R/convertxrf.R

In agryt/xrfr: Converting XRF Data from Kilo-counts per Second to Micromolar

#' Converting XRF data from kcal to micromolar
#'
#' @description This function takes your generated project data frame created with readxrf() and the file containing information from the XRF machine, and converts your raw data from kcal to micromolar.
#'
#' See vignette("xrfr") for more information.
#'
#' @return description The function creates a data frame in the long format with the new columns "Concentration" and "Adjusted_detection_limit" showing the calculated concentration and the respective detection limit (calculated based on volume filtered).
#'
#' @param imported_data The name of the data frame created with readxrf()
#' @param base_info The name of the data frame containing detection limits, crystal drift, molar weights, and calibration constants.
#' @param year The year the drift was measured closest to when your samples were analysed.
#' @param first_element The name of the first column containing kcps values in the generated project data frame.
#' @param last_element The name of the last column containing kcps values in the generated project data frame.
#'
#' @importFrom tidyr pivot_longer
#' @importFrom dplyr filter group_by summarise left_join mutate select distinct relocate
#' @importFrom stringr str_remove
#' @importFrom rlang .data
#' @importFrom magrittr %>%
#'
#' @examples
#' \dontrun{
#' rawdata.df <- read_delim("xrf_rawdata.txt", delim = "\t", locale = readr::locale(decimal_mark = ","))
#' projectinfo.df <- read_excel("xrf_projectinfo.xlsx")
#' baseinfo.df <- read_excel("xrf_setup.xlsx")
#'
#' projectfile.df <- readxrf(raw_data = rawdata.df, project_info = projectinfo.df)
#'
#' project.df <- convertxrf(imported_data = projectfile.df, base_info = baseinfo.df, year = "2019", first_element = "C", last_element = "As")
#' }
#'
#' @export


convertxrf <- function(imported_data, base_info, year, first_element, last_element) {

  filter_area <- 9.078935

  projectfile.df <- as.data.frame(imported_data)

  # making the data frame longer
  pivotproject.df <- projectfile.df %>%
    tidyr::pivot_longer(first_element : last_element,
                 names_to = "Element",
                 values_to = "Count")

  if(sum(grepl("blank", pivotproject.df$Filter_blank)) < 1) {
    stop("ERROR! There are no samples marked blank in your dataset. You must have a column named Filter_blank with 'blank' written in the cell for the blank samples.")
  }

  # making a data frame with the means of the blank samples
  mean.blanks.df <- pivotproject.df %>%
    dplyr::filter(.data$Filter_blank == "blank") %>%
    dplyr::group_by(.data$Filter_type, .data$Filter_size, .data$Filter_box_nr, .data$Element) %>%
    dplyr::summarise(mean_blank = mean(.data$Count))

  # joining the mean blanks data frame with the project file
  adjusted.for.blanks.df <- dplyr::left_join(pivotproject.df, mean.blanks.df, by = c("Filter_type", "Filter_size", "Filter_box_nr", "Element")) %>%
    dplyr::mutate(Net_count = .data$Count - .data$mean_blank)

  basefile.df <- as.data.frame(base_info)

  if(!"PC" %in% names(basefile.df)) {
    stop("ERROR! Your base information file is missing one or more of the following columns: PC, ANO, GFF, DL_PC, DL_ANO, and DL_GFF.")
  }
  if(!"ANO" %in% names(basefile.df)) {
    stop("ERROR! Your base information file is missing one or more of the following columns: PC, ANO, GFF, DL_PC, DL_ANO, and DL_GFF.")
  }
  if(!"GFF" %in% names(basefile.df)) {
    stop("ERROR! Your base information file is missing one or more of the following columns: PC, ANO, GFF, DL_PC, DL_ANO, and DL_GFF.")
  }
  if(!"DL_PC" %in% names(basefile.df)) {
    stop("ERROR! Your base information file is missing one or more of the following columns: PC, ANO, GFF, DL_PC, DL_ANO, and DL_GFF.")
  }
  if(!"DL_ANO" %in% names(basefile.df)) {
    stop("ERROR! Your base information file is missing one or more of the following columns: PC, ANO, GFF, DL_PC, DL_ANO, and DL_GFF.")
  }
  if(!"DL_GFF" %in% names(basefile.df)) {
    stop("ERROR! Your base information file is missing one or more of the following columns: PC, ANO, GFF, DL_PC, DL_ANO, and DL_GFF.")
  }
  if(!"Drift_2008" %in% names(basefile.df)) {
    stop("ERROR! Your base information file is missing the column Drift_2008, which is needed to perform the calculations.")
  }
  if(!"MolarW" %in% names(basefile.df)) {
    stop("ERROR! Your base information file is missing the column MolarW, which is needed to perform the calculations.")
  }

  # joining base info file with the project file
  basefile.df <- basefile.df %>%
    tidyr::pivot_longer(cols = c(.data$PC, .data$ANO, .data$GFF),
                        names_to = "Filter_type",
                        values_to = "Cal_const") %>%
    dplyr::relocate(Filter_type)

  joined.df <- dplyr::left_join(adjusted.for.blanks.df, basefile.df, by = c("Filter_type", "Element"))

  # performing the calculation to convert from kcps to µM
  calculations.df <- joined.df %>%
    dplyr::mutate(Concentration = ((.data$Net_count * .data$Cal_const) * filter_area * (1000 / .data$Volume) * 1000 * (.data$Drift_2008 / .data[[paste0("Drift_", year)]])) / .data$MolarW)

  # making a data frame showing the detection limits of each element
  detectionlimits.df <- basefile.df %>%
    dplyr::select(.data$DL_PC, .data$DL_ANO, .data$DL_GFF, .data$Element) %>%
    tidyr::pivot_longer(cols = c(.data$DL_PC, .data$DL_ANO, .data$DL_GFF),
                        names_to = "Filter_type",
                        values_to = "Detection_limit") %>%
    dplyr::mutate(Filter_type = stringr::str_remove(.data$Filter_type, "DL_"))

  # joining detection limit data frame with the project file
  project.detectionlim.df <- dplyr::left_join(calculations.df, detectionlimits.df, by = c("Filter_type", "Element"))

  # adjusting the detection limit based on filtered volume
  project.adj.detectionlim.df <- project.detectionlim.df %>%
    dplyr::mutate(Adjusted_detection_limit = (1000 / .data$Volume) * .data$Detection_limit) %>%
    dplyr::select(-.data$Detection_limit)

  # removing the columns we don't need anymore
  project.df <- project.adj.detectionlim.df %>%
    dplyr::distinct() %>%
    dplyr::select(.data$Sample : .data$Element, .data$Concentration, .data$Adjusted_detection_limit)

  return(project.df)

}