public MustGenotypeExtractor(string compressedSeqPath, string oneKGenomeVcf, string clinvarVcf, string cosmicVcf, bool isHg19 = false)
{
_compressedSequence = new CompressedSequence();
_dataFileManager = new DataFileManager(new CompressedSequenceReader(FileUtilities.GetReadStream(compressedSeqPath), _compressedSequence), _compressedSequence);
_assembly = _compressedSequence.GenomeAssembly == GenomeAssembly.GRCh37 && isHg19? GenomeAssembly.hg19:_compressedSequence.GenomeAssembly;
if (_assembly == GenomeAssembly.Unknown)
{
throw new Exception("Genome assembly must be either GRCh37 or GRCh38");
}
if (_compressedSequence.GenomeAssembly == GenomeAssembly.GRCh38 && isHg19)
{
throw new Exception("reference sequence is GRCh38 while generating hg19 files");
}
_oneKGenomeReader = string.IsNullOrEmpty(oneKGenomeVcf)? null: GZipUtilities.GetAppropriateStreamReader(oneKGenomeVcf);
_clinvarReader = string.IsNullOrEmpty(clinvarVcf) ? null : GZipUtilities.GetAppropriateStreamReader(clinvarVcf);
_cosmicReader = string.IsNullOrEmpty(cosmicVcf) ? null : GZipUtilities.GetAppropriateStreamReader(cosmicVcf);
}
/// <summary>
/// Retrieves all Exon sequences and concats them together.
/// This includes 5' UTR + cDNA + 3' UTR [Transcript.pm:862 spliced_seq]
/// </summary>
private static string GetSplicedSequence(ICompressedSequence compressedSequence, CdnaCoordinateMap[] cdnaMaps, bool onReverseStrand)
{
var sb = new StringBuilder();
foreach (var exon in cdnaMaps)
{
var exonLength = exon.GenomicEnd - exon.GenomicStart + 1;
// sanity check: handle the situation where no reference has been provided
if (compressedSequence == null)
{
sb.Append(new string('N', exonLength));
continue;
}
sb.Append(compressedSequence.Substring(exon.GenomicStart - 1, exonLength));
}
return(onReverseStrand ? SequenceUtilities.GetReverseComplement(sb.ToString()) : sb.ToString());
}
public void AnnotateVariant(IVariantFeature variant, List <Transcript> transcripts,
IAnnotatedVariant annotatedVariant, ICompressedSequence sequence)
{
if (variant.IsStructuralVariant)
{
return;
}
CreateTranscriptDictionary(transcripts);
foreach (var altAllele in annotatedVariant.AnnotatedAlternateAlleles)
{
foreach (var transcript in altAllele.EnsemblTranscripts)
{
var lofteeOut = LofteeAnalysis(transcript, altAllele, sequence);
if (lofteeOut != null)
{
if (transcript.AdditionalInfo == null)
{
transcript.AdditionalInfo = new Dictionary <string, string>();
}
transcript.AdditionalInfo["loftee"] = lofteeOut;
}
}
foreach (var transcript in altAllele.RefSeqTranscripts)
{
var lofteeOut = LofteeAnalysis(transcript, altAllele, sequence);
if (lofteeOut != null)
{
if (transcript.AdditionalInfo == null)
{
transcript.AdditionalInfo = new Dictionary <string, string>();
}
transcript.AdditionalInfo["loftee"] = lofteeOut;
}
}
}
}
private void CheckNagnagSite(Transcript transcript, IAnnotatedAlternateAllele allele,
HashSet <LofteeFilter.Flag> flags, ICompressedSequence sequence)
{
if (allele.ReferenceBegin == null || allele.ReferenceEnd == null ||
allele.ReferenceBegin.Value != allele.ReferenceEnd.Value)
{
return;
}
int pos = allele.ReferenceBegin.Value;
string upStreamSeq = sequence.Substring(pos - 6, 6);
string downStreamSeq = sequence.Substring(pos, 5);
var combineSeq = transcript.Gene.OnReverseStrand
? SequenceUtilities.GetReverseComplement(upStreamSeq + downStreamSeq)
: upStreamSeq + downStreamSeq;
if (Regex.Match(combineSeq, "[A|T|C|G]AG[A|T|C|G]AG").Success)
