/**
* Checks whether all targets in the input target collection have a designated name.
*
* @param result the query target collection.
* @throws UserException if there are some targets with no name in {@code result}.
*/
private void checkAllTargetsHaveName(TargetCollection<Target> result) {
if (result.targets().stream().anyMatch(t -> t.getName() == null || t.getName().equals(""))) {
throw new UserException(
String.format(
"Target output info requested '%s' requires that each target has a designated unique name/id but there are some with no names: %s",
targetOutInfo.name(),
result
.targets()
.stream()
.filter(t -> t.getName() == null || t.getName().equals(""))
.limit(10)
.map(e -> result.location(e).toString())
.collect(Collectors.joining(", "))));
}
}
/**
* Composes the main output header.
*
* @param commandLine the tool command line.
* @param groupBy the group-by argument used.
* @param countColumnNames the column names.
* @return never {@code null}.
*/
private static String composeMatrixOutputHeader(
final String commandLine,
final TargetOutInfo targetOutInfo,
final GroupBy groupBy,
final List<String> countColumnNames) {
final String countColumnHeaderString =
String.join(COLUMN_SEPARATOR, countColumnNames).replace("%", "%%");
final String formatString =
String.join(
LINE_SEPARATOR,
"##fileFormat = tsv",
"##commandLine = %s",
"##title = Read counts per target and %s",
String.join(COLUMN_SEPARATOR, targetOutInfo.headerString(), countColumnHeaderString));
return String.format(formatString, commandLine, groupBy.toString());
}
/**
* Builds the target collection given the values of user arguments.
*
* @throws UserException if there is some inconsistency in user arguments and inputs.
* @throws GATKException if there was any problem parsing the content of the targets file.
* @return never {@code null}.
*/
private TargetCollection<Target> resolveTargetCollection() {
final TargetCollection<Target> result;
if (targetsFile != null) {
result = resolveTargetsFromFile();
} else if (hasIntervals()) {
final SAMSequenceDictionary sequenceDictionary = getBestAvailableSequenceDictionary();
final List<SimpleInterval> intervals =
intervalArgumentCollection.getIntervals(sequenceDictionary);
// this constructor automatically generates target names from the intervals
final List<Target> targets = intervals.stream().map(Target::new).collect(Collectors.toList());
result = new HashedListTargetCollection<>(targets);
} else {
throw new UserException(
String.format(
"You must indicate the set of target as input intervals (e.g. -L target-intervals.list) or a target feature file (e.g. -%s my-targets.tsv) ",
TARGET_FILE_SHORT_NAME));
}
if (targetOutInfo.requiresUniqueTargetName()) {
checkAllTargetsHaveName(result);
}
return result;
}
/**
* Composes the row summary output header.
*
* @param commandLine the execution command line.
* @param groupBy the value of the group-by argument used.
* @param columnCount number of count-column involved in the analysis.
* @return never {@code null}.
*/
private static String composeRowOutputHeader(
final String commandLine,
final TargetOutInfo targetOutInfo,
final GroupBy groupBy,
final int columnCount) {
return String.format(
String.join(
LINE_SEPARATOR,
"##fileFormat = tsv",
"##commandLine = %s",
"##title = Summary counts per target",
"##metaData = {",
"## groupBy = %s,",
"## columnCount = %d",
"##}",
String.join(
COLUMN_SEPARATOR,
targetOutInfo.headerString(),
SUM_COLUMN_NAME,
AVG_COL_BP_COLUMN_NAME)),
commandLine,
groupBy,
columnCount);
}
/**
* Writes the row in the main matrix output file for a target and, if requested, the corresponding
* row in the row summary output file.
*
* @param countBuffer the counts for the target.
* @param index the index of target within the target collection.
*/
private void writeOutputRows(
final int[] countBuffer, final long[] columnTotals, final int index) {
final String countString =
IntStream.range(0, countBuffer.length)
.mapToObj(i -> transform.apply(countBuffer[i], columnTotals[i]))
.collect(Collectors.joining(COLUMN_SEPARATOR));
final String targetInfoString =
targetOutInfo.composeTargetOutInfoString(index, targetCollection);
outputWriter.println(String.join(COLUMN_SEPARATOR, targetInfoString, countString));
if (rowSummaryOutputWriter != null) {
final long sum = MathUtils.sum(countBuffer);
final SimpleInterval location = targetCollection.location(index);
final int targetSize = location.size();
rowSummaryOutputWriter.println(
String.join(
COLUMN_SEPARATOR,
targetInfoString,
Long.toString(sum),
String.format(
AVERAGE_DOUBLE_FORMAT, sum / ((float) countColumns.columnCount() * targetSize))));
}
}
