return EnsemblSeqType.GENOMIC;
}
+ @Override
+ protected String getObjectType()
+ {
+ return OBJECT_TYPE_GENE;
+ }
+
/**
* Returns an alignment containing the gene(s) for the given gene or
* transcript identifier, or external identifier (e.g. Uniprot id). If given a
* <li>resolves an external identifier by looking up xref-ed gene ids</li>
* <li>fetches the gene sequence</li>
* <li>fetches features on the sequence</li>
- * <li>identifies "transcript" features whose Parent is the requested gene</li>
+ * <li>identifies "transcript" features whose Parent is the requested
+ * gene</li>
* <li>fetches the transcript sequence for each transcript</li>
* <li>makes a mapping from the gene to each transcript</li>
* <li>copies features from gene to transcript sequences</li>
}
if (geneAlignment.getHeight() == 1)
{
+ // ensure id has 'correct' case for the Ensembl identifier
+ geneId = geneAlignment.getSequenceAt(0).getName();
getTranscripts(geneAlignment, geneId);
}
if (al == null)
}
/**
- * Converts a query, which may contain one or more gene or transcript
- * identifiers, into a non-redundant list of gene identifiers.
+ * Converts a query, which may contain one or more gene, transcript, or
+ * external (to Ensembl) identifiers, into a non-redundant list of gene
+ * identifiers.
*
* @param accessions
* @return
*/
List<String> getGeneIds(String accessions)
{
- List<String> geneIds = new ArrayList<String>();
+ List<String> geneIds = new ArrayList<>();
for (String acc : accessions.split(getAccessionSeparator()))
{
- if (isGeneIdentifier(acc))
- {
- if (!geneIds.contains(acc))
- {
- geneIds.add(acc);
- }
- }
-
/*
- * if given a transcript id, look up its gene parent
+ * First try lookup as an Ensembl (gene or transcript) identifier
*/
- else if (isTranscriptIdentifier(acc))
+ String geneId = new EnsemblLookup(getDomain()).getGeneId(acc);
+ if (geneId != null)
{
- String geneId = new EnsemblLookup(getDomain()).getParent(acc);
- if (geneId != null && !geneIds.contains(geneId))
+ if (!geneIds.contains(geneId))
{
geneIds.add(geneId);
}
}
-
- /*
- * if given a gene or other external name, lookup and fetch
- * the corresponding gene for all model organisms
- */
else
{
+ /*
+ * if given a gene or other external name, lookup and fetch
+ * the corresponding gene for all model organisms
+ */
List<String> ids = new EnsemblSymbol(getDomain(), getDbSource(),
- getDbVersion()).getIds(acc);
- for (String geneId : ids)
+ getDbVersion()).getGeneIds(acc);
+ for (String id : ids)
{
- if (!geneIds.contains(geneId))
+ if (!geneIds.contains(id))
{
- geneIds.add(geneId);
+ geneIds.add(id);
}
}
}
}
/**
- * Attempts to get Ensembl stable identifiers for model organisms for a gene
- * name by calling the xrefs symbol REST service to resolve the gene name.
- *
- * @param query
- * @return
- */
- protected String getGeneIdentifiersForName(String query)
- {
- List<String> ids = new EnsemblSymbol(getDomain(), getDbSource(),
- getDbVersion()).getIds(query);
- if (ids != null)
- {
- for (String id : ids)
- {
- if (isGeneIdentifier(id))
- {
- return id;
- }
- }
- }
- return null;
- }
-
- /**
* Constructs all transcripts for the gene, as identified by "transcript"
* features whose Parent is the requested gene. The coding transcript
* sequences (i.e. with introns omitted) are added to the alignment.
