X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fdatamodel%2FSequence.java;h=950ee0515ce493393a9051afae74a768749d6eac;hb=17e4ea278bc9a5fb280db1252ce78b7a295215f5;hp=59c3fb171029c5d075362d0b7f9ad9d8908e1d02;hpb=948bd3bcbacc509da0cefaae3eedd97300a6ccce;p=jalview.git diff --git a/src/jalview/datamodel/Sequence.java b/src/jalview/datamodel/Sequence.java index 59c3fb1..950ee05 100755 --- a/src/jalview/datamodel/Sequence.java +++ b/src/jalview/datamodel/Sequence.java @@ -21,7 +21,6 @@ package jalview.datamodel; import jalview.analysis.AlignSeq; -import jalview.api.DBRefEntryI; import jalview.datamodel.features.SequenceFeatures; import jalview.datamodel.features.SequenceFeaturesI; import jalview.util.Comparison; @@ -43,32 +42,56 @@ import fr.orsay.lri.varna.models.rna.RNA; /** * - * Implements the SequenceI interface for a char[] based sequence object. - * - * @author $author$ - * @version $Revision$ + * Implements the SequenceI interface for a char[] based sequence object */ public class Sequence extends ASequence implements SequenceI { + + /** + * A subclass that gives us access to modCount, which tracks whether there + * have been any changes. We use this to update + * + * @author hansonr + * + * @param + */ + @SuppressWarnings("serial") + public class DBModList extends ArrayList + { + + protected int getModCount() + { + return modCount; + } + + } + SequenceI datasetSequence; - String name; + private String name; private char[] sequence; - String description; + private String description; - int start; + private int start; - int end; + private int end; - Vector pdbIds; + private Vector pdbIds; - String vamsasId; + private String vamsasId; - DBRefEntry[] dbrefs; + private DBModList dbrefs; // controlled access - RNA rna; + /** + * a flag to let us know that elements have changed in dbrefs + * + * @author Bob Hanson + */ + private int refModCount = 0; + + private RNA rna; /** * This annotation is displayed below the alignment but the positions are tied @@ -76,7 +99,7 @@ public class Sequence extends ASequence implements SequenceI * * TODO: change to List<> */ - Vector annotation; + private Vector annotation; private SequenceFeaturesI sequenceFeatureStore; @@ -281,18 +304,18 @@ public class Sequence extends ASequence implements SequenceI { setDatasetSequence(seq.getDatasetSequence()); } - + /* * only copy DBRefs and seqfeatures if we really are a dataset sequence */ if (datasetSequence == null) { - if (seq.getDBRefs() != null) + List dbr = seq.getDBRefs(); + if (dbr != null) { - DBRefEntry[] dbr = seq.getDBRefs(); - for (int i = 0; i < dbr.length; i++) + for (int i = 0, n = dbr.size(); i < n; i++) { - addDBRef(new DBRefEntry(dbr[i])); + addDBRef(new DBRefEntry(dbr.get(i))); } } @@ -356,8 +379,8 @@ public class Sequence extends ASequence implements SequenceI { if (sf.getType() == null) { - System.err.println("SequenceFeature type may not be null: " - + sf.toString()); + System.err.println( + "SequenceFeature type may not be null: " + sf.toString()); return false; } @@ -445,7 +468,7 @@ public class Sequence extends ASequence implements SequenceI @Override public Vector getAllPDBEntries() { - return pdbIds == null ? new Vector() : pdbIds; + return pdbIds == null ? new Vector<>() : pdbIds; } /** @@ -576,8 +599,8 @@ public class Sequence extends ASequence implements SequenceI public char[] getSequence() { // return sequence; - return sequence == null ? null : Arrays.copyOf(sequence, - sequence.length); + return sequence == null ? null + : Arrays.copyOf(sequence, sequence.length); } /* @@ -673,8 +696,8 @@ public class Sequence extends ASequence implements SequenceI public void setGeneLoci(String speciesId, String assemblyId, String chromosomeId, MapList map) { - addDBRef(new DBRefEntry(speciesId, assemblyId, DBRefEntry.CHROMOSOME - + ":" + chromosomeId, new Mapping(map))); + addDBRef(new GeneLocus(speciesId, assemblyId, chromosomeId, + new Mapping(map))); } /** @@ -685,41 +708,14 @@ public class Sequence extends ASequence implements SequenceI @Override public GeneLociI getGeneLoci() { - DBRefEntry[] refs = getDBRefs(); + List refs = getDBRefs(); if (refs != null) { for (final DBRefEntry ref : refs) { - if (ref.isChromosome()) + if (ref instanceof GeneLociI) { - 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 (GeneLociI) ref; } } } @@ -797,6 +793,7 @@ public class Sequence extends ASequence implements SequenceI * preserve end residue column provided cursor