R/snpMatrix.R

In thesushantpatil/maftools: Summarize, Analyze and Visualize MAF Files

#' Creates a Genotype Matrix for every variant
#' @description Creates a Genotype matrix using allele frequcies or by muatation status.
#'
#' @param maf an \code{\link{MAF}} object generated by \code{\link{read.maf}}
#' @param genes create matrix for only these genes. Define NULL
#' @param tsb create matrix for only these tumor sample barcodes/samples. Define NULL
#' @param includeSyn whether to include silent mutations. Default FALSE
#' @param vafCol specify column name for vaf's. Default NULL. If not provided simply assumes all mutations are heterozygous.
#' @param vafCutoff specify minimum and maximum vaf to define mutations as heterozygous. Default range 0.1 to 0.75. Mutations above maximum vafs are defined as homozygous.
#' @return matrix
#' @examples
#' laml.maf <- system.file("extdata", "tcga_laml.maf.gz", package = "maftools")
#' laml <- read.maf(maf = laml.maf)
#' genotypeMatrix(maf = laml, genes = "RUNX1")
#'
#' @export

genotypeMatrix = function(maf, genes = NULL, tsb = NULL, includeSyn = FALSE, vafCol = NULL, vafCutoff = c(0.1, 0.75)){

  mdat = subsetMaf(maf = maf, tsb = tsb, genes = genes, fields = vafCol, includeSyn = includeSyn, mafObj = FALSE)

  if(nrow(mdat) == 0){
    stop("Zero mutations to make table.")
  }

  mdat[,id := paste0(Chromosome, ":", Start_Position)]

  if(!is.null(vafCol)){
    if(vafCol %in% colnames(mdat)){
      colnames(mdat)[which(x = colnames(mdat) == vafCol)] = 't_vaf'
    }else{
      stop(paste0("Column ", vafCol, " not found!"))
    }

    if(max(mdat[,t_vaf], na.rm = TRUE) > 1){
      mdat[,t_vaf := as.numeric(as.character(t_vaf))/100]
    }

    vafMin = vafCutoff[1]
    vafMax = vafCutoff[2]

    mdat.cast = data.table::dcast(data = mdat, formula = id ~ Tumor_Sample_Barcode, value.var = 't_vaf', fill = 0)
    data.table::setDF(x = mdat.cast, rownames = mdat.cast$id)
    mdat.cast = mdat.cast[,-1]

    tnumMat = t(mdat.cast) #transposematrix
    mdat.cast = t(tnumMat[do.call(order, c(as.list(as.data.frame(tnumMat)), decreasing = TRUE)), ]) #sort

    mdat.cast = apply(X = mdat.cast, MARGIN = 2, FUN = function(x){
      ifelse(test = x == 0, yes = "None",
             no = ifelse(test = x > vafMin & x < vafMax, yes = "Het",
                         no = ifelse(test = x > vafMax, yes = "Hom", no = "None")))
    })
  }else{
    mdat.cast = data.table::dcast(data = mdat, formula = id ~ Tumor_Sample_Barcode, value.var = 'id', fill = 0)
    data.table::setDF(x = mdat.cast, rownames = mdat.cast$id)
    mdat.cast = mdat.cast[,-1]
    mdat.cast[mdat.cast != 0] = 1

    tnumMat = t(mdat.cast) #transposematrix
    mdat.cast = t(tnumMat[do.call(order, c(as.list(as.data.frame(tnumMat)), decreasing = TRUE)), ]) #sort

    mdat.cast = apply(X = mdat.cast, MARGIN = 2, FUN = function(x){
      ifelse(test = x == "0", yes = "None",
             no = "Het")
    })
  }

  mdat.cast
}