
Originally Posted by
Angela
"The amount ofsteppe ancestry is about ~13% when the Early/Middle Bronze Age group(“Yamnya/Afnasievo/Poltavka-related”) is used as a source (Steppe_EMBA), which is in harmonywith our finding of ~7% EHG ancestry in Mycenaeans, as this group has about half of its ancestryfrom the EHG1,8,16. The proportion is slightly higher when the Middle/Late Bronze Age(Steppe_MLBA) group (“Srubnaya/Andronovo/Sintashta-related”) is used as a source, and higher stillwhen the Late Neolithic/Bronze Age populations from mainland Europe (Europe_LNBA) are used asa source, reflecting the fact that these have substantial European/Anatolian Neolithic-relatedancestry1,8,20 which dilutes their EHG-related ancestry further. We cannot distinguish which of thesepopulations was a source for Mycenaeans (whether there was a migration directly from the steppe,from populations related to the Early, Middle/Late Bronze Age steppe, or an indirect migration fromcentral Europe from steppe-influenced populations that were formed there during the Late/NeolithicBronze Age)."
"Thus, while we cannot distinguish between the differentsource populations of ‘northern’ ancestry, our results do not depend strongly on the sampledpopulations, as quantitatively similar estimates of their impact on Mycenaeans are inferred when weeither use any of them, or use none of them, but simply infer ancestry from an unsampled “ghost”population from either the eastern European-Iran continuum that formed the early populations of thesteppe1,10,13, or the steppe-European farmer continuum of the Middle/Late Bronze Age8,20."
"While both ‘eastern’ and ‘northern’ 2-way mixture models fit the data statistically, we were curiouswhether a more complicated model could provide additional insight, so we tested 3-way mixturemodels with Anatolia_N or Minoan_Lasithi as the substratum population and both steppe-related‘northern’ ancestry (Steppe_EMBA, Steppe_MLBA, or Europe_LNBA) and Armenia-related‘eastern’ ancestry (Armenia_MLBA or Armenia_ChL). The results are presented in Table S2.26.Anatolian Neolithic/Minoans make up the majority of the ancestry (~59-90%) in all these models.Most of the coefficients for the ‘northern’ and ‘eastern’ ancestry are positive, suggesting that there ismigration from both sources, but many of these positive coefficients do not significantly differ fromzero (explaining why the simpler 2-way mixture models fit the data adequately without taking intoaccount a 3rd ancestral source). Interestingly, the proportion of ‘eastern’ and ‘northern’ ancestry inTable S2.26 are anti-correlated (r=-0.95) suggesting again that they both capture the same underlyingphenomenon.Table S2.26: 3-way mixture models. Left = (Mycenaean, A, B, C). The Right set is All++"
"However, we do notice that the model79%Minoan_Lasithi+21%Europe_LNBA tends to share more drift with Mycenaeans (at the |Z|>2level). Europe_LNBA is a diverse group of steppe-admixed Late Neolithic/Bronze Age individualsfrom mainland Europe, and we think that the further study of areas to the north of Greece mightidentify a surrogate for this admixture event – if, indeed, the Minoan_Lasithi+Europe_LNBA modelrepresents the true history."
There's a large section on the implications for language change...too large to copy and paste. It starts on page 49.
Certain people won't be at all happy.