Identification of Natural Product-based Drug Combination (NPDC) Using Artificial Intelligence

Natural product-based drug combinations (NPDCs) offer unique advantages for treating complex diseases. Although high-throughput screening and traditional computational approaches have accelerated the discovery of synergistic drug combinations to a certain extent, their applications are still limited due to the fragmentation of experimental data, high experimental costs, and massive combinatorial space. Recently, artificial intelligence (AI) methods including traditional machine learning and deep learning algorithms have been developed and widely used in the identification of NPDCs. By integrating multi-source heterogeneous data and autonomously extracting features, the accuracy of prediction has been significantly improved, providing a powerful technical way to discover new NPDCs. Here, this review systematically summarizes the latest advances in AI-driven NPDC prediction, introduces related data resources and algorithmic frameworks, and analyzes current bottlenecks and future opportunities. AI methods are expected to significantly accelerate NPDC discovery and guide experimental validation.