@Override
protected void finish() {
multiCollector.finish();
final MetricsFile<InsertSizeMetrics, Integer> file = getMetricsFile();
multiCollector.addAllLevelsToFile(file);
if (file.getNumHistograms() == 0) {
// can happen if user sets MINIMUM_PCT = 0.5, etc.
log.warn(
"All data categories were discarded because they contained < "
+ MINIMUM_PCT
+ " of the total aligned paired data.");
final InsertSizeMetricsCollector.PerUnitInsertSizeMetricsCollector allReadsCollector =
(InsertSizeMetricsCollector.PerUnitInsertSizeMetricsCollector)
multiCollector.getAllReadsCollector();
log.warn(
"Total mapped pairs in all categories: "
+ (allReadsCollector == null
? allReadsCollector
: allReadsCollector.getTotalInserts()));
} else {
file.write(OUTPUT);
final int rResult;
if (HISTOGRAM_WIDTH == null) {
rResult =
RExecutor.executeFromClasspath(
HISTOGRAM_R_SCRIPT,
OUTPUT.getAbsolutePath(),
HISTOGRAM_FILE.getAbsolutePath(),
INPUT.getName());
} else {
rResult =
RExecutor.executeFromClasspath(
HISTOGRAM_R_SCRIPT,
OUTPUT.getAbsolutePath(),
HISTOGRAM_FILE.getAbsolutePath(),
INPUT.getName(),
String.valueOf(
HISTOGRAM_WIDTH)); // HISTOGRAM_WIDTH is passed because R automatically sets
// histogram width to the last
// bin that has data, which may be less than HISTOGRAM_WIDTH and confuse the user.
}
if (rResult != 0) {
throw new PicardException(
"R script " + HISTOGRAM_R_SCRIPT + " failed with return code " + rResult);
}
}
}
/**
* Asserts that files are readable and writable and then fires off an HsMetricsCalculator instance
* to do the real work.
*/
protected int doWork() {
IoUtil.assertFileIsReadable(getProbeIntervals());
IoUtil.assertFileIsReadable(TARGET_INTERVALS);
IoUtil.assertFileIsReadable(INPUT);
IoUtil.assertFileIsWritable(OUTPUT);
if (PER_TARGET_COVERAGE != null) IoUtil.assertFileIsWritable(PER_TARGET_COVERAGE);
final SAMFileReader samReader = new SAMFileReader(INPUT);
final File probeIntervals = getProbeIntervals();
// Validate that the targets and baits have the same references as the reads file
SequenceUtil.assertSequenceDictionariesEqual(
samReader.getFileHeader().getSequenceDictionary(),
IntervalList.fromFile(TARGET_INTERVALS).getHeader().getSequenceDictionary(),
INPUT,
TARGET_INTERVALS);
SequenceUtil.assertSequenceDictionariesEqual(
samReader.getFileHeader().getSequenceDictionary(),
IntervalList.fromFile(probeIntervals).getHeader().getSequenceDictionary(),
INPUT,
probeIntervals);
ReferenceSequenceFile ref = null;
if (REFERENCE_SEQUENCE != null) {
IoUtil.assertFileIsReadable(REFERENCE_SEQUENCE);
ref = ReferenceSequenceFileFactory.getReferenceSequenceFile(REFERENCE_SEQUENCE);
SequenceUtil.assertSequenceDictionariesEqual(
samReader.getFileHeader().getSequenceDictionary(),
ref.getSequenceDictionary(),
INPUT,
REFERENCE_SEQUENCE);
}
final TargetMetricsCollector collector =
makeCollector(
METRIC_ACCUMULATION_LEVEL,
samReader.getFileHeader().getReadGroups(),
ref,
PER_TARGET_COVERAGE,
TARGET_INTERVALS,
probeIntervals,
getProbeSetName());
// Add each record to the requested collectors
final Iterator<SAMRecord> records = samReader.iterator();
final ProgressLogger progress = new ProgressLogger(log);
while (records.hasNext()) {
final SAMRecord sam = records.next();
collector.acceptRecord(sam, null);
progress.record(sam);
}
// Write the output file
final MetricsFile<HsMetrics, Integer> metrics = getMetricsFile();
collector.finish();
collector.addAllLevelsToFile(metrics);
metrics.write(OUTPUT);
return 0;
}