【摘要】 采用遗传算法研究了一系列药物分子的Caco鄄2细胞表观穿透系数(lgPeff)和分子结构之间的关系.基于51个化合物构成的训练集,计算得到了一组效果较好的定量构效关系(QSAR)模型.这些模型不仅具有较好的回归能力,还能对预测集中的分子进行较好的预测.在计算得到的精华种群中,共有4个分子参数具有较高的出现频率,它们分别是lgD(表观酯水分配系数)、rgyr(回旋半径)、Shadow鄄Xlength(分子在X维上的投影长度)以及NHBD(氢键给体数目).更多还原

【Abstract】 In current work, the correlations between Caco-2 permeability(lgPeff) and molecular structures were investigated using genetic algorithm(GA). A group of multiple regression models with good statistical significance were obtained based on a training set of 51 molecules, from the statistical analyses of the descriptors used in the evolution procedure, four of them, including the distribution coefficient(lgD), radius of gyration(rgyr), shadow-Xlength(length of molecule in the X dimension), and NHBD, may have the largest impact on diffusion through Caco-2 cell.更多还原