu + Mongolian + Oroqen + Xibo @ 4.028243
    13 Japanese + Oroqen + Xibo + Xibo @ 4.030292
    14 Evenki + Tujia + Xibo + Xibo @ 4.059318
    15 Hezhen + Japanese + Tu + Yakut @ 4.059593
    16 Evens + Hezhen + Tujia + Xibo @ 4.073646
    17 Evens + Hezhen + Miaozu + Xibo @ 4.088564
    18 Buryat + Japanese + Oroqen + Tujia @ 4.088934
    19 Xibo + Xibo + Yakut + Yizu @ 4.095832
    20 Hezhen + Hezhen + Oroqen + Tu @ 4.120090

    ReplyDelete
  170. DA39: Y: R-L645 (Z280- Z94-) mtDNA: N9a2'4'5'11 Gedmatch: Z780597

    Using 1 population approximation:
    1 Hezhen @ 7.253376
    2 Xibo @ 8.164757


    Using 3 populations approximation:
    1 50% Japanese +25% Xibo +25% Yakut @ 4.011470

    Ancient R1a-L645* Corded Ware Battle Song

    ReplyDelete
  171. @ Open Genomes
    See "Tocharians"
    China Xiaohe, Xinjiang [106, 111, 115, 120, 121, 136, 139] 2515 ± 43 BC M R1a1a M89+, M9+, M45+, M173+, M198+, Z93-

    ReplyDelete
  172. @Supernord
    Yes, but the "Tocharians" (or Afanasievo?) were 2,150 miles (3,450 km) west of where the Jurchen lived. On the other hand, they had about 4000 years to go those 2150 miles ...

    He could be this R-KMS149 which is found in the Altai, but he's negative for both KMS133 and all the R-Y20787 equivalents.

    ReplyDelete
  173. Interesting. Andronovo Indo-Iranians did have a strong cultural influence on the later steppe tribes from East Asia. Mongols even harbor quite a bit of West Eurasian ancestry (from the latest study Genetic structure, divergence and admixture of Han Chinese, Japanese and Korean populations

    But I don't think the ancestors of Indo-Aryans came from as far east as Manchuria.They probably migrated south from the Altai region through the IAMC, very early on. But this gives you the idea how far some of the bearers of basal Z93 clades went. There is a hotspot of basal Z93 in that region.

    ReplyDelete
  174. @Nirjhar

    "What is your opinion on sites like Bustan and the data we have from there?."

    Asking about anything in particular? I don't have much to say to be honest. Nothing is new or unexpected except the lack of R1a in India, which is kinda disappointing.

    ReplyDelete
  175. @David

    DA27: R-Z93* mtDNA: C4b1 Gedmatch: Z419959

    1240k SNPs in 23andMe format:
    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA27/genome_DA27-1240k.zip
    SNPs read: 919361 percent covered: 76.71%

    DA27 is clearly a R-Z93* (Z94-) Mongolian. The Altai is one place that we know R-Z938 is found.
    R-YP1518 in the Altai

    K13 Oracle4:
    Using 1 population approximation:
    1 Mongolian @ 8.388174
    2 Buryat @ 10.953264
    3 Tuvinian @ 12.499017
    4 Altaian @ 16.284502
    5 Oroqen @ 19.40483

    ReplyDelete
  176. Anthro Survey: Yeah, I'm getting substantial Gonur-like stuff in almost every single monte run for Brahmins and Kshatriya. Glad to see your formal models confirm this!

    the thing is, there tends to be a correlation between high caste ancestry and Gonur shift. So, even if 8% doesn't seem like much overall, it's quite a bit when you consider it arriving in tandem with the 18% Sintashta-like

    If this works out, may be linked to R1a, and absence of R1a in Swat, if that continues to be the case after further reanalysis of the y on the samples.

    Two way ANI+ASI approximation something more like Indo-Aryan+ANI+ASI (where ANI=simple low level of Inner Asian Mountain Corridor Steppe_MLBA+Indus Periphery, and not any particular dynamic favouring intrusive y, while Indo-Aryan = higher level of Steppe_MLBA+BMAC+Indus_Periphery), where North Indian Brahmins enriched for Indo-Aryan? Only Indo-Aryan associated with entry of R1a, and not whatever Steppe_MLBA related ancestry in the Swat samples?

