Discovery of metabolite features for the modelling and analysis of high-resolution NMR spectra

• Type of Material: Article
• Authors: Hyun-Woo Cho, University of Tennessee; Seoung Bum Kim, University of Texas at Arlington; Myong K. Jeong, University of Tennessee; Youngja Park, Emory University; Nana Gletsu-Miller, Emory University; Thomas R Ziegler, Emory University; Dean P Jones, Emory University
• Language: English
• Date: 2008
• Publisher: Inderscience
• Publication Version: Author Accepted Manuscript (After Peer Review)
• Copyright Statement: © 2008 Inderscience Enterprises Ltd.
• Final Published Version (URL): http://www.inderscience.com/dev/search/index.php?mainAction=search&action=record&rec_id=19097&prevQuery=&ps=10&m=or
• Title of Journal or Parent Work: International Journal of Data Mining and Bioinformatics
• ISSN: 1748-5673
• Volume: 2
• Issue: 2
• Start Page: 176
• End Page: 192
• Grant/Funding Information: This study was supported by grants from the National Institutes of Health: R03 DK066008, R03 ES012929, R01 ES011195, R01 DK55850, and the Emory General Clinical Research Center grant M01 RR00039.
• Keywords: NMR; multivariate statistical analysis; metabolomics; OSC; Orthogonal Signal Correction; Nuclear Magnetic Resonance; feature selection
• Research Categories: Health Sciences, Medicine and Surgery; Engineering, Biomedical

Abstract

High-resolution Nuclear Magnetic Resonance (NMR) spectroscopy in combination with multivariate statistical methods has been widely used to investigate metabolic fluctuations in biological systems. This study presents three feature selection methods for identifying the metabolite features that contribute to the distinction of spectral samples among varying nutritional conditions in human plasma. Loading vectors of Principal Component Analysis (PCA), the optimal discriminant direction of Fisher discriminant analysis, and index values of the Variable Importance in Projection (VIP) in a Partial Least Square Discriminant Analysis (PLS-DA) were used to calculate the importance of individual metabolite feature in spectra. In addition, an Orthogonal Signal Correction (OSC) filter was used to eliminate unnecessary variations in NMR spectra and its effectiveness was demonstrated through PCA and kernel PCA. For the evaluation of presented feature selection methods, we compared the ability of classification based on the metabolite features selected by each method. The results have shown that the best classification was achieved using VIP values from an OSC-PLS-DA model.