Get clustering with KMeans

Perform KMeans clustering on a SingleCellExperiment object, with kmeans.

runKMeans(
  inSCE,
  nCenters,
  useReducedDim = "PCA",
  clusterName = "KMeans_cluster",
  nComp = 10,
  nIter = 10,
  nStart = 1,
  seed = 12345,
  algorithm = c("Hartigan-Wong", "Lloyd", "MacQueen")
)

Arguments

inSCE

A SingleCellExperiment object.

nCenters

An integer, the number of centroids (clusters).

useReducedDim

A single character, specifying which low-dimension representation to perform the clustering algorithm on. Default "PCA".

clusterName

A single character, specifying the name to store the cluster label in colData. Default "KMeans_cluster".

nComp

An integer. The number of components to use for K-Means. Default 10. See Detail.

nIter

An integer, the maximum number of iterations allowed. Default 10.

nStart

An integer, the number of random sets to choose. Default 1.

seed

An integer. The seed for the random number generator. Default 12345.

algorithm

A single character. Choose from "Hartigan-Wong", "Lloyd", "MacQueen". May be abbreviated. Default "Hartigan-Wong".

Value

The input SingleCellExperiment object with factor cluster labeling updated in colData(inSCE)[[clusterName]].

Examples

data("mouseBrainSubsetSCE")
mouseBrainSubsetSCE <- runKMeans(mouseBrainSubsetSCE,
                                 useReducedDim = "PCA_logcounts",
                                 nCenters = 2)
#> Sat Mar 18 10:31:04 2023 ... Running 'KMeans clustering' with Hartigan-Wong algorithm.
#> Sat Mar 18 10:31:04 2023 ...   Identified 2 clusters