Plots results of a signature check, as returned by
The list returned by
If a classifier was used in the original call to
Title string for plot. If missing, a default plot title will be generated.
Additional arguments to be passed to the
For results based on survival analysis, the background distribution of p-values (in
derived from the check (either random signatures, known signatures, or performance
using permuted data) is plotted. Up to two vertical red lines are also plotted: a solid red line representing the performance of the primary signature/data,
and a dotted red line representing a p-value of 0.05.
One or both of these may be missing if their performance falls outside the range
of the background distributions.
For results based on classification performance, the x-axis represents the range of classification performance scores computed in the check, and the y-axis representing how many times that score was obtained. In addition, vertical lines are plotted representing the classification performance of the originally specified signature (solid red line) and the performance of a classifier that always predicts the mode value of the training samples (dotted red line).
If the results of
sigCheckAll is passed in, all four results
plots are generated in a 2x2 grid.
Better formance of the signature being checked results in the solid red line being to the right of the background distribution. For survival results, this indicates a lower-value. For classification results,this indicates superior classification performance.
Rory Stark with Justin Norden
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
#Disable parallel so Bioconductor build won't hang library(BiocParallel) register(SerialParam()) library(breastCancerNKI) data(nki) nki <- nki[,!is.na(nki$e.dmfs)] data(knownSignatures) ITERATIONS <- 5 # should be at least 1000 for real checks ## survival analysis with separate training and validation using SVM check <- sigCheck(nki, classes="e.dmfs", survival="t.dmfs", signature=knownSignatures$cancer$VANTVEER, annotation="HUGO.gene.symbol", validationSamples=250:319, scoreMethod="classifier", threshold=.33) results <- sigCheckRandom(check,iterations=ITERATIONS) par(mfrow=c(1,2)) sigCheckPlot(results) sigCheckPlot(results, classifier=TRUE)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.