An Efficient Algorithm for Identifying Mutated Subnetworks Associated with Survival in Cancer.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Protein-protein interaction (PPI) networks model gene interconnections, with functionally related proteins forming pathways or modules.
• Mutations in genes can disrupt these pathways, influencing disease initiation, progression, and severity.
• Integrating mutation data, patient survival information, and PPI networks is crucial for understanding disease mechanisms.

Purpose of the Study:

• To develop a computational method for identifying functionally connected subnetworks within PPI networks that are significantly associated with patient survival.
• To define a scoring system using the log-rank statistic to evaluate the survival impact of mutated subnetworks.
• To propose and validate the Survival Associated Mutated Subnetwork (SAMS) algorithm for efficient subnetwork identification.

Main Methods:

• Utilized a fitness function based on the log-rank statistic to score protein-protein interaction subnetworks.
• Employed a genetic algorithm strategy within the SAMS method to search for optimal subnetworks.
• Validated the SAMS method on both real cancer datasets and synthetic data.

Main Results:

• The SAMS method identified significantly mutated subnetworks associated with patient survival.
• SAMS demonstrated remarkable efficiency, generating solutions in negligible time compared to exponential time for state-of-the-art methods.
• The identified gene sets showed substantial overlap with known cancer driver genes and relevant biological pathways.

Conclusions:

• SAMS is an efficient and effective tool for identifying survival-associated mutated subnetworks in protein-protein interaction networks.
• The method provides biologically relevant insights into cancer driver genes and pathways.
• This approach offers a valuable strategy for cancer research and personalized medicine by linking genetic mutations to patient outcomes.