%0 Journal Article
%T 基于遗传算法神经网络的抗乳腺癌候选药物优化建模<br>Optimization Modeling of Anti-Breast Cancer Drug Candidate Based on Genetic Algorithm Neural Network
%A 辜承梁
%A 毛翊丞
%A 吴雅南
%J Modeling and Simulation
%P 346-357
%@ 2324-870X
%D 2022
%I Hans Publishing
%R 10.12677/MOS.2022.112031
%X 乳腺癌已经成为危害全球女性健康的主要癌症之一。拮抗ERα活性的化合物可能是治疗乳腺癌的候选药物，本文通过对1974个与ERα的生物活性有关的化合物进行研究，对分子描述符进行斯皮尔曼等级相关性分析，为了降低变量之间相关性对结果的影响，还需进行系统聚类分析，提取其中20个对ERα的生物活性影响最大的分子描述符。采用遗传算法优化的BP神经网络建立出ERα生物活性定量预测模型，再利用支持向量机SVM算法构建化合物ADMET性质分类预测模型，最后利用多目标优化思想结合遗传算法寻优，得出了使抗乳腺癌药物具有最优效果的分子描述符及其取值。<br />
Breast cancer has become one of the major cancers endangering women’s health all over the world. Compounds antagonizing the activity of ERα may be candidates for the treatment of breast cancer. In this paper, 1974 compounds related to the biological activity of ERα were studied, and spearman rank correlation analysis of molecular descriptors was conducted. In order to reduce the influence of the correlation between variables on the results, systematic cluster analysis was also needed. Twenty molecular descriptors with the greatest influence on the biological activity of ERα were extracted. Using genetic algorithm to optimize the BP neural network to build out ERα bioactive quantitative prediction model, using support vector machine SVM algorithm to construct compound ADMET properties classification prediction model, finally using combined optimization genetic algorithm, the molecular descriptors and their values for the optimal effect of anti-breast cancer drugs were obtained.&nbsp;
%K 抗乳腺癌药物优化，聚类分析，BP神经网络，遗传算法，支持向量机，多目标优化<br>Anti-Breast Cancer Drug Optimization
%K Cluster Analysis
%K BP Neural Network
%K Genetic Algorithm
%K Support Vector Machine
%K Multi-Objective Optimization
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=49330