Modelling Admixture with D-stats

davef said:
He did a nice job! Yes I read the thread before I even signed up here and its pretty interesting but is the "whg" in jews actual whg from a northern euro population or is it something that resembles it? What gets to me is that in the lazaridis study mentioned in that thread, jews have absolutely none of it, yet they supposedly mixed with Polish who have a ton of it? And the ibd with Poles could be from jews to poles and not vice versa, according to that thread. I'm not quite convinced that jews have actual ancestry from north or Eastern Europe but I'm open to any counter argument.
Click to expand...

Precision is important when discussing these things, yes? It's true that in the Lazaridis et al three ancient population model, and in their graph based on f3 stats, the Ashkenazim, along with other populations, don't score any WHG. That's not the same as saying they don't have WHG, because they do, but it's included within the "farmer" percentage.

In terms of the IBD sharing, why would you assume that the gene flow was all from the Jews into Slavs? I don't see any evidence that was the case. If nothing else, with each pogrom there were reported rapes. Some of those no doubt resulted in pregnancies, much as I would prefer not to discuss it. They may not have happened specifically in Poland, but they were "Slavic" genes nonetheless. That's why ultra-orthodox Jewish women still have very short hair and wear wigs. The custom developed in the Pale, and was to make them less attractive to the surrounding goyim. At least that's what my Orthodox friends have told me.

Regardless, I don't see how the AJ failure to score any "additional" WHG means there wasn't a 3-5% influx of Slavic genes into the Ashkenazim. The Sicilians don't score any WHG, and yet we know that there was a large migration of northern Italians, who do score it (although not was much as in the Slavs), into Sicily in the Middle Ages.

In the last page of the Lazaridis et al paper, in Extended Data Table 3, they discuss the range of the possible WHG ancestry in all of their chosen populations. The Greek range is from .019 to .060 and yet we know there was a Slavic migration into Greece, although it was heavier in the Balkans. That's not very impressive. WHG scores are low all over southern Europe. Look at the Tuscans and the Spaniards: they score 12% as a mean, but the low end of the range for the Spaniards is .066. (A bit off topic, but you might want to take a look at the total EEF scores for some of these populations: Bergamo (72%), Tuscan (75%) , Spanish (76%) , Albanians (78%), Greek (79%), Sardinians 82%. Bulgarians are very close to Bergamo, with .718, but of course Cavalli Sforza proved that 30 years ago.)
https://genetics.med.harvard.edu/re...14_Nature_Lazaridis_EuropeThreeAncestries.pdf

When you're talking about flow into such a bottlenecked group as the Ashkenazim, it's very possible that just chance determined what alleles were passed on. Drift is very important in such groups.

At any rate, even if, for some reason, someone wanted to discount the many and excellently done research papers on the Ashkenazim and other Jewish groups which show such admixture, we now have ancient genomes from the Levant Bronze Age, of a population which is pretty close to being a proto-Canaanite one, as I've said before. The Ashkenazim are pulled away from them toward Europe. We likewise have Samaritan dna, people whom we know both from history and from comparisons to the Ashkenazim, are related to them, and we know where they plot. Again, they're pulled away from the Samaritans. '

So, the admixture happened; what remains to be discovered, only through ancient dna, in my opinion, is when, where, from whom, and precisely how much.

ClJ7V6WWkAA03oR.jpg
 
King Casimir III of Poland had at least 6 illegitimate sons and several illegitimate daughters. Some of those children (including 2 sons) were with a Jewish woman. But I think that they became Christians, so in this case it would be IBD from Jews to Poles.

It is ironic that he did not have any legitimate sons, only "bastards". All of his legitimate children were daughters.
 
Wow, thank you for the long detailed response!

Going by the chart you posted, Ashkenazim are way way north of Bronze Age levant where we have our Canaanite. The difference is about as great as between Ashkenazim and British I would guess.

So if Ashkenazic Jews were originally like these Canaanites, there's no way in my head that they remained like them up until they reached Central or Eastern Europe. Someone who is 95 percent Bronze Age levant and 5 percent polish would never plot as north as Ashkenazim do in the map.
 
davef said:
Wow, thank you for the long detailed response!

