dF, dF/F, and f0 are packaged in DANDI files now

[rcpeene]

here:
https://dandiarchive.org/dandiset/001424/draft/files?location=sub-794237&page=1

The files have been replaced with the latest NWB files on DANDI.

F0 can be accessed with the following paths
nwb.processing['ophys']['DMD1_f0']
nwb.processing['ophys']['DMD2_f0']

df can be accessed as
nwb.processing['ophys']['df'].roi_response_series['DMD1_df']
nwb.processing['ophys']['df'].roi_response_series['DMD2_df']

and dff as
nwb.processing['ophys']['dff'].roi_response_series['DMD1_dff']
nwb.processing['ophys']['dff'].roi_response_series['DMD2_dff']

[Sarruedi]

Thanks a lot for the update, Carter!

[Dedalus9]

@rcpeene Is it safe to compare amplitude differences between the pre- and post- tuning block with 'df/f' data?

[jeromelecoq]

Bleaching should manifest as a gradual decrease in baseline fluorescence as well as fluorescence responses.
In principle, the baseline fluorescence should be linearly proportional to the fluorophor concentration. Obviously the measurement of concentration can be biased. This twitter thread might be helpful : https://x.com/LecoqJerome/status/1466164516463394820

I will let @KasparP or @seyedolmohadesin speak more to it.