plotReducedDims: Plot 2-dimensionsal representation with clusters

In epurdom/clusterExperiment: Compare Clusterings for Single-Cell Sequencing

Plot 2-dimensionsal representation with clusters

Description

Plot a 2-dimensional representation of the data, color-code by a clustering.

Usage

## S4 method for signature 'ClusterExperiment'
plotReducedDims(
  object,
  whichCluster = "primary",
  reducedDim = "PCA",
  whichDims = c(1, 2),
  plotUnassigned = TRUE,
  legend = TRUE,
  legendTitle = "",
  nColLegend = 6,
  clusterLegend = NULL,
  unassignedColor = NULL,
  missingColor = NULL,
  pch = 19,
  xlab = NULL,
  ylab = NULL,
  ...
)

Arguments

object	a ClusterExperiment object
whichCluster	argument that can be a single numeric or character value indicating the single clustering to be used. Giving values that result in more than one clustering will result in an error. See details of getClusterIndex.
reducedDim	What dimensionality reduction method to use. Should match either a value in reducedDimNames(object) or one of the built-in functions of listBuiltInReducedDims()
whichDims	vector of length 2 giving the indices of which dimensions to show. The first value goes on the x-axis and the second on the y-axis.
plotUnassigned	logical as to whether unassigned (either -1 or -2 cluster values) should be plotted.
legend	either logical, indicating whether to plot legend, or character giving the location of the legend (passed to legend)
legendTitle	character value giving title for the legend. If NULL, uses the clusterLabels value for clustering.
nColLegend	The number of columns in legend. If missing, picks number of columns internally.
clusterLegend	matrix with three columns and colnames 'clusterIds','name', and 'color' that give the color and name of the clusters in whichCluster. If NULL, pulls the information from object.
unassignedColor	If not NULL, should be character value giving the color for unassigned (-1) samples (overrides clusterLegend) default.
missingColor	If not NULL, should be character value giving the color for missing (-2) samples (overrides clusterLegend) default.
pch	the point type, passed to plot.default
xlab	Label for x axis
ylab	Label for y axis
...	arguments passed to plot.default

Details

If plotUnassigned=TRUE, and the color for -1 or -2 is set to "white", will be coerced to "lightgrey" regardless of user input to missingColor and unassignedColor. If plotUnassigned=FALSE, the samples with -1/-2 will not be plotted, nor will the category show up in the legend.

If the requested reducedDim method has not been created yet, the function will call makeReducedDims on the FIRST assay of x. The results of this method will be saved as part of the object and returned INVISIBLY (meaning if you don't save the output of the plotting command, the results will vanish). To pick another assay, you should call 'makeReducedDims' directly and specify the assay.

Value

A plot is created. Nothing is returned.

See Also

plot.default, makeReducedDims, listBuiltInReducedDims()

Examples

#clustering using pam: try using different dimensions of pca and different k
data(simData)

cl <- clusterMany(simData, nReducedDims=c(5, 10, 50), reducedDim="PCA",
clusterFunction="pam", ks=2:4, findBestK=c(TRUE,FALSE),
removeSil=c(TRUE,FALSE), makeMissingDiss=TRUE)

plotReducedDims(cl,legend="bottomright")

epurdom/clusterExperiment documentation built on April 23, 2024, 9:09 p.m.