Rethink assays nomenclature

[cvanderaa]

As discussed for the protocol paper with @samgregoire, the fact that the quantitative data in a SummarizedExperiment and the data objects in a QFeatures are both called assays complicates the description of the data structure.

I would suggest renaming the QFeatures' assays.

• sets: it is generic, like data sets (and in honor of the previously supported ExpressionSet class)
• experiments: because each object is a [SingleCell|Summarized]Experiment. Actually, the objects in QFeatures are retrieved using experiments(qf).
• elements: because each object is in fact an element of a list (ExperimentList)
• other?

[lmsimp]

Following @cvanderaa 😀 as you know we use QFeatures extensively in our lab and we have thought the same 👍

CCing @Charl-Hutchings @csdaw

[lgatto]

Could you share your thoughts on this. In our latest paper, we refer to the elements of a QFeatures object as sets. (@samgregoire, could you confirm).

[samgregoire]

I confirm.

• We use assay or quantitative assay to refer to SummarizedExperiment's quantitative data (since they are extracted using the assay() function).
• We use sets to refer to data objects in a QFeatures.

[lmsimp]

This great 👍 I think sets works well but any of Chris' suggestions above would be preferable over assays.

[tavareshugo]

my vote for "experiments" based on discussions on #197