Wednesday, August 06, 2008

DNA SNAFU: Questions erupt about strange matches in FBI's DNA database

Ron Franscell, author of the bestselling true-crime THE DARKEST NIGHT, will be the guest on Burl Barer's Internet radio show at 4 p.m. CDT Saturday (8/9). Listen on your computer by clicking on OutlawCrime.com

By Ron Franscell

You might think from watching the dizzying array of TV crime fare that DNA evidence is the incontrovertible defense killer (or in some cases, the golden key to the jailhouse door for wrongly convicted inmates). In most cases it is definitely the most trustworthy evidence ... except that for the past 7 years, questions have been rising about matches in the FBI's central database that defy the odds and send a little quiver through our faith in this science as a prosecutorial tool.

It all began in 2001, when an Arizona crime lab worker tested the state's DNA database and found two felons with similar genetic profiles. Remarkably, they matched at 9 of the 13 locations on chromosomes, or loci, commonly used to distinguish people from each other. In court, a DNA expert would say that the chance of these two men sharing these same markers would be 1 in 113 billion -- or nearly impossible.

But these two men did. And they weren't related: one was black and one was white.

Crime labs began conducting other searches. In 2 states, nearly 1,000 such cases were found where two criminals matched at 9 or more "loci."

This weekend, the Los Angeles Times reported that this surprising discovery has ignited a legal fight in which the FBI is trying to block similar searches and forestall even court-ordered inquiries into its DNA database known as CODIS (Combined DNA Index System). The FBI asserts the data was misleading and misrepresented, and further mucking around in its system will simply harm crime-fighting. The FBI has even reportedly threatened to cut off some states' access to CODIS if they persist in so-called "Arizona searches."

Nobody knows exactly how rare DNA matches are; they are just FBI estimates. But the dispute here focuses on one word: "profile." Your complete genetic makeup is unique, but your "genetic profile" is just a narrowly focused snapshot of your genes. As the Times said, siblings often share these genetic markers, and unrelated people can share some by coincidence. An exact match of 13 markers by two unrelated people is unlikely. The odds? 1 in 1 quadrillion.

DNA evidence laws have changed since that 2001 search. States now require DNA profiles match at 13 loci instead of nine, enormously strengthening the odds. But in some older, colder cases, 9 loci can still be used, and the Arizona results have thrown a huge wrench into those prosecutions.

What happens now? DNA remains a strong piece of evidence, and an even stronger argument for releasing wrongly convicted people. But the fight over the data is likely to muddy every single case in the near future where DNA is the only evidence against an accused offender.

0 comments: