Wednesday 8 December 2021

Unexpected Y-DNA results - as an Admin, what do you tell the test-taker?

Someone does a Y-DNA test to find out more about their direct male line and "where my surname came from" only to find that the results are not what they expected. So they join the DNA project relevant for their surname and ask the Administrator for an interpretation. What information are they going to find useful? And, as an Admin, what information can you give them?

These were the questions I pondered recently when a new project member joined my O'Malley DNA Project. And I think it would be useful to take you through my response to his email and point out some of the challenges that Admins face when confronted by the unexpected results of one of their project members.

I have privatised any identifying information. My comments are in italics.

Dear Mr O'Malley

I had a look at your results and here are some top line observations …

You have 13 matches at the 111-marker level of comparison, 20 matches at 67 markers, and 20 at 37.
Your top matches are to people called Leyden/Layden/Lyden, Ward & Corcoran.
You have no O’Malley’s among these matches.
You do not match any of the genetic groups within the O'Malley Project.

This suggests that there may have been a Surname or DNA Switch somewhere along your direct male line, which means that your Y-DNA signature was associated with some other surname prior to switching to O’Malley. This could have happened relatively recently, or centuries ago. The previous surname could have been Leyden/Layden/Lyden, Ward, Corcoran ... or some other surname.

Already we may have shattered a lot of the hopes that the test-taker may have had when they first took the Y-DNA test. Some people may be interested in the deeper origins of their surname, but I suspect that the reason most of us do any type of DNA test is to help with our genealogy and perhaps to break down a specific Brick Wall. In the case of Y-DNA, the Brick Wall will be on the direct male line, and in the case of Irish genealogy, this typically halts abruptly at 1800. So, for many test-takers (including perhaps Mr O'Malley above), the hope with Y-DNA may be "can it help me break through my 1800 Brick Wall and push it back by a generation or two into the 1700s?". Alternatively, the Brick Wall may be more recent (say mid-1800s) and the test-taker's genealogy may be stuck in America (for example), in which case the hope may be "can Y-DNA help me jump back across the pond to Ireland?". Another possibility is that the test-taker has explored the relevant surname project and sees that there are several well-defined genetic groups within the project and wants to find out which one of these he belongs to.

In all of the above cases, Mr O'Malley will be disappointed. He was aiming to find an O'Malley in Ireland, and instead he has discovered that he is "not an O'Malley". Has his line of genealogical enquiry been completely derailed? Is his Y-DNA telling him he is not on the right track? Has he been sidelined into completely foreign territory?

As an Admin, what do you do in such a situation? What kind of information might be helpful for the test-taker? What information can help him navigate through this new uncharted territory?

The sort of additional questions that may arise include: 
  • if I am not an O'Malley, what am I? what surname did my ancestors carry before it was switched?
  • where did the name come from?
  • when did the switch happen?
  • why did the switch happen? was there some secret adoption or illegitimacy in the recent past? or did the switch happen centuries ago, for reasons lost in the mists of time?
  • do I have any hope of breaking through my Brick Wall? and what's on the other side - is it an O'Malley or a complete stranger?

The sad truth is that Y-DNA is severely limited in its ability to answer these questions. It will possibly give you some clues, but the answer may not be definitive and you may have to spend a lot of money on expensive tests to arrive at a nebulous conclusion. I know this sounds pessimistic ... but that's because it is. From the perspective of day-to-day genealogy, there is a lot less to Y-DNA than meets the eye ... especially when you compare it to the information you derive from autosomal DNA testing. I'm not saying its useless ... just less useful, from a genealogical point of view. And I feel as frustrated about this as the average test-taker.

