import jalview.datamodel.SequenceI;
import jalview.datamodel.features.SequenceFeatures;
import jalview.exceptions.JalviewException;
+import jalview.io.gff.Gff3Helper;
import jalview.io.gff.SequenceOntologyFactory;
import jalview.io.gff.SequenceOntologyI;
import jalview.util.Comparison;
*/
public abstract class EnsemblSeqProxy extends EnsemblRestClient
{
- private static final String ALLELES = "alleles";
-
- protected static final String NAME = "Name";
-
protected static final String DESCRIPTION = "description";
/*
}
/**
- * Parses a JSON response into a list of sequences
+ * Parses a JSON response for a single sequence ID query
*
* @param br
- * @return
+ * @return a single jalview.datamodel.Sequence
* @see http://rest.ensembl.org/documentation/info/sequence_id
*/
protected List<SequenceI> parseSequenceJson(BufferedReader br)
protected MapList getGenomicRangesFromFeatures(SequenceI sourceSequence,
String accId, int start)
{
- List<SequenceFeature> sfs = sourceSequence.getFeatures()
- .getPositionalFeatures();
+ List<SequenceFeature> sfs = getIdentifyingFeatures(sourceSequence,
+ accId);
if (sfs.isEmpty())
{
return null;
for (SequenceFeature sf : sfs)
{
+ int strand = sf.getStrand();
+ strand = strand == 0 ? 1 : strand; // treat unknown as forward
+
+ if (directionSet && strand != direction)
+ {
+ // abort - mix of forward and backward
+ System.err
+ .println("Error: forward and backward strand for " + accId);
+ return null;
+ }
+ direction = strand;
+ directionSet = true;
+
/*
- * accept the target feature type or a specialisation of it
- * (e.g. coding_exon for exon)
+ * add to CDS ranges, semi-sorted forwards/backwards
*/
- if (identifiesSequence(sf, accId))
+ if (strand < 0)
{
- int strand = sf.getStrand();
- strand = strand == 0 ? 1 : strand; // treat unknown as forward
-
- if (directionSet && strand != direction)
- {
- // abort - mix of forward and backward
- System.err.println(
- "Error: forward and backward strand for " + accId);
- return null;
- }
- direction = strand;
- directionSet = true;
-
- /*
- * add to CDS ranges, semi-sorted forwards/backwards
- */
- if (strand < 0)
- {
- regions.add(0, new int[] { sf.getEnd(), sf.getBegin() });
- }
- else
- {
- regions.add(new int[] { sf.getBegin(), sf.getEnd() });
- }
- mappedLength += Math.abs(sf.getEnd() - sf.getBegin() + 1);
-
- if (!isSpliceable())
- {
- /*
- * 'gene' sequence is contiguous so we can stop as soon as its
- * identifying feature has been found
- */
- break;
- }
+ regions.add(0, new int[] { sf.getEnd(), sf.getBegin() });
}
+ else
+ {
+ regions.add(new int[] { sf.getBegin(), sf.getEnd() });
+ }
+ mappedLength += Math.abs(sf.getEnd() - sf.getBegin() + 1);
}
if (regions.isEmpty())
}
/**
- * Answers true if the sequence being retrieved may occupy discontiguous
- * regions on the genomic sequence.
- */
- protected boolean isSpliceable()
- {
- return true;
- }
-
- /**
- * Returns true if the sequence feature marks positions of the genomic
+ * Answers a list of sequence features that mark positions of the genomic
* sequence feature which are within the sequence being retrieved. For
* example, an 'exon' feature whose parent is the target transcript marks the
- * cdna positions of the transcript.
+ * cdna positions of the transcript. For a gene sequence, this is trivially
+ * just the 'gene' feature with matching gene id.
*
- * @param sf
+ * @param seq
* @param accId
* @return
*/
- protected abstract boolean identifiesSequence(SequenceFeature sf,
- String accId);
+ protected abstract List<SequenceFeature> getIdentifyingFeatures(
+ SequenceI seq, String accId);
/**
* Transfers the sequence feature to the target sequence, locating its start
*/
static void reverseComplementAlleles(SequenceFeature sf)
{
- final String alleles = (String) sf.getValue(ALLELES);
+ final String alleles = (String) sf.getValue(Gff3Helper.ALLELES);
if (alleles == null)
{
return;
reverseComplementAllele(complement, allele);
}
String comp = complement.toString();
- sf.setValue(ALLELES, comp);
+ sf.setValue(Gff3Helper.ALLELES, comp);
sf.setDescription(comp);
/*
String atts = sf.getAttributes();
if (atts != null)
{
- atts = atts.replace(ALLELES + "=" + alleles, ALLELES + "=" + comp);
+ atts = atts.replace(Gff3Helper.ALLELES + "=" + alleles,
+ Gff3Helper.ALLELES + "=" + comp);
sf.setAttributes(atts);
}
}
protected boolean featureMayBelong(SequenceFeature sf, String identifier)
{
String parent = (String) sf.getValue(PARENT);
- // using contains to allow for prefix "gene:", "transcript:" etc
if (parent != null
- && !parent.toUpperCase().contains(identifier.toUpperCase()))
+ && !parent.equalsIgnoreCase(identifier))
{
// this genomic feature belongs to a different transcript
return false;
return true;
}
+ /**
+ * Answers a short description of the sequence fetcher
+ */
@Override
public String getDescription()
{
/**
* Answers true if the feature type is either 'NMD_transcript_variant' or
- * 'transcript' or one of its sub-types in the Sequence Ontology. This is
- * needed because NMD_transcript_variant behaves like 'transcript' in Ensembl
+ * 'transcript' (or one of its sub-types in the Sequence Ontology). This is
+ * because NMD_transcript_variant behaves like 'transcript' in Ensembl
* although strictly speaking it is not (it is a sub-type of
* sequence_variant).
+ * <p>
+ * (This test was needed when fetching transcript features as GFF. As we are
+ * now fetching as JSON, all features have type 'transcript' so the check for
+ * NMD_transcript_variant is redundant. Left in for any future case arising.)
*
* @param featureType
* @return
git://source.jalview.org / jalview.git / blobdiff