E-alert基于1H-NMR的模式识别方法应用于异常黑胆质糖尿病患者的尿液代谢组学研究
1H-NMR Metabonomic Analysis Based on Different Pattern Recognition for Urine Sample of DM Patients with Syndrome of Abnormal Savda
-
摘要: 探讨核磁共振氢谱结合模式识别方法应用于异常黑胆质糖尿病患者的尿液代谢组研究可行性。对32 例异常黑胆质糖尿病患者和29 例健康人尿液进行核磁共振氢谱检测，采用主成分分析(principal component analysis, PCA)、偏最小二乘法判别分析(partial least squares dis-criminant analysis, PLS-DA)、正交偏最小二乘法判别分析(orthogonal to partial least squares discriminant analysis,OPLS-DA)进行模式识别分析，比较3种模式识别方法的判别能力。运用3种模式识别均可以对2组数据进行有效的区分,但OPLS-DA较PCA、P1LS-DA更加有效，不仅提高了模式识别方法的判断能力，可以清楚的判断两组中有差异的代谢物。基于核磁共振氢谱结合模式识别分析方法可以为异常黑胆质糖尿病代谢标志物的寻找提供理论依据。OPLS-DA的模式识别方法较其它2种方法更具优势，在揭示维医理论本质上有着广阔的应用前景。Abstract: To investigate the possibility of using 1H-NMR based on different pattern recognition for the urine samples of DM patients with abnormal savda syndrome. 1H-NMR technique was applied to the examination of the urine samples of 32 abnormal savda syndrome patients with DM and 29 healthy volunteers. Different pattern recognitions of principal component analysis (PCA),partial least squares discriminant analysis (PLS-DA) and orthogonal to partial least squares discriminantanalysis (OPLS-DA) were used to distinguish the metabolic phenotypes. All different pattern recognitions can distinguish the metabolism products in urine of abnormal savda syndrome patients with DM and healthy people. OPLS-DA is more efficienct than PCA and PLS-DA in the metabonomic analysis. OPLS-DA method has more advantages over the other methods, on providing more evidences for probing the essences of DM with syndrome of abnormal savda and on the clinical diagnosis of this disease.
下载: