Applying Kernel Methods on Protein Complexes Detection Problem
During the last years, various methodologies have made possible the detection of large parts of the protein interaction network of various organisms. However, these networks are containing highly noisy data, degrading the quality of information they carry. Various weighting schemes have been applied in order to eliminate noise from interaction data and help bioinformaticians to extract valuable information such as the detection of protein complexes. In this contribution, we propose the addition of an extra step on these weighting schemes by using kernel methods to better assess the reliability of each pairwise interaction. Our experimental results prove that kernel methods clearly help the elimination of noise by producing improved results on the protein complexes detection problem.
Keywordskernel methods protein-protein interactions protein interaction graphs protein complexes
Unable to display preview. Download preview PDF.
- 5.Brun, C., et al.: Functional classification of proteins for the prediction of cellular function from a protein-protein interaction network. Genome Biol. 5(1), R6 (2003)Google Scholar
- 12.Moschopoulos, C.N., et al.: An enchanced Markov clustering method for detecting protein complexes. In: 8th IEEE International Conference on BioInformatics and BioEngineering (BIBE 2008), Athens (2008)Google Scholar
- 15.Mewes, H.W., et al.: MIPS: analysis and annotation of proteins from whole genomes in 2005. Nucleic Acids Res. 34(Database issue), D169–D172 (2006)Google Scholar
- 16.Kandola, N., Cristianini, N., Shawe-Taylor, J.: Learning semantic similarity. In: Advances in Neural Information Processing Systems, pp. 657–664 (2002)Google Scholar
- 17.Kondor, R., Lafferty, J.: Diffusion kernels on graphs and other discrete structures. In: Proceedings of the Nineteenth International Conference on Machine Learning (2002)Google Scholar