A protocol for extracting chlorogenic acid from Eucommia ulmoides leaf was optimized by orthogonal test on the base of noticeable factors test.The preventing effect of extract of Eucommia ulmoides leaf on mice suffere...A protocol for extracting chlorogenic acid from Eucommia ulmoides leaf was optimized by orthogonal test on the base of noticeable factors test.The preventing effect of extract of Eucommia ulmoides leaf on mice suffered from high lipid serum was evaluated.High lipid serum was induced by gavaging fat emulsion to mice everyday.The TC,TG,LDL-C and HDL-C level in the serum of different animal groups were monitored.The result showed that the optimum process was to extract the chlorogenicacid at 90 ℃ with 70% ethanol for 2 h,with solid-liquid ratio 1∶16,and the extract rate was able to reach to 4%.The extract was efficient in defending high lipid serum.展开更多
分别采用支持向量学习机、人工神经网络、调节性逻辑回归和K-最临近等机器学习方法对761个二氢叶酸还原酶抑制剂建立了其活性分类预测模型.采用组成描述符和拓扑描述符表征抑制剂的分子结构及物理化学性质,使用Kennard-Stone方法进行训...分别采用支持向量学习机、人工神经网络、调节性逻辑回归和K-最临近等机器学习方法对761个二氢叶酸还原酶抑制剂建立了其活性分类预测模型.采用组成描述符和拓扑描述符表征抑制剂的分子结构及物理化学性质,使用Kennard-Stone方法进行训练集的设计,并用Metropolis Monte Carlo模拟退火方法作变量选择.结果表明,支持向量学习机优于其它机器学习方法,所得到的最优模型具有较好的预测结果,其预测正确率为91.62%.说明通过合适的训练集设计及变量选择,支持向量学习机方法可以很好地用于二氢叶酸还原酶抑制剂的活性分类预测.展开更多
文摘A protocol for extracting chlorogenic acid from Eucommia ulmoides leaf was optimized by orthogonal test on the base of noticeable factors test.The preventing effect of extract of Eucommia ulmoides leaf on mice suffered from high lipid serum was evaluated.High lipid serum was induced by gavaging fat emulsion to mice everyday.The TC,TG,LDL-C and HDL-C level in the serum of different animal groups were monitored.The result showed that the optimum process was to extract the chlorogenicacid at 90 ℃ with 70% ethanol for 2 h,with solid-liquid ratio 1∶16,and the extract rate was able to reach to 4%.The extract was efficient in defending high lipid serum.
文摘分别采用支持向量学习机、人工神经网络、调节性逻辑回归和K-最临近等机器学习方法对761个二氢叶酸还原酶抑制剂建立了其活性分类预测模型.采用组成描述符和拓扑描述符表征抑制剂的分子结构及物理化学性质,使用Kennard-Stone方法进行训练集的设计,并用Metropolis Monte Carlo模拟退火方法作变量选择.结果表明,支持向量学习机优于其它机器学习方法,所得到的最优模型具有较好的预测结果,其预测正确率为91.62%.说明通过合适的训练集设计及变量选择,支持向量学习机方法可以很好地用于二氢叶酸还原酶抑制剂的活性分类预测.
