In Silico Screening of Zinc (II) Enzyme Inhibitors Using ILP

In silico screening is a powerful drug-discovery tool. However, the application of traditional structure-based and mechanism-based drug design is hampered by the limited availability of three-dimensional structures of target enzymes or proteins. Thus, we propose a new method of screening good inhibitors of target enzymes without using their precise structures, based on machine learning. With this method, the data of ligands and decoys are collected from the inhibitor’s Database of Useful Decoys: Enhanced (DUD-E). We evaluated the accuracy of Inductive-Logic Programing (ILP) by applying a classification model that learned ligands and decoys from DUD-E to ligand candidates that are not included in DUD-E. In the present study, this technique is applied to the screening of inhibitors of carbonic anhydrase. ILP exhibited high classification performance. Furthermore, we visualized the rules from ILP and obtained a clear classification model.