2 * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
3 * Copyright (C) $$Year-Rel$$ The Jalview Authors
5 * This file is part of Jalview.
7 * Jalview is free software: you can redistribute it and/or
8 * modify it under the terms of the GNU General Public License
9 * as published by the Free Software Foundation, either version 3
10 * of the License, or (at your option) any later version.
12 * Jalview is distributed in the hope that it will be useful, but
13 * WITHOUT ANY WARRANTY; without even the implied warranty
14 * of MERCHANTABILITY or FITNESS FOR A PARTICULAR
15 * PURPOSE. See the GNU General Public License for more details.
17 * You should have received a copy of the GNU General Public License
18 * along with Jalview. If not, see <http://www.gnu.org/licenses/>.
19 * The Jalview Authors are detailed in the 'AUTHORS' file.
21 package jalview.datamodel;
23 import jalview.analysis.AlignSeq;
24 import jalview.api.DBRefEntryI;
25 import jalview.datamodel.features.SequenceFeatures;
26 import jalview.datamodel.features.SequenceFeaturesI;
27 import jalview.util.Comparison;
28 import jalview.util.DBRefUtils;
29 import jalview.util.MapList;
30 import jalview.util.StringUtils;
32 import java.util.ArrayList;
33 import java.util.Arrays;
34 import java.util.Collections;
35 import java.util.Enumeration;
36 import java.util.List;
37 import java.util.Vector;
39 import com.stevesoft.pat.Regex;
41 import fr.orsay.lri.varna.models.rna.RNA;
45 * Implements the SequenceI interface for a char[] based sequence object.
50 public class Sequence extends ASequence implements SequenceI
52 private static final Regex limitrx = new Regex(
53 "[/][0-9]{1,}[-][0-9]{1,}$");
55 private static final Regex endrx = new Regex("[0-9]{1,}$");
57 SequenceI datasetSequence;
61 private char[] sequence;
69 Vector<PDBEntry> pdbIds;
78 * This annotation is displayed below the alignment but the positions are tied
79 * to the residues of this sequence
81 * TODO: change to List<>
83 Vector<AlignmentAnnotation> annotation;
86 * The index of the sequence in a MSA
90 private SequenceFeatures sequenceFeatureStore;
93 * A cursor holding the approximate current view position to the sequence,
94 * as determined by findIndex or findPosition or findPositions.
95 * Using a cursor as a hint allows these methods to be more performant for
98 private SequenceCursor cursor;
101 * A number that should be incremented whenever the sequence is edited.
102 * If the value matches the cursor token, then we can trust the cursor,
103 * if not then it should be recomputed.
105 private int changeCount;
108 * Creates a new Sequence object.
111 * display name string
113 * string to form a possibly gapped sequence out of
115 * first position of non-gap residue in the sequence
117 * last position of ungapped residues (nearly always only used for
120 public Sequence(String name, String sequence, int start, int end)
123 initSeqAndName(name, sequence.toCharArray(), start, end);
126 public Sequence(String name, char[] sequence, int start, int end)
129 initSeqAndName(name, sequence, start, end);
133 * Stage 1 constructor - assign name, sequence, and set start and end fields.
134 * start and end are updated values from name2 if it ends with /start-end
141 protected void initSeqAndName(String name2, char[] sequence2, int start2,
145 this.sequence = sequence2;
157 .println("POSSIBLE IMPLEMENTATION ERROR: null sequence name passed to constructor.");
160 // Does sequence have the /start-end signature?
161 if (limitrx.search(name))
163 name = limitrx.left();
164 endrx.search(limitrx.stringMatched());
165 setStart(Integer.parseInt(limitrx.stringMatched().substring(1,
166 endrx.matchedFrom() - 1)));
167 setEnd(Integer.parseInt(endrx.stringMatched()));
171 void checkValidRange()
174 // http://issues.jalview.org/browse/JAL-774?focusedCommentId=11239&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-11239
177 for (int j = 0; j < sequence.length; j++)
179 if (!jalview.util.Comparison.isGap(sequence[j]))
198 * default constructor
202 sequenceFeatureStore = new SequenceFeatures();
206 * Creates a new Sequence object.
