tests/testthat/test-clusterSingle.R
In epurdom/clusterCells: Compare Clusterings for Single-Cell Sequencing
context("clusterSingle")
test_that("`clusterSingle` works with matrix, ClusterExperiment objects,
SummarizedExperiments", {
#---
#Matrix
#---
expect_silent(clustNothing <-
clusterSingle(mat, inputType="X",
subsample=FALSE, sequential=FALSE,
mainClusterArgs=list(clusterArgs=list(k=3),
clusterFunction="pam"),
isCount=FALSE))
expect_equal(clusterLabels(clustNothing),"clusterSingle")
expect_is(clustNothing, "ClusterExperiment")
expect_is(clustNothing, "SummarizedExperiment")
truth<- unname(cluster::pam(x=t(mat),cluster.only=TRUE,k=3))
#in case change in sim, make sure same as from previous tests
expect_equal(truth,c(1,2,3,1,3,1,1,1,1,2,2,1,1,1,1))
expect_equal(primaryCluster(clustNothing),truth)
#test clusterLabel
expect_silent(clustNothing2 <- clusterSingle(mat,
inputType="X",
mainClusterArgs= list(clusterArgs=list(k=3), clusterFunction="pam"),
subsample=FALSE,
sequential=FALSE, isCount=FALSE,
clusterLabel="myownClustering"))
expect_equal(clusterLabels(clustNothing2),"myownClustering")
#test default 01 distance
expect_silent(clusterSingle(inputMatrix=dissMat, inputType="diss", mainClusterArgs= list(clusterArgs=list(alpha=0.1),clusterFunction="tight"),
subsample=FALSE, sequential=FALSE,
isCount=FALSE))
#test default K distance
expect_silent(clusterSingle(dissMat, inputType="diss", mainClusterArgs= list(clusterArgs=list(k=3),clusterFunction="hierarchicalK"),subsample=FALSE, sequential=FALSE, isCount=FALSE))
#---
#SE/SCE
#---
###Apply to SE
expect_silent(clustNothing2 <- clusterSingle(se,
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE))
expect_equal(clusterMatrix(clustNothing2), clusterMatrix(clustNothing))
###Apply to SCE
expect_silent(clustNothing3 <- clusterSingle(sce, mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE))
expect_equal(clusterMatrix(clustNothing2), clusterMatrix(clustNothing3))
expect_silent(clustNothing3 <- clusterSingle(sce,
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE))
#---
#CE
#---
#test running on ClusterExperiment Object -- should add the new clustering
expect_error(clustNothing3 <- clusterSingle(clustNothing2, mainClusterArgs=list(clusterArgs=list(k=4),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE,
isCount=FALSE),"setting 'transFun' or 'isCount' for a 'ClusterExperiment' is not allowed")
expect_silent(clustNothing3 <- clusterSingle(clustNothing2, mainClusterArgs=list(clusterArgs=list(k=4),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE ))
expect_equal(NCOL(clusterMatrix(clustNothing3)),2)
expect_equal(length(table(primaryCluster(clustNothing3))),4,info="Check reset primary cluster after run clusterSingle")
})
test_that("`clusterSingle` warnings/errors work", {
expect_error(clusterSingle(mat, inputType="X",
mainClusterArgs= list(clusterArgs=list(alpha=0.1),
clusterFunction="tight"),
distFunction=function(x){dist(x,method="manhattan")},
subsample=FALSE, checkDiss=TRUE,
sequential=FALSE,isCount=FALSE),
"distance function must give values between 0 and 1")
#test warn with wrong arguments
expect_warning(clusterSingle(dissMat, inputType="diss", mainClusterArgs= list(clusterArgs=list(k=3,alpha=0.1),clusterFunction="tight"),
subsample=FALSE, sequential=FALSE,
,isCount=FALSE),"arguments passed via clusterArgs to the clustering function tight are not all applicable")
#test turning off warning
expect_silent(clusterSingle(dissMat,
inputType="diss",
mainClusterArgs= list(clusterArgs=list(k=3,alpha=0.1), clusterFunction="tight"),
warnings=FALSE,
subsample=FALSE,
sequential=FALSE,
isCount=FALSE))
})
test_that("`clusterSingle` works with reduceMethod", {
#existing dim reduce values
expect_silent(clustNothing4 <- clusterSingle(sceSimDataDimRed,
reduceMethod="PCA", nDims=3,
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE))
expect_equal(sort(reducedDimNames(clustNothing4)),sort(c("PCA","TSNE")))
expect_silent(clustNothing5 <- clusterSingle(t(reducedDim(sceSimDataDimRed,"PCA")[,1:3]),
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE))
expect_equal(clusterMatrix(clustNothing4), clusterMatrix(clustNothing5))
#test nDims=NA
expect_silent(clustNothing1 <- clusterSingle(sceSimDataDimRed,
reduceMethod="PCA", nDims=NA,
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE))
expect_equal(sort(reducedDimNames(clustNothing4)),sort(c("PCA","TSNE")))
expect_silent(clustNothing2 <- clusterSingle(t(reducedDim(sceSimDataDimRed,"PCA")),
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE))
expect_equal(clusterMatrix(clustNothing1), clusterMatrix(clustNothing2))
#blank sceSimData object
##FIXME
#expect_silent(
clustNothing6 <- clusterSingle(sceSimData,
reduceMethod="PCA", nDims=3,
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE)
#)
expect_equal(sort(reducedDimNames(clustNothing6)),sort(c("PCA")))
expect_silent(clustNothing7 <- clusterSingle(t(reducedDim(sceSimDataDimRed,"PCA")[,1:3]),
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE))
expect_equal(clusterMatrix(clustNothing6), clusterMatrix(clustNothing7))
## FIXME
#expect_silent(
clustNothing8 <- clusterSingle(seSimData,
reduceMethod="PCA", nDims=3,
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE)
#)
expect_equal(sort(reducedDimNames(clustNothing8)),sort(c("PCA")))
expect_equal(clusterMatrix(clustNothing8), clusterMatrix(clustNothing7))
#check consistency
expect_equal(primaryCluster(clustNothing8),rep(c(1,2,3),each=100))
## FIXME
#expect_silent(
clusterSingle(sceSimData,
reduceMethod="PCA", nDims=NA,
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE)
#)
})
test_that("all built-in clustering methods",{
#########
## Put all clusterSingle tests that loop over methods HERE!
#########
kClusterArgs<-list(k=3)
aClusterArgs<-list(alpha=0.1)
set.seed(3325)
biggerMat<-matrix(data=rnorm(20*100), ncol=100)
diss<-as.matrix(dist(t(mat)))
biggerDiss<-as.matrix(dist(t(biggerMat)))
getInputMatrix<- function(cf, big=FALSE){
if("X" %in% inputType(cf)){
inputType<-"X"
if(!big) inputMat<-mat
else inputMat<-biggerMat
}
else if("diss" %in% inputType(cf)){
inputType<-"diss"
if(!big) inputMat<-diss
else inputMat<- biggerDiss
}
else stop("error, don't have test for cat type matrix")
return(list(inputType=inputType,inputMatrix=inputMat))
}
## ---------
# check with main clustering
## ---------
#FIXME: Is this duplicative of previous test in hdf5?
