Optional Parents

[Sue_3]

Does anyone know a way to create a tree to link to DNA results on Ancestry that caters for uncertainty about a parental relationship? As in, mother's family fully known but father's family could be one of two alternatives? I have asked Ancestry and they said no ... but I think they were actually saying that you can only link DNA to one tree?

I have tried all of the options I can find in Family Historian, but always one father becomes the default. I want to get the DNA results to link to either potential father - to be able to judge the most likely biological father - but also I want to know the history of both men (because, whatever the biology, both were significant in the family's social history).

When I upload the tree to Ancestry it seems to reflect that the person has two fathers, but only uses one of them to detect common ancestors.

Surely I am not alone in this? Anyone know how best to proceed? TIA for any advice.

 

[peter]

I don't believe you can do what you're suggesting at Ancestry, nor is it currently possible to have a tentative link (as I do in the trees on my own computer). Both would be useful enhancements for those who know what they're doing, but as that's a small minority of Ancestry users it might not be a good idea.

The way to achieve what you're looking for requires three trees, one for each father, and one without a father (that's the one to attach to your DNA results). ThruLines will then reveal any trees with candidate fathers.

 

[Sue_3]

The way to achieve what you're looking for requires three trees, one for each father, and one without a father (that's the one to attach to your DNA results). ThruLines will then reveal any trees with candidate fathers.

I don't understand how that would work - can you please explain further? For ThruLines to find links to either father, wouldn't it require someone else to have the mother and child in their tree with that person as the father? Paper records show the father to be the mother's husband, so no one will have anything other than that in their tree?

What I need is a tree that will highlight any matches to the wider family of either candidate father and I don't see how it can do that unless the families of both candidates are in the tree that the DNA is linked to?

 

[jorghes]

I don't understand how that would work - can you please explain further? For ThruLines to find links to either father, wouldn't it require someone else to have the mother and child in their tree with that person as the father? Paper records show the father to be the mother's husband, so no one will have anything other than that in their tree?

ThruLines doesn't need users to have added ancestors to their tree to create a possible link - it uses all the resources of other people's trees to offer possible relatives to you - for example some of the ThruLine links I have are to people with only 2-3 people on their own trees.

However, Ancestry does require you to have attached a tree to your results - my aunt, who has had her DNA tested, does not appear on any ThruLines, because she has not created a tree.

 

[peter]

I don't understand how that would work - can you please explain further? For ThruLines to find links to either father, wouldn't it require someone else to have the mother and child in their tree with that person as the father? Paper records show the father to be the mother's husband, so no one will have anything other than that in their tree?

What I need is a tree that will highlight any matches to the wider family of either candidate father and I don't see how it can do that unless the families of both candidates are in the tree that the DNA is linked to?

ThruLines offers hints. It only works when you don't have an ancestor in your tree: so as well as allowing for the two candidates you've found, it would allow for any other solution you might not have thought of.

Of course, you can always search other peoples' trees to see who they show as the father. But the advantage of ThruLines is that (like LostCousins) it repeats the same set of searches automatically, so that you don't have to do them one by one.

(The disadvantage of ThruLines is that it will only work when the user at the end of the line is a DNA match for you, but it would appear from your question that you're looking for a DNA-based solution.)

 

[Pauline]

I had an uncertainty in the maternal line of a fairly near ancestor but I nevertheless had the documented mother entered in my tree. When I did my DNA test (before things like ThruLines were available) I linked my results to this tree.

It was readily apparent from my DNA results that in my case the documented mother was the correct one, but had it not been so, I think it would probably have been easy enough to deduce if the other suspected mother was the actual one instead, as I had already researched her ancestry too. I could at that point have changed my usual tree to show the other mother, but I don't think it would have been essential to do so in order to interpret the DNA results.

 

[Helen7]

ThruLines doesn't need users to have added ancestors to their tree to create a possible link - it uses all the resources of other people's trees to offer possible relatives to you

This is true, but for ThruLines to work, it does need someone to have entered the relevant father and child, to make the connection between you and your DNA match, using that other person's tree (who may well not be a DNA match to you).

