Systematic expression profiling analysis mines dys-regulated modules in active tuberculosis based on re-weighted protein-protein interaction network and attract algorithm. (June 2017)

• Record Type: Journal Article
• Authors: Sun, Ying; Weng, Yan; Zhang, Ying; Yan, Xiang; Guo, Lei; Wang, Jia; Song, Xin; Yuan, Ying; Chang, Fu-Ye; Wang, Chun-Ling
• Is Part Of: Microbial pathogenesis. Volume 107(2017)
• Page Start: 48
• Page End: 53
• Publication Date: 2017-06
• Subjects: Active tuberculosis -- Protein-protein interaction network -- Attractors -- Dys-regulated modules; Pathogenic microorganisms -- Periodicals; Pathology, Molecular -- Periodicals; Communicable Diseases -- microbiology -- Periodicals; Communicable Diseases -- parasitology -- Periodicals; Micro-organismes pathogènes -- Périodiques; Pathologie moléculaire -- Périodiques; Electronic journals; 616.9041
• DOI: 10.1016/j.micpath.2017.03.013
• Languages: English
• ISSNs: 0882-4010

Abstract

Abstract: About 90% of tuberculosis (TB) patients latently infected with Mycobacterium tuberculosis (Mtb) show no symptoms, yet have a 10% chance in lifetime to progress active TB. Nevertheless, current diagnosis approaches need improvement in efficiency and sensitivity. The objective of this work was to detect potential signatures for active TB to further improve the understanding of the biological roles of functional modules involved in this disease. First, targeted networks of active TB and control groups were established via re-weighting protein-protein interaction (PPI) networks using Pearson's correlation coefficient (PCC). Candidate modules were detected from the targeted networks, and the modules with Jaccard score >0.7 were defined as attractors. After that, identification of dys-regulated modules was conducted from the attractors using attract method, Subsequently, gene oncology (GO) enrichment analyses were implemented for genes in the dys-regulated modules. We obtained 33 and 65 candidate modules from the targeted networks of control and active TB groups, respectively. Overall, 13 attractors were identified. Using the cut-off criteria of false discovery rate <0.05, there were 4 dys-regulated modules (Module 1, 2, 3, and 4). Based on the GO annotation results, genes in Modules 1, 2 and 4 were only involved in translation. Most genes in Module 1, 2 and 4 were associated with ribosomes. Accordingly, these dys-regulated modules might serve as potential biomarkers of active TB, facilitating the development for a more efficient, and sensitive diagnostic assay for active TB. Highlights: Totally, 33 and 65 candidate modules were identified in control and active TB. Using FDR <0.05, there were 4 dys-regulated modules (Module 1, 2, 3, and 4). Genes in Modules 1, 2 and 4 were only involved in translation. Most genes in Module 1, 2 and 4 were associated with ribosomes.