Comprehensive characterization of respiratory genes based on a computational framework in pan-cancer to develop stratified treatment strategies.

Abnormal cellular respiration plays a critical role in carcinogenesis. However, the molecular mechanisms underlying dysregulation of respiratory gene expression across different cancer types remain unclear. Here, we developed a computational framework that provides an analytical approach for exploring the molecular alterations and clinical relevance of respiratory genes in pan-cancer. We identified a total of 53 gene signatures in the three stages of respiration (including glycolysis, tricarboxylic acid cycle, and oxidative phosphorylation) through this framework and found that they were broadly differentially expressed and genetically altered across 33 cancer types. Pathway analysis manifested that the expression levels of almost all respiratory gene signatures were remarkably associated with the activation or inhibition of numerous oncogenic pathways, such as metabolism, angiogenesis, cell proliferation, and apoptosis. Survival analysis highlighted the oncogenic or tumor suppressor potential of the respiratory gene signatures. In particular, VCAN has shown significant oncogenic features in multiple cancer types. Finally, we identified a number of respiratory gene signatures that could be potential therapeutic targets, including VCAN. We also predicted small-molecule compounds targeting respiratory gene signatures or components of pathways regulated by them. Overall, our comprehensive analysis has greatly enhanced the understanding of molecular alterations of respiratory genes in tumorigenesis and progression, and provided insights into developing new therapeutic strategies.