In Depth DNA Analyzers at the Joint Genome Institute

Published on November 18th, 2014 | by Travis Korte

0

Next Steps for Genomic Data Sharing

Genomic data holds great promise for creating personalized medical treatments, improving population health, and managing costs in health care organizations. But because genomics researchers have traditionally collected and stored their data in decentralized silos—by disease or by country, for example—researchers hoping to draw from past research to inform new work or conduct replication studies must navigate a complex web of access issues, non-standard data formats, and data being stored in multiple places. However, a recent initiative by the U.S. National Institutes of Health (NIH) to implement better genomic data sharing practices will do much to improve the situation. The NIH’s Genomic Data Sharing Policy, which it released in August 2014, states that funding applications for genomic research submitted to the NIH must include a data sharing plan in which applicants estimate how much resources they will need to submit their data to appropriate repositories. Non-human genomic data must then be submitted no later than the date the research is published and human data no later than the date the research is accepted for publication. While NIH’s policy is a huge step forward, there are still opportunities for NIH to improve genomic data sharing, such as by working with its international partners to establish international data sharing standards and placing more weight on data sharing in its grant-making decisions.

First, NIH should encourage more genomic data sharing by working with its international partners to ensure they have similarly broad data sharing policies. Genome Canada, the country’s primary funding body for genomic research, has a liberal sharing policy that is roughly in line with the NIH directive, but the EU has no comparable policy, and the EU’s proposed General Data Protection Regulation imposes a variety of restrictions on data sharing, such as requiring organizations to repeatedly acquire consent from patients to share their data instead of allowing individuals to offer blanket consent for future data sharing. Organizations such as the European Alliance for Personalized Medicine and Science Europe have expressed similar concerns about the General Data Protection Regulation’s implications for data sharing. To ensure that its forward-thinking policies around data sharing are adopted by funding agencies in other countries, the NIH should use its leadership and prestige to encourage other nations to follow suit.

Second, the NIH and the National Human Genome Research Institute (NHGRI) should ensure that their grant reviewers take data sharing plans seriously in their funding decisions. A 2014 survey of life science researchers’ attitudes around data sharing found that nearly a third felt that data sharing plans were not at all important in their review of proposals. In addition, the research found that hardly any scientists had moved to formally sanction a colleague for failing to comply with data sharing agreements and that only a small minority had imposed informal sanctions, such as refusing to collaborate with the non-sharing colleague. While this survey took place before the NIH introduced the Genomic Data Sharing Policy and reviewers attitudes may change as a result of that policy, the NIH should be sensitive to these kinds of cultural barriers to data sharing and strive to increase the percentage of reviewers who take data sharing plans seriously in their funding decisions in the future. Although the study did not address genomics specifically, it points to the larger issue that data sharing must become part of the culture of research (including genomics) if it is to yield benefits. The NIH and the NHGRI should hasten their efforts to change this culture, perhaps calling for reviewers to explicitly take data sharing plans into account in their funding decisions. NIH could also instruct reviewers to take past compliance with the data sharing policy of applicants’ affiliated research institutions into account for future grantmaking decisions, which could help motivate research universities and institutions to hold their researches accountable for compliance.

While the NIH policy does a great deal to advance genomic data sharing, the job is far from over. If lifesaving personalized genomic applications are to be realized, research data sharing infrastructure must be international in scope and taken seriously by grant funders. The NIH has an opportunity to continue to lead the way by collaborating with its partners in the EU and beyond, and working to change research funding culture to prioritizing data sharing.

Photo: Flickr user Berkeley Lab

Tags: , , , , ,


About the Author

Travis Korte is a research analyst at the Center for Data Innovation specializing in data science applications and open data. He has a background in journalism, computer science and statistics. Prior to joining the Center for Data Innovation, he launched the Science vertical of The Huffington Post and served as its Associate Editor, covering a wide range of science and technology topics. He has worked on data science projects with HuffPost and other organizations. Before this, he graduated with highest honors from the University of California, Berkeley, having studied critical theory and completed coursework in computer science and economics. His research interests are in computational social science and using data to engage with complex social systems. You can follow him on Twitter @traviskorte.



Back to Top ↑

Show Buttons
Hide Buttons