Inquiry on KEGG to SBML/SBGN Conversion and Boolean Modeling Workflows

[Muhammad Kabir]

Dear Members of the SBML Community,

I hope you are doing well.

I am currently working on constructing Boolean network models for pathway simulation (e.g., knock-in/knock-out analysis) using BoolNet. My goal is to derive these models from curated pathway resources.

More specifically, I am now focusing on pathways from the KEGG database (for example tyrosine_metabolism pathway at https://www.kegg.jp/pathway/map00350) and would like to convert KEGG pathway representations (KGML format) into either SBML (preferably SBML-qual) or SBGN formats, which are more suitable for downstream logical modeling.

I previously explored using KEGGtranslator for KGML to SBML/SBGN conversion; however, it appears to be outdated and no longer reliably supported. As a result, I am currently facing challenges in establishing a robust and reproducible workflow.

I would greatly appreciate your guidance on the following:

• Are there currently maintained tools or pipelines for converting KEGG KGML files into SBML (especially SBML-qual) or SBGN formats?

• Are there recommended best practices for transforming KEGG pathways into Boolean or logical models?

• Is there an alternative workflow (e.g., via BioPAX, Cytoscape, or other intermediate formats) that you would suggest for this purpose?

My overall aim is to establish a pipeline such as:

KGML (KEGG) → SBML/SBGN → Boolean network → simulation in BoolNet

If anyone has experience with a similar workflow or can recommend updated tools or strategies, I would be very grateful for your insights.

Thank you very much for your time and support.

Kind regards,
Muhammad Kabir

[Andreas Dräger]

Dear Muhammad, dear all,

Many thanks for raising this engaging question.

Since KEGGtranslator was mentioned, I maintain the repository, though I must admit the tool has not been actively updated for some time. That said, to the best of my knowledge, the core functionality of the conversion is still working reliably. The main limitation is that it is based on an older Java stack, which may affect usability (especially the GUI), but it also includes a command-line interface for more automated workflows.

One important aspect of KEGGtranslator is that the mapping from KGML to formats such as SBML or SBGN was developed in close interaction with the KEGG team. This goes beyond a straightforward interpretation of the KGML specification and aims to provide biologically meaningful conversions, which are not always guaranteed by tools that rely solely on the public KGML documentation.

If you encounter specific issues (e.g., failing conversions, unexpected semantics, or incompatibilities with SBML-qual workflows), I would strongly encourage you to report them via GitHub. Concrete examples are extremely helpful and would also guide potential updates of the tool.

More generally, I would be very interested to hear about current best practices in this area as well. In particular, the step from pathway representations (such as KEGG) to Boolean or logical models often requires additional curation and interpretation, regardless of the conversion tool used.

We would certainly welcome contributions or collaboration efforts to modernize KEGGtranslator and better support contemporary workflows.

With best regards

Prof. Dr. Andreas Draeger

---

Martin Luther University Halle-Wittenberg

Faculty of Natural Sciences III

Institute of Computer Science

Chair of Data Analytics and Bioinformatics

Von-Seckendorff-Platz 1 · Office #4.11 · 06120 Halle (Saale) · Germany
Phone: +49-345-55-24728 · Fax: +49-345-55-27039

Web: http://systems-biology.info · 𝕏: @dr_drae

YouTube: https://www.youtube.com/c/systemsbiology

--
For questions or feedback about the sbml-discuss list,
contact sbml...@googlegroups.com
---
You received this message because you are subscribed to the Google Groups "sbml-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sbml-discuss...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/sbml-discuss/1a2748e3-cd0b-48ad-8efd-fbf175e80e6cn%40googlegroups.com.