Collecting DNA: database mission creep.

SINCE DATABASES of DNA taken from convicted criminals have proven such a success in forensic investigations, why not create a universal one covering all citizens? So suggested James Hodge, director of the Center for Law and the Public Health at Johns Hopkins University, during a February forum sponsored by the National Academies, an advisory body for national science policy.

That isn't a serious policy proposal in D.C.--yet--but don't count it out. Forensic DNA databases, the first of which was created in 1989, initially included only violent felons for whom such evidence could be crucial, such as rapists and murderers. But they have expanded to include all felony prisoners, many people on probation, and, in most states, all juvenile offenders and all people convicted of misdemeanors. At least four states take DNA samples from everyone arrested for any reason, even before conviction. In January President George W. Bush signed a law authorizing the reds to do the same with their arrestees and noncitizen detainees.

These DNA samples are used to create computerized searchable "profiles" in the FBI's Combined DNA Index System, which contains nearly 2.8 million such profiles. The samples themselves, which are available for re-analysis at any time, can reveal all sorts of personal information, including genetic predispositions to diseases or addictions.

Testing labs already have enormous backlogs of samples waiting to be turned into profiles, and such overloading can lead to screw-ups. In one of the more notorious cases, a sloppy lab in Houston produced a bad DNA analysis that falsely convicted accused rapist Josiah Sutton in 1999.Widening the DNA database to include all arrestees, much less all Americans, greatly increases the chances of such bad forensic science.