Three-dimensional quantitative structure-activity, relationship modeling of γ-secretase inhibitors using molecular field analysis

Three-dimensional quantitative structure-activity relationship analysis of a set of 79 analogs of gamma-secretase inhibitors was performed by molecular field analysis with genetic partial least squares method to investigate the substitutional requirements to derive a predictive model and for the favorable receptor-drug interaction that may be used for the designing of a novel V-secretase inhibitors. The developed molecular field analysis model has a good fit, with r(2) value of 0.952 and cross-validated coefficient, r(cv)(2), value of 0.931. Predictive ability of the developed model was further assessed using test set of 19 compounds and r(pred)(2) was found to be 0.665.