X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fdatamodel%2FSequence.java;h=1905f421569d5d1b046812c31653d4aa57f64001;hb=5b0949ce83f4ae4f1327fe386340064c83d88178;hp=a442cf063b55044acf047681310a84ff377a5392;hpb=c70905eb5589c7622a2c7cf8a008ba0863b5c010;p=jalview.git diff --git a/src/jalview/datamodel/Sequence.java b/src/jalview/datamodel/Sequence.java index a442cf0..1905f42 100755 --- a/src/jalview/datamodel/Sequence.java +++ b/src/jalview/datamodel/Sequence.java @@ -38,8 +38,6 @@ import java.util.List; import java.util.ListIterator; import java.util.Vector; -import com.stevesoft.pat.Regex; - import fr.orsay.lri.varna.models.rna.RNA; /** @@ -51,11 +49,6 @@ import fr.orsay.lri.varna.models.rna.RNA; */ public class Sequence extends ASequence implements SequenceI { - private static final Regex limitrx = new Regex( - "[/][0-9]{1,}[-][0-9]{1,}$"); - - private static final Regex endrx = new Regex("[0-9]{1,}$"); - SequenceI datasetSequence; String name; @@ -89,7 +82,7 @@ public class Sequence extends ASequence implements SequenceI */ int index = -1; - private SequenceFeatures sequenceFeatureStore; + private SequenceFeaturesI sequenceFeatureStore; /* * A cursor holding the approximate current view position to the sequence, @@ -151,25 +144,49 @@ public class Sequence extends ASequence implements SequenceI checkValidRange(); } + /** + * If 'name' ends in /i-j, where i >= j > 0 are integers, extracts i and j as + * start and end respectively and removes the suffix from the name + */ void parseId() { if (name == null) { - System.err - .println("POSSIBLE IMPLEMENTATION ERROR: null sequence name passed to constructor."); + System.err.println( + "POSSIBLE IMPLEMENTATION ERROR: null sequence name passed to constructor."); name = ""; } - // Does sequence have the /start-end signature? - if (limitrx.search(name)) + int slashPos = name.lastIndexOf('/'); + if (slashPos > -1 && slashPos < name.length() - 1) { - name = limitrx.left(); - endrx.search(limitrx.stringMatched()); - setStart(Integer.parseInt(limitrx.stringMatched().substring(1, - endrx.matchedFrom() - 1))); - setEnd(Integer.parseInt(endrx.stringMatched())); + String suffix = name.substring(slashPos + 1); + String[] range = suffix.split("-"); + if (range.length == 2) + { + try + { + int from = Integer.valueOf(range[0]); + int to = Integer.valueOf(range[1]); + if (from > 0 && to >= from) + { + name = name.substring(0, slashPos); + setStart(from); + setEnd(to); + checkValidRange(); + } + } catch (NumberFormatException e) + { + // leave name unchanged if suffix is invalid + } + } } } + /** + * Ensures that 'end' is not before the end of the sequence, that is, + * (end-start+1) is at least as long as the count of ungapped positions. Note + * that end is permitted to be beyond the end of the sequence data. + */ void checkValidRange() { // Note: JAL-774 : @@ -178,7 +195,7 @@ public class Sequence extends ASequence implements SequenceI int endRes = 0; for (int j = 0; j < sequence.length; j++) { - if (!jalview.util.Comparison.isGap(sequence[j])) + if (!Comparison.isGap(sequence[j])) { endRes++; } @@ -260,9 +277,8 @@ public class Sequence extends ASequence implements SequenceI protected void initSeqFrom(SequenceI seq, AlignmentAnnotation[] alAnnotation) { - char[] oseq = seq.getSequence(); - initSeqAndName(seq.getName(), Arrays.copyOf(oseq, oseq.length), - seq.getStart(), seq.getEnd()); + char[] oseq = seq.getSequence(); // returns a copy of the array + initSeqAndName(seq.getName(), oseq, seq.getStart(), seq.getEnd()); description = seq.getDescription(); if (seq != datasetSequence) @@ -454,15 +470,15 @@ public class Sequence extends ASequence implements SequenceI } /** - * DOCUMENT ME! + * Sets the sequence name. If the name ends in /start-end, then the start-end + * values are parsed out and set, and the suffix is removed from the name. * - * @param name - * DOCUMENT ME! + * @param theName */ @Override - public void setName(String name) + public void setName(String theName) { - this.name = name; + this.name = theName; this.parseId(); } @@ -563,7 +579,9 @@ public class Sequence extends ASequence implements SequenceI @Override public char[] getSequence() { - return sequence; + // return sequence; + return sequence == null ? null : Arrays.copyOf(sequence, + sequence.length); } /* @@ -644,10 +662,10 @@ public class Sequence extends ASequence implements SequenceI } /** - * DOCUMENT ME! + * Sets the sequence description, and also parses out any special formats of + * interest * * @param desc - * DOCUMENT ME! */ @Override public void setDescription(String desc) @@ -655,10 +673,67 @@ public class Sequence extends ASequence implements SequenceI this.description = desc; } + @Override + public void setGeneLoci(String speciesId, String assemblyId, + String chromosomeId, MapList map) + { + addDBRef(new DBRefEntry(speciesId, assemblyId, DBRefEntry.CHROMOSOME + + ":" + chromosomeId, new Mapping(map))); + } + /** - * DOCUMENT ME! + * Returns the gene loci mapping for the sequence (may be null) * - * @return DOCUMENT ME! + * @return + */ + @Override + public GeneLociI getGeneLoci() + { + DBRefEntry[] refs = getDBRefs(); + if (refs != null) + { + for (final DBRefEntry ref : refs) + { + if (ref.isChromosome()) + { + return new GeneLociI() + { + @Override + public String getSpeciesId() + { + return ref.getSource(); + } + + @Override + public String getAssemblyId() + { + return ref.getVersion(); + } + + @Override + public String getChromosomeId() + { + // strip off "chromosome:" prefix to chrId + return ref.getAccessionId().substring( + DBRefEntry.CHROMOSOME.length() + 1); + } + + @Override + public MapList getMap() + { + return ref.getMap().getMap(); + } + }; + } + } + } + return null; + } + + /** + * Answers the description + * + * @return */ @Override public String getDescription() @@ -722,7 +797,10 @@ public class Sequence extends ASequence implements SequenceI */ protected void updateCursor(int residuePos, int column, int startColumn) { - int endColumn = cursor == null ? 0 : cursor.lastColumnPosition; + /* + * preserve end residue column provided cursor was valid + */ + int endColumn = isValidCursor(cursor) ? cursor.lastColumnPosition : 0; if (residuePos == this.end) { endColumn = column; @@ -980,6 +1058,53 @@ public class Sequence extends ASequence implements SequenceI } /** + * {@inheritDoc} + */ + @Override + public Range findPositions(int fromColumn, int toColumn) + { + if (toColumn < fromColumn || fromColumn < 1) + { + return null; + } + + /* + * find the first non-gapped position, if any + */ + int firstPosition = 0; + int col = fromColumn - 1; + int length = sequence.length; + while (col < length && col < toColumn) + { + if (!Comparison.isGap(sequence[col])) + { + firstPosition = findPosition(col++); + break; + } + col++; + } + + if (firstPosition == 0) + { + return null; + } + + /* + * find the last non-gapped position + */ + int lastPosition = firstPosition; + while (col < length && col < toColumn) + { + if (!Comparison.isGap(sequence[col++])) + { + lastPosition++; + } + } + + return new Range(firstPosition, lastPosition); + } + + /** * Returns an int array where indices correspond to each residue in the * sequence and the element value gives its position in the alignment * @@ -1313,8 +1438,9 @@ public class Sequence extends ASequence implements SequenceI @Override public AlignmentAnnotation[] getAnnotation() { - return annotation == null ? null : annotation - .toArray(new AlignmentAnnotation[annotation.size()]); + return annotation == null ? null + : annotation + .toArray(new AlignmentAnnotation[annotation.size()]); } @Override @@ -1395,7 +1521,7 @@ public class Sequence extends ASequence implements SequenceI private boolean _isNa; - private long _seqhash = 0; + private int _seqhash = 0; /** * Answers false if the sequence is more than 85% nucleotide (ACGTU), else @@ -1426,8 +1552,9 @@ public class Sequence extends ASequence implements SequenceI { if (datasetSequence == null) { - Sequence dsseq = new