Announcement

Collapse
No announcement yet.

Small match in X + "4th cousin" autosomal significance

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

  • Small match in X + "4th cousin" autosomal significance

    Ok, I have 3 matches against my solitary X chromosome from my FF, aka autosomal, test. They are on the precise same segment, but only about 1.65 cM. Two of those matches were put up by one person. Looks like they are 1st cousins to each other perhaps. All three also match on another chromosome at about 14 cM and 8 cM respectively for the longest block. None of them appear to be actual "4th cousins" or 5th cousins. My guess would be between 6th and 8th cousins.

    Since they have some match, however tenuous on my X choromosome, I think I'm justified in concluding they are related through my mother's side of the family. What are the odds though that I would get 3 matches on exactly the same 1.65+/- cM?

  • #2
    Isn´t 1.65 a bit too small to be able to conclude anything?

    Comment


    • #3
      Originally posted by LynCra View Post
      Isn´t 1.65 a bit too small to be able to conclude anything?
      Not if it is between kits that Family Finder has deemed matches to yours. They are good as gold. Most experts/professionals have a misunderstanding of the probabilities involved in this situation. They treat each small segment as an independent test. Family Finder is conservative in deciding who matches who. If you have a FF finder match, not too far back you have an ancestor that matched virtually everything with an offspring. The question that you are actually testing is what are the chances of a remnant of that match surviving between two cousins. It is not a question of whether a short matching segment between too people is enough to show that they are related. We already know that. Think conditional probability, where what you already know changes what the probability is. A simple example is if you are part of a group that has a life expectancy of 76 years at birth. When you are seventy-five, your life expectancy is not one more year, but much more than that.

      When I use the FF Chromosome Browser, the first thing I do is to change the segment size to 1 cM. I see no reason not to.
      Last edited by georgian1950; 9th May 2015, 04:21 AM. Reason: clarity

      Comment


      • #4
        I sincerely hope that you are not going to hijack yet another thread with your theories.

        Comment


        • #5
          Originally posted by georgian1950 View Post
          When I use the FF Chromosome Browser, the first thing I do is to change the segment size to 1 cM. I see no reason not to.
          Please ignore georgian1950. He/she is just plain wrong. Below are actual case studies that completely contradict georgian1950. Note "There is no debate in the genetic genealogy community that many small segments are false positive matches."

          See http://www.yourgeneticgenealogist.co...gments-as.html and http://www.thegeneticgenealogist.com...ts-friend-foe/
          Last edited by thetick; 9th May 2015, 09:20 AM.

          Comment


          • #6
            False positives

            Originally posted by thetick View Post
            Please ignore georgian1950. He/she is just plain wrong. Below are actual case studies that completely contradict georgian1950. Note "There is no debate in the genetic genealogy community that many small segments are false positive matches."
            I totally get that small segments aren't a controlling factor. But that wasn't really my question. This is one identical small matching segment displayed by three different people who also match me on a much larger segment (>10cM) in other chromosomes. Granted these look to be remote cousins. All my matches look to be greater than 3rd cousins and likely all greater then 4th cousin. The combined cMs for the best are about 60 cM, and these three are in the 30s. I've been restricting my views to consider only matches that have at least one single 7 to 10 cM at least match.

            It just seemed odd, that 3 and exactly 3 out of 500+ would match my X chromosome and that they all matched in the identical place and nearly identical size. Considering two look to be very closely related, but not immediate family really pushes it to basically 2 matches on the X chromosome.

            An odd coincidence?

            Or is it reasonable to assume these small matches make it more likely they are a match from my mother's side of the tree? With a segment match of 8 or 14 cM on other chromosomes, they clearly are a match of some degree. At least from all I read.

            Thanks,
            Brian

            Comment


            • #7
              Are these matches female or male?
              What about the "Two of those matches were put up by one person"(Both female, both male, or one female and one male)

              Reason I ask is that if one is a male, you will know that algorithm is not making a IBS segment by creating a segment by matching your X with values out of the females 2 separate X's.
              Male to Male X matches will be a little more precise, not saying this 1.65cM X match is valid, just saying it will just strengthen the chance it may be.

              Comment


              • #8
                False positives

                Two females and one male. The two put up by one person, or managed by one person, has the male. The other female doesn't actually match the two who are obvious cousins (they share many of the same surnames but not all). The two cousins and the lone female don't match me in any of the same segments, no matter how small, except on this one segment. Based on the links about warning about using small segments, I gather I have a less than 1/3 chance this might be relevant. Given these two cousins match me in a 14 cM segment, I approximate they are 7th to 9th cousins. If that is even really determinable from autosomal tests.

                On my mother's side have most of that tree complete in depth to 11 generations and beyond. Which should be fairly easy to fan out and drive forward to locate the conjunction. I have many of the necessary parish registers digitized and stored on my computer.

                My father's tree is mostly a nightmare. The 500+ matches I expect are mostly on my father's mother's father's tree. But I do see pinpricks of light here and there, and what started out a test just for grins, may actually prove to be a useful tool. If I can get enough known cousins in the right places of the tree to also do the test. One down, seven to go.

                At least, once I have a handle on what is and what isn't real.

                Comment

                Working...
                X