Shape Matching and Object Recognition Using Shape Contexts
IEEE Transactions on Pattern Analysis and Machine Intelligence
A Metric for Distributions with Applications to Image Databases
ICCV '98 Proceedings of the Sixth International Conference on Computer Vision
| Hi-index | 0.00 |
The study of non-sequential alignments, with different connectivity of the aligned fragments in the proteins being compared can offer a more complete picture of the structural, evolutionary and functional relationship between two proteins, than what is possible purely with sequential alignments. The design of techniques for non-sequential protein structure alignment therefore, constitutes an important direction of research. This paper introduces a novel method for non-sequential protein structure alignment involving three principle technical facets: (1) determination of the seed alignments not just by matching features from a single residue or considering well defined regions in the structure such as α–helices and β-strands, but through rich and robust descriptors that can capture the structural similarities of the local 3D environment around arbitrary residues of interest. (2) Scoring alignments using both geometric criterion (RMSD) as well as the biochemical characteristics of the residues. (3) An iterative chaining process which alternates between refinement and non-sequential extension stages to build a final alignment. The efficacy of the approach is demonstrated using the RIPC reference set which includes 40 structural pairs that are problematic to align. The performance of the method was found to be comparable or better than established techniques across the experiments.