X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fdatamodel%2FSequence.java;h=9c2fa259b5565ab6331d40d0e2a0f7a662f63ee4;hb=d5702bf3268c2688e90d645ec635ad5bb5d3a76e;hp=96a2fa17b5b6c3253130b8df9552b83cfc0a2ad8;hpb=bba9e58f08fc55c0f789f8354d1444e5c5e1b0e6;p=jalview.git diff --git a/src/jalview/datamodel/Sequence.java b/src/jalview/datamodel/Sequence.java index 96a2fa1..9c2fa25 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; @@ -84,12 +77,7 @@ public class Sequence extends ASequence implements SequenceI */ Vector annotation; - /** - * The index of the sequence in a MSA - */ - int index = -1; - - private SequenceFeatures sequenceFeatureStore; + private SequenceFeaturesI sequenceFeatureStore; /* * A cursor holding the approximate current view position to the sequence, @@ -151,25 +139,49 @@ 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 +190,7 @@ 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++; } @@ -396,7 +408,7 @@ public class Sequence extends ASequence implements SequenceI { if (pdbIds == null) { - pdbIds = new Vector(); + pdbIds = new Vector<>(); pdbIds.add(entry); return true; } @@ -453,15 +465,15 @@ 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(); } @@ -1059,6 +1071,27 @@ public class Sequence extends ASequence implements SequenceI return map; } + /** + * Build a bitset corresponding to sequence gaps + * + * @return a BitSet where set values correspond to gaps in the sequence + */ + @Override + public BitSet gapBitset() + { + BitSet gaps = new BitSet(sequence.length); + int j = 0; + while (j < sequence.length) + { + if (jalview.util.Comparison.isGap(sequence[j])) + { + gaps.set(j); + } + j++; + } + return gaps; + } + @Override public int[] findPositionMap() { @@ -1082,7 +1115,7 @@ public class Sequence extends ASequence implements SequenceI @Override public List getInsertions() { - ArrayList map = new ArrayList(); + ArrayList map = new ArrayList<>(); int lastj = -1, j = 0; int pos = start; int seqlen = sequence.length; @@ -1148,7 +1181,7 @@ public class Sequence extends ASequence implements SequenceI } @Override - public void deleteChars(int i, int j) + public void deleteChars(final int i, final int j) { int newstart = start, newend = end; if (i >= sequence.length || i < 0) @@ -1160,62 +1193,75 @@ public class Sequence extends ASequence implements SequenceI boolean createNewDs = false; // TODO: take a (second look) at the dataset creation validation method for // the very large sequence case - int eindex = -1, sindex = -1; - boolean ecalc = false, scalc = false; + int startIndex = findIndex(start) - 1; + int endIndex = findIndex(end) - 1; + int startDeleteColumn = -1; // for dataset sequence deletions + int deleteCount = 0; + for (int s = i; s < j; s++) { - if (jalview.schemes.ResidueProperties.aaIndex[sequence[s]] != 23) + if (Comparison.isGap(sequence[s])) + { + continue; + } + deleteCount++; + if (startDeleteColumn == -1) + { + startDeleteColumn = findPosition(s) - start; + } + if (createNewDs) + { + newend--; + } + else { - if (createNewDs) + if (startIndex == s) { - newend--; + /* + * deleting characters from start of sequence; new start is the + * sequence position of the next column (position to the right + * if the column position is gapped) + */ + newstart = findPosition(j); + break; } else { - if (!scalc) - { - sindex = findIndex(start) - 1; - scalc = true; - } - if (sindex == s) + if (endIndex < j) { - // delete characters including start of sequence - newstart = findPosition(j); - break; // don't need to search for any more residue characters. + /* + * deleting characters at end of sequence; new end is the sequence + * position of the column before the deletion; subtract 1 if this is + * gapped since findPosition returns the next sequence position + */ + newend = findPosition(i - 1); + if (Comparison.isGap(sequence[i - 1])) + { + newend--; + } + break; } else { - // delete characters after start. - if (!ecalc) - { - eindex = findIndex(end) - 1; - ecalc = true; - } - if (eindex < j) - { - // delete characters at end of sequence - newend = findPosition(i - 1); - break; // don't need to search for any more residue characters. - } - else - { - createNewDs = true; - newend--; // decrease end position by one for the deleted residue - // and search further - } + createNewDs = true; + newend--; } } } } - // deletion occured in the middle of the sequence + if (createNewDs && this.datasetSequence != null) { - // construct a new sequence + /* + * if deletion occured in the middle of the sequence, + * construct a new dataset sequence and delete the residues + * that were deleted from the aligned sequence + */ Sequence ds = new Sequence(datasetSequence); + ds.deleteChars(startDeleteColumn, startDeleteColumn + deleteCount); + datasetSequence = ds; // TODO: remove any non-inheritable properties ? // TODO: create a sequence mapping (since there is a relation here ?) - ds.deleteChars(i, j); - datasetSequence = ds; } start = newstart; end = newend; @@ -1364,8 +1410,9 @@ 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 @@ -1379,7 +1426,7 @@ public class Sequence extends ASequence implements SequenceI { if (this.annotation == null) { - this.annotation = new Vector(); + this.annotation = new Vector<>(); } if (!this.annotation.contains(annotation)) { @@ -1477,8 +1524,9 @@ 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; @@ -1545,7 +1593,7 @@ public class Sequence extends ASequence implements SequenceI return null; } - Vector subset = new Vector(); + Vector subset = new Vector<>(); Enumeration e = annotation.elements(); while (e.hasMoreElements()) { @@ -1620,7 +1668,7 @@ 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) { @@ -1663,30 +1711,6 @@ public class Sequence extends ASequence implements SequenceI } } - /** - * @return The index (zero-based) on this sequence in the MSA. It returns - * {@code -1} if this information is not available. - */ - @Override - public int getIndex() - { - return index; - } - - /** - * Defines the position of this sequence in the MSA. Use the value {@code -1} - * if this information is undefined. - * - * @param The - * position for this sequence. This value is zero-based (zero for - * this first sequence) - */ - @Override - public void setIndex(int value) - { - index = value; - } - @Override public void setRNA(RNA r) { @@ -1703,7 +1727,7 @@ public class Sequence extends ASequence implements SequenceI public List getAlignmentAnnotations(String calcId, String label) { - List result = new ArrayList(); + List result = new ArrayList<>(); if (this.annotation != null) { for (AlignmentAnnotation ann : annotation) @@ -1759,7 +1783,7 @@ public class Sequence extends ASequence implements SequenceI } synchronized (dbrefs) { - List primaries = new ArrayList(); + List primaries = new ArrayList<>(); DBRefEntry[] tmp = new DBRefEntry[1]; for (DBRefEntry ref : dbrefs) { @@ -1777,8 +1801,8 @@ 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 @@ -1821,12 +1845,13 @@ public class Sequence extends ASequence implements SequenceI 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; - * remove any that are not 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()) @@ -1847,6 +1872,13 @@ public class Sequence extends ASequence implements SequenceI { it.remove(); } + else if (featureEndColumn > toColumn && sf.isContactFeature()) + { + /* + * remove an enclosing feature if it is a contact feature + */ + it.remove(); + } } } }