Algorithms and tools for the alignment of multiple protein networks

Comparative analysis of protein networks has proven to be a powerful approach for elucidating network structure and predicting protein function and interaction. A fundamental challenge for the success of this approach is the development of efficient algorithms and tools for comparing multiple networks. In this thesis we present two novel tools for comparative network analysis. The first tool is a framework for the alignment of multiple networks. At the heart of the framework is a novel representation of multiple networks that is only linear in their size as opposed to current exponential representations. The alignment algorithm is very efficient, being capable of aligning 10 networks with tens of thousands of proteins each in minutes. We show that our algorithm outperforms previous strategies for the problem, and produces results that are more in line with current biological knowledge. A paper describing this framework appeared in the Proceedings of RECOMB 2008. The second tool is a web-server implementing the NetworkBLAST algorithm for identifying protein complexes in protein-protein interaction networks. The server can analyze a single network or two networks from different species. In the latter case, NetworkBLAST outputs a set of putative complexes that are evolutionarily conserved across the two networks. The server accepts input data in various formats and provides a graphical visualization of the identified complexes. The application report appeared in Bioinformatics.