Clustering gene expression regulators: new approach to disease subtyping

Area of Science:

• Computational Biology
• Genomics
• Systems Biology

Background:

• Personalized medicine requires patient stratification based on molecular mechanisms.
• Current methods may not fully capture pathway-level biological differences.

Purpose of the Study:

• To develop a novel method for disease subtyping using activated expression regulators.
• To enable personalized therapy by understanding patient-specific molecular mechanisms.

Main Methods:

• Utilized the Sub-Network Enrichment Analysis (SNEA) algorithm.
• Analyzed gene subnetworks, identified central regulators, and their downstream genes.
• Clustered regulators per patient and assigned activity scores for grouping.

Main Results:

• Demonstrated superior performance compared to existing methods.
• Successfully stratified neuromuscular disorders based on underlying causes.
• Differentiated colorectal carcinoma from adenoma by identifying key regulator clusters.

Conclusions:

• The SNEA approach offers a biologically meaningful feature selection method.
• Identifies significant expression regulators, enhancing pathway-level biological understanding.
• Facilitates hypothesis generation for molecular mechanisms driving clinical outcomes.