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.util.StringUtils;
27 import java.util.ArrayList;
28 import java.util.Arrays;
29 import java.util.Enumeration;
30 import java.util.List;
31 import java.util.Vector;
33 import fr.orsay.lri.varna.models.rna.RNA;
37 * Implements the SequenceI interface for a char[] based sequence object.
42 public class Sequence extends ASequence implements SequenceI
44 SequenceI datasetSequence;
48 private char[] sequence;
56 Vector<PDBEntry> pdbIds;
60 DBRefEntryI sourceDBRef;
67 * This annotation is displayed below the alignment but the positions are tied
68 * to the residues of this sequence
70 * TODO: change to List<>
72 Vector<AlignmentAnnotation> annotation;
75 * The index of the sequence in a MSA
79 /** array of sequence features - may not be null for a valid sequence object */
80 public SequenceFeature[] sequenceFeatures;
83 * Creates a new Sequence object.
88 * string to form a possibly gapped sequence out of
90 * first position of non-gap residue in the sequence
92 * last position of ungapped residues (nearly always only used for
95 public Sequence(String name, String sequence, int start, int end)
97 initSeqAndName(name, sequence.toCharArray(), start, end);
100 public Sequence(String name, char[] sequence, int start, int end)
102 initSeqAndName(name, sequence, start, end);
106 * Stage 1 constructor - assign name, sequence, and set start and end fields.
107 * start and end are updated values from name2 if it ends with /start-end
114 protected void initSeqAndName(String name2, char[] sequence2, int start2,
118 this.sequence = sequence2;
125 com.stevesoft.pat.Regex limitrx = new com.stevesoft.pat.Regex(
126 "[/][0-9]{1,}[-][0-9]{1,}$");
128 com.stevesoft.pat.Regex endrx = new com.stevesoft.pat.Regex("[0-9]{1,}$");
135 .println("POSSIBLE IMPLEMENTATION ERROR: null sequence name passed to constructor.");
138 // Does sequence have the /start-end signature?
139 if (limitrx.search(name))
141 name = limitrx.left();
142 endrx.search(limitrx.stringMatched());
143 setStart(Integer.parseInt(limitrx.stringMatched().substring(1,
144 endrx.matchedFrom() - 1)));
145 setEnd(Integer.parseInt(endrx.stringMatched()));
149 void checkValidRange()
152 // http://issues.jalview.org/browse/JAL-774?focusedCommentId=11239&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-11239
155 for (int j = 0; j < sequence.length; j++)
157 if (!jalview.util.Comparison.isGap(sequence[j]))
176 * Creates a new Sequence object.
183 public Sequence(String name, String sequence)
185 this(name, sequence, 1, -1);
189 * Creates a new Sequence object with new AlignmentAnnotations but inherits
190 * any existing dataset sequence reference. If non exists, everything is
194 * if seq is a dataset sequence, behaves like a plain old copy
197 public Sequence(SequenceI seq)
199 this(seq, seq.getAnnotation());
203 * Create a new sequence object with new features, DBRefEntries, and PDBIds
204 * but inherits any existing dataset sequence reference, and duplicate of any
205 * annotation that is present in the given annotation array.
208 * the sequence to be copied
209 * @param alAnnotation
210 * an array of annotation including some associated with seq
212 public Sequence(SequenceI seq, AlignmentAnnotation[] alAnnotation)
214 initSeqFrom(seq, alAnnotation);
219 * does the heavy lifting when cloning a dataset sequence, or coping data from
220 * dataset to a new derived sequence.
223 * - source of attributes.
