1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
|
\name{AggExResult-class}
\docType{class}
\alias{AggExResult-class}
\alias{AggExResult}
\alias{aggexresult}
\alias{[,AggExResult,index,missing,missing-method}
\alias{[[,AggExResult,index,missing-method}
\alias{length,AggExResult-method}
\alias{similarity,AggExResult-method}
\title{Class "AggExResult"}
\description{S4 class for storing results of exemplar-based
agglomerative clustering}
\section{Objects}{
Objects of this class can be created by calling \code{\link{aggExCluster}}
for a given similarity matrix.
}
\section{Slots}{
The following slots are defined for \link{AggExResult} objects:
\describe{
\item{\code{l}:}{number of samples in the data set}
\item{\code{sel}:}{subset of samples used for leveraged clustering
(empty for normal clustering)}
\item{\code{maxNoClusters}:}{maximum number of clusters in the
cluster hierarchy, i.e. it
contains clusterings with 1 - \code{maxNoClusters} clusters.}
\item{\code{exemplars}:}{list of length \code{maxNoClusters};
the \code{i}-th component of the list is a vector of \code{i}
exemplars (corresponding to the level with \code{i} clusters).}
\item{\code{clusters}:}{list of length \code{maxNoClusters};
the \code{i}-th component of \code{clusters} is a list of \code{i}
clusters, each of which is a vector of sample indices.}
\item{\code{merge}:}{a \code{maxNoClusters-1} by 2 matrix that
contains the merging hierarchy; fully analogous to the
slot \code{merge} in the class \code{\link{hclust}}.}
\item{\code{height}:}{a vector of length \code{maxNoClusters-1} that
contains the merging objective of each merge; largely analogous to
the slot \code{height} in the class \code{\link{hclust}} except
that the slot \code{height} in \code{AggExResult} objects is
supposed to be non-increasing, since \code{\link{aggExCluster}}
is based on similarities, whereas \code{\link{hclust}} uses
dissimilarities.}
\item{\code{order}:}{a vector containing a permutation of indices
that can be used for plotting proper dendrograms without crossing
branches; fully analogous to the
slot \code{order} in the class \code{\link{hclust}}.}
\item{\code{labels}:}{a character vector containing labels of
clustered objects used for plotting dendrograms.}
\item{\code{sim}:}{similarity matrix; only available if
\code{\link{aggExCluster}} was called with similarity
function and \code{includeSim=TRUE}.}
\item{\code{call}:}{method call used to produce this
clustering result}
}
}
\section{Methods}{
\describe{
\item{plot}{\code{signature(x="AggExResult")}: see
\code{\link{plot-methods}}}
\item{plot}{\code{signature(x="AggExResult", y="matrix")}: see
\code{\link{plot-methods}}}
\item{heatmap}{\code{signature(x="AggExResult")}: see
\code{\link{heatmap-methods}}}
\item{heatmap}{\code{signature(x="AggExResult", y="matrix")}: see
\code{\link{heatmap-methods}}}
\item{show}{\code{signature(object="AggExResult")}: see
\code{\link{show-methods}}}
\item{cutree}{\code{signature(object="AggExResult", k="ANY",
h="ANY")}: see \code{\link{cutree-methods}}}
\item{length}{\code{signature(x="AggExResult")}: gives the number of
clustering levels in the clustering result.}
\item{as.hclust}{\code{signature(x="AggExResult")}: see
\code{\link{coerce-methods}}}
\item{as.dendrogram}{\code{signature(object="AggExResult")}: see
\code{\link{coerce-methods}}}
}
}
\section{Accessors}{
In the following code snippets, \code{x} is an \code{AggExResult} object.
\describe{
\item{[[}{\code{signature(x="AggExResult", i="index", j="missing")}:
\code{x[[i]]} returns an object of class
\code{\linkS4class{ExClust}} corresponding to the clustering level
with \code{i} clusters; synonymous to \code{\link{cutree}(x, i)}.
}
\item{[}{\code{signature(x="AggExResult", i="index", j="missing",
drop="missing")}: \code{x[i]} returns a list of \code{\linkS4class{ExClust}}
objects with all clustering levels specified in vector \code{i}.
So, the list has as many components as the argument \code{i} has
elements. A list is returned even if \code{i} is a single level.
}
\item{similarity}{\code{signature(x="AggExResult")}: gives the similarity
matrix.
}
}
}
\author{Ulrich Bodenhofer, Johannes Palme, and Johannes Palme}
\references{\url{https://github.com/UBod/apcluster}
Bodenhofer, U., Kothmeier, A., and Hochreiter, S. (2011)
APCluster: an R package for affinity propagation clustering.
\emph{Bioinformatics} \bold{27}, 2463-2464.
DOI: \doi{10.1093/bioinformatics/btr406}.
}
\seealso{\code{\link{aggExCluster}}, \code{\link{show-methods}},
\code{\link{plot-methods}}, \code{\link{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)
}
\keyword{classes}
|
