Comprehensive network analysis of different subtypes of molecular disorders in lung cancer

Lung cancer is the leading cause of cancer-related death worldwide. In this study, we attempted to identify the common pathogenesis of lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) based on a modular and comprehensive analysis method. Data were downloaded and the differences analyzed in LUAD samples, LUSC samples, and normal samples, respectively. Co-expression analysis, enrichment analysis, and hypergeometric testing were used to predict transcription factors (TFs) and ncRNAs, as well as target genes. We obtained 4,596 differentially expressed genes which were clustered into 14 modules dysfunction. The 14 clustered genes (including DOK2, COL5A1, and TSPAN8) were identified as the core genes of the module. Module genes are substantially involved in biological processes, such as extracellular matrix, carbohydrate binding and renal system development, and signal transduction as well, including PPAR signal transduction, cGMP-PKG signal transduction, PI3K-Akt signal transduction, and Apelin signal transduction. We identified ncRNA (miR-335-5p, ANCR, TUG1) and transcription factors (RELA, SP1) to regulate dysfunction module genes essentially. The analysis showed that comprehensive co-expression analysis contributes to understanding the TF ncRNA. Moreover, it assisted in further understanding of the molecular pathogenesis of co-expression of modular genes that regulate LUAD and LUSC. It provided a precious resource and theoretical basis for further experiments.