DiME: a scalable disease module identification algorithm with application to glioma progression.

Area of Science:

• Network medicine
• Computational biology
• Systems biology

Background:

• Disease modules, groups of interacting molecules, are crucial for understanding complex human diseases.
• Identifying these modules is a key challenge in network medicine for elucidating disease mechanisms.
• Network analysis offers insights into pathogenesis and disease progression.

Purpose of the Study:

• To propose a novel algorithm, DiME (Disease Module Extraction), for identifying disease modules in biological networks.
• To adapt and optimize community extraction heuristics for biological network analysis.
• To incorporate a statistical measure (B-score) for evaluating the quality of identified modules.

Main Methods:

• Developed DiME algorithm using optimized Community Extraction heuristics.
• Applied DiME to construct and analyze low- and high-grade glioma co-expression networks.
• Utilized B-score for statistical validation of extracted disease modules.

Main Results:

• Identified distinct topological and expression changes in disease modules between low- and high-grade glioma.
• Revealed significant shifts in molecular constitution and cellular physiology during glioma progression.
• Highlighted transcription factors E2F4, AR, and ETS1 as potential key regulators in glioma progression.

Conclusions:

• DiME effectively identifies disease modules and provides insights into complex disease mechanisms.
• The study elucidates molecular changes underlying glioma progression.
• Identified transcription factors offer potential therapeutic targets for glioma treatment.