论文部分内容阅读
目的研究23个色胺类迷幻剂的定量结构-活性关系(QSAR)。方法在DFT/B3LYP/6-311+G(d,p)水平上计算化合物的一系列量化性质;从分子轨道和电子结构方面进行量化分析;结合量化参数和一些物化性质,用多元线性回归(MLR)和人工神经元网络(ANN)方法建立了其与迷幻活性的QSAR模型。结果新建MLR和ANN模型的复相关系数分别为r2=0.721 0和r2=0.998 2。结论色胺类化合物与受体作用时以给电子为主,且吡咯环上的氮原子为主要的活性部位;影响活性的最主要因素是Qr5,Q1和I5;ANN方法建立的QSAR模型具有良好的稳定性和预测能力,优于MLR方法。
Objective To investigate the quantitative structure-activity relationship (QSAR) of 23 tryptamine hallucinogens. Methods A series of quantitative properties of compounds were calculated at the DFT / B3LYP / 6-311 + G (d, p) level. Quantitative analysis was performed on molecular orbital and electronic structures. Combined with quantitative parameters and some physicochemical properties, MLR) and artificial neural network (ANN) method to establish its QSAR model with psychedelic activity. Results The new correlation coefficients of MLR and ANN models were r2 = 0.721 0 and r2 = 0.998 2, respectively. Conclusions The tryptamine compounds are electron donating and the nitrogen atom on the pyrrole ring is the main active site. The most important factors influencing the activity are Qr5, Q1 and I5. The QSAR model established by the ANN method is good The stability and predictive ability is superior to the MLR method.