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.List;
38 import java.util.ListIterator;
39 import java.util.Vector;
41 import com.stevesoft.pat.Regex;
43 import fr.orsay.lri.varna.models.rna.RNA;
47 * Implements the SequenceI interface for a char[] based sequence object.
52 public class Sequence extends ASequence implements SequenceI
54 private static final Regex limitrx = new Regex(
55 "[/][0-9]{1,}[-][0-9]{1,}$");
57 private static final Regex endrx = new Regex("[0-9]{1,}$");
59 SequenceI datasetSequence;
63 private char[] sequence;
71 Vector<PDBEntry> pdbIds;
80 * This annotation is displayed below the alignment but the positions are tied
81 * to the residues of this sequence
83 * TODO: change to List<>
85 Vector<AlignmentAnnotation> annotation;
88 * The index of the sequence in a MSA
92 private SequenceFeatures sequenceFeatureStore;
95 * A cursor holding the approximate current view position to the sequence,
96 * as determined by findIndex or findPosition or findPositions.
97 * Using a cursor as a hint allows these methods to be more performant for
100 private SequenceCursor cursor;
103 * A number that should be incremented whenever the sequence is edited.
104 * If the value matches the cursor token, then we can trust the cursor,
105 * if not then it should be recomputed.
107 private int changeCount;
109 private GeneLoci geneLoci;
112 * Creates a new Sequence object.
115 * display name string
117 * string to form a possibly gapped sequence out of
119 * first position of non-gap residue in the sequence
121 * last position of ungapped residues (nearly always only used for
124 public Sequence(String name, String sequence, int start, int end)
127 initSeqAndName(name, sequence.toCharArray(), start, end);
130 public Sequence(String name, char[] sequence, int start, int end)
133 initSeqAndName(name, sequence, start, end);
137 * Stage 1 constructor - assign name, sequence, and set start and end fields.
138 * start and end are updated values from name2 if it ends with /start-end
145 protected void initSeqAndName(String name2, char[] sequence2, int start2,
149 this.sequence = sequence2;
161 "POSSIBLE IMPLEMENTATION ERROR: null sequence name passed to constructor.");
164 // Does sequence have the /start-end signature?
165 if (limitrx.search(name))
167 name = limitrx.left();
168 endrx.search(limitrx.stringMatched());
169 setStart(Integer.parseInt(limitrx.stringMatched().substring(1,
170 endrx.matchedFrom() - 1)));
171 setEnd(Integer.parseInt(endrx.stringMatched()));
175 void checkValidRange()
178 // http://issues.jalview.org/browse/JAL-774?focusedCommentId=11239&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-11239
181 for (int j = 0; j < sequence.length; j++)
183 if (!jalview.util.Comparison.isGap(sequence[j]))
202 * default constructor
206 sequenceFeatureStore = new SequenceFeatures();
210 * Creates a new Sequence object.
217 public Sequence(String name, String sequence)
219 this(name, sequence, 1, -1);
223 * Creates a new Sequence object with new AlignmentAnnotations but inherits
224 * any existing dataset sequence reference. If non exists, everything is
228 * if seq is a dataset sequence, behaves like a plain old copy
231 public Sequence(SequenceI seq)
233 this(seq, seq.getAnnotation());
237 * Create a new sequence object with new features, DBRefEntries, and PDBIds
238 * but inherits any existing dataset sequence reference, and duplicate of any
239 * annotation that is present in the given annotation array.
242 * the sequence to be copied
243 * @param alAnnotation
244 * an array of annotation including some associated with seq
246 public Sequence(SequenceI seq, AlignmentAnnotation[] alAnnotation)
249 initSeqFrom(seq, alAnnotation);
253 * does the heavy lifting when cloning a dataset sequence, or coping data from
254 * dataset to a new derived sequence.
257 * - source of attributes.
