Run k-means clustering algorithm

Source: R/clustering.R

Run k-means clustering algorithm

cluster_kmeans(fsce, expt = "rnaseq", k, method = "PCA",
  n_dims = NULL, seed = NULL, ...)

Arguments

fsce

An object of class FunctionalSingleCellExperiment
expt

Data to use for calculating variable features (default is rnaseq). Must be present in names(fsce).
k

number of classes
method

dimensionality reduction method for clustering (defaults to PCA)
n_dims

specify the number of dimensions from "dr" to use for clustering, defaults to all dimensions
seed

seed for reproducible result
...

additional arguments to pass to stats::kmeans()

Value

fsce with k_cluster in expt colData.

See also

Other clustering functions: cluster_leiden

Examples

# calculate PCA for k-means default method
fsce <- calc_pca(fsce_small)
#> scaling data
#> calculating pcs

fsce <- cluster_kmeans(fsce, k = 6)

SingleCellExperiment::colData(fsce[["rnaseq"]], "k_cluster")
#> DataFrame with 250 rows and 4 columns
#>                           cell_id   k_cluster leiden_cluster  cell_cycle
#>                       <character> <character>    <character> <character>
#> TGCGGGTGTAGAGTGC TGCGGGTGTAGAGTGC           2              2           S
#> TGGTTCCCATCTATGG TGGTTCCCATCTATGG           5              4           S
#> CATATTCCACGTCAGC CATATTCCACGTCAGC           2              2         G2M
#> TCACAAGTCCTGCTTG TCACAAGTCCTGCTTG           4              3          G1
#> GGAATAATCCAGGGCT GGAATAATCCAGGGCT           5              4          G1
#> ...                           ...         ...            ...         ...
#> CTACGTCCACCACGTG CTACGTCCACCACGTG           5              4          G1
#> TGTGGTAAGGTGCTTT TGTGGTAAGGTGCTTT           6              6          G1
#> AGCGTCGGTCGAGTTT AGCGTCGGTCGAGTTT           2              6          G1
#> CTCGAAAAGTTCGATC CTCGAAAAGTTCGATC           5              5         G2M
#> TGACAACGTCGCATAT TGACAACGTCGCATAT           5              4           S