Three-Dimensional Quantitative Structure−Activity Relationships from Molecular Similarity Matrices and Genetic Neural Networks. 1. Method and Validations

TLDR

The theoretical basis for using molecular similarity in QSAR is discussed. The study demonstrates the utility of genetic neural networks for deriving QSAR models from molecular similarity matrices. The method is applied to predict corticosteroid‑binding globulin affinity using a well‑known steroid dataset. The GNN model achieved excellent predictivity, with a six‑descriptor electrostatic/shape model attaining a cross‑validated r² of 0.94 and outperforming PLS, genetic regression, and other 3D QSAR methods, while statistical validation confirmed the results were not due to chance.

Abstract

The utility of genetic neural network (GNN) to obtain quantitative structure−activity relationships (QSAR) from molecular similarity matrices is described. In this application, the corticosteroid-binding globulin (CBG) binding affinity of the well-known steroid data set is examined. Excellent predictivity can be obtained through the use of either electrostatic or shape properties alone. Statistical validation using a standard randomization test indicates that the results are not due to chance correlations. Application of GNN on the combined electrostatic and shape matrix produces a six-descriptor model with a cross-validated r2 value of 0.94. The model is superior to those obtained from partial least-squares and genetic regressions, and it also compares favorably with the results for the same data set from other established 3D QSAR methods. The theoretical basis for the use of molecular similarity in QSAR is discussed.