高等学校化学学报 ›› 2004, Vol. 25 ›› Issue (6): 1100.

• 研究论文 • 上一篇    下一篇

连接树方法在吡啶类化合物pKa值预测中的应用

齐玉华1, 张庆友2, 罗操操1, 王俊1   

  1. 1. 中国科学院长春应用化学研究所, 长春 130022;
    2. 哈尔滨工业大学, 哈尔滨 150001
  • 收稿日期:2003-07-08 出版日期:2004-06-24 发布日期:2004-06-24
  • 通讯作者: 许 禄(1941年出生),男,研究员,博士生导师,从事计算机化学的研究.E-mail:Luxu@ciac.j.lcn E-mail:Luxu@ciac.j.lcn
  • 基金资助:

    国家自然科学基金(批准号:20077026)资助

Application of Tree Structured Fingerprint to the Prediction of pKa of Pyridine Derivatives

QI Yu-Hua1, ZHANG Qing-You2, LUO Cao-Cao1, WANG Jun1   

  1. 1. Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun 130022, China;
    2. Harbin Institute of Technology, Harbin 150001, China
  • Received:2003-07-08 Online:2004-06-24 Published:2004-06-24

摘要: 将连接树的方法应用到吡啶类化合物pKa值的预测中,并将该方法计算出的参数与量子化学参数相结合.变量的两两相关性检验结果表明,所选择的10个参数相关性较小,同时用交叉验证方法所得到的结果表明,所构造的多元回归模型十分稳定.通过人工神经网法对回归模型进一步优化,得到满意的结果.

关键词: 定量构效关系, 连接树方法, 吡啶类化合物, pKa值, 人工神经网

Abstract: For the prediction of pKa of pyridine derivatives,the combinations of the parameters by using the method of tree structured fingerprint and the quantum-chemical parameters was performed. The results of the cross-correlation among the selected ten parameters showed that they possessed low correlations. And the results of the cross-validation method indicated that the multiple linear regression model was very steady. Furthermore,the pKa values were estimated by using artificial neural networks.

Key words: Quantitative structure-property relationship, Tree structured fingerprint, Pyridine derivatives, pKa value, Artificial neural networks

中图分类号: 

TrendMD: