QSAR study of pharmacological permeabilities

Quantitative Structure-Activity Relationship (QSAR) models are developed for three pharmacological permeabilities, i.e. two PAMPA apparent permeabilities (logP(app)) at different pH values (pH 5.5 and pH 7.4) and Caco-2 cell monolayer apparent permeability (logP(app)(Caco-2)). The compounds are represented by chemical descriptors calculated from their constitutional, geometrical and topological structure, and quantum mechanical wave function. The obtained linear (multilinear regression) and nonlinear (artificial neural network) models link the drug structures to their reported permeabilities. Each multilinear model was tested by leave-one-out and ABC methods whereas the neural networks were assessed using the test sets. All drug structures were investigated by conformational analysis in order to find the low energy conformers.