Introduction: Lung cancer is the leading cause of cancer deaths in the world and is usually divided into non-small cell lung cancer (NSCLC) and small cell lung cancer. NSCLC is dominant and accounts for 85% of the total cases. Currently, the therapeutic method of NSCLC is not so satisfactory, and thus identification of new biomarkers is critical for new clinical therapy for this disease.
Methods: Datasets of miRNA and gene expression were obtained from the NCBI database. The differentially expressed genes (DEGs) and miRNAs (DEMs) were analyzed by GEO2R tools. The DEG-DEM interaction was built via miRNA-targeted genes by miRWalk. Several hub genes were selected via network topological analysis in Cytoscape.
Results: A set of 276 genes were found to be significantly differentially expressed in the three datasets. Functional enrichment by the DAVID tool showed that these 276 DEGs were significantly enriched in the term "cancer", with a statistic p-value of 1.9 × 10-5. The subdivision analysis of the specific cancer types indicated that "lung cancer" occupies the largest category with a p-value of 2 × 10-3. Furthermore, 75 miRNAs were shown to be differentially expressed in three representative datasets. A group of 13 DEGs was selected by analysis of the miRNA-gene interaction of these DEGs and DEMs. The investigation of these 13 genes by GEPIA tools showed that eight of them had consistent results with NSCLC samples in the TCGA database. In addition, we applied the KMplot to conduct the survival analysis of these eight genes and found that seven of them have a significant effect on the prognosis survival of patients. We believe that this study could provide effective research clues for the prevention and treatment of non-small cell lung cancer.
Keywords: biomarker; microRNA–gene interaction network; non-small cell lung cancer.