论文部分内容阅读
采用按氢分类的分子电性距离矢量(H-MEDV)为结构描述筒,对30个脂肪族醛酮类化合物进行了结构表征。采用逐步回归(SMR)方法对变量进行了筛选,多元线性回归(MLR)方法对化合物的结构与离子迁移常数(K_0)之间的关系进行建模。所得模型的复相关系数(R)为0.984,标准偏差(SD)为0.030。采用“留一法(LOO)”对模型进行了交互检验,得到的复相关系数(R_(cv))为0.973,标准偏差(SD_(cv))为0.039。结果显示该模型具有较强的预测能力及良好的稳定性,对于有机化合物离子迁移谱的QSSR研究具有一定的参考价值。
Thirty aliphatic aldehydes and ketones were characterized by using H-MEDV (Molecular Electrical Distance Vector) as structural descriptors. The variables were screened by the stepwise regression (SMR) method and the relationship between the structure of the compound and the ion mobility constant (K0) was modeled by the multiple linear regression (MLR) method. The resulting model had a complex correlation coefficient (R) of 0.984 and a standard deviation (SD) of 0.030. The model was tested by using “LOO” method. The obtained complex correlation coefficient (R cv) was 0.973 and the standard deviation (SD cv) was 0.039. The results show that the model has strong predictive ability and good stability, and has certain reference value for the QSSR study of ion mobility spectra of organic compounds.