Mergeomics (version 1.0.0)

tool.cluster: Hierarchical clustering of nodes

Description

tool.cluster performs agglomerative hierarchical clustering for nodes (genes)

Usage

tool.cluster(edges, cutoff = NULL)

Arguments

edges

edge (weight) list among two group, whose overlapping information (overlapping ratio based on shared entries of two groups, number of members in both group) had been assesed previously

cutoff

cutting level of dendrogram for hierarchical clustering

Value

res

data list including clustering results:

CLUSTER: cluster label
NODE: item (node) name

Details

tool.cluster takes overlapping information between two groups, produces distance matrix based on 1-strength(overlap) ratio between two groups, and apply agglomerative hierarchical clustering based on the distance matrix.

References

Shu L, Zhao Y, Kurt Z, Byars S, Tukiainen T, Kettunen J, Ripatti S, Zhang B, Inouye M, Makinen VP, Yang X. Mergeomics: integration of diverse genomics resources to identify pathogenic perturbations to biological systems. bioRxiv doi: http://dx.doi.org/10.1101/036012

Examples

## read the coexpr module file as an example:
moddata <- tool.read(system.file("extdata",
"modules.mousecoexpr.liver.human.txt", package="Mergeomics"))

## let us cluster the first 10 modules in the module file:
mod.names <- unique(moddata$MODULE)[1:min(length(unique(moddata$MODULE)),
10)]
moddata <- moddata[which(!is.na(match(moddata$MODULE, mod.names))),]
## Find clusters.
rmax = 0.33
edges <- tool.overlap(items=moddata$GENE, groups=moddata$MODULE)
clustdat <- tool.cluster(edges, cutoff=rmax)
nclust <- length(unique(clustdat$CLUSTER))
nnodes <- length(unique(clustdat$NODE))