Removed MPEAR, code cleanup and added documentation

Author: sdentro

Diff for: DESCRIPTION

@@ -14,6 +14,6 @@ Depends:
   ks,
   lattice,
   ggplot2,
-  gridExtra,
-  mcclust
+  gridExtra
 LazyLoad: yes
+RoxygenNote: 6.1.1

Diff for: Dockerfile

@@ -4,7 +4,7 @@ USER root
 
 RUN apt-get update && apt-get -y install r-base
 
-RUN R -q -e 'source("http://bioconductor.org/biocLite.R"); biocLite(c("optparse"","KernSmooth","ks","lattice","ggplot2","gridExtra","mcclust"))'
+RUN R -q -e 'source("http://bioconductor.org/biocLite.R"); biocLite(c("optparse"","KernSmooth","ks","lattice","ggplot2","gridExtra"))'
 
 RUN mkdir -p /opt/dpclust
 COPY . /opt/dpclust/

Diff for: NAMESPACE

@@ -1,8 +1,48 @@
-exportPattern("^[[:alpha:]]+")
-import(lattice)
-import(KernSmooth)
-import(ks)
-import(mcclust)
-import(gridExtra)
-import(ggplot2)
+# Generated by roxygen2: do not edit by hand
 
+export(RunDP)
+export(make_advanced_params)
+export(make_run_params)
+export(make_sample_params)
+import(ggplot2)
+import(gridExtra)
+importFrom(KernSmooth,bkde)
+importFrom(KernSmooth,bkde2D)
+importFrom(grDevices,cm.colors)
+importFrom(grDevices,colorRampPalette)
+importFrom(grDevices,dev.cur)
+importFrom(grDevices,dev.off)
+importFrom(grDevices,dev.set)
+importFrom(grDevices,pdf)
+importFrom(grDevices,png)
+importFrom(grDevices,rainbow)
+importFrom(grDevices,rgb)
+importFrom(graphics,hist)
+importFrom(graphics,legend)
+importFrom(graphics,lines)
+importFrom(graphics,par)
+importFrom(graphics,plot)
+importFrom(graphics,points)
+importFrom(graphics,polygon)
+importFrom(graphics,text)
+importFrom(graphics,title)
+importFrom(ks,kde)
+importFrom(lattice,levelplot)
+importFrom(lattice,lpoints)
+importFrom(lattice,panel.abline)
+importFrom(lattice,panel.levelplot)
+importFrom(stats,density)
+importFrom(stats,median)
+importFrom(stats,pchisq)
+importFrom(stats,quantile)
+importFrom(stats,rbeta)
+importFrom(stats,rbinom)
+importFrom(stats,rgamma)
+importFrom(stats,rmultinom)
+importFrom(stats,runif)
+importFrom(stats,sd)
+importFrom(utils,head)
+importFrom(utils,read.csv)
+importFrom(utils,read.table)
+importFrom(utils,write.csv)
+importFrom(utils,write.table)

Diff for: R/DensityEstimator.R

@@ -183,7 +183,7 @@ Gibbs.subclone.density.est <- function(burden, GS.data, pngFile, density.smooth
       write.csv(density.data,gsub(".png",paste("_densityData",i,".csv",sep=""),pngFile))
     }
     if(num.timepoints==2){
-      d=bkde2D(density.data,bandwidth=density.smooth,gridsize=gridsize,range.x=range)
+      d=KernSmooth::bkde2D(density.data,bandwidth=density.smooth,gridsize=gridsize,range.x=range)
       #if(i==post.burn.in.start){
       if(i==1){	
         xvals=d$x1
@@ -193,7 +193,7 @@ Gibbs.subclone.density.est <- function(burden, GS.data, pngFile, density.smooth
       post.ints[,,i]=d$fhat	
 
     }else{
-      d=bkde(density.data,bandwidth=density.smooth,gridsize=512L,range.x=range)
+      d=KernSmooth::bkde(density.data,bandwidth=density.smooth,gridsize=512L,range.x=range)
       #post.ints[,i - post.burn.in.start + 1]=d$y
       post.ints[,i]=d$y
     }
@@ -336,10 +336,10 @@ Gibbs.subclone.density.est.nd <- function(burden, GS.data, density.smooth = 0.1,
     if (i %% 100 == 0) { print(paste(i, "/", length(sampledIters), sep=" ")) }
     if(num.timepoints>=4){
       #use weights
-      d=kde(pi.h.cols[sampledIters[i]-1,,],H=diag(num.timepoints)*density.smooth,eval.points = evaluation.points,w=C*wts[sampledIters[i],])                   
+      d=ks::kde(pi.h.cols[sampledIters[i]-1,,],H=diag(num.timepoints)*density.smooth,eval.points = evaluation.points,w=C*wts[sampledIters[i],])                   
     }else{
       #use weights
-      d=kde(pi.h.cols[sampledIters[i]-1,,],H=diag(num.timepoints)*density.smooth,gridsize=gridsize,xmin=range[,1],xmax=range[,2],w=C*wts[sampledIters[i],])
+      d=ks::kde(pi.h.cols[sampledIters[i]-1,,],H=diag(num.timepoints)*density.smooth,gridsize=gridsize,xmin=range[,1],xmax=range[,2],w=C*wts[sampledIters[i],])
     }
     if(i==1){       
       eval.points = d$eval.points

Diff for: R/DirichletProcessClustering.R

@@ -52,7 +52,7 @@ load.data <- function(list_of_data_files, cellularity, Chromosome, position, WT.
 