224 * @param alAnnotation
225 * - alignment annotation present on seq that should be copied onto
228 protected void initSeqFrom(SequenceI seq,
229 AlignmentAnnotation[] alAnnotation)
232 char[] oseq = seq.getSequence();
233 initSeqAndName(seq.getName(), Arrays.copyOf(oseq, oseq.length),
237 description = seq.getDescription();
238 sourceDBRef = seq.getSourceDBRef() == null ? null : new DBRefEntry(
239 seq.getSourceDBRef());
240 if (seq != datasetSequence)
242 setDatasetSequence(seq.getDatasetSequence());
244 if (datasetSequence == null && seq.getDBRefs() != null)
246 // only copy DBRefs and seqfeatures if we really are a dataset sequence
247 DBRefEntry[] dbr = seq.getDBRefs();
248 for (int i = 0; i < dbr.length; i++)
250 addDBRef(new DBRefEntry(dbr[i]));
252 if (seq.getSequenceFeatures() != null)
254 SequenceFeature[] sf = seq.getSequenceFeatures();
255 for (int i = 0; i < sf.length; i++)
257 addSequenceFeature(new SequenceFeature(sf[i]));
261 if (seq.getAnnotation() != null)
263 AlignmentAnnotation[] sqann = seq.getAnnotation();
264 for (int i = 0; i < sqann.length; i++)
266 if (sqann[i] == null)
270 boolean found = (alAnnotation == null);
273 for (int apos = 0; !found && apos < alAnnotation.length; apos++)
275 found = (alAnnotation[apos] == sqann[i]);
280 // only copy the given annotation
281 AlignmentAnnotation newann = new AlignmentAnnotation(sqann[i]);
282 addAlignmentAnnotation(newann);
286 if (seq.getAllPDBEntries() != null)
288 Vector<PDBEntry> ids = seq.getAllPDBEntries();
289 for (PDBEntry pdb : ids)
291 this.addPDBId(new PDBEntry(pdb));
298 public void setSequenceFeatures(SequenceFeature[] features)
300 if (datasetSequence == null)
302 sequenceFeatures = features;
307 .println("Warning: JAL-2046 side effect ? Possible implementation error: overwriting dataset sequence features by setting sequence features on alignment");
308 datasetSequence.setSequenceFeatures(features);
313 public synchronized void addSequenceFeature(SequenceFeature sf)
315 // TODO add to dataset sequence instead if there is one?
316 if (sequenceFeatures == null)
318 sequenceFeatures = new SequenceFeature[0];
321 for (int i = 0; i < sequenceFeatures.length; i++)
323 if (sequenceFeatures[i].equals(sf))
329 SequenceFeature[] temp = new SequenceFeature[sequenceFeatures.length + 1];
330 System.arraycopy(sequenceFeatures, 0, temp, 0, sequenceFeatures.length);
331 temp[sequenceFeatures.length] = sf;
333 sequenceFeatures = temp;
337 public void deleteFeature(SequenceFeature sf)
339 if (sequenceFeatures == null)
345 for (index = 0; index < sequenceFeatures.length; index++)
347 if (sequenceFeatures[index].equals(sf))
353 if (index == sequenceFeatures.length)
358 int sfLength = sequenceFeatures.length;
361 sequenceFeatures = null;
365 SequenceFeature[] temp = new SequenceFeature[sfLength - 1];
366 System.arraycopy(sequenceFeatures, 0, temp, 0, index);
368 if (index < sfLength)
370 System.arraycopy(sequenceFeatures, index + 1, temp, index,
371 sequenceFeatures.length - index - 1);
374 sequenceFeatures = temp;
379 * Returns the sequence features (if any), looking first on the sequence, then
380 * on its dataset sequence, and so on until a non-null value is found (or
381 * none). This supports retrieval of sequence features stored on the sequence
382 * (as in the applet) or on the dataset sequence (as in the Desktop version).
387 public SequenceFeature[] getSequenceFeatures()
389 SequenceFeature[] features = sequenceFeatures;
391 SequenceI seq = this;
392 int count = 0; // failsafe against loop in sequence.datasetsequence...
393 while (features == null && seq.getDatasetSequence() != null
396 seq = seq.getDatasetSequence();
397 features = ((Sequence) seq).sequenceFeatures;
403 public void addPDBId(PDBEntry entry)
407 pdbIds = new Vector<PDBEntry>();
409 if (pdbIds.contains(entry))
411 updatePDBEntry(pdbIds.get(pdbIds.indexOf(entry)), entry);
415 pdbIds.addElement(entry);
419 private static void updatePDBEntry(PDBEntry oldEntry, PDBEntry newEntry)
421 if (newEntry.getFile() != null)
423 oldEntry.setFile(newEntry.getFile());
434 public void setPDBId(Vector<PDBEntry> id)
442 * @return DOCUMENT ME!
445 public Vector<PDBEntry> getAllPDBEntries()
447 return pdbIds == null ? new Vector<PDBEntry>() : pdbIds;
453 * @return DOCUMENT ME!
456 public String getDisplayId(boolean jvsuffix)
458 StringBuffer result = new StringBuffer(name);
461 result.append("/" + start + "-" + end);
464 return result.toString();
474 public void setName(String name)
483 * @return DOCUMENT ME!
486 public String getName()
498 public void setStart(int start)
506 * @return DOCUMENT ME!
509 public int getStart()
521 public void setEnd(int end)
529 * @return DOCUMENT ME!
540 * @return DOCUMENT ME!
