sigModules: Retrieve ids for significant gene modules searched by pvclust

In PANR: Posterior association networks and functional modules inferred from rich phenotypes of gene perturbations

Description

The function retrieve ids for significant gene modules that satisfy the given p-value cutoff and module size range.

Usage

sigModules(object, pValCutoff=0.01, minSize=3, maxSize=100, sortby="size", 
decreasing=FALSE, ...)

Arguments

object	an object of S4 class PAN.
pValCutoff	a numeric value specifying the p-value cutoff to tell the significance of a gene module.
minSize	a numeric or integer value giving the minimal size of gene modules.
maxSize	a numeric or integer value giving the maximal size of gene modules.
sortby	a character value specitying how to sort the list of gene modules: by ‘size’ (module size) or ‘pval’ (pvclust p-value).
decreasing	a logical value specifying whether or not the sorting will be conducted decreasingly.
...	not in use, but only for further extension.

Details

This function facilitates the user to retrieve significant gene modules found by pvclust and obtain their ids, which can be subsequently used for visualization (see viewPAN for details).

Value

a numeric vector of ids for significant gene modules

Author(s)

Xin Wang [e-mail]

References

Xin Wang, Mauro Castro, Klaas W. Mulder and Florian Markowetz, Posterior association networks and enriched functional gene modules inferred from rich phenotypic perturbation screens, in preparation.

R. Suzuki and H. Shimodaira. Pvclust: an r package for assessing the uncertainty in hierarchical clustering. Bioinformatics, 22(12):1540, 2006.

Examples

## Not run: 
data(bm, package="PANR")
pan<-new("PAN", bm1=bm1)
pan<-infer(pan, para=list(type="SNR", log=TRUE, sign=TRUE, cutoff=log(5)),
filter=FALSE, verbose=TRUE)
data(Bakal2007Cluster, package="PANR")
pan<-buildPAN(pan, engine="igraph", para=list(nodeColor=nodeColor, 
hideNeg=TRUE), verbose=TRUE)
##need pvclust to search modules
library(pvclust)
pan<-pvclustModule(pan, nboot=1000, metric="cosine",
hclustMethod="average", filter=TRUE, verbose=TRUE, r=c(5:12/7))
inds<-sigModules(pan, pValCutoff=0.01, minSize=5, maxSize=100, sortby="size", 
decreasing=FALSE)
pan@modules$clusters[inds]

## End(Not run)

PANR documentation built on Nov. 8, 2020, 8:15 p.m.