How can I find my mother find her GI father through DNA?

[Paperblank]

Briefly my mother has been searching for her father for 50 years. We have opted to do the Family Finder test and are waiting for the results.

We have a surname but it hasn't checked out and have identified a potential half sibling (same father different mother) who has also kindly agreed to test. I am told that their results will likely correlate to me even though they'd not be my half sibling should there be a connection?

I know we can't isolate DNA that came from the father but is there anything we can do to begin to build a picture/identify DNA that most likely came from the father? For example testing relatives on mother's side - if they share DNA with me the DNA they share will only have come from mother's side.

Lastly I am testing (mother's daughter) rather than my mother and I realise this makes the connection to the father's side less strong and isn't ideal. Hopefully in time my mother will test too.

We have an idea of where the father may have come from so if this shows up in the heritage test that will be interesting.

Is there anything else we can do or other companies I should test with to begin to build the best picture we can? Please let me know.

Also can blood group tell us anything useful - I think not. We don't know the mother's blood type.

[katerennie4]

If you already have a known child (your predicted half aunt/uncle) testing, it will be very clear whether or not that's your relationship. And assuming the two of you don't share any other common ancestors, any DNA you share will be due to your mother's father.

[tomcat]

If the half-sibling is your half-uncle or half-aunt, he or she would resemble you, on average, 12.5%, or the equivalent of a close cousin. If that proves-out -

If the half-sibling of the same father and different mother is male he would carry the Y of the father. He should be tested via FamilyTreeDNA for Y-67 or Y-111 STR's and then enrolled in the appropriate surname, lineage and geographical projects.

If the half-sibling of the same father and different mother is female she would carry the X of the father that would be identical to one of your mother's X's. I believe the X of your mother and her half-sister can be phased, separated, through Gedmatch where the paternal X can be matched to others in the Gedmatch database. FTDNA doesn't do anything with X but 23&Me does do ancestry estimation and segment matching for X. You might consider testing your mother and her half-sister through 23&Me.

[Paperblank]

Thank you very much. Really helpful and useful.

We are not certain that we have the right person - that the person testing is my mother's half sibling.

If we have not got the right person what then is the best course of action?

Is there anything we can do to try to effectively isolate the DNA that has come from my mother's father from my DNA sample?

Will testing other members of my family help? For example if my father tested any DNA we shared would only come from his line (we are trying to identify my maternal grandfather and my mother her father).

Would testing a cousin on my mother's mother's side help at all?

Thanks once more. We appreciate it.

[tomcat]

Exactly half your mother's genome comes from her unknown father. At best, one quarter of your genome comes from that grandfather, could be less. So, your mother is the person that needs to test.

She needs to be tested through 23&Me as they are the only company that offers ancestry analysis and matching on the X chromosome. One of your mother's two X chromosomes came from her unknown father. He got that X from his mother and is a recombined product of his mother's two X's.

That X is the only part of your mother's genome that can be directly attributed to her unknown father and could, with luck, actually lead to his identification.

If your mother consents to be tested through 23&Me she, or whoever is managing her account, should elect to participate in Relative Finder - 23&Me's equivalent to FTDNA's segment-matching program Family Finder. That account manager should also have as complete a genealogy as available for your mother's maternal lines so as to be able to determine which segment matches are attributable to those maternal lines and which are attributable to the unknown father's lines.

The current cost of the 23&Me test is $99. and the 23&Me data can be transferred to FTDNA's Family Finder for $89. Doing that transfer will put the profile in two of the three segment-matching databases.

[tomcat]

If your mother refuses to test, testing your father would allow you to phase your results so as to isolate that half of your genome attributable to your mother, but that would only give you half your mother's genome.

The X thereby attributable to your mother could be wholly from her maternal lines or wholly from her paternal lines or, most likely, some blend of the two. Might be enough but both your mother's X's would be best.

[mkdexter]

Briefly my mother has been searching for her father for 50 years. We have opted to do the Family Finder test and are waiting for the results.

We have a surname but it hasn't checked out and have identified a potential half sibling (same father different mother) who has also kindly agreed to test. I am told that their results will likely correlate to me even though they'd not be my half sibling should there be a connection?

I know we can't isolate DNA that came from the father but is there anything we can do to begin to build a picture/identify DNA that most likely came from the father? For example testing relatives on mother's side - if they share DNA with me the DNA they share will only have come from mother's side.

Lastly I am testing (mother's daughter) rather than my mother and I realise this makes the connection to the father's side less strong and isn't ideal. Hopefully in time my mother will test too.

We have an idea of where the father may have come from so if this shows up in the heritage test that will be interesting.

Is there anything else we can do or other companies I should test with to begin to build the best picture we can? Please let me know.

Also can blood group tell us anything useful - I think not. We don't know the mother's blood type.

