computeAuROC | R Documentation |
This function performs the same rewiring simulation as netSimAndEval
. But computes the AUC and ROC as the result
computeAuROC(g,nDup,dDup,rewProb,simMethods,steps = 2, nThreads=1) ## S3 method for class 'CoReg.auROC' print(CoReg.auROC) ## S3 method for class 'CoReg.auROC' summary(CoReg.auROC) ## S3 method for class 'CoReg.auROC' plot(CoReg.auROC)
g |
The input transcription network. This should be an "igraph" object returned by the function |
nDup |
numeric value specifying number of the genes to be duplicated. These duplicated genes will be rewired |
dDup |
numeric value specifying the minimum degree for the genes to be duplicated. This function first ranks the genes based on their degrees. Then randomly selected |
rewProb |
a numeric value specifying the rewiring probability |
simMethods |
similarity indices used. Available options are: "jaccard" (jaccard similarity index),"geometric" (geometric similarity index),"invlogweighted" (inverse log weighted similarity index), "wt" (walk-trap-based similarity index) |
steps |
Number of steps for random walks, which is used by walk trap algorithm. Only valid when |
nThreads |
number of threads for running the simulation. Only valid when |
CoReg.auROC |
|
This function performes the rewring simulation similar to netSimAndEval
, but uses the AUC values and ROC curves as the evaluation metrics. ROC curves can be plotted using the generic function plot()
.
An object of class CoReg.auROC
Qi Song
rewSim
data(athNet) re <- computeAuROC(athNet,nDup=50,dDup=10,rewProb=0.5,simMethods=c("jaccard","geometric","invlogweighted","wt"),steps = 2, nThreads=1) # Display summary information for rewiring simulation print(re) # Plot ROC curves for the result plot(re) # Get the AUC values for the result re$AUC
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.