Multi-network approach to identify differentially methylated gene communities in cancer

Area of Science:

• Genomics
• Epigenetics
• Bioinformatics

Background:

• High-throughput sequencing generates vast cancer epigenome data.
• Identifying differentially methylated regions in gene networks is crucial for understanding cancer.
• Current methods integrating methylation and expression data have limitations.

Purpose of the Study:

• To develop a novel framework for analyzing cancer epigenome data.
• To improve the identification of epigenetic dysregulations in cancer gene networks.
• To address limitations in current gene network integration approaches.

Main Methods:

• A consensus-based clustering framework, Differentially Methylated Gene Communities based on Multi-network (DMGC-M), was proposed.
• The DMGC-M framework integrates evidence from multiple gene network types.
• It builds a community structure based on consensus clustering.

Main Results:

• Experiments were conducted on six cancer datasets from The Cancer Genome Atlas (TCGA).
• Multi-network approaches demonstrated more discriminative gene communities compared to integrated approaches.
• The proposed DMGC-M method shows superior performance in identifying gene communities.

Conclusions:

• The DMGC-M method is valuable for researchers studying epigenetic dysregulations in cancer pathways.
• Findings can advance research in Molecular Pathologic Epidemiology.
• The study highlights the benefit of multi-network analysis for cancer epigenome research.