Plot Dispersion Versus Mean Depth

Plot dispersion trend and estimates for each cell type

Usage

# S4 method for class 'lucida'
plotDispEsts(object, cluster_ids = names(object), minDispersion = 0.001, ...)

Arguments

object

lucida object

cluster_ids

which cell types to plot

minDispersion

minimum dispersion value plotted

...

arguments passed to facet_wrap()

Value

ggplot2 plot of 1/theta versus mean normalized counts for each for each gene and each cell type. Original values are shown in black, the trend line is shown in red, values after dispersion shrinkage are shown in blue. Grey points indicate dispersion outliers excluded from the trend.

Examples

library(SingleCellExperiment)

# Load example data
data(example_sce, package="muscat")
sce <- example_sce

# Compute library size for each cell
sce$libSize <- colSums(counts(sce))

# Specify regression formula and cell annotation 
form <- ~ group_id 
fit <- lucida(sce, form, "cluster_id", verbose=FALSE)
#> B cells 
#> CD14+ Monocytes 
#> CD4 T cells 
#> CD8 T cells 
#> FCGR3A+ Monocytes 

# Plot the dispersion trend
plotDispEsts(fit)