Carmi, Xue paper on "The Time and Place of European Gene Flow into Ashkenazi Jews"

Angela

Elite member
Messages
21,823
Reaction score
12,327
Points
113
Ethnic group
Italian
Carmi, Xue paper on "The Time and Place of European Gene Flow into Ashkenazi Jews"

We'll have to wait fro the paper, which was also presented at the Cold Spring Harbor conference, but this is the poster, which is pretty informative:

https://shaicarmi.files.wordpress.com/2014/08/aj_admixture_poster.pdf

I don't see anything that different, as the 37% or so figure has been posited already. Of that 37%, they claim that 75% is of southern European origin, and the remainder from eastern Europe.

As for the timing, some of their methods give an estimate of around 20 generations. Using a figure of 30 years per generation, which I think is pretty standard, that's about 1400 AD, which is a pretty good fit for the arrival of the Jews in eastern Europe. They also posit a 30 generation possibility for the admixture, which is around 1100 AD, or when they are attested in Germany. Only with the 30-40 generation estimate they get with Alder do you get to around 800 AD, which is when some sources attest their presence in northern Italy.

It seems to me that all of these dating methods tend to emphasize the "last" date of admixture. So, it's still unclear to me that the admixture event with "southern European" can actually be dated to even 800 AD, although it's possible. What makes me suspect perhaps it was somewhat earlier is the fact that Judaism was a proselytizing religion during late antiquity and there are reports of conversions among Greeks, and, to a lesser extent, among Romans from that period and I wonder if some of this "southern European" admixture may date to that period.

Also, to the best of my recollection, wasn't there an IBD analysis that showed a correlation with Greek dna rather than Italian dna? I'll try to locate that analysis if I have some time.

There's been a lot of uninformed, agenda driven commentary on this paper that I'm not even going to bother to address, since it wasn't posted here. Suffice it to say that in my opinion any theory that Ashkenazi Jews have no Levantine roots is unsupported by autosomal dna, but also by yDna.
 
The source says 50%.

"Based on sequence data, we recently estimated the EU and ME ancestry proportions to be roughly equal"

I thought they were talking about a prior paper, and that's why the number 4 appears after that sentence. In the references, number 4 is a 2014 Carmi paper.

I was going by this:
Carmi graphic of Ashkenazi ancestry.JPG

As with so many of these things, I think we need an ancient sample...Judea, or even just a Jewish sample from a diaspora community. In any case, something from around the first century AD.
 
Ashkenazi didn't have absorbed some Italian and Greek blood in their diaspora as well?
 
They claim it's 50% ME 35% south Euro, 12% east Euro and 3% west Euro, I wonder how they would manage to even get 12% east Euro, maybe during the early days of settlement in Poland when the Poles welcomed the Jews as the latter were fleeing western Europe. As for the south Euro, the Hellenistic/Roman period seems to make most sense IMHO.
 
They claim it's 50% ME 35% south Euro, 12% east Euro and 3% west Euro, I wonder how they would manage to even get 12% east Euro, maybe during the early days of settlement in Poland when the Poles welcomed the Jews as the latter were fleeing western Europe. As for the south Euro, the Hellenistic/Roman period seems to make most sense IMHO.
Don't you think that Ashkenazi have absorbed some East Euro blood in Poland, Lithuania and Russia?Looking at Roman Abramovich I think it's possible.
 
Ashkenazi didn't have absorbed some Italian and Greek blood in their diaspora as well?

Well, that's what this paper purports to say. As I posted upthread, only one of the dating methods puts the admixture at a time when they were in Italy, and that's the one that approximates admixture at around 800 AD. All the other admixture dates are too recent. However, these dating methods are often believed to "capture" the last date of admixture best, so perhaps there was some admixture with other groups took place earlier, in the classical Greek and Roman era.

However, if my memory serves, the IBD sharing is more with Greeks than Italians. I'll have to check my folders for the paper.

As to the East European, I think it might be around 10% or less, but that's just my opinion. As to the source, absorbing some people when they first went east, especially some women, at a time when some of the people were not Christianized, is certainly a possibility. However, as painful as it is to contemplate, Orthodox women shaved their heads and wore wigs for a reason...to wit, the marauding groups of Cossacks and other groups during the various pogroms.

