This article addresses serious questions in DNA science, but gets the science completely wrong and indicates a much larger issue. If this guy, who hopefully did some research before writing this article, cannot figure out what is going on, how are jurors supposed to do so?
First, the analysis of how familial DNA searches are conducted is seemingly not accurate. I don't know what happened in the particular cases he referenced, but a familial DNA search is not based on the “20 critical markers”. And it is absolutely false that matching “10 of 20 critical markers strongly suggests a person is the parent, child or sibling of the suspect.” That is just nonsense and a gross misstatement of DNA science, period.
Second, The familial DNA searches are an examination of the mtDNA (mitochondrial DNA) and/or Y-STR DNA, as opposed to the normal 13-21 markers used in ‘regular' forensic searches. The replication and analysis of mtDNA and Y-STR DNA are different processes than that of ‘regular' nuclear DNA (scientifically the same, but different targets). The mtDNA is not even in the nucleus of the cell but rather in the mitochondria, organelles within our cells.
It so happens that mtDNA is transferred through the matrilineal line in a family, notwithstanding any mutations, and persons can be asserted to be descendants/ascendants of a particular family if their mtDNA matches the females of the same line. Y-STR is the same for the patrimonial line. But these DNA markers have nowhere near the resolution power of ‘regular' nuclear DNA testing results, and should be considered only within the framework of the whole of the evidence in a case.
Third, as for the FBI CODIS database, the reason law enforcement can't use the CODIS database to do familial searches is because it contains the profiles generated through the analysis of the ‘regular' 13-21 nuclear DNA, and not either mtDNA or Y-STR DNA profiles. Familial profiles, mtDNA and Y-STR DNA have not been included in the database and are not regularly performed in crime lab analysis.
Lastly, the author failed to mention the real problems with the Ursy case which were, (1) the law enforcement authority releasing his name as a suspect when they found a familial DNA “match”. It cannot be overstated how personally and professionally devastating it would be to have your name connected to a rape/murder ‘by DNA'. I do not know, but can imagine, this ‘mistake' is still affecting this man and that is a gross injustice. And, (2) doesn't this case beg the question how did they have a “match” for his familial DNA and yet prove conclusively it was not him or any member of his family? This goes back to the difference between the resolution power of familial DNA profiles and ‘regular' nuclear DNA profiles. All the more reason to be responsible about how we use technology and the ever growing field of forensic DNA profiling.
To this end I would assert that not all DNA analysis has the same power for identification and not all DNA results are equal, even among the same analysis. We must consider this evidence as it is and not as we wish it to be.