A cryptographic system could help protect the privacy of people whose data is included in consumer genetics databases while also allowing researchers access to that information, Agence France Presse reports.
Researchers from the Massachusetts Institute of Technology describe in a new paper appearing in Science an secure pharmacological collaboration approach based on a "secret sharing" method. In this way, various organizations can share data without revealing the underlying drugs or targets. Then, the organizations can run a secure computational analysis to train a predictive model based on the pooled dataset. This, the researchers say, then returns that model to the collaborators or returns to each results based on their input.
While they applied the approach described in their Science paper to developing a model of drug–target interactions, AFP notes that the researchers previously said that such an approach could be used for genomic data sharing.
"We're currently at a stalemate in sharing all this genomic data," Berger tells AFP. "It's really hard for researchers to get any of their data, so they're not really helping science."