This project is to use a novel bioinformatics approach for modeling metabolic pathways. There are four steps to the Principal Investigators' strategy: (1) complex biological information is to be organized and integrated into a single consistent metabolic knowledge base (MKB), which will serve as a basis for the remaining steps; (2) a novel biocatalysis classification scheme (BCS) is to be induced from available information about catalyzed biochemical transformations; (3) the correlations between known protein properties and their catalytic function is to be used to learn novel empirical mappings between proteins and their function, which is to form the basis of a biocatalytic assignment tool (BAT); and (4) a metabolic pathway synthesis (MPS) system is to be designed to use induced knowledge about metabolism to generate new pathways between given compounds. Together, these four modules are to form the basis of a system for modeling metabolic pathways that can be used to solve a number of difficult problems.