Going by the chart you posted, Ashkenazim are way way north of Bronze Age levant where we have our Canaanite. The difference is about as great as between Ashkenazim and British I would guess.

So if Ashkenazic Jews were originally like these Canaanites, there's no way in my head that they remained like them up until they reached Central or Eastern Europe. Someone who is 95 percent Bronze Age levant and 5 percent polish would never plot as north as Ashkenazim do in the map.
Click to expand...

Obviously not. That's why the working theory has been admixture with southern Europeans in the interim, with the proposed vector usually being Greeks during the Hellenic period. Some central European admixture might have been picked up once they got to the Rhineland as well.

The only Jews anywhere near the modern Levant are North African Jews. Whether that's because they're still mainly "original" Hebrews with only a small dollop of admixture with Sephardim and a few Ashkenazim, or whether the absorption of quite a bit of Berber ancestry also factors into the equation I can't tell.


Tomenable said:
King Casimir III of Poland had at least 6 illegitimate sons and several illegitimate daughters. Some of those children (including 2 sons) were with a Jewish woman. But I think that they became Christians, so in this case it would be IBD from Jews to Poles.

It is ironic that he did not have any legitimate sons, only "bastards". All of his legitimate children were daughters.
Click to expand...

It happens. The War of the Roses which so devastated England was because the male line ran out. Henry VIII of England had one legitimate son, but he died, so the crown passed to two of his daughters even though he had an illegitimate son. Of course, he blamed his wives for his predicament but the male plays a big role in the determination of the gender of the offspring. Also, past a certain point he seems to have been virtually incapable of begetting children of either gender.

As to Jewish gene flow into gentiles, I know some scholars claim there was a bit of Sephardic ancestry in Ferdinand and Isabella. I'm sure there's controversy about it, though, so I wouldn't bet on it.

"Isabella and Ferdinand expelled the Jews from Spain and made the Inquisition into a powerful institution whose main victims were Catholics of Jewish or Moorish ancestry. However, like a part of Iberians in general and most of Iberian nobility, Isabella had some Jewish ancestry: three of her great-great-grandparents had Iberian (Sephardic) Jewish roots.
http://www.helium.com/items/764266-biogr...

The founder of Trastamara dynasty Henry II of Castile was a son of Castilian King Alfonso XI and his mistress of Jewish converso origin Eleanor of Guzman; his grandson Henry III of Castile married Katherine of Lancaster whose mother was a daughter of Castilian King Pedro the Cruel and his Jewish converso mistress/wife Maria de Padilla: Ines Pirez, a mistress of John I of Portugal and mother of Afonso, 1st Duke of Braganza was a Jewish converso; finally all royal families of Iberia and consequently, most of European royalty descended by female lines from the first Navarrese dynasty which was in part of Iberian Jewish descent, that includes ISABELLA's Castilian ancestors Alfonso XI and Pedro the Cruel, John I of Portugal, John of Gaunt, Eleanor of Aragon. etc.

Peggy K. Liss, "Isabel the Queen," New York: Oxford University Press, 1992, p. 165; Norman Roth, "Conversos, Inquisition, and the Expulsion of the Jews from Spain," Madison, WI, The University of Wisconsin Press, 1995,p. 150; Isabel Violante Pereira, De Mendo da Guarda a D.Manuel I, Lisboa, 2001, Livros Horizonte; James Reston, Jr. "Dogs of God," New York: Doubleday, 2005, p. 18.

As I said, don't take this as gospel, as I haven't tracked down each claim. In particular, I'd have to find the Iberian link to John of Gaunt.
 
Actual Southern Euro admixture sounds legit but if it did happen, it had to have been way back when.

Or maybe the levant became more euro like over time. The Bronze Age levant is tens of thousands of years old right? A lot can happen given that much time. Then it became less euro like after some of the jews left. Feel free to point out any flaws
 
davef said:
Actual Southern Euro admixture sounds legit but if it did happen, it had to have been way back when.

