Announcement

Collapse
No announcement yet.

New Phasing Tool at Gedmatch

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

  • #31
    You can't download the file but you can use the ID's given at the end of the process to do anything you can do with your own ID. That includes looking for matches with the phased data but you'll have to wait 24 hours or so for the analysis to happen.

    You can do admixture analysis without any waiting.

    Comment


    • #32
      Ah, got it! Thanks!

      Comment


      • #33
        Originally posted by thetick View Post
        This is just meaningless paranoia. What can anyone possibly do with some DNA markers? Now a complete genome would be a different story.

        Posting ydna markers is like posting two digits of your credit card number... not useful to anyone. BTW I have a MasterCard that starts with 84. Oh no what will happen now?
        Now? well now you just give us the next two numbers.. and then next week.....

        Just an FYI darroll. Your Internet usage is all tracked and recorded by the government, Google, your service provider, and a whole lot of other companies...... ..... now you have more to worry about.

        Matt.

        Comment


        • #34
          Hello all,

          Can we please keep the posts relevant to the original post (phasing). Thank you much.

          -Darren Marin
          Family Tree DNA

          Comment


          • #35
            Originally posted by 1_mke View Post
            You can't download the file but you can use the ID's given at the end of the process to do anything you can do with your own ID. That includes looking for matches with the phased data but you'll have to wait 24 hours or so for the analysis to happen.

            You can do admixture analysis without any waiting.
            So if I am understanding this correctly, you can phase any two files and the resultant file contains the data that is not in common. You then need to determine whose data that is, which is a function of who you phased.

            For example,

            Say I have 3 kits, child, father and maternal grandmother.

            Step 1 - phase child and father --> resultant file (File #A) is mother's data contributed to child (without any matching data).

            Step 2 - phase #A against maternal grandmother --> resultant file (File #B) is ma data that child is carrying from maternal grandfather.

            In this example, using child, one parent and one grandparent. We were able to separate data from second parent and another grandparent.

            Am I missing something?

            Comment


            • #36
              Originally posted by BenTurin View Post
              Say I have 3 kits, child, father and maternal grandmother.

              Step 1 - phase child and father --> resultant file (File #A) is mother's data contributed to child (without any matching data).

              Step 2 - phase #A against maternal grandmother --> resultant file (File #B) is ma data that child is carrying from maternal grandfather.

              In this example, using child, one parent and one grandparent. We were able to separate data from second parent and another grandparent.

              Am I missing something?
              Step 2 might cause a problem. It depends on how JSW setup his program. I guess the first question is..... does it work for you?

              What I imagine will happen is that you will remove all paternal alleles from the child's kit in step 1, that leaves maternal alleles only and then you are going to remove about half of the maternal alleles (grandmother's) but will have nothing left so you may end up with a kit containing maternal grandfather's alleles (say 1700cm worth) and null data for the rest or just another copy of the mother's alleles which will result in no actual phasing.

              It depends on how the program was setup. If I had a copy of the Pxxxxxx raw files I could tell you what would happen.

              Try it!

              This would be stretching things. The idea is to remove half of the allele pair in each SNP but in the case of a grandparent you are not dealing with a pair of alleles only, you are dealing with half of the cm match too.

              Matt.

              Comment


              • #37
                Originally posted by mkdexter View Post
                Step 2 might cause a problem. It depends on how JSW setup his program. I guess the first question is..... does it work for you?

                What I imagine will happen is that you will remove all paternal alleles from the child's kit in step 1, that leaves maternal alleles only and then you are going to remove about half of the maternal alleles (grandmother's) but will have nothing left so you may end up with a kit containing maternal grandfather's alleles (say 1700cm worth) and null data for the rest or just another copy of the mother's alleles which will result in no actual phasing.

                It depends on how the program was setup. If I had a copy of the Pxxxxxx raw files I could tell you what would happen.

                Try it!

                This would be stretching things. The idea is to remove half of the allele pair in each SNP but in the case of a grandparent you are not dealing with a pair of alleles only, you are dealing with half of the cm match too.

                Matt.
                Thanks Matt,

                I have the kits for the scenario described on order, but only have the results from my own kit at this point. The others were ordered during the sale.

                Comment


                • #38
                  When you phase kits (right now you can only do parent/child phasing) you get a P1 (paternal) file and a M1 (maternal) file. This happens whether you use one parent or both parents.

                  I doubt if you could phase a phased file.

                  In your example phasing father and child would produce

                  PchildkitP1 and PchildkitM1 where "childkit" is the child's GEDmatch kit #

                  Then your plan is to take PchildkitM1 and phase it against the maternal grandmother. I don't think you can do that. What you can do is compare the M1 file to the maternal grandmother's kit to see which segments came from the maternal grandmother (and note where lack of a matching segment means that stretch of child's DNA came from the maternal grandfather) but you could do that without phasing just by using the chromosome browser at FF.

                  I would love to be able to phase half-siblings and other family, not just parent/child.

                  Comment


                  • #39
                    Half sibs are already phased by the fact they only match one parent.

                    Comment

                    Working...
                    X