Sequence(getName(), AlignSeq.extractGaps( - jalview.util.Comparison.GapChars, getSequenceAsString()), + Sequence dsseq = new Sequence(getName(), + AlignSeq.extractGaps(jalview.util.Comparison.GapChars, + getSequenceAsString()), getStart(), getEnd()); datasetSequence = dsseq; @@ -1569,7 +1696,7 @@ public class Sequence extends ASequence implements SequenceI List sfs = entry.getSequenceFeatures(); for (SequenceFeature feature : sfs) { - SequenceFeature sf[] = (mp != null) ? mp.locateFeature(feature) + SequenceFeature sf[] = (mp != null) ? mp.locateFeature(feature) : new SequenceFeature[] { new SequenceFeature(feature) }; if (sf != null) { @@ -1726,8 +1853,8 @@ public class Sequence extends ASequence implements SequenceI } } // whilst it looks like it is a primary ref, we also sanity check type - if (DBRefUtils.getCanonicalName(DBRefSource.PDB).equals( - DBRefUtils.getCanonicalName(ref.getSource()))) + if (DBRefUtils.getCanonicalName(DBRefSource.PDB) + .equals(DBRefUtils.getCanonicalName(ref.getSource()))) { // PDB dbrefs imply there should be a PDBEntry associated // TODO: tighten PDB dbrefs @@ -1763,52 +1890,47 @@ public class Sequence extends ASequence implements SequenceI String... types) { int startPos = findPosition(fromColumn - 1); // convert base 1 to base 0 - int endPos = findPosition(toColumn - 1); - // to trace / debug behaviour: - // System.out - // .println(String - // .format("%s.findFeatures columns [%d-%d] positions [%d-%d] leaves cursor %s", - // getName(), fromColumn, toColumn, startPos, - // endPos, cursor)); - List result = new ArrayList<>(); - if (datasetSequence != null) - { - result = datasetSequence.getFeatures().findFeatures(startPos, endPos, - types); - } - else - { - result = sequenceFeatureStore.findFeatures(startPos, endPos, types); - } + int endPos = fromColumn == toColumn ? startPos + : findPosition(toColumn - 1); + + List result = getFeatures().findFeatures(startPos, + endPos, types); /* - * if the start or end column is gapped, startPos or endPos may be to the - * left or right, and we may have included adjacent or enclosing features; + * if end column is gapped, endPos may be to the right, + * and we may have included adjacent or enclosing features; * remove any that are not enclosing, non-contact features */ - if (endPos > this.end || Comparison.isGap(sequence[fromColumn - 1]) - || Comparison.isGap(sequence[toColumn - 1])) + boolean endColumnIsGapped = toColumn > 0 && toColumn <= sequence.length + && Comparison.isGap(sequence[toColumn - 1]); + if (endPos > this.end || endColumnIsGapped) { ListIterator it = result.listIterator(); while (it.hasNext()) { SequenceFeature sf = it.next(); - int featureStartColumn = findIndex(sf.getBegin()); - int featureEndColumn = findIndex(sf.getEnd()); - boolean noOverlap = featureStartColumn > toColumn - || featureEndColumn < fromColumn; - - /* - * reject an 'enclosing' feature if it is actually a contact feature - */ - if (sf.isContactFeature() && featureStartColumn < fromColumn - && featureEndColumn > toColumn) + int sfBegin = sf.getBegin(); + int sfEnd = sf.getEnd(); + int featureStartColumn = findIndex(sfBegin); + if (featureStartColumn > toColumn) { - noOverlap = true; + it.remove(); } - if (noOverlap) + else if (featureStartColumn < fromColumn) { - it.remove(); + int featureEndColumn = sfEnd == sfBegin ? featureStartColumn + : findIndex(sfEnd); + if (featureEndColumn < fromColumn) + { + it.remove(); + } + else if (featureEndColumn > toColumn && sf.isContactFeature()) + { + /* + * remove an enclosing feature if it is a contact feature + */ + it.remove(); + } } } } @@ -1825,4 +1947,34 @@ public class Sequence extends ASequence implements SequenceI { changeCount++; } + + /** + * {@inheritDoc} + */ + @Override + public int replace(char c1, char c2) + { + if (c1 == c2) + { + return 0; + } + int count = 0; + synchronized (sequence) + { + for (int c = 0; c < sequence.length; c++) + { + if (sequence[c] == c1) + { + sequence[c] = c2; + count++; + } + } + } + if (count > 0) + { + sequenceChanged(); + } + + return count; + } }