In silico prediction of Tetrahymena pyriformis toxicity for diverse industrial chemicals with substructure pattern recognition and machine learning methods

► A total of 1571 diverse unique chemicals were collected from the literature and composed of the largest diverse data set for Tetrahymena pyriformis toxicity. ► Robust and high predictive accuracy classification models for T. pyriformis toxicity prediction were developed by substructure pattern recognition and different machine learning methods. ► Some useful substructure patterns for characterizing T. pyriformis toxicity were also identified via the information gain analysis methods.