258 * @param alAnnotation
259 * - alignment annotation present on seq that should be copied onto
262 protected void initSeqFrom(SequenceI seq,
263 AlignmentAnnotation[] alAnnotation)
265 char[] oseq = seq.getSequence(); // returns a copy of the array
266 initSeqAndName(seq.getName(), oseq, seq.getStart(), seq.getEnd());
268 description = seq.getDescription();
269 if (seq != datasetSequence)
271 setDatasetSequence(seq.getDatasetSequence());
275 * only copy DBRefs and seqfeatures if we really are a dataset sequence
277 if (datasetSequence == null)
279 if (seq.getDBRefs() != null)
281 DBRefEntry[] dbr = seq.getDBRefs();
282 for (int i = 0; i < dbr.length; i++)
284 addDBRef(new DBRefEntry(dbr[i]));
289 * make copies of any sequence features
291 for (SequenceFeature sf : seq.getSequenceFeatures())
293 addSequenceFeature(new SequenceFeature(sf));
297 if (seq.getAnnotation() != null)
299 AlignmentAnnotation[] sqann = seq.getAnnotation();
300 for (int i = 0; i < sqann.length; i++)
302 if (sqann[i] == null)
306 boolean found = (alAnnotation == null);
309 for (int apos = 0; !found && apos < alAnnotation.length; apos++)
311 found = (alAnnotation[apos] == sqann[i]);
316 // only copy the given annotation
317 AlignmentAnnotation newann = new AlignmentAnnotation(sqann[i]);
318 addAlignmentAnnotation(newann);
322 if (seq.getAllPDBEntries() != null)
324 Vector<PDBEntry> ids = seq.getAllPDBEntries();
325 for (PDBEntry pdb : ids)
327 this.addPDBId(new PDBEntry(pdb));
333 public void setSequenceFeatures(List<SequenceFeature> features)
335 if (datasetSequence != null)
337 datasetSequence.setSequenceFeatures(features);
340 sequenceFeatureStore = new SequenceFeatures(features);
344 public synchronized boolean addSequenceFeature(SequenceFeature sf)
346 if (sf.getType() == null)
348 System.err.println("SequenceFeature type may not be null: "
353 if (datasetSequence != null)
355 return datasetSequence.addSequenceFeature(sf);
358 return sequenceFeatureStore.add(sf);
362 public void deleteFeature(SequenceFeature sf)
364 if (datasetSequence != null)
366 datasetSequence.deleteFeature(sf);
370 sequenceFeatureStore.delete(sf);
380 public List<SequenceFeature> getSequenceFeatures()
382 if (datasetSequence != null)
384 return datasetSequence.getSequenceFeatures();
386 return sequenceFeatureStore.getAllFeatures();
390 public SequenceFeaturesI getFeatures()
392 return datasetSequence != null ? datasetSequence.getFeatures()
393 : sequenceFeatureStore;
397 public boolean addPDBId(PDBEntry entry)
401 pdbIds = new Vector<PDBEntry>();
406 for (PDBEntry pdbe : pdbIds)
408 if (pdbe.updateFrom(entry))
413 pdbIds.addElement(entry);
424 public void setPDBId(Vector<PDBEntry> id)
432 * @return DOCUMENT ME!
435 public Vector<PDBEntry> getAllPDBEntries()
437 return pdbIds == null ? new Vector<PDBEntry>() : pdbIds;
443 * @return DOCUMENT ME!
446 public String getDisplayId(boolean jvsuffix)
448 StringBuffer result = new StringBuffer(name);
451 result.append("/" + start + "-" + end);
454 return result.toString();
464 public void setName(String name)
473 * @return DOCUMENT ME!
476 public String getName()
488 public void setStart(int start)
496 * @return DOCUMENT ME!
499 public int getStart()
511 public void setEnd(int end)
519 * @return DOCUMENT ME!
530 * @return DOCUMENT ME!
533 public int getLength()
535 return this.sequence.length;
545 public void setSequence(String seq)
547 this.sequence = seq.toCharArray();
553 public String getSequenceAsString()
555 return new String(sequence);
559 public String getSequenceAsString(int start, int end)
561 return new String(getSequence(start, end));
565 public char[] getSequence()
568 return sequence == null ? null : Arrays.copyOf(sequence,
575 * @see jalview.datamodel.SequenceI#getSequence(int, int)
578 public char[] getSequence(int start, int end)
584 // JBPNote - left to user to pad the result here (TODO:Decide on this
586 if (start >= sequence.length)
591 if (end >= sequence.length)
593 end = sequence.length;
596 char[] reply = new char[end - start];
597 System.arraycopy(sequence, start, reply, 0, end - start);
603 public SequenceI getSubSequence(int start, int end)
609 char[] seq = getSequence(start, end);
614 int nstart = findPosition(start);
615 int nend = findPosition(end) - 1;
616 // JBPNote - this is an incomplete copy.