To be honest, I've wondered what the attitude was to abortion among Orthodox Jews of that time and place. I think it may have been religiously prohibited. Not that religion always stops women from having abortions even in countries where it's against the laws of both God and man. Getting pregnant after a violent, repulsive rape would constitute, I think, one of the situations where a woman might question whether she can bear to live up to her society's rules.

As is usually the case, in my opinion, the data can be interpreted in various ways, and how it's interpreted can be affected by the biases, preconceptions, and even the politics of the person involved, unless they're very careful.

Ed. On reflection, I would think most of it might come from women absorbed in the first centuries. Otherwise we'd see more eastern European yDna. (The R1a has now been found to be the "Asian" variety, yes?)
 
Don't you think that Ashkenazi have absorbed some East Euro blood in Poland, Lithuania and Russia?Looking at Roman Abramovich I think it's possible.
It's possible, I'm just saying that it must have happened in a short timespan.
 
In terms of modern proxies, until we get a Judean genome from the beginning of the modern era, I think the Samaritans are probably a good bet, or even the Druze. I know that some Jews get them as part of their Oracle results.

Just looking at it from a historical perspective, despite the belief of Jews that Samaritans have some Assyrian ancestry, they're like a fossilized genetic remnant of classical era people from the Levant. No input from the Arab invasions and the SSA mixture they brought through slavery, no input from Crusaders or Ottomans. The Druse also are pretty much a remnant population even if they do have some Iranian input, although I wonder how much of that is ascribed to them just because of the founder of their religion.

That should give a general idea, I would think.

From Wiki:
http://en.wikipedia.org/wiki/Samaritans#Genetic_studies

"Recently several genetic studies on the Samaritan population were made using haplogroup comparisons as well as wide-genome genetic studies. Of the 12 Samaritan males used in the analysis, 10 (83%) had Y chromosomes belonging to haplogroup J, which includes three of the four Samaritan families. The Joshua-Marhiv family belongs to haplogroup J1, while the Danfi and Tsedakah families belong to haplogroup J2, and can be further distinguished by M67, the derived allele of which has been found in the Danfi family. The only Samaritan family not found in haplogroup J was the Cohen family (Tradition: Tribe of Levi) which was found in haplogroup E3b1a M78.[59] This article predated the change of the classification of haplogroup E3b1-M78 to E3b1a-M78 and the further subdivision of E3b1a-M78 into 6 subclades based on the research of Cruciani, et al."

The paper to which they refer is the Cruciani et al one:
http://onlinelibrary.wiley.com/doi/10.1002/humu.9445/epdf

Then there is this more recent paper:
"The 2004 article on the genetic ancestry of the Samaritans by Shen et al. concluded from a sample comparing Samaritans to several Jewish populations, all currently living in Israel—representing Ethiopian Jews, Ashkenazi Jews, Iraqi Jews, Libyan Jews, Moroccan Jews, and Yemenite Jews, as well as Israeli Druze and Palestinian Arabs—that "the principal components analysis suggested a common ancestry of Samaritan and Jewish patrilineages. Most of the former may be traced back to a common ancestor in what is today identified as the paternally inherited Israelite high priesthood (Cohanim) with a common ancestor projected to the time of the Assyrian conquest of the kingdom of Israel."

The paper itself:
http://evolutsioon.ut.ee/publications/Shen2004.pdf

I don't see how anyone could claim that the yDna lines of the "western Jews" are not very "Middle Eastern".

Also from the paper:
"For the Y chromosome, all Jewish groups, except for the Ethiopians, are closely related to each other. They do not differ significantly from Samaritans (0.041) and Druze (0.03), butare different from Palestinians (9.163), Africans (0.219), and Europeans (0.111) (Table 4).
(0.219), and Europeans (0.111) (Table 4)"

I also still think that this PCA is pretty informative. You can see how the Ashkenazim, in particular, are pulled toward Europe.
westeurasianpca.jpg

It's interesting to see how far north the Belmonte "Hidden/Converso" Jews plot. Is that the "look" of the original expelled Spanish Jews, or over the centuries did some "non-Jewish" ancestry keep in. Certainly, their religious observances were no longer "orthodox".

