Examples with OverrepresentedSequence ca.corefacility.bioinformatics.irida.model.sequenceFile.OverrepresentedSequence used on opensource projects

Example 1 with OverrepresentedSequence

use of ca.corefacility.bioinformatics.irida.model.sequenceFile.OverrepresentedSequence in project irida by phac-nml.

the class SequencingObjectServiceImplIT method testCreateNotCompressedSequenceFile.

@Test
@WithMockUser(username = "fbristow", roles = "SEQUENCER")
public void testCreateNotCompressedSequenceFile() throws IOException, InterruptedException {
    final Long expectedRevisionNumber = 1L;
    SequenceFile sf = createSequenceFile("file1");
    Path sequenceFile = sf.getFile();
    SingleEndSequenceFile singleEndSequenceFile = new SingleEndSequenceFile(sf);
    logger.trace("About to save the file.");
    SequencingObject sequencingObject = asRole(Role.ROLE_SEQUENCER, "fbristow").objectService.create(singleEndSequenceFile);
    logger.trace("Finished saving the file.");
    assertNotNull("ID wasn't assigned.", sequencingObject.getId());
    // Sleeping for a bit to let file processing run
    Thread.sleep(10000);
    // figure out what the version number of the sequence file is (should be
    // 1; the file wasn't gzipped, but fastqc will have modified it.)
    SequencingObject readObject = null;
    do {
        readObject = asRole(Role.ROLE_ADMIN, "admin").objectService.read(sequencingObject.getId());
        sf = readObject.getFiles().iterator().next();
        if (sf.getFileRevisionNumber() < expectedRevisionNumber) {
            logger.info("Still waiting on thread to finish, having a bit of a sleep.");
            Thread.sleep(1000);
        }
    } while (sf.getFileRevisionNumber() < expectedRevisionNumber);
    assertEquals("Wrong version number after processing.", expectedRevisionNumber, sf.getFileRevisionNumber());
    // verify the file checksum was taken properly
    assertEquals("checksum should be equal", CHECKSUM, sf.getUploadSha256());
    AnalysisFastQC analysis = asRole(Role.ROLE_ADMIN, "admin").analysisService.getFastQCAnalysisForSequenceFile(readObject, sf.getId());
    assertNotNull("FastQCAnalysis should have been created for the file.", analysis);
    Set<OverrepresentedSequence> overrepresentedSequences = analysis.getOverrepresentedSequences();
    assertNotNull("No overrepresented sequences were found.", overrepresentedSequences);
    assertEquals("Wrong number of overrepresented sequences were found.", 1, overrepresentedSequences.size());
    OverrepresentedSequence overrepresentedSequence = overrepresentedSequences.iterator().next();
    assertEquals("Sequence was not the correct sequence.", SEQUENCE, overrepresentedSequence.getSequence());
    assertEquals("The count was not correct.", 2, overrepresentedSequence.getOverrepresentedSequenceCount());
    assertEquals("The percent was not correct.", new BigDecimal("100.00"), overrepresentedSequence.getPercentage());
    // confirm that the file structure is correct
    Path idDirectory = baseDirectory.resolve(Paths.get(sf.getId().toString()));
    assertTrue("Revision directory doesn't exist.", Files.exists(idDirectory.resolve(Paths.get(sf.getFileRevisionNumber().toString(), sequenceFile.getFileName().toString()))));
    // no other files or directories should be beneath the ID directory
    int fileCount = 0;
    Iterator<Path> dir = Files.newDirectoryStream(idDirectory).iterator();
    while (dir.hasNext()) {
        dir.next();
        fileCount++;
    }
    assertEquals("Wrong number of directories beneath the id directory", 1, fileCount);
}

Example 2 with OverrepresentedSequence

use of ca.corefacility.bioinformatics.irida.model.sequenceFile.OverrepresentedSequence in project irida by phac-nml.

the class FastqcFileProcessor method handleOverRepresentedSequences.

/**
 * Handle getting over represented sequences from fastqc.
 *
 * @param seqs
 *            overrepresented sequences.
 * @return a collection of {@link OverrepresentedSequence} corresponding to
 *         the FastQC {@link OverRepresentedSeqs}.
 */
private Set<OverrepresentedSequence> handleOverRepresentedSequences(OverRepresentedSeqs seqs) {
    OverrepresentedSeq[] sequences = seqs.getOverrepresentedSequences();
    if (sequences == null) {
        return Collections.emptySet();
    }
    Set<OverrepresentedSequence> overrepresentedSequences = new HashSet<>(sequences.length);
    for (OverrepresentedSeq s : sequences) {
        String sequenceString = s.seq();
        int count = s.count();
        BigDecimal percent = BigDecimal.valueOf(s.percentage());
        String possibleSource = s.contaminantHit();
        overrepresentedSequences.add(new OverrepresentedSequence(sequenceString, count, percent, possibleSource));
    }
    return overrepresentedSequences;
}

Example 3 with OverrepresentedSequence

use of ca.corefacility.bioinformatics.irida.model.sequenceFile.OverrepresentedSequence in project irida by phac-nml.

the class FastqcFileProcessor method processSingleFile.

