Cellarity published a framework in Nature Communications outlining an integrated multi‑omics and AI approach to predict drug‑induced liver injury (DILI). The company combined transcriptomics, proteomics and computational modeling to identify cellular state changes that precede liver toxicity and trained models to forecast DILI risk from early‑stage compound data. Cellarity says the framework improves sensitivity and specificity over conventional in vitro assays and offers a path to earlier de‑risking during lead optimization. The approach aims to reduce late‑stage attrition by flagging hepatotoxic liabilities earlier and providing mechanistic hypotheses for medicinal chemistry mitigation.