Researchers working on large-scale publicly funded genome-sequencing projects must release their data as soon as possible, without imposing restrictions on the use of the information, say guidelines from two of the main funders of such work.

The US National Human Genome Research Institute (NHGRI) this week issued a draft revision to its policy on the use of data from large-scale genome projects. Existing rules prohibit users from publishing a whole-genome analysis before the sequencers' initial publication on the complete genome. “That restriction will be removed and will not be present in any other genome databases that we support,” says the institute's director, Francis Collins.

The move follows discussions at a meeting of sequencers, bioinformatics researchers and journal editors last month in Fort Lauderdale, Florida, aimed at resolving disagreements over data access. The revision is expected to be approved by the institute's advisory council in May.

The Wellcome Trust, the London-based medical charity that convened the meeting, will also apply similar rules to the large-scale genome-sequencing projects for which it provides funding. Both organizations may also extend the rules to other large-scale ventures, such as protein-structure projects and analyses of gene expression.

The NHGRI's guidelines say users should acknowledge that “sequence producers have a legitimate interest in publishing peer-reviewed reports” about their sequence, and that “the entire scientific community can also help ensure that the system works fairly”. They are an extension of the Bermuda Principle — the 1996 agreement between the leaders of the Human Genome Project to make sequence data available without restrictions within 24 hours.

The move is likely to find favour with bioinformatics researchers, who use the sequence, but sequence producers may worry about being scooped. “Some groups may lack the integrity to play gentlemen's cricket,” says Richard Gibbs, director of the Human Genome Sequencing Center at Baylor College of Medicine in Houston.

The team working on the malaria parasite Plasmodium falciparum was angered when other researchers tried to publish studies of sequence data before the project was complete (see Nature 405, 601–602; 2000). But Collins believes this risk is a small price to pay to ensure the immediate release of data. “We will be scooped on a few occasions,” agrees Jonathan Eisen of The Institute for Genome Research in Rockville. “We'll just have to deal with it.”

The guidelines could also hamper international collaborations. “While we as scientists are willing to release the data, the funding agencies in Japan may not support data release without restrictions,” says Yoshiyuki Sakaki of the RIKEN Genomic Sciences Center in Yokohama City, Japan. “Chinese agencies will face a dilemma,” adds Huanming Yang, head of the Beijing Genomics Institute. “Why should they invest the money if they can get data for free from the United States?”

But some question whether there will be a rush to publish ahead of the sequence centres. “I would much rather the annotation be done at the sequencing centre, rather than a third party who doesn't know how the data were generated,” says Sean Eddy, a bioinformatician at Washington University School of Medicine in St Louis, Missouri. “We can't be biting the hand that feeds us.”