redefining PC space for feature view
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 430
- Forks
- 187
- Avg merge
- 2h 31m
- Merged PRs (30d)
- 2
Description
Hello!
Is it possible to display cluster features in global PCA space in Phy?
I am using silicon probes that have recording sites on both sides, and there is some double counting of spikes between clusters across the two sides of the shank (see attached pic) - spikes of a blue unit on one side of the shank also appear as part of a red unit on the other side.
I would like to put these clusters into a common PC space to see the cluster overlap, and perhaps see if I can cut out the double counted spikes.
I have figured out how to resolve this prior to a call to Kilosort by grouping channels on the two sides of a shank (rather than treating these as independent groups, as is apparent in the image). However, I have many manually curated sessions in which this problem appears, so I wonder if there's a way to redefine the PC space after automated clustering has taken place.
Thank you for any advice!
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
The issue names no files, tests, or entry points. Start by locating how Phy builds feature views and PCA data, then investigate how Kilosort channel groups and existing clusters are represented. Done would mean defining and validating a way to display clusters from both shank sides in one shared PCA space after clustering.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- data-visualization
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100