for(cf in listBuiltInFunctions()){
algType<-algorithmType(cf)
clustArgs<-switch(algType,
"K"=kClusterArgs,
"01"=aClusterArgs
)
input<-getInputMatrix(cf)
expect_silent(clusterSingle(inputMat=input$inputMatrix,
inputType=input$inputType,
mainClusterArgs= list(clusterArgs=clustArgs, clusterFunction=cf),
subsample=FALSE, sequential=FALSE,isCount=FALSE)
)
if(algType=="K"){
input<-getInputMatrix(cf,big=TRUE)
expect_silent(clusterSingle(inputMat=input$inputMatrix,
inputType=input$inputType,
mainClusterArgs= list(clusterArgs=clustArgs,
diss=switch(algType,"01"=NULL,"K"=biggerDiss),
clusterFunction=cf,findBestK=TRUE,removeSil=TRUE),
subsample=FALSE, sequential=FALSE,isCount=FALSE))
}
}
##------
## Check with subsampling
##------
# also force to be opposite type from mainClusterArgs
for(cf in listBuiltInFunctions()){
algType<-algorithmType(cf)
if(algType=="K"){
mainArgs<-list(clusterFunction="hierarchical01",
clusterArgs=aClusterArgs)
subArgs<-list(clusterFunction=cf,resamp.num=20, clusterArgs=kClusterArgs)
}
if(algType=="01"){
mainArgs<-list(clusterFunction="pam", clusterArgs=kClusterArgs)
subArgs<-list(clusterFunction=cf,resamp.num=20,
clusterArgs=aClusterArgs)
}
input<-getInputMatrix(cf,big=TRUE)
set.seed(1045)
##In-sample
expect_silent(clusterSingle(inputMat=input$inputMatrix,
inputType=input$inputType,
subsample=TRUE,
sequential=FALSE,
makeMissingDiss=FALSE,
subsampleArgs=c(list(
classifyMethod="InSample"),subArgs),
mainClusterArgs=mainArgs,isCount=FALSE))
if(!is.null(getBuiltInFunction(cf)@classifyFUN)){
#All
set.seed(1045)
expect_silent(clusterSingle(inputMat=input$inputMatrix,
inputType=input$inputType, subsample=TRUE, sequential=FALSE,
subsampleArgs=c(list(classifyMethod="All"),subArgs),
mainClusterArgs=mainArgs,isCount=FALSE))
#OutOfSample
set.seed(1045)
expect_silent(clusterSingle(inputMat=input$inputMatrix,
inputType=input$inputType, subsample=TRUE, sequential=FALSE,
subsampleArgs=c(list(classifyMethod="OutOfSample"),subArgs),
mainClusterArgs=mainArgs,isCount=FALSE))
}
}
# --------
# Seq cluster -- only works for K methods
# --------
for(cf in listBuiltInTypeK()){
algType<-algorithmType(cf)
subArgs<-list(clusterFunction="pam",n.sample=40)
seqArgs<-list(k0=5,beta=0.9,verbose=FALSE)
input<-getInputMatrix(cf,big=FALSE)
#check subsample=FALSE
expect_silent(clusterSingle(input$inputMatrix, inputType=input$inputType,
mainClusterArgs=list(clusterFunction=cf),
subsample=FALSE, sequential=TRUE,
isCount=FALSE,seqArgs=seqArgs))
#check subsample=TRUE
if("cat" %in% inputType(cf))
expect_silent(clusterSingle(input$inputMatrix,
inputType=input$inputType,
mainClusterArgs=list(clusterFunction=cf),
subsample=TRUE, sequential=TRUE,
subsampleArgs=subArgs,
isCount=FALSE,seqArgs=seqArgs))
}
#-------
## Check that they work with hdf5
#-------
seedValue<-571839
for(cf in listBuiltInFunctions()){
#print(cf)
clArgs<-switch(algorithmType(cf),
"K"=kClusterArgs,
"01"=aClusterArgs)
### Annoyingly, have to separately for "X" and "diss"
### Because for this, no way to give it the diss and still test the objects
if("X" %in% inputType(cf)){
set.seed(seedValue)
expect_silent(clust1<-clusterSingle(sceSimDataDimRed,
reduceMethod = "none",
mainClusterArgs= list(clusterArgs=clArgs,
clusterFunction=cf),
subsample=FALSE,
makeMissingDiss=TRUE,
sequential=FALSE,
isCount=FALSE)
)
set.seed(seedValue)
expect_silent(clust2<-clusterSingle(hdfSCE,
reduceMethod = "none",
mainClusterArgs= list(clusterArgs=clArgs,
clusterFunction=cf),
subsample=FALSE,
makeMissingDiss=TRUE,
sequential=FALSE,
isCount=FALSE)
)
set.seed(seedValue)
expect_silent(clust3<-clusterSingle(hdfObj,
reduceMethod = "none",
mainClusterArgs= list(clusterArgs=clArgs,
clusterFunction=cf),
subsample=FALSE,
makeMissingDiss=TRUE,
sequential=FALSE,
isCount=FALSE)
)
set.seed(seedValue)
expect_silent(clust4<-clusterSingle(assay(hdfSCE),
reduceMethod = "none",
mainClusterArgs= list(clusterArgs=clArgs,
clusterFunction=cf),
subsample=FALSE,
makeMissingDiss=TRUE,
sequential=FALSE,
isCount=FALSE)
)
}
else{
set.seed(seedValue)
expect_message(clust1<-clusterSingle(sceSimDataDimRed,
reduceMethod = "none",
mainClusterArgs= list(clusterArgs=clArgs,
clusterFunction=cf),
subsample=FALSE,
makeMissingDiss=TRUE,
sequential=FALSE,
isCount=FALSE),"Note: Making nxn dissimilarity matrix."
)
set.seed(seedValue)
expect_message(clust2<-clusterSingle(hdfSCE,
reduceMethod = "none",
mainClusterArgs= list(clusterArgs=clArgs,
clusterFunction=cf),
subsample=FALSE,
makeMissingDiss=TRUE,
sequential=FALSE,
isCount=FALSE),"Note: Making nxn dissimilarity matrix."
)
set.seed(seedValue)
expect_message(clust3<-clusterSingle(hdfObj,
reduceMethod = "none",
mainClusterArgs= list(clusterArgs=clArgs,
clusterFunction=cf),
subsample=FALSE,
makeMissingDiss=TRUE,
sequential=FALSE,
isCount=FALSE),"Note: Making nxn dissimilarity matrix."
)
set.seed(seedValue)
expect_message(clust4<-clusterSingle(assay(hdfSCE),
reduceMethod = "none",
mainClusterArgs= list(clusterArgs=clArgs,
clusterFunction=cf),
subsample=FALSE,
makeMissingDiss=TRUE,
sequential=FALSE,
isCount=FALSE),"Note: Making nxn dissimilarity matrix."
)
}
expect_equal(clusterMatrix(clust1) ,clusterMatrix(clust2))
expect_equal(clusterMatrix(clust1) ,clusterMatrix(clust3))
expect_equal(clusterMatrix(clust1) ,clusterMatrix(clust4))
}
})
test_that("`clusterSingle` works with hdf5Matrix",{
# These are slow!
skip_on_os("windows")
skip_on_os("mac")
####Test sequential option,
expect_silent(clustSeq <- clusterSingle(hdfObj,
reduceMethod="none",
subsample=FALSE,
sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE)
))
expect_silent(clustSeq <- clusterSingle(assay(hdfObj),
reduceMethod="none",
subsample=FALSE, sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE)))
####Test subsample option
expect_silent(clusterSingle(hdfObj,
reduceMethod="none",
subsample=TRUE,
sequential=FALSE,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE,
subsampleArgs=list(clusterFunction="pam",resamp.num=3,
clusterArgs=list(k=3))
)
)
expect_silent(clusterSingle(assay(hdfObj), reduceMethod="none",
subsample=TRUE, sequential=FALSE,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE,
subsampleArgs=list(clusterFunction="pam",resamp.num=3, clusterArgs=list(k=3))
)
)
})
test_that("`clusterSingle` works with filtering", {
####Check built in functions ####
for(fs in listBuiltInFilterStats()){
expect_silent(clusterSingle(mat,
subsample=FALSE,
sequential=FALSE,
reduceMethod=fs,
nDims=3,
mainClusterArgs=list(clusterFunction="pam", clusterArgs=list(k=3)),
isCount=FALSE))
}
expect_warning(clusterSingle(mat,
subsample=FALSE,
sequential=FALSE,
reduceMethod="var",
nDims=NROW(mat)+1,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE),
"the number of most features requested after filtering is either missing or larger than the number of features. Will not do any filtering")
expect_warning(clusterSingle(mat,
subsample=FALSE, sequential=FALSE,
reduceMethod="none",nDims=3,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE),
"specifying nDims has no effect if reduceMethod==`none`")
#check returns filterStats
expect_silent(cc <- clusterSingle(mat,
subsample=FALSE, sequential=FALSE,
reduceMethod="var", nDims=3,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)), isCount=FALSE))
expect_equal(sort(filterNames(cc)),"var")
expect_silent(cc2 <- clusterSingle(se,
subsample=FALSE, sequential=FALSE,
reduceMethod="var", nDims=3,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)), isCount=FALSE))
expect_equal(sort(filterNames(cc2)),c("b","var"))
expect_silent(cc3 <- clusterSingle(sceSimData,
subsample=FALSE, sequential=FALSE,
reduceMethod="var", nDims=3,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)), isCount=FALSE))
expect_equal(sort(filterNames(cc3)),c("b","var"))
#test using existing values
expect_silent(cc4 <- clusterSingle(sceFull,
subsample=FALSE, sequential=FALSE,
reduceMethod="var", nDims=3,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)), isCount=FALSE))
expect_equal(sort(filterNames(cc4)),sort(c("var","b","Filter1","Filter2")))
})
test_that("`clusterSingle` consistent results (no transformation)", {
#----
#check get the same result from previous times when transform the data
#----
#make it big enough can do pca and filter...
contData<-simData[,1:20]
testSE<-SummarizedExperiment(contData)
testSCE<-as(testSE,"SingleCellExperiment")
expectTrans1<-round(contData[1,],2)
###
#var filter
###
#matrix
expect_silent(cc<-clusterSingle(contData,
subsample=FALSE, sequential=FALSE, reduceMethod="var",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_equal(primaryCluster(cc),c(3,1,1,3,2,2,3,2,3,1,2,2,1,3,3,2,1,1,2,1))
expect_equal(round(transformData(cc)[1,],2), expectTrans1)
expect_silent(cc<-clusterSingle(contData,
subsample=FALSE, sequential=FALSE, reduceMethod="var",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_equal(primaryCluster(cc),c(3,1,1,3,2,2,3,2,3,1,2,2,1,3,3,2,1,1,2,1))
#SE
expect_silent(cc2<-clusterSingle(testSE,
subsample=FALSE, sequential=FALSE, reduceMethod="var",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_equal(primaryCluster(cc2),primaryCluster(cc))
#SCE
expect_silent(cc3<-clusterSingle(testSCE,
subsample=FALSE, sequential=FALSE, reduceMethod="var",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_equal(primaryCluster(cc3),primaryCluster(cc))
#CE
expect_silent(cc5<-clusterSingle(cc,
subsample=FALSE, sequential=FALSE, reduceMethod="var",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),clusterLabel="redoCS"))
expect_equal(primaryCluster(cc5),primaryCluster(cc))
###
#PCA
###
#matrix
expect_silent(cc<-clusterSingle(contData,
subsample=FALSE, sequential=FALSE, reduceMethod="PCA",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_equal(primaryCluster(cc),c(3,1,3,1,3,2,1,1,1,3,2,2,2,3,1,2,2,1,2,3))
expect_equal(round(transformData(cc)[1,],2), expectTrans1)
#SE
## FIXME
#expect_silent(
cc2<-clusterSingle(testSE,
subsample=FALSE, sequential=FALSE, reduceMethod="PCA",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE)
#)
expect_equal(primaryCluster(cc2),primaryCluster(cc))
#SCE
## FIXME
#expect_silent(
cc3<-clusterSingle(testSCE,
subsample=FALSE, sequential=FALSE, reduceMethod="PCA",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE)
#)
expect_equal(primaryCluster(cc3),primaryCluster(cc))
#CE
expect_silent(cc5<-clusterSingle(cc,
subsample=FALSE, sequential=FALSE, reduceMethod="PCA",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),clusterLabel="redoCS"))
expect_equal(primaryCluster(cc5),primaryCluster(cc))
###
#Nothing
###
#matrix
expect_silent(cc<-clusterSingle(contData,
subsample=FALSE, sequential=FALSE, reduceMethod="none",
nDims=NA, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_equal(primaryCluster(cc),c(3,2,1,3,2,1,3,2,3,2,1,2,1,1,3,1,1,1,2,2))
expect_equal(round(transformData(cc)[1,],2), expectTrans1)
#SE
expect_silent(cc2<-clusterSingle(testSE,
subsample=FALSE, sequential=FALSE, reduceMethod="none",
nDims=NA, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_equal(primaryCluster(cc2),primaryCluster(cc))
#SCE
expect_silent(cc3<-clusterSingle(testSCE,
subsample=FALSE, sequential=FALSE, reduceMethod="none",
nDims=NA, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_equal(primaryCluster(cc3),primaryCluster(cc))
#CE
expect_silent(cc5<-clusterSingle(cc,
subsample=FALSE, sequential=FALSE, reduceMethod="none",
nDims=NA, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),clusterLabel="redoCS"))
expect_equal(primaryCluster(cc5),primaryCluster(cc))
})
test_that("`clusterSingle` consistent results (with transformation)", {
#----
#check get the same result from previous times when transform the data
#----
#make it big enough can do pca and filter...
