private static void WritePredictions(PredictionWriter writer, IReadOnlyList <MutableTranscript> transcripts,
Func <MutableTranscript, string> predictionFunc, IChromosome chromosome)
{
var predictionDict = new Dictionary <string, List <int> >(StringComparer.Ordinal);
for (var transcriptIndex = 0; transcriptIndex < transcripts.Count; transcriptIndex++)
{
var transcript = transcripts[transcriptIndex];
string predictionData = predictionFunc(transcript);
if (predictionData == null)
{
continue;
}
if (predictionDict.TryGetValue(predictionData, out var transcriptIdList))
{
transcriptIdList.Add(transcriptIndex);
}
else
{
predictionDict[predictionData] = new List <int> {
transcriptIndex
}
};
}
writer.Write(chromosome, predictionDict);
}
private static ExitCodes ProgramExecution()
{
var transcriptSource = GetSource(_transcriptSource);
var sequenceReader = new CompressedSequenceReader(FileUtilities.GetReadStream(_inputReferencePath));
var vepRootDirectory = new VepRootDirectory(sequenceReader.RefNameToChromosome);
var refIndexToVepDir = vepRootDirectory.GetRefIndexToVepDir(_inputVepDirectory);
var genomeAssembly = GenomeAssemblyHelper.Convert(_genomeAssembly);
long vepReleaseTicks = DateTime.Parse(_vepReleaseDate).Ticks;
var idToGenbank = GetIdToGenbank(genomeAssembly, transcriptSource);
// =========================
// create the pre-cache file
// =========================
// process each VEP directory
int numRefSeqs = sequenceReader.NumRefSeqs;
var header = new IntermediateIoHeader(_vepVersion, vepReleaseTicks, transcriptSource, genomeAssembly, numRefSeqs);
string siftPath = _outputStub + ".sift.gz";
string polyphenPath = _outputStub + ".polyphen.gz";
string transcriptPath = _outputStub + ".transcripts.gz";
string regulatoryPath = _outputStub + ".regulatory.gz";
using (var mergeLogger = new TranscriptMergerLogger(FileUtilities.GetCreateStream(_outputStub + ".merge_transcripts.log")))
using (var siftWriter = new PredictionWriter(GZipUtilities.GetStreamWriter(siftPath), header, IntermediateIoCommon.FileType.Sift))
using (var polyphenWriter = new PredictionWriter(GZipUtilities.GetStreamWriter(polyphenPath), header, IntermediateIoCommon.FileType.Polyphen))
using (var transcriptWriter = new MutableTranscriptWriter(GZipUtilities.GetStreamWriter(transcriptPath), header))
using (var regulatoryRegionWriter = new RegulatoryRegionWriter(GZipUtilities.GetStreamWriter(regulatoryPath), header))
{
var converter = new VepCacheParser(transcriptSource);
var emptyPredictionDict = new Dictionary <string, List <int> >();
for (ushort refIndex = 0; refIndex < numRefSeqs; refIndex++)
{
var chromosome = sequenceReader.RefIndexToChromosome[refIndex];
if (!refIndexToVepDir.TryGetValue(refIndex, out string vepSubDir))
{
siftWriter.Write(chromosome, emptyPredictionDict);
polyphenWriter.Write(chromosome, emptyPredictionDict);
continue;
}
Console.WriteLine("Parsing reference sequence [{0}]:", chromosome.UcscName);
var rawData = converter.ParseDumpDirectory(chromosome, vepSubDir);
var mergedTranscripts = TranscriptMerger.Merge(mergeLogger, rawData.Transcripts, idToGenbank);
var mergedRegulatoryRegions = RegulatoryRegionMerger.Merge(rawData.RegulatoryRegions);
int numRawTranscripts = rawData.Transcripts.Count;
int numMergedTranscripts = mergedTranscripts.Count;
Console.WriteLine($"- # merged transcripts: {numMergedTranscripts}, # total transcripts: {numRawTranscripts}");
WriteTranscripts(transcriptWriter, mergedTranscripts);
WriteRegulatoryRegions(regulatoryRegionWriter, mergedRegulatoryRegions);
WritePredictions(siftWriter, mergedTranscripts, x => x.SiftData, chromosome);
WritePredictions(polyphenWriter, mergedTranscripts, x => x.PolyphenData, chromosome);
}
}
Console.WriteLine("\n{0} directories processed.", refIndexToVepDir.Count);
return(ExitCodes.Success);
}
