摘要
肝细胞生长因子(HGF)/酪氨酸蛋白激酶(c-Met)介导的细胞信号是导致肿瘤细胞产生和转移的主要途径之一。c-Met抑制剂能够阻断HGF/c-Met信号,抑制人类肿瘤的转移发生。本文选用77个结构多样的吡唑啉酮类衍生物作为c-Met抑制剂分子的数据集,随机选取其中16个分子作为检验集,其余作为训练集,采用多元线性回归(MLR)和主成分回归分析(PCA)法对每个分子的648个参数进行回归分析,分别建立定量构效关系的最优预测模型。结果表明,多元线性回归中的逐步筛选法是最佳的建模方法,其所建模型统计结果良好(R2=0.81,SEE=0.37),应用于检验集的结果也较理想(R2=0.83,SEP=0.42),模型可靠性和预测能力较强,能直观反映影响活性的主要因素。此模型的确立有助于指导新型高效c-Met抑制剂药物的筛选和开发。
Cells signal induced by hepatocyte growth factor（HGF） /tyrosine kinase（c-Met） is one of main approaches to induce the tumor growth and metastasis.c-Met inhibitors can block the HGF / c-Met signaling pathway and inhibit tumor growth and metastasis.A dataset composed of 77 pyrazolone-based c-Met receptor inhibitors with diversified structures was built,in which 16 molecules served as the test set and the rest as the training set.Altogether,648 molecular indices were regressed by multiple linear regression（MLR） and principle component regression analysis（PCA） methods,finally ending up with the optimum predictable mathematic models.The results showed that stepwise regression analysis was the optimal regression method.The asbuilt model showed satisfactory statistical results（R^2 = 0.81,SEE = 0.37）,whose proper predictability was validated by the independent test set（R^2 = 0.83,SEP = 0.42）.This model is helpful for further researching and development of new,efficient cMet inhibitors.
出处
《化学通报》
CAS
CSCD
北大核心
2012年第10期902-902,共1页
Chemistry
