Announcement

Collapse
No announcement yet.

DNA and Prosecutions

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

  • DNA and Prosecutions

    So I've always wondered, when it comes to the courts and DNA how many "points"; 12, 25, 37 or 111, do they test for?

  • #2
    The standard DNA test used in law enforcement investigations and court cases is called CODIS - see https://isogg.org/wiki/CODIS. This test involves 13 autosomal STRs, not yDNA STRs. It's also the same type of test used in paternity testing.

    There may be exceptional cases where a law enforcement agency uses yDNA testing to compare to yDNA databases. This will usually identify either the suspect or a close male relative of the subject, when analyzed properly.

    Law enforcement could also use an autosomal test like Family Finder, but they seem to be content with 13 CODIS markers. Family Finder would be more definitive than CODIS in identifying suspects or close family members of suspects. But it seems that law enforcement hasn't caught on to what genetic genealogists and adoptees have been using successfully for the last several years.

    Comment


    • #3
      Originally posted by MMaddi View Post
      There may be exceptional cases where a law enforcement agency uses yDNA testing to compare to yDNA databases. This will usually identify either the suspect or a close male relative of the subject, when analyzed properly.
      I should have written "This will usually identify either potential suspects or close male relatives of potential suspects..." Since mutations are based on probability, it's possible, although unlikely, that a 37 or 67 marker exact match may be coincidental and identify someone from the wrong paternal line as a suspect.

      As far as the original question, in cases where law enforcement uses yDNA testing, I don't know how many markers tested they would require to regard someone as a suspect. I certainly hope that they would require at least a close 37 marker match or, better yet, a close 67 or 111 marker match.

      Comment


      • #4
        Originally posted by MMaddi View Post
        As far as the original question, in cases where law enforcement uses yDNA testing, I don't know how many markers tested they would require to regard someone as a suspect. I certainly hope that they would require at least a close 37 marker match or, better yet, a close 67 or 111 marker match.
        For better or for worse, law enforcement typically relies on self-proclaimed experts, who may not actually be so. This article describes one particular incident.

        - A self-proclaimed expert pointed to a particular family as suspect based solely on the SMGF database, whose marker panel has a resolution even lower than FTDNA's Y-DNA37. The alleged suspect mismatched at least one marker.

        - The police obtained a secret warrant and deceived the alleged suspect into entering a police station, where they drew a DNA sample. He was never given the opportunity even to consult a lawyer, much less contest the warrant.

        - Even after DNA evidence exonerated the alleged suspect, the self-proclaimed expert refused to admit her error in over-interpreting a weak match.
        Last edited by lgmayka; 30 April 2017, 07:31 AM.

        Comment


        • #5
          Originally posted by MMaddi View Post
          As far as the original question, in cases where law enforcement uses yDNA testing, I don't know how many markers tested they would require to regard someone as a suspect. I certainly hope that they would require at least a close 37 marker match or, better yet, a close 67 or 111 marker match.
          As mentioned above, Y-DNA testing is seldom used. The forensic databases for Y (with results gathered at http://yhrd.org) use some standard mutiplex kits: Y Filer (17 markers), Y Filer Plus (27 markers) or PowerPlex (23 markers). The YHRD database shows the enormous haplotype diversity found with even this limited set of markers, so it is a powerful exclusionary technique.

          There is a very recent article about how to estimate the frequency of a particular profile.

          http://biorxiv.org/content/early/2017/04/28/131920

          Comment

          Working...
          X