validateNetwork: Compare a panda network with a gold standard

In QuackenbushLab/pandaR: PANDA Algorithm

Description

Computes the ROC curve for the panda result, allowing for plotting the against a gold standard. Optionally, allows the inclusion of another network for comparison

Usage

validateNetwork(x, y, ...)

Arguments

x	An object of class "panda" or matrix
y	A gold standard dataset, a data.frame, matrix or exprSet containing 2 or 3 columns. Each row describes an motif associated with a transcription factor (column 1) a gene (column 2) and a score (column 3) for the gold standard score (0 or 1). If column 3 is not given, all rows will be assumed to be gold standard edges.
...	Options

Examples

data(pandaToyData)
pandaRes <- panda(pandaToyData$motif,
           pandaToyData$expression,pandaToyData$ppi,hamming=.001,progress=TRUE)
gold <- melt(pandaRes@regNet)[sample(1000,200),]
gold[,3] <-1
roc <- validateNetwork(pandaRes,gold)
plot(roc)

QuackenbushLab/pandaR documentation built on May 8, 2019, 3:49 a.m.