SOAPy_st.pp.make_network

SOAPy_st.pp.make_network(adata: AnnData, sample_key: str | None = None, sample: str | int | list | None = None, cluster_key: str = 'clusters', method: Literal['radius', 'knn', 'regular', 'neighbor'] = 'knn', cutoff: float | int = 6, max_quantile: float | int = 98, exclude: str | dict | None = None, scale: str | float = 'hires', spatial_in_obsm: str = 'spatial', inplace: bool = True) → AnnData

A function to create a network based on spatial information. We offer four different ways to build a network: KNN network, range radiation network, regular network

and First-order neighbor network

‘exclude’ is a parameter to exclude categories that cannot form an edge, you can set ‘same’ and ‘different’ to specifies the same/different clusters may not be connected.If you want to define a custom class of points that can’t be connected as edges, pass it as a dictionary.

Parameters

adataanndata.AnnData

An AnnData object containing spatial omics data and spatial information.

sample_keystr, optional

Batch’s key in adata.obs.

sampleUnion[str, int, list], optional

The sample number for which the network needs to be built.

cluster_keystr, optional

The column label of clusters in adata.obs.

methodstr, optional

the method to make network, select in ‘Radius’ and ‘KNN’.

cutoffUnion[float, int], optional

In KNN network and regular network, cutoff means number of neighbors to use. In range radiation network, cutoff means range of parameter space to use

max_quantileUnion[float, int], optional

In Neighbor network, Order the distance of all sides, and more than max_quantile% of the sides will be removed

excludeUnion[str, float], optional

Excluding categories that cannot form an edge.

scaleUnion[str, float], optional

Scale used in subsequent analyses. If it’s Visium data it can also be HE image labels (hires or lower). Most of the time you don’t need to change this.

spatial_in_obsmstr, optional

Keyword of coordinate information in obsm.

inplacebool, optional

Whether to change the original adata.

Returns

uns['SOAPy']

SOAPy generated parameters

uns['SOAPy']['indices']

adjacency matrix of network

uns['SOAPy']['distance']

distance matrix of network

uns['SOAPy']['edges']

edges of network