| Autor: |
Lim, Seonjoo, Jaegle, Stephen, Hamel, Lutz |
| Předmět: |
|
| Zdroj: |
Proceedings of the International Conference on Bioinformatics & Computational Biology (BIOCOMP) (American Council on Science & Education); 2016, p10-16, 7p |
| Abstrakt: |
Here we describe an approach for protein structure-function analysis using self-organizing maps based on the structure of a protein's functional site. Our current approach differs from other approaches in that we directly unfold the 3D structure of the functional center of a protein into a suitable high-dimensional feature vector and then use self-organizing maps to discover similarities/dissimilarities between the corresponding feature vectors. We successfully applied our techniques to two large protein families: the protein kinases and the Ras superfamily. Even though a number of different approaches using self-organizing maps for the conformational analysis of molecules have been proposed, our approach is novel in that we apply it to protein families, use an efficient feature vector construction, and use a recently developed selforganizing map package that provides statistical support for evaluating the resulting map quality. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