213 public Sequence(String name, String sequence)
215 this(name, sequence, 1, -1);
219 * Creates a new Sequence object with new AlignmentAnnotations but inherits
220 * any existing dataset sequence reference. If non exists, everything is
224 * if seq is a dataset sequence, behaves like a plain old copy
227 public Sequence(SequenceI seq)
229 this(seq, seq.getAnnotation());
233 * Create a new sequence object with new features, DBRefEntries, and PDBIds
234 * but inherits any existing dataset sequence reference, and duplicate of any
235 * annotation that is present in the given annotation array.
238 * the sequence to be copied
239 * @param alAnnotation
240 * an array of annotation including some associated with seq
242 public Sequence(SequenceI seq, AlignmentAnnotation[] alAnnotation)
245 initSeqFrom(seq, alAnnotation);
249 * does the heavy lifting when cloning a dataset sequence, or coping data from
250 * dataset to a new derived sequence.
253 * - source of attributes.
254 * @param alAnnotation
255 * - alignment annotation present on seq that should be copied onto
258 protected void initSeqFrom(SequenceI seq,
259 AlignmentAnnotation[] alAnnotation)
261 char[] oseq = seq.getSequence();
262 initSeqAndName(seq.getName(), Arrays.copyOf(oseq, oseq.length),
263 seq.getStart(), seq.getEnd());
265 description = seq.getDescription();
266 if (seq != datasetSequence)
268 setDatasetSequence(seq.getDatasetSequence());
272 * only copy DBRefs and seqfeatures if we really are a dataset sequence
274 if (datasetSequence == null)
276 if (seq.getDBRefs() != null)
278 DBRefEntry[] dbr = seq.getDBRefs();
279 for (int i = 0; i < dbr.length; i++)
281 addDBRef(new DBRefEntry(dbr[i]));
286 * make copies of any sequence features
288 for (SequenceFeature sf : seq.getSequenceFeatures())
290 addSequenceFeature(new SequenceFeature(sf));
294 if (seq.getAnnotation() != null)
296 AlignmentAnnotation[] sqann = seq.getAnnotation();
297 for (int i = 0; i < sqann.length; i++)
299 if (sqann[i] == null)
303 boolean found = (alAnnotation == null);
306 for (int apos = 0; !found && apos < alAnnotation.length; apos++)
308 found = (alAnnotation[apos] == sqann[i]);
313 // only copy the given annotation
314 AlignmentAnnotation newann = new AlignmentAnnotation(sqann[i]);
315 addAlignmentAnnotation(newann);
319 if (seq.getAllPDBEntries() != null)
321 Vector<PDBEntry> ids = seq.getAllPDBEntries();
322 for (PDBEntry pdb : ids)
324 this.addPDBId(new PDBEntry(pdb));
330 public void setSequenceFeatures(List<SequenceFeature> features)
332 if (datasetSequence != null)
334 datasetSequence.setSequenceFeatures(features);
337 sequenceFeatureStore = new SequenceFeatures(features);
341 public synchronized boolean addSequenceFeature(SequenceFeature sf)
343 if (sf.getType() == null)
345 System.err.println("SequenceFeature type may not be null: "
350 if (datasetSequence != null)
352 return datasetSequence.addSequenceFeature(sf);
355 return sequenceFeatureStore.add(sf);
359 public void deleteFeature(SequenceFeature sf)
361 if (datasetSequence != null)
363 datasetSequence.deleteFeature(sf);
367 sequenceFeatureStore.delete(sf);
377 public List<SequenceFeature> getSequenceFeatures()
379 if (datasetSequence != null)
381 return datasetSequence.getSequenceFeatures();
383 return sequenceFeatureStore.getAllFeatures();
387 public SequenceFeaturesI getFeatures()
389 return datasetSequence != null ? datasetSequence.getFeatures()
390 : sequenceFeatureStore;
394 public boolean addPDBId(PDBEntry entry)
398 pdbIds = new Vector<PDBEntry>();
403 for (PDBEntry pdbe : pdbIds)
405 if (pdbe.updateFrom(entry))
410 pdbIds.addElement(entry);
421 public void setPDBId(Vector<PDBEntry> id)
429 * @return DOCUMENT ME!
