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.BitSet;
35 import java.util.Collections;
36 import java.util.Enumeration;
37 import java.util.Iterator;
38 import java.util.List;
39 import java.util.ListIterator;
40 import java.util.Vector;
42 import fr.orsay.lri.varna.models.rna.RNA;
46 * Implements the SequenceI interface for a char[] based sequence object.
51 public class Sequence extends ASequence implements SequenceI
53 SequenceI datasetSequence;
57 private char[] sequence;
65 Vector<PDBEntry> pdbIds;
74 * This annotation is displayed below the alignment but the positions are tied
75 * to the residues of this sequence
77 * TODO: change to List<>
79 Vector<AlignmentAnnotation> annotation;
81 private SequenceFeaturesI sequenceFeatureStore;
84 * A cursor holding the approximate current view position to the sequence,
85 * as determined by findIndex or findPosition or findPositions.
86 * Using a cursor as a hint allows these methods to be more performant for
89 private SequenceCursor cursor;
92 * A number that should be incremented whenever the sequence is edited.
93 * If the value matches the cursor token, then we can trust the cursor,
94 * if not then it should be recomputed.
96 private int changeCount;
99 * Creates a new Sequence object.
102 * display name string
104 * string to form a possibly gapped sequence out of
106 * first position of non-gap residue in the sequence
108 * last position of ungapped residues (nearly always only used for
111 public Sequence(String name, String sequence, int start, int end)
114 initSeqAndName(name, sequence.toCharArray(), start, end);
117 public Sequence(String name, char[] sequence, int start, int end)
120 initSeqAndName(name, sequence, start, end);
124 * Stage 1 constructor - assign name, sequence, and set start and end fields.
125 * start and end are updated values from name2 if it ends with /start-end
132 protected void initSeqAndName(String name2, char[] sequence2, int start2,
136 this.sequence = sequence2;
144 * If 'name' ends in /i-j, where i >= j > 0 are integers, extracts i and j as
145 * start and end respectively and removes the suffix from the name
152 "POSSIBLE IMPLEMENTATION ERROR: null sequence name passed to constructor.");
155 int slashPos = name.lastIndexOf('/');
156 if (slashPos > -1 && slashPos < name.length() - 1)
158 String suffix = name.substring(slashPos + 1);
159 String[] range = suffix.split("-");
160 if (range.length == 2)
164 int from = Integer.valueOf(range[0]);
165 int to = Integer.valueOf(range[1]);
166 if (from > 0 && to >= from)
168 name = name.substring(0, slashPos);
173 } catch (NumberFormatException e)
175 // leave name unchanged if suffix is invalid
182 * Ensures that 'end' is not before the end of the sequence, that is,
183 * (end-start+1) is at least as long as the count of ungapped positions. Note
184 * that end is permitted to be beyond the end of the sequence data.
186 void checkValidRange()
189 // http://issues.jalview.org/browse/JAL-774?focusedCommentId=11239&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-11239
192 for (int j = 0; j < sequence.length; j++)
194 if (!Comparison.isGap(sequence[j]))
213 * default constructor
217 sequenceFeatureStore = new SequenceFeatures();
221 * Creates a new Sequence object.
228 public Sequence(String name, String sequence)
230 this(name, sequence, 1, -1);
234 * Creates a new Sequence object with new AlignmentAnnotations but inherits
235 * any existing dataset sequence reference. If non exists, everything is
239 * if seq is a dataset sequence, behaves like a plain old copy
242 public Sequence(SequenceI seq)
244 this(seq, seq.getAnnotation());
248 * Create a new sequence object with new features, DBRefEntries, and PDBIds
249 * but inherits any existing dataset sequence reference, and duplicate of any
250 * annotation that is present in the given annotation array.
253 * the sequence to be copied
254 * @param alAnnotation
255 * an array of annotation including some associated with seq
257 public Sequence(SequenceI seq, AlignmentAnnotation[] alAnnotation)
260 initSeqFrom(seq, alAnnotation);
264 * does the heavy lifting when cloning a dataset sequence, or coping data from
265 * dataset to a new derived sequence.
268 * - source of attributes.
