/// <summary>
/// Creates UTRs for transcripts and intergenic regions after reading gene model
/// </summary>
public void CreateIntergenicRegions()
{
foreach (IntervalTree it in GenomeForest.Forest.Values)
{
Gene previousPositiveStrandGene = null;
Gene previousNegativeStrandGene = null;
// Create intergenic regions on each strand
foreach (Gene gene in it.Intervals.OfType <Gene>().OrderBy(g => g.OneBasedStart))
{
Intergenic intergenic = null;
Gene previous = gene.IsStrandPlus() ? previousPositiveStrandGene : previousNegativeStrandGene;
if (previous != null)
{
// if there's a previous gene, create the intergenic region
intergenic = new Intergenic(gene.Chromosome, gene.ChromosomeID, gene.Source, gene.Strand, previous.OneBasedEnd + 1, gene.OneBasedStart - 1, null);
}
// store previous genes on each strand
if (gene.IsStrandPlus())
{
previousPositiveStrandGene = gene;
}
if (gene.IsStrandMinus())
{
previousNegativeStrandGene = gene;
}
// add the intergenic region to the genome forest if it was created
if (intergenic != null && intergenic.Length() > 0)
{
GenomeForest.Add(intergenic);
}
}
}
}
/// <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>
/// 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
}
}
}
public string ToString(bool useSeqOntology, bool useHgvs)
{
// Get data to show
string geneId = "", geneName = "", transcriptId = "", exonId = "", customId = "";
string bioType = null;
int exonRank = -1;
if (Marker != null)
{
// Gene Id, name and biotype
Gene gene = GetGene();
Transcript tr = GetTranscript();
// CDS size info
if (gene != null)
{
geneId = gene.ID;
//geneName = gene.getGeneName();
//bioType = (getBiotype() == null ? "" : getBiotype().ToString());
}
// Update trId
if (tr != null)
{
transcriptId = tr.ID;
}
// Exon rank information
Exon exon = GetExon();
//if (exon != null)
//{
// exonId = exon.ID;
// exonRank = exon.getRank();
//}
// Regulation
//if (isRegulation()) bioType = ((Regulation)marker).getRegulationType();
}
// Add seqChage's ID
//if (!variant.getId() == "") customId += variant.getId();
// Add custom markers
//if (marker != null && marker is Custom) customId += (customId == "" ? "" : ";") + marker.getId();
// CDS length
long cdsSize = GetCdsLength();
string error = String.Join(",", Error);
string warning = String.Join(",", Warning);
string errWarn = error + (error == "" ? "" : "|") + warning;
//string aaChange = "";
//if (useHgvs) aaChange = getHgvs();
//else aaChange = ((aaRef.Length + aaAlt.Length) > 0 ? aaRef + "/" + aaAlt : "");
string aaChange = ((ReferenceAA.Length + AlternateAA.Length) > 0 ? ReferenceAA + "/" + AlternateAA : "");
return(errWarn //
+ "\t" + geneId //
+ "\t" + geneName //
+ "\t" + bioType //
+ "\t" + transcriptId //
+ "\t" + exonId //
+ "\t" + (exonRank >= 0 ? exonRank.ToString() : "") //
+ "\t" + Effect(false, false, false, useSeqOntology, false) //
+ "\t" + aaChange //
+ "\t" + ((CodonsRef.Length + CodonsAlt.Length) > 0 ? CodonsRef + "/" + CodonsAlt : "") //
+ "\t" + (CodonNum >= 0 ? (CodonNum + 1).ToString() : "") //
//+ "\t" + (codonDegeneracy >= 0 ? codonDegeneracy + "" : "") //
+ "\t" + (cdsSize >= 0 ? cdsSize.ToString() : "") //
+ "\t" + (CodonsAroundOld.Length > 0 ? CodonsAroundOld + " / " + CodonsAroundNew : "") //
+ "\t" + (AroundOldAAs.Length > 0 ? AroundOldAAs + " / " + AroundNewAAs : "") //
+ "\t" + customId //
);
}