Skip to content
New issue

Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? # to your account

Add parameter to sliding_window to end up with windows of (approximately) same number of cells #908

Open
timtreis opened this issue Nov 4, 2024 · 2 comments
Labels
enhancement ✨ New feature or request feature PR introduces a new feature graph 🕸️ squidpy2.0 Everything releated to a Squidpy 2.0 release

Comments

@timtreis
Copy link
Member

timtreis commented Nov 4, 2024

Could add a heuristic that tries to optimise the number of cells per window (and unassigned cells) so that on average, most windows have a number of cells assigned that is as close as possible to what the user specified.

@timtreis timtreis added enhancement ✨ New feature or request feature PR introduces a new feature graph 🕸️ squidpy2.0 Everything releated to a Squidpy 2.0 release labels Nov 4, 2024
@FrancescaDr
Copy link
Contributor

Hi Tim, I agree this would be a really useful feature. More specifically, we could add the following parameters:

  • max_nr_cells: int = None.
    • If not None, it calculates the window_size such that the number of cells in a sliding window is <= max_nr_cells (e.i. 1k).

Considering that given multiple library_key's the FOVs have different sizes or densities we could introduce a separate parameter that differentiates between whether the window sizes should be equal across all library_key's or not. For example:

  • window_size_per_library_key: 'equal', 'unique'
    • 'equal': calculate one window size for all library_keys
    • 'unique': calculate a unique window size for all library keys

@mossishahi
Copy link
Collaborator

Hi @timtreis and @FrancescaDr , I'll work on this.

# for free to join this conversation on GitHub. Already have an account? # to comment
Labels
enhancement ✨ New feature or request feature PR introduces a new feature graph 🕸️ squidpy2.0 Everything releated to a Squidpy 2.0 release
Projects
None yet
Development

No branches or pull requests

3 participants