Confirming Proper Use

[AidanFishenden]

This is my first major project using kilosort, so I wanted to confirm that I ran it properly for my data before I start more analysis.

Information on my data Generated from a 32 channel probe. Probe had two unconnected shanks with 16 channels per shank. Channels were 50 microns apart, from depths of 50-800. The two shanks were 250 microns apart. Sampling rate was 40000, and at specific time points stimuli was administered, there were 420 stimuli per penetration. Recordings averaged around 210-215 seconds long.

Settings used in Kilosort GUI

Number of Channels 32 sampling frequency 40000 Batch size 60000 Nblocks 1 Th(universal) 9 Th(learned) 8 Tmin 0 Tmax inf Nt 61 Shift none Scale none Artifact threshold inf Nskip 25 Whitening range 32 High pass cutoff 300 Binning depth 5 Sig_interp 20 Drift smoothing .5,.5,.5 Nt0min none Dmin 0 Dminx 32 Min template size 10 Template sizes 5 Nearest chans 10 Nearest templates 100 Max channel distance none Templates from data true N templates 6 Npcs 6 Th(single channel) 6 Act threshold .2 Ccg threshold .25 Cluster downsampling 20 Xcenters none Duplicate spike ms .25

[jacobpennington]

The settings look fine. If you ran into an error or something looks strange in Phy, please let us know and attach kilosort4.log from the results directory.

[AidanFishenden]

Ok good, thank you.