Software support for SBGN maps: SBGN-ML and LibSBGN

Abstract

Motivation: LibSBGN is a software library for reading, writing and manipulating Systems Biology Graphical Notation (SBGN) maps stored using the recently developed SBGN-ML file format. The library (available in C++ and Java) makes it easy for developers to add SBGN support to their tools, whereas the file format facilitates the exchange of maps between compatible software applications. The library also supports validation of maps, which simplifies the task of ensuring compliance with the detailed SBGN specifications. With this effort we hope to increase the adoption of SBGN in bioinformatics tools, ultimately enabling more researchers to visualize biological knowledge in a precise and unambiguous manner.

Availability and implementation: Milestone 2 was released in December 2011. Source code, example files and binaries are freely available under the terms of either the LGPL v2.1+ or Apache v2.0 open source licenses from http://libsbgn.sourceforge.net.

Contact: sbgn-libsbgn@lists.sourceforge.net.

Fig. 1.

An example PD map (right) with the corresponding SBGN-ML code (left). This example shows the import of glucose followed by the first step of glycolysis. The colors used have no special meaning in SBGN, here they merely indicate the relation between each SBGN glyph and its SBGN-ML representation; a process node in orange, a simple chemical (ATP) in green, a production arc in cyan, a catalysis arc in purple, a compartment in yellow and a state variable in blue

Fig. 2.

Example of reading a file using the Java version of LibSBGN. Here an SBGN-ML file named ‘adh.sbgn’ (included in the LibSBGN source distribution) is read, and some basic information about each glyph in that file is printed to standard output. The complete program can be found as ReadExample.java in the LibSBGN source distribution

Fig. 3.

Rendering comparison. A series of test-cases is rendered by all supported tools in an automated render comparison pipeline. The rendering results are compared with the reference map (top-left), in this case an ER map. A couple of significant differences have been highlighted with red circles. In the PathVisio case (top-right), arrowheads are drawn where none is expected. In the SBML Layout example (bottom-right), the wrong arrowheads are drawn for absolute inhibition and stimulation arcs. Note that these are historical images for illustration purposes, and the highlighted issues have already been fixed

Fig. 4.

Typical validator benchmark. This particular example tests the software for rule pd10110: in PD maps, catalysis arcs must point to a process node (not to an entity pool node). In the negative test-case on the left, the enzyme GPI appears to ‘catalyze’ a molecule rather than a reaction. This is a logical impossibility. The positive test-case on the right shows correctly how the enzyme GPI catalyzes the reaction from glucose-6P to fructose-6P. Taken together, these test-cases help to prevent bugs in the validation software

Fig. 5.

Screenshots of a number of tools that use LibSBGN. Clockwise, from the top: CellDesigner, SBGN-ED, VISIBIOweb and PathVisio. These tools are able to use SBGN-ML for import, export or both. At the time of writing, for some of these tools a version with SBGN support has not been officially released, but is expected soon