No announcement yet.

Looking for help in solving a problem...

  • Filter
  • Time
  • Show
Clear All
new posts

  • Looking for help in solving a problem...

    First, a quick background:
    The paternity of my great great grandfather has never been known to my family, and that is the question I am trying to answer. I have had a 111 marker YDNA test performed on myself, as well as tests with NatGeo Geno 2.0, 23andme, and Ancestry. The YDNA test has identified for my the probable surname of my gg grandfather's father. There are several people of that surname that have performed tests as well, at each of the above listed sites.

    As far as who my gg grandfather was, I have a couple of prime suspects. I happen to be in communication with the (legitimate) gg grandson of THE prime suspect (prime suspect for reasons in addition to the DNA match). This particular person has not tested with FTDNA (yet). I am also in communication with several others in that same line, but their common ancestor was further back (meaning they shared ggg grandparents or further).
    1. Is it possible for me, via YDNA testing, to definitively answer whether this gg grandfather is common between us?
    2. If it is not possible, what tests could get me as close to definitive as possible?

    I'm thinking that certainly people who share gg grandfathers must be closer genetically than people who share ggg grandfathers, but my knowledge of DNA testing capabilities is very minimal.

    Thanks for any help!
    Last edited by RCrumpton; 28 December 2012, 12:30 AM.

  • #2
    Definitive seems so definite, doesn't it?

    Who needs to be convinced by a definitive result, should you be able to obtain it? (You seem fairly convinced, thereby indicating it is one or more of your Y correspondents that need convincing, but that may not be the case).

    Two individuals who share a GGF "ought" to share some autosomal DNA. They may not, or if they do, it may not be much, and any sharing "could" be explained away by other, even unknown, lines of shared descent. So, autosomal testing could fail to be definitive or, at least, convincing.

    On the other hand, a positive autosomal result might convince you, and/or your closest cousin and/or some of your other Y correspondents that you do, most likely, belong in the lineage (their lineage).

    And such an outcome might inspire you, and possibly the others, to do the deeper research and deeper Y testing, to deepen your understanding of your shared lineage.
    Last edited by tomcat; 28 December 2012, 01:40 AM.


    • #3
      I'm not going to answer your question with certainty but I can give you some ideas based on my experience so far.

      I have the same problem. My surname GG grandfather is unknown.
      I found a surname - 3rd cousins' Y-DNA results on another site and we matched 33 of 34 markers that he tested. (Unfortunately for me he does not return emails). I'm just telling you this to let you know for me there was only one mutation in the 3 generations of Y-DNA between the two of us.

      I also found two atDNA (non surname line) 3rd and 4th cousin matches on gedmatch from my known great grandfather. I am currently using to triangulate matches that may tell me who the unknown GG grandfather was by finding matches to people who match the 3 of us before our great grandfather.

      If you have not done the FF on FTDNA you can use your 23andME atDNA on gedmatch to Triangulate on common matches, or confirm the bunch of you all match. You will need to use both your FF and 23andMe if you have matches on both sites to use the triangulation function.

      I don't think you will be able to "definitively" answer whether your gg grandfather is common between your matches but you can confirm a solid guestimate.
      I am very interested in what the pro's think


      • #4
        Originally posted by tomcat View Post
        Definitive seems so definite, doesn't it?
        Does "most likely" sound better?

        My family and the other lineage are all convinced that my line is a branch of theirs (although previously unknown); no one has doubts about that. There isn't convincing to do per se, let's just say I like to *know*.

        It's has been a mystery (to the last couple generations any way) for quite some time. Now that I know the line, I'd really like to know the -individual-. I have a good idea who it might be, which is based really on not much more than lore. I'm looking for tests to support or cast doubt on whether or not my guess is accurate.

        Say I take 3 individuals. Two of them share great great grandfathers and the third shares a great great great grandfather. If you were charged with identifying which two shared gg grandfathers, how would you go about doing so?

        There are several (8-10 individuals, both male & female) in both lines available for testing (or who have already been tested); I am looking for any suggestions on ways to narrow down the possibilities to an individual.


        • #5
          Originally posted by RCrumpton View Post
          Say I take 3 individuals. Two of them share great great grandfathers and the third shares a great great great grandfather. If you were charged with identifying which two shared gg grandfathers, how would you go about doing so?
          Test the three Y descendants to the same degree at the same lab and hope for datable variances. If none turns-up test them all for additional STR's until datable variances are found.


          • #6
            Thank you sir!

            What is meant by the terms "datable variance"?
            Last edited by RCrumpton; 28 December 2012, 02:26 AM.


            • #7
              A difference in the number of repeats in an STR for which a rate of change has been established.

              For dating one needs a computation that yields a likelihood estimate of relationship within number of generations. FTDNA includes such computation for all matches in their Y database.

              The two should show a high likelihood of a shared ancestor at five generations and the singleton should show a high likelihood of being related to the two at six generations.


              • #8
                This is relevant: