Comparing Alzheimer's and Parkinson's diseases networks using graph communities structure

Area of Science:

• Computational Biology
• Systems Biology
• Neuroscience

Background:

• Modern biology benefits from large dataset analysis for system-level pathology investigation.
• Alzheimer's Disease (AD) and Parkinson's Disease (PD) involve complex molecular alterations.
• Reductionist approaches offer limited insight into multifactorial neurological disorders.

Purpose of the Study:

• To develop a novel computational method for analyzing
• omics
• data in AD and PD.
• To compare neurological disorders using a network perspective to identify common and distinct pathways.
• To leverage network analysis for a deeper understanding of disease etiopathogenesis.

Main Methods:

• Utilized a network community discovery algorithm (InfoMap) based on information theory.
• Quantified functional and topological similarities between AD and PD using two similarity measurements.
• Constructed a Similarity Matrix to visualize common communities and analyzed statistically significant Gene Ontology (GO) terms.

Main Results:

• Identified shared biological processes in AD and PD, including DNA repair, RNA metabolism, and glucose metabolism, some not found by standard GO enrichment.
• Captured the link between mitochondrial dysfunction and metabolism (glucose, glutamate/glutamine).
• Highlighted proteins or pathways common but differently represented between the two pathologies.

Conclusions:

• The network approach effectively identifies shared biological processes between pathologies.
• Distinct communities reveal processes unique to individual diseases, aiding in targeted investigation.
• This strategy is broadly applicable for comparing any biological networks.