Run a topological analysis on an expression dataset using clipper.

Share:

Description

clipper is a package for topological gene set analysis. It implements a two-step empirical approach based on the exploitation of graph decomposition into a junction tree to reconstruct the most relevant signal path. In the first step clipper selects significant pathways according to statistical tests on the means and the concentration matrices of the graphs derived from pathway topologies. Then, it "clips" the whole pathway identifying the signal paths having the greatest association with a specific phenotype.

Usage

1
runClipper(x, expr, classes, method, ...)

Arguments

x

a PathwayList, a list of Pathways or a single Pathway object.

expr

a matrix (size: number p of genes x number n of samples) of gene expression.

classes

a vector (length: n) of class assignments.

method

the kind of test to perform on the cliques. It could be either "mean" or "variance".

...

Additional options; see for details easyClip.

When invoked on a PathwayList, can use the named option "maxNodes" to limit the analysis to those pathways having up to this given number of nodes.

Details

The expression data and the pathway have to be annotated in the same set of identifiers.

References

Martini P, Sales G, Massa MS, Chiogna M, Romualdi C. Along signal paths: an empirical gene set approach exploiting pathway topology. Nucleic Acids Res. 2013 Jan 7;41(1):e19. doi: 10.1093/nar/gks866. Epub 2012 Sep 21. PubMed PMID: 23002139; PubMed Central PMCID: PMC3592432.

See Also

clipper

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
if (require(clipper) & require(ALL)){
  k <- pathways("hsapiens", "kegg")
  path <- convertIdentifiers(k$'Chronic myeloid leukemia', "entrez")
  genes <- nodes(path)
  data(ALL)
  all <- as.matrix(exprs(ALL[1:length(genes),1:20]))
  classes <- c(rep(1,10), rep(2,10))
  rownames(all) <- genes
  runClipper(path, all, classes, "mean", pathThr=0.1)
}