{
flags.Add(LofteeFilter.Flag.nagnag_site);
}
}
/// <summary>
/// constructor
/// </summary>
public VariantAligner(ICompressedSequence compressedSequence)
{
_compressedSequence = compressedSequence;
}
/// <summary>
/// constructor
/// </summary>
public ClinVarXmlReaderTests(ChromosomeRenamerFixture fixture)
{
_sequence = fixture.Sequence;
_reader = fixture.Reader;
}
/// <summary>
/// constructor
/// </summary>
public CacheData(ICompressedSequence compressedSequence, Transcript transcript)
{
_compressedSequence = compressedSequence;
_transcript = transcript;
}
private void GetProteinPosition(TranscriptAnnotation ta, Transcript transcript, ICompressedSequence compressedSequence)
{
const int shift = 0;
if (ta.HasValidCdsStart)
{
ta.ProteinBegin = (int)((ta.CodingDnaSequenceBegin + shift + 2.0) / 3.0);
}
if (ta.HasValidCdsEnd)
{
ta.ProteinEnd = (int)((ta.CodingDnaSequenceEnd + shift + 2.0) / 3.0);
}
// assign our codons and amino acids
Codons.AssignExtended(ta, transcript, compressedSequence);
_aminoAcids.Assign(ta);
}
private void GetCodingAnnotations(Transcript transcript, TranscriptAnnotation ta, ICompressedSequence compressedSequence)
{
// coding annotations
if (!ta.HasValidCdnaStart && !ta.HasValidCdnaEnd)
{
return;
}
CalculateCdsPositions(transcript, ta);
// determine the protein position
if (!ta.HasValidCdsStart && !ta.HasValidCdsEnd)
{
return;
}
GetProteinPosition(ta, transcript, compressedSequence);
}
private void CheckNonCanonicalSpliceSurr(IAnnotatedTranscript ta, Transcript transcript,
HashSet <LofteeFilter.Filter> filters, ICompressedSequence sequence)
{
if (ta.Exons == null)
{
return;
}
int affectedExonIndex = Convert.ToInt32(ta.Exons.Split('/').First().Split('-').First());
var totalExons = transcript.CdnaMaps.Length;
string surrDonor = null;
string surrAcceptor = null;
if (totalExons <= 1)
{
return;
}
var onReverseStrand = transcript.Gene.OnReverseStrand;
if (affectedExonIndex > 1)
{
var intron = onReverseStrand ? transcript.Introns[totalExons - affectedExonIndex] : transcript.Introns[affectedExonIndex - 2];
int acceptorStart = onReverseStrand ? intron.Start : intron.End - 1;
var acceptorSeq = sequence.Substring(acceptorStart - 1, 2);
surrAcceptor = onReverseStrand ? SequenceUtilities.GetReverseComplement(acceptorSeq) : acceptorSeq;
}
if (affectedExonIndex < totalExons)
{
var intron = onReverseStrand ? transcript.Introns[totalExons - affectedExonIndex - 1] : transcript.Introns[affectedExonIndex - 1];
int donorStart = onReverseStrand ? intron.End - 1 : intron.Start;
var donorSeq = sequence.Substring(donorStart - 1, 2);
surrDonor = onReverseStrand ? SequenceUtilities.GetReverseComplement(donorSeq) : donorSeq;
}
if (surrAcceptor != null && surrAcceptor != "AG" || surrDonor != null && surrDonor != "GT")
{
filters.Add(LofteeFilter.Filter.non_can_splice_surr);
}
}
/// <summary>
/// assigns the reference and alternate codons [TranscriptVariationAllele.pm:259 codon]
/// if frameshift variants, add 45 basepair for both ref and alt codon
/// added 45 base pair for stop loss variant
/// </summary>
public static void AssignExtended(TranscriptAnnotation ta, Transcript transcript, ICompressedSequence compressedSequence)
{
// sanity check: make sure this is a coding region
if (!(ta.HasValidCdsEnd && ta.HasValidCdsStart))
{
ta.ReferenceCodon = null;
ta.AlternateCodon = null;
return;
}
// calculate necessary coordinates and lengths
var aminoAcidStart = ta.ProteinBegin * 3 - 2;