432 public Vector<PDBEntry> getAllPDBEntries()
434 return pdbIds == null ? new Vector<PDBEntry>() : pdbIds;
440 * @return DOCUMENT ME!
443 public String getDisplayId(boolean jvsuffix)
445 StringBuffer result = new StringBuffer(name);
448 result.append("/" + start + "-" + end);
451 return result.toString();
461 public void setName(String name)
470 * @return DOCUMENT ME!
473 public String getName()
485 public void setStart(int start)
493 * @return DOCUMENT ME!
496 public int getStart()
508 public void setEnd(int end)
516 * @return DOCUMENT ME!
527 * @return DOCUMENT ME!
530 public int getLength()
532 return this.sequence.length;
542 public void setSequence(String seq)
544 this.sequence = seq.toCharArray();
550 public String getSequenceAsString()
552 return new String(sequence);
556 public String getSequenceAsString(int start, int end)
558 return new String(getSequence(start, end));
562 public char[] getSequence()
570 * @see jalview.datamodel.SequenceI#getSequence(int, int)
573 public char[] getSequence(int start, int end)
579 // JBPNote - left to user to pad the result here (TODO:Decide on this
581 if (start >= sequence.length)
586 if (end >= sequence.length)
588 end = sequence.length;
591 char[] reply = new char[end - start];
592 System.arraycopy(sequence, start, reply, 0, end - start);
598 public SequenceI getSubSequence(int start, int end)
604 char[] seq = getSequence(start, end);
609 int nstart = findPosition(start);
610 int nend = findPosition(end) - 1;
611 // JBPNote - this is an incomplete copy.
612 SequenceI nseq = new Sequence(this.getName(), seq, nstart, nend);
613 nseq.setDescription(description);
614 if (datasetSequence != null)
616 nseq.setDatasetSequence(datasetSequence);
620 nseq.setDatasetSequence(this);
626 * Returns the character of the aligned sequence at the given position (base
627 * zero), or space if the position is not within the sequence's bounds
632 public char getCharAt(int i)
634 if (i >= 0 && i < sequence.length)
651 public void setDescription(String desc)
653 this.description = desc;
659 * @return DOCUMENT ME!
662 public String getDescription()
664 return this.description;
671 public int findIndex(int pos)
674 * use a valid, hopefully nearby, cursor if available
676 if (isValidCursor(cursor))
678 return findIndex(pos, cursor);
683 // Rely on end being at least as long as the length of the sequence.
684 while ((i < sequence.length) && (j <= end) && (j <= pos))
686 if (!Comparison.isGap(sequence[i]))
693 if (j == end && j < pos)
698 updateCursor(pos, i);
703 * Updates the cursor to the latest found residue and column position
710 protected void updateCursor(int residuePos, int column)
712 cursor = new SequenceCursor(this, residuePos, column, this.changeCount);
716 * Answers the aligned column position (1..) for the given residue position
717 * (start..) given a 'hint' of a residue/column location in the neighbourhood.
718 * The hint may be left of, at, or to the right of the required position.
724 protected int findIndex(int pos, SequenceCursor curs)
726 if (!isValidCursor(curs))
729 * wrong or invalidated cursor, compute de novo
731 return findIndex(pos);
734 if (curs.residuePosition == pos)
736 return curs.columnPosition;
740 * move left or right to find pos from hint.position
742 int col = curs.columnPosition - 1; // convert from base 1 to 0-based array
744 int newPos = curs.residuePosition;
745 int delta = newPos > pos ? -1 : 1;
747 while (newPos != pos)
749 col += delta; // shift one column left or right
750 if (col < 0 || col == sequence.length)
754 if (!Comparison.isGap(sequence[col]))
760 col++; // convert back to base 1
761 updateCursor(pos, col);
770 public int findPosition(final int column)
773 * use a valid, hopefully nearby, cursor if available
775 if (isValidCursor(cursor))
777 return findPosition(column + 1, cursor);
780 // TODO recode this more naturally i.e. count residues only
781 // as they are found, not 'in anticipation'
783 int lastPosFound = 0;
784 int lastPosFoundColumn = 0;
785 int seqlen = sequence.length;
786 if (seqlen > 0 && !Comparison.isGap(sequence[0]))
788 lastPosFound = start;
789 lastPosFoundColumn = 0;
795 while (j < column && j < seqlen)
797 if (!Comparison.isGap(sequence[j]))
800 lastPosFoundColumn = j;
805 if (j < seqlen && !Comparison.isGap(sequence[j]))
808 lastPosFoundColumn = j;
812 * update the cursor to the last residue position found (if any)
813 * (converting column position to base 1)
815 if (lastPosFound != 0)
817 updateCursor(lastPosFound, lastPosFoundColumn + 1);
824 * Answers true if the given cursor is not null, is for this sequence object,
825 * and has a token value that matches this object's changeCount, else false.
