The FBI originally created the Combined DNA Index System (CODIS) in 1990 to help investigators search among convicted sex offenders and other violent criminals for matches to evidence from unsolved crimes. Over the years, its use has rapidly widened to include other types of felons, juvenile offenders and some who committed misdemeanors. Five states can collect DNA from some arrests, whereas federal authorities may acquire biological samples from those who are arrested as well as from noncitizens who are detained. Nationally officials have compiled more than 3.6 million profiles based on 13 regions, or loci, of the human genome that vary among individual people.
When labs can show a match is close enough to indicate a likely relative--that is, when at least one of the two versions (alleles) of the gene segment at each locus matches up--and there are no other leads, a new interim plan allows states to disclose identifying information on FBI approval. This expansion follows a paper in the June 2 Science that outlined the power of indirectly detecting suspects through the database. If only 5 percent of criminals had biological relatives in CODIS, then "cold hit" matches could increase by thousands, the authors of the paper estimated. They pointed to U.S. Department of Justice findings that 46 percent of jailed inmates had an incarcerated relative. "We don't venture whether genetic, social environment, socioeconomic, demographics or whether law enforcement practices account for it," says the lead author, medical geneticist Frederick Bieber of Harvard University. "In a way, it doesn't matter."
By widening its net, law enforcement can move more quickly and potentially head off future crimes, Bieber points out. Critics wonder, however, whether extending genetic surveillance from individuals already associated with crime to their families will help catch enough criminals to outweigh its likely intrusion on privacy and civil liberties. "We're talking about innocent people by proxy being included in this database," objects Tania Simoncelli, science adviser for the American Civil Liberties Union.
Princeton University sociologist Bruce Western questions the premise that familial searches will offer up strong leads, particularly as the databases expand under new laws to include more arrests, detentions and convictions for less serious crimes. In these categories, black and Latino young men are disproportionately pursued and become overrepresented in comparison to their actual share of offenses committed, he explains, and familial searches would magnify these racial and ethnic disparities. In a study of California's largest counties, half of felony charges were later dismissed.
Although family members do tend to share criminal behavior, investigators could use other social patterns of crime to make a comparable argument for searching other groups of DNA profiles--for instance, among males age 16 to 24, people with low levels of schooling, or, for the most violent crimes, people related to the victim. The database operates on a premise that Western says is severely at odds with criminological research--that of criminality as a permanent trait.
"We're kind of blundering ahead with this technology," worries William Thompson, a criminologist at the University of California, Irvine, who would like to see the government open up the database for independent scrutiny and statistical analysis. He is especially concerned about reports of faked test results and poor-quality lab work such as cross-contamination and sample mix-ups.
Also, statistical probabilities for unique matches and assumptions about the database population have recently come into question. Arizona compared felons already in its database and reported 20 "hits" between offenders where both alleles matched at 10 loci. Only three would be expected based on established calculations. The extras could be relatives or even entirely coincidental matches.
Bieber cited two high-profile cases in which investigators used partial matches to narrow in on a perpetrator who was a brother or uncle. But unless the shared alleles are quite rare or the matches are extremely close, this approach can be very inefficient. So Bieber suggested the type of kinship analysis and weighted allele probabilities used to connect family members with loved ones lost in the World Trade Center attacks. CODIS software does not allow for such sophisticated searches at the moment, but Bieber predicts an upgrade as soon as investigators see the potential and define clear statistical thresholds that must be met.