论文部分内容阅读
利用半经验量子化学PM3方法计算了106个HEPT类化合物的优势构象以及物理化学和电子结构等参数,讨论了该类化合物的这些参数对HIV-1逆转录酶抑制剂活性的影响,利用穷举回归、偏最小二乘和混沌遗传算法训练的人工神经网络方法建立了物理化学、电子结构等参数和其抑制HIV-1逆转录酶活性之间的定量结构活性关系模型。结果表明:(1)较大的logP、较高的最低空轨道能、较小的分子量均有利于提高该类化合物的抗HIV-1活性;(2)当该类化合物有适当大小的 R1取代基、C-6位苯环的间位被疏水性集团取代时该类化合物有较高的抗HIV-1生物活性。
The semi-empirical quantum chemical method PM3 was used to calculate the dominant conformation, physicochemical and electronic structure parameters of 106 HEPT compounds. The effects of these compounds on the activity of HIV-1 reverse transcriptase inhibitors were discussed. Regression, partial least squares and chaos genetic algorithm training artificial neural network method to establish the physical and chemical, electronic structure parameters and their inhibition of HIV-1 reverse transcriptase activity between the quantitative structure activity relationship model. The results showed that: (1) Larger logP, higher lowest lowest orbital energies, and lower molecular weight all contributed to the enhancement of the anti-HIV-1 activity of these compounds; (2) This compound has a higher anti-HIV-1 biological activity when the meta-position of the C-6 benzene ring is replaced by a hydrophobic group.