Skip to contents

Plot deconvolution results

Source: R/DeconvolutionPlot.R
DeconvolutionPlot.Rd

Plot deconvolution results

Usage

DeconvolutionPlot(
  object = NULL,
  res = NULL,
  plot_type = c("bar", "heatmap", "box"),
  sample_order = NULL,
  cell_type_order = NULL,
  palette = "Chinese",
  palcolor = NULL,
  heatmap_palette = "YlGnBu",
  heatmap_palcolor = NULL,
  sample_annotation = NULL,
  sample_annotation_palette = "Chinese",
  sample_annotation_palcolor = NULL,
  sample_split = NULL,
  cluster_rows = FALSE,
  cluster_columns = FALSE,
  show_row_names = TRUE,
  show_column_names = FALSE,
  theme_use = "theme_scop",
  theme_args = list(),
  legend.position = "right",
  legend.direction = "vertical",
  grid_major = FALSE,
  verbose = TRUE,
  ...
)

Arguments

object

A SummarizedExperiment object containing deconvolution results in metadata(object)[["Deconvolution"]].

res

A deconvolution result data frame. When provided, object is ignored.

plot_type

Plot type. One of "bar", "heatmap", or "box".

sample_order

Optional sample order.

cell_type_order

Optional cell-type order.

palette

Palette used for discrete cell-type colors in "bar" and "box" modes.

palcolor

Optional custom colors for palette.

heatmap_palette

Palette used for continuous proportions in "heatmap" mode.

heatmap_palcolor

Optional custom colors for heatmap_palette.

sample_annotation

Character vector of colData(object) columns used as top annotations in "heatmap" mode.

sample_annotation_palette

Palette(s) used for sample_annotation.

sample_annotation_palcolor

Optional custom colors for sample_annotation_palette. Use a list when multiple annotations are provided.

sample_split

Optional colData(object) column used to split heatmap columns for grouped comparison.

cluster_rows, cluster_columns

Whether to cluster rows or columns in "heatmap" mode.

show_row_names, show_column_names

Whether to show row or column names in "heatmap" mode.

theme_use, theme_args

Theme name or function, plus extra theme arguments.

legend.position

Legend placement ("none", "left", "right", "bottom", "top"), direction, and title. legend.title = NULL uses the group name.

legend.direction

Legend direction: "horizontal" or "vertical".

grid_major

Whether to show major grid lines for "bar" and "box" plots.

verbose

Whether to print the message. Default is TRUE.

...

Reserved for future use.

Value

A ggplot object. For plot_type = "heatmap", returns a ComplexHeatmap::Heatmap object.

See also

RunDeconvolution

Examples

data(islet_bulk)
islet_bulk <- RunDeconvolution(
  islet_bulk,
  method = "CIBERSORT",
  backend = "cpp",
  perm = 0
)
#> ℹ [2026-08-30 04:21:17] Use 400 shared genes for CIBERSORT
DeconvolutionPlot(islet_bulk, plot_type = "bar")


DeconvolutionPlot(islet_bulk, plot_type = "box")


ht <- DeconvolutionPlot(
  islet_bulk,
  plot_type = "heatmap",
  sample_annotation = "condition",
  sample_split = "condition"
)
ComplexHeatmap::draw(ht)