Family Finder - The Big Picture? #dna


Martin Davis (com)
 

I received below a helpful clarification >from Max Blankfeld at FTDNA which I
am sure others would wish to see.

Martin

Martin Davis - London (UK)

---
From: Max - Family Tree DNA [mailto:max@familytreedna.com]

Martin, this is what I sent to Steven Bloom, who had similar questions. You
are welcome to post it to the list:

Here are the main problems that Jews have in trying to identify common
ancestors - as you certainly know?
a) lack of a common surname prior to late 1700?s early 1800?s
b) inbreeding population
c) many unrelated lines adopting same surnames
d) many related lines adopting different surnames
e) many surnames being adapted to the land where descendants moved in

To put it in simple terms it makes the work for us, Jews, a little more
complicated as we cannot trace people as easily as the others would be able
to. Our trees in general are much shorter. It's harder for us to focus on a
certain path/line. To do that we need more people to be tested, and combine
more with Y, mtDNA and X to help identify lines given the combined results
from all tests.
So, in terms of a Family Finder result, a 3rd cousin match, given the
inbreeding, may show a total value of centimorgans composed >from the
combination of different lines due to that inbreeding, and thus, what the
algorithm guesses to be a 3rd cousin, may in fact be a more distant cousin.
Example: I had a case of a person that matched with my nephew (my brother?s
son) as a 2nd cousin, and with me as a 4th cousin. If we were talking about
matching with just one line, he should be 3rd, and I 4th. But because he may
be adding blocks of DNA >from his maternal line (unrelated to me) to the
relationship with that person, those blocks, adding up to the my main block
with him, elevates by one generation his matching to that person.
I hope this was not too confusing.
Therefore, the names that don't make sense to you are there as a match
because of inbreeding, which makes them look more closely related than they
really are, and therefore unknown to us.
Another example >from my own matches related to change of names: I have one
match predicted as 4th cousin with the last name Rubio - typical Hispanic
name (I also had the first and middle Hispanic names). My first reaction was,
what is this name doing there?? Well, I checked his record and there's a
note there: his ancestral name was Rubizewsky? well, I checked JewishGen and
I saw 2 records >from a very small town in Belarus with the name Pinsk. And
guess where my mother is from? Yes, you guessed it right: Pinsk!
So, this is a most probable case of inbreeding, combined with a change,
generations later, to a Hispanic name.
In short, while you should not dismiss those "make no sense matches", you
should certainly weight in all the factors I mentioned above.
As we gather more Jewish samples we will be able to adjust the algorithm to
reflect all those factors.
In the meantime, I can tell you that we've already had several success
stories coming out of the Family Finder test.

E-mail me anytime!

Max Blankfeld
Vice-President, Operations and Marketing
http://www.FamilyTreeDNA.com

Join main@groups.jewishgen.org to automatically receive all group messages.