/**
 * Process a single {@link SequenceFile}
 *
 * @param sequenceFile
 *            file to process
 * @throws FileProcessorException
 *             if an error occurs while processing
 */
private void processSingleFile(SequenceFile sequenceFile) throws FileProcessorException {
    Path fileToProcess = sequenceFile.getFile();
    AnalysisFastQC.AnalysisFastQCBuilder analysis = AnalysisFastQC.builder().executionManagerAnalysisId(EXECUTION_MANAGER_ANALYSIS_ID).description(messageSource.getMessage("fastqc.file.processor.analysis.description", null, LocaleContextHolder.getLocale()));
    try {
        uk.ac.babraham.FastQC.Sequence.SequenceFile fastQCSequenceFile = SequenceFactory.getSequenceFile(fileToProcess.toFile());
        BasicStats basicStats = new BasicStats();
        PerBaseQualityScores pbqs = new PerBaseQualityScores();
        PerSequenceQualityScores psqs = new PerSequenceQualityScores();
        OverRepresentedSeqs overRep = new OverRepresentedSeqs();
        QCModule[] moduleList = new QCModule[] { basicStats, pbqs, psqs, overRep };
        logger.debug("Launching FastQC analysis modules on all sequences.");
        while (fastQCSequenceFile.hasNext()) {
            Sequence sequence = fastQCSequenceFile.next();
            for (QCModule module : moduleList) {
                module.processSequence(sequence);
            }
        }
        logger.debug("Finished FastQC analysis modules.");
        handleBasicStats(basicStats, analysis);
        handlePerBaseQualityScores(pbqs, analysis);
        handlePerSequenceQualityScores(psqs, analysis);
        handleDuplicationLevel(overRep.duplicationLevelModule(), analysis);
        Set<OverrepresentedSequence> overrepresentedSequences = handleOverRepresentedSequences(overRep);
        logger.trace("Saving FastQC analysis.");
        analysis.overrepresentedSequences(overrepresentedSequences);
        sequenceFile.setFastQCAnalysis(analysis.build());
        sequenceFileRepository.saveMetadata(sequenceFile);
    } catch (Exception e) {
        logger.error("FastQC failed to process the sequence file. Stack trace follows.", e);
        throw new FileProcessorException("FastQC failed to parse the sequence file.", e);
    }
}

Example 4 with OverrepresentedSequence

use of ca.corefacility.bioinformatics.irida.model.sequenceFile.OverrepresentedSequence in project irida by phac-nml.

the class SequencingObjectServiceImplIT method testCreateCompressedSequenceFile.

@Test
@WithMockUser(username = "fbristow", roles = "SEQUENCER")
public void testCreateCompressedSequenceFile() throws IOException, InterruptedException {
    final Long expectedRevisionNumber = 2L;
    SequenceFile sf = new SequenceFile();
    Path sequenceFile = Files.createTempFile("TEMPORARY-SEQUENCE-FILE", ".gz");
    OutputStream gzOut = new GZIPOutputStream(Files.newOutputStream(sequenceFile));
    gzOut.write(FASTQ_FILE_CONTENTS);
    gzOut.close();
    sf.setFile(sequenceFile);
    SingleEndSequenceFile singleEndSequenceFile = new SingleEndSequenceFile(sf);
    logger.trace("About to save the file.");
    SequencingObject sequencingObject = objectService.create(singleEndSequenceFile);
    logger.trace("Finished saving the file.");
    assertNotNull("ID wasn't assigned.", sequencingObject.getId());
    // Sleeping for a bit to let file processing run
    Thread.sleep(10000);
    // figure out what the version number of the sequence file is (should be
    // 2; the file was gzipped)
    // get the MOST RECENT version of the sequence file from the database
    // (it will have been modified outside of the create method.)
    SequencingObject readObject = null;
    do {
        readObject = asRole(Role.ROLE_ADMIN, "admin").objectService.read(sequencingObject.getId());
        sf = readObject.getFiles().iterator().next();
        if (sf.getFileRevisionNumber() < expectedRevisionNumber) {
            logger.info("Still waiting on thread to finish, having a bit of a sleep.");
            Thread.sleep(1000);
        }
    } while (sf.getFileRevisionNumber() < expectedRevisionNumber);
    assertEquals("Wrong version number after processing.", expectedRevisionNumber, sf.getFileRevisionNumber());
    assertFalse("File name is still gzipped.", sf.getFile().getFileName().toString().endsWith(".gz"));
    AnalysisFastQC analysis = asRole(Role.ROLE_ADMIN, "admin").analysisService.getFastQCAnalysisForSequenceFile(readObject, sf.getId());
    // verify the file checksum was taken properly
    assertEquals("checksum should be equal", ZIPPED_CHECKSUM, sf.getUploadSha256());
    Set<OverrepresentedSequence> overrepresentedSequences = analysis.getOverrepresentedSequences();
    assertNotNull("No overrepresented sequences were found.", overrepresentedSequences);
    assertEquals("Wrong number of overrepresented sequences were found.", 1, overrepresentedSequences.size());
    OverrepresentedSequence overrepresentedSequence = overrepresentedSequences.iterator().next();
    assertEquals("Sequence was not the correct sequence.", SEQUENCE, overrepresentedSequence.getSequence());
    assertEquals("The count was not correct.", 2, overrepresentedSequence.getOverrepresentedSequenceCount());
    assertEquals("The percent was not correct.", new BigDecimal("100.00"), overrepresentedSequence.getPercentage());
    // confirm that the file structure is correct
    String filename = sequenceFile.getFileName().toString();
    filename = filename.substring(0, filename.lastIndexOf('.'));
    Path idDirectory = baseDirectory.resolve(Paths.get(sf.getId().toString()));
    assertTrue("Revision directory doesn't exist.", Files.exists(idDirectory.resolve(Paths.get(sf.getFileRevisionNumber().toString(), filename))));
    // no other files or directories should be beneath the ID directory
    int fileCount = 0;
    Iterator<Path> dir = Files.newDirectoryStream(idDirectory).iterator();
    while (dir.hasNext()) {
        dir.next();
        fileCount++;
    }
    assertEquals("Wrong number of directories beneath the id directory", 2, fileCount);
}

