头孢菌素类抗生素毒性的定量构效关系预测研究
QSAR prediction of toxicity of cephalosporin antibiotics
本文对 25种头孢菌素类抗生素的半数致死量(LD50)与其分子结构参数之间进行了定量构效关系(QSAR)建模。首先采用CODESSA软件计算了头孢菌素类抗生素关于组成、拓扑、几何、电子和量子化学的554个分子描述符,然后基于启发式算法挑选出不同个数的分子描述符构建多元线性回归方程。结果表明当采用7个描述符时,相关系数和交叉验证系数趋于稳定,模型相关系数R2=0.9470。
his article trys to buid a QSAR model to predict cephalosporin antibiotics toxicity and find out the relationship between 25 cephalosporin antibiotics' structure parameters and their median lethal dose (LD50) data. We firstly use the CODESSA software to calculate 554 descriptors, including composition, topology, geometry, electronics, and quantum chemistry molecular descriptors. Then we use a heuristic algorithm to select the optimal molecular descriptors to build a multi-linear function. The results show that when using 7 descriptors, the higest correlation coefficient and cross-validation coefficient are obtained and the correlation coefficient R2 is 0.9470.
李梦龙、郭延芝、李益洲、王羽
药学基础医学生物科学研究方法、生物科学研究技术
头孢菌素类抗生素定量构效关系(QSAR)毒性预测启发式方法
cephalosporin antibioticsQSARtoxicityheuristic method(HM)
李梦龙,郭延芝,李益洲,王羽.头孢菌素类抗生素毒性的定量构效关系预测研究[EB/OL].(2012-12-11)[2025-08-16].http://www.paper.edu.cn/releasepaper/content/201212-205.点此复制
评论