Or maybe the levant became more euro like over time. The Bronze Age levant is tens of thousands of years old right? A lot can happen given that much time. Then it became less euro like after some of the jews left. Feel free to point out any flaws
Click to expand...

Are you perhaps reading this hypothesis from posts by Sikeliot? I understand from other people that he's pushing this narrative along with dubious ideas about the Greeks, in my opinion.

The sample is from around 2300 BC, so, no, not tens of thousands of years ago.
See Supplementary Table I here:
http://biorxiv.org/content/early/2016/06/16/059311.figures-only

The Hellenic Era would have begun in the first millennium BC, so not a huge stretch of time, in my opinion. More importantly, I'm not aware of any huge migration from Europe to the Levant during that intervening time, which is what it would require to move them that distance.

The only possible such movement would involve people like the coastal Philistines, if they were indeed from the Aegean or other places in Europe, but that's highly speculative as we have no idea really of the "ethnicity" of the Sea Peoples. Also highly speculative is how much admixture there would have been with the Hebrews. The latter were almost as endogamous as the Bedouins if we go by their history as expressed in the Old Testament, which perhaps shouldn't be that surprising as these are highly patriarchal initially pastoral groups. Even with the Canaanites, the fulmination against taking their women to wife covers pages. For some of the Hebrews the appropriate thing to do if they had a conflict with them was to kill all of them, men and women alike. Certainly, any admixture would have mostly entailed incorporating their women, for which there is indeed some evidence. Usually, most of the males over a certain age were put to the sword. One would think the Sea Peoples, Philistines included, would have been a male dominated group, so I doubt there was a huge amount of admixture, although there might have been some.

By the Hellenic and Roman periods, attitudes among the Jews had changed, and from being a very "exclusive" religion, it had become an expansive, proselytizing religion. Of course, the institutionalized repression which they started to encounter in the first centuries AD led to a retreat and endogamy once more.

See:
http://www.jewishideasdaily.com/5189/features/is-judaism-a-proselytizing-religion/

"Historians agree that there were different attitudes toward conversion in the talmudic period, but disagree regarding the historical record. Historian Louis Feldman argues that Judaism took an "active approach" toward proselytizing during the Hellenic period, as shown by the dramatic increase in Jewish population at this time. It is precisely because this Jewish approach continued into the 5th century, Feldman says, even after the Roman conquest and the ascent of Christianity, that Roman-Christian jurists felt the need to issue repeated bans on Jewish proselytism. But Shaye Cohen contends that while the Judaism of late antiquity accepted converts, it never engaged in any outreach, especially institutional outreach. Martin Goodman generally sides with Cohen,arguing that while some 3rd century rabbis, influenced by missionary Christianity, "began assuming the desirability of a mission to proselytize," this notion was never widely held or seriously implemented, especially following the Roman bans on proselytizing.Understandably missing from this debate is the phenomenon of conversion of Gentile slaves to Judaism upon the completion of their servitude. Throughout the centuries, Jews, alas, definitely owned non-Jewish (and sometimes Jewish) slaves. Many talmudic sages argued that a Gentile slave purchased by a Jew must immerse in themikveh and, if male, undergo circumcision in a partial conversion; when the servitude was done, conversion could be completed. In line with the rule that conversion must be voluntary, a slave could reject this process; after 12 months of refusal, the slave would have to be sold. But slaves that agreed to the conversion process were required to keep most commandments; and their Jewish owner was prohibited from selling them to Gentiles, since such a sale would preclude their full conversion.
Because of economic pressures and restrictions on conversion, Jews sometimes bought slaves without converting them, a practice allowed by some talmudic sages. Thus, it is difficult to argue that slavery was used to proselytize; the motivation for purchasing slaves was almost certainly socio-economic, not religious. Yet many historians believe that released slaves became a regular source of new members of the Jewish community, even after the talmudic period."

My research into the archives in my own area echoes some of this. Even two centuries after the bans against conversion of gentiles to Judaism had been implemented, the Pope was writing to the local Bishop of Luni to say that he should not be allowing a local Jewish landowner to convert his slaves. It couldn't have been extremely widespread, however, given that there is no out of the ordinary IBD sharing between Jews and Italians.
 
