Fig. 5.

Nextcast pipeline for biomarkers identification and QSAR models development from toxicogenomics and cheminformatics data. Raw omics data can be preprocessed with eUTOPIA. Then, the preprocessed transcriptomics data, chemical representation data, and the outcome variable can be provided to hyQSAR or MaNGA to identify the optimal predictive model. The red boxes indicate the input for the pipeline while the green box is the output. The dark blue and the yellow box are the individual Nextcast components, and the light blue box represents the intermediate output/input.