269 * @param alAnnotation
270 * - alignment annotation present on seq that should be copied onto
273 protected void initSeqFrom(SequenceI seq,
274 AlignmentAnnotation[] alAnnotation)
276 char[] oseq = seq.getSequence(); // returns a copy of the array
277 initSeqAndName(seq.getName(), oseq, seq.getStart(), seq.getEnd());
279 description = seq.getDescription();
280 if (seq != datasetSequence)
282 setDatasetSequence(seq.getDatasetSequence());
286 * only copy DBRefs and seqfeatures if we really are a dataset sequence
288 if (datasetSequence == null)
290 if (seq.getDBRefs() != null)
292 DBRefEntry[] dbr = seq.getDBRefs();
293 for (int i = 0; i < dbr.length; i++)
295 addDBRef(new DBRefEntry(dbr[i]));
300 * make copies of any sequence features
302 for (SequenceFeature sf : seq.getSequenceFeatures())
304 addSequenceFeature(new SequenceFeature(sf));
308 if (seq.getAnnotation() != null)
310 AlignmentAnnotation[] sqann = seq.getAnnotation();
311 for (int i = 0; i < sqann.length; i++)
313 if (sqann[i] == null)
317 boolean found = (alAnnotation == null);
320 for (int apos = 0; !found && apos < alAnnotation.length; apos++)
322 found = (alAnnotation[apos] == sqann[i]);
327 // only copy the given annotation
328 AlignmentAnnotation newann = new AlignmentAnnotation(sqann[i]);
329 addAlignmentAnnotation(newann);
333 if (seq.getAllPDBEntries() != null)
335 Vector<PDBEntry> ids = seq.getAllPDBEntries();
336 for (PDBEntry pdb : ids)
338 this.addPDBId(new PDBEntry(pdb));
344 public void setSequenceFeatures(List<SequenceFeature> features)
346 if (datasetSequence != null)
348 datasetSequence.setSequenceFeatures(features);
351 sequenceFeatureStore = new SequenceFeatures(features);
355 public synchronized boolean addSequenceFeature(SequenceFeature sf)
357 if (sf.getType() == null)
359 System.err.println("SequenceFeature type may not be null: "
364 if (datasetSequence != null)
366 return datasetSequence.addSequenceFeature(sf);
369 return sequenceFeatureStore.add(sf);
373 public void deleteFeature(SequenceFeature sf)
375 if (datasetSequence != null)
377 datasetSequence.deleteFeature(sf);
381 sequenceFeatureStore.delete(sf);
391 public List<SequenceFeature> getSequenceFeatures()
393 if (datasetSequence != null)
395 return datasetSequence.getSequenceFeatures();
397 return sequenceFeatureStore.getAllFeatures();
401 public SequenceFeaturesI getFeatures()
403 return datasetSequence != null ? datasetSequence.getFeatures()
404 : sequenceFeatureStore;
408 public boolean addPDBId(PDBEntry entry)
412 pdbIds = new Vector<>();
417 for (PDBEntry pdbe : pdbIds)
419 if (pdbe.updateFrom(entry))
424 pdbIds.addElement(entry);
435 public void setPDBId(Vector<PDBEntry> id)
443 * @return DOCUMENT ME!
446 public Vector<PDBEntry> getAllPDBEntries()
448 return pdbIds == null ? new Vector<PDBEntry>() : pdbIds;
454 * @return DOCUMENT ME!
457 public String getDisplayId(boolean jvsuffix)
459 StringBuffer result = new StringBuffer(name);
462 result.append("/" + start + "-" + end);
465 return result.toString();
469 * Sets the sequence name. If the name ends in /start-end, then the start-end
470 * values are parsed out and set, and the suffix is removed from the name.
475 public void setName(String theName)
484 * @return DOCUMENT ME!
487 public String getName()
499 public void setStart(int start)
507 * @return DOCUMENT ME!
510 public int getStart()
522 public void setEnd(int end)
530 * @return DOCUMENT ME!
541 * @return DOCUMENT ME!
544 public int getLength()
546 return this.sequence.length;
556 public void setSequence(String seq)
558 this.sequence = seq.toCharArray();
564 public String getSequenceAsString()
566 return new String(sequence);
570 public String getSequenceAsString(int start, int end)
572 return new String(getSequence(start, end));
576 public char[] getSequence()
579 return sequence == null ? null : Arrays.copyOf(sequence,
586 * @see jalview.datamodel.SequenceI#getSequence(int, int)
589 public char[] getSequence(int start, int end)
595 // JBPNote - left to user to pad the result here (TODO:Decide on this
597 if (start >= sequence.length)
602 if (end >= sequence.length)
604 end = sequence.length;
607 char[] reply = new char[end - start];
608 System.arraycopy(sequence, start, reply, 0, end - start);
614 public SequenceI getSubSequence(int start, int end)
620 char[] seq = getSequence(start, end);
625 int nstart = findPosition(start);
626 int nend = findPosition(end) - 1;
627 // JBPNote - this is an incomplete copy.
