search this blog

Monday, October 15, 2018

ASHG 2018 open thread

The American Society of Human Genetics (ASHG) annual meetings kicks off tomorrow in San Diego. Feel free to post anything near and far related to this event in the comment thread below.

You can explore this year's offerings via the online planner/abstract search located HERE. See anything really interesting? Here's what I found after a quick search using the term "ancient". Hopefully someone tweets from the South Asian talk.

Mount Lebanon provides an opportunity to study DNA from the ancient Near East

Reconstructing the peopling of old world south Asia: From modern to ancient genomes

Tracing the evolution of pigmentation-associated variants in Europe

Intriguingly, the Mount Lebanon abstract says this:

In addition, we found steppe-like ancestry in the Roman Period individuals which we have previously detected in present-day Lebanese but not in Bronze Age individuals. This supports our previous proposition that the steppe ancestry penetrated the region more than 2,000 years ago, and genetic continuity in Lebanon is substantial.

So what are we dealing with here exactly: admixture from the Hittites, Mittani, and/or Romans? Who does the Global25 point to?

See also...

The South Asian cline that no longer exists


Davidski said...


Seven carefully selected Romaniote Jewish samples are now in the Global25 datasheets. All the links are HERE.

This is how they cluster in my Mediterranean PCA...



Razib Khan said...

i'll be asking niraj rai about the south asian stuff at his poster i think

claravallensis said...

This should be interesting too, given how poorly sampled Italy is in general, although it's not mainland.
"PgmNr 2719/T: Investigating Sardinian population history with ancient DNA"
"...To gain further insight, we analyze genome-wide capture data (~1.2 millions SNPs) of 26 ancient Sardinians spanning the Neolithic, Copper Age, and Bronze Age, including individuals from Sardinia's Nuragic culture. Merging this novel data with 998 previously studied aDNA samples from across Europe and throughout the last ten millennia, we are able to place the ancient Sardinian samples into the broader context of the peopling of Europe. We confirm that ancient Sardinian samples show a strong affinity to early Neolithic samples and a near complete absence of the “Steppe” ancestry associated with Bronze Age expansions on the mainland. Interestingly, we also detect elevated affinities with pre-Neolithic peoples of Europe...."

Aniasi said...

That would be amazing, especially his mention of museum samples

Sofia Aurora said...

Guys the ESHE abstracts for 2018 are available:

Along with the European Society of Human Evolution meeting another event came.

The Baltic-Pontic Studies journal is open for the year 2018 finally.
Please see below:

It has very important news about the Proto-Indoeuropeans both from an archaeological and genetics point of interest!

Synome said...

I would guess that Persians and Greeks are more likely candidates for a steppe source in Roman era Lebanese people.

Davidski said...


Persians and Greeks, especially pre-Slavic admixture Greeks, probably had about 15-20% of steppe ancestry, so even a few per cent of steppe ancestry in Roman era and modern Lebanese would require quite a bit of Persian and/or Greek admixture into the northern Levant. I'm not sure if that works?

So I think we're looking at something else here, maybe guys like this...

An early Iranian, obviously

Samuel Andrews said...

None of the ASHG links work for me.

Davidski said...

Seems like you can't view the abstracts now unless you have a password. Oh well, let's wait for the tweets.

Samuel Andrews said...

To me it seems Romaniote Jews are same as all European Jews minus minor northern European admixture in Ashkenazi and Iberian admixture in Sephardic. Do they confirm most Gentile ancestry in Jews is from Greece/Aegean?

PF said...

Thanks for the new samples Davidski. Well, this quick fit looks pretty damn good to me. :-D

[1] "distance%=0.9828"



epoch said...

I am hugely curious about the results of the Les Cottés sample.

Nirjhar007 said...


Will you add the polish data to G25?

Davidski said...


Yeah, when the genotype data become available. Currently the data are only available as BAM files.

Erikl86 said...

@Samuel - I believe it does, since Romaniote Jews do not have any North Italian ancestry, yet they still plot almost exactly as the other Western Jews. Those 7 Romaniote Jewish samples are from a very well isolated community so they also have virtually no Sephardic admixture.

@PF - thanks. Closest single item distances with ancient samples:

Mycenaean Levant_BA_North Anatolia_IA Armenia_MLBA
5.678238 6.174472 10.592026 10.620053


[1] "distance%=2.0563"



[1] "distance%=2.0581"



And you can also model Italian Jews as Romaniote Jews + North Italians, confirming the vast historical proof that the first Jews to arrive to Italy were Greek speaking Jews, most likely ancestors of modern day Romaniote Jews:

[1] "distance%=0.89"



Then Ashkenazi Jews:

[1] "distance%=1.3595"



And Sephardic Jews:

[1] "distance%=1.2399"



Judging by the fact that Romaniote Jews can be modeled as 60% Hellenic-like, ~40% Levant, and they are roughly 80-90% of other Western Jews, this indeed would suggest ~50% Hellenic-like, ~30-35% Levant-derived.