826 * This allows us to ignore a cursor as 'stale' if the sequence has been
827 * modified since the cursor was created.
832 protected boolean isValidCursor(SequenceCursor curs)
834 if (curs == null || curs.sequence != this || curs.token != changeCount)
839 * sanity check against range
841 if (curs.columnPosition < 0 || curs.columnPosition >= sequence.length)
845 if (curs.residuePosition < start || curs.residuePosition > end)
853 * Answers the sequence position (start..) for the given aligned column
854 * position (1..), given a hint of a cursor in the neighbourhood. The cursor
855 * may lie left of, at, or to the right of the column position.
861 protected int findPosition(final int col, SequenceCursor curs)
863 if (!isValidCursor(curs))
866 * wrong or invalidated cursor, compute de novo
868 return findPosition(col - 1);// ugh back to base 0
871 if (curs.columnPosition == col)
873 cursor = curs; // in case this method becomes public
874 return curs.residuePosition; // easy case :-)
878 * move left or right to find pos from cursor position
880 int column = curs.columnPosition - 1; // to base 0
881 int newPos = curs.residuePosition;
882 int delta = curs.columnPosition > col ? -1 : 1;
883 boolean gapped = false;
884 int lastFoundPosition = curs.residuePosition;
885 int lastFoundPositionColumn = curs.columnPosition;
887 while (column != col - 1)
889 column += delta; // shift one column left or right
890 if (column < 0 || column == sequence.length)
894 gapped = Comparison.isGap(sequence[column]);
898 lastFoundPosition = newPos;
899 lastFoundPositionColumn = column + 1;
903 if (cursor == null || lastFoundPosition != cursor.residuePosition)
905 updateCursor(lastFoundPosition, lastFoundPositionColumn);
909 * hack to give position to the right if on a gap
910 * or beyond the length of the sequence (see JAL-2562)
912 if (delta > 0 && (gapped || column >= sequence.length))
924 public Range findPositions(int fromCol, int toCol)
926 if (cursor != null && cursor.sequence == this
927 && cursor.token == changeCount)
929 return findPositions(fromCol, toCol, cursor);
933 * count residues before fromCol
937 int seqlen = sequence.length;
938 while (j < fromCol && j < seqlen)
940 if (!Comparison.isGap(sequence[j]))
948 * find first and last residues between fromCol and toCol
952 boolean foundFirst = false;
954 while (j <= toCol && j < seqlen)
956 if (!Comparison.isGap(sequence[j]))
972 * no residues in this range
978 * adjust for sequence start coordinate
980 firstPos += start - 1;
981 lastPos += start - 1;
983 return new Range(firstPos, lastPos);
987 * Returns the range of sequence positions included in the given alignment
988 * position range. If no positions are included (the range is entirely gaps),
989 * then returns null. The cursor parameter may provide a starting position in
990 * the neighbourhood of the search (which may be left of, right of, or
991 * overlapping the search region).
994 * start column of region (0..)
996 * end column of region (0..)
1000 protected Range findPositions(int fromCol, int toCol, SequenceCursor curs)
1002 if (!isValidCursor(curs))
1005 * wrong or invalidated cursor, compute de novo
1007 return findPositions(fromCol, toCol);
1011 * keep this simple...first step from cursor to fromCol...
