A RING-BASED CHEMICAL STRUCTURAL QUERY SYSTEM - USE OF A NOVEL RING-COMPLEXITY HEURISTIC

A system for extracting, organizing, storing, and retrieving ring systems contained in large structural databases has been developed. For each compound a unique set of ring systems is extracted by using a hash-coding scheme to eliminate duplicates. The set of all the different ring systems in the database is then classified by using a simple heuristic measure of ring complexity. The ring-system data are stored in a database that allows retrieval of compounds based on the size, number, or complexity of the rings they contain. The ring-system connection tables are stored in a separate database that can be used to do standard substructure searches as well as topological similarity searches on the ring systems. Such searches can be a powerful adjunct to conventional database searches. © 1990, American Chemical Society. All rights reserved.