Computation and Visualization of Protein Topology Graphs Including Ligand Information

Abstract

Motivation: Ligand information is of great interest to understand protein function. Protein structure topology can be modeled as a graph with secondary structure elements as vertices and spatial contacts between them as edges. Meaningful representations of such graphs in 2D are required for the visual inspection, comparison and analysis of protein folds, but their automatic visualization is still challenging. We present an approach which solves this task, supports different graph types and can optionally include ligand contacts. Results: Our method extends the field of protein structure description and visualization by including ligand information. It generates a mathematically unique representation and high- quality 2D plots of the secondary structure of a protein based on a protein-ligand graph. This graph is computed from 3D atom coordinates in PDB files and the corresponding SSE assignments of the DSSP algorithm. The related software supports different notations and allows a rapid visualization of protein structures. It can also export graphs in various standard file formats so they can be used with other software. Our approach visualizes ligands in relationship to protein structure topology and thus represents a useful tool for exploring protein structures. Availability: The software is released under an open source license and available at http://www.bioinformatik.uni-frankfurt.de/ in the Software section under Visualization of Protein Ligand Graphs.