package jalview.ext.ensembl;
import jalview.datamodel.SequenceFeature;
-import jalview.io.gff.SequenceOntology;
+import jalview.io.gff.SequenceOntologyFactory;
+import jalview.io.gff.SequenceOntologyI;
+
+import java.util.List;
import com.stevesoft.pat.Regex;
public class EnsemblCdna extends EnsemblSeqProxy
{
+ // TODO modify to accept other species e.g. ENSMUSPnnn
+ private static final Regex ACCESSION_REGEX = new Regex(
+ "(ENST|ENSG|CCDS)[0-9.]{3,}$");
+
/*
- * fetch exon features on genomic sequence (to identify the cdnaregions)
+ * fetch exon features on genomic sequence (to identify the cdna regions)
* and cds and variation features (to retain)
*/
private static final EnsemblFeatureType[] FEATURES_TO_FETCH = {
@Override
public Regex getAccessionValidator()
{
- return new Regex("((ENST|ENSG|CCDS)[0-9.]{3,})");
+ return ACCESSION_REGEX;
}
@Override
}
/**
- * Answers true unless the feature type is 'exon' (or a sub-type of exon in
- * the Sequence Ontology). Exon features are only retrieved in order to
- * identify the exon sequence loci, and are redundant information on the exon
- * sequence itself.
+ * Answers true unless the feature type is 'transcript' (or a sub-type in the
+ * Sequence Ontology).
*/
@Override
protected boolean retainFeature(SequenceFeature sf, String accessionId)
{
- if (SequenceOntology.getInstance().isA(sf.getType(),
- SequenceOntology.EXON))
+ if (isTranscript(sf.getType()))
{
return false;
}
- return super.retainFeature(sf, accessionId);
+ return featureMayBelong(sf, accessionId);
}
/**
@Override
protected boolean identifiesSequence(SequenceFeature sf, String accId)
{
- if (SequenceOntology.getInstance().isA(sf.getType(),
- SequenceOntology.EXON))
+ if (SequenceOntologyFactory.getInstance().isA(sf.getType(),
+ SequenceOntologyI.EXON))
{
String parentFeature = (String) sf.getValue(PARENT);
if (("transcript:" + accId).equals(parentFeature))
return false;
}
+ @Override
+ protected List<String> getCrossReferenceDatabases()
+ {
+ return super.getCrossReferenceDatabases();
+ // 30/01/16 also found Vega_transcript, OTTT, ENS_LRG_transcript, UCSC,
+ // HGNC_trans_name, RefSeq_mRNA, RefSeq_mRNA_predicted
+ }
+
}