88 lines
3.2 KiB
R
88 lines
3.2 KiB
R
#AddShiftVals2Dev.R
|
|
#for use as a command line script (Rscript)
|
|
#need two starting files
|
|
#first file is the finalTable.csv created by clustering
|
|
#second file needs to contain the list of ORFs and Genes and the shift data for L and K for each gene
|
|
#third need to give the name of what the output should be called
|
|
#May want to reorder columns in excel before making heatmaps - otherwise all the shift data will be plotted next to each other.
|
|
version
|
|
timestamp()
|
|
library(plyr)
|
|
library(dplyr)
|
|
library(sos)
|
|
|
|
Args = commandArgs(TRUE)
|
|
#GTF/REMcOutput/20_0310_2KR_vs_WT_EG_Above2SD_NoDAMPs_REMcReady.csv-finalTable.csv ../JoinFiles/Shift_only.csv REMcHeatmaps/REMcWithShift.csv
|
|
|
|
#Args= "GTF/REMcOutput/TESTjwr4exp_REMcReady.csv-finalTable.csv" #20_0310_2KR_vs_WT_EG_Above2SD_NoDAMPs_REMcReady.csv-finalTable.csv"
|
|
#Args[2]= "../JoinFiles/Shift_only.csv"
|
|
#Args[3]= "REMcHeatmaps/REMcWithShift.csv"
|
|
|
|
finaltable = "REMcRdy_lm_only.csv-finalTable.csv" #Args[1]
|
|
ShiftFile = "Shift_only.csv" #Args[2]
|
|
output = "REMcHeatmaps/REMcWithShift.csv" #Args[3]
|
|
|
|
#read in the REMc finalTable data
|
|
X = data.frame(read.csv(file=finaltable,header=TRUE,stringsAsFactors = FALSE))
|
|
#read in the shift data From ../JoinInteractions
|
|
Y = data.frame(read.csv(file=ShiftFile,header=TRUE,stringsAsFactors = FALSE))
|
|
Labels <- read.delim("../Code/StudyInfo.csv",skip=0,as.is=T,row.names=1,strip.white=TRUE)
|
|
|
|
#determine the number of cols - needed to create the correct number of new cols
|
|
Xcolnum <- length(X[1,])
|
|
ADDnum <- Xcolnum + length(Y[1,]) - 2
|
|
|
|
#create new columns filled with NAs to be filled with data
|
|
Xtemp= X
|
|
Xtemp[,(Xcolnum+1):ADDnum] <- NA
|
|
|
|
#match the orf names in each row to a orf name in the shift data file and then add the shift data to the finalTable file
|
|
shiftTbl <-as.data.frame(matrix(nrow=1,ncol=length(Y)-2)) #the df shiftTbl must be initialized before for loop
|
|
|
|
for(i in 1:length(X[,1])){
|
|
Shiftrownum = match(X[i,2],Y[,1])
|
|
shiftTbl[i,]= Y[Shiftrownum,3:length(Y[1,])]
|
|
Xtemp[i,(Xcolnum+1):ADDnum] <- Y[Shiftrownum,3:length(Y[1,])]
|
|
}
|
|
headerX= colnames(Xtemp)
|
|
headerY= colnames(Y)
|
|
shfHdr= headerY[3:length(headerY)]
|
|
combTbl<- X[,1:3]
|
|
lmTbl= select(Xtemp, contains('Z_lm')) #X[,(4:Xcolnum-2)]
|
|
shiftTbl<- select(Xtemp, contains('V'))
|
|
clustTbl<- select(Xtemp, contains('cluster.'))
|
|
|
|
|
|
#give the new column names the same names as in the shift file
|
|
Xcols = colnames(X)
|
|
Ycols = colnames(Y)[3:length(Y[1,])]
|
|
newCols = c(Xcols[1:Xcolnum],Ycols)
|
|
|
|
|
|
#+++++++++++++++++++++REORDER COLUMNS FOR GENERATING HEATMAPS+++++++++++++++++++++++
|
|
|
|
combI= combTbl #Starting Template orf, Genename columns
|
|
headersRemc<-newCols #colnames(X)
|
|
newHeaders= newCols[1:3]
|
|
lmHdr= colnames(lmTbl) #newCols[4:(length(Xcols)-2)]
|
|
clstHdr= colnames(clustTbl) #select(newCols, contains('cluster.')) #newCols[3+length(lmHdr):2]
|
|
|
|
intLvHdr= vector()
|
|
#Reorder columns to produce an interleaved set of Z_lm and Shift data for all the cpps.
|
|
for(i in 1:(length(shiftTbl[1,]))){
|
|
combI=cbind.data.frame(combI, shiftTbl[i])
|
|
combI=cbind.data.frame(combI, lmTbl[i])
|
|
intLvHdrx= c(shfHdr[i],lmHdr[i] )
|
|
intLvHdr= c(intLvHdr,intLvHdrx)
|
|
}
|
|
|
|
combIHdr= c(colnames(combTbl),intLvHdr,clstHdr)
|
|
combI=cbind.data.frame(combI, clustTbl)
|
|
colnames(combI)= combIHdr
|
|
|
|
|
|
write.csv(combI,file=output, row.names=FALSE)
|
|
|
|
timestamp()
|
|
|