PlotROCTest: Plot ROC for the ROC Curve Based Model Creation and...

In xia-lab/MetaboAnalystR: An R Package for Comprehensive Analysis of Metabolomics Data

Plot ROC for the ROC Curve Based Model Creation and Evaluation module

Description

Plot the ROC curve of the biomarker model created using a user-selected subset of features. Pred and auroc are lists containing predictions and labels from different cross-validations.

Usage

PlotROCTest(mSetObj=NA, imgName, format="png", 
dpi=72, mdl.inx, avg.method, show.conf, show.holdout, focus="fpr", cutoff = 1.0)

Arguments

mSetObj	Input the name of the created mSetObj (see InitDataObjects)
imgName	Input a name for the plot
format	Select the image format, "png", of "pdf".
dpi	Input the dpi. If the image format is "pdf", users need not define the dpi. For "png" images, the default dpi is 72. It is suggested that for high-resolution images, select a dpi of 300.
mdl.inx	Model index, 0 means to compare all models, input 1-6 to plot a ROC curve for one of the top six models
avg.method	Input the method to compute the average ROC curve, either "threshold", "vertical" or "horizontal"
show.conf	Logical, if 1, show confidence interval, if 0 do not show
show.holdout	Logical, if 1, show the ROC curve for hold-out validation, if 0 do not show
focus	"fpr"
cutoff	Input the threshold to limit the calculation of the ROC curve, the number must be between 0 and 1.

Author(s)

Jeff Xia [e-mail] McGill University, Canada License: GNU GPL (>= 2)

xia-lab/MetaboAnalystR documentation built on April 15, 2024, 12:16 p.m.