Reoptimization of MDL keys for use in drug discovery

For a number of years MDL products have exposed both 166 bit and 960 bit keysets based on 2D descriptors. These keysets were originally constructed and optimized for substructure searching. We report on improvements in the performance of MDL keysets which are reoptimized for use in molecular similarity. Classification performance for a test data set of 957 compounds was increased from 0.65 for the 166 bit keyset and 0.67 for the 960 bit keyset to 0.71 for a surprisal S/N pruned keyset containing 208 bits and 0.71 for a genetic algorithm optimized keyset containing 548 bits. We present an overview of the underlying technology supporting the definition of descriptors and the encoding of these descriptors into keysets. This technology allows definition of descriptors as combinations of atom properties, bond properties, and atomic neighborhoods at various topological separations as well as supporting a number of custom descriptors. These descriptors can then be used to set one or more bits in a keyset. We constructed various keysets and optimized their performance in clustering bioactive substances. Performance was measured using methodology developed by Briem and Lessel. "Directed pruning" was carried out by eliminating bits from the keysets on the basis of random selection, values of the surprisal of the bit, or values of the surprisal S/N ratio of the bit. The random pruning experiment highlighted the insensitivity of keyset performance for keyset lengths of more than 1000 bits. Contrary to initial expectations, pruning on the basis of the surprisal values of the various bits resulted in keysets which underperformed those resulting from random pruning. In contrast, pruning on the basis of the surprisal S/N ratio was found to yield keysets which performed better than those resulting from random pruning. We also explored the use of genetic algorithms in the selection of optimal keysets. Once more the performance was only a weak function of keyset size, and the optimizations failed to identify a single globally optimal keyset. Instead multiple, equally optimal keysets could be produced which had relatively low overlap of the descriptors they encoded.