A bid by leading researchers to bring into the open data on bird flu that were previously kept behind closed doors has met with cautious optimism from observers.

Some 70 avian-flu scientists from all corners of the globe have signed up to the Global Initiative on Sharing Avian Influenza Data (GISAID). A letter outlining the agreement was published online last week in Nature and appears in this week's issue (see page 981).

The move aims to resolve issues that have seen some countries and organizations come under fire for hoarding genetic information about avian flu strains. The reasons for their reluctance to share are varied: some, for example, fear that others might use the data without properly crediting the researchers involved. And there were concerns over whether countries worst hit by bird flu would benefit from the drugs and vaccines developed from the sequences they provide.

Precise details of the GISAID agreement are still being thrashed out. But the idea is that participants will place genetic data into secure sections of existing online databases as soon as possible after producing and analysing them.

Lining up: access to DNA sequence data is vital for fighting bird flu. Credit: K. CURTIS/SPL

Peter Palese, who studies flu viruses at Mount Sinai School of Medicine in New York, describes GISAID as “a very positive development”. There are signatories from many of the countries affected by bird flu, including China, Indonesia, Thailand and Vietnam. And the Indonesian government has thrown its support behind free access to data. But Palese questions whether the agreement will work without the full involvement of all the scientists and governments in these countries.

Menno de Jong, an avian-flu researcher at the Hospital for Tropical Diseases in Ho Chi Minh City, Vietnam, also welcomes the initiative. But he cautions that care should be taken to ensure GISAID doesn't become “another, albeit bigger, old boys' network”. He suggests that increasing the capacity of poor countries affected by the disease to do their own research, enabling them to be more equal partners in collaborations, might be an extra encouragement for them to share their data.

GISAID proposes pooling its data using the International Nucleotide Sequence Database Collaboration, a network of three major public databases. The data will initially be accessible only to researchers who have signed up to GISAID, but the information will become open to the public no more than six months after it is deposited.

GISAID members have also agreed to collaborate with, and appropriately credit, all relevant researchers in any resulting publications and intellectual-property agreements.

Clinical and epidemiological data are included in the agreement. The hope is that researchers will be able to compare new strains against others quickly, for example to track whether a virus is acquiring mutations or becoming resistant to drugs.

Until now, access to many genetic sequences has been restricted to a global network of flu labs associated with the World Health Organization (WHO). Dick Thompson, WHO spokesman in Geneva, says that the organization supports the sharing of sequence data. He notes that the WHO has sometimes been portrayed as wanting to keep data secret but says that is not the case. “Sometimes countries have legitimate reservations and we have to work with that,” he says.

John Sulston, of the Wellcome Trust Sanger Institute in Cambridge, UK, says that he supports GISAID as a compromise. But he adds that because bird flu is spreading around the globe so fast, “I still think it's a good idea to release data immediately.”

But virologist Ilaria Capua at the Experimental Animal Healthcare Institute in Padua, Italy, who campaigned for the initiative with Hollywood media adviser Peter Bogner, says she is very happy with the result. She adds that the framework could be used for other emerging infectious diseases. “If a new SARS knocks on our door, we have a system in place.”