puma (version 3.14.0)

plotROC: Receiver Operator Characteristic (ROC) plot

Description

Plots a Receiver Operator Characteristic (ROC) curve.

Usage

plotROC(
	scoresList
,	truthValues
,	includedProbesets=1:length(truthValues)
,	legendTitles=1:length(scoresList)
,	main = "PUMA ROC plot"
,	lty = 1:length(scoresList)
,	col = rep(1,length(scoresList))
,	lwd = rep(1,length(scoresList))
,	yaxisStat = "tpr"
,	xaxisStat = "fpr"
,	downsampling = 100
,	showLegend = TRUE
,	showAUC = TRUE
,	...
)

Arguments

scoresList

A list, each element of which is a numeric vector of scores.

truthValues

A boolean vector indicating which scores are True Positives.

includedProbesets

A vector of indices indicating which scores (and truthValues) are to be used in the calculation. The default is to use all, but a subset can be used if, for example, you only want a subset of the probesets which are not True Positives to be treated as False Positives.

legendTitles

Vector of names to appear in legend.

main

Main plot title

lty

Line types.

col

Colours.

lwd

Line widths.

yaxisStat

Character string identifying what is to be plotted on the y-axis. The default is "tpr" for True Positive Rate. See performance function from ROCR package.

xaxisStat

Character string identifying what is to be plotted on the x-axis. The default is "fpr" for False Positive Rate. See performance function from ROCR package.

downsampling

See details for plot.performance from the ROCR package.

showLegend

Boolean. Should legend be displayed?

showAUC

Boolean. Should AUC values be included in legend?

...

Other parameters to be passed to plot.

Value

This function has no return value. The output is the plot created.

See Also

Related method calcAUC

Examples

	class1a <- rnorm(1000,0.2,0.1)
	class2a <- rnorm(1000,0.6,0.2)
	class1b <- rnorm(1000,0.3,0.1)
	class2b <- rnorm(1000,0.5,0.2)
	scores_a <- c(class1a, class2a)
	scores_b <- c(class1b, class2b)
	scores <- list(scores_a, scores_b)
	classElts <- c(rep(FALSE,1000), rep(TRUE,1000))
	plotROC(scores, classElts)