Please use the online documentation website going forward: https://bahanonu.github.io/ciatah/

Preprocessing: Improving SNR

This relates to methods within the calciumImagingAnalysis class for checking ΔF/F during cell sorting and ways to improve SNR (esp. in miniscope movies), also see Preprocessing:-Spatial-filtering.

Movie cell ROI trace during cell sorting

• If users press r while in the interface, it will show users a ROI calculated trace (which will contain more cross-talk, but will give you a direct measure of ΔF/F from your movie compared to ICA/CNMF-E/etc. traces.

Spatial filtering during modelPreprocessMovie preprocessing

• The other is that in the modelPreprocessMovie preprocessing module, when you are on the 1st page of the options, in the filterBeforeRegister setting select matlab divide by lowpass before registering or matlab bandpass before registering.
• See also the spatial filtering wiki page: https://github.com/bahanonu/calciumImagingAnalysis/wiki/Preprocessing:-Spatial-filtering#dark-halos-around-cells.

viewMovieRegistrationTest

• To get a feel for how the different filtering affects SNR/movie data, run viewMovieRegistrationTest module and select either matlab divide by lowpass before registering or matlab bandpass before registering then change filterBeforeRegFreqLow and filterBeforeRegFreqHigh settings, see below.

• You'll get an output like the below (top left is without any filtering, other 3 are with different bandpass filtering options).

• Cell ΔF/F intensity profile from the raw movie

• Same cell ΔF/F intensity profile from the bottom/left movie (not the y-axis is the same as above):