No announcement yet.

Which is the right test?

  • Filter
  • Time
  • Show
Clear All
new posts

  • Which is the right test?

    I have found someone that has the same unusual surname as my great-grandfather before it was changed. None of the surviving males will do a Y DNA test to see if we are related, even though I have offered to pay. I have been told if I have an autosomal test of daughter of a daughter of a daughter of one of my great-grandfather's sisters there should be enough within five generations that they will show up as 5th or closer cousins. Is this correct?
    Last edited by gnome788; 2 August 2017, 05:00 PM.

  • #2
    The rule of thumb for autosomal matches of 4th cousins (5 generations back to the common ancestor) is that about 50 percent of them won't be detectable, because they don't share enough DNA, due to recombination and the random assortment of chromosomes. So, while you might get lucky, the odds of being able to prove or disprove a genealogical hypothesis in this situation don't look good. If there are additional cousins who can be tested, that would increase the odds that some of them will be detectable as matches, but I'm not sure that would actually help you prove your case.


    • #3
      The right test.

      Mr McCoy, I find your response interesting because I was told by a Biology Professor that this is the route to take given the circumstance that the direct males refuse to do testing.


      • #4
        Just to clarify your situation, note that a "daughter of a daughter of a daughter of one of my great-grandfather's sisters" would be your third cousin. (You share a pair of great-great-grandparents.) On average, two third cousins share 53 cM in an autosomal test, but just in my own small family, I've seen a third cousin match as high as 140 cM, while at the same time, I have scores of people matching me up into the low 90s, and we can't find any sort of connection. Then there's the phenomenon of siblings whose match lists overlap only partially.

        So I agree with McCoy that testing one cousin won't give you a reliable answer, and that testing more such cousins will get you closer.


        • #5
          I agree - FamilyFinder is the way to go! Definitely, unequivocally!!

          However, the others are correct also. So - be willing to live with it if you spend your money and still do not know.

          If you and the other person are great-grandchildren of siblings, then you are third cousins. There is about a 90% chance the DNA will show the relationship (or at least, a relationship).

          I just recently tested someone (say XX) I believed was my 3rd cousin -- annoyingly, we do not show on each others matches at all. But a grandchild (say AA) of my relevant greatgrandparent had also tested, as had a grandchild (say BB) of another sibling of that greatgrandparent. XX and AA share 155 cm, longest segment 35 cm -- XX and BB share 218 cm, longest 32 cm -- so I am pretty sure we (XX and I) actually are 3rd cousins.
          Just recently also had a paper-trail known 3rd cousin test - she matched me fine, but of about a half-dozen second cousins to me to whom she is also 3rd cousin, one of them shows as not a match to her.

          If you match nice and clearly, then you know. If you do not match, then do not know, and you wait and search and hope to find some other way or person to test or etc. But for now, in my opinion, taking that test is best bet towards finding answers.
          Last edited by loobster; 8 August 2017, 04:16 PM.