The invention discloses a
traditional Chinese medicine prescription effective
small molecule identification method fusing EGA and PPO algorithms, and relates to the technical field of
traditional Chinese medicine decoction
small molecule identification, and the method comprises the following steps: constructing a molecular seed
bank containing TPACD known components and phytochemicals with similar structures, simulating
material selection, conversion and recombination mechanisms in a
traditional Chinese medicine decoction process through EGA, and identifying the effective small molecules in the traditional Chinese
medicine decoction process. Three parallel modules of a basic multi-objective
genetic algorithm (BMGA), an enhanced multi-objective
genetic algorithm (EMGA) and an intelligent multi-objective
genetic algorithm (IMGA) are established by combining
sequential decision optimization capability of PPO
reinforcement learning, and anti-CRC lead molecules with effectiveness, stability and
druggability are screened out through
molecular evolution,
targeted screening, iterative refining and closed-loop feedback. The molecules generated by the invention are obviously superior to the original component of TPACD in
drug-likeness,
structural stability and anti-CRC activity, the IMGA module has the best comprehensive performance, and an efficient and explainable new strategy is provided for excavation of
active components of traditional Chinese
medicine compounds.