Chemical Research in Chinese Universities ›› 2013, Vol. 29 ›› Issue (3): 519-525.doi: 10.1007/s40242-013-2422-2

• Articles • Previous Articles     Next Articles

Prediction of Enhancement Effect of Nitroimidazoles onIrradiation by Gene Expression Programming

LONG Wei1, ZHANG Xiao-dong1, WANG Hao1, SHEN Xiu1, SI Hong-zong2,3, FAN Sai-jun1, ZHOU Ze-wei1, LIU Pei-xun1   

  1. 1. Key Laboratory of Tianjin Molecular Nuclear Medicine, Institute of Radiation Medicine, Peking Union Medical College, Chinese Academic of Medical Sciences, Tianjin 300192, P. R. China;
    2. Growing Base for State Key Laboratory of New Fibrous Materials and Modern Textile, Institute for Computational Science and Engineering, Qingdao University, Qingdao 266071, P. R. China;
    3. Department of Chemistry, Lanzhou University, Lanzhou 730000, P. R. China
  • Received:2012-11-02 Revised:2012-12-31 Online:2013-06-01 Published:2013-05-15
  • Supported by:

    Supported by the National Natural Science Foundation of China(No.81202153), the Research Fund for Youth Scholars of Union Medical College of China(No.2012D03), the Research Fund for the Doctoral Program of Higher Education of China (No.20121106120042) and the Development Fund of Institute of Radiation Medicine, Chinese Academy of Medical Sciences (No.SF1227).

Abstract:

A novel machine learning method, gene expression programming(GEP), was employed to build quatitative structure-activity relationship(QSAR) models for predicting the enhancement effect of nitroimidazole compounds on irradiation. The models were based on descriptors which were calculated from the molecular structures. Four descriptors were selected from the pool of descriptors by best multiple linear regression(BMLR) method. After that, three regression methods, multiple linear regression(MLR), support vector machine(SVM) and GEP, were used to build QSAR models. Compared to MLR and SVM, GEP produced a better model with the square of correlation coefficient(R2), 0.9203 and 0.9014, and the root mean square error(RMSE), 0.6187 and 0.6875, for training set and test set, respectively. The results show that the GEP model has better predictive ability and more reliable than the MLR and SVM models. This indicates that GEP is a promising method on relevant researches in radiation area.

Key words: Nitroimidazole, Irradiation enhancement effect , Gene expression programming