基于人工神经网络模型的逍遥散功效药队抗抑郁配伍作用研究  

Study on the Compatibility of the Anti-depression Effects of Xiaoyao Powder's Efficacy Groups Based on Artificial Neural Network

作　　者：李孟宇 王雅泽 赵思俊 陈士剑 秦雪梅[1,2,3] 刘晓节[1,2,3] LI Mengyu;WANG Yaze;ZHAO Sijun;CHEN Shijian;QIN Xuemei;LIU Xiaojie(Modern Research Center for Traditional Chinese Medicine,Shanxi University,Taiyuan 030006,China;Key Laboratory of Chemical Biology and Molecular Engineering of Ministry of Education,Shanxi University,Taiyuan 030006,China;Key Laboratory of Effective Substances Research and Utilization in TCM of Shanxi Province,Shanxi University,Taiyuan 030006,China;Department of Pathophysiology,Huazhong University of Science and Technology,Wuhan 430015,China;Shanxi Institute for Food and Drug Control,Taiyuan 030006,China;School of Computer and Information Technology,Shanxi University,Taiyuan 030006,China)

机构地区：[1]山西大学中医药现代研究中心,山西太原030006 [2]山西大学化学生物学与分子工程教育部重点实验室,山西太原030006 [3]山西大学地产中药功效物质研究与利用山西省重点实验室,山西太原030006 [4]华中科技大学基础医学院,湖北武汉430015 [5]山西省食品药品检验所,山西太原030006 [6]山西大学计算机与信息技术学院,山西太原030006

出　　处：《山西大学学报（自然科学版）》2023年第4期969-980,共12页Journal of Shanxi University(Natural Science Edition)

基　　金：国家自然科学基金(82274360,81803962);山西省省筹资金资助回国留学人员科研项目(2022-026)。

摘　　要：中药复方与其药效之间复杂的非线性关系以及中药复方配伍关系的诠释是研究难点。文章采用人工神经网络数学模型解析逍遥散及其功效药队的抗抑郁配伍关系。首先,采用孤养结合慢性温和不可预知应激(CUMS)的方法建立抑郁大鼠模型;然后,明确逍遥散及其功效药队对抑郁大鼠行为学以及学习记忆能力的改善作用。在此基础上,对行为学数据进行收集整理后将所得行为学数据作为训练集,建立BP神经网络药效预测模型,应用模型对所测行为学的重要度进行排序。结果显示,旷场穿越格数>旷场直立格数>强迫游泳不动时间>水迷宫穿越平台次数>糖水偏爱率。同时,还对逍遥散、疏肝药队和健脾药队的行为学数据分别进行了BP神经网络回归分析,表明逍遥散对各指标的回归系数高于疏肝药队和健脾药队。研究结果表明,逍遥散及其功效药队可显著改善CUMS抑郁大鼠快感缺失、活动以及学习记忆能力减弱等核心精神症状,具有显著的抗抑郁作用。而疏肝药队和健脾药队协同发挥抗抑郁作用,验证逍遥散组方精当、配伍合理。文章亦表明人工神经网络模型在研究中药复方配伍上具有很大潜力。It is difficult to interpret the complicated non-linear relationships between traditional Chinese medicine(TCM)prescriptions and their effects,as well as the compatibility of TCM prescriptions.This study aimed to reveal the antidepressant compatibility of Xiaoyao Powder and its efficacy groups by artificial neural network(ANN).Firstly,chronic unpredictable mild stress(CUMS)was used to establish a rat depression model to investigate the effects of Xiaoyao Powder and its efficacy groups on the behaviors,learning,and memory abilities of CUMS rats.On top of this,the collected behavioral data were used as a training set to establish a BP neural network pharmacodynamic prediction model.The importance of the measured behavior was demonstrated by the ANN model.The order of the behavioral parameters is crossing times of open field test rearing of open field test immobility of forced swimming crossing times of Morris water maze test sucrose preference.At the same time,back propagation neural network analysis was performed on the behavioral data of Xiaoyao Powder,the Shugan group,and the Jianpi group.The results showed that the regression coefficients of Xiaoyao Powder in all indexes were higher than that of the Shugan group and the Jianpi group.The results of the study showed that Xiaoyao Powder and its efficacy groups significantly improved the core mental symptoms of CUMS depressed rats,exhibiting significant antidepressant effects.The Shugan group and the Jianpi group exert antidepressant effects in combination.The results suggest that the compatibility of Xiaoyao powder is precise and appropriate.This study also suggested ANN model has great potential in studying the compatibility of TCM prescriptions.

关 键 词：人工神经网络 功效药队 逍遥散 抑郁症 配伍关系 

分 类 号：R285.5[医药卫生—中药学]