countData<-simCount[,1:20]
testSE<-SummarizedExperiment(countData)
testSCE<-as(testSE,"SingleCellExperiment")
expectTrans1<-round(log2(countData[1,]+1),2)
###
#var filter
###
#matrix
expect_silent(cc<-clusterSingle(countData,
subsample=FALSE, sequential=FALSE, reduceMethod="var",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=TRUE))
expect_equal(primaryCluster(cc),c(1,1,2,1,2,1,2,2,3,2,1,3,1,3,1,1,2,3,2,2))
expect_equal(round(transformData(cc)[1,],2), expectTrans1)
#SE
expect_silent(cc2<-clusterSingle(testSE,
subsample=FALSE, sequential=FALSE, reduceMethod="var",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=TRUE))
expect_equal(primaryCluster(cc2),primaryCluster(cc))
#SCE
expect_silent(cc3<-clusterSingle(testSCE,
subsample=FALSE, sequential=FALSE, reduceMethod="var",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=TRUE))
expect_equal(primaryCluster(cc3),primaryCluster(cc))
#CE
expect_silent(cc5<-clusterSingle(cc,
subsample=FALSE, sequential=FALSE, reduceMethod="var",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),clusterLabel="redoCS"))
expect_equal(primaryCluster(cc5),primaryCluster(cc))
###
#PCA
###
#matrix
expect_silent(cc<-clusterSingle(countData,
subsample=FALSE, sequential=FALSE, reduceMethod="PCA",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=TRUE))
expect_equal(primaryCluster(cc),c(3,2,1,1,3,1,3,1,1,1,1,2,1,2,1,1,1,2,2,1))
expect_equal(round(transformData(cc)[1,],2), expectTrans1)
#SE
## FIXME
#expect_silent(
cc2<-clusterSingle(testSE,
subsample=FALSE, sequential=FALSE, reduceMethod="PCA",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=TRUE)
#)
expect_equal(primaryCluster(cc2),primaryCluster(cc))
#SCE
## FIXME
#expect_silent(
cc3<-clusterSingle(testSCE,
subsample=FALSE, sequential=FALSE, reduceMethod="PCA",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=TRUE)
#)
expect_equal(primaryCluster(cc3),primaryCluster(cc))
#CE
expect_silent(cc5<-clusterSingle(cc,
subsample=FALSE, sequential=FALSE, reduceMethod="PCA",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),clusterLabel="redoCS"))
expect_equal(primaryCluster(cc5),primaryCluster(cc))
###
#Nothing
###
#matrix
expect_silent(cc<-clusterSingle(countData,
subsample=FALSE, sequential=FALSE, reduceMethod="none",
nDims=NA, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=TRUE))
expect_equal(primaryCluster(cc),c(2,1,2,1,2,1,2,1,3,2,1,3,1,3,1,1,1,3,3,2))
#SE
expect_silent(cc2<-clusterSingle(testSE,
subsample=FALSE, sequential=FALSE, reduceMethod="none",
nDims=NA, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=TRUE))
expect_equal(primaryCluster(cc2),primaryCluster(cc))
#SCE
expect_silent(cc3<-clusterSingle(testSCE,
subsample=FALSE, sequential=FALSE, reduceMethod="none",
nDims=NA, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=TRUE))
expect_equal(primaryCluster(cc3),primaryCluster(cc))
#CE
expect_silent(cc5<-clusterSingle(cc,
subsample=FALSE, sequential=FALSE, reduceMethod="none",
nDims=NA, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),clusterLabel="redoCS"))
expect_equal(primaryCluster(cc5),primaryCluster(cc))
})
test_that("Different options of mainClustering",{
#check errors and warnings
expect_error(clusterSingle(mat,
subsample=FALSE, sequential=TRUE,
seqArgs=list(verbose=FALSE),
isCount=FALSE,
mainClusterArgs=list(clusterFunction="pam")),
"required argument 'k0' is missing for the sequential clustering step")
expect_error(clusterSingle(mat,
subsample=FALSE, sequential=TRUE,
seqArgs=list(verbose=FALSE), isCount=FALSE,
mainClusterArgs=list(clusterFunction="pam","findBestK"==TRUE)),
"required argument 'k0' is missing for the sequential clustering step")
expect_error(clusterSingle(mat, makeMissingDiss=TRUE,
subsample=FALSE, sequential=FALSE,
mainClusterArgs=list(clusterFunction="tight",clusterArgs=list(k=3)),
isCount=FALSE),
"must supply arguments: alpha")
expect_warning(clusterSingle(mat, makeMissingDiss=TRUE,
subsample=FALSE, sequential=FALSE,
mainClusterArgs=list(clusterFunction="tight",
clusterArgs=list(alpha=0.1),findBestK=TRUE),
isCount=FALSE),
"Some arguments passed via mainClusterArgs in mainClustering step do not match the algorithmType of the given ClusterFunction object")
expect_error(clusterSingle(mat,
subsample=FALSE, sequential=FALSE,
distFunction=function(x){abs(cor(t(x)))},
mainClusterArgs=list(clusterFunction="tight", clusterArgs=list(alpha=0.1)),
checkDiss=TRUE,isCount=FALSE),
"Dissimilarity matrix must have zero values on the diagonal")
#nReducedDims=5,k=NA,findBestK=TRUE
#-------
## Check findBestK=TRUE (i.e. post-processing)
#-------
# Error, because needs calculate diss
expect_error(clusterSingle(simData,
subsample=FALSE, makeMissingDiss = FALSE,
mainClusterArgs=list(clusterFunction="pam",
kRange=2:4, findBestK=TRUE)),
"Cannot do requested post processing")
#Should find best K is K=4
expect_message(out<-clusterSingle(simData,
subsample=FALSE,
makeMissingDiss=TRUE,
mainClusterArgs=list(clusterFunction="pam",
kRange=2:4,
findBestK=TRUE)),
"Making nxn dissimilarity matrix in order to do post-processing")
expect_equal(length(tableClusters(out)),4)
## Should be same as if gave distance, since its pam at main clustering
## (and if give distance, shouldn't get message)
d<-as.matrix(dist(t(simData)))
expect_silent(out2<-clusterSingle(d, inputType="diss",
subsample=FALSE,
mainClusterArgs=list(clusterFunction="pam",
kRange=2:4,
findBestK=TRUE)))
expect_equal(primaryCluster(out),out2$clustering)
## Also same as if gave distance to mainClusterArgs
expect_silent(out3<-clusterSingle(simData,
subsample=FALSE,
mainClusterArgs=list(clusterFunction="pam",
kRange=2:4, diss=d,
findBestK=TRUE)))
expect_equal(clusterMatrix(out),clusterMatrix(out3))
})
test_that("Different options of subsampling",{
#Check saving of subsampling matrix
expect_silent(clustSubsample <- clusterSingle(mat,
inputType="X",
subsample=TRUE,
saveSubsamplingMatrix=TRUE,
sequential=FALSE, makeMissingDiss=FALSE, warnings=TRUE,
subsampleArgs=list(resamp.num=3,
clusterFunction="kmeans",clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="hierarchical01",
clusterArgs=list(alpha=.3)),
isCount=FALSE))
expect_equal(NROW(coClustering(clustSubsample)),nSamples(clustSubsample))
expect_false(is.null(coClustering(clustSubsample)))
expect_message(clustSubsample <- clusterSingle(mat,
subsample=TRUE,
saveSubsamplingMatrix=FALSE,
sequential=FALSE,
subsampleArgs=list(resamp.num=3, clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="pam", clusterArgs=list(k=3)),
isCount=FALSE,warnings=TRUE),
"clusterFunction was not given for subsampleClustering -- set to be the same as the mainClustering step")
expect_true(is.null(coClustering(clustSubsample)))
expect_silent(test <- clusterSingle(sce,
subsample=TRUE,
saveSubsamplingMatrix=TRUE,
sequential=FALSE,
subsampleArgs=list(resamp.num=3, clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="pam", clusterArgs=list(k=3)),
isCount=FALSE,warnings=FALSE))
expect_false(is.null(coClustering(test)))
expect_silent(test <- clusterSingle(sce,
subsample=TRUE,
saveSubsamplingMatrix=FALSE,
sequential=FALSE,
subsampleArgs=list(resamp.num=3, clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="pam", clusterArgs=list(k=3)),
isCount=FALSE,warnings=FALSE))
expect_true(is.null(coClustering(test)))
expect_silent(clustSubsampleCE <- clusterSingle(cc,
subsample=TRUE,
saveSubsamplingMatrix=TRUE,
sequential=FALSE,
subsampleArgs=list(resamp.num=3, clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="pam", clusterArgs=list(k=3)),
warnings=FALSE))
expect_false(is.null(coClustering(clustSubsampleCE)))
expect_silent(test <- clusterSingle(cc,
subsample=TRUE,
saveSubsamplingMatrix=FALSE,
sequential=FALSE,
subsampleArgs=list(resamp.num=3, clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="pam", clusterArgs=list(k=3)),
warnings=FALSE))
expect_true(is.null(coClustering(test)))
expect_silent(test <- clusterSingle(clustSubsampleCE,
subsample=TRUE,
sequential=FALSE,
saveSubsamplingMatrix=FALSE,
subsampleArgs=list(resamp.num=3, clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="pam", clusterArgs=list(k=3)),
warnings=FALSE))
expect_equal(coClustering(test),coClustering(clustSubsampleCE))
## get NA values
#set.seed(1045)
