cellHTS (version 1.42.0)

ROC: Creates an object of class "ROC" which can be plotted as a ROC curve

Description

The function ROC construct an object of S3 class ROC, which represents a receiver-operator-characteristic curve, from the data of the annotated positive and negative controls in a scored cellHTS object.

Usage

ROC(x, positives, negatives)
"plot"(x, col="darkblue", type="l", main = "ROC curve", ...)
"lines"(x, ...)

Arguments

x

a cellHTS object that has already been scored (see details).

positives

a list or vector of regular expressions specifying the name of the positive controls. See the details for the argument posControls of writeReport function.

negatives

a vector of regular expressions specifying the name of the negative controls. See the details for the argument negControls of writeReport function.

col

the graphical parameter for color; see par for details.

type

the graphical parameter giving the type of plot desired; see par for details.

main

the graphical parameter giving the desired title of plot; see par for details.

...

other graphical parameters as in par may be also passed as arguments.

Value

ROC. There are methods plot.ROC and lines.ROC.

Details

The cellHTS object x must contain a slot called score, and selection proceeds from large to small values of this score. Furthermore, x is expected to contain positive and negative controls annotated in the slot wellAnno with the values of the arguments positives and negatives, respectively. If the assay is a two-way experiment, positives should be a list with components act and inh, specifying the name of the activators, and inhibitors, respectively. In this case, the ROC cureve is constructed based on the absolute values of x$score.

Examples

    data(KcViabSmall)
    ## Not run: 
#     x <- normalizePlates(KcViabSmall, normalizationMethod="median", zscore="-")
#     x <- summarizeReplicates(x)
#     y <- ROC(x)
#     plot(y)
#     lines(y)
#     ## End(Not run)