1013 final int seqlen = sequence.length;
1014 int resNo = curs.residuePosition;
1015 int col = curs.columnPosition - 1; // from base 1 to base 0
1018 int delta = col > fromCol ? -1 : 1;
1019 while (col != fromCol && col >= 0 && col < seqlen)
1021 if (!Comparison.isGap(sequence[col]))
1029 if (col < fromCol || col == seqlen)
1032 * sequence lies to the left of the target region
1038 * resNo is now the residue at fromCol (if not gapped), else the one
1039 * before it (if delta == 1), else the one after (if delta == -1);
1040 * we want the residue before fromCol
1042 if (!Comparison.isGap(sequence[fromCol]))
1046 else if (curs.columnPosition > fromCol)
1052 * now first and last residues between fromCol and toCol
1056 boolean foundFirst = false;
1058 while (col <= toCol && col < seqlen)
1060 if (!Comparison.isGap(sequence[col]))
1076 * no residues in this range
1081 return new Range(firstPos, lastPos);
1085 * Returns an int array where indices correspond to each residue in the
1086 * sequence and the element value gives its position in the alignment
1088 * @return int[SequenceI.getEnd()-SequenceI.getStart()+1] or null if no
1089 * residues in SequenceI object
1092 public int[] gapMap()
1094 String seq = jalview.analysis.AlignSeq.extractGaps(
1095 jalview.util.Comparison.GapChars, new String(sequence));
1096 int[] map = new int[seq.length()];
1100 while (j < sequence.length)
1102 if (!jalview.util.Comparison.isGap(sequence[j]))
1114 public int[] findPositionMap()
1116 int map[] = new int[sequence.length];
1119 int seqlen = sequence.length;
1120 while ((j < seqlen))
1123 if (!jalview.util.Comparison.isGap(sequence[j]))
1134 public List<int[]> getInsertions()
1136 ArrayList<int[]> map = new ArrayList<int[]>();
1137 int lastj = -1, j = 0;
1139 int seqlen = sequence.length;
1140 while ((j < seqlen))
1142 if (jalview.util.Comparison.isGap(sequence[j]))
1153 map.add(new int[] { lastj, j - 1 });
1161 map.add(new int[] { lastj, j - 1 });
1168 public void deleteChars(int i, int j)
1170 int newstart = start, newend = end;
1171 if (i >= sequence.length || i < 0)
1176 char[] tmp = StringUtils.deleteChars(sequence, i, j);
1177 boolean createNewDs = false;
1178 // TODO: take a (second look) at the dataset creation validation method for
1179 // the very large sequence case
1180 int eindex = -1, sindex = -1;
1181 boolean ecalc = false, scalc = false;
1182 for (int s = i; s < j; s++)
1184 if (jalview.schemes.ResidueProperties.aaIndex[sequence[s]] != 23)
1194 sindex = findIndex(start) - 1;
1199 // delete characters including start of sequence
1200 newstart = findPosition(j);
1201 break; // don't need to search for any more residue characters.
1205 // delete characters after start.
1208 eindex = findIndex(end) - 1;
1213 // delete characters at end of sequence
1214 newend = findPosition(i - 1);
1215 break; // don't need to search for any more residue characters.
1220 newend--; // decrease end position by one for the deleted residue
1221 // and search further
1227 // deletion occured in the middle of the sequence
1228 if (createNewDs && this.datasetSequence != null)
1230 // construct a new sequence
1231 Sequence ds = new Sequence(datasetSequence);
1232 // TODO: remove any non-inheritable properties ?
1233 // TODO: create a sequence mapping (since there is a relation here ?)
