lohhla icon indicating copy to clipboard operation
lohhla copied to clipboard

unable to run lohhla on test data.

Open iichelhadi opened this issue 3 years ago • 2 comments

I have installed and tried to run the program on hpc conda environment on the test data but encounter the following error image

I tried to locate where the error occurs in the R script and found the first one to be in this chunk of code:

# and filter out reads that have too many events -- has to be done here because some reads map to multiple alleles
      passed.reads <- count.events(paste(regionDir, '/', BAMid, '.type.', allele, '.bam', sep = ''), n = numMisMatch)
     write.table(passed.reads, file = paste(regionDir, '/', BAMid, '.', allele, '.passed.reads.txt', sep = ''), sep = '\t', quote = FALSE, row.names = FALSE, col.names = FALSE)
      extractCMD <- paste("java -jar ",GATKDir,"/FilterSamReads.jar ", "I=",  regionDir, '/', BAMid, '.type.', allele, '.bam' , " FILTER=includeReadList READ_LIST_FILE=", regionDir, "/", BAMid, '.', allele, ".$
      write.table(extractCMD, file = log.name, quote = FALSE, row.names = FALSE, col.names = FALSE, append = TRUE)
      system(extractCMD)

      samtoolsIndex <- paste("samtools index ",regionDir,"/",BAMid,".type.",allele, ".filtered.bam",sep="")
      write.table(samtoolsIndex, file = log.name, quote = FALSE, row.names = FALSE, col.names = FALSE, append = TRUE)
      system(samtoolsIndex)
    }

I commented that part out and proceeded with running the script but encountered the another issue at this part of the code

#HLAtype 1 coverage"
      {
        HLA_A_type1normalLoc <- grep(pattern = HLA_A_type1, x = list.files(workDir, pattern = "normal.mpileup", full.names = TRUE), value = TRUE)
        if(runWithNormal){
          HLA_A_type1normal <- read.table(HLA_A_type1normalLoc ,sep="\t",stringsAsFactors=FALSE,quote="",fill=TRUE)
       	}

help would be greatly appreciated. thank you

iichelhadi avatar Sep 28 '21 05:09 iichelhadi

Did you find any solution?

al3xMlt avatar Feb 08 '23 07:02 al3xMlt

No I just commented things I didnt need and go it working but it wasn't efficient

iichelhadi avatar Feb 09 '23 15:02 iichelhadi