Go Back   Family Tree DNA Forums > Universal Lineage Testing (Autosomal DNA) > Family Finder Basics

Reply
 
Thread Tools Display Modes
  #1  
Old 9th October 2017, 12:39 AM
EndlessResearch EndlessResearch is offline
FTDNA Customer
 
Join Date: Oct 2017
Posts: 8
Jewish Matches Question

Hello, I'm sort of new to genealogy and I just had a question I'm hoping someone can shed some light on.

I recently took a DNA test and uploaded the results to various services such as GEDMatch. On GEDMatch, it displays about 2000 matches, but all of them are not very close - the highest amount of shared cM I even have with a person is 39. However, I was surprised to see that some of my matches appear to be Jewish. The reason this was surprising is because I'm unaware of any Jewish ancestry - my ancestors as far as I know were Catholics and some Protestants from Central Europe. I'm not 100% sure of how many Jewish matches I have, but I estimate about 15-25 (but again, not entirely sure, as I wasn't able to look through all my matches). Some of them appear to be distant cousins of each other. I believe the highest amount of shared cMs I have shared with one individual is about 16. The rest range mostly from 8-11.

I'm curious if this connection is either 'noise', from a shared Jewish ancestor, or from a shared Non-Jewish ancestor. I would really appreciate any insight I can get on this! Thanks so much!
Reply With Quote
  #2  
Old 9th October 2017, 03:02 AM
dna dna is offline
FTDNA Customer
 
Join Date: Aug 2014
Posts: 2,653
You have to think about DNA and history of those lands in more general terms.

Yes, it is quite possible that some or all the matches are just a noise.

However, you had entirely overlooked the following two scenarios:
  1. somebody Jewish became a Christian (very many examples in Poland from centuries ago until 1945) and they are one of your ancestors;
  2. a Christian from your family converted to Judaism and is one of ancestors of your Jewish matches.
The tricky part is that in the 1. case that Jewish person did not need to be an ancestor of your Jewish matches, and similarly in the 2. case that Christian person did not need to be your ancestor. How is that possible? In either case, it could have been a sibling of an ancestor.


Mr. W

P.S.
In the 1. scenario, when centuries ago entire families converted to Christianity, after their conversion they often intermarried within their own group for many generations. Consequently, their Jewish DNA was not getting diluted.
Reply With Quote
  #3  
Old 9th October 2017, 07:11 AM
EndlessResearch EndlessResearch is offline
FTDNA Customer
 
Join Date: Oct 2017
Posts: 8
Thank you for your insight! I did consider that there was a conversion, but I did indeed overlook the possibility that one of my ancestor's siblings was a convert. But like you said it seems to be very difficult to be able to determine which of the different scenarios is the correct one. Is there any test or technique that can help maybe narrow it down?
Reply With Quote
  #4  
Old 9th October 2017, 10:06 AM
khazaria khazaria is offline
FTDNA Customer
 
Join Date: May 2014
Posts: 435
I talked about this before. Yes, there is a technique called segment painting that sometimes suggests the type of common ancestor involved.

Meanwhile, extensive triangulation (and, where possible, phasing) are necessary to prove the matches are not false.
Reply With Quote
  #5  
Old 9th October 2017, 11:42 AM
khazaria khazaria is offline
FTDNA Customer
 
Join Date: May 2014
Posts: 435
Occasionally we get lucky on the types of matches we find to a triangulating cluster and it leaves no room for doubt about the type of ancestor.

In July, I explored the DNA of a Polish person who was suggested to have some "Ashkenazi Jewish" DNA by 23andMe, but whose family only knew of many generations of Polish ancestors. I found 5 autosomal DNA segments where this person matches clusters of people whose recent ancestry is fully Ashkenazic Jewish, and many or most of them show no signals of recent Slavic admixture, and some of the segments painted as "East Mediterranean" and "West Asian".

On one of the segments, which was painted "West Mediterranean", in addition to many Ashkenazic matches, there was a New Mexican Hispano match with no Slavic background and no Ashkenazic background. The ultimate ancestor of this segment had to have been Sephardic Jewish from Spain or Portugal rather than Ashkenazic Jewish from Central or Eastern Europe. It is impossible that the common ancestor of that segment was Polish.

