private string LofteeAnalysis(IAnnotatedTranscript ta, IAnnotatedAlternateAllele allele,
ICompressedSequence sequence)
{
if (!LofteeUtilities.IsApplicable(ta))
{
return(null);
}
var filters = new HashSet <LofteeFilter.Filter>();
var flags = new HashSet <LofteeFilter.Flag>();
var transcript = _transcriptsById[ta.TranscriptID];
CheckSingleExon(transcript, flags);
if (LofteeUtilities.IsInExon(ta))
{
CheckEndTruncation(ta, transcript, filters);
CheckIncompleteCds();
CheckNonCanonicalSpliceSurr(ta, transcript, filters, sequence);
}
var intronIdx = LofteeUtilities.GetIntronIndex(ta, transcript);
if (LofteeUtilities.IsSpliceVariant(ta) && intronIdx != -1)
{
CheckSmallIntron(intronIdx, transcript, filters);
CheckNonCanonicalSplice(intronIdx, transcript, filters, sequence);
CheckNagnagSite(transcript, allele, flags, sequence);
}
return(GetLofteeString(filters, flags));
}
private static IAnnotatedTranscript GetAnnotatedTranscript(IVariant variant, ISequence compressedSequence,
ITranscript transcript, Status annotationStatus, IPredictionCache siftCache, IPredictionCache polyphenCache)
{
IAnnotatedTranscript annotatedTranscript = null;
// ReSharper disable once SwitchStatementMissingSomeCases
switch (annotationStatus)
{
case Status.FlankingAnnotation:
annotatedTranscript =
FlankingTranscriptAnnotator.GetAnnotatedTranscript(variant.End, transcript);
break;
case Status.ReducedAnnotation:
annotatedTranscript = ReducedTranscriptAnnotator.GetAnnotatedTranscript(transcript, variant);
break;
case Status.RohAnnotation:
annotatedTranscript = RohTranscriptAnnotator.GetAnnotatedTranscript(transcript);
break;
case Status.FullAnnotation:
var acidsProvider = variant.Chromosome.UcscName == "chrM"
? MitoAminoAcidsProvider
: AminoAcidsProvider;
annotatedTranscript = FullTranscriptAnnotator.GetAnnotatedTranscript(transcript, variant,
compressedSequence, siftCache, polyphenCache, acidsProvider);
break;
}
return(annotatedTranscript);
}
private void AddConservationScore(IAnnotatedTranscript annotatedTranscript)
{
if (_conservationProvider == null)
{
return;
}
if (annotatedTranscript.MappedPosition == null)
{
return;
}
var scores = new List <double>();
var start = annotatedTranscript.MappedPosition.ProteinStart;
var end = annotatedTranscript.MappedPosition.ProteinEnd;
if (start == -1 || end == -1)
{
return;
}
for (int aaPos = start; aaPos <= end; aaPos++)
{
var transcriptId = annotatedTranscript.Transcript.Id.WithVersion;
var score = _conservationProvider.GetConservationScore(transcriptId, aaPos);
if (score == -1)
{
return; //don't add conservation scores
}
scores.Add(1.0 * score / 100);
}
annotatedTranscript.ConservationScores = scores;
}
public static int GetIntronIndex(IAnnotatedTranscript ta, Transcript transcript)
{
if (ta.Introns == null)
{
return(-1);
}
int affectedIntron = Convert.ToInt32(ta.Introns.Split('/').First().Split('-').First());
var totalIntrons = transcript.Introns.Length;
return(transcript.Gene.OnReverseStrand ? totalIntrons - affectedIntron : affectedIntron - 1);
}
private void CheckEndTruncation(IAnnotatedTranscript ta, Transcript transcript, HashSet <LofteeFilter.Filter> filters)
{
if (!ta.Consequence.Contains("stop_gained") && !ta.Consequence.Contains("frameshift_variant"))
{
return;
}
var cdsPositions = ta.CdsPosition.Split('-');
var startCdPos = Convert.ToInt32(cdsPositions[0]);
var cdsLen = CodingSequence.GetCodingSequenceLength(transcript.CdnaMaps,
transcript.Translation.CodingRegion.GenomicStart, transcript.Translation.CodingRegion.GenomicEnd,
transcript.StartExonPhase);
if ((double)startCdPos / cdsLen > 0.95)
{
filters.Add(LofteeFilter.Filter.end_trunc);
}
}
private static bool IsFlankingTranscript(IAnnotatedTranscript transcript)
{
if (transcript.Consequences == null)
{
return(false);
}
// ReSharper disable once LoopCanBeConvertedToQuery
foreach (var consequence in transcript.Consequences)
{
if (consequence == ConsequenceTag.downstream_gene_variant ||
consequence == ConsequenceTag.upstream_gene_variant)
{
return(true);
}
}
return(false);
}
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);
}
}
public static bool IsInExon(IAnnotatedTranscript ta) => !string.IsNullOrEmpty(ta.Exons);
public static bool IsApplicable(IAnnotatedTranscript ta)
{
return(ta.Consequence.Contains("stop_gained") || ta.Consequence.Contains("frameshift_variant") ||
ta.Consequence.Contains("splice_donor_variant") || ta.Consequence.Contains("splice_acceptor_variant"));
}
private static bool IsSpliceAcceptor(IAnnotatedTranscript ta) => ta.Consequence.Contains("splice_acceptor_variant");
public static bool IsSpliceVariant(IAnnotatedTranscript ta) => IsSpliceDonor(ta) || IsSpliceAcceptor(ta);