论文部分内容阅读
紫杉醇是从紫杉或称红豆杉中提取的1种天然抗癌物质,具有独特的抗癌机理。由于紫杉醇的种种限制,开发具有更高抗癌活性的紫杉醇类似物药物具有广阔的前景。本文选用36个结构多样的紫杉醇类似物分子作为数据集,随机选取其中28个作为训练集,其它为检验集,采用多元线性回归(MLR)法及主成分回归分析(PCA)法分析每个化合物的197个分子参数,分别建立定量构效关系的最优预测模型。并用检验集检验所建模型的预测能力。结果表明:多元线性回归分析法所建模型与主成分回归所建模型相比,发现逐步筛选法为最优建模方法。该方法所建模型统计结果良好(R~2=0.846,SEE=1.060),应用于检验集时,结果也比较满意(R~2=0.841,SEP=1.071),模型的可靠性和预测性较强。建模和确定主要影响因素有助于指导筛选和研发新型类紫杉醇药物。
Paclitaxel is a natural anticancer substance extracted from yew or yew, with a unique anti-cancer mechanism. Because of the limitations of paclitaxel, the development of paclitaxel analogs with higher anticancer activity has broad prospects. In this paper, 36 structurally diverse paclitaxel analogues were selected as data sets, 28 were randomly selected as the training set, the others were test sets, and each compound was analyzed by multiple linear regression (MLR) and principal component analysis (PCA) Of the 197 molecular parameters, respectively, to establish the quantitative structure-activity relationship of the optimal prediction model. The test set was used to test the predictive ability of the model. The results show that the multivariate linear regression analysis model compared with the principal component regression model found that step by step screening method for the optimal modeling method. The statistical results of the proposed method are good (R ~ 2 = 0.846, SEE = 1.060). When applied to the test set, the results are satisfactory (R ~ 2 = 0.841, SEP = 1.071). The reliability and predictive model Strong. Modeling and identifying key influencing factors can help guide the screening and development of new classes of paclitaxel drugs.