# with the new version, we fix NA's in subsampleClustering
# expect_error(clusterSingle(mat,
# subsample=TRUE, sequential=FALSE,
# subsampleArgs=list(resamp.num=20,clusterArgs=list(k=3),clusterFunction="pam",
# classifyMethod="OutOfSample"),
# mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),isCount=FALSE),"NA values found in dissimilarity matrix")
# Check warnings in missing args in subsample -- should borrow from mainClusterArgs .
set.seed(1045)
expect_warning(clusterSingle(mat, subsample=TRUE, sequential=FALSE,
subsampleArgs= list(clusterFunction="pam", resamp.num=3),
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE),
"missing arguments k provided from those in 'mainClusterArgs'")
# Check warnings in missing clusterFunction in subsample -- should borrow from mainClusterArgs .
expect_message(clusterSingle(mat, subsample=TRUE, sequential=FALSE,
subsampleArgs=list(resamp.num=3,clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE),
"a clusterFunction was not given for subsampleClustering")
#Check different function types -- should error out.
expect_error(clusterSingle(mat, subsample=TRUE, sequential=FALSE, subsampleArgs=list(clusterFunction="pam",resamp.num=3), mainClusterArgs=list(clusterFunction="tight",clusterArgs=list(alpha=0.1)), isCount=FALSE),
"must supply arguments: k")
})
test_that("Interactions of mainClustering and subsampling options",{
##This once creates unexpected errors even on master, because subsampleArgs not given (even though mainClusterArgs has k, because k isn't the right one for clusterFunction of main (hierarchical01), they aren't passed on to subsample. Unclear if this is the best option...)
expect_error(ce<-clusterSingle(se,
isCount = FALSE,
reduceMethod="PCA",
nDims=10,
subsample=TRUE,
sequential=FALSE,
mainClusterArgs=list( clusterFunction="hierarchical01",minSize=1, clusterArgs=list(alpha=0.1,k = 4))
)
,"must supply arguments: k"
)
#check totally weird names passed are caught
expect_warning(ce<-clusterSingle(se,
isCount = FALSE,
reduceMethod="PCA",
nDims=10,
subsample=TRUE,
sequential=FALSE,
subsampleArgs=list(clusterArgs=list(k=3)),
mainClusterArgs=list( clusterFunction="hierarchical01",myFunkyName=1, clusterArgs=list(alpha=0.1))
)
,"Some arguments passed via mainClusterArgs in mainClustering step do not match the algorithmType of the given ClusterFunction object: myFunkyName"
)
# check: if provide diss and mainClust okay with that but
# subsampling not, should get error (even if say makeDiss)
expect_error(clustSubsample <- clusterSingle(dissMat,
inputType="diss",
subsample=TRUE,
sequential=FALSE, makeMissingDiss=TRUE, warnings=TRUE,
subsampleArgs=list(resamp.num=3,
clusterFunction="kmeans",clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="hierarchical01",
clusterArgs=list(alpha=.3)),
isCount=FALSE),
"In subsampling clustering step")
# check: should be okay even though main doesn't take type X
# Should NOT trigger calculation of diss matrix.
expect_silent(clustSubsample <- clusterSingle(mat,
inputType="X",
subsample=TRUE,
sequential=FALSE, makeMissingDiss=TRUE, warnings=TRUE,
subsampleArgs=list(resamp.num=3, clusterFunction="kmeans",clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="hierarchical01", clusterArgs=list(alpha=.3)),
isCount=FALSE))
# check: should be okay but must calculate diss
expect_message(clustSubsample <- clusterSingle(mat,
inputType="X",
subsample=TRUE,
sequential=FALSE, makeMissingDiss=TRUE, warnings=TRUE,
subsampleArgs=list(resamp.num=3, classifyMethod="InSample",
clusterFunction="hierarchicalK",clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="hierarchical01",
clusterArgs=list(alpha=.3)),
isCount=FALSE),"Making nxn dissimilarity matrix")
})
test_that("Different passed options of seqCluster",{
#check sequential
expect_silent(clustSeq <- clusterSingle(mat,
subsample=FALSE, sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE))
)
expect_error(clusterSingle(mat,subsample=FALSE, sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE),
"if sequential=TRUE, must give seqArgs so as to identify k0 and beta")
expect_error(clusterSingle(mat,subsample=FALSE, sequential=TRUE,mainClusterArgs=list(clusterFunction="pam"),isCount=FALSE,seqArgs=list(k0=5,verbose=FALSE)), "required argument 'beta' is missing")
expect_error(clusterSingle(mat,subsample=FALSE, sequential=TRUE,mainClusterArgs=list(clusterFunction="pam"),isCount=FALSE,seqArgs=list(beta=0.9,verbose=FALSE)), "required argument 'k0' is missing")
#right clusterFunctions
expect_error(clusterSingle(mat,
mainClusterArgs=list(clusterFunction="kmeans"),
subsample=TRUE, sequential=TRUE,
subsampleArgs=list(clusterFunction="pam",n.sample=40),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE)),
"If choosing subsample=TRUE, the clusterFunction used in the mainClustering step must take input that is a categorical")
expect_error(clusterSingle(dissMat, inputType="diss",
mainClusterArgs=list(clusterFunction="tight"),
subsample=FALSE, sequential=TRUE,
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE)),
"if subsample=FALSE, sequentical clustering can only be implemented with a clusterFunction for mainClustering step with algorithmType 'K'")
#warning if try to set k
expect_warning(clusterSingle(mat,
mainClusterArgs=list(clusterFunction="pam"),
subsample=TRUE, sequential=TRUE,
subsampleArgs=list(clusterFunction="pam",
n.sample=40,clusterArgs=list(k=3)),
isCount=FALSE,seqArgs=list(k0=5,beta=0.9,verbose=FALSE)),
"Setting 'k' in subsampleArgs when sequential=TRUE is called will have no effect.")
expect_warning(clusterSingle(mat,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
subsample=FALSE, sequential=TRUE,
subsampleArgs=list(clusterFunction="pam",n.sample=40),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE)),
"Setting 'k' in mainClusterArgs when sequential clustering is requested will have no effect.")
## Check warnings for esoteric arguments
expect_error(clusterSingle(mat,subsample=FALSE, sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE,top.can=5.5)),
"If top.can parameter of seqArgs is not a whole number, must be between 0 and 1.")
