DNA case study: uploading your DNA to other testing sites


13 August 2021
Photo from Unsplash
Family Tree reader and DNA Bootcamp student Jenny wrote in about her DNA research. Here we share her useful question and Michelle Leonard’s expert answer.

Jenny writes:

Having tested with Ancestry DNA, one of the things I have done is to upload mine and my husband Steve’s DNA results to other sites. We have both made contact with second cousins that we didn’t know directly, which is great and something that wouldn’t have happened otherwise. There have been lots of other matches on both our trees though that have led to new connections and evidence that I have the correct ancestors. I have made connections in Australia, the USA and Finland, as well as finding out that the wife of a relative is also a relative!

Steve and a match named ‘L.’ share a great-grandfather, Patrick Treanor, whose father is a John Treanor. I know this to be correct. Before the DNA test, I had my doubts about my husband’s cousin’s proposal of their joint 2nd great-grandfather, due to paper trail inconsistencies, and I was hoping to find some other matches which would lead to another John Treanor as Steve’s 2nd great-grandfather. I have narrowed down the eight possible options I had to one that I think is the most likely, but I still don’t have any conclusive evidence.

Steve shares 1,098cM with his known first cousin L

He shares 27cM with M who would be his 2C1R if the John Treanor proposed is the correct 2nd great grandfather

He shares 32cM with D and 36cM with J who would both be his 3Cs again if that John Treanor is his 2nd great grandfather

L., M. and J. have all tested with Ancestry and uploaded their results to MyHeritage too.

Do you think the hypothesis is correct or am I right to be sceptical?

Michelle Leonard writes:

Thanks for your question Jenny - I can understand why you are sceptical given you've found paper trail inconsistencies.  It is important to look at all angles, and such discrepancies are certainly red flags. Just looking solely at the DNA side of things [some details have been withheld from publication on the blog] my view is that it is unlikely the hypothesized John Treanor is the father of your husband’s great grandfather Patrick and I would suspect they share a more distant relationship on the same line but the bottom line is that it is possible. Overall, the picture is currently inconclusive and requires further testing and research to gain more clarity.

Read on for useful tips to apply when working with DNA test results…

Content continues after advertisements

Think about probabilities

If your cousin’s hypothesis is correct then M would be Steve’s 2C1R. 27cM is a very low number for a 2C1R but I have seen several very low genuine 2C1Rs before so your cousin’s hypothesis cannot be definitively ruled out simply because the match is lower than expected for the proposed relationship - ranges are wide and outliers, even extreme ones, are possible. I always encourage people to consider the most likely relationships before the lesser likely ones, though, as the most likely are most likely for a reason.

If you look at the different relationships and probabilities on the DNA Painter Shared cM Project Tool (remembering to turn on ‘beta with updated probabilities’ for the most up-to-date data!) you will see that a 2C1R relationship is listed as an 8% probability for 27cM so while it would be a reasonably large outlier, it is still within the realms of possibility.

The 32cM and 36cM potential 3C matches (D and J) are less problematic as approximately 10% of third cousins won't share anything at all. This means it is very common to find some 3Cs who only share around 30cM. 

The only way to obtain a higher level of surety in this case is further testing on both sides of the equation. 

Ask to see your matches’ DNA match list

One thing I would pursue is whether your husband's cousin would share her DNA match list with you so you can see how well she matches these three matches: if they're all equally low for her then that's further evidence in favour of Patrick and this John being more distantly related than father/son.

Test the older generation or those in a different line

What I would recommend doing (if at all possible!) is testing at least one family member who is that magic generation closer to the common ancestors e.g. in this case that would be your husband's father's generation (any living aunts/uncles/first cousins of his father).  If no-one of that generation is alive or willing to test I would turn your attention to others in your husband's generation but perhaps try to get someone from a different line e.g. if his grandmother Sarah had siblings then look to children or grandchildren of those siblings.

Ask your matches if they know suitable candidates who might take a test

I'd also investigate whether ‘M’ has any living aunts/uncles who could test and whether ‘D’s’ father (listed as private on his tree so likely still living) would be willing to test.

Amass that body of evidence

It's all about amassing a body of evidence when it comes to relationships and potential outliers - the more testers there are, especially of the closer/older generation on both sides of the equation, the clearer the picture will become.  Any one match could be an outlier (even an extreme one at times!) but when you have a number of family members on both sides to compare, outliers become easier to identify and eliminate.

Build out other trees

The other thing I would research is the potential 2nd great-grandmother Mary Corrigan's line. Can you build a tree for her back to parents and siblings and can you find anything in your husband's or L’s DNA that links back to that line? If you can't that's another red flag especially if the other descendants who have tested (M and D) do have matches to her ancestry so I'd be asking them if they'd be willing to share their match lists too so you could investigate that.

It's important not to overlook the 2nd great-grandmother as a very similar amount of DNA will have been passed down from her as from the 2nd great-grandfather. Patrick had both a mother and a father and there will be matches to that woman in your husband’s DNA results.

Lastly, remember to avoid confirmation bias!

Finally, try to look at this puzzle as objectively as you can - don't let confirmation bias get in the way.  Always think "can I disprove my preferred hypothesis" and you shouldn't go too far wrong on that front.  Good luck with it!


Would you like to learn more about DNA for family history?

Jenny took part in the DNA Bootcamp run by Family Tree and DNA Detective Michelle Leonard in the spring of 2021. If you’d like to find out more about how to use DNA in your family history, join us for the next DNA Bootcamp, starting 29 September 2021.

About the DNA Bootcamp

The DNA Bootcamp is an 8 week online course run by Family Tree and Michelle Leonard with live presentations, live follow-up group discussions, a dedicated Facebook group solely for DNA Bootcamp students, and a digital DNA Workbook, packed with useful information & practical advice.

To find out more click here.

Totally new to the DNA thing? We're here to help

If you are totally new to DNA, dip your toe in the water with our ‘DNA Absolute Beginners’ webinar, 7pm on Wed 1 September, tickets £10. You’re not too late to learn more! Click here.