Exploratory data analysis for investigating GC-MS biomarkers

Abstract

The detection of reliable biomarkers is a major research activity within the field of proteomics. A biomarker can be a single molecule or set of molecules that can be used to differentiate between normal and diseased states. This paper describes our methods to develop a reliable, automated method of detecting abnormal metabolite profiles from urinary organic acids. These metabolic profiles are used to detect Inborn Errors of Metabolism (IEM) in infants, which are inherited diseases resulting from alterations in genes that code for enzymes. The detection of abnormal metabolic profiles is usually accomplished through manual inspection of the chromatograms by medical experts. The chromatograms are derived by a method called Gas Chromatography - Mass Spectrometry (GC-MS). This combined technique is used to identify presence of different substances in a given sample. Using GC/MS analysis of the urine sample of the patient, the medical experts are able to identify the presence of metabolites which are a result of an IEM. © 2008 Springer Berlin Heidelberg.