apcluster (version 1.4.11)

apcluster-package: APCluster Package

Description

The apcluster package implements affinity propagation according to Frey and Dueck and a method for exemplar-based agglomerative clustering. It further offers various functions for plotting clustering results.

Author

Ulrich Bodenhofer, Andreas Kothmeier, Johannes Palme, Chrats Melkonian & Nikola Kostic

Details

The central function is apcluster. It runs affinity propagation on a given similarity matrix or it creates a similarity matrix for a given data set and similarity measure and runs affinity propagation on this matrix. The function returns an APResult object from which the clustering itself and information about the affinity propagation run can be obtained. Leveraged affinity propagation clustering apclusterL allows efficient clustering of large datasets by using only a subset of the similarities. The package further implements an exemplar-based agglomerative clustering method aggExCluster that can be used for computing a complete cluster hierarchy, but also for joining fine-grained clusters previously obtained by affinity propagation clustering. Further functions are implemented to visualize the results and to create distance matrices.

References

http://www.bioinf.jku.at/software/apcluster/

Bodenhofer, U., Kothmeier, A., and Hochreiter, S. (2011) APCluster: an R package for affinity propagation clustering. Bioinformatics 27, 2463-2464. DOI: tools:::Rd_expr_doi("10.1093/bioinformatics/btr406").

Frey, B. J. and Dueck, D. (2007) Clustering by passing messages between data points. Science 315, 972-976. DOI: tools:::Rd_expr_doi("10.1126/science.1136800").

Examples

## create two Gaussian clouds
cl1 <- cbind(rnorm(100, 0.2, 0.05), rnorm(100, 0.8, 0.06))
cl2 <- cbind(rnorm(100, 0.7, 0.08), rnorm(100, 0.3, 0.05))
x <- rbind(cl1, cl2)

## compute similarity matrix (negative squared Euclidean)
sim <- negDistMat(x, r=2)

## run affinity propagation
apres <- apcluster(sim, details=TRUE)

## show details of clustering results
show(apres)

## plot information about clustering run
plot(apres)

## plot clustering result
plot(apres, x)

## employ agglomerative clustering to join clusters
aggres <- aggExCluster(sim, apres)

## show information
show(aggres)
show(cutree(aggres, 2))

## plot dendrogram
plot(aggres)

## plot clustering result for k=2 clusters
plot(aggres, x, k=2)

## plot heatmap
heatmap(apres, sim)

## leveraged apcluster
apresL <- apclusterL(s=negDistMat(r=2), x=x,  frac=0.2, sweeps=3)

## show details of clustering results
show(apresL)

## plot clustering result
plot(apresL, x)