Combining In Vivo Data with In Silico Predictions for Modeling Hepatic Steatosis by Using Stratified Bagging and Conformal Prediction.

Hepatic steatosis (fatty liver) is a severe liver disease induced by the excessive accumulation of fatty acids in hepatocytes. In this study, we developed reliable in silico models for predicting hepatic steatosis on the basis of an in vivo data set of 1041 compounds measured in rodent studies with repeated oral exposure. The imbalanced nature of the data set (1:8, with the "steatotic" compounds belonging to the minority class) required the use of meta-classifiers-bagging with stratified under-sampling and Mondrian conformal prediction-on top of the base classifier random forest. One major goal was the investigation of the influence of different descriptor combinations on model performance (tested by predicting an external validation set): physicochemical descriptors (RDKit), ToxPrint features, as well as predictions from in silico nuclear receptor and transporter models. All models based upon descriptor combinations including physicochemical features led to reasonable balanced accuracies (BAs between 0.65 and 0.69 for the respective models). Combining physicochemical features with transporter predictions and further with ToxPrint features gave the best performing model (BAs up to 0.7 and efficiencies of 0.82). Whereas both meta-classifiers proved useful for this highly imbalanced toxicity data set, the conformal prediction framework also guarantees the error level and thus might be favored for future studies in the field of predictive toxicology.

A quantitative structure-toxicity relationships model for the dermal sensitization guinea pig maximization assay.

We have developed quantitative structure-toxicity relationship (QSTR) models for assessing dermal sensitization using guinea pig maximization test (GPMT) results. The models are derived from 315 carefully evaluated chemicals. There are two models, one for aromatics (excluding one-benzene-ring compounds), and the other for aliphatics and one-benzene-ring compounds. For sensitizers, the models can resolve whether they are weak/moderate or severe sensitizers. The statistical methodology, based on linear discriminant analysis, incorporates an optimum prediction space (OPS) algorithm. This algorithm ensures that the QSTR model will be used only to make predictions on query structures which fall within its domain. Calculation of the similarities between a query structure and the database compounds from which the applicable model was developed are used to validate each skin sensitization assessment. The cross-validated specificity of the equations ranges between 81 and 91%, and the sensitivity between 85 and 95%. For an independent test set, specificity is 79%, and sensitivity 82%.