A new topological method to measure protein structure similarity |
| |
Authors: | Bostick David Vaisman Iosif I |
| |
Affiliation: | Department of Physics and Program in Molecular/Cell Biophysics, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA. ivaisman@gmu.edu |
| |
Abstract: | ![]() A method for the quantitative evaluation of structural similarity between protein pairs is developed that makes use of a Delaunay-based topological mapping. The result of the mapping is a three-dimensional array which is representative of the global structural topology and whose elements can be used to construe an integral scoring scheme. This scoring scheme was tested for its dependence on the protein length difference in a pairwise comparison, its ability to provide a reasonable means for structural similarity comparison within a family of structural neighbors of similar length, and its sensitivity to the differences in protein conformation. It is shown that such a topological evaluation of similarity is capable of providing insight into these points of interest. Protein structure comparison using the method is computationally efficient and the topological scores, although providing different information about protein similarity, correlate well with the distance root-mean-square deviation values calculated by rigid-body structural alignment. |
| |
Keywords: | Protein structure Protein similarity Computational geometry Delaunay tessellation |
本文献已被 ScienceDirect PubMed 等数据库收录! |