Update: some feedback on this post spurs me to make some additional points:
1) another possibility is that this man comes from a very rare branch of O'Malley's - a branch that arose completely independently of the other major branches and is in no way connected to them in the immediate pre-surname era (e.g. 700-900 AD in Ireland) - and he is currently the only man from this rare branch in the entire FTDNA database. He will have to wait for another member of this rare branch to join the project before he (and this new member) can be allocated together to a new genetic subgroup. I sometimes mention this in feedback to new members with an SDS / NPE but the O'Malley DNA Project is quite mature and a variety of genetic subgroups have already been identified, and there are no outstanding O'Malley groups mentioned in the historical record (e.g. surname dictionaries) so I would be very surprised if any additional rare subgroups were to emerge (it is not impossible but it is improbable) - hence I decided not to mention this rare possibility in this particular reply. I may mention it if the response from the project member raises any questions about the probable SDS, but it is always a balance between giving just enough information to adequately inform the new member and not too much that it overwhelms the recipient. It's a judgement call. What would you have done? Hmmm ...
2) Most people take the prospect of an ancestral surname switch quite well, but some jump to the decision that they have wasted decades of research on their genealogy and half of it is now useless. I go to pains in my presentations to explain how commonplace such SDS / NPE events are and point out that each of us has somewhere between a 33% to 55% chance that our direct male line does NOT go back to the progenitor of the surname that we carry today. So if we insisted that you could only call yourself an O'Malley if you matched the DNA of the progenitor of the surname, then about half of us would not be who we thought we were. For me, anyone who carries a surname, owns that surname, and has every right to claim that surname's history and heritage - "an O'Malley by any other name would smell as sweet".

Here's how I continued my response to my new project member. I veered into deeper origin territory and if you get the impression that I was clutching at straws, you may be right ...

Your matches are relatively distant. The closest matches have a Genetic Distance to you of 6 at the 111-marker level of comparison (i.e. 6 steps away from an exact match), 3 out of 67, and 3/37. Nevertheless, it may be helpful to write to each of them and ask if they have any connection to the O'Malley surname. The geographic origin of their MDKA (Most Distant Known Ancestor) may also provide some clues.

We could get further clues from looking at SNP markers. Some of your matches have done SNP testing (usually the Big Y test) which allows us to see where they sit on the Tree of Mankind. The question is - is there a pattern? Here are the SNP Sequences (i.e. list of ancestral SNP markers) for your matches at 111 markers (generated via the 
SNP to Breadcrumbs tool here):
  • R-L21 > S552 > DF13 > DF21 > S5488 > Z16294 > BY4001 > L130 > BY3308 > BY16783 > BY16785
  • R-L21 > S552 > DF13 > FGC11134 > FGC12055 > Z3026 > Z16250 > A114 > CTS4466
  • R-L21 > S552 > DF13 > DF21 > S5488 > Z16294 > BY4001 > L130 > BY3308 > BY16783 > BY16787 (x2)
  • R-L21 > S552 > DF13 > DF21 > S5488 > Z16294 > BY4001 > L130 > BY3308 > BY16783 > BY16785
  • R-L21 > S552 > DF13 > DF21 > S5488 > Z16294 > BY4001 > L130 > BY3308 > BY16783 > BY16787 > BY16790 (x2)
  • R-L21 > S552 > DF13 > DF21 > S5488 > Z16294 > BY4001 > L130
  • R-L21 > S552 > DF13 > DF21 > S5488 > Z16294 > BY4001 > L130 > BY3308 > BY16783

From the above we can see that most of your SNP-tested 111-marker matches fall on or below the branch characterised by the SNP marker BY16783 (in bold). It is therefore likely that (if you were to do SNP testing) this is where you would sit too.

You can see the path of migration of the people who sit on this particular branch of the Tree here …

This SNP marker arose about 2400 years ago (i.e. 400 BC, well before the advent of surnames), probably in Ireland.

Rob Spencer's tools are great for illustrating the deeper origins of a particular SNP marker and the SNP Tracker tool provides a great visual that captures the imagination. Customers like it. However, this is closer to archaeology than genealogy, and does not address the immediate questions of the test-taker as described above ... but is it the best we have to offer?

Only 7 people fall on or below this branch (according to FTDNA's public Y-DNA Haplotree) - 2 report Ireland as their ancestral homeland, 1 reports USA.

The Big Y Block Tree (which is visible only to those who have done the Big Y test) shows the more fine-detailed branching structure of this portion of the Tree of Mankind, including the number of SNPs associated with each branch.

