COMPASS - PREDICTING BIOLOGICAL-ACTIVITIES FROM MOLECULAR-SURFACE PROPERTIES - PERFORMANCE COMPARISONS ON A STEROID BENCHMARK

We describe a new method, Compass, for predicting the biological activ ities of molecules based on the activities and three-dimensional struc tures of other molecules. The method improves on previous techniques b y representing only the surface of molecules, by incorporating a nonli near statistical method, and by automatically choosing conformations a nd alignments of molecules. We use a benchmark problem of steroid bind ing affinity prediction to compare the performance of the method with that of two previous systems: CoMFA and a molecular similarity method. Compass predicts steroid affinities substantially more accurately tha n the others, which represent the state of the art. We present experim ents showing that the improved performance depends on each of the tech nical innovations.