Add SpaGFT-like method for identification of SVGs #901
Labels
enhancement ✨
New feature or request
graph 🕸️
squidpy2.0
Everything releated to a Squidpy 2.0 release
SpaGFT (Spatial omics representation and functional tissue module inference using graph Fourier transform) uses Fourier transforms to identify SVGs (and then groups them into "tissue modules". At SBI2 the presenter said their method is "similar but better than Moran's I" - so we could potentially implement it as another flavor in https://squidpy.readthedocs.io/en/stable/api/squidpy.gr.spatial_autocorr.html.
The implementation already used AnnData and looks to be low on dependencies:
https://github.com/jxLiu-bio/SpaGFT/blob/master/SpaGFT/gft.py
The text was updated successfully, but these errors were encountered: