Years ago, someone had made a comment to one of my sisters-in-law that she was obviously "more Irish"—than what? Than the rest of the family? Needless to say, that sister-in-law took a shine to the designation. She took that comment as a compliment and ran with it, much to the dismay—and ridicule—of her siblings.
In an unrelated occurrence, along came this woman's particular genealogy-driven sister-in-law (me, if you can't tell) who innocently asked for three siblings from that family to step up and volunteer to take an autosomal DNA test to help said relative with a class project. That sister-in-law, along with another, as well as my husband, became my volunteers.
When the test results came back, there didn't seem to be many surprises. Ethnicity results seemed to fit what we already knew, according to family lore: about fifty percent of each person's result showed Irish ancestry. Small wonder, since my father-in-law could claim three out of four grandparents who were actually born in Ireland, with the fourth the son of Irish immigrants.
Long after I had completed the class project for which I had recruited family volunteers, I almost forgot about the DNA results for those two in-laws. Sure, I served as admin for each of the accounts, and sat down with each volunteer to explain what the test results meant and what we could learn from examining the many matches they each had received. After the initial fascination wore off, everyone set aside the results and, well, pretty much forgot about them.
While I had taken the time to examine the then-current set of matches for each sister-in-law, eventually I, too, set aside the experiment...until I started noticing some unusual results popping up. Since those early days of mounting the steep learning curve for utilizing genetic genealogy, I had met experts in the field both in the United States and in Ireland. I have had some fascinating talks with some of these innovative leaders, and appreciated their accessibility in those early days.
One day, returning to the test results for my in-laws, I was surprised to see a name pop up for one of my husband's sisters: that of a key genetic genealogist I had met in Ireland. It was linked to a DNA match for which he served as test administrator. His Irish ancestor from County Tipperary was our distant relative!
While that was fun news to discover, I also realized something was odd about the finding. Why hadn't I spotted that name when I reviewed my husband's results? After all, I had kept checking my husband's results quite regularly. Besides, why did it only come up for this one sister's test? I checked again—then double-checked by looking at his other sister's test. Not one sign of this DNA connection for either of them; only for the one sister.
I'll give you a moment to guess which of the three siblings was the owner of that test result. Clue: it was the sister who claimed she was "more Irish" than the rest of the family. (I couldn't resist teasing her that her DNA test validated her claim.)
Fast-forward to this month's research project. Now that I've been struggling to find a way to confirm this month's possible connection to my father-in-law's Tully family—the Dennis Tully, born 1830, whose early existence in County Tipperary, Ireland, seems to have been lost to time—it occurred to me that my "more Irish" sister-in-law could once again come to my assistance. I went back to her DNA test to see how her numbers compared to any Tully matches.
In particular, I was looking for one specific match: a descendant of Dennis Tully's daughter Johanna who, of all Tully DNA matches, shared the largest centiMorgan count with my husband. This DNA match (we'll call her TD for Tully descendant) shared 72.8 cMs with my husband. When I compared this match with my "more Irish" sister-in-law, the count bumped up to 91.3 cMs.
I then turned to the Shared centiMorgan interactive page at DNA Painter to enter some numbers and check results. I wanted to see what the probability might be for these matches as third cousins once removed (in other words, if Dennis were son of our Denis Tully and Margaret Flannery) versus fourth cousin once removed (if Dennis were the older couple's nephew). I was hoping the difference between my husband's match level and that of "More Irish" might bump one into the not-possible realm.
Well, it came close, but didn't quite convince me. Here's what the numbers said, first for current version, then for beta version #4 with updated probabilities:
If Dennis were son of the elder Denis:
- TD as 3C1R to hubs = 31%
- TD as 3C1R to "More Irish" = 24%
- TD as 3C1R to hubs on update = 14%
- TD as 3C1R to "More Irish" on update = 8%
If Dennis were nephew of the elder Denis:
- TD as 4C1R to hubs = 4% chance
- TD as 4C1R to "More Irish" = 0.64% chance (but not footnoted as "falls outside the bounds")
- TD as 4C1R to hubs on update = 0.61% chance
- TD as 4C1R to "More Irish" on update = 0.50% chance
What does all that tell me? I was hoping that using my "More Irish" sister-in-law's numbers might help flush out some more telling numbers, as she had more centiMorgans in common with their highest Tully match. The probabilities for the closer scenario—assuming Dennis Tully's father was actually our Denis Tully—sure look better than those for the minuscule chances for the uncle-nephew scenario. But—and here is the part that leaves me twisting in the wind—either relationship could be possible. Slim, but not quite ruled out.
Which leaves me asking the same question I had before starting: now what?
No comments:
Post a Comment