Beispiel #1
0
 //	TODO: HANDLE THE CASE WHERE ANY OF THE START/ENDS ARE 0
 private String[] parseBarcode(FastqSequence r) {
   if (this.IN_READ_NAME) {
     return parseBarcodeFromName(r.getName());
   } else {
     return parseBarcodeFromSequence(r.getSequence());
   }
 }
Beispiel #2
0
  private void setBarcode(FastqSequence r, String[] bcInfo) {
    String compoundBC =
        bcInfo[0] + (bcInfo[1] != null && bcInfo[1].length() > 0 ? "_" + bcInfo[1] : "");
    String newName = buildBarcodeFromReadName(r.getName(), compoundBC);

    r.setName(newName);

    /*String comment = r.getDescription();
    if (comment == null || comment.trim().isEmpty()) {
    	comment = compoundBC;
    } else {
    	comment = comment + ":" + compoundBC;
    }
    r.setDescription(comment);*/
  }
Beispiel #3
0
  @Override
  protected int doWork() {
    sampleBarcodeCounts = new HashMap<String, Integer>();
    randomBarcodeCounts = new HashMap<String, Integer>();
    otherBarcodeCounts = new HashMap<String, Integer>();

    sampleNonBarcodeCounts = new HashMap<String, Integer>();
    randomNonBarcodeCounts = new HashMap<String, Integer>();
    otherNonBarcodeCounts = new HashMap<String, Integer>();

    HashMap<String, String> bcSampleMap = null;
    try {
      bcSampleMap = BCEvaluator.readBCMap(BC_SAMPLE_MAP);
    } catch (FileNotFoundException e) {
      System.out.println("Could not access the barcode sample mapping file");
      e.printStackTrace();
      return 1;
    } catch (IOException e) {
      System.out.println("Error accessing barcode sample mapping file");
      e.printStackTrace();
      return 1;
    }

    if (!OUTDIR.exists()) {
      try {
        System.out.println("Output directory does not exists " + OUTDIR + ", going to create it");
        OUTDIR.mkdir();
      } catch (SecurityException e) {
        System.out.println("Could not create output directory");
        e.printStackTrace();
        return 1;
      }
    }

    init();

    FastqParser parser1 = new FastqParser();
    FastqParser parser2 = new FastqParser();

    HashMap<String, BufferedWriter[]> writerMap = null;
    BufferedWriter unassignedP1 = null;
    BufferedWriter unassignedP2 = null;

    try {
      try {
        parser1.start(FASTQ1);
      } catch (IOException e) {
        System.out.println("Error reading fastq file " + FASTQ1);
        e.printStackTrace();
        return 1;
      }

      if (isPaired) {
        try {
          parser2.start(FASTQ2);
        } catch (IOException e) {
          System.out.println("Error reading fastq file " + FASTQ2);
          e.printStackTrace();
          return 1;
        }
      }
      try {
        writerMap = buildWriterMap(bcSampleMap);
        unassignedP1 = new BufferedWriter(new FileWriter(OUTDIR + "/unassigned.P1.fq"));
        if (isPaired) {
          unassignedP2 = new BufferedWriter(new FileWriter(OUTDIR + "/unassigned.P2.fq"));
        }
      } catch (IOException e) {
        System.out.println("Could not create output files for barcode splitting");
        e.printStackTrace();
        return 1;
      }

      while (parser1.hasNext()) {
        FastqSequence r1 = parser1.next();

        String[] bcInfo = parseBarcode(r1); // First position sample, second UMI, third OTHER

        if (bcInfo[0] == null) {
          logger.error(
              "Obtained a null barcode for read " + r1.getName() + " sequence " + r1.getSequence());
        }

        BufferedWriter out1 = unassignedP1;
        BufferedWriter out2 = unassignedP2;

        String bc = getClosestBc(bcSampleMap, bcInfo[0]);
        if (bc != null) {
          // Need to update the BCInfo with the closest barcode once it is found
          bcInfo[0] = bc;
          BufferedWriter[] out = writerMap.get(bc);
          out1 = out[0];
          if (isPaired) {
            out2 = out[1];
          }

          setBarcode(r1, bcInfo);
          if (!IN_READ_NAME) {
            r1.excise(this.BC_START_POS, this.bcLength);
          }
          updateCounts(bcInfo);
        } else {
          updateNonCounts(bcInfo);
        }

        r1.write(out1);
        if (isPaired) {
          FastqSequence r2 = parser2.next();
          if (bc != null) {
            setBarcode(r2, bcInfo);
          }
          // TODO: Handle when/if there is another barcode on R2
          r2.write(out2);
        }

        if (bc != null) {
          updateCounts(bcInfo);
        }
      }

      writeReports();

    } catch (IOException e) {
      System.out.println("Error writing records to file");
      e.printStackTrace();
      return 1;

    } finally {
      try {
        if (parser1 != null) parser1.close();
      } catch (IOException e) {
        System.out.println("Could not close fastq1 file");
      }
      if (isPaired) {
        try {
          parser2.close();
        } catch (IOException e) {
          System.out.println("Could not close fastq2 file");
        }
      }

      try {
        if (unassignedP1 != null) {
          unassignedP1.close();
        }
        if (unassignedP2 != null) {
          unassignedP2.close();
        }
        if (writerMap != null) {
          closeWriters(writerMap);
        }
      } catch (IOException e) {
        System.err.println("Error closing fastq writers");
        e.printStackTrace();
      }
    }

    return 0;
  }