add MiniscopeDataInterface

[bendichter]

Motivation

fix #162

Checklist

• Have you thoroughly read our Developer Guide?
• Have you ensured the PR description clearly describes the problem and solutions?
• Have you checked to ensure that there aren't other open or previously closed Pull Requests for the same change?
• If this PR fixes an issue, is the first line of the PR description fix #XXX where XXX is the issue number?
• Did you update the CHANGLOG.md file on your branch to describe your changes?

[CodyCBakerPhD]

The OnePhotonSeries here: NeurodataWithoutBorders/nwb-schema#523

has been approved by Oliver but stalled after that. Let's try to get that integrated and make use of that instead of making something ad-hoc here.

[bendichter]

good idea!

[CodyCBakerPhD]

Gonna need some example data together with some example stub files to really understand this, but the behavior video stream is attached to the regular microscope camera? Are they synchronized through the same system?

[bendichter]

They are synchronized through the same system

[bendichter]

http://miniscope.org/index.php/Data_Acquisition_Software#Data_Structure

[CodyCBakerPhD]

That is pretty nice then actually, first time I've seen a format with auto-synchronizing behavior stream!

[h-mayorquin]

@bendichter
For the test that is failing you need to add the name of the interface in the following lines:

neuroconv/tests/imports.py

Lines 109 to 110 in aae9a0d

	"Suite2pSegmentationInterface",
	"TiffImagingInterface",

[h-mayorquin]

I added type hints which were causing tests to fail. I guess we don't have gin data for this, do we?

[bendichter]

We need gin data for this interface

[CodyCBakerPhD]

Found some example files: https://github.com/denisecailab/minian/tree/master/demo_movies

with some corresponding .nc files (not sure what those are): https://github.com/denisecailab/minian/tree/master/demo_data/session1

OnePhotonSeries is under way: NeurodataWithoutBorders/pynwb#1593

If we wanted to get this through faster, I would say we would at least need to adjust the ndx-miniscope extension to use a TwoPhotonSeries named OnePhotonSeries instead of an ImageSeries as it currently is. There's a lot of ophys metadata about the imaging planes and optic channels that isn't linked otherwise...

The behavior cam stuff would be good to go as is, though. I think the VideoInterface should actually work for that as-is, but we could always create a specially named child class just for miniscope if desired

[CodyCBakerPhD]

Looking deeper into the ndx-miniscope extension, its practice of writing the data to a simple ImageSeries causes us to loose a lot of valuable ophys metadata structure (the ImagingPlane, the OpticalChannels, etc.) so I'd prefer if it at least wrote to a TwoPhotonSeries named 'OnePhotonSeries' (as people have had to do for a while now) until the hanging issues on the OnePhotonSeries (NeurodataWithoutBorders/pynwb#1593) are resolved and released

[codecov]

Codecov Report

Merging #163 (522e7e8) into main (f0d5886) will decrease coverage by 0.41%.
The diff coverage is 50.00%.

❗ Current head 522e7e8 differs from pull request most recent head ac6f316. Consider uploading reports for the commit ac6f316 to get more accurate results

@@            Coverage Diff             @@
##             main     #163      +/-   ##
==========================================
- Coverage   91.73%   91.32%   -0.42%     
==========================================
  Files          78       66      -12     
  Lines        3739     3435     -304     
==========================================
- Hits         3430     3137     -293     
+ Misses        309      298      -11     

Flag	Coverage Δ
unittests	91.32% <50.00%> (-0.42%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
...terfaces/ophys/miniscope/miniscopedatainterface.py	48.00% <48.00%> (ø)
src/neuroconv/datainterfaces/__init__.py	100.00% <100.00%> (ø)
...c/neuroconv/tools/spikeinterface/spikeinterface.py	90.21% <0.00%> (-0.92%)	⬇️
...erfaces/ecephys/baserecordingextractorinterface.py	94.00% <0.00%> (-0.12%)	⬇️
src/neuroconv/utils/__init__.py	100.00% <0.00%> (ø)
src/neuroconv/basedatainterface.py	96.00% <0.00%> (ø)
...onv/datainterfaces/ecephys/ced/ceddatainterface.py	85.71% <0.00%> (ø)
...onv/datainterfaces/ecephys/phy/phydatainterface.py	100.00% <0.00%> (ø)
...terfaces/ecephys/spikeglx/spikeglxdatainterface.py	100.00% <0.00%> (ø)
... and 27 more

[CodyCBakerPhD]

Replaced by #468, which uses the OnePhotonSeries, ophys metadata, and ROIExtractors on top of the ndx-miniscope types

GIN data should be made available soon for Miniscope v3