Implementing datatype for new mzTab-M 2.1 format dedicated to metabolomics data

[nguilhot]

Hi,

Since the ELIXIR BioHackathon 2025, the mzTab-M working group has continued its work on defining a new generation of the mzTab-M specification dedicated to metabolomics. The format has now evolved significantly compared both to the original mzTab specification for proteomics and to mzTab-M 2.0.

We would like to start a discussion regarding support for the upcoming mzTab-M 2.1 format in Galaxy.

The current specification draft and schema are available here:

https://github.com/HUPO-PSI/mzTab-M/tree/main/specification_documents

Based on the current state of the specification, it now seems more appropriate to treat mzTab-M 2.1 as a dedicated metabolomics format rather than as a direct extension of the existing MzTab Galaxy datatype.

To avoid impacting existing behavior and workflows, our current idea would be:

• to keep the existing mztab and mztab2 datatypes completely unchanged,
• to introduce a new dedicated datatype implementation for mzTab-M 2.1,
• to place this implementation in a new metabolomics.py datatype module instead of adding it to proteomics.py,
• and to implement a dedicated class that would not inherit from the current MzTab/MzTab2 classes, since the structure and validation rules are now diverging substantially.

At this stage, before starting implementation work and preparing a new PR, we would appreciate your feedback on this proposed approach.

Thanks for your guidance and support.

Regards,

Nicolas

[fgiacomoni]

Hi @bernt-matthias, since you were involved in the previous mzTab-related PR, we’d really appreciate your feedback on this proposed direction for mzTab-M 2.1 in Galaxy. Your input would be very helpful before we start implementation. Thanks!
Regards,
Franck