Improve the Validation of SBGN Diagrams

[cannin]

Background

Systems Biology Graphical Notation (SBGN) is a standard language for representing biological pathways with SBGNML being the exchange and storage format for SBGN. SBGN diagrams are created using the developed SBGN editors (Newt, CellDesigner, SBGN-ED etc.), and they can be saved as the XML-based SBGNML files. The specification that defines the rules governing SBGN can be difficult for many users. Validation tools can the ability of users to create clear SBGN diagrams that are unambiguous. Programmatic validation of SBGN is encoded in the Schematron language that can validate XML documents through an XSLT transformation; XSLT transformation is available in a variety of programming languages.

Goal

The goal of the project improve the validation of SBGN documents (and extend it if needed, by re-examining the SBGN specifications). Furthermore, an objective is also to update validation in common SBGN tools like Newt (written in Javascript) with any improvements. Possible improvements include

1. Clarification to the diagnostic information provided for individual rules that help user understand rule failures,
2. Addition of missing rules (needs re-examination of SBGN specs)

Starting point

Make sure you can run the Schematron validation in either Python or Javascript for example diagrams.

Difficulty Level: Medium

Difficulty is based on the having and understanding basics of SBGN

Size and Length of Project

• Medium: 175 hours

Skills

Essential skills: Javascript or Python, XML

Public Repository

• https://github.com/sbgn/libsbgn/tree/master/validation
• https://github.com/sciluna/sbgn_dataset_generator
• https://sbgn.github.io/ (specification exist here)

Potential Mentors

• Augustin Luna (augustin AT nih.gov)

[Arushi-Gupta13]

Hi @cannin interesting project!
I wanted to know what kinds of SBGN rules cannot be validated using Schematron?

[Drishti160803]

Hi @cannin
Sir, I noticed that some current Schematron asserts provide generic error messages. I plan to use the element in Schematron to provide user-friendly explanations..
Will you plse guide me ?

[roopasreegit]

Hi, @cannin
I’m Roopasree, a third-year CS student with experience in JavaScript, Python, and working with structured data formats like JSON and XML. I’m very interested in contributing to this SBGN validation project.

From my understanding, the project focuses on improving schematron based validation for SBGNML, enhancing diagnostic messages, and extending the validation rules based on the SBGN specification. I find this particularly interesting because I took biology in my high school and I've been looking for opportunities to combine my love for development and biology meaningfully in a project.

As a starting step, I plan to set up the current schematron validation pipeline in both Python and JavaScript and test it on example SBGNML diagrams. I’d like to analyze existing validation rules and compare them with the official SBGN specifications to identify missing or ambiguous constraints.

For improvements, I’m interested in:

Enhancing diagnostic error messages to be more understandable and convenient for users by explaining semantic meaning and not just schema failures

Adding missing validation rules by systematically mapping the SBGN spec to schematron assertions.

Exploring how improved validation can be integrated into Newt and other SBGN editors, possibly by exposing validation feedback directly in the UI.

I also plan to explore the sbgn_dataset_generator to generate diverse test cases and evaluate validation coverage. If possible, I’d like to write automated tests for validation rules to ensure robustness and avoid regressions.

I’m comfortable working with GitHub workflows, reading specs, and collaborating through issues and PRs. I’d love to start by tackling a small validation improvement or documentation task. Please let me know if there’s a good first issue or guidance on setting up the environment.

Looking forward to contributing and learning from this project.