It is a pity that the Big Y Block Tree is not publicly viewable, unlike Alex Williamson's "Big Tree" which has proved to be so useful over the years. Nowadays the Block Tree has much more data than the Big Tree and better delineates the more fine-detailed branching structure of the Tree of Mankind. However, a big drawback of the Big Y Block Tree is that it does not display surnames for each and every branch. I wish it would. Instead it reports the entire name for matches (rather than just the surname) and only for those matches closest to the test-taker in question. This makes it far less useful as a research tool. We don't need to see the entire name of matches on the Block Tree - just the surname would do very well thank you.

The next question is - what surnames do we find most commonly in this portion of the Tree of Mankind?

We already know that common surnames among your matches are Leyden/Layden/Lyden, Ward & Corcoran. We can look at relevant Haplogroup Projects to see if there are any others. And the relevant Haplogroup Projects for this portion of the Tree includes the R-DF21 and subclades project (1664 members). This has the following people grouped under or near BY16783 …

From this we can see that the surname Duffy sits on a neighbouring branch.

It is unfortunate that new members are not added automatically to the Haplogroup Projects relevant to their Y-DNA signature. This would help make Y-DNA testing more useful to FTDNA's customers. In the present case, only a handful of relevant matches have joined the project and this decreases the value of the Haplogroup Project for research purposes.

Is there a pattern for the distribution of these surnames on Surname Distribution Maps?

It is difficult to draw any firm conclusions from this, but there may be a preponderance of these various surnames in the west of Ireland, around Galway & Mayo, and this suggests that this could be where this particular SNP marker arose. Coincidentally, this is close to the origins of the Mayo O'Malley's.

You may benefit from joining the R-DF21 and subclades project - the Admins there could offer additional insights into the origins of your Y-DNA signature. 

There are no DNA Projects for the surnames Corcoran, Duffy or Leyden/Layden/Lyden, otherwise I would have referred him there too. There are a few Leyden's and Layden's in the Lyddon/Lydon/Liddon Project so that might be worth joining, but none of the participants on the public Results Page sit anywhere close to BY16783.

For now, I have moved you into the Ungrouped section of the O'Malley project. This will change if another O’Malley tests and joins the project and is a match to you.

Hope you find this of interest.

Best, Maurice

So there we have it. Have we been able to answer any of the burning questions this test-taker may have had? Let's recap on what they might have been ...
  • if I am not an O'Malley, what am I? what surname did my ancestors carry before it was switched? 
    • we are still none the wiser - our additional analysis has not pinned down a most likely candidate
  • where did the name come from?
    • possibly the west of Ireland.
  • when did the switch happen?
    • we still have no clue. Only (expensive) sequential Y-DNA testing of progressively more distant male cousins could possibly address this question (but only if it was within the last 200 years or so). See an example here from the Gleason DNA Project.
  • why did the switch happen? was there some secret adoption or illegitimacy in the recent past? or did the switch happen centuries ago, for reasons lost in the mists of time?
    • we may never know the answer to this question. If it was a recent switch, there may be some clues in documentary records (e.g. the lodger's name was Corcoran and he lived with the family for 20 years after the husband left).
  • do I have any hope of breaking through my Brick Wall? and what's on the other side - is it an O'Malley or a complete stranger?
    • his close matches may offer some clues.
    • Big Y testing may place him on a specific branch of the Tree of Mankind that is associated with a specific single surname, but with only 7 people sitting on or below BY16783, it may be that a specific single surname cannot be identified and therefore no firm conclusions can be drawn. In which case it is a waiting game for more people to join the databases and do the Big Y test and fall on the same or adjacent branches.
    • autosomal DNA testing may help if the Surname or DNA Switch has been recent (i.e. within the last 200 years).

So the test-taker has been supplied with a huge amount of information to digest ... but at the end of it all, he still remains in limbo. He does not match any of the O'Malley groups within the project. His closest matches are not that close and have completely foreign surnames. He arrived with a whole list of questions and leaves with a whole set of new ones (and the original list unanswered). His options for moving forward are limited and his Brick Wall looms larger than ever.

This unfortunately is the current state of the art.

Maurice Gleeson
Dec 2021

No comments:

Post a Comment