分子结构分区方法及在二硫化合物QSRR研究中的应用  

Quantitative structure-retention relationships study of disulfides compounds by dividing its molecular structure into substructure

作　　者：程彬[1] 刘凤萍[1] 曹晨忠[1] 

机构地区：[1]湖南科技大学化学化工学院,湖南湘潭411201

出　　处：《计算机与应用化学》2011年第1期73-77,共5页Computers and Applied Chemistry

基　　金：国家自然科学基金(20772028)

摘　　要：从二硫化合物分子结构特征出发,并结合色谱分离机制,将分子结构进行分区处理,通过所建构的总拓扑立体效应指数(TTSEI)、顶点度-距离指数(VDI)、奇偶指数(OEI)、极化效应指数(PEI)等拓扑结构参数,用多元线性回归(MLR)方法获得了二硫化合物在4种不同极性固定相上气相色谱保留指数(RI)与4个结构参数之间的定量结构-色谱保留相关(QSRR)的统一模型,相关系数均大于0.98,以留一法(Leave-one-out)进行交互检验,相关系数R_(cv)均大于0.97。模型方程用于预测各二硫化合物的气相色谱保留指数其稳定性和准确性俱佳。In order to investigate the quantitative structure-retention relationship of disulfides compounds, the molecular structure is divided into two parts, R and X, to obtain molecular structure parameters. The topological indices： vertex degree-distance index （VD1）, odd-even index （OEI）, the total topological steric effect index （TTSE1） and polarization effect index （PE1） were extended and used to predict the chromatographic retention indices （R/） for disulfides compounds. Simple linear regressions between the retention indices and molecular structural descriptors were established. The correlation coefficients were larger than 0.98. The performance of the model was tested through cross-validation by the Leave-one-out procedure （LOO）, showing good statistical parameters. Statistical analysis showed that the quantitative structure-retention relationship （QSRR） models have high internal stability and good predictive ability.

关 键 词：分子结构分区 二硫化合物 拓扑指数 定量结构-色谱保留相关 

分 类 号：O658[理学—分析化学]