Proposal: Move MSstats Out of quantms Into a Separate Downstream Workflow

[ypriverol]

Currently, the MSstats step is included as part of the quantms workflow. Based on feedback from previous users and internal discussions, we have observed that this step may be better suited outside the main workflow, in a separate pipeline or tool.

The main reasons for this proposal are:

• Highly manual analysis step
• MSstats requires substantial user input and dataset-specific decisions.
• Strong dependency on experimental design and method selection
• The appropriate statistical method and parameter settings vary significantly depending on the dataset and analysis goals.
• Users often report difficulties understanding the required parameters, default settings, and how to configure the step properly.

Given this, we propose removing MSstats from quantms and instead have the main workflow end with the generation of the identification and quantification tables.

This would make quantms more focused and easier to use:

• RAW files → identification / quantification tables

For downstream analyses, such as:

• baseline analysis
• differential expression / abundance analysis
• other statistical post-processing

We plan to develop a separate dedicated downstream workflow or pipeline, similar in concept to how differential expression is handled in other ecosystems such as nf-core. This could potentially also be offered as a standalone desktop tool in the future.

We believe this change would make quantms:

• smaller
• easier to understand
• easier to maintain
• more modular
• more focused on primary data processing

If anyone has concerns, objections, or alternative suggestions, please let us know — feedback is very welcome before moving forward with this change.