Announcement

Collapse
No announcement yet.

V4 chip

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

  • V4 chip

    Is the 23andMe V4 chip still incompatible with FTDNA? My wife & son tested with 23andMe under the V4 chip. I'd like to transfer their atDNA to FTDNA, but I've read this is not possible.

  • #2
    Retest them, if that is possible.

    W. (Mr.)

    Comment


    • #3
      Originally posted by dna View Post
      Retest them, if that is possible.

      W. (Mr.)
      I believe that you mean the OP should have them submit a DNA sample to FTDNA for the Family Finder test, since the 23andMe V4 chip is still incompatible for a raw data upload to FTDNA.

      Unless the top management at FTDNA changes their mind about the V4 chip, it will remain incompatible. The number of SNPs tested in the V4 chip (reduced from the V3 chip) is too far off the number that the FTDNA algorithm uses to reliably find matches in the Family Finder database. FTDNA doesn't feel comfortable declaring a match with a reduced number of SNPs to compare.

      Comment


      • #4
        Originally posted by MMaddi View Post
        I believe that you mean the OP should have them submit a DNA sample to FTDNA for the Family Finder test, since the 23andMe V4 chip is still incompatible for a raw data upload to FTDNA.

        Unless the top management at FTDNA changes their mind about the V4 chip, it will remain incompatible. The number of SNPs tested in the V4 chip (reduced from the V3 chip) is too far off the number that the FTDNA algorithm uses to reliably find matches in the Family Finder database. FTDNA doesn't feel comfortable declaring a match with a reduced number of SNPs to compare.
        Thank you!

        Yes, I meant please retest with Family Finder at FTDNA, as that not only allows matching at FTDNA, but also allows further testing (not available at 23andme) if desired.

        W. (Mr.)

        Comment


        • #5
          Originally posted by NCroots View Post
          Is the 23andMe V4 chip still incompatible with FTDNA? My wife & son tested with 23andMe under the V4 chip. I'd like to transfer their atDNA to FTDNA, but I've read this is not possible.
          According to Felix :


          There are only 312257 overlap SNPs between FTDNA and 23andme V4. Not enough for comparison in FTDNA.

          Comment


          • #6
            Thank you all for your replies. I still don't quite understand about the V3 and V4 chip. When I first had my DNA tested at 23andMe it was with the V3 chip (my sister-in-law also tested with the V3 chip). We noticed that our results changed when 23andMe switched to the V4 chip. Originally I showed 5% Italian and my sister-in-law showed .5% Sub-Saharan African. When 23andMe changed to the V4 my results changed from 5% Italian to .1% Sub-Saharan African and my sister-in-law's from .5% Sub-Saharan African to .1% East Asian.

            Then my wife & son tested with the V4 chip. My son does not show any Sub-Saharan African, but my results again changed with Sub-Saharan African showing up in all the values (where previously it had only been in the "speculative" view). Likewise, my wife showed Native American, and my sister-in-law's changed to Japanese! (in all values).

            I'm not exactly sure what the change means, but I like having the FTDNA results for comparison. Perhaps I'll have my wife & son test with FTDNA too.

            Comment


            • #7
              Originally posted by MMaddi View Post
              I believe that you mean the OP should have them submit a DNA sample to FTDNA for the Family Finder test, since the 23andMe V4 chip is still incompatible for a raw data upload to FTDNA.

              Unless the top management at FTDNA changes their mind about the V4 chip, it will remain incompatible. The number of SNPs tested in the V4 chip (reduced from the V3 chip) is too far off the number that the FTDNA algorithm uses to reliably find matches in the Family Finder database. FTDNA doesn't feel comfortable declaring a match with a reduced number of SNPs to compare.
              I created an experimental file with v4 data loaded into a v3 template and uploaded it to Family Finder. The main difference I see is that the reduced number of SNPs creates a larger number of small 1-3 cM segments, which are likely to be IBS. The longer segments seem very similar, and the experimental file included all but two out of 843 matches with my regular kit.

              The additional small segments actually "fix" some of the false negatives in cases where a match in a child is not found in either parent, since they increase the total cM count to the (presumed) 20 cM threshold.

              I thus believe FTDNA could accommodate v4 files (and v2 files, too), but it would require adjusting their algorithm to exclude small segments. The calculations required to redo all the pairwise comparisons might be prohibitive, but it would benefit other people, too, e.g. African Americans who receive a miniscule number of matches because it's harder for them to reach the 20 cM threshold.

              Comment


              • #8
                Originally posted by Ann Turner View Post
                I created an experimental file with v4 data loaded into a v3 template and uploaded it to Family Finder. The main difference I see is that the reduced number of SNPs creates a larger number of small 1-3 cM segments, which are likely to be IBS. The longer segments seem very similar, and the experimental file included all but two out of 843 matches with my regular kit.

                The additional small segments actually "fix" some of the false negatives in cases where a match in a child is not found in either parent, since they increase the total cM count to the (presumed) 20 cM threshold.

                I thus believe FTDNA could accommodate v4 files (and v2 files, too), but it would require adjusting their algorithm to exclude small segments. The calculations required to redo all the pairwise comparisons might be prohibitive, but it would benefit other people, too, e.g. African Americans who receive a miniscule number of matches because it's harder for them to reach the 20 cM threshold.
                That's interesting Ann, both in terms of your experiment and the alleged 20 cM threshold.

                If they could accommodate the v4 chip somehow it would help grow the database so everyone wins. I finally got caught out by the alleged 20cM threshold myself this month with a paper 4th cousin. We get over 18 cM total on Gedmatch with a 8.4 cM biggest segment which seemed sensible for a paper fourth cousin. Of course being below 20 cM total, FTDNA does not show it as a match so I can't run the in common with tool here. It does seem odd that if we had has an additional tiny 2cM IBS segment in common FTDNA would then have declared a potential match. We are going to test more additional close relatives anyway so I am hoping to break the 20cM with some of them but it's frustrating not to be able to use the FTDNA tools with matches over 8cM biggest due to the algorithm.

                Earl.

                Comment


                • #9
                  Is a combined file possible?

                  If one had tested at all three vendors, would it be possible (and reasonable) to create a sort of meta file combining all three sets of results? I'm wondering if might have better matching say at gedmatch.com? Rather curious.

                  Comment

                  Working...
                  X