Artificial Intelligence-Driven Multi-Omics Integration in Precision Medicine: Advances, Challenges, and Future Perspectives

Precision medicine is increasingly shaped by the need to interpret biological variation at a depth that conventional clinical models cannot provide. Omics technologies have made this possible, but they have also created a data landscape that is fragmented, high-dimensional, and difficult to analyze. In this setting, artificial intelligence offers a practical route for connecting molecular signals across genomic, transcriptomic, proteomic, metabolomic, and epigenomic layers. This review examines how AI methods are being used to merge these data sources and translate them into clinically useful knowledge. We focus on model classes that are most relevant to current biomedical research, including machine learning, deep learning, graph-based methods, transformer architectures, and foundation models. Their applications are considered across biomarker discovery, disease classification, risk estimation, treatment selection, and therapeutic planning. The review also addresses the expanding role of AI in nanomedicine and intelligent drug delivery, where computational methods are increasingly used to support adaptive therapy design, response prediction, repurposing strategies, and nanocarrier optimization. Alongside these opportunities, we discuss the main barriers to clinical implementation, including data incompleteness, interoperability constraints, privacy issues, bias, interpretability, and regulatory uncertainty. Finally, we outline emerging directions such as generative AI, digital twins, federated learning, and real-time precision therapeutics. Taken together, these developments suggest that AI-enabled multi-omics integration may become a central component of future precision healthcare.

کلیدواژه ها:

Artificial intelligence ، Precision medicine ، Multi-omics integration ، Biomarker discovery ، Nanomedicine ، Personalized therapeutics