Abstract
The NIH's new Genomic Data-Sharing policy, which goes into effect on January 25, 2015, outlines data sharing and informed consent expectations for researchers who generate large amounts of genomic data during NIH-funded studies.
Beginning with applications submitted for the January 25, 2015, deadline, researchers seeking NIH funding for projects that generate large amounts of genomic data—whether involving human or nonhuman subjects—will be expected to share that data through public repositories. In addition, the NIH expects researchers to gain informed consent from study participants to share the data for use in other studies. That consent should also extend to deidentified clinical specimens collected on or after January 25.
These expectations are outlined in the NIH's new Genomic Data-Sharing (GDS) Policy, which expands upon previous policies and broadens data-sharing practices. It also eliminates the embargo period, the time between when a researcher uploads human genomic data to the primary NIH database and when the data are released to the public and to other researchers. The new GDS policy replaces the 2007 Genome-Wide Association Studies policy and was designed to encourage the replication of findings and the use of data in future studies.
Dina Paltoo, PhD, MPH, who directs the Genetics, Health, and Society Program at the NIH's Office of Science Policy, says that the NIH wants to make all genomic data publicly available as quickly as possible, while at the same time protecting the identities and other personal information of study participants. The agency was motivated to draft a new policy, in part, by the rapid growth in the amount of data collected in genomic studies.
“Because of recent advances in DNA sequencing and other high-throughput ‘omics’ technologies—along with the drop in cost—the NIH has been able to fund more research that generates larger volumes of genomic data,” says Paltoo. Enabling a broader population of researchers to use the data, she says, could help expedite efforts to identify genes associated with specific diseases—as well as which genes can help researchers fine-tune new, targeted treatments.
According to the new policy, all nonhuman data from NIH-funded genomic studies can be registered in any widely used repository, and human genomic data should be registered in the NIH database for Genotypes and Phenotypes (dbGaP). Since the establishment of the dbGaP in 2007, more than 300 genomic data studies have been registered, the use of which has resulted in more than 900 papers, according to a report by Paltoo and a team of NIH researchers. One fifth of those papers were devoted to cancer research—a greater number than in any other subject except methods development, the authors reported.
In addition to adding human genomic data to the dbGaP, researchers may also upload their findings to other repositories such as GenBank, an online collection of publicly available DNA sequences, if participants have provided the appropriate consent.