spatialtis.find_neighbors

spatialtis.find_neighbors(data, method='kdtree', r=None, k=None, scale=None, export_key=None, **kwargs)

To find the neighbors of each cell.

KD-tree and Delaunay triangulation are used when cells are points, by default search for k=5.

R-tree is used when cells are polygons, by default search for scale=1.4.

Note

When method="kdtree", you can search neighbors within radius and/or by nearest-neighbors. If you specific r=30, k=5, this will search within 30 while limited the number of neighbors to 5.

Parameters

dataAnnData

The AnnData to work with.

method{‘kdtree’, ‘rtree’, ‘delaunay’}, default: ‘kdtree’

rfloat

The search radius.

kfloat

The (minimum) number of nearest-neighbors.

scalefloat

How much to scale each cell, only if cell has shape.

export_keystr

The key used to store result.

**kwargs

Config for the analysis, for details check spatialtis.abc.AnalysisBase.