vulcan: VULCAN - VirtUaL Chipseq data Analysis using Networks
In vulcan: VirtUaL ChIP-Seq data Analysis using Networks

Description Usage Arguments Value Examples

This function calculates the enrichment of a gene regulatory network over a ChIP-Seq derived signature

1	vulcan(vobj, network, contrast, annotation = NULL, minsize = 10)

`vobj`	a list, the output of the `'vulcan.normalize'` function
`network`	an object of class `'viper::regulon'`
`contrast`	a vector of two fields, containing the condition names to be compared (1 vs 2)
`annotation`	an optional named vector to convert gene identifiers (e.g. entrez ids to gene symbols) Default (NULL) won't convert gene names.
`minsize`	integer indicating the minimum regulon size for the analysis to be run. Default: 10

A list of components:

peakcounts: A matrix of raw peak counts, peaks as rows, samples as columns
peakrpkms: A matrix of peak RPKMs, peaks as rows, samples as columns
rawcounts: A matrix of raw gene counts, genes as rows, samples as columns. The counts are associated to the promoter region of the gene
rpkms: A matrix of RPKMs, genes as rows, samples as columns. The RPKMs are associated to the promoter region of the gene
normalized: A matrix of gene abundances normalized by Variance-Stabilizing Transformation (VST), genes as rows, samples as columns. The abundances are associated to the promoter region of the gene
samples: A vector of sample names and conditions
msviper: a multisample virtual proteomics object from the viper package
mrs: A table of master regulators for a specific signature, indicating their Normalized Enrichment Score (NES) and p-value

library(vulcandata)
# Get an example vulcan object (generated with vulcan.import() using the
# dummy dataset contained in the \textit{vulcandata} package)
vobj<-vulcandata::vulcanexample()
# Annotate peaks to gene names
vobj<-vulcan.annotate(vobj,lborder=-10000,rborder=10000,method='sum')
# Normalize data for VULCAN analysis
vobj<-vulcan.normalize(vobj)
# Detect which conditions are present
names(vobj$samples)

# Load an ARACNe network
# This is a regulon object as specified in the VIPER package, named 'network'
load(system.file('extdata','network.rda',package='vulcandata',mustWork=TRUE))
# Run VULCAN
# We can reduce the minimum regulon size, since in this example only one
# chromosome
# was measured, and the networks would otherwise have too few hits
vobj_analysis<-vulcan(vobj,network=network,contrast=c('t90','t0'),minsize=5)
# Visualize output using the msviper plotting function
plot(vobj_analysis$msviper,mrs=7)