Scientists Discover How to Identify People From ‘Anonymous’ Genomes

Most people participate in genomic research because they hope the DNA they offer up will help scientists uncover the roots of human diversity and disease. They generally expect to remain anonymous. But they may not be.

Researchers armed with little more than an internet connection identified nearly 50 people who participated in a large genomic study based on some of the participants’ genomes and other publicly accessible information.

The researchers do not name these individuals, and they insist their intentions were good. “We are not trying to start a panic,” said Yaniv Erlich of the Whitehead Institute for Biomedical Research, who led the new study. “We are trying to illuminate some of the gaps in privacy we have right now and initiate a public discussion.”

Scientists and research subjects aren’t the only ones who have a stake in that conversation, says genomics pioneer George Church, of Harvard Medical School. Today, a genome can be sequenced for a few thousand dollars, and the cost continues to drop.

“Soon you’ll be sequencing yourself and your dog and your plants,” Church said.

The new study shows that those sequences can sometimes be traced back to the individuals they came from.

Erlich and his team started with the observation that Y-chromosomes and surnames tend to go together. That’s because sons always inherit their father’s Y-chromosome and typically inherit his surname. Certain genetic stutters on the Y-chromosome, in which the letters of the genetic code repeat over and over, vary widely in the general population but tend to be shared by closely related men.

In a few highly publicized cases, people have exploited this to find their sperm donor father. In 2005, for example, a 15-year old boy reportedly found his biological father after having his own Y-chromosome tested and combing a commercial genealogy website for close matches. These matches pointed to a potential surname, which the boy combined with other clues — including the sperm donor’s birth place and date — to track him down.

What Erlich’s team did is conceptually similar, but far more technically sophisticated. Instead of starting with a DNA sample, they started with 10 entire genomes publicly available as part of the international 1000 Genomes Project. An algorithm the researchers developed mined these genomes, all from men, for telltale variations in the Y chromosome. Then they searched two commercial genealogy databases for close matches and identified potential surnames. Finally, using additional details from the research records, such as the participants’ age and state of residence, as well as obituaries and other public documents, the team identified five of the 10 research participants and their entire families, which were also part of the genome research project, they report today in Science.

more at: Wired