* Realigns the given dna to match the alignment of the protein, using codon
* mappings to translate aligned peptide positions to codons.
*
+ * Always produces a padded CDS alignment.
+ *
* @param dna
* the alignment whose sequences are realigned by this method
* @param protein
// todo: implement this
List<AlignedCodonFrame> mappings = protein.getCodonFrames();
int alignedCount = 0;
+ int width = 0; // alignment width for padding CDS
for (SequenceI dnaSeq : dna.getSequences())
{
if (alignCdsSequenceAsProtein(dnaSeq, protein, mappings,
{
alignedCount++;
}
+ width = Math.max(dnaSeq.getLength(), width);
+ }
+ int oldwidth, diff;
+ for (SequenceI dnaSeq : dna.getSequences())
+ {
+ oldwidth = dnaSeq.getLength();
+ diff = width - oldwidth;
+ if (diff > 0)
+ {
+ dnaSeq.insertCharAt(oldwidth, diff, dna.getGapCharacter());
+ }
}
return alignedCount;
}
return false; // should only pass alignments with datasets here
}
- // map from dataset sequence to alignment sequence
- Map<SequenceI, SequenceI> alignedDatasets = new HashMap<SequenceI, SequenceI>();
+ // map from dataset sequence to alignment sequence(s)
+ Map<SequenceI, List<SequenceI>> alignedDatasets = new HashMap<SequenceI, List<SequenceI>>();
for (SequenceI seq : aligned.getSequences())
{
- alignedDatasets.put(seq.getDatasetSequence(), seq);
+ SequenceI ds = seq.getDatasetSequence();
+ if (alignedDatasets.get(ds) == null)
+ {
+ alignedDatasets.put(ds, new ArrayList<SequenceI>());
+ }
+ alignedDatasets.get(ds).add(seq);
}
/*
}
/*
- * second pass - copy aligned sequences
+ * second pass - copy aligned sequences;
+ * heuristic rule: pair off sequences in order for the case where
+ * more than one shares the same dataset sequence
*/
for (SequenceI seq : unaligned.getSequences())
{
- SequenceI alignedSequence = alignedDatasets.get(seq
+ List<SequenceI> alignedSequences = alignedDatasets.get(seq
.getDatasetSequence());
// TODO: getSequenceAsString() will be deprecated in the future
// TODO: need to leave to SequenceI implementor to update gaps
- seq.setSequence(alignedSequence.getSequenceAsString());
+ seq.setSequence(alignedSequences.get(0).getSequenceAsString());
+ if (alignedSequences.size() > 0)
+ {
+ // pop off aligned sequences (except the last one)
+ alignedSequences.remove(0);
+ }
}
return true;