X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fdatamodel%2FSequence.java;h=89c55e64f273b738e56bd47c1e44e3338e0584b6;hb=1b355b40b6f8db41ced6a05a9eab76b44089f7f1;hp=026061bd00969854e10110d5e861141c52efd762;hpb=8c0019f5e35dcfdb8014f280ba1193db239c92dd;p=jalview.git diff --git a/src/jalview/datamodel/Sequence.java b/src/jalview/datamodel/Sequence.java index 026061b..89c55e6 100755 --- a/src/jalview/datamodel/Sequence.java +++ b/src/jalview/datamodel/Sequence.java @@ -21,13 +21,13 @@ 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; import jalview.util.DBRefUtils; import jalview.util.MapList; import jalview.util.StringUtils; +import jalview.workers.InformationThread; import java.util.ArrayList; import java.util.Arrays; @@ -47,25 +47,55 @@ import fr.orsay.lri.varna.models.rna.RNA; */ 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; + + private int start; + + private int end; - int start; + private Vector pdbIds; - int end; + private String vamsasId; - Vector pdbIds; + HiddenMarkovModel hmm; - String vamsasId; + boolean isHMMConsensusSequence = false; + private DBModList dbrefs; // controlled access - DBRefEntry[] dbrefs; + /** + * a flag to let us know that elements have changed in dbrefs + * + * @author Bob Hanson + */ + private int refModCount = 0; - RNA rna; + private RNA rna; /** * This annotation is displayed below the alignment but the positions are tied @@ -73,7 +103,7 @@ public class Sequence extends ASequence implements SequenceI * * TODO: change to List<> */ - Vector annotation; + private Vector annotation; private SequenceFeaturesI sequenceFeatureStore; @@ -228,6 +258,21 @@ public class Sequence extends ASequence implements SequenceI } /** + * Create a new sequence object from a characters array using default values + * of 1 and -1 for start and end. The array used to create the sequence is + * copied and is not stored internally. + * + * @param name + * sequence name + * @param sequence + * list of residues + */ + public Sequence(String name, char[] sequence) + { + this(name, Arrays.copyOf(sequence, sequence.length), 1, -1); + } + + /** * Creates a new Sequence object with new AlignmentAnnotations but inherits * any existing dataset sequence reference. If non exists, everything is * copied. @@ -278,18 +323,18 @@ 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))); } } @@ -335,6 +380,10 @@ public class Sequence extends ASequence implements SequenceI this.addPDBId(new PDBEntry(pdb)); } } + if (seq.getHMM() != null) + { + this.hmm = new HiddenMarkovModel(seq.getHMM(), this); + } } @Override @@ -353,8 +402,8 @@ 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; } @@ -497,6 +546,7 @@ public class Sequence extends ASequence implements SequenceI public void setStart(int start) { this.start = start; + sequenceChanged(); } /** @@ -544,6 +594,14 @@ public class Sequence extends ASequence implements SequenceI return this.sequence.length; } + @Override + public void setSequence(char[] seq) + { + this.sequence = Arrays.copyOf(seq, seq.length); + checkValidRange(); + sequenceChanged(); + } + /** * DOCUMENT ME! * @@ -574,8 +632,8 @@ 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); } /* @@ -683,7 +741,7 @@ 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) @@ -853,7 +911,7 @@ 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' @@ -1042,7 +1100,8 @@ public class Sequence extends ASequence implements SequenceI @Override public ContiguousI findPositions(int fromColumn, int toColumn) { - if (toColumn < fromColumn || fromColumn < 1) + fromColumn = Math.max(fromColumn, 1); + if (toColumn < fromColumn) { return null; } @@ -1158,7 +1217,7 @@ 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)) { @@ -1192,7 +1251,7 @@ 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)) { @@ -1360,24 +1419,22 @@ 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) @@ -1390,6 +1447,7 @@ public class Sequence extends ASequence implements SequenceI @Override public void addDBRef(DBRefEntry entry) { + // TODO JAL-3980 maintain as sorted list if (datasetSequence != null) { datasetSequence.addDBRef(entry); @@ -1398,12 +1456,13 @@ public class Sequence extends ASequence implements SequenceI if (dbrefs == null) { - dbrefs = new DBRefEntry[0]; + dbrefs = new DBModList<>(); } + // TODO JAL-3979 LOOK UP RATHER THAN SWEEP FOR EFFICIENCY - 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 @@ -1413,18 +1472,19 @@ 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 @@ -1537,6 +1597,8 @@ 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 @@ -1554,7 +1616,7 @@ public class Sequence extends ASequence implements SequenceI _isNa = Comparison.isNucleotide(this); } return !_isNa; - }; + } /* * (non-Javadoc) @@ -1668,13 +1730,14 @@ 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())) { /* @@ -1710,8 +1773,9 @@ 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++) @@ -1733,12 +1797,12 @@ 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 @@ -1769,13 +1833,30 @@ public class Sequence extends ASequence implements SequenceI public List getAlignmentAnnotations(String calcId, String label) { + return getAlignmentAnnotations(calcId, label, null, true); + } + + @Override + public List getAlignmentAnnotations(String calcId, + String label, String description) + { + return getAlignmentAnnotations(calcId, label, description, false); + } + + private List getAlignmentAnnotations(String calcId, + String label, String description, boolean ignoreDescription) + { List result = new ArrayList<>(); if (this.annotation != null) { for (AlignmentAnnotation ann : annotation) { - if (ann.calcId != null && ann.calcId.equals(calcId) - && ann.label != null && ann.label.equals(label)) + String id = ann.getCalcId(); + if ((id != null && id.equals(calcId)) + && (ann.label != null && ann.label.equals(label)) + && ((ignoreDescription && description == null) + || (ann.description != null + && ann.description.equals(description)))) { result.add(ann); } @@ -1812,6 +1893,8 @@ public class Sequence extends ASequence implements SequenceI return null; } + private List tmpList; + @Override public List getPrimaryDBRefs() { @@ -1819,16 +1902,29 @@ 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; @@ -1843,8 +1939,8 @@ 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 @@ -1853,25 +1949,59 @@ 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; } } + @Override + public HiddenMarkovModel getHMM() + { + return hmm; + } + + @Override + public void setHMM(HiddenMarkovModel hmm) + { + this.hmm = hmm; + } + + @Override + public boolean hasHMMAnnotation() + { + if (this.annotation == null) { + return false; + } + for (AlignmentAnnotation ann : annotation) + { + if (InformationThread.HMM_CALC_ID.equals(ann.getCalcId())) + { + return true; + } + } + return false; + } /** * {@inheritDoc} */ @@ -2036,4 +2166,10 @@ public class Sequence extends ASequence implements SequenceI // otherwise, sequence was completely hidden return 0; } + + @Override + public boolean hasHMMProfile() + { + return hmm != null; + } }