prcPlot {MetaIntegrator}

Plot the PRC Curve for a Dataset

Description

prcPlot will plot a Precision-Recall curve (and return the AUPRC) that describes how well a gene signature (as defined in a filterObject) classifies groups in a dataset (in the form of a datasetObject).

Usage

prcPlot(filterObject, datasetObject, title = datasetObject$formattedName,
  subtitle = NULL, textSize = NULL, rounding = 3,
  curveColors = "red", legend = TRUE, PRC.lty = 1, PRC.lwd = 1,
  backgroundColor = "gray93", grid.marks = 0.1, grid.color = "white",
  grid.lty = 1, grid.lwd = 0.9, legend.lty = 0, cex.main = 1,
  cex.subtitle = 0.9)

Arguments

filterObject

a metaFilter object containing the signature genes that will be used for calculating the score

datasetObject

a Dataset object for group comparison in the PRC plot. (At least, must have a $expr of probe-level data, $keys of probe:gene mappings, and $class of two-class labels.)

title

title of the plot (default: datasetObject$formattedName)

subtitle

subtitle of the figure

textSize

use this to easily increase or decrease the size of all the text in the plot

rounding

how many digits to round the AUPRC and CI to (default: 3)

curveColors

Graphical: the color for the PRC curves (default: "red")

legend

Graphical: if TRUE, a legend will be included

PRC.lty

Graphical: PRC curve line type

PRC.lwd

Graphical: PRC curve line width

backgroundColor

Graphical: background color of the plot

grid.marks

Graphical: increment between grid lines

grid.color

Graphical: grid line color

grid.lty

Graphical: grid line type

grid.lwd

Graphical: grid line width

legend.lty

Graphical: legend style (0 is no box, 1 is boxed legend)

cex.main

Graphical: title size

cex.subtitle

Graphical: subtitle size

Details

Evaluates the ability of a given gene set to separate two classes. As opposed to ROC curves, PRC curves are more sensitive to class imbalances. The gene set is evaluated as a Z-score of the difference in means between the positive genes and the negative genes (see calculateScore).

Value

Returns a standard PRC plot, plus AUPRC with 95% CI (calculated with the trapezoid method).

Author(s)

Aditya M. Rao, Jiaying Toh

See Also

multiplePRCPlot, rocPlot

Examples

prcPlot(tinyMetaObject$filterResults[[1]], tinyMetaObject$originalData[[1]])