Ticket #486: Illumina conversion.txt

# Set the wd:
setwd("C:/BASE/Illumina imports/BeadStudio output")

# Choose the file:
filename <- "gene_profile.csv"

#Set the number of arrays per chip:
arraysPerChip = 8; # 6
columnsPerArray = 8;

# assign the column headers:
headerColumnNames <-c("MIN_Signal","AVG_Signal","MAX_Signal","NARRAYS","ARRAY_STDEV","BEAD_STDEV","Avg_NBEADS","Detection"

illumina <- read.csv(file=paste(getwd(),"/",filename,sep=""),skip=7,header=TRUE,row.name=1)

ncol(illumina)
nrow(illumina)

numData = ncol(illumina) - 
numArrays = numData / columnsPerArray # 8 data columns per array
numChips = numArrays/arraysPerChip  #

numChips

if numChips%%1 ~= 0
  "ERROR: an incomplete chip is in this file!"

colnames(illumina) -> myCols

# Derive the Chip names from the colnames:
# Format of first entry: "MIN_Signal.XXXXXXXXXX_A" ( 10 Xs: maybe different?)

lenFirstData <-nchar(headerColumnNames[1])

chipNums = NULL
arrayNames = NULL

# Derive names of arrays:
for (k in 1:numChips){
    
  # Get the numbers of the chips:
colNum = 1 + (k-1)* (arraysPerChip *columnsPerArray)
chipNums = c(chipNums, substr(myCols [colNum],lenFirstData+2,nchar(myCols[colNum])-2))

  # Get names of all the arrays
  labels = c("A","B","C","D","E","F","G","H")
  labels = labels[1: arraysPerChip]
  arrayNames = c(arrayNames, paste(chipNums[k],labels,sep="_"))
}
chipNums
arrayNames


# output arrays (groups of columnsPerArray columns, plus gene names)
for (k in 1:length(arrayNames))
{
  outputCols <- c(1: columnsPerArray) + columnsPerArray * (k-1)
  outputData <- illumina[,outputCols]
  outputData <- cbind(rownames(illumina),outputData) 

  colnames(outputData) <- as.character(c("TargetID",headerColumnNames))

  write.table(outputData,file = paste(arrayNames[k],".txt",sep=""),sep = "\t",row.names=FALSE,na="")
}