2300 BC? Interesting! That sounds like its within the Bible range, date wise.

Converted Greek slaves or non slaves may have made them plot where they plot according to your post. That seems possible. There is however debate against that as well.

The IBD sharing with Greeks in your chart is nothing to celebrate. Then again, this is with modern ones.

BUT, something must've turned a subset of a Bronze Age levant population into a population heavy in European /Anatolian farmer genes. A Bronze Age levant population will never make it to Europe genetically without intervention.

Or, maybe this whole thing about Jews being Canaanites is bogus and they always were Euro like. :)

And what theory did this Sikeliot come up with regarding Greek people?
 
This is a double post...
I might add that the mystery Euro population (Greeks seem more likely for historical reasons according to your post) that sent the Jews who then became Ashkenazim to Europe must've had an impact on them so big that a lot of their original Hebrew ancestry may have been lost. A half Bronze Age levant half sicilian (I'm guessing greece was more sicilian plotting back then) would be too south for where Ashkenazim plot today. A little Polish might not help much.
 
Working Models for West Eurasia

With the new Middle Eastern aDNA we can get good fits using D-stats for all West Eurasians only using ancient DNA. I'll post more about it later, but here are some interesting results.

First I'll show the new model for Europe.

Two updates
>Southern, Central Europe has NorthWest Asian ancestry(mixture of Anatolia_N, Levant_N, Caucasus_HG).
>Saami have SHG ancestry

SteppeEurope_MNCaucasus_HGLevant_BANW AfricanSiberianWHGSHG
Italian_Tuscan22.543.85825.65------------0.012477
Sardinian3.267.454.824.55------------0.010596
Spanish28.250.954.2510.55.4---------0.009378
Basque_Spanish2868.953.050------------0.009118
Bulgarian32.6538.48.3520.6------------0.011657
English_Cornwall45.451.053.550------------0.009454
Lithuanian43.3539.252.6------2.7510.851.20.008322
Sami32.5523.10------26.46.911.050.025285


I've been saying for a long time Southern and Central Europeans have Cypriot-like admixture. The new ancient DNA supports this theory. Cypriots and Anatolia_Chl(they're not the same. Anatolia_CHl has more EEF) in D-stats fit well as Anatolia Neo+Levant BA+Caucasus HG. Cypriot and Anatolia_Chl fit well as an ancestor for South/Central Europe. I'm very confident an EEF/Levant/Caucasus admixed population from Anatolia or nearby migrated into Europe after the Neolithic. Jordan_BA can be fit as mostly Anatolia_N, with African and Iran_Neo admixture, and so it's contribution in Southern/Central Europe has been confused as Neolithic European ancestry.

BTW, al Anatolia_Neo ancestry isn't from Europe_MN. Levant_BA and Steppe have it, so if you include that most Europeans are near 50% and Southern Europe is over 50%. The people who came into Europe from NW Asia after the Neolithic were mostly EEF, so....

When using EHG, WHG, EN_Europe, CHG, Levant_Neo these are results I get.

Eastern_HGSatsurbliaLBK_ENHungary_ENIberia_ENLevant_NeolithicLoschbour
Italian_Tuscan15.617.1558.8007.20
English_Cornwall25.316.1542.5507.5508.45
Spanish15.9514.141.908.759.359.25


It's consistent using older genomes. When using more ancient genomes like these I get basically the same results. Yamnaya scores about 60% EHG(over 40%, 30% CHG and 10% Anatolia_Neo) which can be translated as about 50% Yamnaya for English, and 27% for Tuscan and Spanish. Notice Tuscan and Spanish though score as much CHG as English. That's because of Post-Neolithic ancestry from West Asia which had lots of CHG.
 
Summary: I'm not confident yet that such high amounts of Jordan_EBA admixture like above is realistic. I used Europe_MN references who have more WHG admixture than Neolithic Italy and Balkans probably had which rises Jordan_EBA percentages. It looks like Tuscany and Greece is mostly Anatolia_Neolithic(~60%) with significant EHG(15%)/Steppe(25-30%) admixture, and also some WHG(2-5%) and CHG/Iran_Neo(10-20%). Iran_Neo/CHG estimates are rough estimates and change with differnt references, but the EHG/Steppe percentage looks concrete and the Anatolia_Neolithic one looks pretty concrete. The Anatolia_Neolithic admixture arrived at differnt times. It's probable that lots arrived with Steppe admixture and some arrived with CHG/Iran_Neo admixture.

