Visualize column after dimension reduction — dimension_reduction-ConsensusPartition-method • cola

Visualize samples (the matrix columns) after dimension reduction

# S4 method for ConsensusPartition
dimension_reduction(object, k, top_n = NULL,
    method = c("PCA", "MDS", "t-SNE", "UMAP"),
    control = list(), color_by = NULL,
    internal = FALSE, nr = 5000,
    silhouette_cutoff = 0.5, remove = FALSE,
    scale_rows = object@scale_rows, verbose = TRUE, ...)

Arguments

object: A ConsensusPartition-class object.
k: Number of subgroups.
top_n: Top n rows to use. By default it uses all rows in the original matrix.
method: Which method to reduce the dimension of the data. MDS uses cmdscale, PCA uses prcomp. t-SNE uses Rtsne. UMAP uses umap.
color_by: If annotation table is set, an annotation name can be set here.
control: A list of parameters for Rtsne or umap.
internal: Internally used.
nr: If number of matrix rows is larger than this value, random nr rows are used.
silhouette_cutoff: Cutoff of silhouette score. Data points with values less than it will be mapped with cross symbols.
remove: Whether to remove columns which have less silhouette scores than the cutoff.
scale_rows: Whether to perform scaling on matrix rows.
verbose: Whether print messages.
...: Pass to dimension_reduction,matrix-method.

Value

Locations of the points.

Author

Zuguang Gu <z.gu@dkfz.de>

Examples

data(golub_cola)
dimension_reduction(golub_cola["ATC", "skmeans"], k = 3)
#> use UMAP