var aminoAcidEnd = ta.ProteinEnd * 3;
var prefixLen = ta.CodingDnaSequenceBegin - aminoAcidStart;
var suffixLen = aminoAcidEnd - ta.CodingDnaSequenceEnd;
var codingSequence = new CodingSequence(compressedSequence, transcript.Translation.CodingRegion.GenomicStart,
transcript.Translation.CodingRegion.GenomicEnd, transcript.CdnaMaps, transcript.Gene.OnReverseStrand,
transcript.StartExonPhase);
var aminoAcidSeq = codingSequence.Sequence();
var start1 = aminoAcidStart - 1;
var start2 = aminoAcidEnd - suffixLen;
var maxSuffixLen = aminoAcidSeq.Length - start2;
var atTailEnd = false;
if (suffixLen > maxSuffixLen)
{
suffixLen = maxSuffixLen;
atTailEnd = true;
}
if (start1 < 0)
{
start1 = 0;
}
if (start2 < 0)
{
start2 = 0;
}
if (prefixLen < 0)
{
prefixLen = 0;
}
var prefix = start1 + prefixLen < aminoAcidSeq.Length
? aminoAcidSeq.Substring(start1, prefixLen).ToLower()
: "AAA";
var suffix = suffixLen > 0
? aminoAcidSeq.Substring(start2, suffixLen).ToLower()
: "";
var needExtend = !atTailEnd && !IsTriplet(prefixLen + suffixLen + ta.TranscriptAlternateAllele.Length);
var extendedLen = maxSuffixLen - suffixLen > 45 ? 45 : (maxSuffixLen - suffixLen) / 3 * 3;
if (needExtend)
{
suffix = aminoAcidSeq.Substring(start2, suffixLen + extendedLen);
}
ta.HasFrameShift = false;
ta.ReferenceCodon = GetCodon(ta.TranscriptReferenceAllele, prefix, suffix, ref ta.HasFrameShift, atTailEnd);
ta.AlternateCodon = GetCodon(ta.TranscriptAlternateAllele, prefix, suffix, ref ta.HasFrameShift, atTailEnd);
}
// constructor
public CreateSupplementaryDatabase(
string compressedReferencePath,
string nsdBaseFileName,
string dbSnpFileName = null,
string cosmicVcfFile = null,
string cosmicTsvFile = null,
string clinVarFileName = null,
string oneKGenomeAfFileName = null,
string evsFileName = null,
string exacFileName = null,
List <string> customFiles = null,
string dgvFileName = null,
string oneKSvFileName = null,
string clinGenFileName = null,
string chrWhiteList = null)
{
_nsdBaseFileName = nsdBaseFileName;
_dataSources = new List <DataSourceVersion>();
_iSupplementaryDataItemList = new List <IEnumerator <SupplementaryDataItem> >();
_supplementaryIntervalList = new List <SupplementaryInterval>();
Console.WriteLine("Creating supplementary annotation files... Data version: {0}, schema version: {1}", SupplementaryAnnotationCommon.DataVersion, SupplementaryAnnotationCommon.SchemaVersion);
_compressedSequence = new CompressedSequence();
var compressedSequenceReader = new CompressedSequenceReader(FileUtilities.GetReadStream(compressedReferencePath), _compressedSequence);
_renamer = _compressedSequence.Renamer;
_dataFileManager = new DataFileManager(compressedSequenceReader, _compressedSequence);
if (!string.IsNullOrEmpty(chrWhiteList))
{
Console.WriteLine("Creating SA for the following chromosomes only:");
foreach (var refSeq in chrWhiteList.Split(','))
{
InputFileParserUtilities.ChromosomeWhiteList.Add(_renamer.GetEnsemblReferenceName(refSeq));
Console.Write(refSeq + ",");
}
Console.WriteLine();
}
else
{
InputFileParserUtilities.ChromosomeWhiteList = null;
}
if (dbSnpFileName != null)
{
AddSourceVersion(dbSnpFileName);
var dbSnpReader = new DbSnpReader(new FileInfo(dbSnpFileName), _renamer);
var dbSnpEnumerator = dbSnpReader.GetEnumerator();
_iSupplementaryDataItemList.Add(dbSnpEnumerator);
}
if (cosmicVcfFile != null && cosmicTsvFile != null)
{
AddSourceVersion(cosmicVcfFile);
