TCMFP: a novel herbal formula prediction method based on network target's score integrated with semi-supervised learning genetic algorithms.

This article focuses on the development and validation of a herbal formula prediction approach (TCMFP) that integrates traditional Chinese medicine (TCM) therapy experience, artificial intelligence, and network science algorithms to efficiently screen optimal herbal formulas for complex diseases. TCMFP employs three key scores: Hscore, measuring the effectiveness of individual herbs based on their targets’ importance in protein–protein interaction networks; Pscore, reflecting the empirical effectiveness of herb pairs derived from clinical and historical data; and FmapScore, which combines Hscore and Pscore to evaluate entire herbal formulas. Using genetic algorithms, TCMFP generated and validated effective herbal formulas for Alzheimer's disease, asthma, and atherosclerosis, with functional enrichment and network analyses supporting their therapeutic relevance. The study suggests that TCMFP offers a novel strategy for optimizing herbal formulas and advancing TCM-based drug development, while noting the need for further experimental validation and integration of additional TCM theories.