public abstract class EnsemblSeqProxy extends EnsemblRestClient
{
private static final List<String> CROSS_REFERENCES = Arrays
- .asList(new String[] { "CCDS" });
+ .asList(new String[] { "CCDS", "Uniprot/SWISSPROT" });
protected static final String CONSEQUENCE_TYPE = "consequence_type";
protected static final String ID = "ID";
- /*
- * this needs special handling, as it isA sequence_variant in the
- * Sequence Ontology, but behaves in Ensembl as if it isA transcript
- */
- protected static final String NMD_VARIANT = "NMD_transcript_variant";
-
protected static final String NAME = "Name";
+ /*
+ * enum for 'type' parameter to the /sequence REST service
+ */
public enum EnsemblSeqType
{
/**
- * type=genomic for the full dna including introns
+ * type=genomic to fetch full dna including introns
*/
GENOMIC("genomic"),
/**
- * type=cdna for transcribed dna including UTRs
+ * type=cdna to fetch dna including UTRs
*/
CDNA("cdna"),
/**
- * type=cds for coding dna excluding UTRs
+ * type=cds to fetch coding dna excluding UTRs
*/
CDS("cds"),
/**
- * type=protein for the peptide product sequence
+ * type=protein to fetch peptide product sequence
*/
PROTEIN("protein");
+ ")";
System.err.println(msg);
break;
- // if (alignment != null)
- // {
- // break; // return what we got
- // }
- // else
- // {
- // throw new JalviewException(msg, r);
- // }
}
}
+ if (alignment == null)
+ {
+ return null;
+ }
+
/*
* fetch and transfer genomic sequence features,
* fetch protein product and add as cross-reference
* get 'dummy' genomic sequence with exon, cds and variation features
*/
SequenceI genomicSequence = null;
- EnsemblOverlap gffFetcher = new EnsemblOverlap();
+ EnsemblFeatures gffFetcher = new EnsemblFeatures();
EnsemblFeatureType[] features = getFeaturesToFetch();
AlignmentI geneFeatures = gffFetcher.getSequenceRecords(accId,
features);
MapList mapList = mapCdsToProtein(querySeq, proteinSeq);
if (mapList != null)
{
- Mapping map = new Mapping(proteinSeq.getDatasetSequence(), mapList);
+ // clunky: ensure Uniprot xref if we have one is on mapped sequence
+ SequenceI ds = proteinSeq.getDatasetSequence();
+ ds.setSourceDBRef(proteinSeq.getSourceDBRef());
+ Mapping map = new Mapping(ds, mapList);
DBRefEntry dbr = new DBRefEntry(getDbSource(), getDbVersion(),
accId, map);
querySeq.getDatasetSequence().addDBRef(dbr);
}
/**
- * Get Uniprot and PDB xrefs from Ensembl, and attach them to the protein
- * sequence
+ * Get database xrefs from Ensembl, and attach them to the sequence
*
* @param seq
*/
int mappedDnaLength = getCdsRanges(dnaSeq, ranges);
int proteinLength = proteinSeq.getLength();
- List<int[]> proteinRange = new ArrayList<int[]>();
+ int proteinEnd = proteinLength;
int proteinStart = 1;
/*
proteinStart = 2;
proteinLength--;
}
- proteinRange.add(new int[] { proteinStart, proteinLength });
+ List<int[]> proteinRange = new ArrayList<int[]>();
/*
* dna length should map to protein (or protein plus stop codon)
*/
int codesForResidues = mappedDnaLength / 3;
- if (codesForResidues == proteinLength
- || codesForResidues == (proteinLength + 1))
+ if (codesForResidues == (proteinLength + 1))
{
+ MappingUtils.unmapStopCodon(ranges, mappedDnaLength);
+ codesForResidues--;
+ }
+ if (codesForResidues == proteinLength)
+ {
+ proteinRange.add(new int[] { proteinStart, proteinEnd });
return new MapList(ranges, proteinRange, 3, 1);
}
return null;
{
return 0;
}
+ SequenceOntologyI so = SequenceOntologyFactory.getInstance();
int mappedDnaLength = 0;
for (SequenceFeature sf : sfs)
{
/*
* process a CDS feature (or a sub-type of CDS)
*/
- if (SequenceOntologyFactory.getInstance().isA(sf.getType(),
- SequenceOntologyI.CDS))
+ if (so.isA(sf.getType(), SequenceOntologyI.CDS))
{
int phase = 0;
try {
*/
int begin = sf.getBegin();
int end = sf.getEnd();
- if (ranges.isEmpty() && phase > 0)
+ if (ranges.isEmpty())
{
begin += phase;
if (begin > end)
/*
* for sequence_variant, make an additional feature with consequence
*/
- if (SequenceOntologyFactory.getInstance().isA(sf.getType(),
- SequenceOntologyI.SEQUENCE_VARIANT))
- {
- String consequence = (String) sf.getValue(CONSEQUENCE_TYPE);
- if (consequence != null)
- {
- SequenceFeature sf2 = new SequenceFeature("consequence",
- consequence, copy.getBegin(), copy.getEnd(), 0f,
- null);
- targetSequence.addSequenceFeature(sf2);
- }
- }
+ // if (SequenceOntologyFactory.getInstance().isA(sf.getType(),
+ // SequenceOntologyI.SEQUENCE_VARIANT))
+ // {
+ // String consequence = (String) sf.getValue(CONSEQUENCE_TYPE);
+ // if (consequence != null)
+ // {
+ // SequenceFeature sf2 = new SequenceFeature("consequence",
+ // consequence, copy.getBegin(), copy.getEnd(), 0f,
+ // null);
+ // targetSequence.addSequenceFeature(sf2);
+ // }
+ // }
}
}
return false;
}
+ // long start = System.currentTimeMillis();
SequenceFeature[] sfs = sourceSequence.getSequenceFeatures();
MapList mapping = getGenomicRangesFromFeatures(sourceSequence, accessionId,
targetSequence.getStart());
return false;
}
- return transferFeatures(sfs, targetSequence, mapping, accessionId);
+ boolean result = transferFeatures(sfs, targetSequence, mapping,
+ accessionId);
+ // System.out.println("transferFeatures (" + (sfs.length) + " --> "
+ // + targetSequence.getSequenceFeatures().length + ") to "
+ // + targetSequence.getName()
+ // + " took " + (System.currentTimeMillis() - start) + "ms");
+ return result;
}
/**
count++;
}
}
+
+ /*
+ * ugly sort to get sequence features in start position order
+ * - would be better to store in Sequence as a TreeSet instead?
+ */
+ Arrays.sort(peptide.getSequenceFeatures(),
+ new Comparator<SequenceFeature>()
+ {
+ @Override
+ public int compare(SequenceFeature o1, SequenceFeature o2)
+ {
+ int c = Integer.compare(o1.getBegin(), o2.getBegin());
+ return c == 0 ? Integer.compare(o1.getEnd(), o2.getEnd())
+ : c;
+ }
+ });
return count;
}
*/
public static boolean isTranscript(String featureType)
{
- return NMD_VARIANT.equals(featureType)
+ return SequenceOntologyI.NMD_TRANSCRIPT_VARIANT.equals(featureType)
|| SequenceOntologyFactory.getInstance().isA(featureType,
SequenceOntologyI.TRANSCRIPT);
}