Samuel Andrews said...

Um. It matters whether European Jews have actual Greek ancestry or non-Greek Aegean ancestry. It would be interesting if turns out they're a Greek/Hebrew mix.

Erikl86 said...

@Samuel - I agree, which is why I've opened up a thread on this subject on Anthrogenica. I never like to jump to conclusions when there are still so many blanks to fill - but historical evidence as well as autosomal data seem to partially support this, at least from what we've reviewed so far in that thread, IMO.

I still think we need to find where Romaniote, Sephardic and/or Italian Jewish maternal lineages originate from. Ashkenazi Jews are not good to do that, as a result of that genetic bottleneck their ancestors have experienced.

And of course, get some ancient Judean Jewish samples, and perhaps even Hellenistic Jewish samples from one of those cemeteries along the East Mediterranean.

Samuel Andrews said...

@Aric L86,

mtDNA is my focus area. There's not a lot of Greek mtDNA data published. We'll probably have to wait till there is to see if there's shared lineages with European Jews. In the end, ancient DNA will find the answer.

PF said...

@Aric @Samuel

It's still really difficult to tell at this point whether the Aegean-like input entered prior to Judaism forming, with the common proposal being a post Bronze Age collapse migration of Aegeans to the Levant, or later, during actual Greek times.

If Samaritans are taken as representative of early Israelites, then it becomes more likely that the mixture happened in Classical times and outside the Levant. Samaritans appear very similar to Levant_BA_North and don't require anything Aegean-like to model them. Conversely, if we believe they were resettled from elsewhere at some point, then using them for that argument becomes moot.

It's also possible that the earliest Israelites were already somewhat ethnically structured (within the range of regional variation, from more Arabian-like to more Aegean-like). There's the evidence of distinct polities (Israel vs. Judah)... early assimilation of nearby various non-Israelites... it's even not unreasonable to think that there was variation at the individual tribe level.

The question is too complicated and generally there's too much mixing in the region to use Admixture / PCA data + nMonte to make sense of it. I think only a good sampling of direct aDNA or really high-res mtDNA analysis would do the trick, exactly as Aric suggested.

I saw the gargantuan thread on Anthrogenica recently -- it will require a rainy day to go through it all. I should check in though, as I've been a mild proponent of the Aegean+Levantine theory for years (especially since the Anatolia_Chl genome came out), but I'm not confident.

Though that Ashkenazis are nowhere near 50% Italian, that I am sure about. :)

Erikl86 said...

@PF ,

I completely agree with you - in that gargantuan thread on Anthrogenica which I've opened, I actually cover these two options. Of course, the main suspects of that hypothetical post BA collapse Aegean ADMIXTURE would be the Philistines, that Pentapolis (Ashdod, Ashkelon, Gath, Ekron and Gaza) was in adjacent to the ancient Kingdom of Judah and and even according to Jewish religion, seem to have somewhat close (albeit hostile) relationship with the nearby Judeans.

I think that at least the Philistine aDNA which we might get from the Leon-Levy Expedition near Ashkelon (the huge cemetery of ~150 Philistine remains) would help uncovering some more crucial information, for example finally showing the Aegean origin of Philistines (which I'm almost certain they were of) and / or showing whether or not they were mixing with the nearby Canaanites.

However I am for one, as I write in that thread, am more of a supporter of Hellenistic (or classical) admixing event, for there is a plethora of historical evidence and even archaeological evidence suggesting such mixing, which I and other also cover in that thread. It could be of course, both - some Philistine, then later on more substantial Hellenistic ADMIXTURE.

Anyone interested in this subject, I really suggest reading that thread, despite it being as of today ~380 pages long, and although some parts of that thread get a little daunting, it is IMO very informative.

One thing for sure - I think we did manage to prove there to a large degree that the 50% Italian / 50% Levantine is indeed incorrect.

Katharós said...

I think historians who are specialized on the history of early Christianity could possibly piece the data together. The so called sebomenoi +- prosêlutoi , were gentile sympathizers or converts to Hellenistic Judaism in the first centuries. They later play a major role in the splitting of the two groups, because these sebomenoi were targeted by early followers of Christ, namely Paul. By the way , if I think about Samaritans , I think of Simon Magnus the
Gnostic , which of course interjavascript:void(0)communicates with the Druze.

Erikl86 said...

You actually have in some synagogues in West Anatolia, such as in the ancient ruins of Sardis and the remains of the ancient synagogue of Aphrodisias, dated as late as the 4th century AD, you see names of the members of the communities with the Greek term theosebeis - "godfearers". These are gentiles who were sympathetic to Judaism, but didn't fully convert.

Also, many synagogues from late antiquity Galilee and Judea (4th-6th centuries AD) like the Hamat Tiberias and Bet Alpha synagogues (and the Naaran synagogue next to Jericho), show extensive Greco-Roman cultural influence such as mosaics with the zodiac, the Greco-pagan god Helios (in a synagogue !) and extensive inscriptions in Greek.

