vulcan: VULCAN - VirtUaL Chipseq data Analysis using Networks

In federicogiorgi/vulcan: VirtUaL ChIP-Seq data Analysis using Networks

Description

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

Usage

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

Arguments

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

Value

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

Examples

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)

federicogiorgi/vulcan documentation built on May 27, 2019, 7:25 a.m.