Single-Sample Networks Reveal Intra-Cytoband Co-Expression Hotspots in Breast Cancer Subtypes

Area of Science:

• Genomics
• Bioinformatics
• Cancer Biology

Background:

• Breast cancer is a heterogeneous disease with diverse subtypes, posing challenges for personalized medicine.
• Traditional gene co-expression analyses often miss individual-specific interactions crucial for targeted therapies.
• Understanding subtype-specific genomic alterations is key to improving breast cancer diagnosis, prognosis, and treatment.

Purpose of the Study:

• To investigate structural and functional genomic alterations in breast cancer subtypes using single-sample gene co-expression network analysis.
• To compare gene co-expression patterns between breast cancer subtypes and normal breast tissue.
• To identify subtype-specific genomic features and potential therapeutic targets.

Main Methods:

• Utilized RNA-Seq gene expression data to infer gene co-expression networks.
• Employed the LIONESS algorithm to construct individual patient gene co-expression networks.
• Analyzed the top 10,000 gene interactions and calculated network topological properties.

Main Results:

• Breast cancer subtypes exhibit fragmented co-expression networks characterized by a shift from interchromosomal (TRANS) to intrachromosomal (CIS) interactions.
• This transition signifies disrupted long-range genomic communication, leading to localized regulation and increased genomic instability.
• Single-sample analyses confirmed individual-level consistency of these genomic patterns, underscoring breast cancer's molecular heterogeneity.

Conclusions:

• Single-sample co-expression network analysis is a powerful tool for uncovering individual-specific genomic interactions in breast cancer.
• Identified subtype-specific high-degree genes and critical cytobands offer insights into regulatory networks and potential therapeutic targets.
• While genomic alterations are pronounced, the proportion of CIS interactions did not significantly correlate with survival, suggesting limited prognostic value.