Data mining to identify diagnostic and prognostic markers in lung adenocarcinoma

Abstract Background: Lung adenocarcinoma (LUAD), the most common type of lung cancers, is a leading cause of cancer-related death. Early diagnosis is essential to improve the outcome of LUAD, especially to reduce its mortality. This study aims to explore individual and regulatory network of mRNA, lncRNA and miRNA with diagnostic or prognostic value in LUAD. Methods: Differentially expressed RNAs of LUAD based on intersection of The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO) databases were analyzed by bioinformatic methods. Result: The competitive endogenous RNA (ceRNA) network was obtained with 40 differentially expressed (DE) mRNA nodes, 125 differentially expressed lncRNAs (DElncRNAs), and 7 differentially expressed miRNAs (DEmiRNAs), among which 37 were associated with prognosis and 11 were identified as diagnostic marker. The immune infiltration analysis suggested that components in the ceRNA may be related to changes in tumor immune microenvironment and development of LUAD. Conclusion: This comprehensive analysis of the regulatory RNA network in LUAD has identified a list of RNA markers with implications for clinical use, RNAs in our constructed regulatory network are potential prognostic and diagnostic markers in LUAD.