But a segment this obvious is rare, since most Jewish segments in Polish people are of Ashkenazic origin and therefore don't have matches from isolated Latin American and Iberian communities.
Reply With Quote
  #6  
Old 9th October 2017, 09:50 PM
EndlessResearch EndlessResearch is offline
FTDNA Customer
 
Join Date: Oct 2017
Posts: 8
Thank you for your information, Khazaria, your knowledge appears to be very extensive. I'm definitely interested in trying to work out of these matches are related from a Non-Jewish or Jewish ancestor. Could you perhaps explain how one would 'triangulate' DNA segments to accomplish this?

Another thing, I may be misreading, but you mentioned that primary Ashkenazi components are East Med and West Asian. On the Eurogenes K15 test I get 0% East Med scores, and only a bit of West Asian (I don't know the exact number because I'm a different computer and I don' remember my GEDmatch number at the moment). Would these results be able to help narrow it down? I know you mentioned that West Mediterranean was also a component in Sephardic ancestry, but since none of my ancestors came from Spain or any region where Sephardic peoples lived in high numbers I'm not sure if I should disregard this step or not.

Last edited by EndlessResearch; 9th October 2017 at 09:52 PM.
Reply With Quote
  #7  
Old 10th October 2017, 03:15 AM
dna dna is offline
FTDNA Customer
 
Join Date: Aug 2014
Posts: 2,653
Quote:
Originally Posted by EndlessResearch View Post
Thank you for your information, Khazaria, your knowledge appears to be very extensive. I'm definitely interested in trying to work out of these matches are related from a Non-Jewish or Jewish ancestor. Could you perhaps explain how one would 'triangulate' DNA segments to accomplish this?

Another thing, I may be misreading, but you mentioned that primary Ashkenazi components are East Med and West Asian. On the Eurogenes K15 test I get 0% East Med scores, and only a bit of West Asian (I don't know the exact number because I'm a different computer and I don' remember my GEDmatch number at the moment). Would these results be able to help narrow it down? I know you mentioned that West Mediterranean was also a component in Sephardic ancestry, but since none of my ancestors came from Spain or any region where Sephardic peoples lived in high numbers I'm not sure if I should disregard this step or not.
Many Sefardim did migrate to Poland, however unlike in the other places this time they were absorbed. Please remember that the Polish–Lithuanian Commonwealth was the place with the largest European (and most likely the world) population of Jews.

Why is the above rarely mentioned? I could offer an explanation, but I think it would be out of scope of the FTDNA forum.

Oh, Sefardim also migrated to Prague (Bohemia), but all Jews were expelled from Prague a couple of decades later, so it is difficult to state whether Sefardim population was absorbed there or in the places they went to after the expulsion.

Let me return to the conversions in Central Europe of the 16th and 17th century, as they are a real problem in the genealogical research.
  • Unlike Spain or Portugal at that time, Poland was not a place with hundreds of thousands fast conversions. Even if a conversion could be suspected, there is no way to trace the original Jewish family, unless some other source (say a city chronicle) has all the info. Only new Christian names were recorded in the baptismal books. Etc.
  • If a Christian man converted to Judaism, that was done in utmost secrecy, as whatever freedoms were available, they had never included a conversion to Judaism. For example, of what we know!, in Poland such a man would assume a new Jewish identity and travel the Kingdom of the Netherlands. Essentially untraceable.

Mr. W
Reply With Quote
  #8  
Old 10th October 2017, 07:38 PM
khazaria khazaria is offline
FTDNA Customer
 
Join Date: May 2014
Posts: 435
Quote:
Originally Posted by EndlessResearch View Post
I'm definitely interested in trying to work out of these matches are related from a Non-Jewish or Jewish ancestor. Could you perhaps explain how one would 'triangulate' DNA segments to accomplish this?
As Jim Bartlett wrote at segmentology.org, genetic triangulation is a "mechanical process" whereby we compare all members of a segment area with each other to confirm that they are all related from a common ancestor. 4 full rounds of comparison are enough, that is, choose yourself and 3 other people belonging to the segment and compare those 4 people to all the other members, and stack the One-to-One comparisons that involve them for easy visualization. They all must overlap in the same general area of the same chromosome number.

