It's the machine that matters: Predicting gene function and phenotype from protein networks

Area of Science:

• Computational biology
• Systems biology
• Bioinformatics

Background:

• Knowledge of protein organization into complexes, systems, and pathways is expanding.
• Model organisms are crucial for developing and testing theoretical approaches in biological research.

Purpose of the Study:

• To review computational methods for integrating proteomics and genomics data into protein networks.
• To discuss the application of guilt-by-association and network modularity for identifying genes related to traits and diseases.

Main Methods:

• Integrating proteomics and genomics observations into protein networks.
• Applying guilt-by-association methods within these networks.
• Utilizing network modularity and algorithms like PageRank for gene identification.

Main Results:

• Protein network analysis successfully identifies genes underlying phenotypic traits.
• Network modularity is increasingly recognized as important for understanding traits and phenotypes.
• Predictions from model organisms have been experimentally validated and applied to human diseases.

Conclusions:

• Computational integration of biological data into networks is a powerful approach for functional genomics.
• Network-based methods, including PageRank, are effective for identifying disease-related genes.
• Successful translation of these methods from model organisms to human disease analysis, such as cancer prognosis, is demonstrated.