Gene correlation network analysis to identify regulatory factors in idiopathic pulmonary fibrosis. Issue 2 (26th October 2018)

• Record Type: Journal Article
• Authors: McDonough, John E; Kaminski, Naftali; Thienpont, Bernard; Hogg, James C; Vanaudenaerde, Bart M; Wuyts, Wim A
• Is Part Of: Thorax. Volume 74:Issue 2(2019)
• Page Start: 132
• Page End: 140
• Publication Date: 2018-10-26
• Subjects: idiopathic pulmonary fibrosis; Chest -- Diseases -- Periodicals; Thorax; Chest -- Diseases; Periodicals; Periodicals; 617.54
• DOI: 10.1136/thoraxjnl-2018-211929
• Languages: English
• ISSNs: 0040-6376

Abstract

Abstract : Background: Idiopathic pulmonary fibrosis (IPF) is a severe lung disease characterised by extensive pathological changes. The objective for this study was to identify the gene network and regulators underlying disease pathology in IPF and its association with lung function. Methods: Lung Tissue Research Consortium dataset with 262 IPF and control subjects (GSE47460) was randomly divided into two non-overlapping groups for cross-validated differential gene expression analysis. Consensus weighted gene coexpression network analysis identified overlapping coexpressed gene modules between both IPF groups. Modules were correlated with lung function (diffusion capacity, DLCO ; forced expiratory volume in 1 s, FEV1 ; forced vital capacity, FVC) and enrichment analyses used to identify biological function and transcription factors. Module correlation with miRNA data (GSE72967) identified associated regulators. Clinical relevance in IPF was assessed in a peripheral blood gene expression dataset (GSE93606) to identify modules related to survival. Results: Correlation network analysis identified 16 modules in IPF. Upregulated modules were associated with cilia, DNA replication and repair, contractile fibres, B-cell and unfolded protein response, and extracellular matrix. Downregulated modules were associated with blood vessels, T-cell and interferon responses, leucocyte activation and degranulation, surfactant metabolism, and cellular metabolic and catabolic processes. Lung function correlated with nine modules (eight with DLCO, five with FVC). Intermodular network of transcription factors and miRNA showed clustering of fibrosis, immune response and contractile modules. The cilia-associated module was able to predict survival (p=0.0097) in an independent peripheral blood IPF cohort. Conclusions: We identified a correlation gene expression network with associated regulators in IPF that provides novel insight into the pathological process of this disease.