PDA

View Full Version : Questions having read the paper on Levite origins



genealogist_jewish
02-06-2021, 05:38 PM
I have read the paper: "Phylogenetic applications of whole Y-chromosome sequences and the Near Eastern origin of Ashkenazi Levites (https://www.nature.com/articles/ncomms3928)". It is fascinating. I have a question though, that I would appreciate some help with.


The introduction states that "While our survey of one of these, M582, in 2,834 R1a samples reveals its absence in 922 Eastern Europeans, we show it is present in all sampled R1a Ashkenazi Levites
". Page 5 also says that "all Ashkenazi Levite R1a samples belonged to haplogroup R1a-M582". However, table 2 on page 5 indicates that M582 is present in 64.9% of R1a Ashkenazi Levites. Which one is true?

Thank you. I have zero education in genetics, so please forgive me for misunderstanding.

Table two:

43096

Megalophias
02-06-2021, 05:44 PM
The table shows that 63/97 Ashkenazi Levites (64.9%) had R1a-M198, and that all 63 of these men had R1a-M582. It is saying that 65% of all the Ashkenazi Levites had M582, not that 65% of the Ashkenazi Levites with R1a had M582.

genealogist_jewish
02-06-2021, 06:13 PM
Ahhhh. Thank you so much.

genealogist_jewish
02-06-2021, 06:16 PM
The table shows that 63/97 Ashkenazi Levites (64.9%) had R1a-M198, and that all 63 of these men had R1a-M582. It is saying that 65% of all the Ashkenazi Levites had M582, not that 65% of the Ashkenazi Levites with R1a had M582.


Strange though, as a previous study has said the frequency overall is 52% https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1180600/pdf/AJHGv73p768.pdf

Why was the sample from the 2017 sample so much higher?

eolien
02-07-2021, 01:38 PM
Might be of Iranian (non-jewish) origin? Not surprising in view of jewish history.

genealogist_jewish
02-07-2021, 11:15 PM
R1a Ashkenazi Levites?

Coldmountains
02-07-2021, 11:26 PM
Might be of Iranian (non-jewish) origin? Not surprising in view of jewish history.

It is definitely of Indo-Iranian origin and closest to R1a in modern day West Iran. That makes Ashkenazi Levites one of richest groups in Proto-Indo-Iranian paternal markers (founder effect)

CopperAxe
02-07-2021, 11:40 PM
I think it somehow has a relation to the second temple and how it came to be, especially if you consider the origin of the Levites. Also the pylogeny of the subclade points toward a West Asian (Iranian) origin.

A conduit for potential Zoroastrian influences perhaps?

Coldmountains
02-11-2021, 10:49 PM
I think it somehow has a relation to the second temple and how it came to be, especially if you consider the origin of the Levites. Also the pylogeny of the subclade points toward a West Asian (Iranian) origin.

A conduit for potential Zoroastrian influences perhaps?

Jewish R-Z2122>R-F1345R>R-CTS6>R-Y2619 is a fascinating clade and i was wondering if it is ultimately Maryannu (Mitanni Indo-Aryan) related. It formed according to Yfull 2800 ybp and has closest relatives under CTS6 in northwest Iran, Mesopotamia and Armenia. CTS6 itself formed according to Yfull 3700 ybp. So way or another it probably entered Mesopotamia and ultimately Israel via NW Iran. Afaik it was not found in ancient dna yet but F1019 also under F1345.


I found rather rare deeper subclade for two more samples from Damgaard paper which were also labeled as males, but didn't have Y-haplogroup. DA229 has less reliable calls (transitions covered with only one read), but neither of them has contradicting calls at upstream levels.


