Nucleic Acids Research Advance Access published online on November 11, 2009
Nucleic Acids Research, doi:10.1093/nar/gkp980
Database Issue |
PTGL: a database for secondary structure-based protein topologies
1Max Planck Institute for Molecular Plant Physiology, Bioinformatics, Am Muehlenberg 1, 14476 Potsdam-Golm, 2Leibniz-Institut fuer Molekulare Pharmakologie, Structural Bioinformatics, Robert-Roessle-Strasse 10, 13125 Berlin, 3Zuse Institute Berlin, Computer Science Research, Takustrasse 7, 14195 Berlin, 4Beuth University for Technology Berlin, FB VI, Bioinformatics, Seestrasse 64, 13347 Berlin and 5Max Planck Institute for Molecular Genetics, Computational Molecular Biology, Ihnestrasse 73, 14195 Berlin, Germany
*To whom correspondence should be addressed. Fax: +49 (0)331 5678136; Email: may{at}mpimp-golm.mpg.de Correspondence may also be addressed to Ina Koch. Tel: +49 (0)30/8413-1168; Fax: +49 (0)30/8413-1152; Email: koch_i{at}molgen.mpg.de
Received August 15, 2009. Revised October 14, 2009. Accepted October 15, 2009.
With growing amount of experimental data, the number of known protein structures also increases continuously. Classification of protein structures helps to understand relationships between protein structure and function. The main classification methods based on secondary structures are SCOP, CATH and TOPS, which all classify under different aspects, and therefore can lead to different results. We developed a mathematically unique representation of protein structure topologies at a higher abstraction level providing new aspects of classification and enabling for a fast search through the data. Protein Topology Graph Library (PTGL; http://ptgl.zib.de) aims at providing a database on protein secondary structure topologies, including search facilities, the visualization as intuitive topology diagrams as well as in the 3D structure, and additional information. Secondary structure-based protein topologies are represented uniquely as undirected labeled graphs in four different ways allowing for exploration under different aspects. The linear notations, and the 2D and 3D diagrams of each notation facilitate a deeper understanding of protein topologies. Several search functions for topologies and sub-topologies, BLAST search possibility, and links to SCOP, CATH and PDBsum support individual and large-scale investigation of protein structures. Currently, PTGL comprises topologies of 54 859 protein structures. Main structural patterns for common structural motifs like TIM-barrel or Jelly Roll are pre-implemented, and can easily be searched.