optional post-processing step for duplicate spike removal #234
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
1. remove_ks2_duplicate_spikes - removes duplicate spikes resulting from multiple templates explaining variance from a single waveform
This reverts commit c4ed558.
optional (off by default) step for removing double-counted spikes
agbondy
commented
Sep 17, 2020
| % OPTIONAL: remove double-counted spikes - solves issue in which individual spikes are assigned to multiple templates. | ||
| % See issue 29: https://github.com/MouseLand/Kilosort2/issues/29 | ||
| %rez = remove_ks2_duplicate_spikes(rez); | ||
|
|
There was a problem hiding this comment.
commented out by default
agbondy
commented
Sep 17, 2020
| @@ -0,0 +1,52 @@ | |||
| function rez = remove_spikes(rez,remove_idx,label,varargin) | |||
There was a problem hiding this comment.
this function encapsulates the code for removing spikes from a "rez" structure, so that it can easily be shared by set_cutoff.m and remove_ks2_duplicated_spikes.m. It also adds a field to the rez structure called "removed" which saves the removed spikes and provides a reason why they were removed in postprocessing.
agbondy
commented
Sep 17, 2020
| [ccbsort, iorig] = sortBatches2(ccb0); | ||
| [ccbsort, iorig, xs] = sortBatches2(rez.ccb); | ||
| rez.iorig = gather(iorig); | ||
| rez.ccbsort = gather(ccbsort); | ||
|
|
||
| % some mandatory diagnostic plots to understand drift in this dataset |
There was a problem hiding this comment.
plot the 1D drift map
…avior was to cause harder to decipher errors down the line).
|
Great, thanks Adrian! |
Merged
cnuahs
pushed a commit
to cnuahs/Kilosort
that referenced
this pull request
Mar 9, 2024
optional post-processing step for duplicate spike removal Former-commit-id: d703473
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I added a function "remove_ks2_duplicate_spikes" which takes in a "rez" structure during post-processing and removes spikes that are likely double-counted, based on their spatiotemporal overlap with other spikes. Specifically, it two spikes occur within less than X ms and belong to templates with peaks less than Y um apart, the spike belonging to the smaller-amplitude template is removed. X is 0.5ms by default, and Y is 100 um by default. We have implemented this in our kilosort master script for some time and I thought it may be useful for others. Right now, it's commented out so as not to disrupt the default behavior but could be uncommented to enable it (or in future enabled programmatically with a flag of some kind).
One other minor change in this PR is the addition of a 1D drift map to the figure illustrating drift in the recording. I find it a useful visual tool in addition to the distance matrices. It plots batch position on the 1D embedding performed already by ks2.
