waterfall: Waterfall plot coloured by discrete categories

In tobyjohnson/gtx: Genetics ToolboX

Description

Generates a waterfall plot, which summarises the distribution of a continuous variable (such as a quantitative trait) by discrete categories (such as genotype).

Usage

## S3 method for class 'character'
waterfall(object, x, data, ylab, ylim, bylevel, col, ...)
## S3 method for class 'numeric'
waterfall(object, x, data, ylab, ylim, bylevel, col, ...)

Arguments

object	The name or values of the continuous variable
x	The name or values of the discrete category variable
data	A data frame containing values
ylab	A label for the y axis
ylim	Limits for the y axis
bylevel	Whether to group subjects by levels of x
col	A vector of colours
...	Other arguments

Details

Waterfall plots are widely used in clinical oncology, to display the distribution of a specific quantitative trait: best on-study change in tumour burden, relative to baseline. The plot is typically coloured by levels of a factor, such as treatment arm or categorical best response. In principle the plot could be used for any quantitative trait, but the display is most visually appealing when the trait takes both positive and negative values.

Value

Returns an invisible null. The plot is generated as a side effect.

Author(s)

Toby Johnson [e-mail]

Examples

data(aoex1)
par(mfrow = c(1, 2))
with(aoex1, waterfall(bmi-mean(bmi, na.rm = TRUE), rs123456, bylevel = TRUE))
with(aoex1, waterfall(bmi-mean(bmi, na.rm = TRUE), rs123456, bylevel = FALSE))

tobyjohnson/gtx documentation built on Aug. 30, 2019, 8:07 p.m.