2024 Dev/GlowwormIntegrate.R

In Hannahglover/Glowworm: Mapping genes to single cell RNAseq

#' @title GlowwormIntegrate
#' @description Integrate GlowwormScale objects
#' @param DataList a list of named GlowwormScale objects processed for the same target and secondary gene lists
#' @return This function takes a list of named GlowwormScale objects and produces a single GlowwormScale object containing:
#' \itemize{
#'     \item MetaData - A copy of the metadata
#'     \item GlowwormScaleOutput - A scaled data frame of n genes by n populations in metadata
#'     \item RankScores - A data frame indicating the significantly enriched cell populations in the target gene list compared to the background lust
#'     \item SumStats - A data frame showing expression level and specificity for each gene.
#'     \item Settings - A list of settings
#'         }
#'
#' @export

GlowwormIntegrate = function(DataList){
  for(LL in 1:length(DataList)){
    if(is.null(names(DataList[LL]))){
      DATANAME = paste("Data", LL)  
    }else{
      DATANAME = names(DataList[LL])
    }   
    
    if(LL == 1){
      DataRecompile = DataList[[LL]]
      DataRecompile@MetaData$Dataset = DATANAME
      DataRecompile@RankScores$Dataset = DATANAME
      row.names(DataRecompile@GlowwormScaleOutput) = paste(row.names(DataRecompile@GlowwormScaleOutput), DATANAME, sep="^")
    }else{
      DataList[[LL]]@MetaData$Dataset = DATANAME
      DataRecompile@MetaData = dplyr::bind_rows(DataRecompile@MetaData, DataList[[LL]]@MetaData)
      row.names(DataList[[LL]]@GlowwormScaleOutput) = paste(row.names(DataList[[LL]]@GlowwormScaleOutput), DATANAME, sep="^")
      DataRecompile@GlowwormScaleOutput = dplyr::bind_rows(DataRecompile@GlowwormScaleOutput, DataList[[LL]]@GlowwormScaleOutput)
      DataList[[LL]]@RankScores$Dataset = DATANAME
      DataRecompile@RankScores = dplyr::bind_rows(DataRecompile@RankScores, DataList[[LL]]@RankScores)  
      DataRecompile@SumStats = dplyr::bind_rows(DataRecompile@SumStats, DataList[[LL]]@SumStats)  
      DataRecompile@Settings[["InputFile"]] = paste(DataRecompile@Settings[["InputFile"]], DataList[[LL]]@Settings[["InputFile"]], sep = "/")
      DataRecompile@Settings[["TargetGenes"]] = unique(c(DataRecompile@Settings[["TargetGenes"]], DataList[[LL]]@Settings[["TargetGenes"]]))
      #DataRecompile@Settings[["SecondaryGenes"]] = DataRecompile@Settings[["SecondaryGenes"]]
      DataRecompile@Settings[["UniqueSecondaryGenes"]] = intersect(DataRecompile@Settings[["UniqueSecondaryGenes"]], DataList[[LL]]@Settings[["UniqueSecondaryGenes"]])
      DataRecompile@Settings[["N_Classes"]] = DataRecompile@Settings[["N_Classes"]] +  DataList[[LL]]@Settings[["N_Classes"]]
      DataRecompile@Settings[["N_Subclasses"]] = DataRecompile@Settings[["N_Subclasses"]] +  DataList[[LL]]@Settings[["N_Subclasses"]]
      DataRecompile@Settings[["N_TotalClusters"]] = DataRecompile@Settings[["N_TotalClusters"]] +  DataList[[LL]]@Settings[["N_TotalClusters"]]
    }
  }
  DataRecompile@Settings[["MissingTargetGenes"]] = NULL
  DataRecompile@GlowwormScaleOutput = DataRecompile@GlowwormScaleOutput %>% dplyr::select(c( DataRecompile@Settings[["TargetGenes"]], DataRecompile@Settings[["UniqueSecondaryGenes"]]))
  DataRecompile@SumStats = subset(DataRecompile@SumStats, row.names(DataRecompile@SumStats) %in% c(DataRecompile@Settings[["TargetGenes"]], DataRecompile@Settings[["UniqueSecondaryGenes"]]))
  
  setMethod("show",signature = signature(object = "GlowwormObj"),
            function(object){
              print_message <- paste(c("\nGlowwormObj for ", DataRecompile@Settings$InputFile, " for ", length(unique(DataRecompile@Settings$TargetGenes)), "target genes and ", length(unique(DataRecompile@Settings$UniqueSecondaryGenes)), " background genes for ",  DataRecompile@Settings$N_TotalClusters, " total clusters.\n", sep=""))
              cat(print_message,sep = "")
            })
  
  return(DataRecompile)
}