Protein Solvent Accessibility Prediction Using Support Vector Machines and Sequence Conservations
A two-stage method is developed for the single sequence prediction of protein solvent accessibility from solely its amino acid sequence. The first stage classifies each residue in a protein sequence as exposed or buried using support vector machine (SVM). The features used in the SVM are physico-chemical properties of the amino acid to be predicted as well as the information coming from its neighboring residues. The SVM-based predictions are refined using pairwise conservative patterns, called maximal unique matches (MUMs). The MUMs are identified by an efficient data structure called suffix tree. The baseline predictions, SVM-based predictions and MUM-based refinements are tested on a nonredundant protein data set and 7̃3% prediction accuracy is achieved for a solvent accessibility threshold that provides an evenly distribution between buried and exposed classes. The results demonstrate that the new method achieves slightly better accuracy than recent methods using single sequence prediction.
KeywordsSupport Vector Machine Solvent Accessibility Suffix Tree Efficient Data Structure Remote Homology Detection
Unable to display preview. Download preview PDF.
- 3.Chen, H., Zhou, H., Hu, X., Yoo, I.: Classification comparison of prediction of solvent accessibility from protein sequences. In: 2nd Asia-Pacific Bioinformatics Conference, Dunedin, New Zelland (2004)Google Scholar
- 6.Horton, H.B., Moran, L.A., Ochs, R.S., Rawn, J.D., Scrimgeour, K.G.: Principles of Biochemistry. Prentice Hall, Englewood Cliffs (2002)Google Scholar
- 9.Liao, L., Noble, W.S.: Combining pairwise sequence similarity and support vector machines for remote homology detection. In: Proc. 6th. Int. Conf. on Computational Molecular Biology, pp. 225–232 (2002)Google Scholar
- 10.Oğul, H., Erciyes, K.: Identifying all local and global alignments between two DNA sequences. In: Proc. 17th Int. Sym. on Computer and Information Sciences, pp. 468–475 (2001)Google Scholar