plotInfReps: Plot inferential replicates for a gene or transcript

In fishpond: Fishpond: differential transcript and gene expression with inferential replicates

Description

For datasets with inferential replicates, boxplots are drawn for the two groups and potentially grouped by covariates. For datasets with only mean and variance, points and intervals (95 approximation) are drawn.

Usage

plotInfReps(
  y,
  idx,
  x,
  cov = NULL,
  colsDrk = c("dodgerblue", "goldenrod4", "royalblue4", "red3", "purple4", "darkgreen"),
  colsLgt = c("lightblue1", "goldenrod1", "royalblue1", "salmon1", "orchid1",
    "limegreen"),
  xaxis,
  xlab,
  ylim,
  main,
  mainCol,
  legend = FALSE,
  legendPos = "topleft",
  legendTitle = FALSE,
  legendCex = 1,
  useMean = TRUE,
  applySF = FALSE,
  reorder,
  thin
)

Arguments

y	a SummarizedExperiment (see swish)
idx	the name or row number of the gene or transcript
x	the name of the condition variable for splitting and coloring the samples or cells. Also can be a numeric, e.g. pseudotime, in which case, cov can be used to designate groups for coloring
cov	the name of the covariate for adjustment
colsDrk	dark colors for the lines of the boxes
colsLgt	light colors for the inside of the boxes
xaxis	logical, whether to label the sample numbers. default is TRUE if there are less than 30 samples
xlab	the x-axis label
ylim	y limits
main	title
mainCol	name of metadata column to use for title (instead of rowname)
legend	logical, show simple legend (default FALSE)
legendPos	character, position of the legend (default "topleft")
legendTitle	logical, whether to add the name of the grouping variable as a title on the legend (default FALSE)
legendCex	numeric, size of the legend (default 1)
useMean	logical, when inferential replicates are not present, use the mean assay or the counts assay for plotting
applySF	logical, when inferential replicates are not present, should y$sizeFactor be divided out from the mean and interval plots (default FALSE)
reorder	logical, should points within a group defined by condition and covariate be re-ordered by their count value (default is FALSE, except for alevin data)
thin	integer, should the mean and interval lines be drawn thin (the default switches from 0 [not thin] to 1 [thinner] at n=150 cells, and from 1 [thinner] to 2 [thinnest] at n=400 cells)

Value

nothing, a plot is displayed

Examples

y <- makeSimSwishData()
plotInfReps(y, 3, "condition")

y <- makeSimSwishData(n=40)
y$batch <- factor(rep(c(1,2,3,1,2,3),c(5,10,5,5,10,5)))
plotInfReps(y, 3, "condition", "batch")

fishpond documentation built on Nov. 8, 2020, 7:54 p.m.