cell.SubClustering {immunoClust} | R Documentation |
These function tests each cell-cluster of a model for refining it into more
sub-clusters and returns the refined model parameter in an object of class
immunoClust
.
cell.SubClustering( x, dat, B=50, tol=1e-5, thres=0.1, bias=0.5, sample.weights=1, sample.EM="MEt", sample.number=1500, sample.standardize=TRUE, extract.thres=0.8, modelName="mvt") cell.TestSubCluster(x, y, t, cluster, J=8, B=500, tol=1e-5, bias=0.5, sample.EM="MEt", sample.df=5, sample.number=1500, sample.standardize=TRUE, modelName="mvt")
x |
An immunoClust object with the initial model parameter (K, w, mu, sigma). |
dat |
A numeric matrix, data frame of observations, or object of class flowFrame. |
B |
The maximum number of EM(t)-iterations in Sub-Clustering. |
tol |
The tolerance used to assess the convergence of the EM(t)-algorithms in Sub-Clustering. |
thres |
Defines the threshold, below which an ICL-increase is meaningless. The threshold is given as the multiple (or fraction) of the costs for a single cluster. |
bias |
The ICL-bias used in the EMt-algorithm. |
sample.weights |
Power of weights applied to hierarchical clustering, where the used weights are the probabilities of cluster membership. |
sample.EM |
Used EM-algorithm; either |
sample.number |
The number of samples used for initial hierarchical clustering. |
sample.standardize |
A numeric indicating whether the samples for hierarchical clustering are standardized (mean=0, SD=1). |
extract.thres |
The threshold used for cluster data extraction. |
modelName |
Used mixture model; either |
y |
A numeric matrix of the observations beloning to the particular cluster. |
t |
A numeric vector with the probability weights for the observations belonining to the particular cluster. |
cluster |
An integer index of the particular cluster |
J |
The number of sub-models to be builded and tested for a particular cluster. |
sample.df |
Degree of freedom for the t-distibutions in a t-mixture model. Has to be 5 in immunoClust. |
These function are used internally by the cell-clustering procedures of
cell.process
in immunoClust and are not intended to be used
directly.
The cluster parameters of the refined model in an object of class
immunoClust
.
Till Sörensen till-antoni.soerensen@charite.de
Sörensen, T., Baumgart, S., Durek, P., Grützkau, A. and Häupl, T. immunoClust - an automated analysis pipeline for the identification of immunophenotypic signatures in high-dimensional cytometric datasets. Cytometry A (accepted).
data(dat.fcs) data(dat.exp) dat.trans <- trans.ApplyToData(dat.exp[[1]], dat.fcs) #need to re-calculate the cluster membership probabilities # not stored in dat.exp r1 <- cell.Classify(dat.exp[[1]], dat.trans) summary(r1) r2 <- cell.SubClustering(r1, dat.trans) summary(r2)