617 SequenceI nseq = new Sequence(this.getName(), seq, nstart, nend);
618 nseq.setDescription(description);
619 if (datasetSequence != null)
621 nseq.setDatasetSequence(datasetSequence);
625 nseq.setDatasetSequence(this);
631 * Returns the character of the aligned sequence at the given position (base
632 * zero), or space if the position is not within the sequence's bounds
637 public char getCharAt(int i)
639 if (i >= 0 && i < sequence.length)
650 * Sets the sequence description, and also parses out any special formats of
656 public void setDescription(String desc)
658 this.description = desc;
663 * Parses and saves fields of an Ensembl-style description e.g.
664 * chromosome:GRCh38:17:45051610:45109016:1
666 protected void parseDescription()
668 if (description == null)
672 String[] tokens = description.split(":");
673 if (tokens.length == 6 && "chromosome".equals(tokens[0])) {
674 String ref = tokens[1];
675 String chrom = tokens[2];
677 int chStart = Integer.parseInt(tokens[3]);
678 int chEnd = Integer.parseInt(tokens[4]);
679 boolean forwardStrand = "1".equals(tokens[5]);
680 String species = ""; // dunno yet!
681 int[] from = new int[] { start, end };
682 int[] to = new int[] { chStart, chEnd };
683 MapList map = new MapList(from, to, 1, 1);
684 GeneLoci gl = new GeneLoci(species, ref, chrom, map, forwardStrand);
686 } catch (NumberFormatException e)
688 System.err.println("Bad integers in description " + description);
693 public void setGeneLoci(GeneLoci gl)
699 * Returns the gene loci mapping for the sequence (may be null)
703 public GeneLoci getGeneLoci()
709 * Answers the description
714 public String getDescription()
716 return this.description;
723 public int findIndex(int pos)
726 * use a valid, hopefully nearby, cursor if available
728 if (isValidCursor(cursor))
730 return findIndex(pos, cursor);
738 * traverse sequence from the start counting gaps; make a note of
739 * the column of the first residue to save in the cursor
741 while ((i < sequence.length) && (j <= end) && (j <= pos))
743 if (!Comparison.isGap(sequence[i]))
754 if (j == end && j < pos)
759 updateCursor(pos, i, startColumn);
764 * Updates the cursor to the latest found residue and column position
771 * column position of the first sequence residue
773 protected void updateCursor(int residuePos, int column, int startColumn)
776 * preserve end residue column provided cursor was valid
778 int endColumn = isValidCursor(cursor) ? cursor.lastColumnPosition : 0;
779 if (residuePos == this.end)
784 cursor = new SequenceCursor(this, residuePos, column, startColumn,
785 endColumn, this.changeCount);
789 * Answers the aligned column position (1..) for the given residue position
790 * (start..) given a 'hint' of a residue/column location in the neighbourhood.
791 * The hint may be left of, at, or to the right of the required position.
797 protected int findIndex(int pos, SequenceCursor curs)
799 if (!isValidCursor(curs))
802 * wrong or invalidated cursor, compute de novo
804 return findIndex(pos);
807 if (curs.residuePosition == pos)
809 return curs.columnPosition;
813 * move left or right to find pos from hint.position
815 int col = curs.columnPosition - 1; // convert from base 1 to 0-based array
817 int newPos = curs.residuePosition;
818 int delta = newPos > pos ? -1 : 1;
820 while (newPos != pos)
822 col += delta; // shift one column left or right
823 if (col < 0 || col == sequence.length)
827 if (!Comparison.isGap(sequence[col]))
833 col++; // convert back to base 1
834 updateCursor(pos, col, curs.firstColumnPosition);
843 public int findPosition(final int column)
846 * use a valid, hopefully nearby, cursor if available
848 if (isValidCursor(cursor))
850 return findPosition(column + 1, cursor);
853 // TODO recode this more naturally i.e. count residues only
854 // as they are found, not 'in anticipation'
857 * traverse the sequence counting gaps; note the column position
858 * of the first residue, to save in the cursor
860 int firstResidueColumn = 0;
861 int lastPosFound = 0;
862 int lastPosFoundColumn = 0;
863 int seqlen = sequence.length;
865 if (seqlen > 0 && !Comparison.isGap(sequence[0]))
867 lastPosFound = start;
868 lastPosFoundColumn = 0;
874 while (j < column && j < seqlen)
876 if (!Comparison.isGap(sequence[j]))
879 lastPosFoundColumn = j;
880 if (pos == this.start)
882 firstResidueColumn = j;
888 if (j < seqlen && !Comparison.isGap(sequence[j]))
891 lastPosFoundColumn = j;
892 if (pos == this.start)
894 firstResidueColumn = j;
899 * update the cursor to the last residue position found (if any)
900 * (converting column position to base 1)
902 if (lastPosFound != 0)
904 updateCursor(lastPosFound, lastPosFoundColumn + 1,
905 firstResidueColumn + 1);
912 * Answers true if the given cursor is not null, is for this sequence object,
913 * and has a token value that matches this object's changeCount, else false.
