We have developed an integrated computer system for analysis of nucleic acid and protein sequences, which consists of sequence and structure databases, a relational database, and software for structural analysis. The system is potentially applicable to a number of problems in structural biology including predictive classification of the function and location of oncogene products.