Molecular simulation of the (GPx)-like antioxidant activity of ebselen derivatives through machine learning techniques

The selenoenzyme glutathione peroxidase (GPx) like activity of stable organoselenium compounds has been evaluated through the initial rate (ν0) reduction reaction H2O2, Cum-OOH, and t-BuOOH. A Quantitative Structure–Activity Relationships (QSAR) analysis based on different machine learning techniques was performed by employing atom-weighed algebraic maps indexes as descriptors. predictive capability obtained models statistically validated mean correlation coefficient for adjusting (R2), leave one out cross validation (Q2LOO), bootstrapping (Q2boot). For case H2O2 reduction, a model with six attributes (M2) values R2 = 0.907, Q2LOO 0.867, Q2boot 0.852. cum-OOH five (M15) statistical parameters: 0.925, 0.894, 0.873. t-BuOOH four descriptors (M19) found 0.938, 0.897, 0.856. parameters these three suggest that they are robust enough good capability. Finally, screening some related containing selenium two possible lead were (16 53), which can be used searching candidates GPx-like activity.