Ant Colony Optimization as a Powerful Tool for Descriptor Selection in QSPR Study of Infinite Dilution Activity Coefficients of Halogenated Hydrocarbons in Water

• 作者: Morteza Atabati
• 来源期刊: 结构化学
• 学科: 化学
• 关键词: 无限稀释活度系数 蚁群优化 QSPR 卤代烃 半经验量子化学计算 HYPERCHEM 工具 广义
• 年，卷（期）: 2012,（7）
• 页码范围: 953-958
• 页数: 6页
• 分类号: O623.11|TQ463.5
• 语种: 英文

摘要

A quantitative structure-property relationship （QSPR） study was suggested for the prediction of infinite dilution activity coefficients of halogenated hydrocarbons, γ∞ , in water at 298.15 K. After optimization of 3D geometry of the halogenated hydrocarbons with semi-empirical quantum chemical calculations at the AM1 level, different descriptors （1514 descriptors） were calculated by the HyperChem and Dragon softwares. A major problem of QSPR is the high dimensionality of the descriptor space; therefore, descriptor selection is the most important step. In this paper, an ant colony optimization （ACO） algorithm was proposed to select the best descriptors. Then the selected descriptors were applied for model development using multiple linear regression. The average absolute relative deviation and correlation coefficient for the training set were obtained as 4.36% and 0.951, respectively, while the corresponding values for the test set were 5.96% and 0.929, respectively. The results showed that the applied procedure is suitable for the prediction of γ∞ of halogenated hydrocarbons in water.