R Graphical Manual

Browse All

Last data update: 2014.03.03

R: Hypergeometric tests for pathway analysis

Arora2010.hypergeo

R Documentation

Hypergeometric tests for pathway analysis

Description

The function is used to perform hypergeometric tests of overrepresentation of KEGG pathways in inferred functional modules in the application to Ewing's sarcoma.

Usage

Arora2010.hypergeo(pan, mod.pval.cutoff=0.05, mod.size.cutoff=4, 
avg.degree.cutoff=0.5, filter.effects=TRUE)

Arguments

`pan`	an object of class `PAN`, which includes the inferred posterior association network and functional modules.
`mod.pval.cutoff`	a numeric value specifying the p-value cutoff for selecting significant functional modules.
`mod.size.cutoff`	an integer value specifying the minimal size of functional modules.
`avg.degree.cutoff`	a numeric value specifying the cutoff of module density, which is the ratio of predicted significant associations to all possible associations.
`filter.effects`	a logical value specifying whether or not to filter out modules that are not of interest. In this application to Ewing's sarcoma, modules associated with positive loss-of-function can be filtered out because we are only interested in modules related to inhibiting growth and proliferation of cancer cells.

Details

The function is used to do pathway analyses in the application of PANs to Ewing's sarcoma. Hypergeometric tests are performed to test the overrepresentation of KEGG pathways in inferred functional modules. Four arguments can be specified to filter modules. More details about module filtering procedures can be found in the Methods section of Wang X et al., 2012.

Value

a list including hypergeometric test result for each module.

Author(s)

Xin Wang xw264@cam.ac.uk

References

Arora S, Gonzales IM, Hagelstrom RT, et al. (2010). RNAi phenotype profiling of kinases identifies potential therapeutic targets in Ewing's sarcoma. Molecular Cancer, 9(1), 218.

Wang X, Castro MA, Mulder KW and Markowetz F (2012), Posterior association networks and enriched functional gene modules inferred from rich phenotypic perturbation screens, PLoS Computational Biology, doi:10.1371/journal.pcbi.1002566.

Examples

## Not run: 
data(pan_Arora2010, package="Mulder2012")
Arora2010.hypergeo(pan_Arora2010, mod.pval.cutoff=0.05, mod.size.cutoff=4, 
avg.degree.cutoff=0.5, filter.effects=TRUE)

## End(Not run)