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.

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.

Three-dimensional structure databases, and their accompanying searching software, have been available for several years, both from commercial software vendors and by in-house development. Commercially available systems include MACCS-I1/3D and ISIS/3D (MDL Information Systems, Inc.), Aladdin (Abbott Laboratories and Daylight Chemical Information Systems, Inc.), ChemDBS-3D (Chemical Design Ltd.), SYBYL/3DB Unity (Tripos Associates), and Catalyst (BioCad Corp.). With the exception of ChemDBS-3D and, most recently, SY BYL/3D Unity, these software products apply geometric searching algorithms to static 3D models. This is acceptable for relatively rigid structures but fails to take into account the inherent flexibility of many molecules of biological interest. Approaches to this problem have included the registration of multiple conformations (all products), conformational analysis at registration and search time (ChemDBS-3D), development of 3D queries that can accommodate limited flexibility in the target structures (MACCS-I1/3D, ISIS/3D), and, most recently, application of the directed tweak approach (SYBYL/3DB Unity). This paper will discuss a new approach to the problem, implemented within the ISIS/3D software. The method uses a multilevel screening and constraint-fitting approach, applying torsional optimization with van der Waals energy contributions in the later stages. Methodology and examples are covered.

No abstract

Using a small database of defined substrates in humans for cytochrome P450 mixed function oxidases, a series of descriptors and classification methods were evaluated with respect to how well they correctly classified substrates. The descriptors ranged from structural keys to topological to electronic. A variety of classification schemes were examined in terms of their ability to point out which descriptors are important for predicting the cytochrome P450 specificity for a substrate. Results illustrate the relative effectiveness of the various kinds of descriptors and classification methods, as well as the value of using as well-defined data set as possible.