论文部分内容阅读
建立预测类黄酮化合物抑制恶性疟原虫株活性定量的模型,并确定影响类黄酮化合物活性的主要因素。本文选用了38个结构不同的类黄酮化合物作为数据集,采用多元线性同归法及主成分分析法分析每个化合物的220个分子参数,建立最优的预测模型。比较用不同方法建立的模型,结果发现带logP参数的向后筛选法为最优方法,所建模型统计结果良好(训练集相关系数R~2=0.81,标准训练误差SEE=0.27),模型代入检验集数据时结果也令人满意(检验集相关系数R~2=0.83,标准检验误差SEP=0.39),可靠性和预测性较强。脂水分配系数的对数logP为模型重要影响参数。建模和确定影响因素有助于筛选新型类黄酮抗疟疾药物和研发。
To establish a model for predicting the activity of flavonoid compounds in inhibiting the activity of Plasmodium falciparum and to identify the main factors influencing the activity of flavonoids. In this paper, 38 different flavonoid compounds were selected as data sets, and 220 molecular parameters of each compound were analyzed by multivariate linear regression and principal component analysis to establish the optimal prediction model. The results showed that the backward screening method with logP parameters was the best method, and the model was statistically good (training set correlation coefficient R ~ 2 = 0.81, standard training error SEE = 0.27) The results of the test set data are also satisfactory (test set correlation coefficient R ~ 2 = 0.83, standard test error SEP = 0.39), reliability and predictability. The logarithm of the lipid partition coefficient, logP, is an important parameter affecting the model. Modeling and determining influencing factors help screen new flavonoids anti-malarial drugs and research and development.