search

AllenInstitute / brain_observatory_qc

Other
2 stars 0 forks source link

Mouseseeks Datacube Cortical/Fullfield column #225

Open samiamseid opened 1 year ago

samiamseid commented 1 year ago

Feature Request Motivation More projects are using data collected in cortical and fullfield stacks, which are "optional" datasets in ophys sessions. This means that not every session is associated with one of these files, and often times only one stack per container exists (or even zero). Because it is not collected every time, it is not only difficult to track down the experiment which contains these files, but difficult to know whether an operator has already collected one for a given mouse. Therefor, we are proposing a new feature which would exist in the Datacube view for a mouse, that would help operators and scientists with these issues.

Requested Feature/Solution Two new columns in the datacube view for a mouse could solve this issue. Mouse-seeks could walk through the session folders for a given mouse ID and return a check mark if any file contains "cortical" or "fullfield" in the name. This alone would solve the issue of operators forgetting to collect stacks for a mouse.

An additional feature which would assist scientists in finding these key files could be an expanded menu which would list out the sessions IDs associated with each returned file matching the above criteria. A button that copies the filepath could then make it easy to access each of the relevant files.

Scope This work would affect all mesoscope projects

matchings commented 8 months ago

@seanmcculloch this sounds useful, however i wonder if this is no longer needed given the move to Code Ocean. It should be possible to query CO for a particular type of data (like a cortical stack), right? Do you think it should still be displayed in mouse QC?

samiamseid commented 8 months ago

currently the "datacube" part of mouse_qc is mostly abandoned by users. We currently only fill it in with mouse_ids so that the containers will generate. I think for this feature to be useful, we would need to fix the other datacube issues. With upcoming code-ocean it may be worthwhile to instead start from scratch with the datacubes.