A Querying Method with Feedback Mechanism for Protein Interaction Network

Comparative analysis of protein interaction networks plays a key role in the study of evolution and molecular etiology of diseases. A topology-based method is proposed in this paper to search biologically significant sub networks in protein interaction networks with relatively high accuracy. Different with other methods, the present method takes into account incorrectness and incompleteness of biological experiment data. It reduces influence made by these errors and introduces feedback mechanism to obtain more accurate results in the sense of biology. The effectiveness of the present method is verified by studying a Parkinson Disease related network. The results indicate that the present method is more efficient in terms of computational accuracy compared with other methods for network querying.