plotCategoryBar: Plot horizontal bars for category enrichment analysis results

In jotsetung/unsoRted: Unsorted R functions and utilities

Description

These functions draw horizontal bars that allow to visualize category enrichment results.

Usage

plotCategoryBar( x=0, y=0, width=1, size, count=NULL,
subcount=NULL, col.size="lightgrey", border.size="grey",
col.count=brewer.pal( 9, "Greys" )[ 7 ], border.count=col.count,
col.subcount=NULL, border.subcount, add=FALSE, xlab="count", ylab="",
legend=FALSE, xlim, cex=par( "cex" ), ... )

plotCategoryBars( x=0, y=0, width=1, size, count=NULL,
subcount=NULL, col.size="lightgrey", border.size="grey",
col.count=brewer.pal( 9 , "Greys" )[ 7 ], border.count=col.count,
col.subcount=NULL, border.subcount, add=FALSE, xlab="count", ylab="",
legend=FALSE, asc=TRUE, spacer=width/10, xlim, cex.legend=par( "cex" ), cex=par( "cex" ), ... )

Arguments

x	The x-coordinate of the rectangle's left side. For plotCategoryBars, x and y specify the position of the bottom rectangle.
y	The y-coordinate of the rectangle's left side (middle position).
width	The width of the rectangle (rectangle will be plotted at y-width/2 to y+width/2 on the y-axis).
size	the total number of counts (genes) in the category. For plotCategoryBar a single value, for plotCategoryBars a numeric vector.
count	the count of significant genes in the category. This can also be a (ideally named) vector of counts, in which case separate rectangles, one next to the other on the x-axis, will be plotted for each count. For plotCategoryBars, to achieve the same (latter) behavious, a list of (named) numerical vectors has to be submitted.
subcount	for plotCategoryBar: a numerical vector (ideally named) if sub-counts. For plotCategoryBars: a list of named numerical vectors. In contrast to the above parameter, these counts will be represented by vertically splitted rectangles instead of rectangles one next to the other along the x-axis. See below for examples.
col.size	the color for the (background) rectangle.
border.size	the border color for the (background) rectangle.
col.count	the color(s) for the significant count rectangle(s).
border.count	the border color for the significant count rectangle.
col.subcount	the colors for the subcount rectangles. The colors have to be named, and the names have to match the names of the subcounts.
border.subcount	the border color.
add	whether the plot should be added to an existing plot.
xlab	the label for the x-axis.
ylab	the label for the y-axis.
legend	whether a legend should be drawn.
asc	if the bars should be plotted in ascending order, i.e. the first element at the top of the plot.
spacer	the spacing between the rectangles on the y-axis.
xlim	the xlim for the plot region. Rectangles will be cropped if size is larger than xlim[2] with the actual count of the size being plotted into the rectangle.
cex	graphical parameter for the plot.
cex.legend	allows to adjust in addition the size of the legend.
...	additional parameters to be passed to the plot function.

Details

The names of the submitted numerical values are required to match the counts to the colors used for the rectangles. For plotCategoryBars the names of the size parameter will be used for the y-axis.

Value

A plot.

Author(s)

Johannes Rainer

References

http://bioinfo.i-med.ac.at

Examples

## Classical view of a Category enrichment: plot size and count.
plotCategoryBar( 0, 0, size=10, count=2, legend=TRUE )
## If count can be further subdivided, plot a rectangle for each count.
plotCategoryBar( 0, 0, size=10, count=c( first=2, second=5 ), legend=TRUE)
## multiple bars: example would be the results of a category enrichment analysis for target
## genes of a set of microRNAs. In such a case, a gene could be target of more then one
## miRNA, thus, separate bars, placed horizontally next to each other would not be correct,
## since the total count of target genes would exceed the unique number of target genes.
## defining the number of target genes for 3 miRNAs in the category.
mirna.counts <- c( mir2=1, mir5=4, mir6=5)
plotCategoryBar( 0, 0, size=10, count=7, subcount=mirna.counts, legend=TRUE )
plotCategoryBar( 0, 0, size=10, count=c(a=7, b=2), subcount=mirna.counts, legend=TRUE )

## multiple stuff...
Sizes <- c( small=10, large=123, medium_category=69 )
## simple bars
plotCategoryBars( size=Sizes )
Counts <- c( 9, 25, 30 )
plotCategoryBars( size=Sizes, count=Counts, legend=TRUE )
## multiple counts per category.
Counts2 <- list( c( a=1, b=4 ), c( c=4, a=9 ), c( a=12, e=4, f=2 ) )
plotCategoryBars( size=Sizes, count=Counts2, legend=TRUE )
## that's now the full thing. Again on an example of miRNA target genes...
mirna.counts <- list( c( mir2=1, mir5=4, mir6=5 ),
                     c( mir3=12, mir6=3, mir7=23, mir9=3 ),
                     c( mir2=8, mir9=14 ) )
plotCategoryBars( size=Sizes, count=Counts, subcount=mirna.counts, legend=TRUE )

jotsetung/unsoRted documentation built on May 19, 2019, 9:42 p.m.