1234 ds.deleteChars(i, j);
1235 datasetSequence = ds;
1244 public void insertCharAt(int i, int length, char c)
1246 char[] tmp = new char[sequence.length + length];
1248 if (i >= sequence.length)
1250 System.arraycopy(sequence, 0, tmp, 0, sequence.length);
1251 i = sequence.length;
1255 System.arraycopy(sequence, 0, tmp, 0, i);
1265 if (i < sequence.length)
1267 System.arraycopy(sequence, i, tmp, index, sequence.length - i);
1275 public void insertCharAt(int i, char c)
1277 insertCharAt(i, 1, c);
1281 public String getVamsasId()
1287 public void setVamsasId(String id)
1293 public void setDBRefs(DBRefEntry[] dbref)
1295 if (dbrefs == null && datasetSequence != null
1296 && this != datasetSequence)
1298 datasetSequence.setDBRefs(dbref);
1304 DBRefUtils.ensurePrimaries(this);
1309 public DBRefEntry[] getDBRefs()
1311 if (dbrefs == null && datasetSequence != null
1312 && this != datasetSequence)
1314 return datasetSequence.getDBRefs();
1320 public void addDBRef(DBRefEntry entry)
1322 if (datasetSequence != null)
1324 datasetSequence.addDBRef(entry);
1330 dbrefs = new DBRefEntry[0];
1333 for (DBRefEntryI dbr : dbrefs)
1335 if (dbr.updateFrom(entry))
1338 * found a dbref that either matched, or could be
1339 * updated from, the new entry - no need to add it
1346 * extend the array to make room for one more
1348 // TODO use an ArrayList instead
1349 int j = dbrefs.length;
1350 DBRefEntry[] temp = new DBRefEntry[j + 1];
1351 System.arraycopy(dbrefs, 0, temp, 0, j);
1352 temp[temp.length - 1] = entry;
1356 DBRefUtils.ensurePrimaries(this);
1360 public void setDatasetSequence(SequenceI seq)
1364 throw new IllegalArgumentException(
1365 "Implementation Error: self reference passed to SequenceI.setDatasetSequence");
1367 if (seq != null && seq.getDatasetSequence() != null)
1369 throw new IllegalArgumentException(
1370 "Implementation error: cascading dataset sequences are not allowed.");
1372 datasetSequence = seq;
1376 public SequenceI getDatasetSequence()
1378 return datasetSequence;
1382 public AlignmentAnnotation[] getAnnotation()
1384 return annotation == null ? null : annotation
1385 .toArray(new AlignmentAnnotation[annotation.size()]);
1389 public boolean hasAnnotation(AlignmentAnnotation ann)
1391 return annotation == null ? false : annotation.contains(ann);
1395 public void addAlignmentAnnotation(AlignmentAnnotation annotation)
1397 if (this.annotation == null)
1399 this.annotation = new Vector<AlignmentAnnotation>();
1401 if (!this.annotation.contains(annotation))
1403 this.annotation.addElement(annotation);
1405 annotation.setSequenceRef(this);
1409 public void removeAlignmentAnnotation(AlignmentAnnotation annotation)
1411 if (this.annotation != null)
1413 this.annotation.removeElement(annotation);
1414 if (this.annotation.size() == 0)
1416 this.annotation = null;
1422 * test if this is a valid candidate for another sequence's dataset sequence.
1425 private boolean isValidDatasetSequence()
1427 if (datasetSequence != null)
1431 for (int i = 0; i < sequence.length; i++)
1433 if (jalview.util.Comparison.isGap(sequence[i]))
1442 public SequenceI deriveSequence()
1444 Sequence seq = null;
1445 if (datasetSequence == null)
1447 if (isValidDatasetSequence())
1449 // Use this as dataset sequence
1450 seq = new Sequence(getName(), "", 1, -1);
1451 seq.setDatasetSequence(this);
1452 seq.initSeqFrom(this, getAnnotation());
1457 // Create a new, valid dataset sequence
1458 createDatasetSequence();
1461 return new Sequence(this);
1464 private boolean _isNa;
1466 private long _seqhash = 0;
1469 * Answers false if the sequence is more than 85% nucleotide (ACGTU), else
1473 public boolean isProtein()
1475 if (datasetSequence != null)
1477 return datasetSequence.isProtein();
1479 if (_seqhash != sequence.hashCode())
1481 _seqhash = sequence.hashCode();
1482 _isNa = Comparison.isNucleotide(this);
1490 * @see jalview.datamodel.SequenceI#createDatasetSequence()
1493 public SequenceI createDatasetSequence()
1495 if (datasetSequence == null)
1497 Sequence dsseq = new Sequence(getName(), AlignSeq.extractGaps(
1498 jalview.util.Comparison.GapChars, getSequenceAsString()),
