A multiobjective approach for identifying protein complexes and studying their association in multiple disorders

Area of Science:

• Bioinformatics
• Systems Biology
• Computational Biology

Background:

• Protein-protein interaction (PPI) networks are crucial for understanding biological processes.
• Identifying protein complexes within these networks remains a significant challenge.
• High-throughput experimental techniques generate extensive PPI data for network construction.

Purpose of the Study:

• To develop a novel multiobjective framework for detecting human protein complexes.
• To integrate topological and biological properties for improved complex identification.
• To explore the association of identified protein complexes with human diseases.

Main Methods:

• A multiobjective optimization framework was developed for human PPI networks.
• Network partitioning into modules identified potential protein complexes.
• Objective functions utilized PPI network topology and Gene Ontology semantic similarity.
• Biological validation employed Gene Ontology and pathway-based analyses.
• Association with disease classes was visualized using bipartite networks.

Main Results:

• The proposed method effectively detects human protein complexes.
• Identified complexes demonstrate significant associations with disease classes including 'Cancer' and 'Endocrine'.
• Biological validation confirmed the functional relevance of the predicted complexes.

Conclusions:

• Protein complex identification framed as a multiobjective optimization problem.
• Discovered associations between protein complexes and disorders suggest potential for multi-target drug development.
• This approach offers new insights into disease mechanisms and therapeutic strategies.