What would you like to see added to MatNWB?

Many NWB types are containers for other types, with two examples being ProcessingModule and Fluorescence.

The syntax for setting/getting data for these containers is cumbersome:

ophys_module = types.core.ProcessingModule('description', 'contains optical physiology data')

ophys_module.nwbdatainterface.set('ImageSegmentation', img_seg);

and

fluorescence = types.core.Fluorescence();

fluorescence.roiresponseseries.set('RoiResponseSeries', roi_response_series);
ophys_module.nwbdatainterface.set('Fluorescence', fluorescence);

A simpler way of doing this would be:

ophys_module = types.core.ProcessingModule('description', 'contains optical physiology data')

ophys_module('ImageSegmentation') = img_seg;

and

fluorescence = types.core.Fluorescence();

fluorescence('RoiResponseSeries') = roi_response_series;
ophys_module('Fluorescence') = fluorescence;

A similar suggestion was raised in pynwb (issue #1993) and it would be great to mirror this on the matnwb side.

Edge cases to consider:

• Spaces in neurodata type names. Use matlab.lang.makeValidName('my dataset') and internal mapping of names
• Reading datatypes where name is the same as the generated property name in matnwb. (See issue [Bug]: Constructing "container" class with name-value pair where name matches property name does not work #676 )
• Type with multiple un-named subgroups (currently only relevant for ProcessingModule). What if the same name is added to more than one subgroup? Is this related to this comment?

Is your feature request related to a problem?

No response

What solution would you like?

Redefine container classes to have dictionary/Set like behavior, while also restricting the types it is allowed to hold

Do you have any interest in helping implement the feature?

Yes.

Code of Conduct

• I agree to follow this project's Code of Conduct
• Have you ensured this change was not already requested?