plotDR: Plot reduced dimensions

In CATALYST: Cytometry dATa anALYSis Tools

Description

Dimension reduction plot colored by expression, cluster, sample or group ID.

Usage

plotDR(
  x,
  dr = NULL,
  color_by = "condition",
  facet_by = NULL,
  ncol = NULL,
  assay = "exprs",
  scale = TRUE,
  q = 0.01,
  dims = c(1, 2),
  k_pal = CATALYST:::.cluster_cols,
  a_pal = hcl.colors(10, "Viridis")
)

Arguments

x	a SingleCellExperiment.
dr	character string specifying which dimension reduction to use. Should be one of reducedDimNames(x); default to the 1st available.
color_by	character string specifying the color coding; valid values are rownames(sce) and names(colData(x)).
facet_by	character string specifying a non-numeric cell metadata column to facet by; valid values are names(colData(x)).
ncol	integer scalar specifying number of facet columns; ignored unless coloring by multiple features without facetting or coloring by a single feature with facetting.
assay	character string specifying which assay data to use when coloring by marker(s); valid values are assayNames(x).
scale	logical specifying whether assay data should be scaled between 0 and 1 using lower (1%) and upper (99%) expression quantiles; ignored if !all(color_by %in% rownames(x)).
q	single numeric in [0,0.5) determining the quantiles to trim when scale = TRUE.
dims	length 2 numeric specifying which dimensions to plot.
k_pal	character string specifying the cluster color palette; ignored when color_by is not one of names(cluster_codes(x)). If less than nlevels(cluster_ids(x, k)) are supplied, colors will be interpolated via colorRampPalette.
a_pal	character string specifying the assay data palette when coloring by feature(s), i.e. all(color_by %in% rownames(x)).

Value

a ggplot object.

Author(s)

Helena L Crowell [e-mail]

References

Nowicka M, Krieg C, Crowell HL, Weber LM et al. CyTOF workflow: Differential discovery in high-throughput high-dimensional cytometry datasets. F1000Research 2017, 6:748 (doi: 10.12688/f1000research.11622.1)

Examples

# construct SCE & run clustering
data(PBMC_fs, PBMC_panel, PBMC_md)
sce <- prepData(PBMC_fs, PBMC_panel, PBMC_md)

# run clustering & dimension reduction
sce <- cluster(sce)
sce <- runDR(sce, dr = "UMAP", cells = 100)

# color by single marker, split by sample
plotDR(sce, color_by = "CD7", facet_by = "sample_id", ncol = 4)

# color by a set of markers using custom color palette
cdx <- grep("CD", rownames(sce), value = TRUE)
plotDR(sce, color_by = cdx, ncol = 4, 
  a_pal = rev(hcl.colors(10, "Spectral")))

# color by scaled expression for 
# set of markers, split by condition
plotDR(sce, 
  scale = TRUE, 
  facet_by = "condition",
  color_by = sample(rownames(sce), 4))

# color by 8 metaclusters using custom 
# cluster color palette, split by sample
p <- plotDR(sce, 
  color_by = "meta8", 
  facet_by = "sample_id", 
  k_pal = c("lightgrey", "cornflowerblue", "navy")) 
p$facet$params$ncol <- 4; p

CATALYST documentation built on Nov. 8, 2020, 6:53 p.m.