QSAR Study of EC50 of Anti-HIV Drugs
Gene expression programming (GEP), a novel machine learning algorithm, was used to develop a quantitative model of anti-HIV compounds for the first time. Each compound was represented by several calculated structural descriptors, which include constitutional, topological, geometrical, electrostatic and quantum-chemical features of this compound. Descriptors were searched and selected by the heuristic method. This approach produces a nonlinear, five-descriptor quantitative model based on GEP with mean errors and correlation coefficient (R) being 0.41 and 0.91, respectively. The predicted results for both training set and test set of GEP are better than those of SVM and HM.
gene expression programming QSAR support vector machine,heuristic method HIV nucleoside
Hong Zong Si Ke Jun Zhang Shu Ping Yuan Ai Ping Fu Yun Bo Duan Zhi De Hu
Institute of Computer Science and Engineering Technology, Qingdao University,266071, Qingdao, P.R.Ch Department of Computer Science and Technology, Zhejiang University, 310027,Hangzhou, P.R. China Institute of Computer Science and Engineering Technology, Qingdao University,266071, Qingdao, P.R.Ch Department of Chemistry, Lanzhou University, 730000, Lanzhou, P.R.China
国际会议
杭州
英文
1253-1283
2007-04-01（万方平台首次上网日期，不代表论文的发表时间）