R/create_nmrdata.R

In brgordon17/coralclass: Metabolomics Classification of Stressed Coral From Heron Island 2011

#' create nmrdata.
#'
#' \code{create_nmrdata()} preprocesses the 1H-NMR data used for modelling
#' in this PhD chapter.
#'
#' \code{create_nmrdata()} takes the binned data from Bruker's Amix and creates
#' new categorical variables based on the sample ID's. It then removes any
#' unwanted variables as defined by the user (e.g. residual methanol or water
#' peaks) before saving the results to \code{./data}
#'
#' @param savecsv Logical indicating if output should be saved as a \code{.csv}
#' file to the current working directory
#' @param saverda Logical indicating if a .rda file should be saved to /data
#' @param csvname The name of the output .csv file to be saved if TRUE
#' @param remove.vars Logical indicating if user-defined variables should be
#' removed
#' @param vars A character vector of variables to be removed
#' @param ... Other arguments passed on to individual methods
#'
#' @return Returns a dataframe of class tbl_df
#'
#' @note \code{create_nmrdata()} was not intended to be used outside of this
#' package.
#'
#' @seealso
#' \href{https://www.bruker.com/products/mr/nmr/nmr-software/nmr-software/amix/overview.html}{Bruker AMIX}
#'
#' @author Benjamin R. Gordon
#'
#' @export
#'
create_nmrdata <- function(savecsv = FALSE,
                           saverda = TRUE,
                           csvname = "nmrdata",
                           remove.vars = TRUE,
                           vars = c("3.31"),
                           ...) {

  nmrdata  <-  readr::read_csv("./data-raw/nmrdata-raw-0.02.csv", na = "0")
  colnames(nmrdata)[1] <- "sample_ids"
  sample_ids <- nmrdata[1]
  var_ids <- round(as.numeric(colnames(nmrdata[-1])),2)
  colnames(nmrdata)[2:ncol(nmrdata)] <- var_ids

  # create categorical variables -----------------------------------------------
  class <- c(rep("eCO2", 6), # 1000-0902-C
             rep("eCO2eT", 6), # 1000-0902-H
             rep("eCO2", 6), # 1000-1202-C
             rep("eCO2eT", 6), # 1000-1202-H
             rep("eCO2", 6), # 1000-1402-C
             rep("eCO2eT", 6), # 1000-1402-H
             rep("eCO2", 6), #1000-d14-C
             rep("eCO2eT", 6), # 1000-d14-H
             rep("control", 6), # 380-0902-C
             rep("eT", 6), # 380-0902-H
             rep("control", 6), # 380-1202-C
             rep("eT", 3), # 380-1202-H
             rep("control", 6), # 380-1402-C
             rep("eT", 6),# 380-1402-H
             rep("control", 6), # 380-d14-C
             rep("eT", 6)) # 380-d14-H
  class <- factor(class, levels = c("control", "eT", "eCO2", "eCO2eT"))

  day <- c(rep("day1", 12),
           rep("day4", 12),
           rep("day6", 12),
           rep("day14", 12),
           rep("day1", 12),
           rep("day4", 9),
           rep("day6", 12),
           rep("day14", 12)
           )
  day <- factor(day, levels = c("day1", "day4", "day6", "day14"))

  tank <- c(rep(c("L","L","L", "R","R","R"), 11),
            c("L","L","L"),
            rep(c("L","L","L", "R","R","R"), 4)
            )
  tank <- factor(tank, levels = c("L", "R"))

  rep <- c(rep(c(1, 2, 3), 31))
  rep <- factor(rep, levels = (c(1:3)))

  class_day <- interaction(class,
                           day,
                           drop = TRUE,
                           sep = "."
                           )

  nmrdata <- tibble::as_tibble(data.frame(sample_ids,
                                          class,
                                          day,
                                          tank,
                                          rep,
                                          class_day,
                                          nmrdata[2:ncol(nmrdata)],
                                          check.names = FALSE)
                               )
  nmrdata <- dplyr::arrange(nmrdata, class, day)

  if(remove.vars) {
    nmrdata <- nmrdata %>%
      dplyr::select(-dplyr::one_of(vars))
  }

  # write data -----------------------------------------------------------------
  if(saverda) {
    save(nmrdata, file = "./data/nmrdata.rda", compress = "bzip2")
  }

  if(savecsv) {
    readr::write_csv(nmrdata, paste(c("./inst/extdata/", csvname, ".csv"),
                                    collapse = "")
                     )
  }

  nmrdata
}

## Document data ---------------------------------------------------------------

#' Preprocessed NMR data
#'
#' A dataset of 1H-NMR variables (0.02 ppm wide bins) for coral samples
#' subjected to experimental ocean warming and acidification at Heron Island in
#' Feb 2011.
#'
#' @format A tibble with 93 rows and 505 variables:
#' \describe{
#'   \item{sample_ids}{ID's for each sample}
#'   \item{class}{Treatment group. One of, control, eT (eleveated temperature),
#'   eCO2 (elevated CO2), eCO2eT (elevated CO2 and temperature), PBQC}
#'   \item{day}{Days of exposure to treatment}
#'   \item{tank}{Tank membership. Either L (left) or R (right)}
#'   \item{rep}{Sample replicate number}
#'   \item{class_day}{interaction variable between class and day variables}
#' }
#' @source Benjamin R. Gordon
"nmrdata"