Leveraging Machine Learning for Network Pharmacology

Abstract

Network pharmacology is a growing area in drug discovery research that applies systems biology with network-based analysis to identify drug actions and relationships with multiple targets. Network pharmacology has transformed the traditional “one-target one-drug” paradigm to a highly effective “multi-target drug” hypothesis. However, this process is highly challenging, specifically analysing the humongous heterogeneous data and retrieving effective information such as drug targets, drugs, and mechanisms of action. The advancement of computer science and the development of various machine learning (ML) algorithms have remarkably transformed Big Data analytics and enabled the extraction of meaningful patterns from high-dimensional, heterogeneous datasets. Its applications span across multiple stages of the network pharmacology pipeline, including the screening of bioactive compounds (e.g., phytochemicals, microbial metabolites), target prediction, metabolic and signalling pathway analysis, protein–protein interaction (PPI) network analysis, host–pathogen interaction studies, hub gene identification, compound–target binding affinity prediction, etc. This review presents a comprehensive survey of core ML techniques applied in network pharmacology, outlining their current applications, future directions, and key challenges that must be addressed for broader implementation. Additionally, representative case studies are included to illustrate the practical impact of AI/ML integration in biomedical research and pharmaceutical innovation, emphasizing its potential to accelerate data-driven drug discovery.