Announcement

Collapse
No announcement yet.

Example where a 12 marker match was misleading and paternal ancestry wrong?

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Example where a 12 marker match was misleading and paternal ancestry wrong?

    Is there a documented instance where two males with the same surname and a known shared direct paternal line ancestry within 20 generations matched at 12 markers, but were later found to not match match sufficiently at 25 or more markers for their known direct paternal line relationship?

    That is to say their additional STR markers revealed they were not related in a genealogical time frame and so their direct paternal line ancestry was mistaken and their 12 marker match was a fluke.

    Thank you and sincerely, Peter

  • #2
    This depends on how many markers you are referring to.

    At 67 or at least 111 markers, they should match even if the GD distance is high.

    My Father and his paternal 1st cousin are a GD of 4 at the 67 marker level.

    before assuming paternal ancestry for one is incorrect, deeper testing maybe need. Whether it is testing both for 111 marker or both with Big Y.

    Have you talk to any of your Surname or haplogroup admins?

    Comment


    • #3
      While not sharing a common surname, I have 12 Marker GD0 matches with people who are R-Z94 descendants(SNP tested) while I am descended from R-L260(SNP tested). Most of my fellow L260's are GD1, GD2, or even further out at 12 markers in my case. So having GD0 matches in general at 12 markers who are not related anytime within the past several thousand years is certainly in the realm of the possible.

      Comment


      • #4
        Hello prairielad and bartarl1260, The original question relates to 12 markers AND sharing the same surname AND a known shared direct paternal line ancestry with a most recent direct paternal line ancestor within 20 generations. Additional testing (e.g. Y-DNA37) showed they were unrelated and so that known direct paternal line ancestry was incorrect.

        Comment


        • #5
          I have a similar situation. We have a common ancestor and our surname is only one letter different, but we don't match at 25 or 37. I haven't tried asking them for their exact numbers to do a manual check. I wish it was already available.

          Comment


          • #6
            Hello Serket, Is there a known ancestry back to their most recent direct paternal line ancestor?

            Comment


            • #7
              Originally posted by peterebay View Post
              Hello prairielad and bartarl1260, The original question relates to 12 markers AND sharing the same surname AND a known shared direct paternal line ancestry with a most recent direct paternal line ancestor within 20 generations. Additional testing (e.g. Y-DNA37) showed they were unrelated and so that known direct paternal line ancestry was incorrect.
              I haven't heard of any case like that, but it always possible. I think a key factor is how strong the genealogy is. If the genealogy is well supported, then I think the odds are very pretty good that they do share a common ancestor. However, once you go back 400+ years, a lot of genealogies get very speculative and based on wishful thinking. In the Ashley surname group, there are men with the same surname who both claim ancestry from the same ancestor in 1660 Virginia. However, the genealogies back to the ancestor are very speculative. Although STR tests show they are not a match, even at 12 STRs, they are both M269, so it does not seem farfetched that they could have matched at 12 and yet not been related.
              Last edited by TwiddlingThumbs; 31st October 2018, 03:37 PM.

              Comment


              • #8
                Originally posted by peterebay View Post
                Hello Serket, Is there a known ancestry back to their most recent direct paternal line ancestor?
                I don't remember if I saw a full tree going back to the common ancestor, but the common ancestor is one of the first people in America with our surname.

                Comment


                • #9
                  Originally posted by peterebay View Post
                  Hello prairielad and bartarl1260, The original question relates to 12 markers AND sharing the same surname AND a known shared direct paternal line ancestry with a most recent direct paternal line ancestor within 20 generations. Additional testing (e.g. Y-DNA37) showed they were unrelated and so that known direct paternal line ancestry was incorrect.
                  IIRC, Y37 cuts out at GD4, which under certain circumstances can be achieved as quickly as 2 generations later(1st cousins, if my math is right, it is improbable but it is not impossible) if you assume 1 mutation per generation for each person involved. FTDNA currently does poorly at weighting "high mutation" markers against "low mutation" ones. From what I have heard: They don't. That's what the report generation tools are for.

                  What you really need to be looking at is the SNPs, and what the STR results predict SNP wise. If the SNP predictions wildly diverge, or more particularly, the SNPs don't match when tested, then relatedness is distant.

                  There are "terminal SNP" matches who fail to match on the STR filtering criteria that FTDNA is using, sometimes at ANY level, never mind on the 12 marker vs 111 marker tier.

                  Now granted, completely failing to get even close on the STRs at every level, even with a "close" SNP match, is a strong indication the relationship is (very) distant all the same. It doesn't change the matter that things aren't as reliably straightforward as many would like it to be.

                  Basically the "quick and dirty" way to (dis)prove a link is spend $40 on a single SNP test for one of the nearest terminal markers available(just be mindful of which "near" one you pick) and see if they match there. If they don't match, they're not recently related, if they do, you might want to dig deeper on the STR side. Of course that means you need to have completed either a SNP Backbone pack(and/or more) or have a BigY result to compare SNPs against.

                  Comment

                  Working...
                  X