Construct Similarity Network

This function generates a similarity network (SN) for a given embeddings. Specifically, the function computes K-nearest neighbor (KNN) graphs based on the provided embedding.

Usage

getSN(emb, npcs = 20, k = 50)

Arguments

emb

A matrix contains the embedding, with rows correspond to principal components (PCs) or features, and columns correspond to spatial neighborhoods.

npcs

An integer specifying the number of principal components or features for the analysis. Default is 20.

k

An integer specifying the number of nearest neighbors to consider when constructing the KNN graph. Default is 50.

Value

A matrix for the similarity network (SN). The rows and columns represent spatial neighborhood, and the values represent inverted distance.

Details

The function computes a KNN graph using the input embedding and inverts the resulting distances to similarity scores.