High-level function for package
SigCheck that runs a default
set of checks against a predictive signature.
1 2 3
Number of iterations to run to generate background distributions. This is how many random signatures the primary signature will be compared to, as well as how many of each type of permuted dataset will be generated for comparison.
Specification of a set of known (previously identified) signatures to
compare to. See
By default, plots of the results will be generated
unless this is set or
Extra parameters to pass through
This high-level function will run four checks, plot the results, and return a consolidated result set.
First, it calls
sigCheckRandom to compare the performance
interations randomly selected signatures.
Next, it calls
sigCheckKnown to check the
performance of the signature against a database of signatures previously
identified to discriminate in other domains.
Finally, two calls are made to
sigCheckPermuted to check the
performance of randomly permuted metadata and expression data.
The first call permutes the survival data if they
are available (
otherwise it permutes the
category assignments (
The second call permuted the expression value for each gene (permuting
each row in the
TRUE, the results are plotted. If a classifier
is involved, a set of four classification results are plotted in
a 2x2 grid, showing how the classification performance of the main
signature compares to that of a mode classifier and to the
distribution of performance values observed for
the random and known signature sets, as well as how it performs using the two
type of permuted dataset.
If survival data is available, another 2x2 grid is plotted showing how the
baseline survival p-value compares to a p-value of 0.05 and to the
distribution of p-values observed for the random and known signatures, as well
as for the permuted data.
A list containing four elements, each containing the result of a check.
$checkRandom is the result list returned by
$checkKnown is the result list returned by
The third element of the result list will be one of the following:
$checkPermutedSurvival is the result list returned by
$checkPermutedCategories is the result list returned by
The fourth element of the list will be:
$checkPermutedFeatures is the result list returned by
Venet, David, Jacques E. Dumont, and Vincent Detours. "Most random gene expression signatures are significantly associated with breast cancer outcome." PLoS Computational Biology 7.10 (2011): e1002240.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
#Disable parallel so Bioconductor build won't hang library(BiocParallel) register(SerialParam()) library(breastCancerNKI) data(nki) nki <- nki[,!is.na(nki$e.dmfs)] data(knownSignatures) ITERATIONS <- 5 # should be at least 20, 1000 for real checks ## survival analysis check <- sigCheck(nki, classes="e.dmfs", survival="t.dmfs", signature=knownSignatures$cancer$VANTVEER, annotation="HUGO.gene.symbol", validationSamples=150:319) results <- sigCheckAll(check,iterations=ITERATIONS, known=knownSignatures$cancer[1:20]) ## classification analysis check <- sigCheck(nki, classes="e.dmfs", signature=knownSignatures$cancer$VANTVEER, annotation="HUGO.gene.symbol", validationSamples=275:319, scoreMethod="classifier") results <- sigCheckAll(check,iterations=ITERATIONS, known=knownSignatures$cancer[1:20])
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.