private static MetadataListItem <List <string> > CDSFeatureMetadata(CDS cds, Exon exon)
{
string cdsAttributes = exon.GetGtfAttributes() + " protein_id \"" + (cds.Parent as Transcript).ProteinID + "\";";
var feature = new MetadataListItem <List <string> >(cds.FeatureType, cdsAttributes);
feature.SubItems["source"] = new List <string> {
cds.Source.ToString()
};
feature.SubItems["start"] = new List <string> {
cds.OneBasedStart.ToString()
};
feature.SubItems["end"] = new List <string> {
cds.OneBasedEnd.ToString()
};
if (cds.Strand != ".")
{
feature.SubItems["strand"] = new List <string> {
cds.Strand.ToString()
};
} // might take in features without strand later on
return(feature);
}
/// <summary>
/// Processes a feature from a GTF gene model file.
/// </summary>
/// <param name="feature"></param>
/// <param name="oneBasedStart"></param>
/// <param name="oneBasedEnd"></param>
/// <param name="chrom"></param>
/// <param name="attributes"></param>
public void ProcessGtfFeature(MetadataListItem <List <string> > feature, long oneBasedStart, long oneBasedEnd, Chromosome chrom, Dictionary <string, string> attributes)
{
bool hasGeneId = attributes.TryGetValue("gene_id", out string geneId);
bool hasTranscriptId = attributes.TryGetValue("transcript_id", out string transcriptId);
bool hasProteinId = attributes.TryGetValue("protein_id", out string proteinId);
bool hasExonId = attributes.TryGetValue("exon_id", out string exonId);
bool hasSource = feature.SubItems.TryGetValue("source", out List <string> sourceish);
bool hasStrand = feature.SubItems.TryGetValue("strand", out List <string> strandish);
bool hasFrame = feature.SubItems.TryGetValue("frame", out List <string> framey);
string source = hasSource ? sourceish[0] : "";
if (!hasStrand)
{
return;
} // strand is a required to do anything in this program
string strand = strandish[0];
int frame = 0;
if (hasFrame)
{
int.TryParse(framey[0], out frame);
}
// Trim prefixes from the IDs
string genePrefix = "gene:";
string transcriptPrefix = "transcript:";
if (hasGeneId && geneId.StartsWith(genePrefix))
{
string newGeneId = geneId.Substring(genePrefix.Length);
feature.FreeText.Replace(geneId, newGeneId);
geneId = newGeneId;
}
if (hasTranscriptId && transcriptId.StartsWith(transcriptPrefix))
{
string newTranscriptId = transcriptId.Substring(transcriptPrefix.Length);
feature.FreeText.Replace(transcriptId, newTranscriptId);
transcriptId = newTranscriptId;
}
if (hasProteinId && proteinId.StartsWith(transcriptPrefix))
{
proteinId = proteinId.Substring(transcriptPrefix.Length); // transcript id is used for protein id sometimes
}
// Catch the transcript features before they go by if available, i.e. if the file doesn't just have exons
if (feature.Key == "transcript" && (currentTranscript == null || hasTranscriptId && transcriptId != currentTranscript.ID))
{
if (currentGene == null || hasGeneId && geneId != currentGene.ID)
{
currentGene = new Gene(geneId, chrom, source, strand, oneBasedStart, oneBasedEnd, feature);
Genes.Add(currentGene);
GenomeForest.Add(currentGene);
}
currentTranscript = new Transcript(transcriptId, currentGene, source, strand, oneBasedStart, oneBasedEnd, null, null, feature);
currentGene.Transcripts.Add(currentTranscript);
GenomeForest.Add(currentTranscript);
}
if (feature.Key == "exon" || feature.Key == "CDS")
