clValid-class {clValid} | R Documentation |
The class "clValid"
contains the clustering results
and validation measures from the accompanying call to the function
clValid
.
Objects can be created using the function clValid
.
clusterObjs
:"list"
. A list
containing the results from the clustering methods.measures
:"array"
. A 3-dimensional array which contains the
validation measures for the clustering results. The first
dimension indicates the validation measures, the second the number of
clusters, and the third the clustering methods.measNames
:"character"
. The
names of the validation measures.clMethods
:"character"
. A
character vector giving the clustering methods.nClust
:"numeric"
. A numeric vector giving the numbers of clusters
which were evaluated.validation
:"character"
. A character vector giving the type of
validation measures used, consisting of some combination of
"internal", "stability", or "biological".metric
:"character"
. The metric used to determine the distance
matrix.method
:"character"
. For
hierarchical clustering, the agglomeration method used.neighbSize
:"numeric"
. For internal validation, the neighborhood size used for the
connectivity measure.annotation
:"character or array
or list"
.
Either a character string naming the Bioconductor annotation
package for mapping genes to GO categories, or a list with the names of the functional classes
and the observations belonging to each class.GOcategory
:"character"
. For biological validation, gives which GO
categories to use for biological validation. Can be one of "BP",
"MF", "CC", or "all"goTermFreq
:"numeric"
. For the
BSI, what threshold frequency of GO terms to use for functional annotation.call
:"call"
. Gives the call
to clValid
used to create the clValid
object.signature(object = "clValid")
: Returns the
names of the clustering methods. signature(object = "clValid")
: Returns the
results from the clustering methods.
Additional arguments:
method = clMethods(object)
signature(object = "clValid")
: Returns the
names of the validation measures.signature(object = "clValid")
: Returns
the validation measures.
Additional arguments:
measures = measNames(object)
signature(object = "clValid")
: Returns the
numbers of clusters evaluated.signature(object = "clValid")
:
Returns the optimal value for each validation measure, along with
the corresponding clustering method and number of clusters.
Additional arguments:
measures = measNames(object)
signature(x = "clValid", y = "missing")
: Plots
the validation measures.
Additional arguments:
measures=measNames(x)
legend=TRUE
legendLoc="topright"
main=NULL
pch=NULL
type="b"
ask=prod(par("mfcol")) < length(measures) &&
dev.interactive()
TRUE
the user is
prompted before each plot.signature(x = "clValid")
: Print method for class
clValid
.signature(object = "clValid")
: Same as print.signature(object = "clValid")
: Summary method
for class clValid
.
Additional arguments:
digits = max(3,getOption("digits")-3)
See the vignette for an illustration of the class.
Guy Brock, Vasyl Pihur, Susmita Datta, Somnath Datta
Datta, S. and Datta, S. (2003). Comparisons and validation of statistical clustering techniques for microarray gene expression data. Bioinformatics 19(4): 459-466.
Datta, S. and Datta, S. (2006). Methods for evaluating clustering algorithms for gene expression data using a reference set of functional classes. BMC Bioinformatics 7:397.
Handl, J., Knowles, K., and Kell, D. (2005). Computational cluster validation in post-genomic data analysis. Bioinformatics 21(15): 3201-3212.
For a description of the function 'clValid' see clValid
.
For help on the clustering methods see hclust
and
kmeans
in package stats,
kmeans
in package stats,
agnes
, clara
, diana
,
fanny
, and pam
in package cluster,
som
in package kohonen, Mclust
in package mclust, and sota
.
For additional help on the validation measures see
connectivity
, dunn
,
stability
,
BHI
, and
BSI
.
## to delete library(clValid) data(mouse) ## internal validation express <- mouse[1:25,c("M1","M2","M3","NC1","NC2","NC3")] rownames(express) <- mouse$ID[1:25] intern <- clValid(express, 2:6, clMethods=c("hierarchical","fanny","model"), validation="internal") slotNames(intern) ## view results intern summary(intern) optimalScores(intern) plot(intern) ## Extract objects from slots measures(intern) hierClust <- clusters(intern,"hierarchical") plot(hierClust) measNames(intern) nClusters(intern)