GENEVA--In May 1996, Amos Bairoch had only two months of funding remaining for work on his SWISS-PROT and PROSITE databases. Today, as a result of his campaign for support, Switzerland has a world-class research center and the bioinformatics industry has two new members: the Swiss Institute of Bioinformatics (SIB) and the company Geneva Bioinformatics (GeneBio) (see BioInform, March 16, 1998).
Bairoch put out a plea for help to his bioinformatics colleagues in early 1996. SWISS-PROT, the premier database of protein sequences developed jointly with the Euro pean Bioinformatics Institute, was threatened with extinction as a result of Swiss funding mandates and a foiled attempt at gaining a European Union grant. News of the crisis spread quickly around the world by e-mail, electronic BIOSCI newsgroups, and articles in Science and Nature. The situation was particularly disquieting for the bioinformatics community, since many valuable protein resources rely on the data in SWISS-PROT and PROSITE for their ongoing development.
Support came in from all walks of bioinformatics. After exploring many options in the commercial sector and receiving interim funding from the state of Geneva and gifts from several companies, Bairoch and his colleagues, together with Swiss scientific authorities, decided it was time to develop a longer-term approach.
The Swiss National Science Foundation generally directs funds toward specific research projects for a limited period of time. For projects such as SWISS-PROT that provide a long-term service to the Swiss, European, or international communities, only soft funding options were available in Switzerland. After much lobbying, a consortium of three Swiss scientific funding agencies took the scientists' pleas to heart and pledged temporary funds for 1997-1999, until a permanent solution could be instituted.
A self-sustaining solution
The ultimate answer was the formation of two groups: the SIB, a nonprofit foundation established to support national research and training in bioinformatics, and GeneBio, a private-sector entity founded to develop cutting-edge bioinformatics tools and to commercialize the groundbreaking technologies developed by the academics at SIB.
SIB, proposed in late 1997 and officially created on March 30, 1998, includes five founding research groups comprising about 50 researchers. The institute expects to grow through training and recruiting efforts to 100 researchers before the turn of the century. In addition to research and teaching activities, SIB is responsible for maintaining the ExPASy server, used by researchers around the world, and hosts the Swiss node of EMBnet.
SIB founding members are:
* Ron Appel, director of the molecular imaging and bioinformatics laboratory at Geneva University Hospital. His group works on the Melanie software for 2-D analysis, currently available through BioRad; the SWISS-2DPAGE database; and toward applying artificial intelligence to problems such as scheduling and traversing the worldwide web.
* Amos Bairoch, group leader in the department of medical biochemistry of the University of Geneva. He is responsible for the ongoing development of the SWISS-PROT, PROSITE, and ENZYME databases and wrote the PC/Gene sequence analysis software. Bairoch currently works on devising novel tools for proteomic analysis.
* Philipp Bucher, group leader and head of the biocomputing unit at the Swiss Institute for Experimental Cancer Research near Lausanne. His group develops profile-based methods for analyzing distantly related protein and DNA sequences and for cataloguing highly divergent protein domains and families. He also works on protein structural prediction and is responsible for the development of the Eukaryotic Promoter Database.
* Victor Jongeneel, director of the Swiss node of EMBnet and director of the Office of Information Technology for the Ludwig Institute for Cancer Research near Lausanne. He is responsible for bioinformatics services and training courses for academic and industrial users throughout Switzerland. His research interests are in the structural and functional analysis of several families of sequences.
* Manuel Peitsch, head of Glaxo Wellcome Experimental Research here, and director of scientific computing worldwide for Glaxo Well come Research and Develop ment. His group focuses on structural computing and quantitative visualization of protein information, and recently announced a large-scale project to systematically predict, by homology modeling, the structure of every protein in known protein families. He is responsible for the SWISS-MODEL server, the Swiss PDB Viewer software tool, and the Swiss 3D IMAGE database.
SIB provides an umbrella organization for funding and administrative support, while still allowing each group to continue its own teaching and research activities. SIB's members have a heavy emphasis on protein bioinformatics and are leading the way in the development of databases and analysis tools for the budding field of proteomics.
SIB's mission includes developing software and databases through high-quality research programs, providing bioinformatics services to the Swiss life science community, and training and developing bioinformatics research scientists through courses and seminars at pre- and postgraduate levels. Funding for the organization comes from several sources, including researchers' individual grants, a special grant from the Swiss government, and revenue from any future product licensing.
GeneBio was created by the same forces that sparked the formation of SIB. No strangers to the world of commercial bioinformatics, Bairoch, Appel, and Denis Hochstrasser of Geneva University Hospital knew that to fund an ongoing research effort in the face of waning government funding, they must have an industrial outlet for the sought-after bioinformatics products developed by Swiss researchers. Having licensed software previously through US companies, the three were keen to keep intellectual property within the country.
Tapping his network of contacts from the 2-D world, Hochstrasser approached Jenny Fichmann at Pharmacia Biotech in San Francisco in December 1997 and offered her a job leading GeneBio. With a biochemistry PhD from the University of California Santa Cruz, Fichmann understood the growing importance and commercial potential bioinformatics offered. The timing of the offer was perfect, luring Fichmann to return to her native country as chief operating officer of GeneBio this past February.
GeneBio has a threefold mission: to build products, primarily specialized databases and proteomic tools; to commercialize technologies developed at SIB; and to provide high-end custom development for clients. Although details will not be finalized until late summer, it appears that up to 75 percent of the revenue from products derived from SIB-initiated technology, minus any overhead costs, go back into SIB to fund further research, linking the two groups symbiotically. GeneBio is privately funded by a small group of stockholders.
Setting the trend
The foundation of these two entities evidences a growing trend in the bioinformatics industry: the formation of national bioinformatics research centers along with matching companies for technology commercialization. The trend is driven by two forces: increased pressure worldwide for life-science researchers to become less reliant on government funding, and the pharmaceutical industry's intense demand for cutting-edge bioinformatics tools. Although Switzerland is the first to tackle the challenge on such a large scale, similar efforts are occurring in other countries as well, including the US, South Africa, Australia, and Japan.
Commenting on the trend, Fichmann said she has "come to no conclusions yet. We are still setting up the standards." But, she added, the approach is a natural one. "In a perfect world, you don't have to pay to access information. Not being a perfect world, one needs funding for the long-term maintenance and enhancement of raw biological data. The good thing is that if the enterprise is successful, everyone benefits."
All parties seem to agree that putting bioinformatics research in the commercial context increases stability for the academics involved and raises development standards and quality across the board. What is yet to be discovered is how these collaborations will affect creativity, output, and perhaps even the existence of pure research in bioinformatics over the next decade.