* (it is a sub-type of sequence_variant)
*/
String[] soTerms = new String[] {
- SequenceOntologyI.NMD_TRANSCRIPT_VARIANT, SequenceOntologyI.EXON,
+ SequenceOntologyI.NMD_TRANSCRIPT_VARIANT,
+ SequenceOntologyI.TRANSCRIPT, SequenceOntologyI.EXON,
SequenceOntologyI.CDS };
List<SequenceFeature> sfs = gene.getFeatures().getFeaturesByOntology(
soTerms);
* the parent gene sequence, with features
* @return
*/
- SequenceI makeTranscript(SequenceFeature transcriptFeature,
- AlignmentI al, SequenceI gene)
+ SequenceI makeTranscript(SequenceFeature transcriptFeature, AlignmentI al,
+ SequenceI gene)
{
String accId = getTranscriptId(transcriptFeature);
if (accId == null)
int transcriptLength = 0;
final char[] geneChars = gene.getSequence();
int offset = gene.getStart(); // to convert to 0-based positions
- List<int[]> mappedFrom = new ArrayList<int[]>();
+ List<int[]> mappedFrom = new ArrayList<>();
for (SequenceFeature sf : splices)
{
mappedFrom.add(new int[] { sf.getBegin(), sf.getEnd() });
}
- Sequence transcript = new Sequence(accId, seqChars, 1, transcriptLength);
+ Sequence transcript = new Sequence(accId, seqChars, 1,
+ transcriptLength);
/*
* Ensembl has gene name as transcript Name
* transfer features to the new sequence; we use EnsemblCdna to do this,
* to filter out unwanted features types (see method retainFeature)
*/
- List<int[]> mapTo = new ArrayList<int[]>();
+ List<int[]> mapTo = new ArrayList<>();
mapTo.add(new int[] { 1, transcriptLength });
MapList mapping = new MapList(mappedFrom, mapTo, 1, 1);
EnsemblCdna cdna = new EnsemblCdna(getDomain());
/**
* Returns a list of the transcript features on the sequence whose Parent is
* the gene for the accession id.
+ * <p>
+ * 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
protected List<SequenceFeature> getTranscriptFeatures(String accId,
SequenceI geneSequence)
{
- List<SequenceFeature> transcriptFeatures = new ArrayList<SequenceFeature>();
+ List<SequenceFeature> transcriptFeatures = new ArrayList<>();
String parentIdentifier = GENE_PREFIX + accId;
- // todo optimise here by transcript type!
+
List<SequenceFeature> sfs = geneSequence.getFeatures()
- .getPositionalFeatures();
+ .getFeaturesByOntology(SequenceOntologyI.TRANSCRIPT);
+ sfs.addAll(geneSequence.getFeatures().getPositionalFeatures(
+ SequenceOntologyI.NMD_TRANSCRIPT_VARIANT));
for (SequenceFeature sf : sfs)
{
- if (isTranscript(sf.getType()))
+ String parent = (String) sf.getValue(PARENT);
+ if (parentIdentifier.equalsIgnoreCase(parent))
{
- String parent = (String) sf.getValue(PARENT);
- if (parentIdentifier.equals(parent))
- {
- transcriptFeatures.add(sf);
- }
+ transcriptFeatures.add(sf);
}
}
if (SequenceOntologyFactory.getInstance().isA(sf.getType(),
SequenceOntologyI.GENE))
{
- String id = (String) sf.getValue(ID);
- if ((GENE_PREFIX + accId).equals(id))
+ // 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;
}
if (isTranscript(type))
{
String parent = (String) sf.getValue(PARENT);
- if (!(GENE_PREFIX + accessionId).equals(parent))
+ if (!(GENE_PREFIX + accessionId).equalsIgnoreCase(parent))
{
return false;
}
@Override
public boolean isFeatureDisplayed(String type)
{
- return (so.isA(type, SequenceOntologyI.EXON) || so.isA(type,
- SequenceOntologyI.SEQUENCE_VARIANT));
+ return (so.isA(type, SequenceOntologyI.EXON)
+ || so.isA(type, SequenceOntologyI.SEQUENCE_VARIANT));
}
@Override