Identification of hub genes, modules and biological pathways associated with lung adenocarcinoma: A system biology approach

Abstract

Lung adenocarcinoma (LUAD) is the most common non-small cell lung cancer with an alarming trend within non-smokers at younger ages. This study is designed to identify mechanisms of LUAD compared to normal lung tissues using systems biology approaches. Four GSE datasets (GSE75037, GSE63459, GSE32863 and GSE10072) were selected from the Gene Expression Omnibus database. Data processing and meta-analysis were performed using the R statistical programming language. The common and unique DEGs between stages of LUAD and normal lung tissues were initiated by the Venny tool. Common genes were then analyzed in a STRING database to obtain protein-protein interaction. STRING output was analyzed by MCODE and CytoHubba applications of Cytoscape to identify modules of co-expression and hub genes, respectively. The correlation between hub genes and survival of LUAD patients was explored using UALCAN. The hub genes from the meta-analysis were confirmed using an independent, comprehensive LUAD RNA-sequencing dataset collected from The Cancer Genome Atlas (TCGA). The shared upregulated and downregulated DEGs among LUAD stages were 22 and 140 genes, respectively, when compared to normal lung tissues. Unique genes for each stage were also identified. The hub genes were PECAM1, TEK, CDH5, VWF and ANGPT1. An inverse correlation was found between the expression levels of ANGPT1 gene and overall survival of LUAD patients (p-value = 0.02). Most of the top cluster genes were enriched for Gα(s) signalling events, GPCR ligand binding, class B/2 (Secretin family receptors), platelet activation, signalling and aggregation in the main three co-expression clusters. Most of the shared genes were enriched in the biological pathway of hemostasis. TCGA analysis showed that hub genes derived from common genes were found statistically significant throughout all stages. In conclusion, unique upregulated and downregulated DEGs at each stage were identified with FERMT1 and SIX1 as specific early-stage diagnostic biomarkers for stage IB and IIB. 5 hub genes were observed, including PECAM1, TEK, CDH5, vWF and ANGPT1 which might be crucial for the onset and progression of LUAD. Interestingly, ANGPT1 can be considered as a potential and valuable prognostic marker in LUAD patients.