public void Write(Dictionary <string, List <ISuppGeneItem> > geneToEntries)
{
using (var memStream = new MemoryStream())
using (var memWriter = new ExtendedBinaryWriter(memStream))
using (var writer = new BinaryWriter(_nsaStream))
{
_version.Write(memWriter);
memWriter.WriteOptAscii(_jsonKey);
memWriter.Write(_isArray);
memWriter.WriteOpt(_schemaVersion);
memWriter.WriteOpt(geneToEntries.Count);
foreach ((string geneSymbol, var entries) in geneToEntries)
{
memWriter.WriteOptAscii(geneSymbol);
memWriter.WriteOpt(entries.Count);
foreach (ISuppGeneItem geneItem in entries)
{
memWriter.Write(geneItem.GetJsonString());
}
}
var uncompressedBytes = memStream.ToArray();
var compressedBytes = new byte[uncompressedBytes.Length + 32];
var compressor = new Zstandard();
var compressSize = compressor.Compress(uncompressedBytes, uncompressedBytes.Length, compressedBytes,
compressedBytes.Length);
writer.Write(compressedBytes, 0, compressSize);
Console.WriteLine("Number of gene entries written:" + geneToEntries.Count);
}
}
private void WriteHeader()
{
_writer.Write(OmimDatabaseCommon.DataHeader);
_writer.Write(OmimDatabaseCommon.SchemaVersion);
_writer.Write(DateTime.UtcNow.Ticks);
_version.Write(_writer);
_writer.Write(OmimDatabaseCommon.GuardInt);
}
private void WriteHeader()
{
_writer.Write(CustomIntervalCommon.DataHeader);
_writer.Write(CustomIntervalCommon.SchemaVersion);
_writer.Write(DateTime.UtcNow.Ticks);
_writer.Write(_referenceName);
_writer.Write(_intervalType);
_version.Write(_writer);
// marking end of header with guard int
_writer.Write(CustomIntervalCommon.GuardInt);
}
private void WriteHeader(DataSourceVersion version, GenomeAssembly assembly, string jsonKey, ReportFor reportFor, int schemaVersion)
{
using (var writer = new ExtendedBinaryWriter(_stream, Encoding.UTF8, true))
{
writer.WriteOptAscii(SaCommon.NsiIdentifier);
version.Write(writer);
writer.Write((byte)assembly);
writer.WriteOptAscii(jsonKey);
writer.Write((byte)reportFor);
writer.Write(schemaVersion);
writer.Write(SaCommon.GuardInt);
}
}
public NsiWriter(BinaryWriter writer, DataSourceVersion version,
GenomeAssembly assembly, string jsonKey, ReportFor reportFor, int schemaVersion)
{
_writer = writer;
_memStream = new MemoryStream();
_memWriter = new ExtendedBinaryWriter(_memStream);
version.Write(_memWriter);
_memWriter.Write((byte)assembly);
_memWriter.WriteOptAscii(jsonKey);
_memWriter.Write((byte)reportFor);
_memWriter.WriteOpt(schemaVersion);
}
public ChunkedIndex(ExtendedBinaryWriter indexWriter, GenomeAssembly assembly, DataSourceVersion version, string jsonKey, bool matchByAllele, bool isArray, int schemaVersion, bool isPositional)
{
_writer = indexWriter;
MatchByAllele = matchByAllele;
JsonKey = jsonKey;
Version = version;
Assembly = assembly;
IsArray = isArray;
IsPositional = isPositional;
indexWriter.Write((byte)assembly);
version.Write(indexWriter);
indexWriter.WriteOptAscii(jsonKey);
indexWriter.Write(matchByAllele);
indexWriter.Write(isArray);
indexWriter.WriteOpt(schemaVersion);
indexWriter.Write(isPositional);
_chromChunks = new Dictionary <ushort, List <Chunk> >();
}
/// <summary>
/// Writeout the header of the nirvana phylop database
/// </summary>
private void WriteHeader()
{
_writer.Write(PhylopCommon.Header);
_writer.Write(PhylopCommon.SchemaVersion);
_writer.Write(PhylopCommon.DataVersion);
_writer.Write((byte)_genomeAssembly);
if (_version == null)
{
throw new MissingFieldException("Phylop data version cannot be null");
}
_version.Write(_writer);
_writer.Write(_refSeqName);
// space holder for chromosome interval list position
_intervalListOffset = _writer.BaseStream.Position;
_writer.Write(_intervalListPosition); // this is just a temp value. We will come back and write the real one before closing
_writer.Write(CacheConstants.GuardInt);
}
public void Write(IEnumerable <ProteinConservationItem> items)
{
if (items == null)
{
return;
}
_writer.WriteOpt(ProteinConservationCommon.SchemaVersion);
_writer.Write((byte)_assembly);
_version.Write(_writer);
var alignedProteinsAndScores = GetProteinWithUniqueScores(items);
var nirvanaProteins = new HashSet <string>(_transcriptCacheData.PeptideSeqs);
CheckProteinSetOverlap(alignedProteinsAndScores, nirvanaProteins);
var transcriptScores = new Dictionary <string, byte[]>();
//protein sequence -> transcript ids mapping
var transcriptGroupsByProtein = new Dictionary <string, List <string> >(alignedProteinsAndScores.Count);
foreach (var protein in alignedProteinsAndScores.Keys)
{
transcriptGroupsByProtein.Add(protein, new List <string>());
}
foreach (var transcriptIntervalArray in _transcriptCacheData.TranscriptIntervalArrays)
{
if (transcriptIntervalArray == null)
{
continue; //may happen since for GRCh38 decoy contigs, there may be none
}
foreach (var transcriptInterval in transcriptIntervalArray.Array)
{
var transcript = transcriptInterval.Value;
if (transcript.Translation == null)
{
continue;
}
var peptideSeq = transcript.Translation.PeptideSeq;
if (!alignedProteinsAndScores.TryGetValue(transcript.Translation.PeptideSeq, out var scores))
{
continue;
}
transcriptScores.TryAdd(transcript.Id.WithVersion, scores);
transcriptGroupsByProtein[peptideSeq].Add(transcript.Id.WithVersion);
}
}
foreach (var(transcriptId, scores) in transcriptScores)
{
var transcriptScore = new TranscriptConservationScores(transcriptId, scores);
transcriptScore.Write(_writer);
}
WriteTranscriptGroups(transcriptGroupsByProtein);
Console.WriteLine($"Recorded conservation scores for {transcriptScores.Count} transcripts.");
//writing an empty item to indicate end of records
var endOfRecordItem = TranscriptConservationScores.GetEmptyItem();
endOfRecordItem.Write(_writer);
}