Molecular databases are the main source of information for computational biology, but they also contain data that are extremely valuable for molecular medicine. Swiss-Prot, undoubtedly the best-annotated protein sequence database, is a prime example. It can be accessed at hundreds of sites via the world-wide web and a measure of its popularity is that its server at the Geneva site alone is accessed more than 50,000 times daily. In the following article, Amos Bairoch, director of Swiss-Prot in Geneva, and Rolf Appweiler, the Head of a Swiss-Prot group at the European Bioinformatics Institute, describe features of the database that are of particular interest to practioners of molecular medicine.