Announcement

Collapse
No announcement yet.

novice mistrusts FTDNA. any comments?

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

  • #16
    Thanks I appreciate sharing you sharing your insights vineviz but are you saying it's not valid to take the probabilites that FTDNA gives us for comparing 2 individuals and projecting them upon an abitrarily larger group of applicable individuals?

    Comment


    • #17
      Originally posted by Bertp
      Thanks I appreciate sharing you sharing your insights vineviz but are you saying it's not valid to take the probabilites that FTDNA gives us for comparing 2 individuals and projecting them upon an abitrarily larger group of applicable individuals?
      I think that is what I'm saying. The MRCA calculations are for pair-wise comparisons, and the math in that circumstance is challenging enough that I wouldn't want to derive the equation for your situation.

      I'm also saying that looking ONLY at the 37/37 matches is not as insightful as looking at ALL the tested descendants of the same man. In addition to the eight perfect matches, it looks to me like there are at least nine (and maybe eleven) members of the same family in the project. In aggregate, these 17 (or 19) participants have ten or more differences from the modal, which is about what you'd expect to see.

      I'm attaching a network diagram I built using Fluxus software that shows the most parsimonious series of mutations for these possibly related Pitmans from your project. I excluded a couple of kits that were obviously unrelated, but there are still a couple (19535, 62894) that are probably not related and a few (52981, 31591) that might not be but likely are related.

      As I alluded to earlier, the only truly accurate way to evaluate your projects's results is to build a full family tree and map the mutations to the tree. Failing that, everything else is just a guess but the math I used in my prior post suggests that the results you see are consistent with these 17 or 19 folks having a common ancestor 200 to 400 years ago.
      Attached Files

      Comment


      • #18
        Thanks for the network diagram. I don't have access to kit numbers other than my own but I will be sure to forward this to our project leader.

        Comment


        • #19
          Originally posted by Bertp
          Thanks for the network diagram. I don't have access to kit numbers other than my own but I will be sure to forward this to our project leader.
          This is a very interesting project, and I agree with the notion that it seems a good candidate for the 67 marker test. And I hope that the paper trail will catch up to the DNA so you can start forming branches instead of just clusters.

          Keep in mind that the modal haplotype is not necessarily the founder's haplotype, even though the diagram software makes it look like it is.

          Thus, the structure of the diagram does not necessarily reflect the historical chain of mutations in their actual order. It does, however, illustrate which haplotypes are probably most closely related.

          Comment

          Working...
          X