Split Boxplot by Group
Description
Plots a boxplot with each data split into sub-boxplots, defined according a factor variable.
The default method plots each column using one boxplot for each level in a factor.
The method for NMF-class objects plots use
the data from the coefficient matrix.
Usage
boxplotBy(x, ...) S3 (default) `boxplotBy`(x, by, scale = FALSE, col = rainbow(nlevels(by)), legend = TRUE, names = colnames(x), ...) S3 (NMF) `boxplotBy`(x, by, scale = FALSE, ...)
Arguments
- x
- data matrix, with samples in column.
NB: in
boxplotx has samples in rows. - by
- grouping variable, coerced to a factor if necessary. It is recycled if not of length the number of columns in x
- col
- spcecifies the colour to use for each group by argument by.
- legend
- legend specification. If a logical, it
turns on/off the legend, if a character string, it
indicates the position of the legend, e.g.
'topright'(seelegendfor possible values). - ...
- extra arguments passed to the base
boxplotfunction. - scale
- logical that indicate if each data column should be scaled so that it sums up to one.
- names
- group labels which will be printed under each boxplot. Can be a character vector or an expression (see plotmath).
Examples
x <- rmatrix(3, 20)
boxplotBy(x, gl(2, 10))
