Re: Autosomal DNA Information for European Jews which is Not Being Revealed #dna

Adam Turner

I have found numerous cases of this - people with ancestors from the same ancestral towns (or very close by), but who I have so far been unable to trace.

But I don't think it's really because there's meaningful information about geographic regions that the testing companies aren't making available to us. The "regions" that each company has created are just a function of various SNPs that are atypically common in particular populations. Perhaps the companies could make the regions a bit more granular than they are already, but that doesn't really tell us very much that a little research on a given match won't turn up anyway. It's interesting to know that a match came from the same sub-population as I did, but it's not particularly useful in a genealogical sense.

Instead, I suspect these cases of same town/significant match/no known connection fall into one of the following categories:

  • These people are actually your cousins in the 5th-7th cousin range, and the connection is simply too far back to trace with the available sources. If you were born in the 1950s, assuming about a 25-year age gap between generations, your fifth cousins will stem from branches that diverged from yours around the 1820s - right around the cusp of the era of surname adoption. Although autosomal testing will not register many (even most) of your 6th cousins as matches, this is counterbalanced by the fact that each generation you go back, you have an exponentially greater number of cousins - increasing the likelihood that you'll come across one with a DNA test.
  • The same thing as above, except even further back - people in the 8th-10th cousin range or even more distant. Since you have a huge number of cousins in this range, at least a handful of them will be outlier cases that share unusually high amounts of DNA with you, relative to most cousins in this range.
(The most distant of these outlier cases who I've come across so far on AncestryDNA is a probable 5C1R of mine. After being run through TIMBER and whatever else Ancestry uses to do its thing, the system gives her match numbers as a total match of 47 cM across 7 segments, unweighted shared DNA 70 cM, longest matching segment 12 cM. Generally, I would interpret a 47 cM match on AncestryDNA as a signal of a possible third or fourth cousin and a 12 cM longest block as too short to be anything meaningful, and write this match off as probably just endogamy - but subsequent research suggests that she is likely an outlier case with an unusually high amount of matching DNA for such a distant relationship: she is descended from a man born in the 1810s who I suspect was my gggg-grandfather's brother, and AncestryDNA is somehow registering a significant total match for her...even though it registers no match between me and her father, who also tested, and who is my probable 4C2R.)

Adam Turner

Join to automatically receive all group messages.