论文部分内容阅读
采用同源建模法对α1A-,α1B-和α1D-AR的三维结构进行了模拟,并采用分子力学、分子动力学方法对所得同源模型进行优化,然后分别采用训练集拮抗剂对接的方法得到拮抗状态下的α1A-,α1B-和α1D-AR三维结构模型.得到的模型再采用FRED对接软件对测试集中的18个化合物进行对接并打分,再将所得打分结果与其活性进行线性回归,其回归结果具有良好的拟合效果,由此回归方程预测的活性与化合物实验值较吻合,说明我们建立的拮抗状态下的α1A-,α1B-和α1D-AR的三维同源模型具有一定的合理性,可作为化合物虚拟筛选模型,对新化合物进行对接虚拟筛选.
The three-dimensional structures of α1A-, α1B- and α1D-AR were simulated by homology modeling, and the homology models were optimized by molecular mechanics and molecular dynamics method. Then, Obtained three-dimensional structure model of α1A-, α1B- and α1D-AR in antagonistic state.The obtained model was then used to connect and score 18 compounds in the test set by FRED docking software, and then the result of scoring was linear regression with its activity The regression results have a good fitting effect, and the predicted activity of the regression equation is in good agreement with the experimental values of the compounds, indicating that the three-dimensional homology model of α1A-, α1B- and α1D-AR in the antagonistic state established by us is reasonable , Which can be used as a virtual screening model for compound to virtual screen new compounds.