DNA and Endogamy: What to expect?

Question

I am working on a historically endogamous region of Southern Italy (Caulonia, Reggio Calabria). The DNA results are sometimes confusing, and I'm trying to get more information on what to expect.

One problem I have run into is that even when we get a close match in DNA terms, it seems harder to find the MRCA than I would expect...in one case I'm working on now, in terms of the paper trail, it's really looking like there isn't one within the last 7 generations -- but that defies logic, since the amount of DNA shared puts these two people much closer than that -- a predicted 3.2 generations to MRCA.

 

I'm a little unsure how to even ask this question, but it seems like with an endogamous population, it would make sense that the apparent strength of this relationship might be deceptive. Perhaps the connection is actually weaker, but it looks strong because they share large segments from several different common ancestors.

Does that question even make sense? I'll be happy to provide examples of what I'm seeing to support my question, I am just trying to figure out if I'm even thinking in the right direction. If anyone has articles or videos specifically about DNA in endogamous populations, I would be very interested. I haven't yet found anything that answers my budding questions.

Thank you!

Answer 1

Hi, Marci. Here's a topic from a few days ago that's applicable: https://www.wikitree.com/g2g/618239/dna-results-for-collapsed-pedigree. (If you want some of the math mentioned, PM me from my profile page and I can send you an MS Word document; I haven't gotten around to editing the Wikipedia page yet.) 

Here's a YouTube-archived webinar from FTDNA a couple of years ago that discusses, in part, addressing DNA in the endogamous Ashkenazi populations...though the almost two-hour webinar talks about yDNA and mtDNA as well as autosomal DNA: https://www.youtube.com/watch?v=BCjQnOONT8I&t=144s.

There's a link in that previous G2G thread to Jim Bartlett's blog where he talks about endogamy in a two-part post, but here's one by Roberta Estes, another from Kitty Cooper with guest author Paul Woodbury, and another by Lara Diamond.

In a truly endogamous population, the use of DNA evidence can be quite challenging, and autosomal DNA sometimes of little use beyond a generation or two. More generationally distant instances of isolated pedigree collapse have much less effect.

Comments

Thank you so much, Edison. I am midway through the video and have looked at the posts you provided. Everything is helping to answer my questions! I think they key phrase here that I did not know prior to your reply was "pedigree collapse". This explains why my normally good research skills were failing in my efforts to learn more. Now that I know what to look for, there's a whole new world of exploring to do. Thank you!!

Answer 2

Hi Marci, It looks like Edison has given you plenty of sources so I will try to provide what information I can from my personal experience. I've been working on my maternal grandmother's family for a couple years now and they are mostly Quebecois, a somewhat endogamous population. I'm not sure how they would compare to the level of endogamy in your ancestors but I have found that many of my bigger total DNA matches match over several small segments and that some of my more distant ancestors had the same last names before they were married. Whereas in other branches of my family it can be difficult to find common ancestors with matches, in my maternal GM's family it is typically more difficult to choose between the multiple CAs that occur.

One way I address the issue is to focus on matches that are on one segment rather than potentially larger matches that are spread across several segments. When I do research larger matches that are spread across several smaller segment I tend to treat each segment as a unique match, particularly if they don't share any common matches with the other segments. There are some cases when it becomes very clear that the individual matches are from different ancestors. I'm fortunate enough to already have determined common ancestors for about a dozen matches on that branch so I can often determine if the different matches with one individual are from different ancestors or not based on whether they match other known relatives. So there are some techniques that I think can help, though none are perfect, as with most things in genetic genealogy there are always exceptions.

All in all in my experience I think endogamy definitely influences estimates of distance to CA and this seems to be more true for matches comprised of several smaller matches than for large individual matches.

Hope that helps and thanks for asking, it's always good to have company. I'd be interested to hear your experiences after you've worked a few matches.

Comments

Paul, this is great information and perspective. It serves to confirm what my gut-level response was to what I'm seeing in my data, and how to interpret it. "Matches on one segment" and "treat each segment as a unique match" describes (better than I could have) where my gut was leading me. I am going to work on putting those ideas to use, and also trying to identify more common ancestors on the "easy" lines, so they can help with the tougher ones. Many thanks for sharing your strategies!