Announcement

Collapse
No announcement yet.

Branch mutation example and question...

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

  • Branch mutation example and question...

    I have a question regarding mutations within different branches of the same main line. I'm very curious as to the plausibility of the following example that I came up with.

    Suppose a guy has a DYS458 value of 19. He has two sons (branch 1 and branch 2). One gets the 19 and the other has a mutation to 20 at DYS458. The son who got the 19 has several male descendants over a few generations, all of whom KEEP the 19 value at DYS458. We'll call this branch 1.

    Now back to branch 2. The son who had the DYS458 mutation to 20 has three sons. Two of them get the 20, but the third son has a mutation at DYS458 BACK TO 19. So what you end up with is this third son appearing to belong to branch 1 based on the DNA evidence, while ACTUALLY belonging to branch 2.

    This seems very plausible to me. Anyone else care to weigh in on this? I would love to hear other's opinions on this.
    jdchess
    Registered User
    Last edited by jdchess; 3 June 2008, 05:47 PM.

  • #2
    Originally posted by jdchess
    I have a question regarding mutations within different branches of the same main line. I'm very curious as to the plausibility of the following example that I came up with.

    Suppose a guy has a DYS458 value of 19. He has two sons (branch 1 and branch 2). One gets the 19 and the other has a mutation to 20 at DYS458. The son who got the 19 has several male descendants over a few generations, all of whom KEEP the 19 value at DYS458. We'll call this branch 1.

    Now back to branch 2. The son who had the DYS458 mutation to 20 has three sons. Two of them get the 20, but the third son has a mutation at DYS458 BACK TO 19. So what you end up with is this third son appearing to belong to branch 1 based on the DNA evidence, while ACTUALLY belonging to branch 2.

    This seems very plausible to me. Anyone else care to weigh in on this? I would love to hear other's opinions on this.
    This is theoretically plausible, especially with that particular marker, which is a rather volatile one (fast-mutating). Back-mutations do happen all the time. Statistically though, it's not very likely for two different mutation events to happen with any given marker within that short a time span. DYS458 may be fast-mutating, but no markers mutate that fast. Still, it is possible.

    Comment


    • #3
      Originally posted by GhostX
      This is theoretically plausible, especially with that particular marker, which is a rather volatile one (fast-mutating). Back-mutations do happen all the time. Statistically though, it's not very likely for two different mutation events to happen with any given marker within that short a time span. DYS458 may be fast-mutating, but no markers mutate that fast. Still, it is possible.
      That's a good point about the time frame, but I suppose I was just using that for the example. The split between the 19 and 20 values could have happened several generations back of where the example started, and branch 1 and branch 2 could have continued with 19 and 20 respectively for several generations before the back mutation to 19 (from 20) in branch 2. You would still have someone who appeared to belong to branch 1 while actually belonging to branch 2. This would lengthen the time frame and make it even more plausible, would it not?
      jdchess
      Registered User
      Last edited by jdchess; 4 June 2008, 03:41 PM.

      Comment


      • #4
        Originally posted by jdchess
        That's a good point about the time frame, but I suppose I was just using that for the example. The split between the 19 and 20 values could have happened several generations back of where the example started, and branch 1 and branch 2 could have continued with 19 and 20 respectively for several generations before the back mutation to 19 (from 20) in branch 2. You would still have someone who appeared to belong to branch 1 while actually belonging to branch 2. This would lengthen the time frame and make it even more plausible, would it not?
        Yes, Jdchess, you're right, hypothetically speaking, but we'd have to extend that timeframe back quite a bit further before we enter the realm of statistical likelihood. Since we're only talking hypothetically here, let's continue that DYS458 marker discussion a bit further...

        I've seen an estimated mutation rate for that marker listed as about 123 generations per mutation. So for that marker to mutate from 19 to 20, and then back again to 19, we're looking at an estimated 246 generations (over 6000-7000 years!).

        If we assume that the first mutation had already happened a long time before your example, as you said, we're still looking at at least 123 generations before the back mutation would occur, statistically speaking.

        That's just statistics though, and it's also possible that those two events could have happened in just 2 generations (though highly unlikely). Mutations certainly don't happen on a set schedule, and can happen anytime.

        This is all assuming that back mutations for that marker occur at the same frequency as forward mutations, which is probably not quite the case (as there is a tendency for long STR repetitions to shorten, and for short ones to lengthen), but it's probably close enough for the sake of this discussion.

        It's an interesting thought exercise.

        Comment


        • #5
          Anyone ever run across a real-world example of a similar situation?

          Comment

          Working...
          X