AUTHOR=Wang Hao , Liu Ruifeng , Schyman Patric , Wallqvist Anders TITLE=Deep Neural Network Models for Predicting Chemically Induced Liver Toxicity Endpoints From Transcriptomic Responses JOURNAL=Frontiers in Pharmacology VOLUME=Volume 10 - 2019 YEAR=2019 URL=https://www.frontiersin.org/journals/pharmacology/articles/10.3389/fphar.2019.00042 DOI=10.3389/fphar.2019.00042 ISSN=1663-9812 ABSTRACT=Improving the accuracy of toxicity prediction models for liver injuries is a key element in evaluating the safety of drugs and chemicals. Mechanism-based information derived from expression (transcriptomic) data, in combination with machine-learning methods, promises to improve the accuracy and robustness of current toxicity prediction models. Deep neural networks (DNNs) have the advantage of automatically assembling the relevant features from a large number of input features. This makes them especially suitable for modeling transcriptomic data, which typically contain thousands of features. Here, we gauged gene- and pathway-level feature selection schemes using the DNN approach in predicting chemically induced liver injuries (biliary hyperplasia, fibrosis, and necrosis) from whole-genome DNA microarray data. The DNN models showed high predictive accuracy and endpoint specificity, with Matthews correlation coefficients for the three endpoints on 10-fold cross validation ranging from 0.69 to 0.90, with an average of 0.83 in the five best curated sets of either preselected genes or pathways. The DNN models performed more robustly than did random forest and support vector machine classifiers. The effects of feature selection scheme were negligible among the best-curated sets. Further evaluation of the models on their ability to predict the injury phenotype per se for non-chemically induced injuries revealed the robust performance of the DNN models across these additional external testing datasets. Thus, the DNN models learned features specific to the injury phenotype contained in the gene expression data.