    In either case, seems like BMAC related ancestry would shift vectors away from using Steppe_MLBA_West for rest of ancestry because of more Anatolian related input in BMAC than InPe?

    ReplyDelete
  177. @ Open Genomes

    The Tarim mummies do not belong to Afanasievo culture.

    ReplyDelete
  178. @Open Genomes,
    "DA221....
    mtDNA J1b1a1 is a typical European Neolithic mtDNA haplogroup. "

    J1b1a1 is actually a Steppe lineage.

    ReplyDelete
  179. DA38 is a high coverage female genome.

    DA38: mtDNA D4b2b4 Gedmatch: Z777394

    1240k SNPs in 23andMe format:
    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA38/genome_DA38-1240k.zip
    SNPs read: 1008889 percent covered: 84.18%

    mtDNA haplogroup D4b2b4 is quite rare, and found in Russia and China.

    In K13 Oracle4, she has an interesting mix of Turkic, South Asian, Uralic, and Far East Paleo-Siberian. This is another puzzling combination:

    Using 1 population approximation:
    1 Afghan_Turkmen @ 13.177837
    2 Uzbeki @ 14.692655
    3 Aghan_Hazara @ 17.337414
    4 Hazara @ 18.874575
    5 Uygur @ 19.402395
    6 Nogay @ 21.575409
    7 Kazakh @ 23.264153
    8 Shors @ 23.582987
    9 Tatar @ 24.686546
    10 Tadjik @ 25.743828
    11 Afghan_Tadjik @ 26.193800
    12 Kirgiz @ 26.624533
    13 Hakas @ 27.532385
    14 Mari @ 29.527866
    15 Chuvash @ 29.548113
    16 Turkmen @ 30.775146
    17 Altaian @ 33.711174
    18 Afghan_Pashtun @ 35.069805
    19 MA-1 @ 35.625000
    20 Kabardin @ 37.343269

    Using 2 populations approximation:
    1 50% Shors +50% Tadjik @ 8.471065

    Using 3 populations approximation:
    1 50% Altaian +25% Balochi +25% Estonian @ 6.505335

    Using 4 populations approximation:
    1 Balochi + Chukchi + East_Finnish + Uygur @ 4.344760
    2 Balochi + Estonian + Koryak + Uygur @ 4.398308
    3 Balochi + East_Finnish + Koryak + Uygur @ 4.522153
    4 Balochi + Finnish + Koryak + Uygur @ 4.570068
    5 Chukchi + Finnish + Hazara + Kalash @ 4.580729
    6 Balochi + Chukchi + Finnish + Uygur @ 4.586906
    7 Balochi + Koryak + Lithuanian + Uygur @ 4.645467
    8 Brahui + Estonian + Koryak + Uygur @ 4.658649
    9 Brahui + Chukchi + East_Finnish + Uygur @ 4.670489
    10 Balochi + Chukchi + Estonian + Uygur @ 4.697381
    11 Finnish + Kalash + Koryak + Uygur @ 4.734208
    12 Estonian + Hazara + Kalash + Koryak @ 4.747745
    13 Finnish + Hazara + Kalash + Koryak @ 4.751423
    14 Estonian + Kalash + Koryak + Uygur @ 4.784140
    15 Balochi + Chukchi + Finnish + Hazara @ 4.801906
    16 Chukchi + Finnish + Kalash + Uygur @ 4.803104
    17 Balochi + Estonian + Hazara + Koryak @ 4.804730
    18 Aghan_Hazara + Finnish + Kalash + Koryak @ 4.827250
    19 Chukchi + East_Finnish + Kalash + Uygur @ 4.834157
    20 Balochi + Chukchi + Estonian + Hazara @ 4.853336

    ReplyDelete
  180. @David

    DA223 is a high-coverage genome with another unusual combination of ancestries, a bit different from the others.

    DA223: Y: R-Z93* mtDNA: J1c5a Gedmatch: Z698544

    1240k SNPs in Gedmatch format:
    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA223/genome_DA223-1240k.zip
    SNPs read: 1025678 percent covered: 85.58%

    mtDNA J1c5a is European.

    However, his K13 Oracle4 shows a combination of Central / South Asian and Uralic. He has less of a "Turkic" mix than than the others with such a combination.

