检索规则说明:AND代表“并且”;OR代表“或者”;NOT代表“不包含”;(注意必须大写,运算符两边需空一格)
检 索 范 例 :范例一: (K=图书馆学 OR K=情报学) AND A=范并思 范例二:J=计算机应用与软件 AND (U=C++ OR U=Basic) NOT M=Visual
作 者:林兆洲[1] 史新元[1,2] 郭明星[1,3] 高晓燕[3] 乔延江[1,2]
机构地区:[1]北京中医药大学中药学院,北京100102 [2]国家中医药管理局中药信息工程重点研究室,北京100102 [3]北京中医药大学科研试验中心,北京100029
出 处:《中华中医药杂志》2014年第5期1328-1333,共6页China Journal of Traditional Chinese Medicine and Pharmacy
基 金:国家"重大新药创制"科技重大专项(No.2010ZX09502-002);国家自然科学基金面上项目(No.81373958);北京中医药大学自主课题(No.2013-JYBZZ-XS-112)~~
摘 要:目的:建立符合中药非线性作用特征的生物标记物辨识方法,初步探讨其与线性辨识方法的差异。方法:以清开灵注射液对酵母菌致热大鼠代谢物的影响为研究载体,基于支持向量机(SVM)建立非线性分类模型,利用非线性双标轨迹图对生物标记物的重要性进行可视化的分析,并与基于偏最小二乘判别(PLS-DA)算法辨识的生物标记物进行对比。结果:两种方法识别的生物标记物有明显差异,但所建立分类模型的预测性能无明显区别。结论:非线性生物标记物辨识方法揭示了与线性方法完全不同的代谢特征,模型交叉验证误差不能有效的区分辨识方法性能的优劣。Objective: To develop a nonlinear biomarker selection method and investigate its effectiveness compared with linear approach. Methods: An ultra performance liquid chromatography quadrupole time-of-flight mass spectrometry (UPLC Q-TOF/MS) metabonomics method was developed to explore the biochemical substances changes in rats of yeast-induced pyrexia treated with and without Qingkailing Injection. Support vector machine (SVM) was used to learn the data structure of UPLC Q-TOF/ MS to distinguish the pyrexia model group and the pyexia model group treated by Qingkailing Injection. The potential biomarkers related to pyrexia were visualized using nonlinear biplot. Partial least square discriminate analysis (PLS-DA) algorithm was also used to explore the metabolic differences between the groups of rats treated with and without Qingkailing Injection. The leave one out cross validation (LOO-CV) errors were recorded for both SVM and PLS-DA algorithm. Results: It was found that the LOO- CV errors of both methods reached as lower as zero and the differences between the potential biomarkers identified by SVM and PLS-DA was apparent. Conclusion: The potential biomarkers selected by SVM depict metabolic character are different from that described by PLS-DA. Moreover, LOO-CV error can not distinguish the two algorithms effectively.
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在链接到云南高校图书馆文献保障联盟下载...
云南高校图书馆联盟文献共享服务平台 版权所有©
您的IP:18.218.61.200