Metabolites can serve as biomarkers and their identification has significant importance in the study of biochemical reaction and signalling networks. Incorporating metabolic and gene expression data to reveal biochemical networks is a considerable challenge, which attracts a lot of attention in recent research. In this paper, we propose a promising approach to identify metabolic biomarkers through integrating available biomedical data and disease-specific gene expression data. A Linear Programming (LP) based method is then utilized to determine flux variability intervals, therefore enabling the analysis of significant metabolic reactions. A statistical approach is also presented to uncover these metabolites. The identified metabolites are then verified by comparing with the results in the existing literature. The proposed approach here can also be applied to the discovery of potential novel biomarkers.
展开▼