You also have the Jewish catacombs in Rome itself, showing how the original Jewish community of Rome was Greek speaking (thus arriving there from Greek-speaking regions - either Magna Graecia in South Italy, or directly from the East Mediterranean) and had Hellenized names which dominate the graves dated to 1st-4th centuries AD, then a clinal transition to Latin.

So there's a great deal of archaeological evidence supporting what historical sources are saying regarding the popularity of Judaism in the East Mediterranean Greco-Roman world.

Genetic uniparental evidence does show that the vast majority of these proselytes would have been women, as most paternal lineages in Ashkenazi, Sephardi etc. are of Near Eastern origin. This would also go along with the known historical deter from circumcision in the ancient Greco-Roman world, which so this as self mutilation, and destruction of the body. Also, circumcision would make it almost impossible for a person to compete in the Greco-Roman gymnasium, a daily routine for most men at the time that would be in complete nude.

The main problem is that the religious narrative that all Jewish diasporas hold, of direct lineage from Judean exiles after 70 AD, seem to not mention at any point the absorption of proselytes - which obviously regardless of where Western Jews (Ashkenazi, Sephardi, Romaniote etc. Jews) got it from, it's there, or they'd cluster with other Levantines instead of inside the East Mediterranean Continuum. So historically, it's hard to draw a direct line - most of the evidence is circumstantial, like for example the documented migration of the Jewish Kalonymos family from Lombardy, Italy to Maintz, Germany around the 9th century AD and becoming one of the leading Rabbinate families of early Ashkenazi Jewish communities there. We know Kalonymus is a Greek name - and it's actually a Greek translation of a very common Jewish last name "Shem Tov". Or for example the high similarity of Romaniote and Italian rites - showing that Italian liturgical rite most likely derive from Romaniote. Add to that the existence of Romaniote communities in Greek-speaking South Italy well after it was lost to the Byzantine Empire in the 11th century AD, and you could draw a potential scenario:

Hellenistic Jews mixed with Hellenic proselytes arrive to Magna Graecia -> Migrate North and mix with some Italian proselytes -> splitting into two branches - one moving West along the Mediterranean and settling in Iberia, while the other one moving North to France and Germany, later on moving to Poland. These would be Sephardic and Ashkenazi Jews.

And just to show how preposterous the often qpAdm model of Ashkenazi Jews as half Levantine, half North Italians, here are some nMonte ADMIXTURE modeling of Romaniote Jews, Sephardic Jews and Ashkenazi Jews:


[1] "distance%=1.6254"




[1] "distance%=2.0385"




[1] "distance%=1.9863"



Ric Hern said...

@ Sofia Aurora

Very interesting. If I understand it correctly it looks like Late Yamnaya in Western Ukraine was basically contemporary with Bell Beaker in the West...?

Ric Hern said...

Does this possibly prove that Central European Bell Beaker people arrived with an earlier expansion from the Steppe ?

capra internetensis said...

New, Large Sequence Panel Enables Population Genetics Research in Africa

"The researchers collected and analyzed genome-wide data from 6,400 individuals from rural Uganda, including the whole genome sequences of 1,978 people within that group. They combined this with data from an additional 7,784 individuals from other African countries, in the first genome-wide association study (GWAS) to analyze multiple traits across Africa.... They also found a complex mixture of similarities between the Ugandan genomes, ancient East African populations, and Eurasian genomes. These findings suggest that modern Ugandan genomes represent the complex history of migration in the region, and that they have been shaped by multiple mixture events over thousands of years, including mixture with genomes resembling those in modern Europe."

Holy shit, that's a damn fine start.

Anshuman said...

So nobody found anything worth reporting on Niraj Rai'S Poster

Samuel Andrews said...

"It's also possible that the earliest Israelites were already somewhat ethnically structured (within the range of regional variation, from more Arabian-like to more Aegean-like). "

But eastern Jews appear to have the same kind of Near Eastern pull that western Jews do. IMO, the most pure reference for western Jews are Greek Jews and the most pure reference for eastern Jews are Iraqi Jews.

Persian Jews=Iraqi Jews+Persian admixture.
Georgian Jews=Persian Jew+Georgian admixture.
European Jews=Greek Jews+fill in the blank.

Bronze age Lebanon/Israel genomes look like a good match for the common thread between Greek Jews & Iraqi Jews. Both look like 50% Lebanon BA+50% local stuff.

So, IMO Iraqi Jews discredit the idea that ancient Jews were diverse because the have the same kind of Near Eastern ancestry that western Jews do. But, historians & then finally ancient DNA will answer that question.

Davidski said...


The results from the Indian aDNA talk or poster may have been embargoed pending formal publication. In other words, maybe no Tweeting allowed.

But I don't know, because I wasn't there. And anyway, like I said recently, I'm generally over it, as in the whole South Asian genetics thing.