KEGGtranslator

KEGGtranslator is a Java desktop application for visualizing and translating pathway files from the KEGG PATHWAY database. It reads KEGG's native KGML XML markup (KGML) and converts it into a range of functional and graphical output formats, giving researchers a single tool to move KEGG pathways into simulation, network-analysis, and drawing environments. It was developed at the Center for Bioinformatics Tübingen (Andreas Zell's group) primarily by Clemens Wrzodek, with Finja Wrzodek, Manuel Ruff, and Andreas Dräger, and published in Bioinformatics in 2011. The code is open source under the LGPL and is maintained by the Dräger Lab.

The tool distinguishes two families of targets. On the functional side it emits SBML (SBML) — including qualitative models via the SBML "qual" extension and layout/groups extensions — plus BioPAX Level 2/3 (BioPAX), enabling stoichiometric analysis, dynamic simulation, and semantic pathway integration. On the graphical side it produces GraphML (GraphML), GML (GML), SIF (SIF), SBGN-ML (SBGN-ML), TGF, and yWorks YGF, as well as rendered raster images (JPG, GIF) and LaTeX reports through SBML2LaTeX. Internally, functional conversion is built on the JSBML library, mapping KGML entries and reactions into SBML data structures before serialization.

A defining feature is that KEGGtranslator does not merely transcode: it enriches the comparatively sparse KGML source by querying the KEGG API and adds standardized, MIRIAM-compliant cross-references to Ensembl, UniProt, ChEBI, and other databases for each element. It can autocomplete fragmentary reactions with missing reactants or products, check atom balance, remove "white" nodes (genes absent in the target organism) and orphaned elements, and customize node naming. Both an interactive GUI and a command-line/batch mode are provided for bulk conversion of pathway collections.

Its strengths are breadth of output formats and rich annotation, which set it apart from narrower converters such as KEGGconverter, KEGGgraph, and kgml2sif; it complements the KEGG-access ecosystem around KEGG. Limitations follow from its scope: it is strictly KGML-in, so it depends on KEGG's data quality and API availability, autocompletion and atom-balancing require careful configuration for complex metabolic pathways, and as an older Java/Swing application it is no longer actively evolving.