Greece, Tuscany, and the rest of Italy and Balkans are fairlly similar but Sardinia is as we know differnt. They look like WHG admixed Neolithic European with 10-20% EHG and Iran_Neo/CHG admixture. Their WHG signal is stronger than Italy's and Balkan's, it's very real. Their EHG signal on the other hand is very weak but real, while Italy/Balkan's EHG is very strong. Like with Italy/Balkans their EEF admixture arrived at differnt times. Lots could have arrived with Iran_Chl/CHG admixture and EHG admixture.

-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

The model above is good but not perfect. For Europe the only mystery is the Southern half. I'll move on to Middle East later. There's reason to put all of Europe under one entity, because everyone has a big chunk of Steppe and EEF and everyone has at least some WHG even in Greece. IMO, Greeks are largely maybe 50% from people who arrived from North of Greece after the Neolithic and brought with them Steppe and WHG admixture.

I chose to use Greece, Tuscan, Sardinia to get good models for Southern Europe because Southern Italy/Spain as hard to model North African admixture and Northern Italy is shifted too far North. Balkans and Italy do have fairly similar proportions of ancestry even though they have few Y DNA connections. Spain is a bit differnt.

Eastern_HGSatsurbliaAnatolia_NeolithicJordan_EBAIran_NeolithicLoschbourHungary_HGLaBrana1VillabrunaIndia_SouthNganasanYoruba
Italian_Tuscan16.0514.1560.6504.8004.350---------0.014227
Greek18.413.857.907002.90---------0.015684
Sardinian4.156.457502.750011.650---------0.013782

Yamnaya_SamaraSatsurbliaAnatolia_NeolithicJordan_EBAIran_NeolithicLoschbourHungary_HGLaBrana1VillabrunaIndia_SouthNganasanYoruba
Italian_Tuscan27.755.6555.11.25.5004.80---------0.012863
Greek31.63.952.408.6003.50---------0.014237
Sardinian7.953.873.703.20.25011.10---------0.013508


So Southern Europe can be modeled mostly mostly Anatolia_Neolithic(50-60%) with significant Steppe/EHG admixture and minor WHG, CHG, and Iran_Neo admixture. When I take out Anatolia_Neolithic and replace it with "EF10"(10% Hungary WHG and 90% Anatolia Neolithic) or "EF15" to represent WHG admixed Neolithic Southern Europe(maybe not in Greece) here are the results.

Yamnaya_SamaraSatsurbliaEF10Jordan_EBAIran_NeolithicLoschbourHungary_HGLaBrana1VillabrunaIndia_SouthNganasanYoruba
Italian_Tuscan26.256.958.153.355.350000---------0.01354
Greek28.75.356.10.059.850000---------0.014951
Sardinian5.95.0580.0504.45004.550---------0.014315
Corded_Ware_GermanySatsurbliaEF10Jordan_EBAIran_NeolithicLoschbourHungary_HGLaBrana1VillabrunaIndia_SouthNganasanYoruba
Italian_Tuscan36.155.7545.67.64.90000---------0.013112
Greek39.254.05433.959.75---------------------0.014677
Sardinian9.754.0577.74.9003.60---------0.014176

Bell_Beaker_GermanySatsurbliaEF10Jordan_EBAIran_NeolithicLoschbourHungary_HGLaBrana1VillabrunaIndia_SouthNganasanYoruba
Italian_Tuscan45.59.6536.355.23.30000---------0.01459
Greek48.98.3533.7508.35---------------------0.01651
Sardinian17.24.1572.8503.9001.90---------0.013967

Bell_Beaker_GermanySatsurbliaEF15Jordan_EBAIran_Neolithic
Italian_Tuscan44.4510.9531.8512.7500.015322
Greek49.7510.125.6511.92.60.017207
Sardinian12.95.675.32.93.30.014627


As WHG in the "EF" ancestor rises the CHG/Iran_Neo proportions rise. When EF15 is added for the first time Southern Europe needs Jordan_EBA admixture. When I add Anatolia_Chl no one needs admixture from them which is surprising.