1499 getStart(), getEnd());
1501 datasetSequence = dsseq;
1503 dsseq.setDescription(description);
1504 // move features and database references onto dataset sequence
1505 dsseq.sequenceFeatureStore = sequenceFeatureStore;
1506 sequenceFeatureStore = null;
1507 dsseq.dbrefs = dbrefs;
1509 // TODO: search and replace any references to this sequence with
1510 // references to the dataset sequence in Mappings on dbref
1511 dsseq.pdbIds = pdbIds;
1513 datasetSequence.updatePDBIds();
1514 if (annotation != null)
1516 // annotation is cloned rather than moved, to preserve what's currently
1518 for (AlignmentAnnotation aa : annotation)
1520 AlignmentAnnotation _aa = new AlignmentAnnotation(aa);
1521 _aa.sequenceRef = datasetSequence;
1522 _aa.adjustForAlignment(); // uses annotation's own record of
1523 // sequence-column mapping
1524 datasetSequence.addAlignmentAnnotation(_aa);
1528 return datasetSequence;
1535 * jalview.datamodel.SequenceI#setAlignmentAnnotation(AlignmmentAnnotation[]
1539 public void setAlignmentAnnotation(AlignmentAnnotation[] annotations)
1541 if (annotation != null)
1543 annotation.removeAllElements();
1545 if (annotations != null)
1547 for (int i = 0; i < annotations.length; i++)
1549 if (annotations[i] != null)
1551 addAlignmentAnnotation(annotations[i]);
1558 public AlignmentAnnotation[] getAnnotation(String label)
1560 if (annotation == null || annotation.size() == 0)
1565 Vector<AlignmentAnnotation> subset = new Vector<AlignmentAnnotation>();
1566 Enumeration<AlignmentAnnotation> e = annotation.elements();
1567 while (e.hasMoreElements())
1569 AlignmentAnnotation ann = e.nextElement();
1570 if (ann.label != null && ann.label.equals(label))
1572 subset.addElement(ann);
1575 if (subset.size() == 0)
1579 AlignmentAnnotation[] anns = new AlignmentAnnotation[subset.size()];
1581 e = subset.elements();
1582 while (e.hasMoreElements())
1584 anns[i++] = e.nextElement();
1586 subset.removeAllElements();
1591 public boolean updatePDBIds()
1593 if (datasetSequence != null)
1595 // TODO: could merge DBRefs
1596 return datasetSequence.updatePDBIds();
1598 if (dbrefs == null || dbrefs.length == 0)
1602 boolean added = false;
1603 for (DBRefEntry dbr : dbrefs)
1605 if (DBRefSource.PDB.equals(dbr.getSource()))
1608 * 'Add' any PDB dbrefs as a PDBEntry - add is only performed if the
1609 * PDB id is not already present in a 'matching' PDBEntry
1610 * Constructor parses out a chain code if appended to the accession id
1611 * (a fudge used to 'store' the chain code in the DBRef)
1613 PDBEntry pdbe = new PDBEntry(dbr);
1614 added |= addPDBId(pdbe);
1621 public void transferAnnotation(SequenceI entry, Mapping mp)
1623 if (datasetSequence != null)
1625 datasetSequence.transferAnnotation(entry, mp);
1628 if (entry.getDatasetSequence() != null)
1630 transferAnnotation(entry.getDatasetSequence(), mp);
1633 // transfer any new features from entry onto sequence
1634 if (entry.getSequenceFeatures() != null)
1637 List<SequenceFeature> sfs = entry.getSequenceFeatures();
1638 for (SequenceFeature feature : sfs)
1640 SequenceFeature sf[] = (mp != null) ? mp.locateFeature(feature)
1641 : new SequenceFeature[] { new SequenceFeature(feature) };
1644 for (int sfi = 0; sfi < sf.length; sfi++)
1646 addSequenceFeature(sf[sfi]);
1652 // transfer PDB entries
1653 if (entry.getAllPDBEntries() != null)
1655 Enumeration<PDBEntry> e = entry.getAllPDBEntries().elements();
1656 while (e.hasMoreElements())
1658 PDBEntry pdb = e.nextElement();
1662 // transfer database references
1663 DBRefEntry[] entryRefs = entry.getDBRefs();
1664 if (entryRefs != null)
1666 for (int r = 0; r < entryRefs.length; r++)
1668 DBRefEntry newref = new DBRefEntry(entryRefs[r]);
1669 if (newref.getMap() != null && mp != null)
1671 // remap ref using our local mapping
1673 // we also assume all version string setting is done by dbSourceProxy
1675 * if (!newref.getSource().equalsIgnoreCase(dbSource)) {
1676 * newref.setSource(dbSource); }
1684 * @return The index (zero-based) on this sequence in the MSA. It returns
1685 * {@code -1} if this information is not available.
