public NsaIndex(Stream stream)
{
using (var memStream = new MemoryStream())
using (var memReader = new ExtendedBinaryReader(memStream))
{
stream.CopyTo(memStream);//reading all bytes in stream to memStream
memStream.Position = 0;
Assembly = (GenomeAssembly)memReader.ReadByte();
Version = DataSourceVersion.Read(memReader);
JsonKey = memReader.ReadAsciiString();
MatchByAllele = memReader.ReadBoolean();
IsArray = memReader.ReadBoolean();
SchemaVersion = memReader.ReadOptInt32();
IsPositional = memReader.ReadBoolean();
var chromCount = memReader.ReadOptInt32();
_chromBlocks = new Dictionary <ushort, List <NsaIndexBlock> >(chromCount);
for (var i = 0; i < chromCount; i++)
{
var chromIndex = memReader.ReadOptUInt16();
var chunkCount = memReader.ReadOptInt32();
_chromBlocks[chromIndex] = new List <NsaIndexBlock>(chunkCount);
for (var j = 0; j < chunkCount; j++)
{
_chromBlocks[chromIndex].Add(new NsaIndexBlock(memReader));
}
}
}
}
public ChunkedIndex(Stream stream)
{
//reading the index in one shot
var buffer = new byte[1048576];
var indexLength = stream.Read(buffer, 0, 1048576);
using (var memStream = new MemoryStream(buffer, 0, indexLength))
using (var memReader = new ExtendedBinaryReader(memStream))
{
Assembly = (GenomeAssembly)memReader.ReadByte();
Version = DataSourceVersion.Read(memReader);
JsonKey = memReader.ReadAsciiString();
MatchByAllele = memReader.ReadBoolean();
IsArray = memReader.ReadBoolean();
SchemaVersion = memReader.ReadOptInt32();
IsPositional = memReader.ReadBoolean();
var chromCount = memReader.ReadOptInt32();
_chromChunks = new Dictionary <ushort, List <Chunk> >(chromCount);
for (var i = 0; i < chromCount; i++)
{
var chromIndex = memReader.ReadOptUInt16();
var chunkCount = memReader.ReadOptInt32();
_chromChunks[chromIndex] = new List <Chunk>(chunkCount);
for (var j = 0; j < chunkCount; j++)
{
_chromChunks[chromIndex].Add(new Chunk(memReader));
}
}
}
}
public static IChromosome Read(ExtendedBinaryReader reader)
{
string ucscName = reader.ReadAsciiString();
string ensemblName = reader.ReadAsciiString();
string refseqAccession = reader.ReadAsciiString();
string genBankAccession = reader.ReadAsciiString();
int length = reader.ReadOptInt32();
ushort refIndex = reader.ReadOptUInt16();
return(new Chromosome(ucscName, ensemblName, refseqAccession, genBankAccession, length, refIndex));
}
public SuppInterval(ExtendedBinaryReader reader)
{
string ensemblName = reader.ReadAsciiString();
string ucscName = reader.ReadAsciiString();
ushort chromIndex = reader.ReadOptUInt16();
Chromosome = new Chromosome(ucscName, ensemblName, chromIndex);
Start = reader.ReadOptInt32();
End = reader.ReadOptInt32();
_jsonString = reader.ReadString();
}
public static ITranscriptRegion Read(ExtendedBinaryReader reader)
{
TranscriptRegionType type = (TranscriptRegionType)reader.ReadByte();
ushort id = reader.ReadOptUInt16();
int genomicStart = reader.ReadOptInt32();
int genomicEnd = reader.ReadOptInt32();
int cdnaStart = reader.ReadOptInt32();
int cdnaEnd = reader.ReadOptInt32();
return(new TranscriptRegion(type, id, genomicStart, genomicEnd, cdnaStart, cdnaEnd));
}
private static ushort GetObservedShort(ushort expectedShort)
{
ushort observedShort;
using (var memoryStream = GetMemoryStream(expectedShort))
using (var reader = new ExtendedBinaryReader(memoryStream))
{
observedShort = reader.ReadOptUInt16();
}
return(observedShort);
}
public static SuppInterval Read(ExtendedBinaryReader reader)
{
string ensemblName = reader.ReadAsciiString();
string ucscName = reader.ReadAsciiString();
ushort chromIndex = reader.ReadOptUInt16();
var chromosome = new Chromosome(ucscName, ensemblName, null, null, 1, chromIndex);
var start = reader.ReadOptInt32();
var end = reader.ReadOptInt32();
var jsonString = reader.ReadString();
return(new SuppInterval(chromosome, start, end, jsonString));
}
public static ITranscript Read(ExtendedBinaryReader reader,
IDictionary <ushort, IChromosome> chromosomeIndexDictionary, IGene[] cacheGenes,
ITranscriptRegion[] cacheTranscriptRegions, IInterval[] cacheMirnas, string[] cachePeptideSeqs)
{
// transcript
var referenceIndex = reader.ReadOptUInt16();
var start = reader.ReadOptInt32();
var end = reader.ReadOptInt32();
var id = CompactId.Read(reader);
// gene
var geneIndex = reader.ReadOptInt32();
var gene = cacheGenes[geneIndex];
// encoded data
var encoded = EncodedTranscriptData.Read(reader);
// transcript regions
var transcriptRegions = encoded.HasTranscriptRegions ? ReadIndices(reader, cacheTranscriptRegions) : null;
ushort numExons = reader.ReadOptUInt16();
// protein function predictions
int siftIndex = encoded.HasSift ? reader.ReadOptInt32() : -1;
int polyphenIndex = encoded.HasPolyPhen ? reader.ReadOptInt32() : -1;
// translation
var translation = encoded.HasTranslation ? DataStructures.Translation.Read(reader, cachePeptideSeqs) : null;
// attributes
var mirnas = encoded.HasMirnas ? ReadIndices(reader, cacheMirnas) : null;
var rnaEdits = encoded.HasRnaEdits ? ReadItems(reader, RnaEdit.Read) : null;
var selenocysteines = encoded.HasSelenocysteines ? ReadItems(reader, x => x.ReadOptInt32()) : null;
return(new Transcript(chromosomeIndexDictionary[referenceIndex], start, end, id, translation,
encoded.BioType, gene, ExonUtilities.GetTotalExonLength(transcriptRegions), encoded.StartExonPhase,
encoded.IsCanonical, transcriptRegions, numExons, mirnas, siftIndex, polyphenIndex,
encoded.TranscriptSource, encoded.CdsStartNotFound, encoded.CdsEndNotFound, selenocysteines, rnaEdits));
}
public NgaReader(Stream stream)
{
_nsaStream = stream;
// read the whole file. Currently they are well under 2MB
var compressedBytes = new byte[2 * 1024 * 1024];
var decompressedBytes = new byte[20 * 1024 * 1024];
var compressedSize = _nsaStream.Read(compressedBytes, 0, compressedBytes.Length);
var zstd = new Zstandard();
var decompressedSize = zstd.Decompress(compressedBytes, compressedSize, decompressedBytes, decompressedBytes.Length);
_memStream = new MemoryStream(decompressedBytes, 0, decompressedSize);
_reader = new ExtendedBinaryReader(_memStream);
Version = DataSourceVersion.Read(_reader);
JsonKey = _reader.ReadAsciiString();
_isArray = _reader.ReadBoolean();
ushort schemaVersion = _reader.ReadOptUInt16();
if (schemaVersion != SaCommon.SchemaVersion)
{
throw new UserErrorException($"Expected schema version: {SaCommon.SchemaVersion}, observed: {schemaVersion} for {JsonKey}");
}
}
