PDA

View Full Version : How much DNA do two randomly picked Ashkenazi people share?



Riverman
04-06-2021, 10:04 AM
I recently came across the question of genealogically not recently related Ashkenazi people still sharing a significant amount of DNA at a level which would suggest a closer genealogical relationship in non-endogamous populations. In a quick search I found this comment:

But for Ashkenazi Jews, heritage is pretty clear. Pick a street, Feldman said. Then pick any two Ashkenazi Jews at random walking down it.
“They’d be fifth to ninth cousins at the genetic level,” Feldman said. Ashkenazi Jews are actually that closely related, all descended from a small group of people.

https://www.jweekly.com/2019/11/29/who-is-a-jew-dna-home-testing-trend-adds-new-wrinkle-to-age-old-debate/

Basically does this mean that a fully Ashkenazi person should share significant amounts of DNA with (almost) every randomly picked other Ashkenazi Jewish person? What are the odds for segments above 7 or even 10 cM being shared by randomly picked Ashkenazi people? Like if comparing one person with 100 other, randomly picked Ashkenazi people, what percentage of these 100 individuals would share segments with more than 10 cM of DNA?
Are there any statistics out there on the topic?

There is of course this comparison:
https://larasgenealogy.blogspot.com/2019/09/ashkenazic-shared-dna-survey-september.html

Which however doesn't go far enough and would rather suggest a lower threshold for very distantly, genealogically not recently related Ashkenazi people.

jkotl0327
04-06-2021, 01:10 PM
I recently came across the question of genealogically not recently related Ashkenazi people still sharing a significant amount of DNA at a level which would suggest a closer genealogical relationship in non-endogamous populations. In a quick search I found this comment:


https://www.jweekly.com/2019/11/29/who-is-a-jew-dna-home-testing-trend-adds-new-wrinkle-to-age-old-debate/

Basically does this mean that a fully Ashkenazi person should share significant amounts of DNA with (almost) every randomly picked other Ashkenazi Jewish person? What are the odds for segments above 7 or even 10 cM being shared by randomly picked Ashkenazi people? Like if comparing one person with 100 other, randomly picked Ashkenazi people, what percentage of these 100 individuals would share segments with more than 10 cM of DNA?
Are there any statistics out there on the topic?

There is of course this comparison:
https://larasgenealogy.blogspot.com/2019/09/ashkenazic-shared-dna-survey-september.html

Which however doesn't go far enough and would rather suggest a lower threshold for very distantly, genealogically not recently related Ashkenazi people.

I can speak from my experience. It does depend on the regions the people hail from, many Ashkenazim nowadays are mixed between German, Ukrainian, Lithuanian etc Jews and will match many other AJs, but generally when you take a Lithuanian Jew and a French AJ there is a smaller chance that they will match than two Lithuanian Jews. At the same time a Lithuanian and a French Jew could easily match while two Lithuanian Jews do not match. Almost all of my ancestry happens to hail from one particular region, and I have connections I've made with people, all of whose ancestry also hails from the same region, and we didn't share any DNA, at least by 23andme standards. Based on what I've seen, the average amount of DNA shared by Ashkenazim from the same region is around 30cM, and the average for Ashkenazim from different regions is around 10 cM. It will also certainly depend on what region we are talking about, my ancestral region, Podolia in Ukraine, underwent a twentyfold Jewish population boom during the 19th century alone, with relatively little immigration from other regions to bolster that, so two Podolian Jews are likely to share more DNA than two Khersonian Jews, because the Kherson region's Jews were mostly descendants of migrants from many different areas in Eastern Europe.

Riverman
04-07-2021, 01:45 PM
I can speak from my experience. It does depend on the regions the people hail from, many Ashkenazim nowadays are mixed between German, Ukrainian, Lithuanian etc Jews and will match many other AJs, but generally when you take a Lithuanian Jew and a French AJ there is a smaller chance that they will match than two Lithuanian Jews. At the same time a Lithuanian and a French Jew could easily match while two Lithuanian Jews do not match. Almost all of my ancestry happens to hail from one particular region, and I have connections I've made with people, all of whose ancestry also hails from the same region, and we didn't share any DNA, at least by 23andme standards. Based on what I've seen, the average amount of DNA shared by Ashkenazim from the same region is around 30cM, and the average for Ashkenazim from different regions is around 10 cM. It will also certainly depend on what region we are talking about, my ancestral region, Podolia in Ukraine, underwent a twentyfold Jewish population boom during the 19th century alone, with relatively little immigration from other regions to bolster that, so two Podolian Jews are likely to share more DNA than two Khersonian Jews, because the Kherson region's Jews were mostly descendants of migrants from many different areas in Eastern Europe.

Thank you for that response, seems very reasonable to me.
An additional question arising from the first one is how the relationship between two modes of homogenisation within the Ashkenazi Jewish population is, namely that of descending from a few founders and constant gene flow between the various subpopulations. Because I think both played a role, but I'm not entirely sure about the importance of either. This results in the next question, namely whether a lot of the distant Ashkenazi cousins can be attributed to these founders and very common segments, and whether there are many uncommon, probably more regional and marriage network specific, much rarer ones too.
One way to solve this is, if anyone of the readers is fully Ashkenazi, especially those from a smaller, not that mobile community, whether he or she is sharing primarily the same segments on a given chromosome over and over again, or whether the complete genome is filled with and covered by all kind of matches.

Because I always wondered about the more recent admixture events or more region-specific communities, and how these would contribute to the overall genetic variation in the Ashkenazi population as a whole, because this variation shoudln't be distributed evenly, contrary to the the founder segments. My hunch, but this might be wrong, is that there should be quite a differentiation between very common founder segments and less frequent, more region-specific ones. But probably that's wrong because of the panmixture which happened in modernity and there is little of these more specific genetic profiles left?