bug normalise function

[toobiwankenobi]

Hi there,

thank you for the work on this package. I realized that there seems to be a bug (unless this behavior is intended) in the normalise() function.

If the data.frame you use as input was previously grouped (by dplyr::group_by()), the function seems to apply median normalization to the previous grouping and not to the sample column. I didn't check the source code but maybe you need to include dplyr::ungroup() to make sure that previous groupings are removed. Anyway, the normalisation works if I do dplyr::ungroup() before using normalise(). Let me know if something is unclear or if you need an example data set.

Best wishes,
Tobi

Version Info
protti version 0.9.0
R version 4.4

[jpquast]

Hi @toobiwankenobi,
thanks for your question.

This is indeed an intended behaviour. We decided that the function should work with a grouped data.frame as it would be expected. So using dplyr::ungroup() before is indeed the right thing to do.

The reason for keeping it compliant with group_by() is that this would allow users to normalise certain batches in a big datasets separately (by grouping by the batch column).

[toobiwankenobi]

Hi @jpquast,
Thanks for your answer! This makes sense. It would be useful to describe this behavior in the help section of the function because the sample argument will be ignored if the data.frame was grouped before. It took me some time to understand why the normalization was not working properly.
Best wishes, Tobi