expect_error(clusterSingle(mat,subsample=FALSE, sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE,top.can=-1)),
"top.can parameter of seqArgs cannot be NA or negative value")
expect_error(clusterSingle(mat,subsample=FALSE, sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE,remain.n=3.4)),
"remain.n argument of seqArgs must be a positive whole number")
expect_error(clusterSingle(mat,subsample=FALSE, sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE,k.min=3.4)),
"k.min argument of seqArgs must be a positive whole number")
expect_error(clusterSingle(mat,subsample=FALSE, sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE,k.max=20.5)),
"k.max argument of seqArgs must be a positive whole number")
expect_silent(clusterSingle(mat,subsample=FALSE, sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE,top.can=0.1)))
})
test_that("Different direct options of `clusterSingle` ", {
#check isCount
expect_silent(clusterSingle(smSimCount,
subsample=FALSE, sequential=FALSE,
mainClusterArgs=list(clusterArgs=list(k=3), clusterFunction="pam"),
isCount=TRUE) )
#suppressWarnings because in addition to error, R prints warning about NAs
expect_error(suppressWarnings(clusterSingle(smSimData,
subsample=FALSE, sequential=FALSE,
mainClusterArgs=list(clusterArgs=list(k=3), clusterFunction="pam"),isCount=TRUE)),"User-supplied `transFun` produces NA values",info="test error handling for isCount=TRUE when can't take log")
})
test_that("`clusterSingle` preserves the colData and rowData of SE", {
expect_silent(cl<-clusterSingle(se,
subsample=FALSE, sequential=FALSE,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_equal(colData(cl),colData(se))
expect_equal(rownames(cl),rownames(se))
expect_equal(colnames(cl),colnames(se))
expect_equal(metadata(cl),metadata(se))
expect_equal(rowData(cl),rowData(se))
})
test_that("`clusterSingle` works with parallel subsampling", {
## Windows does not support mclapply
skip_on_os("windows")
expect_silent(clustSubsample2 <-
clusterSingle(mat, subsample=TRUE,sequential=FALSE,
subsampleArgs=list(clusterFunction="kmeans",resamp.num=3,
clusterArgs=list(k=3),ncores=2),
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_silent(clustSubsample1 <-
clusterSingle(mat, subsample=TRUE,sequential=FALSE,
subsampleArgs=list(clusterFunction="kmeans",resamp.num=3,
clusterArgs=list(k=3),ncores=1),
mainClusterArgs=list(clusterFunction="pam", clusterArgs=list(k=3)),
isCount=FALSE))
})
epurdom/clusterCells documentation built on April 28, 2024, 8:14 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
context("clusterSingle")
test_that("`clusterSingle` works with matrix, ClusterExperiment objects,
SummarizedExperiments", {
#---
#Matrix
#---
expect_silent(clustNothing <-
clusterSingle(mat, inputType="X",
subsample=FALSE, sequential=FALSE,
mainClusterArgs=list(clusterArgs=list(k=3),
clusterFunction="pam"),
isCount=FALSE))
expect_equal(clusterLabels(clustNothing),"clusterSingle")
expect_is(clustNothing, "ClusterExperiment")
expect_is(clustNothing, "SummarizedExperiment")
truth<- unname(cluster::pam(x=t(mat),cluster.only=TRUE,k=3))
#in case change in sim, make sure same as from previous tests
expect_equal(truth,c(1,2,3,1,3,1,1,1,1,2,2,1,1,1,1))
expect_equal(primaryCluster(clustNothing),truth)
#test clusterLabel
expect_silent(clustNothing2 <- clusterSingle(mat,
inputType="X",
mainClusterArgs= list(clusterArgs=list(k=3), clusterFunction="pam"),
subsample=FALSE,
sequential=FALSE, isCount=FALSE,
clusterLabel="myownClustering"))
expect_equal(clusterLabels(clustNothing2),"myownClustering")
#test default 01 distance
expect_silent(clusterSingle(inputMatrix=dissMat, inputType="diss", mainClusterArgs= list(clusterArgs=list(alpha=0.1),clusterFunction="tight"),
subsample=FALSE, sequential=FALSE,
isCount=FALSE))
#test default K distance
expect_silent(clusterSingle(dissMat, inputType="diss", mainClusterArgs= list(clusterArgs=list(k=3),clusterFunction="hierarchicalK"),subsample=FALSE, sequential=FALSE, isCount=FALSE))
#---
#SE/SCE
#---
###Apply to SE
expect_silent(clustNothing2 <- clusterSingle(se,
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE))
expect_equal(clusterMatrix(clustNothing2), clusterMatrix(clustNothing))
###Apply to SCE
expect_silent(clustNothing3 <- clusterSingle(sce, mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE))
expect_equal(clusterMatrix(clustNothing2), clusterMatrix(clustNothing3))
expect_silent(clustNothing3 <- clusterSingle(sce,
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE))
#---
#CE
#---
#test running on ClusterExperiment Object -- should add the new clustering
expect_error(clustNothing3 <- clusterSingle(clustNothing2, mainClusterArgs=list(clusterArgs=list(k=4),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE,
isCount=FALSE),"setting 'transFun' or 'isCount' for a 'ClusterExperiment' is not allowed")
expect_silent(clustNothing3 <- clusterSingle(clustNothing2, mainClusterArgs=list(clusterArgs=list(k=4),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE ))
expect_equal(NCOL(clusterMatrix(clustNothing3)),2)
expect_equal(length(table(primaryCluster(clustNothing3))),4,info="Check reset primary cluster after run clusterSingle")
})
test_that("`clusterSingle` warnings/errors work", {
expect_error(clusterSingle(mat, inputType="X",
mainClusterArgs= list(clusterArgs=list(alpha=0.1),
clusterFunction="tight"),
distFunction=function(x){dist(x,method="manhattan")},
subsample=FALSE, checkDiss=TRUE,
sequential=FALSE,isCount=FALSE),
"distance function must give values between 0 and 1")
#test warn with wrong arguments
expect_warning(clusterSingle(dissMat, inputType="diss", mainClusterArgs= list(clusterArgs=list(k=3,alpha=0.1),clusterFunction="tight"),
subsample=FALSE, sequential=FALSE,
,isCount=FALSE),"arguments passed via clusterArgs to the clustering function tight are not all applicable")
#test turning off warning
expect_silent(clusterSingle(dissMat,
inputType="diss",
mainClusterArgs= list(clusterArgs=list(k=3,alpha=0.1), clusterFunction="tight"),
warnings=FALSE,
subsample=FALSE,
sequential=FALSE,
isCount=FALSE))
})
test_that("`clusterSingle` works with reduceMethod", {
#existing dim reduce values
expect_silent(clustNothing4 <- clusterSingle(sceSimDataDimRed,
reduceMethod="PCA", nDims=3,
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE))
expect_equal(sort(reducedDimNames(clustNothing4)),sort(c("PCA","TSNE")))
expect_silent(clustNothing5 <- clusterSingle(t(reducedDim(sceSimDataDimRed,"PCA")[,1:3]),
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE))
expect_equal(clusterMatrix(clustNothing4), clusterMatrix(clustNothing5))
#test nDims=NA
expect_silent(clustNothing1 <- clusterSingle(sceSimDataDimRed,
reduceMethod="PCA", nDims=NA,
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE))
expect_equal(sort(reducedDimNames(clustNothing4)),sort(c("PCA","TSNE")))
expect_silent(clustNothing2 <- clusterSingle(t(reducedDim(sceSimDataDimRed,"PCA")),
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE))
expect_equal(clusterMatrix(clustNothing1), clusterMatrix(clustNothing2))
#blank sceSimData object
##FIXME
#expect_silent(
clustNothing6 <- clusterSingle(sceSimData,
reduceMethod="PCA", nDims=3,
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE)
#)
expect_equal(sort(reducedDimNames(clustNothing6)),sort(c("PCA")))
expect_silent(clustNothing7 <- clusterSingle(t(reducedDim(sceSimDataDimRed,"PCA")[,1:3]),
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE))
expect_equal(clusterMatrix(clustNothing6), clusterMatrix(clustNothing7))
## FIXME
#expect_silent(
clustNothing8 <- clusterSingle(seSimData,
reduceMethod="PCA", nDims=3,
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE)
#)
expect_equal(sort(reducedDimNames(clustNothing8)),sort(c("PCA")))
expect_equal(clusterMatrix(clustNothing8), clusterMatrix(clustNothing7))
#check consistency
expect_equal(primaryCluster(clustNothing8),rep(c(1,2,3),each=100))
## FIXME
#expect_silent(
clusterSingle(sceSimData,
reduceMethod="PCA", nDims=NA,
mainClusterArgs=list(clusterArgs=list(k=3),clusterFunction="pam"),
subsample=FALSE, sequential=FALSE, isCount=FALSE)
#)
})
test_that("all built-in clustering methods",{
#########
## Put all clusterSingle tests that loop over methods HERE!
#########
kClusterArgs<-list(k=3)
aClusterArgs<-list(alpha=0.1)
set.seed(3325)
biggerMat<-matrix(data=rnorm(20*100), ncol=100)
diss<-as.matrix(dist(t(mat)))
biggerDiss<-as.matrix(dist(t(biggerMat)))
getInputMatrix<- function(cf, big=FALSE){
if("X" %in% inputType(cf)){
inputType<-"X"
if(!big) inputMat<-mat
else inputMat<-biggerMat
}
else if("diss" %in% inputType(cf)){
inputType<-"diss"
if(!big) inputMat<-diss
else inputMat<- biggerDiss
}
else stop("error, don't have test for cat type matrix")
return(list(inputType=inputType,inputMatrix=inputMat))
}
## ---------
# check with main clustering
## ---------
#FIXME: Is this duplicative of previous test in hdf5?