914 * This allows us to ignore a cursor as 'stale' if the sequence has been
915 * modified since the cursor was created.
920 protected boolean isValidCursor(SequenceCursor curs)
922 if (curs == null || curs.sequence != this || curs.token != changeCount)
927 * sanity check against range
929 if (curs.columnPosition < 0 || curs.columnPosition > sequence.length)
933 if (curs.residuePosition < start || curs.residuePosition > end)
941 * Answers the sequence position (start..) for the given aligned column
942 * position (1..), given a hint of a cursor in the neighbourhood. The cursor
943 * may lie left of, at, or to the right of the column position.
949 protected int findPosition(final int col, SequenceCursor curs)
951 if (!isValidCursor(curs))
954 * wrong or invalidated cursor, compute de novo
956 return findPosition(col - 1);// ugh back to base 0
959 if (curs.columnPosition == col)
961 cursor = curs; // in case this method becomes public
962 return curs.residuePosition; // easy case :-)
965 if (curs.lastColumnPosition > 0 && curs.lastColumnPosition < col)
968 * sequence lies entirely to the left of col
969 * - return last residue + 1
974 if (curs.firstColumnPosition > 0 && curs.firstColumnPosition > col)
977 * sequence lies entirely to the right of col
978 * - return first residue
983 // todo could choose closest to col out of column,
984 // firstColumnPosition, lastColumnPosition as a start point
987 * move left or right to find pos from cursor position
989 int firstResidueColumn = curs.firstColumnPosition;
990 int column = curs.columnPosition - 1; // to base 0
991 int newPos = curs.residuePosition;
992 int delta = curs.columnPosition > col ? -1 : 1;
993 boolean gapped = false;
994 int lastFoundPosition = curs.residuePosition;
995 int lastFoundPositionColumn = curs.columnPosition;
997 while (column != col - 1)
999 column += delta; // shift one column left or right
1000 if (column < 0 || column == sequence.length)
1004 gapped = Comparison.isGap(sequence[column]);
1008 lastFoundPosition = newPos;
1009 lastFoundPositionColumn = column + 1;
1010 if (lastFoundPosition == this.start)
1012 firstResidueColumn = column + 1;
1017 if (cursor == null || lastFoundPosition != cursor.residuePosition)
1019 updateCursor(lastFoundPosition, lastFoundPositionColumn,
1020 firstResidueColumn);
1024 * hack to give position to the right if on a gap
1025 * or beyond the length of the sequence (see JAL-2562)
1027 if (delta > 0 && (gapped || column >= sequence.length))
1039 public Range findPositions(int fromColumn, int toColumn)
1041 if (toColumn < fromColumn || fromColumn < 1)
1047 * find the first non-gapped position, if any
1049 int firstPosition = 0;
1050 int col = fromColumn - 1;
1051 int length = sequence.length;
1052 while (col < length && col < toColumn)
1054 if (!Comparison.isGap(sequence[col]))
1056 firstPosition = findPosition(col++);
1062 if (firstPosition == 0)
1068 * find the last non-gapped position
1070 int lastPosition = firstPosition;
1071 while (col < length && col < toColumn)
1073 if (!Comparison.isGap(sequence[col++]))
1079 return new Range(firstPosition, lastPosition);
1083 * Returns an int array where indices correspond to each residue in the
1084 * sequence and the element value gives its position in the alignment
1086 * @return int[SequenceI.getEnd()-SequenceI.getStart()+1] or null if no
1087 * residues in SequenceI object
1090 public int[] gapMap()
1092 String seq = jalview.analysis.AlignSeq.extractGaps(
1093 jalview.util.Comparison.GapChars, new String(sequence));
