A topology-sharing based method for protein function prediction via analysis of protein functional association networks

The sequencing of many genomes has brought to light the discovery of thousands of possible open reading frames which are potentially transcribed and translated into gene products, but a great majority of these have yet to be characterized. Since proteins seldom act alone; rather, they must interact with other biomolecular units to execute their functions. Thus, the function of unknown proteins may be discovered through studying their interaction with known proteins having known functions. Here, we developed a new network topology-based method in which the functions of the uncharacterized proteins can be predicted based on the functions of the proteins sharing similar topology structure. We explored to incorporate the inter-relationship among functional labels into the function prediction framework. We evaluated the performance of the new method with other representative network-based function prediction method using E. coli protein networks. Our results showed that the method has better prediction performance in E. coli protein function prediction.