Search

The novel severe acute respiratory syndrome coronavirus (SARS CoV-2) was introduced as an epidemic in 2019 and had millions of deaths worldwide. Given the importance of this disease, the recommendation and design of new active compounds are crucial. 3-chymotrypsin-like protease (3 CLpro) inhibitors have been identified as potent compounds for treating SARS-CoV-2 disease. So, the design of new 3 CLpro inhibitors was proposed using a quantitative structure-activity relationship (QSAR) study. In this context, a powerful adaptive least absolute shrinkage and selection operator (ALASSO) penalized variable selection method with inherent advantages coupled with a nonlinear artificial neural network (ANN) modelling method were used to provide a QSAR model with high interpretability and predictability. After evaluating the accuracy and validity of the developed ALASSO-ANN model, new compounds were proposed using effective descriptors, and the biological activity of the new compounds was predicted. Ligand-receptor (LR) interactions were also performed to confirm the interaction strength of the compounds using molecular docking (MD) study. The pharmacokinetics properties and calculated Lipinski's rule of five were applied to all proposed compounds. Due to the ease of synthesis of these suggested new compounds, it is expected that they have acceptable pharmacological properties. [ABSTRACT FROM AUTHOR]

The inhibitory activities (pIC50) of N2 and O6 substituted guanine derivatives as cyclin-dependent kinase 2 (CDK2) inhibitors have been successfully modeled using calculated molecular descriptors. Two linear (MLR) and nonlinear (ANN) methods were utilized for construction ofmodels to predict the pIC50 activities of those compounds.The QSARmodels were validated by cross-validation (leave-one-out) as well as application of the models for prediction of pIC50 of external set compounds. Also, the models were validated by calculation of statistical parameters and Y-randomization test. Two methods provided accurate predictions, although more accurate results were obtained by ANNmodel.Themean-squared errors (MSEs) for validation and test sets ofMLR are 0.065, 0.069 and of ANN are 0.017 and 0.063, respectively. [ABSTRACT FROM AUTHOR]