DA229; 250 BC-250 AD; Kaynar Bulak 2, Kazakhstan; Kangju; R1a1a1b2a-Z93>Z94>Z2124>Z2122>F1345>F2935>F1019* (xY37903) (https://www.yfull.com/tree/R-F1019/)


Z2122 level: Z2122/S4576+ G>A (1A)

Z2122>F1345 level: *no calls*

Z2122>F1345>F2935 level: *no calls*

Z2122>F1345>F2935>F1019 level: F1019+ C>T (1T)

Z2122>F1345>F2935>F1019>Y37903 level: Y37909- A>G (1A); Y37910- A>C (1A); Y37905- C>G (2C); Y37916- C>A (1C); Y37917- G>C (2G)



DA52; 403-434 AD; Keden, Kyrgyzstan; Tian_Shan_Hun; R1a1a1b2a-Z93>Z94>Z2124>Z2122>F1345>F2935>F1019* (xY37903) (https://www.yfull.com/tree/R-F1019/)


Z2122 level: *no calls*

Z2122>F1345 level: F1345/Z2472/FGC99+ C>A (1A)

Z2122>F1345>F2935 level: YP5501+ C>T (1T)

Z2122>F1345>F2935>F1019 level: F1019+ C>T (2T)

Z2122>F1345>F2935>F1019>Y37903 level: Y37903- C>T (1C); Y38010- A>C (1A); Y37911- G>A (1G)


R1548, Monterotondo 27 BCE-300 CE (Imperial Rome), belongs to R1a-Z93>F1345>F1019 https://www.yfull.com/tree/R-F1019/

Considering that F1345>F2935>F1019 was found among Kangju, Tian_Shan_Hun and is today found in China, the Caucasus and Turkey i don' think this parallel clade is Maryannu/Indo-Aryan-related and probably of (Proto-)Iranic origin. The common ancestor of R-F1019 and CTS6 lived according to Yfull around 3700 ybp too, but we can expect Yfull to underestimate this age by 10-20%. So likely in reality somewhere around 2000-2500 B.C if not slightly earlier. This would be either around the split of Proto-Indo-Iranians or slightly before, so F1019 could be part of Proto-Iranics and CTS6 part of a Proto-Indo-Aryans theoretically.

But on the other side, i am not aware of any South Asians or other modern-day Indo-Aryans under F1345 subclades, what again would speak against a Mitanni/Maryannu origin of this clade. But neither are classical Indo-Aryan clades under R1a-Z2123 and R1a-L657 found in modern-day Palestine, Israel or Syria among Non-Muslims, so there is a possibility that Maryannu Indo-Aryans of the Near East were under different R1a clades than that of modern day Indo-Aryans or Indo-Aryan Z2123 and L657 simply died out in the Near East, what i think is also very possible considering their small numbers and quickly assimilation into local cultures. So i would not exclude a Maryannu/Mitanni Indo-Aryan origin for modern-day CTS6 and Y2619 , but so far an Iranic and slightly later origin seems to be more likely in my eyes, but this could turn out wrong of course an i hope it will be found among ancient Maryannu and Mitanni Indo-Aryans (sadly the genomes of the Megiddo outliners were of too low quality to check the R1a clades)

CopperAxe
02-12-2021, 09:50 AM
Jewish R-Z2122>R-F1345R>R-CTS6>R-Y2619 is a fascinating clade and i was wondering if it is ultimately Maryannu (Mitanni Indo-Aryan) related. It formed according to Yfull 2800 ybp and has closest relatives under CTS6 in northwest Iran, Mesopotamia and Armenia. CTS6 itself formed according to Yfull 3700 ybp. So way or another it probably entered Mesopotamia and ultimately Israel via NW Iran. Afaik it was not found in ancient dna yet but F1019 also under F1345.





Considering that F1345>F2935>F1019 was found among Kangju, Tian_Shan_Hun and is today found in China, the Caucasus and Turkey i don' think this parallel clade is Maryannu/Indo-Aryan-related and probably of (Proto-)Iranic origin. The common ancestor of R-F1019 and CTS6 lived according to Yfull around 3700 ybp too, but we can expect Yfull to underestimate this age by 10-20%. So likely in reality somewhere around 2000-2500 B.C if not slightly earlier. This would be either around the split of Proto-Indo-Iranians or slightly before, so F1019 could be part of Proto-Iranics and CTS6 part of a Proto-Indo-Aryans theoretically.

