int toOffset = alignTo.getStart() - 1;
int sourceGapMappedLength = 0;
boolean inExon = false;
- final char[] thisSeq = alignTo.getSequence();
- final char[] thatAligned = alignFrom.getSequence();
- StringBuilder thisAligned = new StringBuilder(2 * thisSeq.length);
+ final int toLength = alignTo.getLength();
+ final int fromLength = alignFrom.getLength();
+ StringBuilder thisAligned = new StringBuilder(2 * toLength);
/*
* Traverse the 'model' aligned sequence
*/
- for (char sourceChar : thatAligned)
+ for (int i = 0; i < fromLength; i++)
{
+ char sourceChar = alignFrom.getCharAt(i);
if (sourceChar == sourceGap)
{
sourceGapMappedLength += ratio;
*/
int intronLength = 0;
while (basesWritten + toOffset < mappedCodonEnd
- && thisSeqPos < thisSeq.length)
+ && thisSeqPos < toLength)
{
- final char c = thisSeq[thisSeqPos++];
+ final char c = alignTo.getCharAt(thisSeqPos++);
if (c != myGapChar)
{
basesWritten++;
int gapsToAdd = calculateGapsToInsert(preserveMappedGaps,
preserveUnmappedGaps, sourceGapMappedLength, inExon,
trailingCopiedGap.length(), intronLength, startOfCodon);
- for (int i = 0; i < gapsToAdd; i++)
+ for (int k = 0; k < gapsToAdd; k++)
{
thisAligned.append(myGapChar);
}
* At end of model aligned sequence. Copy any remaining target sequence, optionally
* including (intron) gaps.
*/
- while (thisSeqPos < thisSeq.length)
+ while (thisSeqPos < toLength)
{
- final char c = thisSeq[thisSeqPos++];
+ final char c = alignTo.getCharAt(thisSeqPos++);
if (c != myGapChar || preserveUnmappedGaps)
{
thisAligned.append(c);
SequenceI peptide = mapping.findAlignedSequence(cdsSeq, protein);
if (peptide != null)
{
- int peptideLength = peptide.getLength();
+ final int peptideLength = peptide.getLength();
Mapping map = mapping.getMappingBetween(cdsSeq, peptide);
if (map != null)
{
{
mapList = mapList.getInverse();
}
- int cdsLength = cdsDss.getLength();
+ final int cdsLength = cdsDss.getLength();
int mappedFromLength = MappingUtils.getLength(mapList
.getFromRanges());
int mappedToLength = MappingUtils
* walk over the aligned peptide sequence and insert mapped
* codons for residues in the aligned cds sequence
*/
- char[] alignedPeptide = peptide.getSequence();
- char[] nucleotides = cdsDss.getSequence();
int copiedBases = 0;
int cdsStart = cdsDss.getStart();
int proteinPos = peptide.getStart() - 1;
int cdsCol = 0;
- for (char residue : alignedPeptide)
+
+ for (int col = 0; col < peptideLength; col++)
{
+ char residue = peptide.getCharAt(col);
+
if (Comparison.isGap(residue))
{
cdsCol += CODON_LENGTH;
{
for (int j = codon[0]; j <= codon[1]; j++)
{
- char mappedBase = nucleotides[j - cdsStart];
+ char mappedBase = cdsDss.getCharAt(j - cdsStart);
alignedCds[cdsCol++] = mappedBase;
copiedBases++;
}
* append stop codon if not mapped from protein,
* closing it up to the end of the mapped sequence
*/
- if (copiedBases == nucleotides.length - CODON_LENGTH)
+ if (copiedBases == cdsLength - CODON_LENGTH)
{
for (int i = alignedCds.length - 1; i >= 0; i--)
{
break;
}
}
- for (int i = nucleotides.length - CODON_LENGTH; i < nucleotides.length; i++)
+ for (int i = cdsLength - CODON_LENGTH; i < cdsLength; i++)
{
- alignedCds[cdsCol++] = nucleotides[i];
+ alignedCds[cdsCol++] = cdsDss.getCharAt(i);
}
}
cdsSeq.setSequence(new String(alignedCds));
List<SequenceI> unmappedProtein)
{
/*
- * Prefill aligned sequences with gaps before inserting aligned protein
- * residues.
+ * prefill peptide sequences with gaps
*/
int alignedWidth = alignedCodons.size();
char[] gaps = new char[alignedWidth];
Arrays.fill(gaps, protein.getGapCharacter());
- String allGaps = String.valueOf(gaps);
+ Map<SequenceI, char[]> peptides = new HashMap<>();
for (SequenceI seq : protein.getSequences())
{
if (!unmappedProtein.contains(seq))
{
- seq.setSequence(allGaps);
+ peptides.put(seq, Arrays.copyOf(gaps, gaps.length));
}
}
+ /*
+ * Traverse the codons left to right (as defined by CodonComparator)
+ * and insert peptides in each column where the sequence is mapped.
+ * This gives a peptide 'alignment' where residues are aligned if their
+ * corresponding codons occupy the same columns in the cdna alignment.
+ */
int column = 0;
for (AlignedCodon codon : alignedCodons.keySet())
{
.get(codon);
for (Entry<SequenceI, AlignedCodon> entry : columnResidues.entrySet())
{
- // place translated codon at its column position in sequence
- entry.getKey().getSequence()[column] = entry.getValue().product
- .charAt(0);
+ char residue = entry.getValue().product.charAt(0);
+ peptides.get(entry.getKey())[column] = residue;
}
column++;
}
+
+ /*
+ * and finally set the constructed sequences
+ */
+ for (Entry<SequenceI, char[]> entry : peptides.entrySet())
+ {
+ entry.getKey().setSequence(new String(entry.getValue()));
+ }
+
return 0;
}
.getInverse());
}
- char[] fromChars = fromSeq.getSequence();
int toStart = seq.getStart();
- char[] toChars = seq.getSequence();
/*
* traverse [start, end, start, end...] ranges in fromSeq
* of the next character of the mapped-to sequence; stop when all
* the characters of the range have been counted
*/
- while (mappedCharPos <= range[1] && fromCol <= fromChars.length
+ while (mappedCharPos <= range[1] && fromCol <= fromSeq.getLength()
&& fromCol >= 0)
{
- if (!Comparison.isGap(fromChars[fromCol - 1]))
+ if (!Comparison.isGap(fromSeq.getCharAt(fromCol - 1)))
{
/*
* mapped from sequence has a character in this column
seqsMap = new HashMap<SequenceI, Character>();
map.put(fromCol, seqsMap);
}
- seqsMap.put(seq, toChars[mappedCharPos - toStart]);
+ seqsMap.put(seq, seq.getCharAt(mappedCharPos - toStart));
mappedCharPos++;
}
fromCol += (forward ? 1 : -1);