AggExResult-class: Class "AggExResult"

In apcluster: Affinity Propagation Clustering

Class "AggExResult"

Description

S4 class for storing results of exemplar-based agglomerative clustering

Objects

Objects of this class can be created by calling aggExCluster for a given similarity matrix.

Slots

The following slots are defined for AggExResult objects:

l:

number of samples in the data set

sel:

subset of samples used for leveraged clustering (empty for normal clustering)

maxNoClusters:

maximum number of clusters in the cluster hierarchy, i.e. it contains clusterings with 1 - maxNoClusters clusters.

exemplars:

list of length maxNoClusters; the i-th component of the list is a vector of i exemplars (corresponding to the level with i clusters).

clusters:

list of length maxNoClusters; the i-th component of clusters is a list of i clusters, each of which is a vector of sample indices.

merge:

a maxNoClusters-1 by 2 matrix that contains the merging hierarchy; fully analogous to the slot merge in the class hclust.

height:

a vector of length maxNoClusters-1 that contains the merging objective of each merge; largely analogous to the slot height in the class hclust except that the slot height in AggExResult objects is supposed to be non-increasing, since aggExCluster is based on similarities, whereas hclust uses dissimilarities.

order:

a vector containing a permutation of indices that can be used for plotting proper dendrograms without crossing branches; fully analogous to the slot order in the class hclust.

labels:

a character vector containing labels of clustered objects used for plotting dendrograms.

sim:

similarity matrix; only available if aggExCluster was called with similarity function and includeSim=TRUE.

call:

method call used to produce this clustering result

Methods

plot

signature(x="AggExResult"): see plot-methods

plot

signature(x="AggExResult", y="matrix"): see plot-methods

heatmap

signature(x="AggExResult"): see heatmap-methods

heatmap

signature(x="AggExResult", y="matrix"): see heatmap-methods

show

signature(object="AggExResult"): see show-methods

cutree

signature(object="AggExResult", k="ANY", h="ANY"): see cutree-methods

length

signature(x="AggExResult"): gives the number of clustering levels in the clustering result.

as.hclust

signature(x="AggExResult"): see coerce-methods

as.dendrogram

signature(object="AggExResult"): see coerce-methods

Accessors

In the following code snippets, x is an AggExResult object.

[[

signature(x="AggExResult", i="index", j="missing"): x[[i]] returns an object of class ExClust corresponding to the clustering level with i clusters; synonymous to cutree(x, i).

[

signature(x="AggExResult", i="index", j="missing", drop="missing"): x[i] returns a list of ExClust objects with all clustering levels specified in vector i. So, the list has as many components as the argument i has elements. A list is returned even if i is a single level.

similarity

signature(x="AggExResult"): gives the similarity matrix.

Author(s)

Ulrich Bodenhofer, Johannes Palme, and Johannes Palme

References

https://github.com/UBod/apcluster

Bodenhofer, U., Kothmeier, A., and Hochreiter, S. (2011) APCluster: an R package for affinity propagation clustering. Bioinformatics 27, 2463-2464. DOI: \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1093/bioinformatics/btr406")}.

See Also

aggExCluster, show-methods, plot-methods, cutree-methods

Examples

## create two Gaussian clouds
cl1 <- cbind(rnorm(50, 0.2, 0.05), rnorm(50, 0.8, 0.06))
cl2 <- cbind(rnorm(50, 0.7, 0.08), rnorm(50, 0.3, 0.05))
x <- rbind(cl1, cl2)

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

## compute agglomerative clustering from scratch
aggres1 <- aggExCluster(sim)

## show results
show(aggres1)

## plot dendrogram
plot(aggres1)

## plot heatmap along with dendrogram
heatmap(aggres1, sim)

## plot level with two clusters
plot(aggres1, x, k=2)

## run affinity propagation
apres <- apcluster(sim, q=0.7)

## create hierarchy of clusters determined by affinity propagation
aggres2 <- aggExCluster(sim, apres)

## show results
show(aggres2)

## plot dendrogram
plot(aggres2)

## plot heatmap
heatmap(aggres2, sim)

## plot level with two clusters
plot(aggres2, x, k=2)

apcluster documentation built on May 29, 2024, 2:25 a.m.