628 SequenceI nseq = new Sequence(this.getName(), seq, nstart, nend);
629 nseq.setDescription(description);
630 if (datasetSequence != null)
632 nseq.setDatasetSequence(datasetSequence);
636 nseq.setDatasetSequence(this);
642 * Returns the character of the aligned sequence at the given position (base
643 * zero), or space if the position is not within the sequence's bounds
648 public char getCharAt(int i)
650 if (i >= 0 && i < sequence.length)
667 public void setDescription(String desc)
669 this.description = desc;
675 * @return DOCUMENT ME!
678 public String getDescription()
680 return this.description;
687 public int findIndex(int pos)
690 * use a valid, hopefully nearby, cursor if available
692 if (isValidCursor(cursor))
694 return findIndex(pos, cursor);
702 * traverse sequence from the start counting gaps; make a note of
703 * the column of the first residue to save in the cursor
705 while ((i < sequence.length) && (j <= end) && (j <= pos))
707 if (!Comparison.isGap(sequence[i]))
718 if (j == end && j < pos)
723 updateCursor(pos, i, startColumn);
728 * Updates the cursor to the latest found residue and column position
735 * column position of the first sequence residue
737 protected void updateCursor(int residuePos, int column, int startColumn)
740 * preserve end residue column provided cursor was valid
742 int endColumn = isValidCursor(cursor) ? cursor.lastColumnPosition : 0;
743 if (residuePos == this.end)
748 cursor = new SequenceCursor(this, residuePos, column, startColumn,
749 endColumn, this.changeCount);
753 * Answers the aligned column position (1..) for the given residue position
754 * (start..) given a 'hint' of a residue/column location in the neighbourhood.
755 * The hint may be left of, at, or to the right of the required position.
761 protected int findIndex(int pos, SequenceCursor curs)
763 if (!isValidCursor(curs))
766 * wrong or invalidated cursor, compute de novo
768 return findIndex(pos);
771 if (curs.residuePosition == pos)
773 return curs.columnPosition;
777 * move left or right to find pos from hint.position
779 int col = curs.columnPosition - 1; // convert from base 1 to 0-based array
781 int newPos = curs.residuePosition;
782 int delta = newPos > pos ? -1 : 1;
784 while (newPos != pos)
786 col += delta; // shift one column left or right
787 if (col < 0 || col == sequence.length)
791 if (!Comparison.isGap(sequence[col]))
797 col++; // convert back to base 1
798 updateCursor(pos, col, curs.firstColumnPosition);
807 public int findPosition(final int column)
810 * use a valid, hopefully nearby, cursor if available
812 if (isValidCursor(cursor))
814 return findPosition(column + 1, cursor);
817 // TODO recode this more naturally i.e. count residues only
818 // as they are found, not 'in anticipation'
821 * traverse the sequence counting gaps; note the column position
822 * of the first residue, to save in the cursor
824 int firstResidueColumn = 0;
825 int lastPosFound = 0;
826 int lastPosFoundColumn = 0;
827 int seqlen = sequence.length;
829 if (seqlen > 0 && !Comparison.isGap(sequence[0]))
831 lastPosFound = start;
832 lastPosFoundColumn = 0;
838 while (j < column && j < seqlen)
840 if (!Comparison.isGap(sequence[j]))
843 lastPosFoundColumn = j;
844 if (pos == this.start)
846 firstResidueColumn = j;
852 if (j < seqlen && !Comparison.isGap(sequence[j]))
855 lastPosFoundColumn = j;
856 if (pos == this.start)
858 firstResidueColumn = j;
863 * update the cursor to the last residue position found (if any)
864 * (converting column position to base 1)
866 if (lastPosFound != 0)
868 updateCursor(lastPosFound, lastPosFoundColumn + 1,
869 firstResidueColumn + 1);
876 * Answers true if the given cursor is not null, is for this sequence object,
877 * and has a token value that matches this object's changeCount, else false.
