search

MouseLand / Kilosort

Fast spike sorting with drift correction
https://kilosort.readthedocs.io/en/latest/
GNU General Public License v3.0
478 stars 248 forks source link

NPX 1.0 tetrode channel map #563

Closed chris-angeloni closed 8 months ago

chris-angeloni commented 1 year ago

Hi, I used the .imro file from the SpikeGLX site to configure my recording channels as tetrodes across a larger span of a NPX 1.0 probe (ie. it records 4 channels, then skips 4 channels, effectively doubling the recording span). A consequence of using this file is that the first four channels are at the tip of the probe (bank0), then the next 4 are 384 channels up the probe (bank1), and so on.

I attempted to correct this by creating a channelMap file that modifies the y coordinates accordingly, but when I load this in kilosort GUI, it does not appear to apply the new coordinates from the map to the visualization of the data: image (note the alternating bands of 4 channels)

Is this just a visualization thing? Will the true positions of the channels in the map be used for spike sorting?

chris-angeloni commented 1 year ago

It looks like it did not reorder the channel position accordingly during spike sorting... image