摘要
P2Y12受体拮抗剂是一类重要的抗血小板药物,研究分子活性与其结构参数的关系,对于合成新的P2Y12受体拮抗剂具有一定指导作用.选用178个结构多样的P2Y12受体拮抗剂分子作为数据集,随机选取了143个P2Y12受体拮抗剂作为训练集,剩余分子作为检验集.采用多元线性回归(MLR)方法和主成分回归分析(PCA)方法对每个分子的636个分子参数进行线性回归分析.MLR所建模型的结果为:训练集R2=0.800,检验集R2=0.834;PCA模型结果为:训练集R2=0.545,检验集R2=0.665.相比之下MLR法所建模型具有良好的预测性和可靠性.通过模型分析,确定了影响分子活性的关键因素.以上模型对筛选和合成新型高效P2Y12受体拮抗剂提供了一定理论指导.
P2Y12 receptor antagonists are an important class of anti-platelet drugs.It has an important role in guiding the new synthesis of the P2Y12 receptor antagonists if we know the relationship between activities and structural parameters.In our work,178 P2Y12 receptor antagonist molecules were selected as the date set,with 143 molecules randomly selected as the training set and the rest as the test set.Using the multiple liner regression(MLR) and principle component analysis(PCA) methods,a 2D-QSAR study was carried out to explore the key factors that affect the activities of the P2Y12 receptor antagonist based on the 636 parameters of each molecule.The results show that MLR model presents a statistics result of training set R2=0.800,test set R2=0.834 and PCA model training set R2=0.545,test set R2=0.665.Comparing the two methods,it is shown that MLR exhibits good statistic results and the MLR model will be useful in predicting and developing P2Y12 receptor antagonists.
出处
《分子科学学报》
CAS
CSCD
北大核心
2012年第5期389-394,共6页
Journal of Molecular Science
基金
国家自然科学基金资助项目(10801025)