Cluster-class {flowMatch} | R Documentation |
An object of class "Cluster
" represents a cluster or a cell population. We model a cluster with a normal distribution. An object of class "Cluster
" therefore represents a cluster with a mean vector, a covariance matrix and the size of the cluster.
An object of class Cluster
is usually created when constructing an object of class ClusteredSample
.
Unless you know exactly what you are doing, creating an object of class "Cluster
" using new
or using the constructor is discouraged.
An object of class "Cluster
" can be created using the following constructor
Cluster(size, center, cov, cluster.id = NA_integer_, sample.id=NA_integer_)
The arguments of the constructor bear usual meaning as described in the value section above.
An object of class "Cluster
" contains the following slots:
size
:An integer denoting the number of points (cells) present in the cluster.
center
:A numeric vector denoting the center of the cluster.
cov
:A matrix denoting the covariances of the underlying normal distribution of the cluster.
cluster.id
:The index of the cluster (relative to other clusters in same sample). Default is NA_integer_
.
sample.id
:The index of sample in which the cluster belongs to. Default is NA_integer_
.
All the slot accessor functions take an object of class Cluster
. I show usage of the first accessor function. Other functions can be called similarly.
get.size
:Returns the number of cells in the cluster.
Usage: get.size(object)
here object
is a Cluster
object.
get.center
:Returns the center of the cluster.
get.cov
:Returns the covariances matrix of the cluster.
get.cluster.id
:Returns the index of the cluster (relative to other clusters in same sample).
get.sample.id
:Returns the index of sample in which the cluster belongs to.
sample.id<-
:Set the index of sample in which the cluster belongs to.
Display details about the Cluster
object.
Return descriptive summary for each Cluster
object.
Usage: summary(Cluster)
Ariful Azad
## An object of class "Cluster"" is usually created when constructing a "ClusteredSample". ## Unless you know exactly what you are doing, creating an object of class "Cluster" ## using new or using the constructor is discouraged. ## ------------------------------------------------ ## load data and retrieve a sample ## ------------------------------------------------ library(healthyFlowData) data(hd) sample = exprs(hd.flowSet[[1]]) ## ------------------------------------------------ ## cluster sample using kmeans algorithm ## and retrive the parameters of the first cluster ## ------------------------------------------------ km = kmeans(sample, centers=4, nstart=20) center1 = km$centers[1,] # compute the covariance matrix of the first cluster cov1 = cov(sample[km$cluster==1,]) size1 = length(which(km$cluster==1)) ## ------------------------------------------------ ## Create an object of class "Cluster" ## and show summary ## ------------------------------------------------ clust = Cluster(size=size1, center=center1, cov=cov1) summary(clust)