R/Normalization.R

In jtcasemajor/CPRD: Comparison of DEG and coexpression tools

###########################
#####  Normalization ######
###########################

library(nanostringr)
library(NanoStringNorm)
library(NAPPA)
library(nanoR)

#' Normalize a Nanostring dataset
#'
#' Normalization through different methods in several packages
#'
#' @param raw.data rcc type file. List of 4 elements: 
#' Samples.IDs is a dataframe with four columns: sample name, status (WildType, Mutated), filename. 
#' rcc.df contains expression levels with samples in column and genes in rows. 
#' annots.df is a dataframe with 3 columns: geneclass (endogenous, housekeeping, negative, positive), gene name, and the accession number. 
#' FOV is a dataframe that contains Field of views information.
#' 
#' @param tool Normalization tool. "nappa.NS", "nappa.param1","nappa.param2","nappa.param3" are different parameters used with the NAPPA() function from the NAPPA package.
#'  "nanostringnorm.default","nanostringnorm.param1","nanostringnorm.param2" use the NanoStringNorm() normalization function from the package NanoStringNorm
#'  "nanostringR" uses the HKnorm() function from the package nanostringr.
#'  "nanoR.top100","nanoR.total" uses the nsNormalize() function from the nanoR package.
#'  For the nanoR package, it is needed to give the file path to rcc files
#'
#' @param nanoR Logical value. TRUE is necessary if the tool used is "nanoR.top100" or "nanoR.total". By default, the value is set on FALSE
#' @param dir directory of rcc files. 
#' This parameter is only necessary if the nanoR normalizations are wanted.
#'
#' @return dataframe of normalized expression values
#' 
#' @import "nanostringr" "NanoStringNorm" "NAPPA" "nanoR"
#' @export
#'
#' @examples
#' # Retrieve Nanostring Data
#' # Data = Simul.data(type = "Nanostring")
#' # Normalize data using one method :
#' # Norm.data = tools.norm.Nanostring(raw.data = Data, tool = "nappa.NS",nanoR=F)
#' #with nanoR : Give the rcc files location
#' #RCC.dir <- file.path("./DATA/NANOSTRING","GSE146204_RAW")
#' #Norm.data = tools.norm.Nanostring(raw.data = Data, tool = "nanoR.top100",dir = RCC.dir,nanoR=T)
tools.norm.Nanostring <- function(raw.data,tool,nanoR=F,dir = NULL){
  # if the method is "nanoR.top100" or "nanoR.total", the function needs other informations contained in other files 
  if(!nanoR){
    rcc.samples <- raw.data$rcc.df
    annots.df <- raw.data$annots.df
    samples.IDs <- raw.data$samples.IDs
    FOV <- raw.data$FOV
    dir = NULL
  }
  
  if (missing(dir) & (nanoR = T)) {
    data.dir <- "./DATA/NANOSTRING"
    RCC.dir <- file.path(data.dir,"GSE146204_RAW")
    samples.IDs <- raw.data$samples.IDs
  }
  else if(nanoR){
    samples.IDs <- raw.data$samples.IDs
    raw.data = dir
  }
  
  tools.fnc <- switch(tool,
                      nanostringR={
                        rcc.samples <- cbind(annots.df,rcc.samples)
                        colnames(rcc.samples)[1:3] <- c("Code.Class", "Name", "Accession")
                        res.norm <- HKnorm(rcc.samples)
                        row.names(res.norm) <- res.norm$Name
                        res.norm <- res.norm[,-(1:3)]
                        colnames(res.norm) <- samples.IDs$ID
                        res.norm
                      },
                      nappa.NS={
                        rcc.samples <- cbind(annots.df,rcc.samples)
                        hk <- rcc.samples[grep("Housekeeping",rcc.samples$CodeClass),]
                        hk$CodeClass <- "Endogenous"
                        rcc.samples <- rbind(rcc.samples,hk)
                        res.norm <- NAPPA(rcc.samples,tissueType = "tumour",nposcontrols = 6,scaleFOV = F,output="All",raise.low.counts=1,poscontrol.method = "geometric.mean",background.method = "none",hk.method = "subtract")$GeneExpression
                        colnames(res.norm) <- samples.IDs$ID
                        res.norm
                      },  
                      nappa.default={
                        rcc.samples <- cbind(annots.df,rcc.samples)
                        rcc.samples <- rbind(FOV,rcc.samples)
                        hk <- rcc.samples[grep("Housekeeping",rcc.samples$CodeClass),]
                        hk$CodeClass <- "Endogenous"
                        rcc.samples <- rbind(rcc.samples,hk)
                        res.norm <- NAPPA(rcc.samples,tissueType = "tumour")
                        colnames(res.norm) <- samples.IDs$ID
                        res.norm
                        