 #' This inner function takes a list of loaded data tables and transforms them into
 #' a dataset, which is a list that contains a table per data type
-#' @noRD
+#' @noRd
 load.data.inner = function(list_of_tables, cellularity, Chromosome, position, WT.count, mut.count, subclonal.CN, no.chrs.bearing.mut, mutation.copy.number, subclonal.fraction, phase, is.male, min.depth, min.mutreads, supported_chroms, mutation_type="SNV") {
   no.subsamples = length(list_of_tables)
   no.muts = nrow(list_of_tables[[1]])
@@ -173,7 +173,7 @@ load.indel.data = function(infiles) {
 }
 
 #' Function that adds copy number as a series of SNVs into a data set
-add.in.cn.as.snv.cluster = function(dataset, cndata, add.conflicts=T, conflicting.events.only=F, num.clonal.events.to.add=0, min.cna.size=10) {
+add.in.cn.as.snv.cluster = function(dataset, cndata, cellularity, add.conflicts=T, conflicting.events.only=F, num.clonal.events.to.add=0, min.cna.size=10) {
   # If clonal events are to be added, make a selection. For now this just takes the largest event
   if (num.clonal.events.to.add > 0) {
     # Save the largest clonal event to add to the CNAs
@@ -247,7 +247,7 @@ add.in.cn.as.snv.cluster = function(dataset, cndata, add.conflicts=T, conflictin
 }
 
 #' Function that adds copy number as a single SNV into a data set
-add.in.cn.as.single.snv = function(dataset, cndata, add.conflicts=T) {
+add.in.cn.as.single.snv = function(dataset, cndata, cellularity, add.conflicts=T) {
 
   # The subclonal events can be used for clustering
   allowed.cn = c("sHD", "sLOH", "sAmp", "sGain", "sLoss")
@@ -470,14 +470,19 @@ get.conflicting.indices = function(dataset, cndata) {
 
 #' Replace the pseudo SNV clusters that represent a CNA event during clustering
 #' by amalgamated assignments for each CNA
+#' @param dataset A data set object
+#' @return a data set in which the pseudo SNVs are removed
 remove_pseudo_snv_cna_clusters = function(dataset) {
   # Remove the pseudo CNAs
   pseudo_snv_index = which(dataset$mutationType=="CNA")
   dataset = remove_mutations(dataset, pseudo_snv_index)
-  
+  return(dataset)
 }
 
 #' Helper function to remove a set of mutations from a dataset by their index
+#' @param dataset A data set object
+#' @param mutation_index Index of mutations to be removed
+#' @return a data set in which the pseudo SNVs are removed
 remove_mutations = function(dataset, mutation_index) {
   dataset$chromosome = dataset$chromosome[-mutation_index,,drop=F]
   dataset$position = dataset$position[-mutation_index,,drop=F]
@@ -498,6 +503,10 @@ remove_mutations = function(dataset, mutation_index) {
 }
 
 #' Add mutation phasing information to a dataset
+#' @param dataset A data set object
+#' @param mutphasing data.frame with phasing data
+#' @param add.conflicts Supply TRUE when a mutation-to-mutation confict matrix is to be built (Default: FALSE)
+#' @return a dataset with field mutphasing and potentially field conflict.array added
 add.mutphasing = function(dataset, mutphasing, add.conflicts=F) {
   dataset$mutphasing = mutphasing
 
@@ -535,6 +544,12 @@ add.mutphasing = function(dataset, mutphasing, add.conflicts=F) {
 }
 
 #' Add indels to an existing dataset
+#' @param dataset A data set object
+#' @param indeldata A list of read in data.frames (one per sample) with output from dpclust3p
+#' @param min.depth Minimum depth required to keep an indel
+#' @param min.mutreads Minimum number of mutant reads to keep an indel
+#' @param supported_chroms Chromosomes that are supported
+#' @return the provided dataset with indels added
 add.in.indels = function(dataset, indeldata, is.male, min.depth, min.mutreads, supported_chroms) {
   indelset = load.data.inner(list_of_tables=indeldata, 
                              cellularity=dataset$cellularity, 
@@ -560,6 +575,9 @@ add.in.indels = function(dataset, indeldata, is.male, min.depth, min.mutreads, s
 
 
 #' Convenience function to append two datasets
+#' @param a dataset a
+#' @param b dataset b
+#' @return a dataset in which a and b are appended
 append.dataset = function(a, b) {
   if (dim(a$mutCount) != dim(b$mutCount) & a$cellularity!=b$cellularity) {
     stop("Cannot append two datasets of different sizes or with different cellularities")