helper_io.R

#------------------------------------------------------------------------------#
#                                 I/O helper                                   #
#                                 functions                                    #
#------------------------------------------------------------------------------#

#' Return elapsed time
#'
#' Given a start and end time in seconds, returns elapsed time in sec., min. or
#' hrs, as appropriate.
#' @param t0 Numeric start time representing seconds. No default.
#' @param t1 Numeric end time representing continuous second count from
#' \code{t0}. No default.
#' @return Elapsed time as \code{string}.
#' @examples
#' \dontrun{
#' start <- proc.time()[3]
#' end <- proc.time()[3]
#' elapsed_time(t0=start, t1=end)
#'
#' start <- proc.time()
#' elapsed_time(start[3], proc.time()[3])
#' }
#' @export

elapsed_time <- function(t0, t1) {
  # Calculate elapsed time
  e_time <- t1 - t0
  
  # Convert to appropriate unit
  if (e_time < 60) {
    e_time <- paste0(round(e_time, 1), " sec.")
  } else if (e_time >= 60 & e_time < 3600) {
    e_time <- paste0(round(e_time / 60, 1), " min.")
  } else {
    e_time <- paste0(round(e_time / 60 / 60, 2), " hrs.")
  }
  return(e_time)
}


#' Import bedGraph file
#'
#' Imports data from bedGraph file.
#' \cr\strong{Note 1:} In the bedGraph format chromosome coordinates are
#' zero-based and half-open (the first chromosome position is 0, and the last
#' position in a chromosome of length N is N - 1). Imported data here are always
#' converted to one-based format, irrespective of the format of the imported
#' file.
#' \cr\strong{Note 2:} This function imports the \code{bedGraph} table using the
#' \code{readr} package and then converts it to a \code{GRanges} object. This is
#' faster than using \code{rtracklayer}'s \code{import.bedGraph()} function.
#' @param path Path to a bedGraph file. No default.
#' @return Data as \code{GRanges} object.
#' @examples
#' \dontrun{
#' x <- import_bedGraph('data.bdg')
#' }
#' @export

import_bedGraph <- function(path) {
  t0  <- proc.time()[3]
  
  # Import bedGraph data
  message('   Importing bedGraph file...')
  df <- readr::read_tsv(path, col_names=FALSE)
  names(df) <- c('chr','start','end', 'score')
  message('   Converting to "GRanges" object...')
  gr <- with(df, GenomicRanges::GRanges(chr, IRanges::IRanges(start + 1, end),
                                        score=score))
  
  # Drop zeros (missing data)
  message('   Dropping zero scores...')
  gr <- gr[gr$score > 0]
  
  message('   ---')
  message('   Completed in ', elapsed_time(t0, proc.time()[3]))
  
  return(gr)
}


#' Import MACS2 peaks file
#'
#' Imports data from a peaks file in \coe{BED6+4} format, generated by MACS2.
#' @param path Path to a MACS2 peaks file. No default.
#' @param type Type of peaks file: one of \code{narrow} (default value) and
#' \code{broad}.
#' @return Data as \code{GRanges} object.
#' @examples
#' \dontrun{
#' x <- import_bedGraph('data.bdg')
#' }
#' @export

import_MACS2_peaks <- function(path, type='narrow') {
  # Import peaks
  message('   Importing peaks file...')
  if (type == 'narrow') {
    extra_cols <- c(signalValue = "numeric", pValue = "numeric",
                    qValue = "numeric", summit = "integer")
  } else if (type == 'broad') {
    extra_cols <- c(signalValue = "numeric", pValue = "numeric",
                    qValue = "numeric")
  } else stop('"type" argument must be one of "narrow" or "broad".')
  
  peaks <- rtracklayer::import.bed(path, extraCols=extra_cols)
  
  message('   Completed.')
  return(peaks)
}