1094 int[] map = new int[seq.length()];
1098 while (j < sequence.length)
1100 if (!jalview.util.Comparison.isGap(sequence[j]))
1112 public int[] findPositionMap()
1114 int map[] = new int[sequence.length];
1117 int seqlen = sequence.length;
1118 while ((j < seqlen))
1121 if (!jalview.util.Comparison.isGap(sequence[j]))
1132 public List<int[]> getInsertions()
1134 ArrayList<int[]> map = new ArrayList<int[]>();
1135 int lastj = -1, j = 0;
1137 int seqlen = sequence.length;
1138 while ((j < seqlen))
1140 if (jalview.util.Comparison.isGap(sequence[j]))
1151 map.add(new int[] { lastj, j - 1 });
1159 map.add(new int[] { lastj, j - 1 });
1166 public BitSet getInsertionsAsBits()
1168 BitSet map = new BitSet();
1169 int lastj = -1, j = 0;
1171 int seqlen = sequence.length;
1172 while ((j < seqlen))
1174 if (jalview.util.Comparison.isGap(sequence[j]))
1200 public void deleteChars(int i, int j)
1202 int newstart = start, newend = end;
1203 if (i >= sequence.length || i < 0)
1208 char[] tmp = StringUtils.deleteChars(sequence, i, j);
1209 boolean createNewDs = false;
1210 // TODO: take a (second look) at the dataset creation validation method for
1211 // the very large sequence case
1212 int eindex = -1, sindex = -1;
1213 boolean ecalc = false, scalc = false;
1214 for (int s = i; s < j; s++)
1216 if (jalview.schemes.ResidueProperties.aaIndex[sequence[s]] != 23)
1226 sindex = findIndex(start) - 1;
1231 // delete characters including start of sequence
1232 newstart = findPosition(j);
1233 break; // don't need to search for any more residue characters.
1237 // delete characters after start.
1240 eindex = findIndex(end) - 1;
1245 // delete characters at end of sequence
1246 newend = findPosition(i - 1);
1247 break; // don't need to search for any more residue characters.
1252 newend--; // decrease end position by one for the deleted residue
1253 // and search further
1259 // deletion occured in the middle of the sequence
1260 if (createNewDs && this.datasetSequence != null)
1262 // construct a new sequence
1263 Sequence ds = new Sequence(datasetSequence);
1264 // TODO: remove any non-inheritable properties ?
1265 // TODO: create a sequence mapping (since there is a relation here ?)
1266 ds.deleteChars(i, j);
1267 datasetSequence = ds;
1276 public void insertCharAt(int i, int length, char c)
1278 char[] tmp = new char[sequence.length + length];
1280 if (i >= sequence.length)
1282 System.arraycopy(sequence, 0, tmp, 0, sequence.length);
1283 i = sequence.length;
1287 System.arraycopy(sequence, 0, tmp, 0, i);
1297 if (i < sequence.length)
1299 System.arraycopy(sequence, i, tmp, index, sequence.length - i);
1307 public void insertCharAt(int i, char c)
1309 insertCharAt(i, 1, c);
1313 public String getVamsasId()
1319 public void setVamsasId(String id)
1325 public void setDBRefs(DBRefEntry[] dbref)
1327 if (dbrefs == null && datasetSequence != null
1328 && this != datasetSequence)
1330 datasetSequence.setDBRefs(dbref);
1336 DBRefUtils.ensurePrimaries(this);
1341 public DBRefEntry[] getDBRefs()
1343 if (dbrefs == null && datasetSequence != null
1344 && this != datasetSequence)
1346 return datasetSequence.getDBRefs();
1352 public void addDBRef(DBRefEntry entry)
1354 if (datasetSequence != null)
1356 datasetSequence.addDBRef(entry);
1362 dbrefs = new DBRefEntry[0];
1365 for (DBRefEntryI dbr : dbrefs)
1367 if (dbr.updateFrom(entry))
1370 * found a dbref that either matched, or could be
1371 * updated from, the new entry - no need to add it
1378 * extend the array to make room for one more
