Replies: 2 comments 2 replies
-
Hi @Aryo-Zare Thanks for opening up the discussion. We tried to ask them for the code, with no luck... :( Indeed in SpikeInterface we now have several ways to do motion estimation and correction (including a similar approach to Kilosort4, which I don't believe has a new method for drift correction, but is a port of pykilosort there). Our experimental |
Beta Was this translation helpful? Give feedback.
-
thanks!!!looking forward to that.
Alessio Buccino ***@***.***> 于 2023年10月12日周四 12:49写道:
… Of course, but kilosort4 code is not our yet. As soon as they publish the
code we'll start the integration
—
Reply to this email directly, view it on GitHub
<#1473 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ALY2FQT2L2JI7V3LXVMIUZLX67DLNANCNFSM6AAAAAAWQ55X2E>
.
You are receiving this because you commented.Message ID:
<SpikeInterface/spikeinterface/repo-discussions/1473/comments/7262178@
github.com>
|
Beta Was this translation helpful? Give feedback.
-
You may have already been informed, yet I wanted to mention a new publication regarding a new version of kilosort :
https://www.biorxiv.org/content/10.1101/2023.01.07.523036v1
"Solving the spike sorting problem with Kilosort"
They mention in the manuscript that the new version is written in python from ground up. And that it will be shown in the current github page of kilosort :
https://github.com/MouseLand/Kilosort
Yet so far there is no code available on github for the new version. So how do they work on it ?
kilosort has a built-in drift correction, but seeminlgy does not have an auto-merge option.
In constrast cicus does not have a built-in drift correction but can automatically merge similar units.
Yet spike-interface can do both drift correction and automatic merging independently.
Beta Was this translation helpful? Give feedback.
All reactions