iplot: Visualize gene level behavior of genes within a geneset...

In lianos/multiGSEA: A unified interface to a plethora of gene set enrichment analysis methods

Description

It is informative to look at the individual log fold changes of the genes within a gene set to explore the degree to which they (1) are coherent with respect to each other; and (2) see how the compare to the background distribution of log fold changes of the entire transcriptome.

You can visualize this behavior via a type = "density" plot, or a type = "boxplot". It is also common to plot either the individual log fold changes value = "logFC"or t-statisticsvalue = "t"'.

Usage

iplot(
  x,
  y,
  j,
  value = "logFC",
  type = c("density", "gsea", "boxplot"),
  tools = c("wheel_zoom", "box_select", "reset", "save"),
  main = NULL,
  with.legend = TRUE,
  shiny_source = "mggenes",
  width = NULL,
  height = NULL,
  ggtheme = theme_bw(),
  trim = 0.005,
  ...
)

Arguments

x	A MultiGSEAResult() object
y	the name of the gene set collection
j	the name of the gene set name
value	A string indicating the column name for the value of the gene-level metadata to plot. Default is "logFC". Anoter often used choice might also be "t", to plot t-statistics (if they're in the result). But this can be any numeric column found in the data.frame returned by geneSet(x, y, j). If this is a named string (vector), then the value in names(value) will be used on the axis when plotted.
type	plot the distributions as a "density" plot or "boxplot".
tools	the tools to display in the rbokeh plot
main	A title to display. If not specified, the gene set name will be used, otherwise you can pass in a custom title, or NULL will disable the title altogether.
with.legend	Draws a legend to map point color to meaning. There are three levels a point (gene level statistic) can be color as, "notsig", "psig", and "sig". "notsig" implies that the FDR >= 10%, "psig" means that FDR <= 10%, but the logFC is "unremarkable" (< 1), and "sig" means that both the FDR <= 10% and the logFC >= 1

Value

the ploty plot ojbect

Examples

mgr <- exampleMultiGSEAResult()
iplot(mgr, "c2", "BURTON_ADIPOGENESIS_PEAK_AT_2HR", c("t-statistic" = "t"),
      type = "density")
iplot(mgr, "c2", "BURTON_ADIPOGENESIS_PEAK_AT_2HR", c("t-statistic" = "t"),
      type = "gsea")

lianos/multiGSEA documentation built on Nov. 17, 2020, 1:26 p.m.