var cosmicReader = new MergedCosmicReader(cosmicVcfFile, cosmicTsvFile, _renamer);
var cosmicEnumerator = cosmicReader.GetEnumerator();
_iSupplementaryDataItemList.Add(cosmicEnumerator);
}
if (oneKGenomeAfFileName != null)
{
AddSourceVersion(oneKGenomeAfFileName);
var oneKGenReader = new OneKGenReader(new FileInfo(oneKGenomeAfFileName), _renamer);
var oneKGenEnumerator = oneKGenReader.GetEnumerator();
_iSupplementaryDataItemList.Add(oneKGenEnumerator);
}
if (oneKSvFileName != null)
{
if (oneKGenomeAfFileName == null)
{
AddSourceVersion(oneKSvFileName);
}
var oneKGenSvReader = new OneKGenSvReader(new FileInfo(oneKSvFileName), _renamer);
var oneKGenSvEnumerator = oneKGenSvReader.GetEnumerator();
_iSupplementaryDataItemList.Add(oneKGenSvEnumerator);
}
if (evsFileName != null)
{
AddSourceVersion(evsFileName);
var evsReader = new EvsReader(new FileInfo(evsFileName), _renamer);
var evsEnumerator = evsReader.GetEnumerator();
_iSupplementaryDataItemList.Add(evsEnumerator);
}
if (exacFileName != null)
{
AddSourceVersion(exacFileName);
var exacReader = new ExacReader(new FileInfo(exacFileName), _renamer);
var exacEnumerator = exacReader.GetEnumerator();
_iSupplementaryDataItemList.Add(exacEnumerator);
}
if (clinVarFileName != null)
{
AddSourceVersion(clinVarFileName);
var clinVarReader = new ClinVarXmlReader(new FileInfo(clinVarFileName), compressedSequenceReader, _compressedSequence);
var clinVarList = clinVarReader.ToList();
clinVarList.Sort();
Console.WriteLine($"{clinVarList.Count} clinvar items read form XML file");
IEnumerator <ClinVarItem> clinVarEnumerator = clinVarList.GetEnumerator();
_iSupplementaryDataItemList.Add(clinVarEnumerator);
}
if (dgvFileName != null)
{
AddSourceVersion(dgvFileName);
var dgvReader = new DgvReader(new FileInfo(dgvFileName), _renamer);
var dgvEnumerator = dgvReader.GetEnumerator();
_iSupplementaryDataItemList.Add(dgvEnumerator);
}
if (clinGenFileName != null)
{
AddSourceVersion(clinGenFileName);
var clinGenReader = new ClinGenReader(new FileInfo(clinGenFileName), _renamer);
var clinGenEnumerator = clinGenReader.GetEnumerator();
_iSupplementaryDataItemList.Add(clinGenEnumerator);
}
if (customFiles != null)
{
foreach (var customFile in customFiles)
{
AddSourceVersion(customFile);
var customReader = new CustomAnnotationReader(new FileInfo(customFile), _renamer);
var customEnumerator = customReader.GetEnumerator();
_iSupplementaryDataItemList.Add(customEnumerator);
}
}
// initializing the IEnumerators in the list
foreach (var iDataEnumerator in _iSupplementaryDataItemList)
{
if (!iDataEnumerator.MoveNext())
{
_iSupplementaryDataItemList.Remove(iDataEnumerator);
}
}
_additionalItemsList = new List <SupplementaryDataItem>();
}
/// <summary>
/// constructor
/// </summary>
public DataFileManager(CompressedSequenceReader reader, ICompressedSequence compressedSequence)
{
_compressedSequence = compressedSequence;
_compressedSequenceReader = reader;
}
/// <summary>
/// returns the correct start value when retrieving a substring of a substring
/// where the top level might be reverse complemented
/// </summary>
public static string GetSubSubstring(int seqStart, int seqEnd, bool seqOnReverseStrand, int subStart, int subEnd, ICompressedSequence cs)
{
var start = seqOnReverseStrand ? seqEnd - subEnd : seqStart + subStart;
var precedingBases = cs.Substring(start - 1, subEnd - subStart + 1);
if (seqOnReverseStrand)
{
precedingBases = GetReverseComplement(precedingBases);
}
return(precedingBases);
}