for(cf in listBuiltInFunctions()){
algType<-algorithmType(cf)
clustArgs<-switch(algType,
"K"=kClusterArgs,
"01"=aClusterArgs
)
input<-getInputMatrix(cf)
expect_silent(clusterSingle(inputMat=input$inputMatrix,
inputType=input$inputType,
mainClusterArgs= list(clusterArgs=clustArgs, clusterFunction=cf),
subsample=FALSE, sequential=FALSE,isCount=FALSE)
)
if(algType=="K"){
input<-getInputMatrix(cf,big=TRUE)
expect_silent(clusterSingle(inputMat=input$inputMatrix,
inputType=input$inputType,
mainClusterArgs= list(clusterArgs=clustArgs,
diss=switch(algType,"01"=NULL,"K"=biggerDiss),
clusterFunction=cf,findBestK=TRUE,removeSil=TRUE),
subsample=FALSE, sequential=FALSE,isCount=FALSE))
}
}
##------
## Check with subsampling
##------
# also force to be opposite type from mainClusterArgs
for(cf in listBuiltInFunctions()){
algType<-algorithmType(cf)
if(algType=="K"){
mainArgs<-list(clusterFunction="hierarchical01",
clusterArgs=aClusterArgs)
subArgs<-list(clusterFunction=cf,resamp.num=20, clusterArgs=kClusterArgs)
}
if(algType=="01"){
mainArgs<-list(clusterFunction="pam", clusterArgs=kClusterArgs)
subArgs<-list(clusterFunction=cf,resamp.num=20,
clusterArgs=aClusterArgs)
}
input<-getInputMatrix(cf,big=TRUE)
set.seed(1045)
##In-sample
expect_silent(clusterSingle(inputMat=input$inputMatrix,
inputType=input$inputType,
subsample=TRUE,
sequential=FALSE,
makeMissingDiss=FALSE,
subsampleArgs=c(list(
classifyMethod="InSample"),subArgs),
mainClusterArgs=mainArgs,isCount=FALSE))
if(!is.null(getBuiltInFunction(cf)@classifyFUN)){
#All
set.seed(1045)
expect_silent(clusterSingle(inputMat=input$inputMatrix,
inputType=input$inputType, subsample=TRUE, sequential=FALSE,
subsampleArgs=c(list(classifyMethod="All"),subArgs),
mainClusterArgs=mainArgs,isCount=FALSE))
#OutOfSample
set.seed(1045)
expect_silent(clusterSingle(inputMat=input$inputMatrix,
inputType=input$inputType, subsample=TRUE, sequential=FALSE,
subsampleArgs=c(list(classifyMethod="OutOfSample"),subArgs),
mainClusterArgs=mainArgs,isCount=FALSE))
}
}
# --------
# Seq cluster -- only works for K methods
# --------
for(cf in listBuiltInTypeK()){
algType<-algorithmType(cf)
subArgs<-list(clusterFunction="pam",n.sample=40)
seqArgs<-list(k0=5,beta=0.9,verbose=FALSE)
input<-getInputMatrix(cf,big=FALSE)
#check subsample=FALSE
expect_silent(clusterSingle(input$inputMatrix, inputType=input$inputType,
mainClusterArgs=list(clusterFunction=cf),
subsample=FALSE, sequential=TRUE,
isCount=FALSE,seqArgs=seqArgs))
#check subsample=TRUE
if("cat" %in% inputType(cf))
expect_silent(clusterSingle(input$inputMatrix,
inputType=input$inputType,
mainClusterArgs=list(clusterFunction=cf),
subsample=TRUE, sequential=TRUE,
subsampleArgs=subArgs,
isCount=FALSE,seqArgs=seqArgs))
}
#-------
## Check that they work with hdf5
#-------
seedValue<-571839
for(cf in listBuiltInFunctions()){
#print(cf)
clArgs<-switch(algorithmType(cf),
"K"=kClusterArgs,
"01"=aClusterArgs)
### Annoyingly, have to separately for "X" and "diss"
### Because for this, no way to give it the diss and still test the objects
if("X" %in% inputType(cf)){
set.seed(seedValue)
expect_silent(clust1<-clusterSingle(sceSimDataDimRed,
reduceMethod = "none",
mainClusterArgs= list(clusterArgs=clArgs,
clusterFunction=cf),
subsample=FALSE,
makeMissingDiss=TRUE,
sequential=FALSE,
isCount=FALSE)
)
set.seed(seedValue)
expect_silent(clust2<-clusterSingle(hdfSCE,
reduceMethod = "none",
mainClusterArgs= list(clusterArgs=clArgs,
clusterFunction=cf),
subsample=FALSE,
makeMissingDiss=TRUE,
sequential=FALSE,
isCount=FALSE)
)
set.seed(seedValue)
expect_silent(clust3<-clusterSingle(hdfObj,
reduceMethod = "none",
mainClusterArgs= list(clusterArgs=clArgs,
clusterFunction=cf),
subsample=FALSE,
makeMissingDiss=TRUE,
sequential=FALSE,
isCount=FALSE)
)
set.seed(seedValue)
expect_silent(clust4<-clusterSingle(assay(hdfSCE),
reduceMethod = "none",
mainClusterArgs= list(clusterArgs=clArgs,
clusterFunction=cf),
subsample=FALSE,
makeMissingDiss=TRUE,
sequential=FALSE,
isCount=FALSE)
)
}
else{
set.seed(seedValue)
expect_message(clust1<-clusterSingle(sceSimDataDimRed,
reduceMethod = "none",
mainClusterArgs= list(clusterArgs=clArgs,
clusterFunction=cf),
subsample=FALSE,
makeMissingDiss=TRUE,
sequential=FALSE,
isCount=FALSE),"Note: Making nxn dissimilarity matrix."
)
set.seed(seedValue)
expect_message(clust2<-clusterSingle(hdfSCE,
reduceMethod = "none",
mainClusterArgs= list(clusterArgs=clArgs,
clusterFunction=cf),
subsample=FALSE,
makeMissingDiss=TRUE,
sequential=FALSE,
isCount=FALSE),"Note: Making nxn dissimilarity matrix."
)
set.seed(seedValue)
expect_message(clust3<-clusterSingle(hdfObj,
reduceMethod = "none",
mainClusterArgs= list(clusterArgs=clArgs,
clusterFunction=cf),
subsample=FALSE,
makeMissingDiss=TRUE,
sequential=FALSE,
isCount=FALSE),"Note: Making nxn dissimilarity matrix."
)
set.seed(seedValue)
expect_message(clust4<-clusterSingle(assay(hdfSCE),
reduceMethod = "none",
mainClusterArgs= list(clusterArgs=clArgs,
clusterFunction=cf),
subsample=FALSE,
makeMissingDiss=TRUE,
sequential=FALSE,
isCount=FALSE),"Note: Making nxn dissimilarity matrix."
)
}
expect_equal(clusterMatrix(clust1) ,clusterMatrix(clust2))
expect_equal(clusterMatrix(clust1) ,clusterMatrix(clust3))
expect_equal(clusterMatrix(clust1) ,clusterMatrix(clust4))
}
})
test_that("`clusterSingle` works with hdf5Matrix",{
# These are slow!
skip_on_os("windows")
skip_on_os("mac")
####Test sequential option,
expect_silent(clustSeq <- clusterSingle(hdfObj,
reduceMethod="none",
subsample=FALSE,
sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE)
))
expect_silent(clustSeq <- clusterSingle(assay(hdfObj),
reduceMethod="none",
subsample=FALSE, sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE)))
####Test subsample option
expect_silent(clusterSingle(hdfObj,
reduceMethod="none",
subsample=TRUE,
sequential=FALSE,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE,
subsampleArgs=list(clusterFunction="pam",resamp.num=3,
clusterArgs=list(k=3))
)
)
expect_silent(clusterSingle(assay(hdfObj), reduceMethod="none",
subsample=TRUE, sequential=FALSE,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE,
subsampleArgs=list(clusterFunction="pam",resamp.num=3, clusterArgs=list(k=3))
)
)
})
test_that("`clusterSingle` works with filtering", {
####Check built in functions ####
for(fs in listBuiltInFilterStats()){
expect_silent(clusterSingle(mat,
subsample=FALSE,
sequential=FALSE,
reduceMethod=fs,
nDims=3,
mainClusterArgs=list(clusterFunction="pam", clusterArgs=list(k=3)),
isCount=FALSE))
}
expect_warning(clusterSingle(mat,
subsample=FALSE,
sequential=FALSE,
reduceMethod="var",
nDims=NROW(mat)+1,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE),
"the number of most features requested after filtering is either missing or larger than the number of features. Will not do any filtering")
expect_warning(clusterSingle(mat,
subsample=FALSE, sequential=FALSE,
reduceMethod="none",nDims=3,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE),
"specifying nDims has no effect if reduceMethod==`none`")
#check returns filterStats
expect_silent(cc <- clusterSingle(mat,
subsample=FALSE, sequential=FALSE,
reduceMethod="var", nDims=3,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)), isCount=FALSE))
expect_equal(sort(filterNames(cc)),"var")
expect_silent(cc2 <- clusterSingle(se,
subsample=FALSE, sequential=FALSE,
reduceMethod="var", nDims=3,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)), isCount=FALSE))
expect_equal(sort(filterNames(cc2)),c("b","var"))
expect_silent(cc3 <- clusterSingle(sceSimData,
subsample=FALSE, sequential=FALSE,
reduceMethod="var", nDims=3,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)), isCount=FALSE))
expect_equal(sort(filterNames(cc3)),c("b","var"))
#test using existing values
expect_silent(cc4 <- clusterSingle(sceFull,
subsample=FALSE, sequential=FALSE,
reduceMethod="var", nDims=3,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)), isCount=FALSE))
expect_equal(sort(filterNames(cc4)),sort(c("var","b","Filter1","Filter2")))
})
test_that("`clusterSingle` consistent results (no transformation)", {
#----
#check get the same result from previous times when transform the data
#----
#make it big enough can do pca and filter...
contData<-simData[,1:20]
testSE<-SummarizedExperiment(contData)
testSCE<-as(testSE,"SingleCellExperiment")
expectTrans1<-round(contData[1,],2)
###
#var filter
###
#matrix
expect_silent(cc<-clusterSingle(contData,
subsample=FALSE, sequential=FALSE, reduceMethod="var",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_equal(primaryCluster(cc),c(3,1,1,3,2,2,3,2,3,1,2,2,1,3,3,2,1,1,2,1))
expect_equal(round(transformData(cc)[1,],2), expectTrans1)
expect_silent(cc<-clusterSingle(contData,
subsample=FALSE, sequential=FALSE, reduceMethod="var",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_equal(primaryCluster(cc),c(3,1,1,3,2,2,3,2,3,1,2,2,1,3,3,2,1,1,2,1))
#SE
expect_silent(cc2<-clusterSingle(testSE,
subsample=FALSE, sequential=FALSE, reduceMethod="var",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_equal(primaryCluster(cc2),primaryCluster(cc))
#SCE
expect_silent(cc3<-clusterSingle(testSCE,
subsample=FALSE, sequential=FALSE, reduceMethod="var",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_equal(primaryCluster(cc3),primaryCluster(cc))
#CE
expect_silent(cc5<-clusterSingle(cc,
subsample=FALSE, sequential=FALSE, reduceMethod="var",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),clusterLabel="redoCS"))
expect_equal(primaryCluster(cc5),primaryCluster(cc))
###
#PCA
###
#matrix
expect_silent(cc<-clusterSingle(contData,
subsample=FALSE, sequential=FALSE, reduceMethod="PCA",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_equal(primaryCluster(cc),c(3,1,3,1,3,2,1,1,1,3,2,2,2,3,1,2,2,1,2,3))
expect_equal(round(transformData(cc)[1,],2), expectTrans1)
#SE
## FIXME
#expect_silent(
cc2<-clusterSingle(testSE,
subsample=FALSE, sequential=FALSE, reduceMethod="PCA",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE)
#)
expect_equal(primaryCluster(cc2),primaryCluster(cc))
#SCE
## FIXME
#expect_silent(
cc3<-clusterSingle(testSCE,
subsample=FALSE, sequential=FALSE, reduceMethod="PCA",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE)
#)
expect_equal(primaryCluster(cc3),primaryCluster(cc))
#CE
expect_silent(cc5<-clusterSingle(cc,
subsample=FALSE, sequential=FALSE, reduceMethod="PCA",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),clusterLabel="redoCS"))
expect_equal(primaryCluster(cc5),primaryCluster(cc))
###
#Nothing
###
#matrix
expect_silent(cc<-clusterSingle(contData,
subsample=FALSE, sequential=FALSE, reduceMethod="none",
nDims=NA, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_equal(primaryCluster(cc),c(3,2,1,3,2,1,3,2,3,2,1,2,1,1,3,1,1,1,2,2))
expect_equal(round(transformData(cc)[1,],2), expectTrans1)
#SE
expect_silent(cc2<-clusterSingle(testSE,
subsample=FALSE, sequential=FALSE, reduceMethod="none",
nDims=NA, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_equal(primaryCluster(cc2),primaryCluster(cc))
#SCE
expect_silent(cc3<-clusterSingle(testSCE,
subsample=FALSE, sequential=FALSE, reduceMethod="none",
nDims=NA, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_equal(primaryCluster(cc3),primaryCluster(cc))
#CE
expect_silent(cc5<-clusterSingle(cc,
subsample=FALSE, sequential=FALSE, reduceMethod="none",
nDims=NA, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),clusterLabel="redoCS"))
expect_equal(primaryCluster(cc5),primaryCluster(cc))
})
test_that("`clusterSingle` consistent results (with transformation)", {
#----
#check get the same result from previous times when transform the data
#----
#make it big enough can do pca and filter...
