Nature Biotechnology responds: It is the policy of this journal, and of all Nature journals, that any supporting data sets for which there is no public repository should be made available to any interested reader on and after the publication date from the authors directly. As a condition of publication, authors are required to make all materials and methods used promptly available to researchers for their own use. They are also required to state in the Methods section of their published manuscript any conditions for use of materials and to provide full disclosure of the conditions to the editors at submission and on a freely accessible, identified web site after publication. Authors of published papers are allowed to charge a small amount for materials distributed to other researchers to cover the authors' costs in producing the materials. Those who refuse to share data required to reproduce or build on the findings of a paper published in our pages are thus in breach of our guidelines and behave contrary to the spirit of scientific conduct. In this regard, we applaud the new rules introduced by the US National Institutes of Health (NIH, Bethesda, MD, USA) this month to require grant applications to outline a specific plan for sharing model organisms and materials that may result from the funded research (Nature 430, 953, 2004). Perhaps most importantly, this plan has teeth: an applicant's track record in sharing materials will be taken into consideration when their NIH grant is up for renewal.