新蚁群算法在烷基酚类化合物的QSAR中变量选择的研究

Study of variable options based on modified ant colony algorithm in QSAR of Alkyl phenol compounds

采用量子化学从头算方法在HF/6-311+G(d)水平上计算8种烷基酚类化合物的分子结构描述符,选用修正过的C_p统计量为目标函和新蚁群优化算法,于烷基酚类化合物的定量结构——活性相关研究中的变量选择,建立烷基酚类化合物的生物降解速率常数与其量化参数之间的QSAR模型。结果表明,新蚁群优化算法用于定量构效中的变量选择比较简单,而且需要调节的参数少,是变量选择的有用方法,且应用量子化学结构参数建模的相关系数R =0.994,与文献中R=0.982相比相关性更好。

The quantum chemical descriptors of Alkylphenols were obtained at the HF/6-311 ＋ G （d） level with the ab initio method of quantum chemistry. The modified Cp statistics was chose to be the objective function, QSAR models were obtained for the biodegradation rate constant of Alkylphenols based on improved ant colony algorithm. The results showed that the modified ant colony method was a useful for variable selection. The regulative parameter was less and the algorithm was simpler. The correlation coefficient of the QSAR model based on some quantum chemical descriptors was 0. 994, which is better than the correlation coefficient of 0. 982 mentioned in the literature.