R/canonicalFormat.R

In SwathXtend: SWATH extended library generation and statistical data analysis

###########################################################################
#' Standardise a sprectrum library data frame
#' @param dat a data frame of a spectrum library
#' @param format a character string, representing the format of the input
#' spectrum library. One of "peakview" (default) and "openswath"
#' #' @return a data frame of the library in canonical format
#' @examples 
#' file <- paste(system.file("files",package="SwathXtend"),"Lib1.txt",sep="/")
#' dat <- read.delim2(file,sep="\t",stringsAsFactor = FALSE, header=TRUE)
#' dat <- try(canonicalFormat(dat, format = "peakview"))
############################################################################ 

canonicalFormat = function(dat, format=c("peakview", "openswath") )
{
  format <- match.arg(format) 
  if(format == "peakview"){
  
    dat.res <- dat
    dat.res$Q1 <- as.numeric(as.character(dat.res$Q1))
    dat.res$Q3 <- as.numeric(as.character(dat.res$Q3))    
    dat.res$RT_detected <- as.numeric(as.character(dat.res$RT_detected))
#     dat.res$score <- as.numeric(as.character(dat.res$score))
    dat.res$isotype <- as.factor(dat.res$isotype) 
    dat.res$iRT <- as.numeric(as.character(dat.res$iRT))
    dat.res$relative_intensity <- as.numeric(as.character(dat.res$relative_intensity))
    dat.res$decoy <- as.factor(dat.res$decoy)
    dat.res$shared <- as.factor(dat.res$shared)
    dat.res$confidence <- as.numeric(as.character(dat.res$confidence))

  } else { # OpenSWATH
    
    Q1 <- as.numeric(as.character(dat$PrecursorMz))
    Q3 <- as.numeric(as.character(dat$ProductMz))
    RT_detected <- as.numeric(as.character(dat$Tr_recalibrated))
    protein_name <- dat$ProteinName
    isotype <- as.factor(dat$GroupLabel)
    iRT <- as.numeric(as.character(dat$Tr_recalibrated))
    relative_intensity <- as.numeric(as.character(dat$LibraryIntensity))
    stripped_sequence <- dat$PeptideSequence
    modification_sequence <- dat$FullUniModPeptideName
    uniprot_id <- dat$UniprotID
    decoy <- as.factor(dat$decoy)
    shared <- c(rep(FALSE, nrow(dat)))
    confidence <- c(rep(1, nrow(dat)))
    prec_z <- as.numeric(as.character(dat$PrecursorCharge))
    frg_type <- dat$FragmentType
    frg_z <- as.numeric(as.character(dat$FragmentCharge))
    frg_nr <- as.numeric(as.character(dat$FragmentSeriesNumber))
    N <- c(rep(1, nrow(dat)))
    
    
    dat.res <- data.frame(Q1,Q3,RT_detected,protein_name,isotype,
                          iRT,relative_intensity,stripped_sequence,
                          modification_sequence,uniprot_id,decoy,shared,
                          confidence,prec_z,frg_type,frg_z,
                          frg_nr,N)
    
    if(is.factor(dat.res$protein_name))
      dat.res$protein_name <- as.character(dat.res$protein_name)
    if(is.factor(dat.res$stripped_sequence))
      dat.res$stripped_sequence <- as.character(dat.res$stripped_sequence)
    if(is.factor(dat.res$modification_sequence))
      dat.res$modification_sequence <- as.character(dat.res$modification_sequence)
    if(is.factor(dat.res$uniprot_id))
      dat.res$uniprot_id <- as.character(dat.res$uniprot_id)
    if(is.factor(dat.res$frg_type))
      dat.res$frg_type <- as.character(dat.res$frg_type)
  
  }
  
dat.res


}