1688 public int getIndex()
1694 * Defines the position of this sequence in the MSA. Use the value {@code -1}
1695 * if this information is undefined.
1698 * position for this sequence. This value is zero-based (zero for
1699 * this first sequence)
1702 public void setIndex(int value)
1708 public void setRNA(RNA r)
1720 public List<AlignmentAnnotation> getAlignmentAnnotations(String calcId,
1723 List<AlignmentAnnotation> result = new ArrayList<AlignmentAnnotation>();
1724 if (this.annotation != null)
1726 for (AlignmentAnnotation ann : annotation)
1728 if (ann.calcId != null && ann.calcId.equals(calcId)
1729 && ann.label != null && ann.label.equals(label))
1739 public String toString()
1741 return getDisplayId(false);
1745 public PDBEntry getPDBEntry(String pdbIdStr)
1747 if (getDatasetSequence() != null)
1749 return getDatasetSequence().getPDBEntry(pdbIdStr);
1755 List<PDBEntry> entries = getAllPDBEntries();
1756 for (PDBEntry entry : entries)
1758 if (entry.getId().equalsIgnoreCase(pdbIdStr))
1767 public List<DBRefEntry> getPrimaryDBRefs()
1769 if (datasetSequence != null)
1771 return datasetSequence.getPrimaryDBRefs();
1773 if (dbrefs == null || dbrefs.length == 0)
1775 return Collections.emptyList();
1777 synchronized (dbrefs)
1779 List<DBRefEntry> primaries = new ArrayList<DBRefEntry>();
1780 DBRefEntry[] tmp = new DBRefEntry[1];
1781 for (DBRefEntry ref : dbrefs)
1783 if (!ref.isPrimaryCandidate())
1789 MapList mp = ref.getMap().getMap();
1790 if (mp.getFromLowest() > start || mp.getFromHighest() < end)
1792 // map only involves a subsequence, so cannot be primary
1796 // whilst it looks like it is a primary ref, we also sanity check type
1797 if (DBRefUtils.getCanonicalName(DBRefSource.PDB).equals(
1798 DBRefUtils.getCanonicalName(ref.getSource())))
1800 // PDB dbrefs imply there should be a PDBEntry associated
1801 // TODO: tighten PDB dbrefs
1802 // formally imply Jalview has actually downloaded and
1803 // parsed the pdb file. That means there should be a cached file
1804 // handle on the PDBEntry, and a real mapping between sequence and
1805 // extracted sequence from PDB file
1806 PDBEntry pdbentry = getPDBEntry(ref.getAccessionId());
1807 if (pdbentry != null && pdbentry.getFile() != null)
1813 // check standard protein or dna sources
1815 DBRefEntry[] res = DBRefUtils.selectDbRefs(!isProtein(), tmp);
1816 if (res != null && res[0] == tmp[0])
1830 public List<SequenceFeature> findFeatures(int from, int to,
1833 if (datasetSequence != null)
1835 return datasetSequence.findFeatures(from, to, types);
1837 return sequenceFeatureStore.findFeatures(from, to, types);
1841 * Invalidates any stale cursors (forcing recalculation) by incrementing the
1842 * token that has to match the one presented by the cursor
1845 public void sequenceChanged()