Matching crime-scene DNA to data in a genealogy site recently helped cops nab a suspect in a decades-old murder case referred to as the Golden State Killer. Now, scientists say, the technique could be used to identify more than 60 percent of US citizens with European ancestry by a given DNA sample.
“In a few years, it’s really going to be everyone,” study coauthor Yaniv Erlich, a computational geneticist at Columbia University, tells Science.
Erlich and his colleagues were able to home in on the identity of a person from an anonymous DNA sample using only a basic characteristic about the person, such as her age, and a genetic database of 1.28 million individuals. The team winnowed the results from more than 1 million people to roughly 20 with the technique and could narrow it even further using census data and other publicly available records, according to a paper published today (October 11) in Science. A separate group of researchers report similar results today in Cell.
For a number of years now, people have been using genealogy and genetic databases to identify their birth parents, sperm donors, and long-lost family members, genealogist Debbie Kennett tells The Atlantic. So the results weren’t “a surprise to us at all. . . . We have been using it for years and years,” she says. But the method’s application to an infamous cold case, the Golden State Killer, got people’s attention, she notes. Since its resolution became public, the technique has been used to identify suspects in at least 12 other cold cases.
Matching DNA to a single person isn’t as simple as the scientists writing the paper suggest, especially in criminal cases, CeCe Moore, a genealogist working with Parabon NanoLabs, tells Buzzfeed News. “It’s very difficult, because every case has unique challenges to overcome.”
Still, Erlich and his colleagues pinpointed the identity of the anonymous DNA donor within a matter of hours. They uploaded the DNA of a woman who had donated her genetic material to the 1000 Genomes Project to GEDmatch, a personal genomics database. They then identified her closest genetic matches in the database and used family trees to ultimately expose her identity.
What’s potentially worrisome about the technique is its misuse. Everyone from immigrants to study volunteers could be potentially targeted and tracked using the technique, which has Erlich and others calling for better privacy protections for sites such as GEDmatch and others.
Some legal scholars agree. As Natalie Ram, a law professor at the University of Baltimore in Maryland, and colleagues recently wrote in Science, “At a minimum . . . policymakers should delineate under what circumstances such searches are acceptable.”
Correction (October 13): The context of the initial quote by Natalie Ram was incorrect and has been updated accordingly. The Scientist regrets the error.