Abstract
Traditional Medicine (TM), especially Traditional Chinese Medicine (TCM), is renowned for its distinctive "multi-component-multi-target-multi-pathway" mode of action, which exhibits a unique overall regulatory therapeutic effect. However, the intricate nature of TCM poses significant challenges in identifying active components, elucidating mechanisms of action, and standardizing clinical practices. The advancement of modern science and technology has led to the gradual modernization of TCM research. Network pharmacology (NP) has emerged as a pivotal framework for comprehending the holistic mechanisms of TCM, offering a crucial avenue for unveiling intricate biological networks by integrating chemical information, omics data, and clinical efficacy evidence. Nevertheless, conventional NP approaches exhibit notable limitations, including substantial noise, high dimensionality, challenges in capturing dynamics and time series, and inadequate cross-scale integration, thereby constraining their utility in precise mechanism analysis and clinical translation. In recent years, artificial intelligence (AI), particularly machine learning (ML), deep learning (DL), and graph neural networks (GNN), have empowered NP in an unprecedented way, enabling it to systematically and accurately analyze the cross-scale mechanisms of TCM from molecular interactions to patient efficacy. This review will systematically examine the latest developments in artificial intelligence-network pharmacology (AI-NP) methodology, with a focus on typical research cases of multi-scale mechanism analysis at the molecular, cellular, tissue, and patient levels. It will systematically summarize the challenges currently faced and explore future development directions to fully unlock the systemic therapeutic wisdom of TCM.