首页 | 本学科首页   官方微博 | 高级检索  
     

大黄素衍生物抗肿瘤活性的神经网络模型
引用本文:赵蔡斌,王占领,闵锁田,赖普辉.大黄素衍生物抗肿瘤活性的神经网络模型[J].计算机与应用化学,2007,24(3):341-344.
作者姓名:赵蔡斌  王占领  闵锁田  赖普辉
作者单位:陕西理工学院化学与环境科学学院,陕西,汉中,723000;陕西理工学院机械工程学院,陕西,汉中,723003
基金项目:陕西理工学院专项科研项目(SLGQD0508)
摘    要:目的:建立大黄素衍生物抗肿瘤活性的神经网络模型。方法:采用量子化学的AM1算法,计算了12个大黄素衍生物分子的结构参数,并用逐步回归分析,筛选结构参数。结果:利用筛选后的结构参数,建立大黄素衍生物抗肿瘤活性的神经网络模型。结论:抽一法交叉预报结果表明,本文建立的大黄素衍生物抗肿瘤活性的神经网络模型,预报结果可靠,具有一定的应用价值。

关 键 词:大黄素衍生物  抗肿瘤活性  构效关系  AM1算法  神经网络
文章编号:1001-4160(2007)03-341-344
修稿时间:2006-03-212006-07-28

The ANN model of antitumor activities of emodin erivatives
Zhao Caibin,Wang Zhanling,Min Suotian,Lai Puhui.The ANN model of antitumor activities of emodin erivatives[J].Computers and Applied Chemistry,2007,24(3):341-344.
Authors:Zhao Caibin  Wang Zhanling  Min Suotian  Lai Puhui
Affiliation:1. College of Chemistry and Environment Science, Shaan Xi University of Technology, Hanzhong, 723000, Shannxi, China; 2. College of Mechanical Engineering, Shann Xi University of Technology, Hanzhong, 723003, Shannxi, China
Abstract:AIM:to set up the neural network model of antitumor activities of emodin erivatives.METHOD:12 emodin erivatives' structural parameters were calculated using AM1 method of quantum chemistry.In addition,these parameters were selected by stepwise regression method.RESULTS:the neural network model was set up successfully using the preselected parameters.CONCLUSION:the results of the leave-one-out-cross-validation show the model is creditable and valuable for developing emodin drugs with high antitumor activities.
Keywords:emodin erivatives  antitumor activities  QSAR  AM1 method  artificial neural network
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号