    He's yet another Scythian, but with more Yamnaya / Poltavka input than the others.

    Using 1 population approximation:
    1 Tadjik @ 20.538246
    2 Tabassaran @ 22.110958
    3 Nogay @ 23.390511
    4 Chechen @ 24.650429
    5 Lezgin @ 25.197790
    6 Afghan_Tadjik @ 26.141256
    7 Afghan_Pashtun @ 26.392275
    8 Kabardin @ 26.561010
    9 Tatar @ 27.386696
    10 Kumyk @ 28.199272
    11 Balkar @ 29.258989
    12 Afghan_Turkmen @ 29.359095
    13 North_Ossetian @ 29.912498
    14 Adygei @ 30.521013
    15 Turkmen @ 30.732210
    16 Ossetian @ 31.926556
    17 Aghan_Hazara @ 32.465721
    18 Uzbeki @ 32.555702
    19 Kalash @ 33.689545
    20 Chuvash @ 34.201897

    Using 2 populations approximation:
    1 50% Brahui +50% Finnish @ 11.428927

    Using 3 populations approximation:
    1 50% Balochi +25% Chuvash +25% Southwest_Finnish @ 10.374825

    Using 4 populations approximation:
    1 Balochi + Mari + North_Swedish + Tabassaran @ 9.280260
    2 Brahui + Mari + North_Swedish + Tabassaran @ 9.311368
    3 Brahui + Chuvash + North_Swedish + Tabassaran @ 9.408560
    4 Balochi + Chuvash + North_Swedish + Tabassaran @ 9.411836
    5 Balochi + Mari + Swedish + Tabassaran @ 9.432848
    6 Brahui + Mari + Swedish + Tabassaran @ 9.468179
    7 Brahui + Chuvash + Southwest_Finnish + Tabassaran @ 9.483870
    8 Balochi + Chuvash + Southwest_Finnish + Tabassaran @ 9.486706
    9 Balochi + Mari + Southwest_Finnish + Tabassaran @ 9.487204
    10 Balochi + Chuvash + Finnish + Tabassaran @ 9.508200
    11 Brahui + Chuvash + Finnish + Tabassaran @ 9.514761
    12 Brahui + Mari + Southwest_Finnish + Tabassaran @ 9.517778
    13 Balochi + Finnish + Tabassaran + Tatar @ 9.559561
    14 Balochi + Finnish + Mari + Tabassaran @ 9.593977
    15 Balochi + East_Finnish + Tabassaran + Tatar @ 9.598396
    16 Brahui + Finnish + Tabassaran + Tatar @ 9.604749
    17 Brahui + Finnish + Mari + Tabassaran @ 9.629777
    18 Balochi + Lezgin + Mari + North_Swedish @ 9.655762
    19 Brahui + East_Finnish + Tabassaran + Tatar @ 9.661352
    20 Balochi + Chuvash + Swedish + Tabassaran @ 9.709421

    ReplyDelete
  181. Interesting to see so much "Uralic" admixture on the steppe. This ancestry must have expanded after the IE steppe component during the brinze age.

    Also interestingly the 2 N1c samples we have (still waiting on the last) seem non-uralic. One being a Buryat, the other a Lithuanian.

    ReplyDelete
  182. @David

    DA23 is a Siberian genome.

    DA23: Y: C-Y11990 mtDNA: F1b1b Gedmatch: Z577705

    1240k SNPs in 23andMe format:
    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA23/genome_DA23-1240k.zip
    SNPs read: 931251 percent covered: 77.71%

    His Y is in a very ancient and rare clade, C-Y11990.
    His mtDNA F1b1b is Tungusic (Duggan) and Uyghur.

    DA23 is clearly a Shor, a Siberian Turkic-speaking people of mixed Turkic, Ugric, Samoyedic, and Ket origins.

    K13 Oracle4:

    Using 1 population approximation:
    1 Shors @ 6.412794
    2 Hakas @ 9.038127
    3 Kazakh @ 13.870853
    4 Altaian @ 15.216789
    5 Kirgiz @ 17.223745
    6 Ket @ 22.627415
    7 Tuvinian @ 23.205286

    Using 2 populations approximation:
    1 50% Shors +50% Shors @ 6.412794

    ReplyDelete
  183. @ Anthony

    N1c-VL29 is a very important yDNA in Finns and Estonians. If N1c-VL29 spoke a Baltic language, then we know who may have brought an important part of the Balto-Slavic/Indo-Iranian lexical heritage to Western Uralics.