1380 // TODO use an ArrayList instead
1381 int j = dbrefs.length;
1382 DBRefEntry[] temp = new DBRefEntry[j + 1];
1383 System.arraycopy(dbrefs, 0, temp, 0, j);
1384 temp[temp.length - 1] = entry;
1388 DBRefUtils.ensurePrimaries(this);
1392 public void setDatasetSequence(SequenceI seq)
1396 throw new IllegalArgumentException(
1397 "Implementation Error: self reference passed to SequenceI.setDatasetSequence");
1399 if (seq != null && seq.getDatasetSequence() != null)
1401 throw new IllegalArgumentException(
1402 "Implementation error: cascading dataset sequences are not allowed.");
1404 datasetSequence = seq;
1408 public SequenceI getDatasetSequence()
1410 return datasetSequence;
1414 public AlignmentAnnotation[] getAnnotation()
1416 return annotation == null ? null
1418 .toArray(new AlignmentAnnotation[annotation.size()]);
1422 public boolean hasAnnotation(AlignmentAnnotation ann)
1424 return annotation == null ? false : annotation.contains(ann);
1428 public void addAlignmentAnnotation(AlignmentAnnotation annotation)
1430 if (this.annotation == null)
1432 this.annotation = new Vector<AlignmentAnnotation>();
1434 if (!this.annotation.contains(annotation))
1436 this.annotation.addElement(annotation);
1438 annotation.setSequenceRef(this);
1442 public void removeAlignmentAnnotation(AlignmentAnnotation annotation)
1444 if (this.annotation != null)
1446 this.annotation.removeElement(annotation);
1447 if (this.annotation.size() == 0)
1449 this.annotation = null;
1455 * test if this is a valid candidate for another sequence's dataset sequence.
1458 private boolean isValidDatasetSequence()
1460 if (datasetSequence != null)
1464 for (int i = 0; i < sequence.length; i++)
1466 if (jalview.util.Comparison.isGap(sequence[i]))
1475 public SequenceI deriveSequence()
1477 Sequence seq = null;
1478 if (datasetSequence == null)
1480 if (isValidDatasetSequence())
1482 // Use this as dataset sequence
1483 seq = new Sequence(getName(), "", 1, -1);
1484 seq.setDatasetSequence(this);
1485 seq.initSeqFrom(this, getAnnotation());
1490 // Create a new, valid dataset sequence
1491 createDatasetSequence();
1494 return new Sequence(this);
1497 private boolean _isNa;
1499 private int _seqhash = 0;
1502 * Answers false if the sequence is more than 85% nucleotide (ACGTU), else
1506 public boolean isProtein()
1508 if (datasetSequence != null)
1510 return datasetSequence.isProtein();
1512 if (_seqhash != sequence.hashCode())
1514 _seqhash = sequence.hashCode();
1515 _isNa = Comparison.isNucleotide(this);
1523 * @see jalview.datamodel.SequenceI#createDatasetSequence()
1526 public SequenceI createDatasetSequence()
1528 if (datasetSequence == null)
1530 Sequence dsseq = new Sequence(getName(),
1531 AlignSeq.extractGaps(jalview.util.Comparison.GapChars,
1532 getSequenceAsString()),
1533 getStart(), getEnd());
1535 datasetSequence = dsseq;
1537 dsseq.setDescription(description);
1538 // move features and database references onto dataset sequence
1539 dsseq.sequenceFeatureStore = sequenceFeatureStore;
1540 sequenceFeatureStore = null;
1541 dsseq.dbrefs = dbrefs;
1543 // TODO: search and replace any references to this sequence with
1544 // references to the dataset sequence in Mappings on dbref
1545 dsseq.pdbIds = pdbIds;
1547 datasetSequence.updatePDBIds();
1548 if (annotation != null)
1550 // annotation is cloned rather than moved, to preserve what's currently
1552 for (AlignmentAnnotation aa : annotation)
1554 AlignmentAnnotation _aa = new AlignmentAnnotation(aa);
1555 _aa.sequenceRef = datasetSequence;