Bell_Beaker_GermanySatsurbliaEF15Anatolia_ChalcolithicJordan_EBAIran_Neolithic
Italian_Tuscan44.4510.9531.85012.7500.015322
Greek49.810.0525.6011.952.60.017207
Sardinian12.955.675.2033.250.014627
 
Assuming that the Bronze Age Levants are representative of how ancient Jews were genetically, I'm still open to the possibilities that modern European Jews either are true descendants of these people but absorbed a ton of South Euro like ancestry up to the point where they don't resemble their Biblical forefathers, or they started out from a South Euro like population. The former idea draws from the fact that they are extremely far from Bronze Age Levant (far as in the genetic distance between them and the British) and a half far South Euro (Sicily, Malta) half Bronze age Levantine wouldn't plot nearly as far into Europe where Euro Jews are now. 3-5 percent additional north/east euro won't do much to assist. The D-stats that Fire-Haired did an excellent job on show only 14 percent Bronze Age Jordan (don't know if that's the same as BA Levant, I'll have to see).

I'm sure these opinions would make me the bad guy on most other anthro forums, as they aren't very "politically correct" but I don't set out to offend. I'll still be happy if someone is willing to spot out the holes and offer something that sides with the idea that Euro Jews resemble the Jews of the Bible genetically. I'm not counting on my opinions to be accurate.
 
The last results I posted for Ashkenazi Jews were of the form (Chimp, Row)(Mbuti, Colmn) this is of (Chimp, Colmn)(Mbuti, Row) and as far as I've seen the later creates better fits than the former.

@davef,

This is the best fit I've had for Ashkenazi Jew. It might be realistic. First admixture in the Balkans, then immigration to Eastern Europe and admixture there.

BTW, Jordan_EBA is very similar to modern SW Asians but not the same. SW Asians have extra African ancestry and many have an extra load of Iran_Neo-like ancestry. Ashkenazi_Jews and Druze are ones with this extra load of Iran_Neo.

I doubt Jordan_EBA is a good reference for ancient Jews. IMO, Modern Levanties are, like Druze, Palestinans, Samartians, etc. The Jews tell of migration in the Bible. They arrived in Israel only by the Late Bronze age about 1,500 years after Mr. Jordan_EBA was born. Who knows what type of admixture they picked up along the way and who knows what admixture arrived in the Levant in general after Jordan_EBA.

distance=0.005375"
Ashkenazi_Jew
"Druze" 50.05
"Greek" 34.5
"Belarusian" 11.05
"Jordan_EBA" 4.4
"Satsurblia" 0
"Iran_Neolithic" 0
 
OK, my patience is at an end, Tomenable. Now I'm going to have to waste time removing all these off topic posts to new threads. I don't appreciate it, and on a holiday no less.

Cut it out, and I mean it!
 
Fire Haired14 said:
The last results I posted for Ashkenazi Jews were of the form (Chimp, Row)(Mbuti, Colmn) this is of (Chimp, Colmn)(Mbuti, Row) and as far as I've seen the later creates better fits than the former.

@davef,

This is the best fit I've had for Ashkenazi Jew. It might be realistic. First admixture in the Balkans, then immigration to Eastern Europe and admixture there.

BTW, Jordan_EBA is very similar to modern SW Asians but not the same. SW Asians have extra African ancestry and many have an extra load of Iran_Neo-like ancestry. Ashkenazi_Jews and Druze are ones with this extra load of Iran_Neo.

I doubt Jordan_EBA is a good reference for ancient Jews. IMO, Modern Levanties are, like Druze, Palestinans, Samartians, etc. The Jews tell of migration in the Bible. They arrived in Israel only by the Late Bronze age about 1,500 years after Mr. Jordan_EBA was born. Who knows what type of admixture they picked up along the way and who knows what admixture arrived in the Levant in general after Jordan_EBA.

distance=0.005375"
Ashkenazi_Jew
"Druze" 50.05
"Greek" 34.5
"Belarusian" 11.05
"Jordan_EBA" 4.4
"Satsurblia" 0
"Iran_Neolithic" 0
Click to expand...