    But of course you prefer a situation in which there is a Baltic speaking autosomally Lithuanian N1c-VL29 who borrows Balto-Slavic/Indo-Iranian words to Uralic speaking N1c-VL29.

    However, I have seen this analysis for Eastern Finns
    Finnish_East

    Latvian 34.05
    Saami 33.70
    Icelandic 19.80
    Lithuanian 8.70
    Comb_Ceramic 3.75

    It seems obvious that Balts and Finns share anccestry (in this analysis c. 43%), and I wonder if it can be disconnected from N1c.

    ReplyDelete
  184. DA8 is a female DA100! Balto-Slavic-Indo-Iranian? Score 1 out of 4!

    DA8: mtDNA: U2e1h Gedmatch: Z330228

    1240k SNPs in 23andMe format:
    http://www.open-genomes.org/genomes/Eurasian%20Steppe/DA8/genome_DA8-1240k.zip
    SNPs read: 897817 percent covered: 74.92%

    U2e1h is one of the signature steppe mtDNA haplogroups.

    K15 Oracle4:

    The "Venedi"? ;)

    Using 1 population approximation:
    1 Tatar @ 17.034996
    2 South_Polish @ 21.172087 (Tak!)
    3 Southwest_Finnish @ 21.281363
    4 Ukrainian_Lviv @ 21.299383
    5 Finnish @ 21.517849
    6 Kargopol_Russian @ 21.623228
    7 Hungarian @ 21.731464
    8 Ukrainian_Belgorod @ 21.943901
    9 North_Swedish @ 21.945002
    10 Ukrainian @ 22.429651
    11 Southwest_Russian @ 22.515331
    12 Polish @ 22.616821 (Twoje zdrowie!)

    13 East_Finnish @ 22.741804
    14 East_German @ 22.994944
    15 Croatian @ 23.047771
    16 Moldavian @ 23.364923
    17 Erzya @ 23.380066
    18 Chuvash @ 23.406013
    19 Austrian @ 23.478498
    20 Russian_Smolensk @ 23.504206

    ReplyDelete
  185. A correction:
    But of course you prefer a situation in which there is a Baltic speaking autosomally Lithuanian N1c-VL29 who LENDS Balto-Slavic/Indo-Iranian words to Uralic speaking N1c-VL29.

    For those who want to maintain the separation between Balto-Slavic N1c and Uralic N1c, I want to add that the TMRCA of VL29 is only 3500 years, i.e. it dates to c. 1500 BC. The Trzciniec culture is dated to 1900 – 1200 BC.

    ReplyDelete
  186. @ Davidski

    I find it a very interesting theory that there would be Uralic speaking Q1a2 people on the Kazakhstan steppe, Proto-Balto-Slavic speaking N1c people in Western Russia / Belarus and Yukaghir-speaking N1c in Fennoscandia/Karelia. This gives a completely new spin to Uralic roots.

    That Kola Peninsula line may have become extinct, but we know for sure that L1025 of DA171 is found in modern Finns.

    ReplyDelete
  187. @Kristiina

    Do not write Balto-Slavic N1c!

    N1c is not Balto-Slavic, but only Baltic (from FU), no Slavs. The Slavs never had its. It appears among the Slavs and other Indo-European peoples from assimilated Baltic and Finns exclusive very late time.

    ReplyDelete
  188. @Anthony Haken and re:

    "Interesting to see so much "Uralic" admixture on the steppe."

    Still, the question is whether we're dealing with a real "Uralic" admixture or some sort of combinations which happen to be similar to those of different Uralic speaking groups, from the calculator's point of view. Related to that, what is "Uralic" in terms of genetics? Uralic speaking groups are very diverse, even within linguistic sublineages and it is somewhat challenging to predict the nature of the founding population just by looking at the descendants.

    ReplyDelete
  189. Wow, we have a very large number of "Fenno-Scythians" here! This may explain the ethnic origin of certain Turkic groups around the Volga region that are very similar to Volga Uralics autosomally, like Kryashen Tatars and Chuvash, and also the pulse of more Siberian and East Asian-shifted ancestry that is associated with the Uralics there that distinguishes them from Saami and Finnics.

