Announcement

Collapse
No announcement yet.

Avoid the So-Called "Sibling" Tests

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

  • Avoid the So-Called "Sibling" Tests

    SAVE YOUR MONEY. I revised my web page on the DNA Sibling Test to strengthen my warning about this old-technology. The so-called "sibling" or "kinship" test is almost always inconclusive. And in my personal case it led me to the wrong conclusion. Read more...http://www.dna-testing-adviser.com/DNASiblingTest.html

  • #2
    Is this a CODIS type STR test ? Similar to what's offered by DNA tribes ?
    So the markers are labeled similar to Penta5, Thor ? I can't remember off hand the exact names. Just curious.

    Comment


    • #3
      Yes CODIS uses STR testing..

      Comment


      • #4
        Code:
        Yes CODIS uses STR testing..
        Hi Matt - yes the question I was asking though is this sibling test the op
        mentions using CODIS

        Comment


        • #5
          Yes. Richard used a STR based siblingship test in 2007. I used one myself in 2009 however only to confirm my sister and I were not full siblings, not to test for a parent. My method required the mother be included, Richard didn't have that opportunity hence the reason he says STRs are to be avoided and I agree. I would ONLY use a STR sibling or paternity test if one parent is known and tested too but that said I certainly see it is better to use FF instead; however back then we didn't have FF testing.

          SNP based FTDNA and 23andMe show much better results for siblingship because they use so many more data points as well as segment analysis. Full siblings will share a average of 2800cM while half siblings will share half of that in the FF test. This method is much more detailed than using 15 CODIS markers and trying to determine siblingship where you are not guaranteed so many markers will match, especially for half sibs.

          STRs use a lot of probability readings while with SNP segment comparison you will know without using probability that two of you share a bunch of DNA.

          Here's a good (though extreme) example. With a STR test two half siblings might share just a few markers depending on how they inherited alleles from their one parent. They might share the same amount of STR markers and be cousins too. With a FF SNP test they will share just about half of all of their segments summed in cM, while half of that if first cousins, and half of that if half cousins, leaving less doubt as to the relationship. So by the summed segment cm alone you can easily tell if a person is a full, half or cousin relation while with a STR test a few matching markers will reveal nothing definitive. It is much harder to be mislead with a FF test being used as a siblingship test compared to a STR siblingship test.

          Matt.
          Last edited by mkdexter; 19 June 2012, 05:48 PM.

          Comment


          • #6
            Originally posted by RichardH View Post
            SAVE YOUR MONEY. I revised my web page on the DNA Sibling Test to strengthen my warning about this old-technology. The so-called "sibling" or "kinship" test is almost always inconclusive. And in my personal case it led me to the wrong conclusion. Read more...http://www.dna-testing-adviser.com/DNASiblingTest.html
            Hi Richard, now I am very confused. Are you saying that FTDNA's Family Finder test is unreliable when testing for siblings, cousins, half cousins etc?
            Thanks

            Comment

            Working...
            X