Hello Anne, RE:  > "The Chen case you mentioned was based on Y STRs, not CODIS (autosomal STRs)." Thank you. And the court convicted Chen because it was told that he was a DNA match. While we have "a" probability (theory) on CODIS markers, I do not know what that would be for Y-DNA, although I understand that that has been worked out by halpotype group by some organizations. I believe that probability (of CODIS marker identification, one in 3 million) would be meant to apply to unrelated individuals in a population. I would not understand if 3 million out of 300 million would be the equivalent of one in 100.... My thoughts were more along the lines of what was missing in this discussion was the chance that two related people would have identical DNA profiles. For example, we do not know how many 4th cousins may share identical CODIS markers, as I do not think that topic has ever been studied in any depth. Which, of course, leads me to have problems with familial testing. The thought occurred to me most here would not know why the CODIS markers were upgraded to 20 markers. If memory serves me, I think I read somewhere that there were 250 CODIS type markers available. At any rate, I am not sure that statistics on CODIS marker accuracy are kept anywhere. The Innocence Project states that 45% of their exonerations involved misapplication of forensic science, but DNA is only a subset of that statistic. Which is to say, although statistics have been used for formulating CODIS markers (for use), I have not yet seen statistics regarding performance under practical use. If interested in the stats on which markers to use, see "NIST builds statistical foundation for next-generation forensic DNA profiling" https://www.sciencedaily.com/releases/2018/07/180723132034.htm So now, here we are, with law enforcement using CODIS asking us to believe that a person is guilty because that person may be a match. Much like the Chen case. My thoughts are that we do not know much more about CODIS markers than the court knew about Y-DNA (in the Chen case). Personally, I did not want to discuss microscopic DNA, because when I start looking at that stuff, it gets messy very quickly. There is a good paper by John Butler regarding his thoughts on forensic DNA analysis: "The future of forensic DNA analysis" https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4580997/  - Dave Hamm On 11/27/2018, Ann Turner wrote: > The Chen case you mentioned was based on Y STRs, not CODIS (autosomal > STRs). There was the additional complicating factor of the sample having a > mixture of contributors. The analysis of DNA mixtures is controversial in > general. Another problematic sample type is low copy number DNA ("touch" > DNA). Prosecution of cases like this would be challenged by defense > attorneys, and convictions should be based on the weight of other types of > evidence. > > Ann Turner > > On Tue, Nov 27, 2018 at 6:14 AM Dave Hamm via GENEALOGY-DNA < > genealogy-dna@rootsweb.com> wrote: > >> WJohnson, >> >> Easy for you to say, without citing any sources. >> >> One of the examples I gave was regarding the CODIS markers, which is the >> DNA used for conviction in court. >> >> That happens regardless of the type of DNA samples have been examined. >> >> The CODIS markers convicted Chen, and the jury ignored the remaining >> evidence. >> >> "Five years later, Chen was exonerated when a second DNA test that found >> he was not a match after all. In the years he lived as a convicted >> rapist, he had lost his wife, his business and most of his life." >> >> Gizmodo: >> >> "When Bad DNA Tests Lead to False Convictions" >> >> https://gizmodo.com/when-bad-dna-tests-lead-to-false-convictions-1797915655 >> >> His original 17 genetic markers were overturned when testing more >> markers showed that he was not a match. >> >> This was not sloppy police work, this was the presumption that DNA >> analysis is accurate to within millions of suspects. This was at a time >> when 13 CODIS markers were believed to be accurate to one in millions. >> >> In Chen's case, CODIS was not accurate to within one bar that night. A >> cynic could presume that means nearly doubling the number of markers to >> 20 CODIS markers would not be not accurate to the people within two >> night clubs. (... snip)