Designing libraries with CNS activity

Library design is an important and difficult task. In this paper we describ e one possible solution to designing a CNS-active library. CNS-act;ives and -inactives were selected from the CMC and the MDDR databases based on whet her they were described as having some kind of CNS activity in the database s, This classification scheme results in over 15 000 actives and over 50 00 0 inactives. Each molecule is described by 7 ID descriptors (molecular weig ht, number of donors, number of accepters, etc.) and 166 2D descriptors (pr esence/absence of functional groups such as NH2). A neural network trained using Bayesian methods can correctly predict about 75% of the actives and 6 5% of the inactives using the 7 1D descriptors. The performance improves to a prediction accuracy on the active set of 83% and 79% on the inactives on adding the 2D descriptors. On a database with 275 compounds where the CNS activity is known (from the literature) for each compound, we achieve 92% a nd 71% accuracy on the actives and inactives, respectively. The models we c onstruct can therefore be used as a "filter" to examine any set of proposed molecules in a chemical-library. As an example of the utility of our metho d, we describe the generation of a small library of potentially CNS-active molecules that would be amenable to combinatorial chemistry. This was done by building and analyzing a large database of a million compounds construct ed from frameworks and side chains frequently found in drug molecules.