878 * This allows us to ignore a cursor as 'stale' if the sequence has been
879 * modified since the cursor was created.
884 protected boolean isValidCursor(SequenceCursor curs)
886 if (curs == null || curs.sequence != this || curs.token != changeCount)
891 * sanity check against range
893 if (curs.columnPosition < 0 || curs.columnPosition > sequence.length)
897 if (curs.residuePosition < start || curs.residuePosition > end)
905 * Answers the sequence position (start..) for the given aligned column
906 * position (1..), given a hint of a cursor in the neighbourhood. The cursor
907 * may lie left of, at, or to the right of the column position.
913 protected int findPosition(final int col, SequenceCursor curs)
915 if (!isValidCursor(curs))
918 * wrong or invalidated cursor, compute de novo
920 return findPosition(col - 1);// ugh back to base 0
923 if (curs.columnPosition == col)
925 cursor = curs; // in case this method becomes public
926 return curs.residuePosition; // easy case :-)
929 if (curs.lastColumnPosition > 0 && curs.lastColumnPosition < col)
932 * sequence lies entirely to the left of col
933 * - return last residue + 1
938 if (curs.firstColumnPosition > 0 && curs.firstColumnPosition > col)
941 * sequence lies entirely to the right of col
942 * - return first residue
947 // todo could choose closest to col out of column,
948 // firstColumnPosition, lastColumnPosition as a start point
951 * move left or right to find pos from cursor position
953 int firstResidueColumn = curs.firstColumnPosition;
954 int column = curs.columnPosition - 1; // to base 0
955 int newPos = curs.residuePosition;
956 int delta = curs.columnPosition > col ? -1 : 1;
957 boolean gapped = false;
958 int lastFoundPosition = curs.residuePosition;
959 int lastFoundPositionColumn = curs.columnPosition;
961 while (column != col - 1)
963 column += delta; // shift one column left or right
964 if (column < 0 || column == sequence.length)
968 gapped = Comparison.isGap(sequence[column]);
972 lastFoundPosition = newPos;
973 lastFoundPositionColumn = column + 1;
974 if (lastFoundPosition == this.start)
976 firstResidueColumn = column + 1;
981 if (cursor == null || lastFoundPosition != cursor.residuePosition)
983 updateCursor(lastFoundPosition, lastFoundPositionColumn,
988 * hack to give position to the right if on a gap
989 * or beyond the length of the sequence (see JAL-2562)
991 if (delta > 0 && (gapped || column >= sequence.length))
1003 public Range findPositions(int fromColumn, int toColumn)
1005 if (toColumn < fromColumn || fromColumn < 1)
1011 * find the first non-gapped position, if any
1013 int firstPosition = 0;
1014 int col = fromColumn - 1;
1015 int length = sequence.length;
1016 while (col < length && col < toColumn)
1018 if (!Comparison.isGap(sequence[col]))
1020 firstPosition = findPosition(col++);
1026 if (firstPosition == 0)
1032 * find the last non-gapped position
1034 int lastPosition = firstPosition;
1035 while (col < length && col < toColumn)
1037 if (!Comparison.isGap(sequence[col++]))
1043 return new Range(firstPosition, lastPosition);
1047 * Returns an int array where indices correspond to each residue in the
1048 * sequence and the element value gives its position in the alignment
1050 * @return int[SequenceI.getEnd()-SequenceI.getStart()+1] or null if no
1051 * residues in SequenceI object
1054 public int[] gapMap()
1056 String seq = jalview.analysis.AlignSeq.extractGaps(
1057 jalview.util.Comparison.GapChars, new String(sequence));
1058 int[] map = new int[seq.length()];
1062 while (j < sequence.length)
1064 if (!jalview.util.Comparison.isGap(sequence[j]))
1076 * Build a bitset corresponding to sequence gaps
1078 * @return a BitSet where set values correspond to gaps in the sequence
1081 public BitSet gapBitset()
1083 BitSet gaps = new BitSet(sequence.length);