Paintings of the segment area should be looked at only after this verification is successful.

Checking the ethnic profiles of many of the matches should also be confirmed after successful verification.

Quote:
Another thing, I may be misreading, but you mentioned that primary Ashkenazi components are East Med and West Asian. [...] I know you mentioned that West Mediterranean was also a component in Sephardic ancestry
They are very common on Ashkenazic segments, and I tend to use Eurogenes' old version of EUtest for paintings. In addition to the above 3, a fourth very common element is called "Middle Eastern" and has a more Arabian concentration whereas "East Med" is more centered on the Levant region while "West Asian" is more about the northern Middle East.

"Atlantic", "South Baltic", "North-Central European", and "East European" elements are also somewhat common in Ashkenazic Jews, and some even have "Siberian" and "East Asian" DNA.
Reply With Quote
  #9  
Old 10th October 2017, 08:51 PM
EndlessResearch EndlessResearch is offline
FTDNA Customer
 
Join Date: Oct 2017
Posts: 8
Thanks for your interest guys. I've still been very interested in this question and I've tried looking into it a little bit more. Granted, I'm still an amateur however so I'm not entirely sure if what I'm reading is right.

On a Family Tree DNA FAQ, in regards to a question about Jewish ancestry, one of the answers was that if you have Jewish descent then you will have a lot of Jewish matches on things like Family Finder, GEDMatch, etc. While I do have some Jewish matches like I said, the grand majority of my matches all appear to be European Non-Jews. I don't know if this gives any indication towards the answer to my question.

Also, khazaria, I recently learned that I could run matches GEDMatch numbers through various calculators to look at their composition. All the Jewish matches I looked at had varying amount of Atlantic, Baltic, North Sea, etc. components in their admixture, like you said. These are all components that are the primary make up of my own admixture. Could these "European" admixtures be what is part of our shared DNA?

As a final question, what amount of shared centimorgans is usually considered to be a good chance of a real match, and not one that is just by noise?
Reply With Quote
  #10  
Old 10th October 2017, 10:10 PM
khazaria khazaria is offline
FTDNA Customer
 
Join Date: May 2014
Posts: 435
Quote:
Originally Posted by EndlessResearch View Post
All the Jewish matches I looked at had varying amount of Atlantic, Baltic, North Sea, etc. components in their admixture, like you said. These are all components that are the primary make up of my own admixture. Could these "European" admixtures be what is part of our shared DNA?
A Jewish segment can have 1 ethnic/regional element or more than 1 element.
The element(s) can have Middle Eastern or European tendencies or both.

Quote:
Originally Posted by EndlessResearch View Post
As a final question, what amount of shared centimorgans is usually considered to be a good chance of a real match, and not one that is just by noise?
Extensive enough triangulation, especially in combination with parent-child phasing, cuts out all the noise.

But, to answer your question, almost all segments above 15 cM are real, most above 10 cM are real, and like half of 7 cM segments are real.
Reply With Quote
Reply

Bookmarks


Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Question about Jewish matches in Non-Jewish Person EndlessResearch Family Finder Basics 2 9th October 2017 03:03 AM
A lot of Jewish matches - but I don't have Jewish ancestors (as far as I know). Dora myOrigins Basics 92 11th March 2016 11:54 AM
A Jewish Question rachelleleclaire Family Finder Advanced Topics 0 16th January 2015 01:14 PM
Jewish Haplogroups Question & Possible Hidden German-Jewish heritage? ashimmin mtDNA - Advanced Topics 6 4th September 2012 11:19 AM
Question on Jewish lineage need help to know goodheart DNA and Genealogy for Beginners 5 15th January 2006 05:09 PM


All times are GMT -5. The time now is 09:20 AM.


Family Tree DNA - World Headquarters

1445 North Loop West, Suite 820
Houston, Texas 77008, USA

Phone: (713) 868-1438 | Fax: (832) 201-7147
Copyright 2001-2010 Genealogy by Genetics, Ltd.
Powered by vBulletin® Version 3.8.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.