NEW PROJECT - VERIFY new input (.csv .xlsx) / FOLDER RESTRUCTURE / INPUT_DICT.json

[jjb-hub]

each project will come with minimum:

1. a csv (# columns : 'mouse_id' 'group_id' 'compound_regions' ...)
2. input_dicts.py (see jasmine branch)

df needs to be verified: each project will use some of its own acronyms i.e. DO == DOPAC = 3,4-Dihydroxyphenylacetic acid hence in input_dicts, a dict of compounds and regions exists specifying each this needs to be used to then map on a standard set of nomenclature defined in constants (still working to get acronyms from allensdk)

this then feeds into subselections: ~connect to issue #40 which are defined using the standard acronyms and will need to be saved as JSON to be used for plotting/subsetting ect as I would like to be able to use these sub selectors I need them to work with plotters like correlogram or network or PCA

[jjb-hub]

@remicorne This issue has been broken down into parts: VERIFY function FILE RESTRUCTURING (image attached) INPUT_DICT.JSON - create and implement based on input_dict.py

VERIFY function - check the raw file is consistent with the code:

• check its .csv or .xlsx (xlsx probably better csv can be annoying with missing data)
• check mouse_id and group_id cols exists
• check every other col is compound_region (will use input dicts referring to region/compound constants here) and force user to rename to be consistent with input dicts ... if input_dict is lacking, write Something like "new compound/region acronym fullname" and both column and input dict would be updated)
• check data is Always number (probably overkill but common practice when dealing with humans)

[jjb-hub]

once remi_new_architecture to main redundant