public InterimTsvsMerger(IEnumerable <string> annotationFiles, IEnumerable <string> intervalFiles, string miscFile, IEnumerable <string> geneFiles, string compressedReference, string outputDirectory)
{
_outputDirectory = outputDirectory;
var refSequenceProvider = new ReferenceSequenceProvider(FileUtilities.GetReadStream(compressedReference));
_genomeAssembly = refSequenceProvider.GenomeAssembly;
_refNameToChromosome = refSequenceProvider.RefNameToChromosome;
_tsvReaders = ReaderUtilities.GetSaTsvReaders(annotationFiles);
_miscReader = ReaderUtilities.GetMiscTsvReader(miscFile);
_geneReaders = ReaderUtilities.GetGeneReaders(geneFiles);
_intervalReaders = ReaderUtilities.GetIntervalReaders(intervalFiles);
_saHeaders = new List <SaHeader>();
_saHeaders.AddRange(ReaderUtilities.GetTsvHeaders(_tsvReaders));
_saHeaders.AddRange(ReaderUtilities.GetTsvHeaders(_intervalReaders));
_geneHeaders = ReaderUtilities.GetTsvHeaders(_geneReaders)?.ToList();
_refNames = new HashSet <string>();
_refNames.UnionWith(ReaderUtilities.GetRefNames(_tsvReaders));
_refNames.UnionWith(ReaderUtilities.GetRefNames(_intervalReaders));
if (_miscReader != null)
{
_refNames.UnionWith(_miscReader.RefNames);
}
DisplayDataSources(_saHeaders, _geneHeaders);
MergeUtilities.CheckAssemblyConsistancy(_saHeaders);
}
private void MergeChrom(string refName)
{
var creationBench = new Benchmark();
var currentChrAnnotationCount = 0;
int refMinorCount;
var saEnumerators = GetSaEnumerators(refName);
//return;
var globalMajorAlleleInRefMinors = GetGlobalMajorAlleleForRefMinors(refName);
var dataSourceVersions = MergeUtilities.GetDataSourceVersions(_saHeaders);
var ucscRefName = _refNameToChromosome[refName].UcscName;
var header = new SupplementaryAnnotationHeader(ucscRefName, DateTime.Now.Ticks,
SaDataBaseCommon.DataVersion, dataSourceVersions, _genomeAssembly);
//we need a list because we will enumerate over it multiple times
var intervals = MergeUtilities.GetIntervals(_intervalReaders, refName).OrderBy(x => x.Start).ThenBy(x => x.End).ToList();
var svIntervals = MergeUtilities.GetSpecificIntervals(ReportFor.StructuralVariants, intervals);
var allVariantsIntervals = MergeUtilities.GetSpecificIntervals(ReportFor.AllVariants, intervals);
var smallVariantIntervals = MergeUtilities.GetSpecificIntervals(ReportFor.SmallVariants, intervals);
var saPath = Path.Combine(_outputDirectory, $"{ucscRefName}.nsa");
using (var stream = FileUtilities.GetCreateStream(saPath))
using (var idxStream = FileUtilities.GetCreateStream(saPath + ".idx"))
using (var blockSaWriter = new SaWriter(stream, idxStream, header, smallVariantIntervals, svIntervals, allVariantsIntervals, globalMajorAlleleInRefMinors))
{
int position;
ISaPosition saPosition;
(position, saPosition) = GetNextInterimPosition(saEnumerators);
while (saPosition != null)
{
blockSaWriter.Write(saPosition, position);
currentChrAnnotationCount++;
(position, saPosition) = GetNextInterimPosition(saEnumerators);
}
refMinorCount = blockSaWriter.RefMinorCount;
}
Console.WriteLine($"{ucscRefName,-23} {currentChrAnnotationCount,10:n0} {intervals.Count,6:n0} {refMinorCount,6:n0} {creationBench.GetElapsedIterationTime(currentChrAnnotationCount, "variants", out double _)}");
}
private static (int, ISaPosition) GetNextInterimPosition(List <IEnumerator <IInterimSaItem> > iSaEnumerators)
{
var minItems = MergeUtilities.GetMinItems(iSaEnumerators);
return(MergeUtilities.GetSaPosition(minItems));
}