1085 while (j < sequence.length)
1087 if (jalview.util.Comparison.isGap(sequence[j]))
1097 public int[] findPositionMap()
1099 int map[] = new int[sequence.length];
1102 int seqlen = sequence.length;
1103 while ((j < seqlen))
1106 if (!jalview.util.Comparison.isGap(sequence[j]))
1117 public List<int[]> getInsertions()
1119 ArrayList<int[]> map = new ArrayList<>();
1120 int lastj = -1, j = 0;
1122 int seqlen = sequence.length;
1123 while ((j < seqlen))
1125 if (jalview.util.Comparison.isGap(sequence[j]))
1136 map.add(new int[] { lastj, j - 1 });
1144 map.add(new int[] { lastj, j - 1 });
1151 public BitSet getInsertionsAsBits()
1153 BitSet map = new BitSet();
1154 int lastj = -1, j = 0;
1156 int seqlen = sequence.length;
1157 while ((j < seqlen))
1159 if (jalview.util.Comparison.isGap(sequence[j]))
1185 public void deleteChars(final int i, final int j)
1187 int newstart = start, newend = end;
1188 if (i >= sequence.length || i < 0)
1193 char[] tmp = StringUtils.deleteChars(sequence, i, j);
1194 boolean createNewDs = false;
1195 // TODO: take a (second look) at the dataset creation validation method for
1196 // the very large sequence case
1197 int startIndex = findIndex(start) - 1;
1198 int endIndex = findIndex(end) - 1;
1199 int startDeleteColumn = -1; // for dataset sequence deletions
1200 int deleteCount = 0;
1202 for (int s = i; s < j; s++)
1204 if (Comparison.isGap(sequence[s]))
1209 if (startDeleteColumn == -1)
1211 startDeleteColumn = findPosition(s) - start;
1219 if (startIndex == s)
1222 * deleting characters from start of sequence; new start is the
1223 * sequence position of the next column (position to the right
1224 * if the column position is gapped)
1226 newstart = findPosition(j);
1234 * deleting characters at end of sequence; new end is the sequence
1235 * position of the column before the deletion; subtract 1 if this is
1236 * gapped since findPosition returns the next sequence position
1238 newend = findPosition(i - 1);
1239 if (Comparison.isGap(sequence[i - 1]))
1254 if (createNewDs && this.datasetSequence != null)
1257 * if deletion occured in the middle of the sequence,
1258 * construct a new dataset sequence and delete the residues
1259 * that were deleted from the aligned sequence
1261 Sequence ds = new Sequence(datasetSequence);
1262 ds.deleteChars(startDeleteColumn, startDeleteColumn + deleteCount);
1263 datasetSequence = ds;
1264 // TODO: remove any non-inheritable properties ?
1265 // TODO: create a sequence mapping (since there is a relation here ?)
1274 public void insertCharAt(int i, int length, char c)
1276 char[] tmp = new char[sequence.length + length];
1278 if (i >= sequence.length)
1280 System.arraycopy(sequence, 0, tmp, 0, sequence.length);
1281 i = sequence.length;
1285 System.arraycopy(sequence, 0, tmp, 0, i);
1295 if (i < sequence.length)
1297 System.arraycopy(sequence, i, tmp, index, sequence.length - i);
1305 public void insertCharAt(int i, char c)
1307 insertCharAt(i, 1, c);
1311 public String getVamsasId()
1317 public void setVamsasId(String id)
1323 public void setDBRefs(DBRefEntry[] dbref)
1325 if (dbrefs == null && datasetSequence != null
1326 && this != datasetSequence)
1328 datasetSequence.setDBRefs(dbref);
1334 DBRefUtils.ensurePrimaries(this);
1339 public DBRefEntry[] getDBRefs()
1341 if (dbrefs == null && datasetSequence != null
1342 && this != datasetSequence)
1344 return datasetSequence.getDBRefs();
1350 public void addDBRef(DBRefEntry entry)
1352 if (datasetSequence != null)
1354 datasetSequence.addDBRef(entry);
1360 dbrefs = new DBRefEntry[0];
1363 for (DBRefEntryI dbr : dbrefs)
1365 if (dbr.updateFrom(entry))
1368 * found a dbref that either matched, or could be
1369 * updated from, the new entry - no need to add it
1376 * extend the array to make room for one more
1378 // TODO use an ArrayList instead
