testSubnet: Test the significance of high-scoring subnetworks found using...

In signet: signet: Selection Inference in Gene NETworks

Description

Test the significance of high-scoring subnetworks found using simulated annealing.

Usage

testSubnet(sigObj, null)

Arguments

sigObj	A list of signet objects obtained using the searchSubnet function.
null	Vector of null subnetwork scores generated using the nullDist function.

Value

For each signet object, a p-value is computed given the provided emnpirical null distribution.

Examples

# Get KEGG pathways from the package graphite:
# library(graphite)
# kegg <- pathways("hsapiens", "kegg")
# kegg_human <- lapply(kegg, pathwayGraph)

data(daub13) # load the gene scores from Daub et al. (2013)

#run the search in all the pathways with 2500 iterations (default)
example <- searchSubnet(kegg_human, scores)

# generate the null distribution (here, only 5 values, but
# at least 1000 are advised)
null <- nullDist(kegg_human, scores, n = 5)
example <- testSubnet(example, null) #now, 'example' includes p-values
summary(example)

signet documentation built on April 28, 2020, 7:54 p.m.