ClusterCells: Cluster the Cells

In mkarikom/RSoptSC: Low-Rank Factorization for Subspace-clustering

Cluster the Cells

Description

Use NMF to cluster the cells. Currently, the Lee and Seung algorithm is implemented via a call to renozao/NMF. If cluster-number is inferred then output relevent diagnostics (eg the eigenvalue spaceing of the ensemble method).

Usage

ClusterCells(similarityMatrix = NULL, n_clusters = NULL, n_comp = 3, ...)

Arguments

similarityMatrix	a symmetric nonnegative similarity matrix
n_clusters	default is NULL, ensemble method will be called
n_comp	the number of similarity components to use for ensemble clustering
...	additional parameters for NMF

Value

a list containing

H	the cluster weight matrix, the factorization of the similarity matrix
labels	the cluster labels
ensemble	results of model selection

mkarikom/RSoptSC documentation built on May 10, 2023, 1:10 a.m.