was valid */ int endColumn = isValidCursor(cursor) ? cursor.lastColumnPosition : 0; + if (residuePos == this.end) { endColumn = column; @@ -815,7 +812,7 @@ public class Sequence extends ASequence implements SequenceI * @param curs * @return */ - protected int findIndex(int pos, SequenceCursor curs) + protected int findIndex(final int pos, SequenceCursor curs) { if (!isValidCursor(curs)) { @@ -833,16 +830,20 @@ public class Sequence extends ASequence implements SequenceI /* * move left or right to find pos from hint.position */ - int col = curs.columnPosition - 1; // convert from base 1 to 0-based array - // index + int col = curs.columnPosition - 1; // convert from base 1 to base 0 int newPos = curs.residuePosition; int delta = newPos > pos ? -1 : 1; while (newPos != pos) { col += delta; // shift one column left or right - if (col < 0 || col == sequence.length) + if (col < 0) + { + break; + } + if (col == sequence.length) { + col--; // return last column if we failed to reach pos break; } if (!Comparison.isGap(sequence[col])) @@ -852,7 +853,14 @@ public class Sequence extends ASequence implements SequenceI } col++; // convert back to base 1 - updateCursor(pos, col, curs.firstColumnPosition); + + /* + * only update cursor if we found the target position + */ + if (newPos == pos) + { + updateCursor(pos, col, curs.firstColumnPosition); + } return col; } @@ -870,7 +878,7 @@ public class Sequence extends ASequence implements SequenceI { return findPosition(column + 1, cursor); } - + // TODO recode this more naturally i.e. count residues only // as they are found, not 'in anticipation' @@ -1057,7 +1065,7 @@ public class Sequence extends ASequence implements SequenceI * {@inheritDoc} */ @Override - public Range findPositions(int fromColumn, int toColumn) + public ContiguousI findPositions(int fromColumn, int toColumn) { if (toColumn < fromColumn || fromColumn < 1) { @@ -1175,7 +1183,7 @@ public class Sequence extends ASequence implements SequenceI { ArrayList map = new ArrayList<>(); int lastj = -1, j = 0; - int pos = start; + // int pos = start; int seqlen = sequence.length; while ((j < seqlen)) { @@ -1209,7 +1217,7 @@ public class Sequence extends ASequence implements SequenceI { BitSet map = new BitSet(); int lastj = -1, j = 0; - int pos = start; + // int pos = start; int seqlen = sequence.length; while ((j < seqlen)) { @@ -1251,12 +1259,13 @@ 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 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++) + for (int s = i; s < j && s < sequence.length; s++) { if (Comparison.isGap(sequence[s])) { @@ -1376,24 +1385,22 @@ public class Sequence extends ASequence implements SequenceI vamsasId = id; } + @Deprecated @Override - public void setDBRefs(DBRefEntry[] dbref) + public void setDBRefs(DBModList newDBrefs) { if (dbrefs == null && datasetSequence != null && this != datasetSequence) { - datasetSequence.setDBRefs(dbref); + datasetSequence.setDBRefs(newDBrefs); return; } - dbrefs = dbref; - if (dbrefs != null) - { - DBRefUtils.ensurePrimaries(this); - } + dbrefs = newDBrefs; + refModCount = 0; } @Override - public DBRefEntry[] getDBRefs() + public DBModList getDBRefs() { if (dbrefs == null && datasetSequence != null && this != datasetSequence) @@ -1414,12 +1421,12 @@ public class Sequence extends ASequence implements SequenceI if (dbrefs == null) { - dbrefs = new DBRefEntry[0]; + dbrefs = new DBModList<>(); } - for (DBRefEntryI dbr : dbrefs) + for (int ib = 0, nb = dbrefs.size(); ib < nb; ib++) { - if (dbr.updateFrom(entry)) + if (dbrefs.get(ib).updateFrom(entry)) { /* * found a dbref that either matched, or could be @@ -1429,18 +1436,19 @@ public class Sequence extends ASequence implements SequenceI } } - /* - * extend the array to make room for one more - */ - // TODO use an ArrayList instead - int j = dbrefs.length; - DBRefEntry[] temp = new DBRefEntry[j + 1]; - System.arraycopy(dbrefs, 0, temp, 0, j); - temp[temp.length - 1] = entry; - - dbrefs = temp; + // /// BH OUCH! + // /* + // * extend the array to make room for one more + // */ + // // TODO use an ArrayList instead + // int j = dbrefs.length; + // List temp = new DBRefEntry[j + 1]; + // System.arraycopy(dbrefs, 0, temp, 0, j); + // temp[temp.length - 1] = entry; + // + // dbrefs = temp; - DBRefUtils.ensurePrimaries(this); + dbrefs.add(entry); } @Override @@ -1553,6 +1561,8 @@ public class Sequence extends ASequence implements SequenceI private int _seqhash = 0; + private List primaryRefs; + /** * Answers false if the sequence is more than 85% nucleotide (ACGTU), else * true @@ -1570,7 +1580,7 @@ public class Sequence extends ASequence implements SequenceI _isNa = Comparison.isNucleotide(this); } return !