As for the Palestinians, despite the political agenda to see them as some sort of relic of the Canaanites, and therefore entitled to throw the Jews into the sea, I think it's clear that they are not a remnant population, although I think there is a distinction between Christian Palestinians and Muslim Palestinians. If comparing them to other Levant people whether inbred or not isn't enough, there is documented movement of Arabian people into the area after the Muslim conquests, a movement that brought a big increase of SSA. Gaza, in particular has seen a lot of migration from the Sinai.

I'm sure there are other analyses elsewhere on the web on the Palestinians specifically, but here on the Dodecad Globe 13 which I could easily access, there is a distinct although not huge difference between the Muslim Jordanians and the Christian Lebanese in terms not only of "SSA" components, but of SW Asian and WAsian. I think it's not unreasonable to see that as in part a result of the Muslim conquests.
https://docs.google.com/spreadsheet/ccc?key=0ArAJcY18g2GadF9CLUJnTUdSbkVJaDR2UkRtUE9kaUE#gid=2

Some Jews do overlap with the Cypriots. It's not clear to me what should be made of that...Cypriots would be an EEF population with additional input from the Bronze Age migrations, yes? So, at least partly, are the Askenazim and Sephardim. So, is it just that they have similar components? Or is it possible that Cypriot Jews had a big impact on the ethnogenesis of more "western" Jews? The expulsions scattered many Jews throughout the world, but the diaspora communities which were not directly involved in the rebellions often escaped those consequences and could have formed the nucleus for the "western Jews".

By the way, does anyone have a citation for the study that shows IBD sharing between Greeks and Ashkenazim? I still haven't hit upon the "tag" that I used in my folder files.
 
Wouldn't really call it a study... More of a map, But here:

https://verenich.files.wordpress.com/2014/07/ashkenaziibd1.png

Also, I think Behar 2013 has something to say about it.

Thank you John,

Yes, Behar et al discusses it in terms of broad regions. This is Figure 6:
IBD for Ashkenazim Behar et al 2013.JPG

Of course, this doesn't necessarily tell you the direction of gene flow, so it can also be showing gene flow from Askenazim into East Europeans, and it says nothing necessarily about PCA plotting or admixture components or overall genetic similarity.

Dienekes also did a fast IBD analysis of them. In his case, he says it should cover about the last 2000 years. Ralph and Coop were supposedly able to go back further.
http://dienekes.blogspot.com/2012/08/fastibd-analysis-of-several-jewish-and.html

For all that the Askenazim may plot near southeastern Europeans there doesn't seem to have been very much gene sharing in the last 2,000 years. It might have taken place before that, of course, but these results would seem to argue against a major influx from Italians in the late Roman empire, and and gene sharing either way in the last thousand years, yes?
 
I wonder how they would manage to even get 12% east Euro

For a few centuries most of Ashkenazi Jews lived in Poland-Lithuania. The peak was in the 1st half of the 1800s, when - according to Salo W. Baron - ca. 42-46% of Jewish (not just Ashkenazi, but of all Jews) population of the world lived in lands of the PLC (at that time already partitioned between Russia, Austria and Prussia). Later they started to emigrate, so that in the 1930s there were already 4,2 million Jews in the USA and 0,4 million in Palestine - most of them Ashkenazi.

Jews became so concentrated in Poland-Lithuania relatively late, during the period from ca. 1500 to ca. 1800.

According to Salo W. Baron, in year 1490 only 5% of the Jewish population of just Europe (not of the world), lived in Poland-Lithuania:

http://www.eupedia.com/forum/thread...nd-haplogroups?p=456253&viewfull=1#post456253

Although for religious reasons Jewish-Christian marriages were rare, and in most cases a Jew was converting to Christianity rather than a Christian to Judaism (so Jewish DNA was entering Christian gene pool), there gradually appeared some sects of Judaism, which facilitated intermarriage. Notably, Frankism:

http://en.wikipedia.org/wiki/Frankism

There is also the Khazar issue, as a possible source for some of East Euro admixture (provided that Khazar DNA had this component).

South Euro admixture is probably from conversions of people to Judaism in Hellenistic times and in Roman times.

Some Jews do overlap with the Cypriots.