countData<-simCount[,1:20]
testSE<-SummarizedExperiment(countData)
testSCE<-as(testSE,"SingleCellExperiment")
expectTrans1<-round(log2(countData[1,]+1),2)
###
#var filter
###
#matrix
expect_silent(cc<-clusterSingle(countData,
subsample=FALSE, sequential=FALSE, reduceMethod="var",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=TRUE))
expect_equal(primaryCluster(cc),c(1,1,2,1,2,1,2,2,3,2,1,3,1,3,1,1,2,3,2,2))
expect_equal(round(transformData(cc)[1,],2), expectTrans1)
#SE
expect_silent(cc2<-clusterSingle(testSE,
subsample=FALSE, sequential=FALSE, reduceMethod="var",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=TRUE))
expect_equal(primaryCluster(cc2),primaryCluster(cc))
#SCE
expect_silent(cc3<-clusterSingle(testSCE,
subsample=FALSE, sequential=FALSE, reduceMethod="var",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=TRUE))
expect_equal(primaryCluster(cc3),primaryCluster(cc))
#CE
expect_silent(cc5<-clusterSingle(cc,
subsample=FALSE, sequential=FALSE, reduceMethod="var",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),clusterLabel="redoCS"))
expect_equal(primaryCluster(cc5),primaryCluster(cc))
###
#PCA
###
#matrix
expect_silent(cc<-clusterSingle(countData,
subsample=FALSE, sequential=FALSE, reduceMethod="PCA",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=TRUE))
expect_equal(primaryCluster(cc),c(3,2,1,1,3,1,3,1,1,1,1,2,1,2,1,1,1,2,2,1))
expect_equal(round(transformData(cc)[1,],2), expectTrans1)
#SE
## FIXME
#expect_silent(
cc2<-clusterSingle(testSE,
subsample=FALSE, sequential=FALSE, reduceMethod="PCA",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=TRUE)
#)
expect_equal(primaryCluster(cc2),primaryCluster(cc))
#SCE
## FIXME
#expect_silent(
cc3<-clusterSingle(testSCE,
subsample=FALSE, sequential=FALSE, reduceMethod="PCA",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=TRUE)
#)
expect_equal(primaryCluster(cc3),primaryCluster(cc))
#CE
expect_silent(cc5<-clusterSingle(cc,
subsample=FALSE, sequential=FALSE, reduceMethod="PCA",
nDims=3, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),clusterLabel="redoCS"))
expect_equal(primaryCluster(cc5),primaryCluster(cc))
###
#Nothing
###
#matrix
expect_silent(cc<-clusterSingle(countData,
subsample=FALSE, sequential=FALSE, reduceMethod="none",
nDims=NA, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=TRUE))
expect_equal(primaryCluster(cc),c(2,1,2,1,2,1,2,1,3,2,1,3,1,3,1,1,1,3,3,2))
#SE
expect_silent(cc2<-clusterSingle(testSE,
subsample=FALSE, sequential=FALSE, reduceMethod="none",
nDims=NA, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=TRUE))
expect_equal(primaryCluster(cc2),primaryCluster(cc))
#SCE
expect_silent(cc3<-clusterSingle(testSCE,
subsample=FALSE, sequential=FALSE, reduceMethod="none",
nDims=NA, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=TRUE))
expect_equal(primaryCluster(cc3),primaryCluster(cc))
#CE
expect_silent(cc5<-clusterSingle(cc,
subsample=FALSE, sequential=FALSE, reduceMethod="none",
nDims=NA, mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),clusterLabel="redoCS"))
expect_equal(primaryCluster(cc5),primaryCluster(cc))
})
test_that("Different options of mainClustering",{
#check errors and warnings
expect_error(clusterSingle(mat,
subsample=FALSE, sequential=TRUE,
seqArgs=list(verbose=FALSE),
isCount=FALSE,
mainClusterArgs=list(clusterFunction="pam")),
"required argument 'k0' is missing for the sequential clustering step")
expect_error(clusterSingle(mat,
subsample=FALSE, sequential=TRUE,
seqArgs=list(verbose=FALSE), isCount=FALSE,
mainClusterArgs=list(clusterFunction="pam","findBestK"==TRUE)),
"required argument 'k0' is missing for the sequential clustering step")
expect_error(clusterSingle(mat, makeMissingDiss=TRUE,
subsample=FALSE, sequential=FALSE,
mainClusterArgs=list(clusterFunction="tight",clusterArgs=list(k=3)),
isCount=FALSE),
"must supply arguments: alpha")
expect_warning(clusterSingle(mat, makeMissingDiss=TRUE,
subsample=FALSE, sequential=FALSE,
mainClusterArgs=list(clusterFunction="tight",
clusterArgs=list(alpha=0.1),findBestK=TRUE),
isCount=FALSE),
"Some arguments passed via mainClusterArgs in mainClustering step do not match the algorithmType of the given ClusterFunction object")
expect_error(clusterSingle(mat,
subsample=FALSE, sequential=FALSE,
distFunction=function(x){abs(cor(t(x)))},
mainClusterArgs=list(clusterFunction="tight", clusterArgs=list(alpha=0.1)),
checkDiss=TRUE,isCount=FALSE),
"Dissimilarity matrix must have zero values on the diagonal")
#nReducedDims=5,k=NA,findBestK=TRUE
#-------
## Check findBestK=TRUE (i.e. post-processing)
#-------
# Error, because needs calculate diss
expect_error(clusterSingle(simData,
subsample=FALSE, makeMissingDiss = FALSE,
mainClusterArgs=list(clusterFunction="pam",
kRange=2:4, findBestK=TRUE)),
"Cannot do requested post processing")
#Should find best K is K=4
expect_message(out<-clusterSingle(simData,
subsample=FALSE,
makeMissingDiss=TRUE,
mainClusterArgs=list(clusterFunction="pam",
kRange=2:4,
findBestK=TRUE)),
"Making nxn dissimilarity matrix in order to do post-processing")
expect_equal(length(tableClusters(out)),4)
## Should be same as if gave distance, since its pam at main clustering
## (and if give distance, shouldn't get message)
d<-as.matrix(dist(t(simData)))
expect_silent(out2<-clusterSingle(d, inputType="diss",
subsample=FALSE,
mainClusterArgs=list(clusterFunction="pam",
kRange=2:4,
findBestK=TRUE)))
expect_equal(primaryCluster(out),out2$clustering)
## Also same as if gave distance to mainClusterArgs
expect_silent(out3<-clusterSingle(simData,
subsample=FALSE,
mainClusterArgs=list(clusterFunction="pam",
kRange=2:4, diss=d,
findBestK=TRUE)))
expect_equal(clusterMatrix(out),clusterMatrix(out3))
})
test_that("Different options of subsampling",{
#Check saving of subsampling matrix
expect_silent(clustSubsample <- clusterSingle(mat,
inputType="X",
subsample=TRUE,
saveSubsamplingMatrix=TRUE,
sequential=FALSE, makeMissingDiss=FALSE, warnings=TRUE,
subsampleArgs=list(resamp.num=3,
clusterFunction="kmeans",clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="hierarchical01",
clusterArgs=list(alpha=.3)),
isCount=FALSE))
expect_equal(NROW(coClustering(clustSubsample)),nSamples(clustSubsample))
expect_false(is.null(coClustering(clustSubsample)))
expect_message(clustSubsample <- clusterSingle(mat,
subsample=TRUE,
saveSubsamplingMatrix=FALSE,
sequential=FALSE,
subsampleArgs=list(resamp.num=3, clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="pam", clusterArgs=list(k=3)),
isCount=FALSE,warnings=TRUE),
"clusterFunction was not given for subsampleClustering -- set to be the same as the mainClustering step")
expect_true(is.null(coClustering(clustSubsample)))
expect_silent(test <- clusterSingle(sce,
subsample=TRUE,
saveSubsamplingMatrix=TRUE,
sequential=FALSE,
subsampleArgs=list(resamp.num=3, clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="pam", clusterArgs=list(k=3)),
isCount=FALSE,warnings=FALSE))
expect_false(is.null(coClustering(test)))
expect_silent(test <- clusterSingle(sce,
subsample=TRUE,
saveSubsamplingMatrix=FALSE,
sequential=FALSE,
subsampleArgs=list(resamp.num=3, clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="pam", clusterArgs=list(k=3)),
isCount=FALSE,warnings=FALSE))
expect_true(is.null(coClustering(test)))
expect_silent(clustSubsampleCE <- clusterSingle(cc,
subsample=TRUE,
saveSubsamplingMatrix=TRUE,
sequential=FALSE,
subsampleArgs=list(resamp.num=3, clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="pam", clusterArgs=list(k=3)),
warnings=FALSE))
expect_false(is.null(coClustering(clustSubsampleCE)))
expect_silent(test <- clusterSingle(cc,
subsample=TRUE,
saveSubsamplingMatrix=FALSE,
sequential=FALSE,
subsampleArgs=list(resamp.num=3, clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="pam", clusterArgs=list(k=3)),
warnings=FALSE))
expect_true(is.null(coClustering(test)))
expect_silent(test <- clusterSingle(clustSubsampleCE,
subsample=TRUE,
sequential=FALSE,
saveSubsamplingMatrix=FALSE,
subsampleArgs=list(resamp.num=3, clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="pam", clusterArgs=list(k=3)),
warnings=FALSE))
expect_equal(coClustering(test),coClustering(clustSubsampleCE))
## get NA values
#set.seed(1045)
