R/random_chop.R

In adsteen/genomechop:

##' Chops sequences randomly
##'
##' @param seq The raw sequence to be chopped
##' @param len The length of the chopped pieces (for now, this a constant)
##' @param n The number of pieces to return
##' @export
##'
random_chop <- function(seq_data, len, n) {

  browser()

  # Calculate number of sequences
  if(n == "complete") {
    n <- ceiling(length(seq_data) / len)
  }

  # Check if seq has length n. This turns out to be a useful test if you are a crummy programmer like some people whose names I won't mention, but which rhyme with Bambrew Fleen.
  if(length(seq_data) <= 1) {
    warning("Sequence has length less than 1 which is messed up.")
    return(NA)
  }

  ### Make a vector of start values for the cuts
  start.pos <- sample(length(seq_data), n, replace=TRUE)

  # Preallocate a list to contain the "chopped" sequences
  # NB I have no idea whether this actually helps
  empt.vec <- rep(NA, len)
  chopped.seq.list <- rep(list(empt.vec), n)

  # Should use vapply for safety
  # The problem is that R is trying to apply circular_cut to each element of seq (maybe?)
  chopped.seq.list <- lapply(start.pos, FUN = circular_cut, seq_data = seq_data, len = len)
  #chopped.seq.df <- purrr::map_dfr(start.pos, .f = circular_cut, seq = seq, len = len)

  chopped.seq.list
}