QSAR Modeling of 1,3,4-Oxadiazole Derivatives for Predicting Anticancer Activity Using SO-PLS Analysis.

Abstract

Introduction: The study aimed to develop a QSAR model for a series of 1,3,4- oxadiazole derivatives to identify molecular determinants influencing anticancer activity and provide a predictive framework for rational drug design.

Methods: A QSAR model was constructed using the Simulated Orthogonal Projections to Latent Structures (SO-PLS) approach. Five molecular field descriptors, steric (gauss_s), electrostatic (gauss_e), hydrophobic (gauss_h), hydrogen-bond acceptor (gauss_a), and hydrogen-bond donor (gauss_d), were employed. The model was developed on a training set of 36 compounds and evaluated using Partial Least Squares regression with up to five latent factors. Y-randomization was applied to assess model robustness.

Results: The model demonstrated an excellent fit (R² = 0.931) and acceptable predictive ability (R²-CV = 0.478), with Y-randomization confirming robustness (R²-scrambled = 0.679). Field contribution analysis revealed steric (29.52%) and hydrophobic (24.92%) effects as the primary determinants of biological activity, followed by hydrogen-bond donor (18.81%) and acceptor (18.02%) contributions, while electrostatic interactions (8.73%) were less influential. Regression coefficient analysis identified molecular regions where targeted substitutions could enhance activity.

Discussion: The findings highlight the critical role of steric and hydrophobic interactions in modulating the anticancer activity of oxadiazole derivatives. Strategic molecular modifications guided by field contributions and regression analysis can improve potency, supporting rational design of novel oxadiazole-based anticancer agents.

Conclusion: This QSAR model provides a robust predictive tool for designing 1,3, 4-oxadiazole derivatives with enhanced anticancer activity.