HerbSyner_Finder: a network community-based model for identifying synergistic combinations from herbal medicines and complex systems

Herbal medicine is a valuable resource for disease treatment, with enhanced synergistic efficacy and fewer side effects through combined herbal formulations. However, the synergistic mechanisms of action (MOAs) of these herbal medicines remain largely unexplored. Given the complexity of herbal systems, it is impractical to evaluate all possible drug/ingredient pairs experimentally. In this study, we propose a network-based model, HerbSyner_Finder, to prioritize synergistic ingredients in herbal medicine. By integrating network proximity and community analyses, HerbSyner_Finder could construct a multidimensional combinatorial atlas for complex biological systems to quantify herb-disease, ingredient-disease, herb-herb, and ingredient-ingredient interactions. Using cough variant asthma (CVA)-related herbal formulae as examples, kaempferol-quercetin and berberine-luteolin were successfully prioritized as synergistic for CVA among thousands of potential pairs. Further network analysis revealed that berberine and luteolin synergistically modulate the NLRP3/NF-κB signaling pathway, thereby alleviating CVA-associated inflammation. In summary, HerbSyner_Finder offers a tailored computational framework that efficiently identifies synergistic compounds from complex systems, and herbal medicines through high-throughput screening.