Function to read the binary output of salmon alevin in R (single-cell RNA-seq)

readSalmonAlevin.R

readSalmonAlevin <- function(alevinFolder){
  matrixFile <- paste0(alevinFolder, "/quants_mat.gz")
  matrixFile <- gzfile(description = matrixFile, open = 'rb')
  genes <- readLines(paste0(alevinFolder,"/quants_mat_cols.txt"))
  cells <- readLines(paste0(alevinFolder,"/quants_mat_rows.txt"))
  n <- length(genes)*length(cells)
  counts <- readBin(matrixFile,what = 'numeric', n = n)
  close.connection(matrixFile)
  countMatrix <- matrix(data = counts, 
                        nrow = length(cells), 
                        ncol = length(genes), 
                        byrow = TRUE)
  colnames(countMatrix) <- genes
  rownames(countMatrix) <- cells
    require(Matrix)
  countMatrix <- as(countMatrix, "sparseMatrix")
  countMatrix <- t(countMatrix)
  return(countMatrix)
}