Correctly merging of multiple assays when accessing

[c-mertes]

It is not completely clear what the function should return. It allows for multiple assays but it is not specified how the data should be merged. The groupedIDs.update(xxx) will just replace any existing keys with the new values. To me this looks like we should rather merge uniquely all IDs into the given groups, which is not happening, as only the top level (keys) are merged and if already present overwritten.

https://github.com/gagneurlab/drop/blob/master/drop/config/SampleAnnotation.py#L275-L291

A use case would be: I would like to have all IDs for RNA samples (RNA) and externally provided samples (GENE_COUNTS).

The question now is how do we merge the lists within one entry? Concatenating or making it unique. Is the order important or not. Would it be ok to have the same ID within different assays?

What do you think @vyepez88 @mumichae