But on the other side, i am not aware of any South Asians or other modern-day Indo-Aryans under F1345 subclades, what again would speak against a Mitanni/Maryannu origin of this clade. But neither are classical Indo-Aryan clades under R1a-Z2123 and R1a-L657 found in modern-day Palestine, Israel or Syria among Non-Muslims, so there is a possibility that Maryannu Indo-Aryans of the Near East were under different R1a clades than that of modern day Indo-Aryans or Indo-Aryan Z2123 and L657 simply died out in the Near East, what i think is also very possible considering their small numbers and quickly assimilation into local cultures. So i would not exclude a Maryannu/Mitanni Indo-Aryan origin for modern-day CTS6 and Y2619 , but so far an Iranic and slightly later origin seems to be more likely in my eyes, but this could turn out wrong of course an i hope it will be found among ancient Maryannu and Mitanni Indo-Aryans (sadly the genomes of the Megiddo outliners were of too low quality to check the R1a clades)

I'm not sure if the Mitanni of the middle bronze age are the answer here. I think we would see a more varied distribution of said haplogroup if it entered the Canaanite genepool around 1500 bc.

The Levites specifically have paternal descent from the priestly classes of the Second Temple, which was constructed after they were set free by Cyrus the Great. Some theologists also argue that this is the period where zoroastrianism had an influence on Judaism.

The relatively young MRCAs, the strong distribution amongst priestly classes and the western Iranian origin based on pylogeny make me think that this is the conext of Y2619 amongst Jewish Levites.

Also I definitely do not think that even if the Mitanni were Indo-Aryan proper, instead of another Indo-Iranian branch hard to classify due to a severe lack of proper textual evidence, that they must've been rich in Y3 or L657.

We still havent figured out if the Indo-Aryan affinities with those haplogroups are the result of bottlenecking amongst Proto-Indo-Aryans, a bottleneck amongst the Indo-Aryans which migrated into South Asia, or a bottleneck which occurred after the initial migrations into South Asia.

One interesting feature about the Megiddo outliers is that they show the same type of mixing processes you see with the modern day Ror. Same amount of native ancestry, same amount of steppe ancestry, just two different central asian substrates.

43235

Coldmountains
02-12-2021, 10:08 AM
I'm not sure if the Mitanni of the middle bronze age are the answer here. I think we would see a more varied distribution of said haplogroup if it entered the Canaanite genepool around 1500 bc.

The Levites specifically have paternal descent from the priestly classes of the Second Temple, which was constructed after they were set free by Cyrus the Great. Some theologists also argue that this is the period where zoroastrianism had an influence on Judaism.

The relatively young MRCAs, the strong distribution amongst priestly classes and the western Iranian origin based on pylogeny make me think that this is the conext of Y2619 amongst Jewish Levites.

Also I definitely do not think that even if the Mitanni were Indo-Aryan proper, instead of another Indo-Iranian branch hard to classify due to a severe lack of proper textual evidence, that they must've been rich in Y3 or L657.

We still havent figured out if the Indo-Aryan affinities with those haplogroups are the result of bottlenecking amongst Proto-Indo-Aryans, a bottleneck amongst the Indo-Aryans which migrated into South Asia, or a bottleneck which occurred after the initial migrations into South Asia.

One interesting feature about the Megiddo outliers is that they show the same type of mixing processes you see with the modern day Ror. Same amount of native ancestry, same amount of steppe ancestry, just two different central asian substrates.

43235

Considering that L657 is only present among modern day Indo-Aryans or regions with known Indo-Aryan substrates since the Bronze Age, was not found among any ancient Iranic samples from Ukraine to the Altai it unlikely played a significant role among Proto-Iranics if any and unlikely was not already significant among Proto-Indo-Aryans. L657 and its nearest downstream branches (Y6, Y7, Y9, Y920) being old clades forming likely around or even before the split of PIIs, L657 being so clearly and strongly associated with modern-day Indo-Aryan and dissociated with ancient Iranics makes it in my eyes central for the spread of Proto-Indo-Aryans.