[WIP] Annotation validation function

[mstaniak]

Motivation and Context

112, moreover, some tools export annotation in a way that doesn't match expectation of MSstats

[in short, such outputs require concatenation of BioReplicate and Condition to match repeated measures design]

Changes

Please provide a detailed bullet point list of your changes.

• added a function validateAnnotation which counts conditions per biological replicate. If each biorep matches a single condition, it's a proper group comparison design. If each biorep matches several conditions, it's a proper repeated measures design.

Questions:

• is it enough?
• should we report it in a different data structure? For example some summary table + information that it's OK or not
• should we report balanced/imbalanced design info or something more?

Testing

None yet, can add unit tests after we decide the functionalities.

Checklist Before Requesting a Review

• [x] I have read the MSstats contributing guidelines
• [x] My changes generate no new warnings
• [x] Any dependent changes have been merged and published in downstream modules

[tonywu1999]

I think this is a good start - I'm thinking we should eventually work with users to see what would be ideal in their eyes from a user experience perspective before adding features like reporting a summary table or displaying balanced/imbalanced design info.

I'm also wondering if annotation files should be validated inside the converters, but that would add an additional parameter of "design type" to each converter, which I don't think we want.

As an additional thing - validating that there are at least 2 bioreplicates per condition would be good to have here as well.

[mstaniak]

I think it's OK to just merge user's annotation at the converter step without checking. If there is an error in it, fixing it is just a matter of re-labeling one column

[mstaniak]

TODO:

• throw an error if there is only condition
• produce a warning if there is only one biological replicate per condition