Ancient Cyprus had a large Jewish Diaspora community, which existed there since the 3rd century BC.

There were also some tensions and mutual violence between Jews and ethnic Greeks in Cyprus in the 1st century AD.

For all that the Askenazim may plot near southeastern Europeans there doesn't seem to have been very much gene sharing in the last 2,000 years.

This seems to suggest that most of conversions to Judaism took place in Hellenistic times, rather than in Roman times.

Which seems to be valid because in Roman times there was less tolerance for Jews and also tensions appeared between Greeks and Jews.

Those tensions between Greeks and Jews resulted in pogroms of Diaspora communities in the 1st century AD (though not only Greeks killing Jews, also the other way around). During the Jewish War of 66 - 73 AD (which took place in Palestine), Diaspora communities in Alexandria, Cyrenaica, Damascus, Cesarea and Scithopolis suffered from pogroms - during which Greeks/Romans killed many Jews. On the other hand, in Cyprus those were Jews who killed Greeks.
 
Hellenistic times I don't know, but if it took place during the empire and especially around 800 AD in Italy as some have speculated, it's odd that no significant IBD sharing can be detected. Behar's run doesn't show it and neither does Dienekes' even though Dienekes' run shows the IBD sharing for peninsular Italians with Germans and for Greeks with Slavs, both of which took place around 1600 to 1200 years ago.
http://dienekes.blogspot.com/2012/08/fastibd-analysis-of-several-jewish-and.html

This contrasts with the situation for Ashkenazim and Eastern Europeans, where it's clear that it exists.
 
About mass conversions of people to Judaism in Hellenistic times - one can read for example here:

http://www.theopavlidis.com/MidEast/part10.htm

Jewish communities existed in Anatolia, Greece, Caucasus, Cyprus, Crete, North Africa, Italy, Crimea, etc. already long before 70 AD.

Of course those people were partially from the Diaspora and partially local converts.

Map showing the Jewish Diaspora from 719 BC (beginning of the Assyrian Captivity) to 70 AD (the destruction of Jerusalem), and later:

http://s12.postimg.org/omaza8osr/Jewish_Diaspora.png

Jewish_Diaspora.png


And some citations about Jews by ancient authors (including Josephus, who was himself Jewish):

attachment.php
 
About conversions of people to Judaism in Hellenistic times - one can read for example here:

http://www.theopavlidis.com/MidEast/part10.htm

Yes, I know, but in what numbers? How many adult men, in a time before anesthesia or antibiotics for infection, are going to choose to be circumcised? That would require an extraordinary commitment it seems to me. It doesn't mean it didn't happen, of course.

Women would be a different situation. A Jewish man in the Hellenistic period could take a "pagan" wife, and his children would still be Jews.

As I said, the IBD analyses being done show it didn't happen within the last 2,000 years, so if there was admixture in the prior period it seems it might have been whatever number of men were willing to be circumcised, and whatever women married into the community. That would explain the yDna distributions.


I think it could be argued that it was only when Christianity decided that men didn't need to undergo it to become Christians that the religion really started to spread.
 
BTW - regarding those citations posted above (such as the one by Seneca) - some Romans and Greeks, who were not very familiar with those issues, could not see any obvious distinction between Jews and early Christians, or at least between Jews and some of Christian sects. For example until the 4th century AD within Christianity there existed a sect known as Judeochristians - they worshipped Jesus Christ, but practiced circumcision, and some other Jewish customs.

As for circumcision - Jews weren't the only people in the Ancient Eastern Mediterranean who practiced that custom. But Greeks didn't.

It was a Semitic custom, but some non-Semites (Egyptians?) probably also practiced it. Not Greeks or Romans, though.

it was only when Christianity decided that men didn't need to undergo it to become Christians that the religion really started to spread.

Early Christianity was indeed spreading throughout the Empire mostly thanks to migrations of Christians - conversions played a secondary role.

It was a bit later when conversions became so numerous, that "Christians of the old stock" became only a small minority of all Christians.
 
That would explain the yDna distributions.

What do the latest data say about Jewish Y-DNA - how much of it is of ME and how much of EU origin?
 
 
 

This thread has been viewed 24793 times.

Back
Top