mapTo = new ArrayList<>();
mapTo.add(new int[] { 1, transcriptLength });
MapList mapping = new MapList(mappedFrom, mapTo, 1, 1);
EnsemblCdna cdna = new EnsemblCdna(getDomain());
cdna.transferFeatures(gene.getFeatures().getPositionalFeatures(),
transcript.getDatasetSequence(), mapping, parentId);
/*
* fetch and save cross-references
*/
cdna.getCrossReferences(transcript);
/*
* and finally fetch the protein product and save as a cross-reference
*/
cdna.addProteinProduct(transcript);
return transcript;
}
/**
* Returns the 'transcript_id' property of the sequence feature (or null)
*
* @param feature
* @return
*/
protected String getTranscriptId(SequenceFeature feature)
{
return (String) feature.getValue("transcript_id");
}
/**
* Returns a list of the transcript features on the sequence whose Parent is
* the gene for the accession id.
*
* Transcript features are those of type "transcript", or any of its sub-types
* in the Sequence Ontology e.g. "mRNA", "processed_transcript". We also
* include "NMD_transcript_variant", because this type behaves like a
* transcript identifier in Ensembl, although strictly speaking it is not in
* the SO.
*
* @param accId
* @param geneSequence
* @return
*/
protected List getTranscriptFeatures(String accId,
SequenceI geneSequence)
{
List transcriptFeatures = new ArrayList<>();
String parentIdentifier = GENE_PREFIX + accId;
List sfs = geneSequence.getFeatures()
.getFeaturesByOntology(SequenceOntologyI.TRANSCRIPT);
sfs.addAll(geneSequence.getFeatures().getPositionalFeatures(
SequenceOntologyI.NMD_TRANSCRIPT_VARIANT));
for (SequenceFeature sf : sfs)
{
String parent = (String) sf.getValue(PARENT);
if (parentIdentifier.equalsIgnoreCase(parent))
{
transcriptFeatures.add(sf);
}
}
return transcriptFeatures;
}
@Override
public String getDescription()
{
return "Fetches all transcripts and variant features for a gene or transcript";
}
/**
* Default test query is a gene id (can also enter a transcript id)
*/
@Override
public String getTestQuery()
{
return "ENSG00000157764"; // BRAF, 5 transcripts, reverse strand
// ENSG00000090266 // NDUFB2, 15 transcripts, forward strand
// ENSG00000101812 // H2BFM histone, 3 transcripts, forward strand
// ENSG00000123569 // H2BFWT histone, 2 transcripts, reverse strand
}
/**
* Answers true for a feature of type 'gene' (or a sub-type of gene in the
* Sequence Ontology), whose ID is the accession we are retrieving
*/
@Override
protected boolean identifiesSequence(SequenceFeature sf, String accId)
{
if (SequenceOntologyFactory.getInstance().isA(sf.getType(),
SequenceOntologyI.GENE))
{
// NB features as gff use 'ID'; rest services return as 'id'
String id = (String) sf.getValue("ID");
if ((GENE_PREFIX + accId).equalsIgnoreCase(id))
{
return true;
}
}
return false;
}
/**
* Answers true unless feature type is 'gene', or 'transcript' with a parent
* which is a different gene. We need the gene features to identify the range,
* but it is redundant information on the gene sequence. Checking the parent
* allows us to drop transcript features which belong to different
* (overlapping) genes.
*/
@Override
protected boolean retainFeature(SequenceFeature sf, String accessionId)
{
SequenceOntologyI so = SequenceOntologyFactory.getInstance();
String type = sf.getType();
if (so.isA(type, SequenceOntologyI.GENE))
{
return false;
}
if (isTranscript(type))
{
String parent = (String) sf.getValue(PARENT);
if (!(GENE_PREFIX + accessionId).equalsIgnoreCase(parent))
{
return false;
}
}
return true;
}
/**
* Answers false. This allows an optimisation - a single 'gene' feature is all
* that is needed to identify the positions of the gene on the genomic
* sequence.
*/
@Override
protected boolean isSpliceable()
{
return false;
}
/**
* Override to do nothing as Ensembl doesn't return a protein sequence for a
* gene identifier
*/
@Override
protected void addProteinProduct(SequenceI querySeq)
{
}
@Override
public Regex getAccessionValidator()
{
return ACCESSION_REGEX;
}
/**
* Returns a descriptor for suitable feature display settings with
*
* - only exon or sequence_variant features (or their subtypes in the
* Sequence Ontology) visible
* - variant features coloured red
* - exon features coloured by label (exon name)
* - variants displayed above (on top of) exons
*
*/
@Override
public FeatureSettingsModelI getFeatureColourScheme()
{
return new FeatureSettingsAdapter()
{
SequenceOntologyI so = SequenceOntologyFactory.getInstance();
@Override
public boolean isFeatureDisplayed(String type)
{
return (so.isA(type, SequenceOntologyI.EXON)
|| so.isA(type, SequenceOntologyI.SEQUENCE_VARIANT));
}
@Override
public FeatureColourI getFeatureColour(String type)
{
if (so.isA(type, SequenceOntologyI.EXON))
{
return new FeatureColour()
{
@Override
public boolean isColourByLabel()
{
return true;
}
};
}
if (so.isA(type, SequenceOntologyI.SEQUENCE_VARIANT))
{
return new FeatureColour()
{
@Override
public Color getColour()
{
return Color.RED;
}
};
}
return null;
}
/**
* order to render sequence_variant after exon after the rest
*/
@Override
public int compare(String feature1, String feature2)
{
if (so.isA(feature1, SequenceOntologyI.SEQUENCE_VARIANT))
{
return +1;
}
if (so.isA(feature2, SequenceOntologyI.SEQUENCE_VARIANT))
{
return -1;
}
if (so.isA(feature1, SequenceOntologyI.EXON))
{
return +1;
}
if (so.isA(feature2, SequenceOntologyI.EXON))
{
return -1;
}
return 0;
}
};
}
}