Application of an expert system based on Genetic Algorithm–Adaptive Neuro-Fuzzy Inference System (GA–ANFIS) in QSAR of cathepsin K inhibitors

One strategy to potentially improve the success of drug design and development is to use chemometrics methods early in the process to propose molecules and scaffolds with ideal binding and to clarify physicochemical features influencing in their activity. Adaptive Neuro-Fuzzy Interference System (ANFIS) was used to construct the nonlinear quantitative structure–activity relationship (QSAR) model. The Genetic Algorithm (GA) was used to select descriptors which are responsible for the cathepsin K inhibitory activity of studied compounds. ANFIS regression is a nonlinear regression technique developed to relate many regressors to one or several response variables. The accuracy of the generated QSAR model (R2 = 0.916) is described using various evaluation techniques, such as leave-one-out procedure (RLOO2=0.875) and validation through an external test set (Rpred2=0.932).

► An expert system is developed based on ANFIS and Genetic Algorithm (GA).
► Combination of Genetic Algorithm and ANFIS in QSAR studies is very rare.
► Based on the results obtained, GA-ANFIS has been demonstrated to be an effective QSAR model.