- AI-integrated PBPK and Quantitative Systems Pharmacology (QSP) for systems-level modeling of drug action, pharmacokinetics, and translational prediction - Multi-omics–driven graph learning for disease target identification and patient outcome prediction - Data-driven drug repurposing through the integration of genomic resources, real-world data, and molecular evidence - Risk factor discovery and patient stratification using prospective and retrospective cohorts, with emphasis on patient similarity modeling and cluster-based phenotyping - Generative AI for molecular design, coupled with docking simulation and in vitro property prediction for early-stage drug discovery - Quantitative reanalysis and reproducibility assessment of randomized clinical trials to evaluate therapeutic benefit–risk profiles - Domain-specific large language model development for biomedical knowledge extraction, clinical interpretation, and regulatory science applications