Is a distance of .005 good? Also if you use south italy in place of greek, would the distance increase? That would be interesting
 
davef said:
Is a distance of .005 good? Also if you use south italy in place of greek, would the distance increase? That would be interesting
Click to expand...

Doesn't matter, although it might be about the same given that South Italians/Sicilians and Greeks aren't all that different autosomally; there's no IBD sharing with South Italians. There is with Greeks.
 
Angela said:
OK, my patience is at an end, Tomenable. Now I'm going to have to waste time removing all these off topic posts to new threads. I don't appreciate it, and on a holiday no less.

Cut it out, and I mean it!
Click to expand...

Well... you only forgot to remove your own Off-Topic posts, Angela. :)

Anyway, I wish you a wonderful vacation.
 
Tomenable said:
Well... you only forgot to remove your own Off-Topic posts, Angela. :)

Anyway, I wish you a wonderful vacation.
Click to expand...

Tomenable, you know very well that we are more liberal about posting off topic material than a lot of other sites. However, more than eight posts in a row about something that's off topic? That's a bit different. Just watch it.

Anyway, let's not quarrel. Thanks for the good wishes, but our three day Fourth of July week end is almost over. One more barbecue, for which I am leaving now! :)
 
davef said:
Is a distance of .005 good? Also if you use south italy in place of greek, would the distance increase? That would be interesting
Click to expand...

Anything under 0.02 is a good fit. So 0.005 is very very good. Southern Italy is a bit of an outlier but maybe the Balkans were like Southern Italy(minus NW African ancestry) before Slavic migrations.

Ashkenazi Jews get an equally good fit with South Italy.

distance=0.005037"
Ashkenazi_Jew
"Italian_South" 51.1
"Druze" 34.95
"Belarusian" 13.95

Now assuming Ashkeanzi Jews came to Central/East Europe with no Southern European admixture. This fit works as well. It's not consistent with Y DNA. I haven't looked at Ashkenazi mtDNA though.

distance=0.007639"


Ashkenazi_Jew
"Druze" 68.9
"Belarusian" 31.1
 
South Asia

HighLights:

>Iran_Chalcolithic: West Eurasian side is a mixture of Steppe and something high in CHG and EEF. Iran_Neo doesn't fit this profile. Iran_Chl fits well, but most South Asian's still need extra CHG.
>Yamnaya: Yamnaya is a better proxy for Steppe ancestry than Srubnaya. It doesn't look like South Asians have WHG ancestry which excludes the possibility of Srubnaya-like people.

This is the best model I've found for South Asians. Iran_Neo is not a good ancestor proxy for them suprisingly. When I modeled them with Iran_Neo they usually scored 0, and instead scored equal amounts in CHG(Satsurblia) and Anatolia_N. According to these D-stats their non-ASI and Steppe ancestor was a CHG/EEF mix, and Iran_Neo doesn't fit that profile. It has too low CHG affinity not just too low EEF affinity.

Iran_Late_Neo had some EEF_like admixture, but doesn't work either. Iran_Chl though had lots of EEF and CHG admixture and fits well as the none ASI and Steppe ancestor of South Asians, but most still need some extra CHG.

Yamnaya_SamaraSrubnayaSatsurbliaAnatolia_NeolithicLevant_NeolithicIran_ChalcolithicHungary_HGIndia_SouthHanNganasanYoruba
Kalash30.25010.40021.55037.8---------------0.010657
GujaratiA15.310.72.90017.85053.25---------------0.00596
Tajik_Ishkashim27.7517.16.556.35014.55017.84.355.55-----0.013662
Sindhi16.805.650027.5050.05---------------0.006482
Pathan22.75.656.43.3019.15042.8---------------0.009229
 
You must log in or register to reply here.

This thread has been viewed 44129 times.

Back
Top