1556 _aa.adjustForAlignment(); // uses annotation's own record of
1557 // sequence-column mapping
1558 datasetSequence.addAlignmentAnnotation(_aa);
1562 return datasetSequence;
1569 * jalview.datamodel.SequenceI#setAlignmentAnnotation(AlignmmentAnnotation[]
1573 public void setAlignmentAnnotation(AlignmentAnnotation[] annotations)
1575 if (annotation != null)
1577 annotation.removeAllElements();
1579 if (annotations != null)
1581 for (int i = 0; i < annotations.length; i++)
1583 if (annotations[i] != null)
1585 addAlignmentAnnotation(annotations[i]);
1592 public AlignmentAnnotation[] getAnnotation(String label)
1594 if (annotation == null || annotation.size() == 0)
1599 Vector<AlignmentAnnotation> subset = new Vector<AlignmentAnnotation>();
1600 Enumeration<AlignmentAnnotation> e = annotation.elements();
1601 while (e.hasMoreElements())
1603 AlignmentAnnotation ann = e.nextElement();
1604 if (ann.label != null && ann.label.equals(label))
1606 subset.addElement(ann);
1609 if (subset.size() == 0)
1613 AlignmentAnnotation[] anns = new AlignmentAnnotation[subset.size()];
1615 e = subset.elements();
1616 while (e.hasMoreElements())
1618 anns[i++] = e.nextElement();
1620 subset.removeAllElements();
1625 public boolean updatePDBIds()
1627 if (datasetSequence != null)
1629 // TODO: could merge DBRefs
1630 return datasetSequence.updatePDBIds();
1632 if (dbrefs == null || dbrefs.length == 0)
1636 boolean added = false;
1637 for (DBRefEntry dbr : dbrefs)
1639 if (DBRefSource.PDB.equals(dbr.getSource()))
1642 * 'Add' any PDB dbrefs as a PDBEntry - add is only performed if the
1643 * PDB id is not already present in a 'matching' PDBEntry
1644 * Constructor parses out a chain code if appended to the accession id
1645 * (a fudge used to 'store' the chain code in the DBRef)
1647 PDBEntry pdbe = new PDBEntry(dbr);
1648 added |= addPDBId(pdbe);
1655 public void transferAnnotation(SequenceI entry, Mapping mp)
1657 if (datasetSequence != null)
1659 datasetSequence.transferAnnotation(entry, mp);
1662 if (entry.getDatasetSequence() != null)
1664 transferAnnotation(entry.getDatasetSequence(), mp);
1667 // transfer any new features from entry onto sequence
1668 if (entry.getSequenceFeatures() != null)
1671 List<SequenceFeature> sfs = entry.getSequenceFeatures();
1672 for (SequenceFeature feature : sfs)
1674 SequenceFeature sf[] = (mp != null) ? mp.locateFeature(feature)
1675 : new SequenceFeature[] { new SequenceFeature(feature) };
1678 for (int sfi = 0; sfi < sf.length; sfi++)
1680 addSequenceFeature(sf[sfi]);
1686 // transfer PDB entries
1687 if (entry.getAllPDBEntries() != null)
1689 Enumeration<PDBEntry> e = entry.getAllPDBEntries().elements();
1690 while (e.hasMoreElements())
1692 PDBEntry pdb = e.nextElement();
1696 // transfer database references
1697 DBRefEntry[] entryRefs = entry.getDBRefs();
1698 if (entryRefs != null)
1700 for (int r = 0; r < entryRefs.length; r++)
1702 DBRefEntry newref = new DBRefEntry(entryRefs[r]);
1703 if (newref.getMap() != null && mp != null)
1705 // remap ref using our local mapping
1707 // we also assume all version string setting is done by dbSourceProxy
1709 * if (!newref.getSource().equalsIgnoreCase(dbSource)) {
1710 * newref.setSource(dbSource); }
1718 * @return The index (zero-based) on this sequence in the MSA. It returns
1719 * {@code -1} if this information is not available.
1722 public int getIndex()
1728 * Defines the position of this sequence in the MSA. Use the value {@code -1}
1729 * if this information is undefined.
