In silico prediction of dermal penetration rate of chemicals from their molecular structural descriptors

The dermal penetration rate of some volatile and non-volatile organic compounds was estimated by quantitative structure–activity relationship approaches by using interpretable molecular descriptors. Linear and nonlinear models were developed using multiple linear regressions (MLR) and artificial neural network (ANN) methods. Robustness and reliability of the constructed MLR and ANN models were evaluated by using the leave-one-out cross-validation method, which produces the statistics of QMLR2=0.786, Q ANN2=0.833 for non-volatiles and QMLR2=0.639, Q ANN2=0.712 for volatile compounds. Furthermore, the chemical applicability domains of these models were determined via leverage approach. The results of this study indicated the ability of developed QSAR models in the prediction of dermal penetration rate of various chemicals from their calculated molecular descriptors.

The term “artificial neural network” denotes a computational structure intended to model the properties and behavior of the brain structures, particular self-adaptation, learning and parallel processing.Figure optionsDownload as PowerPoint slideHighlights
► The results of developed models revealed that there is no significant difference between non-linear and linear models in the prediction of dermal penetration rate of various chemicals.
► The results of this study indicated the ability of QSAR in prediction of dermal penetration rate of various chemicals from their theoretically derived molecular descriptors.
► The result of this study reveals the superiority of developed model over the previous reported model.