{
if (currentGene == null || hasGeneId && geneId != currentGene.ID)
{
currentGene = new Gene(geneId, chrom, source, strand, oneBasedStart, oneBasedEnd, feature);
Genes.Add(currentGene);
GenomeForest.Add(currentGene);
}
if (currentTranscript == null || hasTranscriptId && transcriptId != currentTranscript.ID)
{
if (currentTranscript != null)
{
Transcript.SetRegions(currentTranscript);
currentTranscript.FrameCorrection();
}
currentTranscript = new Transcript(transcriptId, currentGene, source, strand, oneBasedStart, oneBasedEnd, null, null, feature);
currentGene.Transcripts.Add(currentTranscript);
GenomeForest.Add(currentTranscript);
}
if (feature.Key == "exon")
{
ISequence exon_dna = chrom.Sequence.GetSubSequence(oneBasedStart - 1, oneBasedEnd - oneBasedStart + 1);
Exon exon = new Exon(currentTranscript, currentTranscript.IsStrandPlus() ? exon_dna : exon_dna.GetReverseComplementedSequence(),
source, oneBasedStart, oneBasedEnd, chrom.Sequence.ID, strand, null, feature);
if (exon.Length() > 0)
{
currentTranscript.Exons.Add(exon);
}
}
else if (feature.Key == "CDS")
{
CDS cds = new CDS(currentTranscript, chrom.Sequence.ID, source, strand, oneBasedStart, oneBasedEnd, null, frame);
if (hasProteinId)
{
currentTranscript.ProteinID = proteinId;
}
if (cds.Length() > 0)
{
currentTranscript.CodingDomainSequences.Add(cds);
}
}
else
{ // nothing to do
}
}
}
/// <summary>
/// Processes a feature from a GFF3 gene model file.
/// </summary>
/// <param name="feature"></param>
/// <param name="oneBasedStart"></param>
/// <param name="oneBasedEnd"></param>
/// <param name="chrom"></param>
/// <param name="attributes"></param>
public void ProcessGff3Feature(MetadataListItem <List <string> > feature, long oneBasedStart, long oneBasedEnd, Chromosome chrom, Dictionary <string, string> attributes)
{
bool hasGeneId = attributes.TryGetValue("gene_id", out string geneId);
bool hasTranscriptId = attributes.TryGetValue("transcript_id", out string transcriptId);
bool hasExonId = attributes.TryGetValue("exon_id", out string exonId);
bool hasProteinId = attributes.TryGetValue("protein_id", out string proteinId);
bool hasSource = feature.SubItems.TryGetValue("source", out List <string> sourceish); // false if empty ("." in GFF format)
bool hasStrand = feature.SubItems.TryGetValue("strand", out List <string> strandish); // false if empty ("." in GFF format)
bool hasFrame = feature.SubItems.TryGetValue("frame", out List <string> framey); // false if empty ("." in GFF format)
string source = hasSource ? sourceish[0] : "";
if (!hasStrand)
{
return;
} // strand is a required to do anything in this program
string strand = strandish[0];
int frame = 0;
if (hasFrame)
{
int.TryParse(framey[0], out frame);
}
if (hasGeneId && (currentGene == null || hasGeneId && geneId != currentGene.ID))
{
currentGene = new Gene(geneId, chrom, source, strand, oneBasedStart, oneBasedEnd, feature);
Genes.Add(currentGene);
GenomeForest.Add(currentGene);
}
if (hasTranscriptId && (currentTranscript == null || hasTranscriptId && transcriptId != currentTranscript.ID))
{
if (currentTranscript != null)
{
Transcript.SetRegions(currentTranscript);
currentTranscript.FrameCorrection();
}
currentTranscript = new Transcript(transcriptId, currentGene, source, strand, oneBasedStart, oneBasedEnd, null, null, feature);