Example 5 with OverrepresentedSequence

use of ca.corefacility.bioinformatics.irida.model.sequenceFile.OverrepresentedSequence in project irida by phac-nml.

the class FastqcFileProcessorTest method testHandleFastqFile.

@Test
public void testHandleFastqFile() throws IOException, IllegalArgumentException, IllegalAccessException {
    // fastqc shouldn't barf on a fastq file.
    Path fastq = Files.createTempFile(null, null);
    Files.write(fastq, FASTQ_FILE_CONTENTS.getBytes());
    Runtime.getRuntime().addShutdownHook(new DeleteFileOnExit(fastq));
    ArgumentCaptor<SequenceFile> argument = ArgumentCaptor.forClass(SequenceFile.class);
    SequenceFile sf = new SequenceFile(fastq);
    sf.setId(1L);
    SingleEndSequenceFile so = new SingleEndSequenceFile(sf);
    try {
        fileProcessor.process(so);
    } catch (Exception e) {
        e.printStackTrace();
        fail();
    }
    verify(sequenceFileRepository).saveMetadata(argument.capture());
    SequenceFile updatedFile = argument.getValue();
    final Field fastqcAnalysis = ReflectionUtils.findField(SequenceFile.class, "fastqcAnalysis");
    ReflectionUtils.makeAccessible(fastqcAnalysis);
    AnalysisFastQC updated = (AnalysisFastQC) fastqcAnalysis.get(updatedFile);
    assertEquals("GC Content was not set correctly.", Short.valueOf((short) 50), updated.getGcContent());
    assertEquals("Filtered sequences was not 0.", Integer.valueOf(0), updated.getFilteredSequences());
    assertEquals("File type was not correct.", "Conventional base calls", updated.getFileType());
    assertEquals("Max length was not correct.", Integer.valueOf(SEQUENCE.length()), updated.getMaxLength());
    assertEquals("Min length was not correct.", Integer.valueOf(SEQUENCE.length()), updated.getMinLength());
    assertEquals("Total sequences was not correct.", Integer.valueOf(2), updated.getTotalSequences());
    assertEquals("Encoding was not correct.", "Illumina <1.3", updated.getEncoding());
    assertEquals("Total number of bases was not correct.", Long.valueOf(SEQUENCE.length() * 2), updated.getTotalBases());
    assertNotNull("Per-base quality score chart was not created.", updated.getPerBaseQualityScoreChart());
    assertTrue("Per-base quality score chart was created, but was empty.", ((byte[]) updated.getPerBaseQualityScoreChart()).length > 0);
    assertNotNull("Per-sequence quality score chart was not created.", updated.getPerSequenceQualityScoreChart());
    assertTrue("Per-sequence quality score chart was created, but was empty.", ((byte[]) updated.getPerSequenceQualityScoreChart()).length > 0);
    assertNotNull("Duplication level chart was not created.", updated.getDuplicationLevelChart());
    assertTrue("Duplication level chart was not created.", ((byte[]) updated.getDuplicationLevelChart()).length > 0);
    Iterator<OverrepresentedSequence> ovrs = updated.getOverrepresentedSequences().iterator();
    assertTrue("No overrepresented sequences added to analysis.", ovrs.hasNext());
    OverrepresentedSequence overrepresentedSequence = updated.getOverrepresentedSequences().iterator().next();
    assertEquals("Sequence was not the correct sequence.", SEQUENCE, overrepresentedSequence.getSequence());
    assertEquals("The count was not correct.", 2, overrepresentedSequence.getOverrepresentedSequenceCount());
    assertEquals("The percent was not correct.", BigDecimal.valueOf(100.), overrepresentedSequence.getPercentage());
}