Deciphering Gene Expression Patterns to Differentiate Among Leading Cancer Types

Abstract

While individual cancers have been extensively researched in terms of prognostic genes, comprehensive studies comparing these across different cancer types remain scarce. Proper cancer classification into subtypes is pivotal for accurate diagnosis and effective treatment strategies. This study delves into gene co-expression networks across five cancer types using patient-to-patient correlation network analysis and Weighted Gene Correlation Network Analysis (WGCNA), utilizing data from UC Irvine. We conduct a thorough comparison of network characteristics such as degree, centrality, and betweenness for each cancer type. Additionally, we employ multinomial logistic regression to pinpoint a crucial subset of genes. Our research provides insights into the unique and overlapping gene expression patterns among various cancer types.