An example from my own experience might help illustrate the point.

Some time ago, I noticed a DNA match (of sufficient strength to convince me it must be genuine) where the linked tree only showed the mother's side. The father's side was blank. There were no names I recognised in the (extensive) mother's family, and there were no ThruLines hints, so I reasoned that the connection was probably via the missing father's line. The mother was born in the same place as my great-grandmother (and none of my other ancestral lines were from anywhere near there), so I was able to narrow it down to that eighth of my ancestry. However, this was a branch where I'd identified no other DNA matches and there were no 'Shared Matches', so it was a bit of a conundrum - a theory but nothing more. I tried contacting the tree owner, but got no reply.

Then more recently, I noticed that this person's tree included my 2x great-grandparents but they were 'floating around' and not linked to the home person. I contacted the tree owner again, and this time started a very useful correspondence. It seems the father on the birth certificate was the mother's husband but not the biological father. There were also a couple of shared matches on MyHeritage, and a few other DNA matches on Ancestry that convinced us that we shared the same 2x great-grandparents. He was then able to fill in the father's side in his tree, which he had been unable to do before. As a result of this, he now appears in my 'Common Ancestor' list, as a 3rd cousin which I believe is correct.

It seems to me that if you have 2 trees, one for each father, you could attach one of these to your DNA results and see if you get any 'Common Ancestor' hits. Then if nothing useful appears, you can then change your linked tree to the other father and see if it makes a difference (but you have to wait a few days after changing trees for the change to take effect). The alternative strategy of having no father in your linked tree relies on someone else having made the connection - to one of these 2 possible fathers or indeed to someone else, as Peter mentions. This is certainly possible, but wasn't the case for me in the example I describe above.

 

[peter]

This is true, but for ThruLines to work, it does need someone to have entered the relevant father and child, to make the connection between you and your DNA match, using that other person's tree

True - but what I'm suggesting has the advantage that it will work even if neither of the candidates Sue_3 has identified is correct.

Helen7's method will work better in the circumstances where nobody has entered the specific father-child combination, provided that that there are DNA matches who are descended from a different child. But surely you could achieve the same result more quickly by searching user trees for the two possible fathers? That's what I do, and there are often spin-off benefits from this approach. Searching user trees will also find people who aren't a DNA match for you but might be a match for one of your cousins - neither Common Ancestors nor Thrulines will pick them up.

The advantage of the ThruLines approach is that Ancestry will keep on searching without you doing anything - you're using the system as it was designed to be used.

 

[Pauline]

what I'm suggesting has the advantage that it will work even if neither of the candidates Sue_3 has identified is correct

I guess it may depend on how certain Sue is that one of the two possible fathers is correct. The chances of a 3rd possibility may be pretty remote and not worth considering unless DNA matching fails to confirm either of the other two.

 

[Helen7]

But surely you could achieve the same result more quickly by searching user trees for the two possible fathers?

I agree this is a good approach, and one I take if I have a choice of parent in my tree. However, in the case I described above, my cousin had no idea who the father was and had to work it out from the DNA matches. The father wasn't in my tree as I hadn't worked down the generations from my great-grandmother's siblings as their surname is very common (though my g-grandmother's forename is very unusual, enabling me to pinpoint her with her family in censuses).

 

[peter]

I guess it may depend on how certain Sue is that one of the two possible fathers is correct. The chances of a 3rd possibility may be pretty remote and not worth considering unless DNA matching fails to confirm either of the other two.

If she takes the ThruLines route and supplements it with searches of user trees then all possibilities are covered. My over-arching aim is to develop approaches that work in a wide range of situations with different levels of knowledge. Based on what Sue_3 has told us any of the suggested approaches will work in her example, but it's better to have methods that can be used more generally.

ThruLines is an innovative feature designed by Ancestry with the specific aim of knocking down 'brick walls' by taking a line back an extra generation - it's important for users to understand what it does and how it does it. Getting too involved in the detail of individual cases can sometimes obscure the broader principles.

