import jalview.datamodel.Mapping;
import jalview.datamodel.SearchResults;
import jalview.datamodel.Sequence;
-import jalview.datamodel.SequenceFeature;
import jalview.datamodel.SequenceGroup;
import jalview.datamodel.SequenceI;
import jalview.schemes.ResidueProperties;
final SequenceI ds = dnaSeq.getDatasetSequence();
List<AlignedCodonFrame> seqMappings = MappingUtils
.findMappingsForSequence(ds, mappings);
- if (!seqMappings.isEmpty())
+ for (AlignedCodonFrame acf : seqMappings)
{
- /*
- * We assume here that only one protein mapping is expected per dna
- * sequence. Mapping to multiple protein sequences is conceivable but
- * undefined. Splitting a mapping to one protein sequence across
- * multiple mappings is possible but pathological. Need closer
- * constraints on the contents of AlignedCodonFrame.
- */
AlignedCodonFrame newMapping = new AlignedCodonFrame();
- final SequenceI exonSequence = makeExonSequence(ds,
- seqMappings.get(0), newMapping);
- exonSequences.add(exonSequence);
- newMappings.add(newMapping);
+ final List<SequenceI> mappedExons = makeExonSequences(ds, acf,
+ newMapping);
+ if (!mappedExons.isEmpty())
+ {
+ exonSequences.addAll(mappedExons);
+ newMappings.add(newMapping);
+ }
}
}
AlignmentI al = new Alignment(
}
/**
- * Helper method to make an exon-only sequence and populate its mapping to
- * protein
+ * Helper method to make exon-only sequences and populate their mappings to
+ * protein products
* <p>
* For example, if ggCCaTTcGAg has mappings [3, 4, 6, 7, 9, 10] to protein
* then generate a sequence CCTTGA with mapping [1, 6] to the same protein
* residues
+ * <p>
+ * Typically eukaryotic dna will include exons encoding for a single peptide
+ * sequence i.e. return a single result. Bacterial dna may have overlapping
+ * exon mappings coding for multiple peptides so return multiple results
+ * (example EMBL KF591215).
*
* @param dnaSeq
* a dna dataset sequence
* @param mapping
- * the current mapping of the sequence to protein
+ * containing one or more mappings of the sequence to protein
* @param newMapping
- * the new mapping to populate, from the exon-only sequence
+ * the new mapping to populate, from the exon-only sequences to their
+ * mapped protein sequences
* @return
*/
- protected static SequenceI makeExonSequence(SequenceI dnaSeq,
- AlignedCodonFrame acf, AlignedCodonFrame newMapping)
+ protected static List<SequenceI> makeExonSequences(SequenceI dnaSeq,
+ AlignedCodonFrame mapping, AlignedCodonFrame newMapping)
{
- Mapping mapping = acf.getMappingForSequence(dnaSeq);
+ List<SequenceI> exonSequences = new ArrayList<SequenceI>();
+ List<Mapping> seqMappings = mapping.getMappingsForSequence(dnaSeq);
final char[] dna = dnaSeq.getSequence();
- StringBuilder newSequence = new StringBuilder(dnaSeq.getLength());
-
- /*
- * Get the codon regions as { [2, 5], [7, 12], [14, 14] etc }
- */
- List<int[]> exonRanges = mapping.getMap().getFromRanges();
- for (int[] range : exonRanges)
+ for (Mapping seqMapping : seqMappings)
{
- for (int pos = range[0]; pos <= range[1]; pos++)
+ StringBuilder newSequence = new StringBuilder(dnaSeq.getLength());
+
+ /*
+ * Get the codon regions as { [2, 5], [7, 12], [14, 14] etc }
+ */
+ List<int[]> exonRanges = seqMapping.getMap().getFromRanges();
+ for (int[] range : exonRanges)
{
- newSequence.append(dna[pos - 1]);
+ for (int pos = range[0]; pos <= range[1]; pos++)
+ {
+ newSequence.append(dna[pos - 1]);
+ }
}
- }
- SequenceI exon = new Sequence(dnaSeq.getName(), newSequence.toString());
+ SequenceI exon = new Sequence(dnaSeq.getName(),
+ newSequence.toString());
- /*
- * Locate any xrefs to CDS database on the protein product and attach to the
- * CDS sequence. Also add as a sub-token of the sequence name.
- */
- // default to "CDS" if we can't locate an actual gene id
- String cdsAccId = FeatureProperties.getCodingFeature(DBRefSource.EMBL);
- DBRefEntry[] cdsRefs = DBRefUtils.selectRefs(
- mapping.getTo().getDBRef(), DBRefSource.CODINGDBS);
- if (cdsRefs != null)
- {
- for (DBRefEntry cdsRef : cdsRefs)
+ /*
+ * Locate any xrefs to CDS database on the protein product and attach to
+ * the CDS sequence. Also add as a sub-token of the sequence name.
+ */
+ // default to "CDS" if we can't locate an actual gene id
+ String cdsAccId = FeatureProperties
+ .getCodingFeature(DBRefSource.EMBL);
+ DBRefEntry[] cdsRefs = DBRefUtils.selectRefs(seqMapping.getTo()
+ .getDBRef(), DBRefSource.CODINGDBS);
+ if (cdsRefs != null)
{
- exon.addDBRef(new DBRefEntry(cdsRef));
- cdsAccId = cdsRef.getAccessionId();
+ for (DBRefEntry cdsRef : cdsRefs)
+ {
+ exon.addDBRef(new DBRefEntry(cdsRef));
+ cdsAccId = cdsRef.getAccessionId();
+ }
}
- }
- exon.setName(exon.getName() + "|" + cdsAccId);
- exon.createDatasetSequence();
+ exon.setName(exon.getName() + "|" + cdsAccId);
+ exon.createDatasetSequence();
- /*
- * Build new mappings - from the same protein regions, but now to contiguous
- * exons
- */
- List<int[]> exonRange = new ArrayList<int[]>();
- exonRange.add(new int[]
- { 1, newSequence.length() });
- MapList map = new MapList(exonRange, mapping.getMap().getToRanges(), 3, 1);
- newMapping.addMap(exon.getDatasetSequence(), mapping.getTo(), map);
+ /*
+ * Build new mappings - from the same protein regions, but now to
+ * contiguous exons
+ */
+ List<int[]> exonRange = new ArrayList<int[]>();
+ exonRange.add(new int[]
+ { 1, newSequence.length() });
+ MapList map = new MapList(exonRange, seqMapping.getMap()
+ .getToRanges(),
+ 3, 1);
+ newMapping.addMap(exon.getDatasetSequence(), seqMapping.getTo(), map);
- return exon;
+ exonSequences.add(exon);
+ }
+ return exonSequences;
}
}