An explainable hypothesis-driven approach to Drug-Induced Liver Injury with HADES
For researchers in predictive toxicology and drug development, this work provides a new benchmark and an explainable approach for DILI prediction, though the performance on hypothesis generation is modest.
The authors argue that Drug-Induced Liver Injury (DILI) prediction should be framed as an explainable hypothesis-generation problem rather than binary classification. They introduce the DILER Benchmark with mechanistic hypotheses and present HADES, an agentic system that achieves a ROC-AUC of 0.68 on the Test Set and 0.59 on the Post-2021 Set, outperforming DILI-Predictor (0.63 and 0.50), and establishes a baseline for mechanistic hypothesis generation with a Hypothesis Alignment Fuzzy Jaccard Index of 0.16.
Drug-induced liver injury (DILI) remains a leading cause of late-stage clinical trial attrition. However, existing computational predictors primarily rely on binary classification, a framing that limits generalization and yields no mechanistic insight to guide translational decisions. We argue that DILI prediction is better posed as an explainable hypothesis-generation problem. To support this shift, we introduce the DILER Benchmark, a dataset that extends beyond binary labels by augmenting a curated set of molecules with mechanistic hepatotoxicity hypotheses derived from biomedical literature. We further present HADES, an agentic system designed to generate transparent and auditable reasoning traces. By combining molecular-level predictions, metabolite decomposition, structural understanding, and toxicity pathway evidence, HADES mechanistically assesses DILI risk. Evaluated on the DILER Benchmark, HADES outperforms existing models in binary classification, achieving a ROC-AUC of 0.68 on the Test Set and 0.59 on the challenging Post-2021 Set, compared with 0.63 and 0.50 for DILI-Predictor, respectively. More importantly, we establish a baseline for mechanistic hypothesis generation, where HADES achieves a Hypothesis Alignment Fuzzy Jaccard Index of 0.16. This result underscores the inherent complexity of the task while highlighting the need for advanced explainable approaches in predictive toxicology.