Abstract
BACKGROUND: Despite the high prevalence of lung cancer, with a five-year survival rate of only 23%, the underlying molecular mechanisms of non-small cell lung cancer (NSCLC) remain unknown. There is a great need to identify reliable candidate biomarker genes for early diagnosis and targeted therapeutic strategies to prevent cancer progression. METHODS: In this study, four datasets obtained from the Gene Expression Omnibus were evaluated for NSCLC- associated differentially expressed genes (DEGs) using bioinformatics analysis. About 10 common significant DEGs were shortlisted based on their p-value and FDR (DOCK4, ID2, SASH1, NPR1, GJA4, TBX2, CD24, HBEGF, GATA3, and DDR1). The expression of significant genes was validated using experimental data obtained from TCGA and the Human Protein Atlas database. The human proteomic data for post- translational modifications was used to interpret the mutations in these genes. RESULTS: Validation of DEGs revealed a significant difference in the expression of hub genes in normal and tumor tissues. Mutation analysis revealed 22.69%, 48.95%, and 47.21% sequence predicted disordered regions of DOCK4, GJA4, and HBEGF, respectively. The gene-gene and drug-gene network analysis revealed important interactions between genes and chemicals suggesting they could act as probable drug targets. The system-level network showed important interactions between these genes, and the drug interaction network showed that these genes are affected by several types of chemicals that could serve as potential drug targets. CONCLUSIONS: The study demonstrates the importance of systemic genetics in identifying potential drug- targeted therapies for NSCLC. The integrative system- level approach should contribute to a better understanding of disease etiology and may accelerate drug discovery for many cancer types.