    ReplyDelete
  190. @Anthro Survey Not sure what you mean by practicing "dharmic" faiths. The pre-dominant faith in Sindh prior to Islam was Hinayana Buddhism. Bactria, Sogdia and Khotan were Buddhist as well in addition to being east Iranian, so practicing 'dharmic' faiths is not limited to Indo-Aryans.

    ReplyDelete
  191. @Ryu: From a layman's point of view we are dealing with at least three eastern biases, related to Uralic speakers. First, the one related to ANE-WSHG, possibly the original one? Second, an other one related to arctic admixture and northern expansions of Uralic speakers, visible in groups such as Saami and Nenets. Third, Turko-Mongol type of admixture of the Volga area, visible in groups such as Mari, which are very Chuvash like.

    To state the obvious, not all of these these biases are to be found in every Uralic group, at least with same amounts.

    ReplyDelete
  192. Interesting factoid: In the Narasimhan et al F3s tables, the most negative for some Central Asian and SC Asian Neolithics when compared against Iran_N is not West_Siberia_N but Afontova Gora, which edges the Siberian out significantly. Wonder if its possible to tease out the different contributions in some way.

    We still need C and S Asian HGs from mesolithic contexts.

    @ Finn

    Yep. West Siberia_N may represent a fit for the Central Siberian cluster peaking in Kets and Nganasan and representing a strong substrate in Khanties and etc.

    The release of all these new genomes will be very exciting... I guess we'll see the slow bleeding of West Asian ancestry into the Steppes, making the prior Scythians look more Sarmartian-like, and the appearance of large quantities (>20%) of East Asian ancestry proper only with the Turks and not before, on top of the more Siberian-like ENA from the prior Scythians.

    One interesting thing is how 'splashy' the populations in Siberia are... 40% ENA people reached Bolshoy Oleni Ostrov many millenia ago via a tundra route, but at the same time West Siberia was dominated by West Siberia_N that was at most 20% ENA and whose autosomal contribution persisted till very late.

    ReplyDelete
  193. @All

    Here are the Global25 coordinates for the Scythians and whoever else they might be.

    https://drive.google.com/file/d/1IuHoFKUvgQHK1OC6GIYPm9-NSVn0Lf0-/view?usp=sharing

    ReplyDelete
  194. Ryukendo: the appearance of large quantities (>20%) of East Asian ancestry proper only with the Turks and not before, on top of the more Siberian-like ENA from the prior Scythians

    One thing I was wondering, which I wanted to sound off of you: I believe Siberian ENA ancestry looks the dominant component in lots of Central Asian/steppe groups today?

    But could this actually be a result of these groups being a fusion of A: Steppe+Central Asian+Siberian_ENA, with B: East Asian proper+Siberian_ENA?

    That is, say: A- 50:50 Siberian_ENA:Steppe+early Central Asian and B- 50:50 Siberian_ENA:East Asian proper, and C- 50:50 A:B, then C- 50:25:25 Siberian_ENA:Steppe+early Central Asian:East Asian proper.

    So that it might look like the Siberian_ENA is a "main component" that picked up some ancestry from other populations, but this is not necessarily so.

    ReplyDelete
  195. @ Matt

    It varies from group to group, for example in the East Altaians, Shors and Khakhass in Siberia are more Siberian, south of them Kyrgyz and Uyghur have a more even mix, and in the West e.g. Karakalpaks are more Siberian and Turkmens have a more even mix. It probably varied from group to group.

    Pugach et al however make it clear that Siberian Turks like Yakut and so on are distinguished from the peoples around them by having ENA proper plus some Steppic ancestry. The appearance of large quantities of ENA proper, resulting in preference for input from those like DA45 or Scythian Zevakinochilikta or Altai_IA (instead of Pazyryk and so on) is probably a distinguishing feature of those migrations that are properly Turkic.

    ReplyDelete
  196. Hmm a correction, looking at the qpAdm of West Siberia N, they are only 20% ENA with the rest ANE and EHG, which makes them very West Eurasian and very unlike Nganasan, Ket and so on.