In the end you need to test your mother for best results. Your test compared to this suspected half brother of your mother will most likely yield a prediction of half sibs to 1st cousins if you really are related to him but it can't be used as much as your mother's test for future phasing and comparisons.

[tomcat]

When and where was your mother born? How it is known that her father was a GI and in what branch of the service?

[Lincoln]

Do you know your mother's mother's family? Could you find someone from that side who might have clues?

[Bertp]

One can infer from your messages that your mother's mother is not still around. If that is not true, then the best you can do is test both your mother and your maternal grandmother and then subtract your grandmother's DNA from you mother DNA leaving only the DNA passed from you mother's father to your mother. (Some here can give you instructions on how to do that, but I think both FTDNA and 23andMe will be, in effect, doing this automatically in the near future )

Also if test results seem to confirm that the other person tested is indeed your mother's paternal line half brother then. if his mother is still around, test her and then subtract her DNA from his leaving DNA passed his father to him
If that's not possible, then you're left looking for people that match both your paternal half-uncle and yourself.

Of course, having other close maternal relatives tested can help identify your maternal line matches but, by far, the most important person to test is your mother
------
I also agree with one of the previous posters on testing strategy; the more economical and effective way is to test at 23andMe and then transfer to FTDNA. That way your DNA is in both databases for less than price of the FTDNA test. (Also consider uploading results to Gedmatch.com)

If you're testing elderly relatives, I still would advise getting an actual test of some sort at FTDNA so that you will have the possibility of ordering an upgrade from FTDNA even after the elderly relative passes on.

[Paperblank]

Thanks so much for help to date. Truly very grateful.

It feels like you need a Phd to understand the science and steps to take but
hopefully we'll get there . It does seem confusing.

My maternal grandmother (mother's mother) has passed away and has no living immediate family.

There is a second cousin (?) on my mother's maternal side who may be willing to test - mother's mother and cousin had the same female relative who was an aunt (in mother's mother's case) and grandmother in cousin's case.

My father may test and my mother may in time. I think she is worried about security of results and privacy etc. It's all such cutting edge technology in the eyes of many and can seem overwhelming.

As for the father being an American serviceman we have letters on file dating back to 1945 which give name, age and colouring. The name, an unusual surname, has never checked out with any National Personnel Record Centre (NPRC) for a WW2 serviceman. It's possible he came originally (?) from Canada. We assumed that the surname given may be a mis-spelling of a much more common surname but having said that this is a surname in its own right. (We even got a dialect expert in the region of the UK we think the serviceman was based to look into our case and let us know how the name might have sounded and hence different permutations of spelling etc).

We checked out the records - as far as possible - for every single serviceman at NPRC with this surname which took more than 2 years (even though there weren't many) and checking out the more common surname took even longer. Records at NPRC are sadly incomplete due to a fire so it's possible he was a serviceman of this name but not on any official record.

In terms of the search for the serviceman we really have turned over every stone you can possibly imagine and checked out every avenue of research open to us in painstaking detail and then a few more. We have searched for many years.

I travelled to the area and have gleaned as much information as possible. Anyone who might have been able to help was long since deceased we found. We tried the papers etc all to no avail.

The potential half sibling we are testing is female not male.

Thanks again for all help we are truly grateful. DNA is all we have to go on really.

Just to finally add I do have some earlier results from a DNA test Mum took - half sibling ship test which proved negative - lots of incomprehensible looking numbers and letters. If I can find this old paperwork would it be helpful?

If so what do I do with these results and is there anyone I can send them to who might be able to make sense of them if I don't want to share them publicly?

I am thinking I can cross reference with new potential half sibling etc.

Thanks again. Sorry, it just all seems more than a bit confusing.
I am sure I am not alone in thinking that .

[Sue Crowley]

Now that the 1940 census is available, have you checked that to see if a person of the right age and surname pops up anywhere in the US?

While you are not adopted, you are essentially using the same techniques to find a birth father, so this site may be of use to you:

http://dna-explained.com/2012/07/30/adoptee-resources-and-genetic-genealogy/

Good Luck!

[Paperblank]

Thank you - and yes we've looked at the Census but can't find a fit. There were a few who looked possible candidates in earlier censuses but I've found surnames to be transcription errors on looking in more detail.

Will look at the link. Thanks for your interest.

[tomcat]

Paperblank,

I think I recall your "case" from a few years back. Think you you were on this forum under another username and think the earlier test you reference was either an X-STR or a CODIS STR. Sorry that was inconclusive hope your new attempt succeeds. The Family Finder test samples SNP's rather than STR's so they can't be cross-referenced.

[Paperblank]

Thank you again. SNPs etc? Oh dear fear I will have to do a lot of reading.

No, first posting on here. There were apparently more than 100,000 babies fathered by GIs during WW2 so there are many 'GI babies' out here. Many with even less information than we have sadly.

We did the initial test with a company called DNA Bioscience and not sure they are still going. They gave some sort of break down/DNA details in paperwork sent re: half siblingship test.