Multiple differential expression networks identify key genes in rectal cancer

Area of Science:

• Genomics
• Bioinformatics
• Cancer Research

Background:

• Rectal cancer significantly contributes to global cancer mortality.
• Understanding molecular differences between rectal cancer phenotypes is crucial for targeted therapies.

Purpose of the Study:

• To identify key genes associated with distinct rectal cancer phenotypes (fungating, polypoid, polypoid & small-ulcer).
• To leverage differential expression networks (DENs) for gene discovery.
• To evaluate the potential of identified genes as biomarkers.

Main Methods:

• Construction of DENs using Spearman correlation coefficient (SCC) for differential and non-differential interactions.
• Topological analysis to identify hub genes within the largest network components.
• Intersection of hub genes with known rectal cancer genes from Genecards.
• Classification of rectal cancer phenotypes using Support Vector Machines (SVM) based on identified hub genes.

Main Results:

• 19 hub genes and 12 common key genes were identified across the three largest DEN components.
• Epidermal Growth Factor Receptor (EGFR) was a common key gene.
• Support Vector Machine (SVM) analysis confirmed that hub genes can effectively classify rectal cancer phenotypes, validating the DEN methodology.

Conclusions:

• Significant genes, including EGFR and UBC (Ubiquitin C), were identified across major rectal cancer phenotypes.
• These identified genes hold promise as potential biomarkers for rectal cancer classification, early detection, and therapeutic strategies.