1379 int j = dbrefs.length;
1380 DBRefEntry[] temp = new DBRefEntry[j + 1];
1381 System.arraycopy(dbrefs, 0, temp, 0, j);
1382 temp[temp.length - 1] = entry;
1386 DBRefUtils.ensurePrimaries(this);
1390 public void setDatasetSequence(SequenceI seq)
1394 throw new IllegalArgumentException(
1395 "Implementation Error: self reference passed to SequenceI.setDatasetSequence");
1397 if (seq != null && seq.getDatasetSequence() != null)
1399 throw new IllegalArgumentException(
1400 "Implementation error: cascading dataset sequences are not allowed.");
1402 datasetSequence = seq;
1406 public SequenceI getDatasetSequence()
1408 return datasetSequence;
1412 public AlignmentAnnotation[] getAnnotation()
1414 return annotation == null ? null
1416 .toArray(new AlignmentAnnotation[annotation.size()]);
1420 public boolean hasAnnotation(AlignmentAnnotation ann)
1422 return annotation == null ? false : annotation.contains(ann);
1426 public void addAlignmentAnnotation(AlignmentAnnotation annotation)
1428 if (this.annotation == null)
1430 this.annotation = new Vector<>();
1432 if (!this.annotation.contains(annotation))
1434 this.annotation.addElement(annotation);
1436 annotation.setSequenceRef(this);
1440 public void removeAlignmentAnnotation(AlignmentAnnotation annotation)
1442 if (this.annotation != null)
1444 this.annotation.removeElement(annotation);
1445 if (this.annotation.size() == 0)
1447 this.annotation = null;
1453 * test if this is a valid candidate for another sequence's dataset sequence.
1456 private boolean isValidDatasetSequence()
1458 if (datasetSequence != null)
1462 for (int i = 0; i < sequence.length; i++)
1464 if (jalview.util.Comparison.isGap(sequence[i]))
1473 public SequenceI deriveSequence()
1475 Sequence seq = null;
1476 if (datasetSequence == null)
1478 if (isValidDatasetSequence())
1480 // Use this as dataset sequence
1481 seq = new Sequence(getName(), "", 1, -1);
1482 seq.setDatasetSequence(this);
1483 seq.initSeqFrom(this, getAnnotation());
1488 // Create a new, valid dataset sequence
1489 createDatasetSequence();
1492 return new Sequence(this);
1495 private boolean _isNa;
1497 private int _seqhash = 0;
1500 * Answers false if the sequence is more than 85% nucleotide (ACGTU), else
1504 public boolean isProtein()
1506 if (datasetSequence != null)
1508 return datasetSequence.isProtein();
1510 if (_seqhash != sequence.hashCode())
1512 _seqhash = sequence.hashCode();
1513 _isNa = Comparison.isNucleotide(this);
1521 * @see jalview.datamodel.SequenceI#createDatasetSequence()
1524 public SequenceI createDatasetSequence()
1526 if (datasetSequence == null)
1528 Sequence dsseq = new Sequence(getName(),
1529 AlignSeq.extractGaps(jalview.util.Comparison.GapChars,
1530 getSequenceAsString()),
1531 getStart(), getEnd());
1533 datasetSequence = dsseq;
1535 dsseq.setDescription(description);
1536 // move features and database references onto dataset sequence
1537 dsseq.sequenceFeatureStore = sequenceFeatureStore;
1538 sequenceFeatureStore = null;
1539 dsseq.dbrefs = dbrefs;
1541 // TODO: search and replace any references to this sequence with
1542 // references to the dataset sequence in Mappings on dbref
1543 dsseq.pdbIds = pdbIds;
1545 datasetSequence.updatePDBIds();
1546 if (annotation != null)
1548 // annotation is cloned rather than moved, to preserve what's currently
1550 for (AlignmentAnnotation aa : annotation)
1552 AlignmentAnnotation _aa = new AlignmentAnnotation(aa);
1553 _aa.sequenceRef = datasetSequence;
1554 _aa.adjustForAlignment(); // uses annotation's own record of
1555 // sequence-column mapping
1556 datasetSequence.addAlignmentAnnotation(_aa);
1560 return datasetSequence;
1567 * jalview.datamodel.SequenceI#setAlignmentAnnotation(AlignmmentAnnotation[]
1571 public void setAlignmentAnnotation(AlignmentAnnotation[] annotations)