I believe in keeping things as simple as possible.

 

[Sue_3]

Thank you all, that gives me a lot of food for thought.

I must admit that it hadn't occurred to me that the father might not be either of my two candidates, and I now appreciate that I ought at least to check that out by adopting Peter's suggestion of having a tree linked to the DNA that has neither of my two candidates shown as the father in it - whether permanently or for a limited time? I am wondering whether any such tree should exclude the mother's marriage completely, or retain it but not link the child to its mother's husband?

I do think it is very unlikely that anyone else will have put the child in a tree with anyone other than the mother's husband as the father. However, in three decades of research I've learned many things about my family that have astonished me so I can only agree that all possibilities have to be considered.

What I didn't mention above is that I have had the DNA linked to a tree with one of the candidates shown as the father ever since I got the results back in 2018. In all that time I have only found one possible match to that candidate's family. That match was thrown up fairly recently by ThruLines, is only a 6.5 cM across 1 segment DNA match, and making the trees match at all requires assuming that someone - who is currently neither in my tree nor the match's tree - changed their surname in the mid 1700s. As I haven't got my research into the relevant branch back quite far enough yet I currently have this flagged as requiring further investigation.

I have, meanwhile, identified several matches to the other candidate's family (17 and counting on Ancestry, without the help of ThruLines, and a few more on other sites that the DNA has been uploaded to). However the closest are potential 4th cousins and I haven't been able to quite convince myself that they are conclusive proof of the relationship? This was why I wanted to try to force Ancestry to use ThruLines for that candidate's tree, in case that could find closer matches. I didn't want to lose the possibility of finding any for the first candidate, though!

So, having gone to the trouble of combining all the trees and stripping personal data from them and uploading the results, to try to make Ancestry answer all of the questions by reference to one tree, I now see that that will not suffice. I know that I could have asked the questions here before starting work on it, but I had to figure out what was possible using my FH software as well. I've learned a lot while doing that, so I'm content with the time spent on it.

I now have to decide whether to follow Peter's suggestion and just have three separate trees, full stop, or adopt Helen's idea of using different trees at different times. In both cases, you can't know which fish will be in the sea at the time?!

I think I need to sleep on it, and perhaps for more than one night?

Happy to hear any other thoughts readers may have in the meantime.

Many thanks again to all who have contributed their ideas,

Sue

 

[jorghes]

This is true, but for ThruLines to work, it does need someone to have entered the relevant father and child, to make the connection between you and your DNA match, using that other person's tree (who may well not be a DNA match to you).

I fully understand your point - but it still remains that Ancestry can compile ThruLines from DNA matches with 3 people on their tree - themselves and their parents, regardless whether or not you have that particular branch on your tree. I have a couple of examples of exactly that on my tree, one of which where my research has stopped some 4-5 generations above the person with the DNA link.

 

[Helen7]

I fully understand your point - but it still remains that Ancestry can compile ThruLines from DNA matches with 3 people on their tree - themselves and their parents, regardless whether or not you have that particular branch on your tree. I have a couple of examples of exactly that on my tree, one of which where my research has stopped some 4-5 generations above the person with the DNA link.

Oh , I agree. Of my 60+ matches in my 'Common Ancestors' list, 10 have <10 people in their trees, with none of them appearing in my tree as the connection is further back.

 

[Sue_3]

My current Ancestry tree, which is the first one I have linked to my cousin's DNA as well as to my own DNA, is now throwing up a few previously unidentified 'common ancestors' in the 6-8 cM range (which I'm just adding to groups for now), so I'm going to let well alone until Ancestry stops showing us matches below 8 cM.

Still pondering what to do after that.

 

[peter]

Oh , I agree. Of my 60+ matches in my 'Common Ancestors' list, 10 have <10 people in their trees, with none of them appearing in my tree as the connection is further back.

It's brilliant news for your cousins - I contact mine and they're invariably delighted as they're often just starting out.

One of my latest finds is half-Melanesian but she knows much more about that side of her tree than the English side, so she was really pleased to get my message.