中药最优配方筛选的图论方法

A Graphic Algorithm for Mining Optimal Ingredient Pattern of Traditional Chinese Medicine

中药在治疗有些疾病时有独特的疗效,疗效的好坏主要取决于方剂中各药物的适当配伍。但是由于中药各药物之间的排列组合方式数目巨大,再加上药物剂量的变化,它们之间的组合数更是难以计数,对这些组合一一进行药效学实验难以实现,这就制约了中药的进一步研发。针对这一问题,该文将中药最优配方筛选问题转化为求一顶点赋权图的最大权团问题,利用求图的最大权团的算法实现中药最优配方筛选。该方法工作量小、简便易学、容易使用。

Traditional Chinese Medicine(TCM) exert unique effect in treating some sort of disease,and the effect depend on appropriate ingredient pattern of decoction.Though the number produced by randomly combining Chinese herbs in different dosage in a decoction is enormous,inspecting every combination in a way of pharmacodynamics is impossible.In result,the development of TCM is encumbered by this condition.In order to improve this condition,this paper presents converting mining optimal ingredient pattern of TCM into solving Maximum Weight Clique Problem of a vertex-weighted graph.This method can be applied in mining ingredient pattern of TCM easily and conveniently.