LOCATING BIOLOGICALLY-ACTIVE COMPOUNDS IN MEDIUM-SIZED HETEROGENEOUS DATASETS BY TOPOLOGICAL AUTOCORRELATION VECTORS - DOPAMINE AND BENZODIAZEPINE AGONISTS

Electronic properties located on the atoms of a molecule such as parti al atomic charges as well as electronegativity and polarizability valu es are encoded by an autocorrelation vector accounting for the constit ution of a molecule. This encoding procedure is able to distinguish be tween compounds being dopamine agonists and those being benzodiazepine receptor agonists even after projection into a two-dimensional self-o rganizing network. The two types of compounds can still be distinguish ed if they are buried in a dataset of 8323 compounds of a chemical sup plier catalog comprising a wide structural variety. The maps obtained by this sequence of events, calculation of empirical physicochemical e ffects, encoding in a topological autocorrelation vector, and projecti on by a self-organizing neural network, can thus be used for searching for structural similarity, and, in particular, for finding new lead s tructures with biological activity.