R/amp_export_fasta.R

In MadsAlbertsen/ampvis2: Tools for visualising amplicon data

#' Export raw DNA sequences
#'
#' Export sequences from an ampvis2 object with added taxonomy.
#'
#' @param data (\emph{required}) Data list as loaded with \code{\link{amp_load}}.
#' @param filename File name of the exported FASTA file. (\emph{default:} \code{"exported_sequences.fa"})
#' @param tax (\emph{logical}) Add taxonomic strings to the output or not. (\emph{default:} \code{TRUE})
#' @param ... Additional arguments directly passed to \code{\link[ape]{write.dna}}.
#'
#' @importFrom ape write.dna
#' @importFrom stringr str_detect str_split
#' @export
#'
#' @seealso
#' \code{\link{amp_load}}, \code{\link{amp_export_otutable}}
#'
#' @examples
#' # Load example data
#' data("AalborgWWTPs")
#'
#' # Export fasta sequences as "exported_sequences.fa"
#' \dontrun{
#' amp_export_fasta(AalborgWWTPs)
#' }
#'
#' @author Kasper Skytte Andersen \email{ksa@@bio.aau.dk}
#' @author Mads Albertsen \email{MadsAlbertsen85@@gmail.com}

amp_export_fasta <- function(data,
                             filename = "exported_sequences.fa",
                             tax = TRUE,
                             ...) {

  ### Data must be in ampvis2 format
  is_ampvis2(data)

  ### Reference sequences must be there!
  if (is.null(data$refseq)) {
    stop("No \"refseq\" element in the provided data.", call. = FALSE)
  }

  ### Check if refseq data is in the right format
  if (!is.null(data$refseq) & isFALSE(inherits(data$refseq, "DNAbin"))) {
    stop("The refseq element is not of class \"DNAbin\". The reference sequences must be loaded with ape::read.dna().", call. = FALSE)
  }

  t <- data[["refseq"]]

  if (tax == TRUE) {
    # dont want duplicate taxonomy, if the refseq loaded already contains taxonomy.
    if (any(stringr::str_detect(names(t), ";"))) {
      names(t) <- stringr::str_split(names(t), ";", simplify = TRUE)[, 1]
    }
    tax <- as.data.frame(data[["tax"]][, 1:7])
    tax <- tax[names(t), ]
    tax_s <- data.frame(lapply(tax, as.character), stringsAsFactors = FALSE)
    df_args <- c(tax_s, sep = "; ")
    tax_sf <- do.call(paste, df_args)

    names(t) <- paste(names(t), tax_sf, sep = "; ")
  }

  ape::write.dna(t, file = filename, format = "fasta", ...)
}