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

Small variations in timing of spikes produce several clusters #510

Closed guilhemibos closed 8 months ago

guilhemibos commented 2 years ago

Hi

I'm new to Kilosort. I'm using KS3, recording from 32 channels linear stereotrodes (50um between contacts), with open pehys. I have a hard time calibrating KS in order to have decent results which do not need very long manual curation.

Often, I observe some sorting like this: A large spike is separated arbitrarily into 2 or more units because their is a small temporal drift in the waveform, messing up with correlagrams of neurons. Is there a reason why this happens and how can I get rid of this issue? Thanks Screenshot from 2022-07-06 15-59-17

Screenshot from 2022-07-06 16-00-01

Screenshot from 2022-07-06 15-50-01

ananmoran commented 1 year ago

Hi, I just want to add that I encounter exactly the same phenomenon sorting Neuropixels data recorded with SpikeGLX. I hope someone can help to resolve this issue. Thanks Anan

shirquinn commented 1 year ago

@guilhemibos Did you manage to solve the problem? can you share what you did? I'm having a similar issue