insertExtraData.R

library(tidyverse)

#' Add arbitrary data to record from an infolist column
#' 
#' Normal behaviour of RMassBank is that only specific columns from infolist.csv
#' are added to records. Any extra data is discarded.
#' 
#' This function provides a possibility to add or overwrite data in a record
#' in a per-compound manner (as opposed to a global value specified in the
#' RMassBank settings). For example, the user may overwrite the retention time
#' or add a CCS value.
#' 
#' Note that tags are not reordered and new tags are added after all existing tags.
#' Therefore, only use this for pre-existing tags or optional tags (that don't have
#' a prespecified order).
#' 
#'
#' @param mb `mbWorkspace` after step 4 of mbWorkflow (i.e. after record compilation)
#' @param infolist a CSV infolist with extra columns. These need to be named
#'  according to the target tag. I.e. for a tag `AC$CHROMATOGRAPHY: RETENTION_TIME`,
#'  the column must be called `AC$CHROMATOGRAPHY.RETENTION_TIME` (it can be obtained
#'  with `make.names("AC$CHROMATOGRAPHY: RETENTION_TIME")`.
#' @param column The column to be inserted. Note that an infolist may have multiple
#'  extra columns. Each one needs to be added separately. 
#'
#' @return
#' The modified `mbWorkspace`
#' @export
#'
#' @examples
#' \dontrun{
#' 
#' mb <- newMbWorkspace(w)
#' mb <- resetInfolists(mb)
#' mb <- loadInfolists(mb, "../infolists") #FOLDER
#' mb <- mbWorkflow(mb, steps = c(1:4))
#' # Currently only inserting from single infolists is supported (not from an entire
#' # folder of infolists)
#' mb <- insertExtraData(mb, "Infolist_PNEG_20eV_modified.csv", "AC$CHROMATOGRAPHY.RETENTION_TIME")
#' mb <- insertExtraData(mb, "Infolist_PNEG_20eV_modified.csv", "AC$CHROMATOGRAPHY.CCS")
#' mb <- mbWorkflow(mb, steps = c(5:8))
#' }
#' 
insertExtraData <- function(mb, infolist, column) {
  infolist_ <- read.csv(infolist)
  mb@compiled_ok <- mb@compiled_ok %>%
    map(function(cpd) {
      
      infolist_row <- infolist_ %>%
        filter(id == as.numeric(cpd@id)) %>%
        as.list()
      cpd@children <- cpd@children %>%
        as.list() %>%
        map(function(sp) {
          
          target <- str_split(column, fixed("."))
          source <- str_replace_all(column, fixed("$"), ".")
          
          
          # Note: We don't take care of reordering here,
          # so use at your own peril
          if(length(target) == 2) {
            sp@info[[target[[1]]]][[target[[2]]]] <- 
              infolist_row[[source]]
          } else if(length(target) == 1) {
            sp@info[[target[[1]]]] <- 
              infolist_row[[source]]
          }
          sp
        }) %>%
        as("SimpleList")
      cpd
    })
  mb
}