1573 if (annotation != null)
1575 annotation.removeAllElements();
1577 if (annotations != null)
1579 for (int i = 0; i < annotations.length; i++)
1581 if (annotations[i] != null)
1583 addAlignmentAnnotation(annotations[i]);
1590 public AlignmentAnnotation[] getAnnotation(String label)
1592 if (annotation == null || annotation.size() == 0)
1597 Vector<AlignmentAnnotation> subset = new Vector<>();
1598 Enumeration<AlignmentAnnotation> e = annotation.elements();
1599 while (e.hasMoreElements())
1601 AlignmentAnnotation ann = e.nextElement();
1602 if (ann.label != null && ann.label.equals(label))
1604 subset.addElement(ann);
1607 if (subset.size() == 0)
1611 AlignmentAnnotation[] anns = new AlignmentAnnotation[subset.size()];
1613 e = subset.elements();
1614 while (e.hasMoreElements())
1616 anns[i++] = e.nextElement();
1618 subset.removeAllElements();
1623 public boolean updatePDBIds()
1625 if (datasetSequence != null)
1627 // TODO: could merge DBRefs
1628 return datasetSequence.updatePDBIds();
1630 if (dbrefs == null || dbrefs.length == 0)
1634 boolean added = false;
1635 for (DBRefEntry dbr : dbrefs)
1637 if (DBRefSource.PDB.equals(dbr.getSource()))
1640 * 'Add' any PDB dbrefs as a PDBEntry - add is only performed if the
1641 * PDB id is not already present in a 'matching' PDBEntry
1642 * Constructor parses out a chain code if appended to the accession id
1643 * (a fudge used to 'store' the chain code in the DBRef)
1645 PDBEntry pdbe = new PDBEntry(dbr);
1646 added |= addPDBId(pdbe);
1653 public void transferAnnotation(SequenceI entry, Mapping mp)
1655 if (datasetSequence != null)
1657 datasetSequence.transferAnnotation(entry, mp);
1660 if (entry.getDatasetSequence() != null)
1662 transferAnnotation(entry.getDatasetSequence(), mp);
1665 // transfer any new features from entry onto sequence
1666 if (entry.getSequenceFeatures() != null)
1669 List<SequenceFeature> sfs = entry.getSequenceFeatures();
1670 for (SequenceFeature feature : sfs)
1672 SequenceFeature sf[] = (mp != null) ? mp.locateFeature(feature)
1673 : new SequenceFeature[] { new SequenceFeature(feature) };
1676 for (int sfi = 0; sfi < sf.length; sfi++)
1678 addSequenceFeature(sf[sfi]);
1684 // transfer PDB entries
1685 if (entry.getAllPDBEntries() != null)
1687 Enumeration<PDBEntry> e = entry.getAllPDBEntries().elements();
1688 while (e.hasMoreElements())
1690 PDBEntry pdb = e.nextElement();
1694 // transfer database references
1695 DBRefEntry[] entryRefs = entry.getDBRefs();
1696 if (entryRefs != null)
1698 for (int r = 0; r < entryRefs.length; r++)
1700 DBRefEntry newref = new DBRefEntry(entryRefs[r]);
1701 if (newref.getMap() != null && mp != null)
1703 // remap ref using our local mapping
1705 // we also assume all version string setting is done by dbSourceProxy
1707 * if (!newref.getSource().equalsIgnoreCase(dbSource)) {
1708 * newref.setSource(dbSource); }
1716 public void setRNA(RNA r)
1728 public List<AlignmentAnnotation> getAlignmentAnnotations(String calcId,
1731 List<AlignmentAnnotation> result = new ArrayList<>();
1732 if (this.annotation != null)
1734 for (AlignmentAnnotation ann : annotation)
1736 if (ann.calcId != null && ann.calcId.equals(calcId)
1737 && ann.label != null && ann.label.equals(label))
1747 public String toString()
1749 return getDisplayId(false);
1753 public PDBEntry getPDBEntry(String pdbIdStr)
1755 if (getDatasetSequence() != null)
1757 return getDatasetSequence().getPDBEntry(pdbIdStr);
1763 List<PDBEntry> entries = getAllPDBEntries();
1764 for (PDBEntry entry : entries)
1766 if (entry.getId().equalsIgnoreCase(pdbIdStr))
1775 public List<DBRefEntry> getPrimaryDBRefs()
1777 if (datasetSequence != null)
1779 return datasetSequence.getPrimaryDBRefs();
1781 if (dbrefs == null || dbrefs.length == 0)
1783 return Collections.emptyList();
1785 synchronized (dbrefs)
