Comparison of the performance of different “local reactive descriptors” in 3D-QSAR analysis of enantioselective molecules

The interaction between Ligand-Receptor (L-R) to solve the pharmacological and toxicological problems of enantioselective molecules was studied as 3D Quantitative Structure-Activity Relationship (3D-QSAR). The biological activity of enantiomers of stereo isomeric molecules that differ in 3D can only be determined by the Local Reactive Descriptor (LRD). The parameters of the Pharmacophore (Pha) on the receptor side can be determined using LRDs of atoms of the molecule in 3D space. “Klopman index,” which was brought to the literature as LRD by us, was compared with typical LRD descriptors such as atom charge, Fukui index, frontier orbital coefficients (HOMO/LUMO) in our Molecular Conformer Electron Topological (MCET) program. After predicting the 3D-QSAR model with the Leave One Out-Cross Validation (LOO-CV) technique on the molecules in the training set, the model was validated on the molecules in the external test set. Statistical results from both sets were evaluated for all LRDs, and the best results were obtained as Q2 = 0.955 and R2= 0.964 in the Klopman index. © 2021, The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.