currentGene.Transcripts.Add(currentTranscript);
GenomeForest.Add(currentTranscript);
}
if (hasExonId)
{
ISequence exon_dna = chrom.Sequence.GetSubSequence(oneBasedStart - 1, oneBasedEnd - oneBasedStart + 1);
Exon exon = new Exon(currentTranscript, currentTranscript.IsStrandPlus() ? exon_dna : exon_dna.GetReverseComplementedSequence(),
source, oneBasedStart, oneBasedEnd, chrom == null ? "" : chrom.ChromosomeID, strand, null, feature);
if (exon.Length() > 0)
{
currentTranscript.Exons.Add(exon);
}
}
else if (hasProteinId)
{
CDS cds = new CDS(currentTranscript, chrom.Sequence.ID, source, strand, oneBasedStart, oneBasedEnd, null, frame);
if (cds.Length() > 0)
{
currentTranscript.CodingDomainSequences.Add(cds);
currentTranscript.ProteinID = proteinId;
}
}
else // nothing to do
{
}
}
/// <summary>
/// Creates coding domains based on another annotated transcript
/// </summary>
/// <param name="withCDS"></param>
/// <returns>true if this transcript was annotated; false if the transcript with CDS did not lead to an annotation</returns>
public bool CreateCDSFromAnnotatedStartCodons(Transcript withCDS)
{
// Nothing to do if null input
if (withCDS == null)
{
return(false);
}
// Figure out the start position
CDS firstCds = withCDS.CdsSortedStrand.First();
long cdsStartInChrom = IsStrandPlus() ? firstCds.OneBasedStart : firstCds.OneBasedEnd;
long cdsStartInMrna = BaseNumber2MRnaPos(cdsStartInChrom);
if (cdsStartInMrna < 0)
{
return(false);
} // the coding start wasn't within any of the exons of this transcript
// Figure out the stop codon from translation
ISequence spliced = SplicedRNA();
ISequence translateThis = spliced.GetSubSequence(cdsStartInMrna, spliced.Count - cdsStartInMrna);
ISequence proteinSequence = Translation.OneFrameTranslation(translateThis, Gene.Chromosome.Mitochondrial);
int stopIdx = proteinSequence.Select(x => x).ToList().IndexOf(Alphabets.Protein.Ter);
if (stopIdx < 0)
{
return(false);
} // no stop codon in sight
long endInMrna = cdsStartInMrna + (stopIdx + 1) * GeneModel.CODON_SIZE - 1; // include the stop codon in CDS
long lengthInMrna = endInMrna - cdsStartInMrna + 1;
// Figure out the stop index on the chromosome
long utr5ishstart = IsStrandPlus() ? Exons.Min(x => x.OneBasedStart) : cdsStartInChrom + 1;
long utr5ishend = IsStrandPlus() ? cdsStartInChrom - 1 : Exons.Max(x => x.OneBasedEnd);
Interval utr5ish = new Interval(null, "", Source, Strand, utr5ishstart, utr5ishend);
var intervals = SortedStrand(Exons.SelectMany(x => x.Minus(utr5ish)).ToList());
long lengthSoFar = 0;
foreach (Interval y in intervals)
{
long lengthSum = lengthSoFar + y.Length();
if (lengthSum <= lengthInMrna) // add this whole interval
{
var toAdd = new CDS(this, ChromosomeID, Source, Strand, y.OneBasedStart, y.OneBasedEnd, 0);
CodingDomainSequences.Add(toAdd);
lengthSoFar += toAdd.Length();
}
else if (lengthSoFar < lengthInMrna) // chop off part of this interval
{
long chopLength = lengthSum - lengthInMrna;
long start = IsStrandPlus() ?
y.OneBasedStart :
y.OneBasedStart + chopLength;
long end = IsStrandPlus() ?
y.OneBasedEnd - chopLength :
y.OneBasedEnd;
var toAdd = new CDS(this, ChromosomeID, Source, Strand, start, end, 0);
CodingDomainSequences.Add(toAdd);
lengthSoFar += toAdd.Length();
}
}
SetRegions(this);
return(true);
}
public CDS(CDS cds)
: base(cds)
{
}