<p>Drug-Induced Liver Injury (DILI) is a major challenge in drug development, occurring due to liver damage caused by the adverse effects of drugs or xenobiotics. High-throughput transcriptomics (HTTr) provides mechanistic insights into drug-induced hepatotoxicity, complementing traditional chemical structure-based methods. To address the challenges posed by DILI, this study aimed to evaluate the suitability of HTTr data for DILI classification and prediction. Initially, we reviewed the current landscape of HTTr-based DILI research, focusing on public datasets, computational tools, and bioinformatics techniques. Building on this foundation, we analyzed HTTr data from the Open TG-GATEs database, which includes primary human hepatocytes treated with 146 drugs at three concentrations. Gene expression data alone had limited ability to classify DILI phenotypes, performing similarly to chemical structure-based models. However, targeted gene sets improved clustering performance, and changes in clustering performance across concentration levels indicated that concentration information influences toxicity analysis. Machine learning models showed that integrating gene expression and chemical structure data enhanced predictive accuracy, emphasizing the need for multi-modal approaches. These findings underscore HTTr as a valuable tool for advancing DILI classification and prediction, contributing to more reliable drug safety assessments.</p>