Metabolomics

, Volume 4, Issue 1, pp 30–38

Plasma fatty acid metabolic profile coupled with uncorrelated linear discriminant analysis to diagnose and biomarker screening of type 2 diabetes and type 2 diabetic coronary heart diseases

  • Lun Zhao Yi
  • Da Lin Yuan
  • Zhi Hong Che
  • Yi Zeng Liang
  • Zhi Guang Zhou
  • Hai Yan Gao
  • Ya Min Wang
Original Article

DOI: 10.1007/s11306-007-0098-7

Cite this article as:
Yi, L.Z., Yuan, D.L., Che, Z.H. et al. Metabolomics (2008) 4: 30. doi:10.1007/s11306-007-0098-7

Abstract

Type 2 diabetes mellitus (T2DM) and type 2 diabetic coronary heart diseases (T2DM–CHD) are directly associated with metabolism disorder of lipid. In the present study, GC–MS followed by multivariate statistical analysis has been successfully applied to plasma free fatty acid metabolic profiling in T2DM and T2DM–CHD. Because principal component analysis and partial least squares-linear discriminant analysis both failed to the class separation among T2DM, T2DM–CHD, and control, uncorrelated linear discriminant analysis (ULDA) was proposed and successfully discriminated these three groups. The predictive correct rates were 81.03%, 85.37%, 88.89% for control and T2DM, control and T2DM–CHD, T2DM and T2DM–CHD, respectively. Furthermore, three potential biomarkers were screened. ULDA are much more efficient than PCA and PLS for discrimination analysis of complex data set. It is undoubtedly that such newly multivariate analysis method will promote and widen the application of metabonome analysis in disease clinical diagnosis.

Keywords

Metabolic profilingUncorrelated linear discriminant analysisBiomarkerType 2 diabetesDiabetic coronary heart disease

Abbreviations

FFA

Free fatty acid

GC–MS

Gas chromatography–mass spectrometry

FBG

Fasting blood glucose

PBG

2 h Postprandial blood glucose

T2DM

Type 2 diabetes mellitus

T2DM–CHD

Type 2 diabetic coronary heart diseases

PCA–LDA

Principal component analysis–linear discriminant analysis

PC

Principal component

PLS–LDA

Partial least squares–linear discriminant analysis

ULDA–LDA

Uncorrelated linear discriminant analysis–linear discriminant analysis

UDV

Uncorrelated discriminant vector

Copyright information

© Springer Science+Business Media, LLC 2007

Authors and Affiliations

  • Lun Zhao Yi
    • 1
  • Da Lin Yuan
    • 1
  • Zhi Hong Che
    • 2
  • Yi Zeng Liang
    • 1
  • Zhi Guang Zhou
    • 2
  • Hai Yan Gao
    • 1
  • Ya Min Wang
    • 1
  1. 1.Research Center of Modernization of Chinese MedicinesCentral South University ChangshaP.R. China
  2. 2.Diabetes Center, Institute of Metabolism and Endocrinology, Department of Endocrinology, The Second Xiangya HospitalCentral South UniversityChangshaP.R. China