# with the new version, we fix NA's in subsampleClustering
# expect_error(clusterSingle(mat,
# subsample=TRUE, sequential=FALSE,
# subsampleArgs=list(resamp.num=20,clusterArgs=list(k=3),clusterFunction="pam",
# classifyMethod="OutOfSample"),
# mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),isCount=FALSE),"NA values found in dissimilarity matrix")
# Check warnings in missing args in subsample -- should borrow from mainClusterArgs .
set.seed(1045)
expect_warning(clusterSingle(mat, subsample=TRUE, sequential=FALSE,
subsampleArgs= list(clusterFunction="pam", resamp.num=3),
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE),
"missing arguments k provided from those in 'mainClusterArgs'")
# Check warnings in missing clusterFunction in subsample -- should borrow from mainClusterArgs .
expect_message(clusterSingle(mat, subsample=TRUE, sequential=FALSE,
subsampleArgs=list(resamp.num=3,clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE),
"a clusterFunction was not given for subsampleClustering")
#Check different function types -- should error out.
expect_error(clusterSingle(mat, subsample=TRUE, sequential=FALSE, subsampleArgs=list(clusterFunction="pam",resamp.num=3), mainClusterArgs=list(clusterFunction="tight",clusterArgs=list(alpha=0.1)), isCount=FALSE),
"must supply arguments: k")
})
test_that("Interactions of mainClustering and subsampling options",{
##This once creates unexpected errors even on master, because subsampleArgs not given (even though mainClusterArgs has k, because k isn't the right one for clusterFunction of main (hierarchical01), they aren't passed on to subsample. Unclear if this is the best option...)
expect_error(ce<-clusterSingle(se,
isCount = FALSE,
reduceMethod="PCA",
nDims=10,
subsample=TRUE,
sequential=FALSE,
mainClusterArgs=list( clusterFunction="hierarchical01",minSize=1, clusterArgs=list(alpha=0.1,k = 4))
)
,"must supply arguments: k"
)
#check totally weird names passed are caught
expect_warning(ce<-clusterSingle(se,
isCount = FALSE,
reduceMethod="PCA",
nDims=10,
subsample=TRUE,
sequential=FALSE,
subsampleArgs=list(clusterArgs=list(k=3)),
mainClusterArgs=list( clusterFunction="hierarchical01",myFunkyName=1, clusterArgs=list(alpha=0.1))
)
,"Some arguments passed via mainClusterArgs in mainClustering step do not match the algorithmType of the given ClusterFunction object: myFunkyName"
)
# check: if provide diss and mainClust okay with that but
# subsampling not, should get error (even if say makeDiss)
expect_error(clustSubsample <- clusterSingle(dissMat,
inputType="diss",
subsample=TRUE,
sequential=FALSE, makeMissingDiss=TRUE, warnings=TRUE,
subsampleArgs=list(resamp.num=3,
clusterFunction="kmeans",clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="hierarchical01",
clusterArgs=list(alpha=.3)),
isCount=FALSE),
"In subsampling clustering step")
# check: should be okay even though main doesn't take type X
# Should NOT trigger calculation of diss matrix.
expect_silent(clustSubsample <- clusterSingle(mat,
inputType="X",
subsample=TRUE,
sequential=FALSE, makeMissingDiss=TRUE, warnings=TRUE,
subsampleArgs=list(resamp.num=3, clusterFunction="kmeans",clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="hierarchical01", clusterArgs=list(alpha=.3)),
isCount=FALSE))
# check: should be okay but must calculate diss
expect_message(clustSubsample <- clusterSingle(mat,
inputType="X",
subsample=TRUE,
sequential=FALSE, makeMissingDiss=TRUE, warnings=TRUE,
subsampleArgs=list(resamp.num=3, classifyMethod="InSample",
clusterFunction="hierarchicalK",clusterArgs=list(k=3)),
mainClusterArgs=list(clusterFunction="hierarchical01",
clusterArgs=list(alpha=.3)),
isCount=FALSE),"Making nxn dissimilarity matrix")
})
test_that("Different passed options of seqCluster",{
#check sequential
expect_silent(clustSeq <- clusterSingle(mat,
subsample=FALSE, sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE))
)
expect_error(clusterSingle(mat,subsample=FALSE, sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE),
"if sequential=TRUE, must give seqArgs so as to identify k0 and beta")
expect_error(clusterSingle(mat,subsample=FALSE, sequential=TRUE,mainClusterArgs=list(clusterFunction="pam"),isCount=FALSE,seqArgs=list(k0=5,verbose=FALSE)), "required argument 'beta' is missing")
expect_error(clusterSingle(mat,subsample=FALSE, sequential=TRUE,mainClusterArgs=list(clusterFunction="pam"),isCount=FALSE,seqArgs=list(beta=0.9,verbose=FALSE)), "required argument 'k0' is missing")
#right clusterFunctions
expect_error(clusterSingle(mat,
mainClusterArgs=list(clusterFunction="kmeans"),
subsample=TRUE, sequential=TRUE,
subsampleArgs=list(clusterFunction="pam",n.sample=40),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE)),
"If choosing subsample=TRUE, the clusterFunction used in the mainClustering step must take input that is a categorical")
expect_error(clusterSingle(dissMat, inputType="diss",
mainClusterArgs=list(clusterFunction="tight"),
subsample=FALSE, sequential=TRUE,
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE)),
"if subsample=FALSE, sequentical clustering can only be implemented with a clusterFunction for mainClustering step with algorithmType 'K'")
#warning if try to set k
expect_warning(clusterSingle(mat,
mainClusterArgs=list(clusterFunction="pam"),
subsample=TRUE, sequential=TRUE,
subsampleArgs=list(clusterFunction="pam",
n.sample=40,clusterArgs=list(k=3)),
isCount=FALSE,seqArgs=list(k0=5,beta=0.9,verbose=FALSE)),
"Setting 'k' in subsampleArgs when sequential=TRUE is called will have no effect.")
expect_warning(clusterSingle(mat,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
subsample=FALSE, sequential=TRUE,
subsampleArgs=list(clusterFunction="pam",n.sample=40),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE)),
"Setting 'k' in mainClusterArgs when sequential clustering is requested will have no effect.")
## Check warnings for esoteric arguments
expect_error(clusterSingle(mat,subsample=FALSE, sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE,top.can=5.5)),
"If top.can parameter of seqArgs is not a whole number, must be between 0 and 1.")
expect_error(clusterSingle(mat,subsample=FALSE, sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE,top.can=-1)),
"top.can parameter of seqArgs cannot be NA or negative value")
expect_error(clusterSingle(mat,subsample=FALSE, sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE,remain.n=3.4)),
"remain.n argument of seqArgs must be a positive whole number")
expect_error(clusterSingle(mat,subsample=FALSE, sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE,k.min=3.4)),
"k.min argument of seqArgs must be a positive whole number")
expect_error(clusterSingle(mat,subsample=FALSE, sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE,k.max=20.5)),
"k.max argument of seqArgs must be a positive whole number")
expect_silent(clusterSingle(mat,subsample=FALSE, sequential=TRUE,
mainClusterArgs=list(clusterFunction="pam"),
isCount=FALSE,
seqArgs=list(k0=5,beta=0.9,verbose=FALSE,top.can=0.1)))
})
test_that("Different direct options of `clusterSingle` ", {
#check isCount
expect_silent(clusterSingle(smSimCount,
subsample=FALSE, sequential=FALSE,
mainClusterArgs=list(clusterArgs=list(k=3), clusterFunction="pam"),
isCount=TRUE) )
#suppressWarnings because in addition to error, R prints warning about NAs
expect_error(suppressWarnings(clusterSingle(smSimData,
subsample=FALSE, sequential=FALSE,
mainClusterArgs=list(clusterArgs=list(k=3), clusterFunction="pam"),isCount=TRUE)),"User-supplied `transFun` produces NA values",info="test error handling for isCount=TRUE when can't take log")
})
test_that("`clusterSingle` preserves the colData and rowData of SE", {
expect_silent(cl<-clusterSingle(se,
subsample=FALSE, sequential=FALSE,
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_equal(colData(cl),colData(se))
expect_equal(rownames(cl),rownames(se))
expect_equal(colnames(cl),colnames(se))
expect_equal(metadata(cl),metadata(se))
expect_equal(rowData(cl),rowData(se))
})
test_that("`clusterSingle` works with parallel subsampling", {
## Windows does not support mclapply
skip_on_os("windows")
expect_silent(clustSubsample2 <-
clusterSingle(mat, subsample=TRUE,sequential=FALSE,
subsampleArgs=list(clusterFunction="kmeans",resamp.num=3,
clusterArgs=list(k=3),ncores=2),
mainClusterArgs=list(clusterFunction="pam",clusterArgs=list(k=3)),
isCount=FALSE))
expect_silent(clustSubsample1 <-
clusterSingle(mat, subsample=TRUE,sequential=FALSE,
subsampleArgs=list(clusterFunction="kmeans",resamp.num=3,
clusterArgs=list(k=3),ncores=1),
mainClusterArgs=list(clusterFunction="pam", clusterArgs=list(k=3)),
isCount=FALSE))
})
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.