#' @title Permutations and Clustering
#' @description This function executes a ubuntu docker that produces a specific number of permutation to evaluate clustering.
#' @param group, a character string. Two options: sudo or docker, depending to which group the user belongs
#' @param scratch.folder, a character string indicating the path of the scratch folder
#' @param file, a character string indicating the path of the file, with file name and extension included
#' @param nPerm, number of permutations to perform the pValue to evaluate clustering
#' @param permAtTime, number of permutations that can be computes in parallel
#' @param percent, percentage of randomly selected cells removed in each permutation
#' @param range1, first number of cluster for k means algorithm
#' @param range2, last number of cluster for k means algorithm
#' @param separator, separator used in count file, e.g. '\\t', ','
#' @param logTen, 1 if the count matrix is already in log10, 0 otherwise
#' @param clustering, clustering method to use : "SIMLR" , "tSne", "griph"
#' @param perplexity, Number of close neighbors for each point. This parameter is specific for tSne. Default value is 10.Setting this parameter when use a clustering method different by tSne will be ignored.
#' @param seed, important value to reproduce the same results with same input
#' @param rK, 1 for rankGene algorithm 0 otherwise WARNING, very slow with this feature. This parameter is specific for SIMLR. Setting this parameter to 1 with other clustering methods will not give any different result to set the parameter to 0.
#' @author Luca Alessandri, alessandri [dot] luca1991 [at] gmail [dot] com, University of Torino
#'
#' @return VioPlot of silhouette cells value for each number of cluster used,clusterP file with clustering results for each permutation, killedCell file with removed cells in each permutation, clustering.output a sommarize file with general information for each cells.
#' @examples
#' \dontrun{
#' permutationClustering("docker","/home/lucastormreig/CASC2.0/permutationClustering/scratch/","/home/lucastormreig/CASC2.0/permutationClustering/Data/TOTAL.csv",4,2,10,3,4,separator=",",logTen=0,clustering="SIMLR",perplexity=0)
#'}
#' @export
permutationClustering <- function(group=c("sudo","docker"), scratch.folder, file, nPerm, permAtTime, percent, range1=3, range2=3, separator, logTen=0, clustering, perplexity=10 , seed=1111, rK=0){
data.folder=dirname(file)
positions=length(strsplit(basename(file),"\\.")[[1]])
matrixNameC=strsplit(basename(file),"\\.")[[1]]
matrixName=paste(matrixNameC[seq(1,positions-1)],collapse="")
format=strsplit(basename(basename(file)),"\\.")[[1]][positions]
#running time 1
ptm <- proc.time()
#setting the data.folder as working folder
if (!file.exists(data.folder)){
cat(paste("\nIt seems that the ",data.folder, " folder does not exist\n"))
return(2)
}
#storing the position of the home folder
home <- getwd()
setwd(data.folder)
#initialize status
system("echo 0 > ExitStatusFile 2>&1")
#testing if docker is running
test <- dockerTest()
if(!test){
cat("\nERROR: Docker seems not to be installed in your system\n")
system("echo 10 > ExitStatusFile 2>&1")
setwd(home)
return(10)
}
#check if scratch folder exist
if (!file.exists(scratch.folder)){
cat(paste("\nIt seems that the ",scratch.folder, " folder does not exist\n"))
system("echo 3 > ExitStatusFile 2>&1")
setwd(data.folder)
return(3)
}
tmp.folder <- gsub(":","-",gsub(" ","-",date()))
scrat_tmp.folder=file.path(scratch.folder, tmp.folder)
writeLines(scrat_tmp.folder,paste(data.folder,"/tempFolderID", sep=""))
cat("\ncreating a folder in scratch folder\n")
dir.create(file.path(scrat_tmp.folder))
#preprocess matrix and copying files
if(separator=="\t"){
separator="tab"
}
dir.create(paste(scrat_tmp.folder,"/",matrixName,sep=""))
dir.create(paste(data.folder,"/Results",sep=""))
system(paste("cp ",data.folder,"/",matrixName,".",format," ",scrat_tmp.folder,"/",matrixName,sep=""))
#executing the docker job
params <- paste("--cidfile ",data.folder,"/dockerID -v ",scrat_tmp.folder,":/scratch -v ", data.folder, ":/data -d docker.io/repbioinfo/permutationclustering Rscript /home/main.R ",matrixName," ",nPerm," ",permAtTime," ",percent," ",range1," ",range2," ",format," ",separator," ",logTen," ",clustering," ",rK," ",perplexity," ", seed, sep="")
resultRun <- runDocker(group=group, params=params)
#waiting for the end of the container work
if(resultRun==0){
#system(paste("cp ", scrat_tmp.folder, "/* ", data.folder, sep=""))
}
#running time 2
ptm <- proc.time() - ptm
dir <- dir(data.folder)
dir <- dir[grep("run.info",dir)]
if(length(dir)>0){
con <- file("run.info", "r")
tmp.run <- readLines(con)
close(con)
tmp.run[length(tmp.run)+1] <- paste("user run time mins ",ptm[1]/60, sep="")
tmp.run[length(tmp.run)+1] <- paste("system run time mins ",ptm[2]/60, sep="")
tmp.run[length(tmp.run)+1] <- paste("elapsed run time mins ",ptm[3]/60, sep="")
writeLines(tmp.run,"run.info")
}else{
tmp.run <- NULL
tmp.run[1] <- paste("run time mins ",ptm[1]/60, sep="")
tmp.run[length(tmp.run)+1] <- paste("system run time mins ",ptm[2]/60, sep="")
tmp.run[length(tmp.run)+1] <- paste("elapsed run time mins ",ptm[3]/60, sep="")
writeLines(tmp.run,"run.info")
}
#saving log and removing docker container
container.id <- readLines(paste(data.folder,"/dockerID", sep=""), warn = FALSE)
system(paste("docker logs ", substr(container.id,1,12), " &> ",data.folder,"/", substr(container.id,1,12),".log", sep=""))
system(paste("docker rm ", container.id, sep=""))
#Copy result folder
cat("Copying Result Folder")
system(paste("cp -r ",scrat_tmp.folder,"/* ",data.folder,"/Results",sep=""))
#removing temporary folder
cat("\n\nRemoving the temporary file ....\n")
system(paste("rm -R ",scrat_tmp.folder))
system("rm -fR out.info")
system("rm -fR dockerID")
system("rm -fR tempFolderID")
system(paste("cp ",paste(path.package(package="rCASC"),"containers/containers.txt",sep="/")," ",data.folder, sep=""))
setwd(home)
}
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.