_isNa; - }; + } /* * (non-Javadoc) @@ -1684,13 +1694,14 @@ public class Sequence extends ASequence implements SequenceI // TODO: could merge DBRefs return datasetSequence.updatePDBIds(); } - if (dbrefs == null || dbrefs.length == 0) + if (dbrefs == null || dbrefs.size() == 0) { return false; } boolean added = false; - for (DBRefEntry dbr : dbrefs) + for (int ib = 0, nb = dbrefs.size(); ib < nb; ib++) { + DBRefEntry dbr = dbrefs.get(ib); if (DBRefSource.PDB.equals(dbr.getSource())) { /* @@ -1726,8 +1737,9 @@ public class Sequence extends ASequence implements SequenceI List sfs = entry.getSequenceFeatures(); for (SequenceFeature feature : sfs) { - SequenceFeature sf[] = (mp != null) ? mp.locateFeature(feature) - : new SequenceFeature[] { new SequenceFeature(feature) }; + SequenceFeature sf[] = (mp != null) ? mp.locateFeature(feature) + : new SequenceFeature[] + { new SequenceFeature(feature) }; if (sf != null) { for (int sfi = 0; sfi < sf.length; sfi++) @@ -1749,12 +1761,12 @@ public class Sequence extends ASequence implements SequenceI } } // transfer database references - DBRefEntry[] entryRefs = entry.getDBRefs(); + List entryRefs = entry.getDBRefs(); if (entryRefs != null) { - for (int r = 0; r < entryRefs.length; r++) + for (int r = 0, n = entryRefs.size(); r < n; r++) { - DBRefEntry newref = new DBRefEntry(entryRefs[r]); + DBRefEntry newref = new DBRefEntry(entryRefs.get(r)); if (newref.getMap() != null && mp != null) { // remap ref using our local mapping @@ -1828,6 +1840,8 @@ public class Sequence extends ASequence implements SequenceI return null; } + private List tmpList; + @Override public List getPrimaryDBRefs() { @@ -1835,16 +1849,29 @@ public class Sequence extends ASequence implements SequenceI { return datasetSequence.getPrimaryDBRefs(); } - if (dbrefs == null || dbrefs.length == 0) + if (dbrefs == null || dbrefs.size() == 0) { return Collections.emptyList(); } synchronized (dbrefs) { - List primaries = new ArrayList<>(); - DBRefEntry[] tmp = new DBRefEntry[1]; - for (DBRefEntry ref : dbrefs) + if (refModCount == dbrefs.getModCount() && primaryRefs != null) { + return primaryRefs; // no changes + } + refModCount = dbrefs.getModCount(); + List primaries = (primaryRefs == null + ? (primaryRefs = new ArrayList<>()) + : primaryRefs); + primaries.clear(); + if (tmpList == null) + { + tmpList = new ArrayList<>(); + tmpList.add(null); // for replacement + } + for (int i = 0, n = dbrefs.size(); i < n; i++) + { + DBRefEntry ref = dbrefs.get(i); if (!ref.isPrimaryCandidate()) { continue; @@ -1859,8 +1886,8 @@ 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 (DBRefSource.PDB_CANONICAL_NAME + .equals(ref.getCanonicalSourceName())) { // PDB dbrefs imply there should be a PDBEntry associated // TODO: tighten PDB dbrefs @@ -1869,21 +1896,28 @@ public class Sequence extends ASequence implements SequenceI // handle on the PDBEntry, and a real mapping between sequence and // extracted sequence from PDB file PDBEntry pdbentry = getPDBEntry(ref.getAccessionId()); - if (pdbentry != null && pdbentry.getFile() != null) + if (pdbentry == null || pdbentry.getFile() == null) { - primaries.add(ref); + continue; } - continue; } - // check standard protein or dna sources - tmp[0] = ref; - DBRefEntry[] res = DBRefUtils.selectDbRefs(!isProtein(), tmp); - if (res != null && res[0] == tmp[0]) + else { - primaries.add(ref); - continue; + // check standard protein or dna sources + tmpList.set(0, ref); + List res = DBRefUtils.selectDbRefs(!isProtein(), + tmpList); + if (res == null || res.get(0) != tmpList.get(0)) + { + continue; + } } + primaries.add(ref); } + + // version must be not null, as otherwise it will not be a candidate, + // above + DBRefUtils.ensurePrimaries(this, primaries); return primaries; } } @@ -1901,6 +1935,15 @@ public class Sequence extends ASequence implements SequenceI List result = getFeatures().findFeatures(startPos, endPos, types); + if (datasetSequence != null) + { + result = datasetSequence.getFeatures().findFeatures(startPos, endPos, + types); + } + else + { + result = sequenceFeatureStore.findFeatures(startPos, endPos, types); + } /* * if end column is gapped, endPos may be to the right,