三维全息原子场作用矢量(3D-HoVAIF)用于神经氨酸酶抑制剂的结构表征与设计及定量构效关系(QSAR)研究

Chemical Structural Characterization and Design of Influenza Neuraminidase Inhibitors Using Three-dimensional Holographicvector of Atomic Interaction Field Through Quantitative Structure-Activity Relationship Studies

对100个神经氨酸酶抑制剂抗禽流感药物结构并与其活性建立定量构效关系模型。采用本实验室提出的三维全息原子场作用矢量(3D-HoVAIF)对100个神经氨酸酶抑制剂进行结构表征,然后采用逐步回归对变量进行筛选后,运用偏最小二乘建立3D-HoVAIF描述子与神经氨酸酶抑制剂活性之间的QSAR模型。结果表明:复相关系数(R),交互校验的复相关系数(Q2)和模型的标准偏差(SD)分别为R2=0.805、Q2=0.657和SD=0.936,模型具有良好的稳定性和预测能力,并对文献中23个药物和设计的32个化合物进行了预测。表明三维全息原子场作用矢量能较好表征该类分子结构信息值得进一步推广应用。

To study the quantitative structure-activity relationship （QSAR） of 100 influenza neuraminidase inhibitors, the methods of three-dimensional holographicvector of atomic interaction field （3D-HoVAIF） was used to describe the chemical structure of influenza neuraminidase inhibitors. After the structural characterization, the descriptors obtained were screened by least square regression （PLS）. It was found that the obtained model with the cumulative multiple correlation coefficient （R^2cum）, cumulative cross-validated （Q^2cum） and standard error of estimation （SD） were R^2cum = 0. 805, Q^2cum = 0. 657 and SD =0.936 respectively. The result showed the model had favorable stability and good prediction capability. It has been used for predicting the 23 drugs from the published neport as well as 32 compounds designed from our lab.. The 3D-HoVAIF was applicable to the molecular structural charac-terization and biologicalactivity prediction.