In the latest salvo in a legal row between the Global Initiative on Sharing Avian Influenza Data (GISAID) – an international group created by leading flu researchers in August 2006 to promote data sharing – and the Swiss Institute of Bioinformatics (SIB) in Geneva, GISAID today launched its own version of the Epiflu database.

GISAID announced a contract with the SIB in December 2006 to build the first EpiFlu database, an international database which opened last year, and which scientists have since used to help monitor the spread and evolution of influenza viruses and to select the strains used to makes flu vaccines. In July, however, SIB made EpiFlu unavailable via the GISAID website, and available only to users redirected to a SIB website (Nature).

At the time, SIB and GISAID officials declined to discuss the details because of the ongoing legal dispute. But, in short, the SIB alleged that GISAID has breached its contract by failing to pay its bills on time, and that, under Swiss law, a default on payment renders a contract null and void, giving the SIB the rights to the database it built. GISAID claimed that SIB had “misappropriated” the database.

That Epiflu database remains accessible to users on the SIB site, but today GISAID launched its own independent version on the EpiFluDB tab on its own website. The new database was built by the the computational biology department of the Max Planck Institute for Informatics near Saarbrücken in Germany, and a3systems GmbH.

Meanwhile both parties continue to reiterate their claims, and also accuse the other of jeopardizing pandemic control efforts. “We regret that GISAID is diverting users to a new database” that is only at an early developmental phase, says Ron Appel, head of SIB. “It is unfortunate that such an important service as the EpiFlu database is being jeopardized by this legal issue, especially at a time when A(H1N1) influenza 2009 is such an important health issue.”

In a statement, GISAID argues that “SIB’s misappropriation of the first Database flagrantly violated numerous provisions of SIB’s Agreement with GISAID; more importantly, SIB’s illegal actions greatly undermined the usefulness of the first EpiFlu Database and threatened the worldwide effort to combat the pandemic H1N1 swine flu, the H5N1 avian flu and seasonal influenza.”

Leaving aside whatever is any eventual legal outcome, in the interim the battle seems to be for the hearts and minds of those submitting data. Before the creation of GISAID and the Epiflu, several countries — including Indonesia, an avian flu hotspot — had refused to share sequence data on the grounds that they got little in return. But they were persuaded to share by GISAID’s terms of access to Epiflu, similar to those of open-source software, in which all users agree to share their own data and give due credit to the originators.

In an email to Epiflu users this morning, GISAID said “thanks to the strong commitment and resilience of the GISAID community, we are pleased to provide you again with access to GISAID’s EpiFlu database, as part of our pledge to offer the most complete and publicly accessible influenza database. While the influenza data GISAID offers is just as comprehensive as before, please understand the new database is at an early developmental phase. Our team of developers are currently enhancing the software to include more features and integrated analysis tools. ”

In a statement, GISAID added: “Leading veterinary reference laboratories of the Food and Agriculture Organization of the United Nations and the World Organization for Animal Health, as well as the influenza collaborating centers of the World Health Organization, have uploaded data onto the new GISAID EpiFlu database and are using it to share information. Indeed WHO collaborating centers are using GISAID’s new database as a tool to make their recommendations for the Southern Hemisphere’s 2010 seasonal flu vaccine.”

SIB sent its own email to Epiflu users this afternoon, saying that SIB was “committed to continue operate the EpiFlu Database, ” and that “if you are satisfied with the SIB’s work and competence in respect to EpiFlu and you would like the SIB to continue to manage the established database, rather than using GISAID’s new database that is “at an early developmental phase”, please let us know.”