DINeuroTpredictor: in silico prediction for drug-induced Neurotoxicity (DINeuroT) with machine learning
Abstract
Neurotoxicity is one of the main reasons for the withdrawal of drugs from the market. Thus, it is of great importance to assess drug-induced neurotoxicity of compounds accurately and rapidly. In this study, we focused on the electronic prediction of neurotoxic chemicals about people and analyzed the mechanisms of neurotoxic action of chemicals. The model developed with the support vector machine (SVM) achieved a prediction accuracy of 76.50% on the training set and 83.70% on test set. We hope the model could be useful complementary tool for chemical neurotoxicity evaluation.