Abstract
Metabolomics is concerned with characterizing the large number of metabolites present in a biological system using nuclear magnetic resonance (NMR) and HPLC/MS (high-performance liquid chromatography with mass spectrometry). Multivariate analysis is one of the most important tools for metabolic biomarker identification in metabolomic studies. However, analyzing the large-scale data sets acquired during metabolic fingerprinting is a major challenge. As a posterior probability that the features of interest are not affected, the local false discovery rate (LFDR) is a good interpretable measure. However, it is rarely used to when interrogating metabolic data to identify biomarkers. In this study, we employed the LFDR method to analyze HPLC/MS data acquired from a metabolomic study of metabolic changes in rat urine during hepatotoxicity induced by Genkwa flos (GF) treatment. The LFDR approach was successfully used to identify important rat urine metabolites altered by GF-stimulated hepatotoxicity. Compared with principle component analysis (PCA), LFDR is an interpretable measure and discovers more important metabolites in an HPLC/MS-based metabolomic study.
Highlights
As a newly emerging field of the ‘omics’ domain based on the exhaustive profiling of metabolites, metabolomics has been widely employed to monitor global metabolic changes taking place in biological systems
Considering the metabolic biomarker identification problem from the perspective of metabolic fingerprinting using HPLC/MS technology, we usually study m/z values at different retention times simultaneously
We presented the local false discovery rate (LFDR) computational method based on Bayesian theory
Summary
As a newly emerging field of the ‘omics’ domain based on the exhaustive profiling of metabolites, metabolomics has been widely employed to monitor global metabolic changes taking place in biological systems. When HPLC/MS technology is used for metabolic fingerprinting [9,10], the unique mass-charge (m/z) value and retention time of compounds are used to construct a metabolic fingerprint that will undergo statistical analysis. This procedure includes biomarker identification by multivariate analysis of metabolic data sets [11]. The major challenges confronting researchers are the analysis of large-scale data sets produced from metabolic fingerprinting and the selection of appropriate multivariate methods to find biomarkers effectively and precisely
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