                      },  
                      
                      nappa.param1={
                        rcc.samples <- cbind(annots.df,rcc.samples)
                        rcc.samples <- rbind(FOV,rcc.samples)
                        hk <- rcc.samples[grep("Housekeeping",rcc.samples$CodeClass),]
                        hk$CodeClass <- "Endogenous"
                        rcc.samples <- rbind(rcc.samples,hk)
                        res.norm <- NAPPA(rcc.samples,tissueType = "tumour",nposcontrols = 6,scaleFOV = TRUE,output="All",raise.low.counts=25)$GeneExpression
                        colnames(res.norm) <- samples.IDs$ID
                        res.norm
                      },
                      nappa.param3={
                        rcc.samples <- cbind(annots.df,rcc.samples)
                        rcc.samples <- rbind(FOV,rcc.samples)
                        hk <- rcc.samples[grep("Housekeeping",rcc.samples$CodeClass),]
                        hk$CodeClass <- "Endogenous"
                        rcc.samples <- rbind(rcc.samples,hk)
                        res.norm <- NAPPA(rcc.samples,tissueType = "tumour",nposcontrols = 6,scaleFOV = TRUE,output="All",raise.low.counts=1)$GeneExpression
                        colnames(res.norm) <- samples.IDs$ID
                        res.norm
                      },
                      nappa.param2={
                        rcc.samples <- cbind(annots.df,rcc.samples)
                        rcc.samples <- rbind(FOV,rcc.samples)
                        hk <- rcc.samples[grep("Housekeeping",rcc.samples$CodeClass),]
                        hk$CodeClass <- "Endogenous"
                        rcc.samples <- rbind(rcc.samples,hk)
                        res.norm <- NAPPA(rcc.samples,tissueType = "tumour",sampleNumber=10,hk.method="shrunken.subtract",nposcontrols = 6,scaleFOV = TRUE,output="All",raise.low.counts=25)$GeneExpression
                        colnames(res.norm) <- samples.IDs$ID
                      },
                      nanostringnorm.default={
                        rcc.samples <- cbind(annots.df,rcc.samples)
                        rcc.samples$CodeClass[rcc.samples$CodeClass=="Housekeeping"] <- "Endogenous"
                        res.norm <- NanoStringNorm(rcc.samples,take.log = TRUE,return.matrix.of.endogenous.probes = TRUE)
                      },
                      
                      nanostringnorm.param1={
                        rcc.samples <- cbind(annots.df,rcc.samples)
                        rcc.samples$CodeClass[rcc.samples$CodeClass=="Housekeeping"] <- "Endogenous"
                        res.norm <- NanoStringNorm(rcc.samples,CodeCount = 'geo.mean',Background = 'mean.2sd',SampleContent = 'low.cv.geo.mean',round.values = TRUE,take.log = TRUE,return.matrix.of.endogenous.probes = TRUE)
                      },
                      nanostringnorm.param2={
                        design <- model.matrix(~0+samples.IDs$tp53.status)
                        colnames(design) <- c("Mutated","WildType")
                        rcc.samples <- cbind(annots.df,rcc.samples)
                        rcc.samples$CodeClass[rcc.samples$CodeClass=="Housekeeping"] <- "Endogenous"
                        res.norm <- NanoStringNorm(rcc.samples,Background = 'mean.2sd',SampleContent = 'low.cv.geo.mean',CodeCount="sum",traits=design,round.values = TRUE,take.log = TRUE,return.matrix.of.endogenous.probes = TRUE)
                        
                      },
                      nanoR.top100={
                        nano <- parseRCC(dir = raw.data)
                        nano <- nsBackgroundCorrect(nano)
                        nano <- nsPositiveControlNormalization(nano)
                        res.norm <- nsNormalize(nano, method="top100")$bg.corr.counts
                        res.norm <- log2(res.norm[!grepl("Neg|Pos",res.norm$CodeClass),-c(1:3)]+1)
                        colnames(res.norm) <- samples.IDs$title
                        res.norm
                        
                      },
                      nanoR.total={
                        nano <- parseRCC(dir = raw.data)
                        nano <- nsBackgroundCorrect(nano)
                        nano <- nsPositiveControlNormalization(nano)
                        res.norm <- nsNormalize(nano, method="total")$bg.corr.counts
                        res.norm <- log2(res.norm[!grepl("Neg|Pos",res.norm$CodeClass),-c(1:3)]+1)
                        colnames(res.norm) <- samples.IDs$title
                        res.norm
                      },
                      stop("Enter something that switches me!")          
  )
  return(res.norm)
}