ZK-119010

A series of 37 substituted indoles for which estrogen receptor binding data had been reported were studied from a quantitative structure–activity relationship (QSAR) viewpoint. The structures were represented by an established unitary mathematical index, the molecular transform (FTm), whose derivation is entirely from the geometry‐optimized structure. Data clusters were generated by a primary numerically descending sort of the structure indices and a concurrent secondary numerically descending sort of the binding data. Reversal of numerical descent of the latter served to generate the cluster boundaries. Analysis within and across the clusters permitted a definitive and detailed picture of the influence of substituents and substituent positions on the molecule, and their effect on the biodata, and suggested the structure of a molecule of greater binding ability than any in the database. The relationship of the electronic and charge nature of the molecules to their binding ability was considered. © 2003 Wiley Periodicals, Inc. Int J Quantum Chem, 2003

Quantitative structure–activity relationships in a series of 37 substituted indoles with endocrine disruptor potential were performed using the structural indices FTe (electronic) and FTc (charge), in conjunction with a clustering technique, to relate substitution patterns to reported relative binding affinities for the calf estrogen receptor. Data clusters were generated by a primary numerically descending sort of the structure indices with a concurrent secondary numerically descending sort of the binding data. Reversal of the numerical descent of the latter served to delineate cluster boundaries. Analysis within the clusters defined the effect of substituents and their molecular positions on the pharmacological data. These results confirmed in detail a similar previous study in the same series using the more general FTm index and again suggested the same structure of a molecule with greater receptor binding ability than any in the database. The methodology used in these studies permits a rational presentation and subsequent interpretation of data that initially appear to be totally random and devoid of recoverable information content. © 2003 Wiley Periodicals, Inc. Int J Quantum Chem, 2003