1787 List<DBRefEntry> primaries = new ArrayList<>();
1788 DBRefEntry[] tmp = new DBRefEntry[1];
1789 for (DBRefEntry ref : dbrefs)
1791 if (!ref.isPrimaryCandidate())
1797 MapList mp = ref.getMap().getMap();
1798 if (mp.getFromLowest() > start || mp.getFromHighest() < end)
1800 // map only involves a subsequence, so cannot be primary
1804 // whilst it looks like it is a primary ref, we also sanity check type
1805 if (DBRefUtils.getCanonicalName(DBRefSource.PDB)
1806 .equals(DBRefUtils.getCanonicalName(ref.getSource())))
1808 // PDB dbrefs imply there should be a PDBEntry associated
1809 // TODO: tighten PDB dbrefs
1810 // formally imply Jalview has actually downloaded and
1811 // parsed the pdb file. That means there should be a cached file
1812 // handle on the PDBEntry, and a real mapping between sequence and
1813 // extracted sequence from PDB file
1814 PDBEntry pdbentry = getPDBEntry(ref.getAccessionId());
1815 if (pdbentry != null && pdbentry.getFile() != null)
1821 // check standard protein or dna sources
1823 DBRefEntry[] res = DBRefUtils.selectDbRefs(!isProtein(), tmp);
1824 if (res != null && res[0] == tmp[0])
1838 public List<SequenceFeature> findFeatures(int fromColumn, int toColumn,
1841 int startPos = findPosition(fromColumn - 1); // convert base 1 to base 0
1842 int endPos = fromColumn == toColumn ? startPos
1843 : findPosition(toColumn - 1);
1845 List<SequenceFeature> result = getFeatures().findFeatures(startPos,
1849 * if end column is gapped, endPos may be to the right,
1850 * and we may have included adjacent or enclosing features;
1851 * remove any that are not enclosing, non-contact features
1853 boolean endColumnIsGapped = toColumn > 0 && toColumn <= sequence.length
1854 && Comparison.isGap(sequence[toColumn - 1]);
1855 if (endPos > this.end || endColumnIsGapped)
1857 ListIterator<SequenceFeature> it = result.listIterator();
1858 while (it.hasNext())
1860 SequenceFeature sf = it.next();
1861 int sfBegin = sf.getBegin();
1862 int sfEnd = sf.getEnd();
1863 int featureStartColumn = findIndex(sfBegin);
1864 if (featureStartColumn > toColumn)
1868 else if (featureStartColumn < fromColumn)
1870 int featureEndColumn = sfEnd == sfBegin ? featureStartColumn
1872 if (featureEndColumn < fromColumn)
1876 else if (featureEndColumn > toColumn && sf.isContactFeature())
1879 * remove an enclosing feature if it is a contact feature
1891 * Invalidates any stale cursors (forcing recalculation) by incrementing the
1892 * token that has to match the one presented by the cursor
1895 public void sequenceChanged()
1904 public int replace(char c1, char c2)
1911 synchronized (sequence)
1913 for (int c = 0; c < sequence.length; c++)
1915 if (sequence[c] == c1)
1931 public String getSequenceStringFromIterator(Iterator<int[]> it)
1933 StringBuilder newSequence = new StringBuilder();
1934 while (it.hasNext())
1936 int[] block = it.next();
1939 newSequence.append(getSequence(block[0], block[1] + 1));
1943 newSequence.append(getSequence(block[0], block[1]));
1947 return newSequence.toString();
1951 public int firstResidueOutsideIterator(Iterator<int[]> regions)
1955 if (!regions.hasNext())
1957 return findIndex(getStart()) - 1;
1960 // Simply walk along the sequence whilst watching for region
1962 int hideStart = getLength();
1964 boolean foundStart = false;
1966 // step through the non-gapped positions of the sequence
1967 for (int i = getStart(); i <= getEnd() && (!foundStart); i++)
1969 // get alignment position of this residue in the sequence
1970 int p = findIndex(i) - 1;
1972 // update region start/end
1973 while (hideEnd < p && regions.hasNext())
1975 int[] region = regions.next();
1976 hideStart = region[0];
1977 hideEnd = region[1];
1981 hideStart = getLength();
1983 // update boundary for sequence
1995 // otherwise, sequence was completely hidden