Research ArticleOpen AccessGoogle Scholar indexed
Visualization of protein structure relationships using constrained twin kernel embedding
School of Science and Technology University of New England
School of Computer Science Charles Sturt University
School of Science and Technology University of New England
School of Biological, Biomedical and Molecular Sciences University of New England
- 1 School of Science and Technology University of New England
- 2 School of Computer Science Charles Sturt University
- 3 School of Science and Technology University of New England
- 4 School of Biological, Biomedical and Molecular Sciences University of New England
Journal of Biomedical Science and Engineering·Volume 01 (2008)·Pages 133–140·Published 1 August 2008·DOI10.4236/jbise.2008.12022
Copy link · social · email
Abstract
In this paper, a recently proposed dimensional-ity reduction method called Twin Kernel Em-bedding (TKE) [10] is applied in 2-dimensional visualization of protein structure relationships. By matching the similarity measures of the input and the embedding spaces expressed by their respective kernels, TKE ensures that both local and global proximity information are preserved simultaneously. Experiments conducted on a subset of the Structural Classification Of Pro-tein (SCOP) database confirmed the effective-ness of TKE in preserving the original relation-ships among protein structures in the lower di-mensional embedding according to their simi-larities. This result is expected to benefit sub-sequent analyses of protein structures and their functions.
KeywordsVisualization of protein structureConstrained twin kernel embedding
- Dimitris K. Agrafiotis. (1997) A new method for analyzing protein sequence relationships based on sammon maps. Protein Science, 6(2):287–293.
- Izydor Apostol and Wojciech Szpankowski. (1999) Indexing and mapping of proteins using a modified nonlinear sammon projection. Journal of Computational Chemistry, 20 (10):1049– 1059.
- Mikhail Belkin and Partha Niyogi. (2003) Laplacian eigenmaps for dimensionality reduction and data representation. Neural Computa-tion, 15(6):1373–1396.
- Samarasena Buchala, Neil Davey, Ray J. Frank, and Tim M. Gale. (2004) Dimensionality reduction of face images for gender classifi-cation. In Proceedings of 2nd International IEEE Conference on Intelligent Systems, volume 1, pages 88–93.
- Payel Das, Mark Moll, Hern an Stamati, Lydia E. Kavraki, and Cecilia Clementi. (2006) Lowdimensional free energy landscapes of protein folding reactions by nonlinear dimensionality reduction. In Proceedings of the National Academy of Sciences, volume 103, pages 9885?9890, USA
- Michael A. Farnum, Huafeng Xu, and Dimitris K. Agrafiotis. (2003) Exploring the nonlinear geometry of protein homology. Protein Science, 12(1604-1612) .
- Ronald A. Fisher. (1936) The use of multiple measurements in taxonomic problems. Annals of Eugenics, 7:179–188 .
- Amir Globerson, Gal Chechik, Fernando Pereira, and Naftali Tishby.(2005) Euclidean embedding of co-occurrence data. In Lawrence K. Saul, YairWeiss, and L eon Bottou, editors, Advances in Neural Information Processing Systems 17, pages 497?04. MIT Press, Cambridge, MA .
- Yi Guo, Junbin Gao, and Paul W. Kwan. (2006) Kernel Laplacian eigenmaps for visualization of non-vectorial data. In Lecture Notes on Artificial Intelligence, volume 4304, pages 1179– 1183.
- Yi Guo, Junbin Gao, and Paul W. Kwan. (2008) Twin kernel embedding. IEEE Transaction ofPattern Analysis and Machine Intelligence, submitted.
- Jihun Ham, Yuanqing Lin, and Daniel. D. Lee. (2005) Learning nonlinear appearance manifolds for robot localization. In IEEE/RSJ International Conference on Intelligent Robots and Systems, pages 2971– 2976.
- Jens Hanke and Jens G. Reich. (1996) Kohonen map as a visuali-zation tool for the analysis of protein sequences: multiple align-ments, domains and segments of secondary structures. Bioinfor-matics, 12(6):447–454.
- M. Jolliffe. (1986) Principal Component Analysis. Springer-Verlag, New York.