Application of weighted gene co‑expression network analysis to explore the potential diagnostic biomarkers for colorectal cancer.

Abstract

Colorectal cancer (CRC) is one of the most common malignant diseases in the world. Although mechanistic studies have been conducted on the pathogenesis of CRC, the molecular mechanism of CRC tumorigenesis remains unclear. In the present study, the weighted gene co-expression network analysis was performed for the Gene Expression Omnibus (GEO) dataset GSE87211, in order to analyze the key modules involved in the pathogenesis of CRC. Next, Gene Ontology and Kyoto Encyclopedia of Genes and Genomes enrichment analyses were performed on the key module genes to analyze the functional pathways involved. The hub genes were screened using the Cytoscape platform and verified by a second GEO dataset, GSE21510. Finally, 10 hub genes were identified in 2 key modules (the green and brown modules) as the genes most significantly associated with the tumorigenesis of CRC. The 5 hub genes from the green module included collagen type I α1 chain, collagen type XII α1 chain, collagen triple helix repeat containing 1, inhibin subunit βa (INHBA) and chromobox 2 (CBX2), while the 5 hub genes from the brown module included bestrophin 2 (BEST2), carbonic anhydrase 2, glucagon, solute carrier family 4 member 4 and gliomedin. The 2 key modules with the 10 hub genes identified may regulate the occurrence and development of CRC through the extracellular matrix pathway, PI3K-Akt and chemokine signaling pathways, thus providing a reference for understanding the complex mechanism of tumorigenesis in CRC. Of note, few studies have reported the pathogenesis of CRC with the 3 identified hub genes, INHBA, CBX2 and BEST2. Further investigation of the molecular mechanism of these genes in CRC is recommended.