Network analysis of kampo formulas based on crude drug composition and indications

Abstract

Aim Many kampo formulas have been developed by combining crude drugs in various ways, providing diverse medicinal effects. Similarity between kampo formulas is usually determined based on their compositions and medicinal effects, but the entire network structure of Kampo formulas has not been analyzed systematically. In this study, I investigated the associations between kampo formulas based on their compositions and indications in order to provide an overview of the complex network structure of kampo formulas. In addition, relationships between crude drugs and indications were also investigated in a similar manner. Methods The list of the 294 over‐the‐counter kampo formulas defined by the Ministry of Health, Labour and Welfare, Japan was analyzed to construct networks of kampo formulas. Kampo formulas were connected by edges if their crude drug compositions or indications had significant overlap. A network of crude drugs and indications was also constructed by linking crude drugs and indications if they co‐occurred in significantly many kampo formulas. Results Three groups related to keishito, shosaikoto, and rikkunshito were identified in the network of kampo formulas based on crude drug composition. Three groups related to the treatment of gastric disorders, women's diseases, and neuralgia were also identified in the network of kampo formulas based on indications. Major groups were also identified in the network of crude drugs and indications. Conclusion The networks of kampo formulas as well as crude drugs and indications were successfully obtained, which may help to facilitate our understanding of kampo formulas and crude drugs.