1732 * position for this sequence. This value is zero-based (zero for
1733 * this first sequence)
1736 public void setIndex(int value)
1742 public void setRNA(RNA r)
1754 public List<AlignmentAnnotation> getAlignmentAnnotations(String calcId,
1757 List<AlignmentAnnotation> result = new ArrayList<AlignmentAnnotation>();
1758 if (this.annotation != null)
1760 for (AlignmentAnnotation ann : annotation)
1762 if (ann.calcId != null && ann.calcId.equals(calcId)
1763 && ann.label != null && ann.label.equals(label))
1773 public String toString()
1775 return getDisplayId(false);
1779 public PDBEntry getPDBEntry(String pdbIdStr)
1781 if (getDatasetSequence() != null)
1783 return getDatasetSequence().getPDBEntry(pdbIdStr);
1789 List<PDBEntry> entries = getAllPDBEntries();
1790 for (PDBEntry entry : entries)
1792 if (entry.getId().equalsIgnoreCase(pdbIdStr))
1801 public List<DBRefEntry> getPrimaryDBRefs()
1803 if (datasetSequence != null)
1805 return datasetSequence.getPrimaryDBRefs();
1807 if (dbrefs == null || dbrefs.length == 0)
1809 return Collections.emptyList();
1811 synchronized (dbrefs)
1813 List<DBRefEntry> primaries = new ArrayList<DBRefEntry>();
1814 DBRefEntry[] tmp = new DBRefEntry[1];
1815 for (DBRefEntry ref : dbrefs)
1817 if (!ref.isPrimaryCandidate())
1823 MapList mp = ref.getMap().getMap();
1824 if (mp.getFromLowest() > start || mp.getFromHighest() < end)
1826 // map only involves a subsequence, so cannot be primary
1830 // whilst it looks like it is a primary ref, we also sanity check type
1831 if (DBRefUtils.getCanonicalName(DBRefSource.PDB)
1832 .equals(DBRefUtils.getCanonicalName(ref.getSource())))
1834 // PDB dbrefs imply there should be a PDBEntry associated
1835 // TODO: tighten PDB dbrefs
1836 // formally imply Jalview has actually downloaded and
1837 // parsed the pdb file. That means there should be a cached file
1838 // handle on the PDBEntry, and a real mapping between sequence and
1839 // extracted sequence from PDB file
1840 PDBEntry pdbentry = getPDBEntry(ref.getAccessionId());
1841 if (pdbentry != null && pdbentry.getFile() != null)
1847 // check standard protein or dna sources
1849 DBRefEntry[] res = DBRefUtils.selectDbRefs(!isProtein(), tmp);
1850 if (res != null && res[0] == tmp[0])
1864 public List<SequenceFeature> findFeatures(int fromColumn, int toColumn,
1867 int startPos = findPosition(fromColumn - 1); // convert base 1 to base 0
1868 int endPos = fromColumn == toColumn ? startPos
1869 : findPosition(toColumn - 1);
1871 List<SequenceFeature> result = getFeatures().findFeatures(startPos,
1875 * if end column is gapped, endPos may be to the right,
1876 * and we may have included adjacent or enclosing features;
1877 * remove any that are not enclosing, non-contact features
1879 if (endPos > this.end || Comparison.isGap(sequence[toColumn - 1]))
1881 ListIterator<SequenceFeature> it = result.listIterator();
1882 while (it.hasNext())
1884 SequenceFeature sf = it.next();
1885 int sfBegin = sf.getBegin();
1886 int sfEnd = sf.getEnd();
1887 int featureStartColumn = findIndex(sfBegin);
1888 if (featureStartColumn > toColumn)
1892 else if (featureStartColumn < fromColumn)
1894 int featureEndColumn = sfEnd == sfBegin ? featureStartColumn
1896 if (featureEndColumn < fromColumn)
1900 else if (featureEndColumn > toColumn && sf.isContactFeature())
1903 * remove an enclosing feature if it is a contact feature
1915 * Invalidates any stale cursors (forcing recalculation) by incrementing the
1916 * token that has to match the one presented by the cursor
1919 public void sequenceChanged()
1928 public int replace(char c1, char c2)
1935 synchronized (sequence)
1937 for (int c = 0; c < sequence.length; c++)
1939 if (sequence[c] == c1)