SHARING SCIENTIFIC RICHES

Michael Gottesman

This issue of The Catalyst includes a review of some of the research achievements of the NIH intramural programs in the past year.

Even a casual glance reveals a rich trove of basic science discoveries, technological advances, and new strategies to prevent, diagnose, and treat human ailments. True to the NIH mission to "do research to improve the public health," there are numerous examples of disease-oriented and patient-oriented research that may soon become the standard of care. We should be proud of these achievements, make the public aware of them, and be generous in our efforts to help all of our scientific colleagues benefit from our research tools and our knowledge.

Thus, it is the policy of the NIH to share research resources and, wherever possible, research databases with the rest of the scientific community (see guidelines for recipients of NIH research grants and a report from the NIH working group on research tools). In keeping with this policy, the scientific directors and their Committee on Large Database Sharing recently issued a statement supporting an NIH Draft Statement on Sharing Research Data. The SDs concluded that the "general principles outlined in the statement are applicable to scientists in the NIH intramural research program involved in the collection and analysis of large-scale databases."

A few examples serve to illustrate the important effects of providing research tools and data to our scientific colleagues. Research tools such as transgenic animals have become an integral part of research advances in many fields of biomedicine; it is inefficient and unjustifiable to use public monies to recreate transgenic mice, for example, that have already been characterized and described in the public literature. We expect our scientists, and our colleagues in the extramural community, to provide such animals to other legitimate laboratories within a reasonable time frame after publication (see policy statement). Similarly, the NCBI’s GenBank provides genome data and analytic software at no charge to the public; the entire human genome sequence, arguably one of the great scientific advances of recent years, is freely available. Such should be the case, too, with published structural data, such as X-ray coordinates. The reward for such openness is the enormous acceleration of progress in biomedical research.

What are the responsiblities of investigators who oversee large databases that have not yet been fully mined for publication? Microarray analyses and clinical and epidemiologic databases come to mind. In these cases, the duration of data collection, the human subjects research (privacy) concerns, and the investment of the individual investigator are all factors that influence the degree and pace of public release. In the statement by the SDs, it is suggested that the Boards of Scientific Counselors should weigh in on how public release of such large databases should be handled; with long-term clinical studies, specific committees are often involved in assuring release of accurate data that do not violate individual privacy.

NIH-supported investigators have an obligation to publish their findings and make their data available so that our achievements can advance science and improve public health. That’s the NIH mandate. I welcome the comments of the intramural community on how best to share data and research tools.

—Michael Gottesman
Deputy Director for Intramural Research

Return to Table of Contents