    If anyone wants to try nMonte fits with West Siberia N and an array of late local EHGs (Combed Ware, Narva) and Steppe sources (Baltic BA, Srubnaya, Sintashta, Hunnic, Scythian, Sarmatian) would make for some interesting fits for Uralics. A mix of local EHGs and West_Siberia_N may make for a nice source for the HG side of proto-Uralic populations (which will have much less ENA than a naive reading of the "Nganassan" ADMIXTURE component in Uralics will have us expect). The West_Siberia_N are very old in their region (8000-6000 ya) and a stable HG continuum of the kind we find in other places (E-S Africa, Anatolia) between them and the Combed Ware peoples probably existed in the Taiga around the Volga.

    ReplyDelete
  197. @ryu, thanks for your thoughts on this. should have figured it wouldn't be that easy to work out. I'm mainly interested thinking about, if there ever was a single proto-Turkic, whether this population was

    - originally like Siberian ENA HG
    - more of a balance of "Yellow River Farmer" and Siberian ENA HG ( in a process somewhat analogous to what the formation of Yamnaya looks like from fairly roughly balanced proportions of a mainly CHG/Iranian related population and EHG related population)
    - and whether they would have had to have any Western Eurasian steppe related genetic input from the beginning or not

    And whether this tells us anything about where and how they formed.

    But this kind of proportional thinking is probably beyond what we can think about usefully now.

    ReplyDelete
  198. @Rob

    If things went down the way models suggest they did(of which OldEurope is so very fond of)----Ukraine_Eneo-like folks mixing with Khvalynsk and CHG to make Yamnaya proper---I made a rather fitting analogy. Don't have the exact numbers, but suppose Ukr_Eneo groups were about 30% EEF. I'm sure you and I both agree about this EEF portion exerting disproportionately large cultural influence. So, for the sake of argument, let's assign an arbitrary 75% EEF cultural ancestry to these Eneolithic groups.
    Now, per models, suppose Yamnaya proper were about 30% Ukr_Eneo ancestrally. Let's suppose, again, that Ukr_Neo groups, in turn, exerted disproportionate influence on Yamnaya's package---75%.
    So, even though actual EEF ancestry is in the single digits by now, it's easy to how this leveraging effect makes Yamna more Old European than might be expected.

    6/16, or ~40%, is NOT a trivial ratio if it's real and if it came in tandem with Sintashta-like. Given that migrant-descended proto-Brahmin groups effectively defined India's cultural future, that Gonur ancestry suddenly isn't so trivial. Kshatriya(warrior caste) also get a healthy Gonur signal, btw.

    Not sure what to say about Swat because that's a transitional region and some of the BMAC acestry there may be pre-existing, owing itself to geographical proximity.

    ReplyDelete
  199. @ Anthro

    At this stage i disagree. I think it was groups from Kuban-Don steppe mixing with western Sredni Stog to form Yamnaya. Khvalynsk was peripheral in the process.
    As I was saying, you're calculations are wrong about EEF in Yamnaya, as you're basing it solely on Russian Yamnaya (& the Ukrainian Yamnaya females).

    Look at the western end of things.

    Sredni Stog average
    Ukraine_Mesolithic 65.1 %
    Balkans_N 26.6 %
    CHG 8.2 %
    WHG 0.1 %

    Yamnaya_Bulgaria
    Balkans_N 37.1 %
    CHG 26.4 %
    Ukraine_Mesolithic 18.3 %
    EHG 18.2 %


    I also disagree about modern north Indian groups. To me it seems as if steppe groups arrived straight to IVC populations.

    ReplyDelete
  200. @Rob

    Yamnaya_Ukraine
    Khvalynsk_Eneolithic 52.6%
    Ukraine_Eneolithic:I6561 25%
    CHG 22.4%
    EHG 0%
    Ukraine_N 0%
    Ukraine_N_o 0%

    Kalmykia and Samara are similar to this. Even if it wasn't Khvalynsk per se, it was still an interaction axis involving an early EHG---CHG continuum and EEF-admixed groups from Eastern Ukraine.

    Which samples are you including in your S.S. average, btw? They date from different periods and shouldn't be treated as a continuum even though they might deceptively appear as such.

    Bulgarian Yamna, an outgrowth of the culture's Ukranian/Russia core, can be modeled as migrants from the Ukraine+native Balkan admixture.

    ReplyDelete

Read the rules before posting.

Comments by people with the nick "Unknown" are no longer allowed.

See also...


New rules for comments

Banned commentators list