Key points are not available for this paper at this time.
The integration of multiomics technologies with artificial intelligence (AI) has become a transformative force in modern precision medicine, particularly within drug discovery. Multiomics approaches, including genome-wide association studies, transcriptomic profiling, proteomic interaction mapping, and metabolomic sequencing, provide unparalleled insights into the molecular dynamics of disease pathogenesis. Advanced AI methodologies, which leverage deep learning architectures, exhibit extraordinary capabilities in deciphering these intricate biological datasets, elucidating latent patterns, and constructing high-fidelity predictive models. The combined application of multiomics and AI has significant potential to accelerate target identification, streamline lead optimization processes, and enhance the precision of clinical trial designs. However, challenges persist, such as the need to harmonize disparate omics data streams, ensure reproducibility, and mitigate algorithmic biases. This review offers an in-depth analysis of multiomics applications across the drug development pipeline, covering target deconvolution, drug repositioning, and de novo compound discovery. It also explores the critical role of AI in drug discovery, focusing on virtual screening, pharmacokinetic modeling, and safety assessment frameworks. The fusion of multiomics with AI provides distinct advantages in hypothesis generation and data-driven discovery, opening new pathways for therapeutic innovation. By examining cases in oncology, neurodegenerative diseases, and cardiovascular conditions supported by robust technological infrastructures, this review presents a forward-thinking vision for future drug development. The convergence of these technologies not only enables comprehensive molecular understanding but also allows for more precise therapeutic interventions, marking the beginning of a new era in bench-to-clinic translational medicine.
Liu et al. (Tue,) studied this question.