Integrative Network Pharmacology: Multilayer Omics, Bioinformatics, and Polypharmacological Drug Design

Abstract

Network pharmacology represents a transformative shift from "one drug–one target" paradigm to a systems-level approach that integrates biological networks, omics data, and computational intelligence. This comprehensive review articulates the conceptual foundations, architecture, and practical applications of network pharmacology in drug discovery and translational medicine. By incorporating multilayer omics—genomics, transcriptomics, proteomics, metabolomics, and single-cell data—with bioinformatics tools and graph theory, researchers can delineate disease modules, predict polypharmacological interactions, and repurpose drugs with increased precision. The work discusses cutting-edge computational strategies including graph neural networks, similarity network fusion, and explainable AI for robust modeling and interpretation. Case studies in oncology, neurodegeneration, infectious diseases, autoimmunity, and ethnopharmacology underscore the method’s potential in revealing novel mechanisms and multi-target therapeutics. Challenges such as incomplete interactomes, data heterogeneity, and model opacity are acknowledged, with solutions proposed through dynamic modeling, FAIR data practices, and ethical AI frameworks. Ultimately, network pharmacology emerges as a pivotal discipline enabling precision medicine through integrative, predictive, and mechanistically informed drug design.