Skip to contents

SEM Graph with wilcox test on single cell level

Source: R/DO.Barplot.R
DO.Barplot.Rd

Perform SEM-based graphs with Wilcox test on single-cell level for Seurat and SCE objects. Calculates mean expression values and SEM for the selected feature, and visualizes them. Performs pairwise Wilcox tests comparing conditions, with optional custom control condition and clustering. Optionally returns a summary data frame, statistical test results, and the generated plot.

Usage

DO.Barplot(
  sce_object,
  Feature,
  ListTest = NULL,
  returnValues = FALSE,
  ctrl.condition = NULL,
  group.by = "condition",
  test_use = "wilcox",
  correction_method = "fdr",
  p_values = NULL,
  bar_colours = NULL,
  stat_pos_mod = 1.15,
  step_mod = 0.2,
  x_label_rotation = 45,
  plot_raw_pvalue = FALSE,
  y_limits = NULL,
  log1p_nUMI = TRUE,
  random_seed = 42
)

Arguments

sce_object

combined SCE object or Seurat

Feature

name of the feature/gene

ListTest

List for which conditions wilcoxon test will be performed, if NULL always CTRL group against everything

returnValues

return data frames needed for the plot, containing df.melt, df.melt.sum, df.melt.orig and wilcoxstats

ctrl.condition

set your ctrl condition, relevant if running with empty comparison List

group.by

select the seurat object slot where your conditions can be found, default conditon

test_use

perform one of c( "wilcox", "wilcox_limma", "bimod", "t", "negbinom", "poisson", "LR", "MAST", "DESeq2", "none" ). default "wilcox"

correction_method

correction for p-value calculation. One of c("BH", "bonferroni", "holm", "BY", "fdr", "none")

p_values

Manually providing p-values for plotting, be aware of group size and if necessary make your test return the same amount of values

bar_colours

colour vector

stat_pos_mod

Defines the distance to the graphs of the statistic

step_mod

Defines the distance between each statistics bracket

x_label_rotation

Rotation of x-labels

plot_raw_pvalue

plot the non adjusted p-value without correcting for multiple tests

y_limits

set limits for y-axis

log1p_nUMI

If nUMIs should be log1p transformed

random_seed

parameter for random state initialisation

Value

a ggplot or a list with plot and data frame

Author

Mariano Ruz Jurado

Examples

sce_data <-
    readRDS(system.file("extdata", "sce_data.rds", package = "DOtools"))

ListTest <- list()
ListTest[[1]] <- c("healthy", "disease")

DO.Barplot(
    sce_object = sce_data,
    Feature = "NKG7",
    test_use = "wilcox",
    correction_method="fdr",
    ListTest = ListTest,
    ctrl.condition = "healthy",
    group.by = "condition"
)
#> Using condition, orig.ident as id variables
#> For a (much!) faster implementation of the Wilcoxon Rank Sum Test,
#> (default method for FindMarkers) please install the presto package
#> --------------------------------------------
#> install.packages('devtools')
#> devtools::install_github('immunogenomics/presto')
#> --------------------------------------------
#> After installation of presto, Seurat will automatically use the more 
#> efficient implementation (no further action necessary).
#> This message will be shown once per session