543 public int getLength()
545 return this.sequence.length;
555 public void setSequence(String seq)
557 this.sequence = seq.toCharArray();
562 public String getSequenceAsString()
564 return new String(sequence);
568 public String getSequenceAsString(int start, int end)
570 return new String(getSequence(start, end));
574 public char[] getSequence()
582 * @see jalview.datamodel.SequenceI#getSequence(int, int)
585 public char[] getSequence(int start, int end)
591 // JBPNote - left to user to pad the result here (TODO:Decide on this
593 if (start >= sequence.length)
598 if (end >= sequence.length)
600 end = sequence.length;
603 char[] reply = new char[end - start];
604 System.arraycopy(sequence, start, reply, 0, end - start);
610 public SequenceI getSubSequence(int start, int end)
616 char[] seq = getSequence(start, end);
621 int nstart = findPosition(start);
622 int nend = findPosition(end) - 1;
623 // JBPNote - this is an incomplete copy.
624 SequenceI nseq = new Sequence(this.getName(), seq, nstart, nend);
625 nseq.setDescription(description);
626 if (datasetSequence != null)
628 nseq.setDatasetSequence(datasetSequence);
632 nseq.setDatasetSequence(this);
638 * Returns the character of the aligned sequence at the given position (base
639 * zero), or space if the position is not within the sequence's bounds
644 public char getCharAt(int i)
646 if (i >= 0 && i < sequence.length)
663 public void setDescription(String desc)
665 this.description = desc;
671 * @return DOCUMENT ME!
674 public String getDescription()
676 return this.description;
682 * @see jalview.datamodel.SequenceI#findIndex(int)
685 public int findIndex(int pos)
687 // returns the alignment position for a residue
690 // Rely on end being at least as long as the length of the sequence.
691 while ((i < sequence.length) && (j <= end) && (j <= pos))
693 if (!jalview.util.Comparison.isGap(sequence[i]))
701 if ((j == end) && (j < pos))
712 public int findPosition(int i)
716 int seqlen = sequence.length;
717 while ((j < i) && (j < seqlen))
719 if (!jalview.util.Comparison.isGap(sequence[j]))
731 * Returns an int array where indices correspond to each residue in the
732 * sequence and the element value gives its position in the alignment
734 * @return int[SequenceI.getEnd()-SequenceI.getStart()+1] or null if no
735 * residues in SequenceI object
738 public int[] gapMap()
740 String seq = jalview.analysis.AlignSeq.extractGaps(
741 jalview.util.Comparison.GapChars, new String(sequence));
742 int[] map = new int[seq.length()];
746 while (j < sequence.length)
748 if (!jalview.util.Comparison.isGap(sequence[j]))
760 public int[] findPositionMap()
762 int map[] = new int[sequence.length];
765 int seqlen = sequence.length;
769 if (!jalview.util.Comparison.isGap(sequence[j]))
780 public List<int[]> getInsertions()
782 ArrayList<int[]> map = new ArrayList<int[]>();
783 int lastj = -1, j = 0;
785 int seqlen = sequence.length;
788 if (jalview.util.Comparison.isGap(sequence[j]))
799 map.add(new int[] { lastj, j - 1 });
807 map.add(new int[] { lastj, j - 1 });
814 public void deleteChars(int i, int j)
816 int newstart = start, newend = end;
817 if (i >= sequence.length || i < 0)
822 char[] tmp = StringUtils.deleteChars(sequence, i, j);
823 boolean createNewDs = false;
824 // TODO: take a (second look) at the dataset creation validation method for
825 // the very large sequence case
826 int eindex = -1, sindex = -1;
827 boolean ecalc = false, scalc = false;
828 for (int s = i; s < j; s++)
830 if (jalview.schemes.ResidueProperties.aaIndex[sequence[s]] != 23)
840 sindex = findIndex(start) - 1;
845 // delete characters including start of sequence
846 newstart = findPosition(j);
847 break; // don't need to search for any more residue characters.
851 // delete characters after start.
854 eindex = findIndex(end) - 1;
859 // delete characters at end of sequence
860 newend = findPosition(i - 1);
861 break; // don't need to search for any more residue characters.
866 newend--; // decrease end position by one for the deleted residue
867 // and search further
873 // deletion occured in the middle of the sequence
874 if (createNewDs && this.datasetSequence != null)
876 // construct a new sequence
877 Sequence ds = new Sequence(datasetSequence);
878 // TODO: remove any non-inheritable properties ?
879 // TODO: create a sequence mapping (since there is a relation here ?)
880 ds.deleteChars(i, j);
881 datasetSequence = ds;
889 public void insertCharAt(int i, int length, char c)
891 char[] tmp = new char[sequence.length + length];
893 if (i >= sequence.length)
895 System.arraycopy(sequence, 0, tmp, 0, sequence.length);
900 System.arraycopy(sequence, 0, tmp, 0, i);
910 if (i < sequence.length)
912 System.arraycopy(sequence, i, tmp, index, sequence.length - i);
919 public void insertCharAt(int i, char c)
921 insertCharAt(i, 1, c);
925 public String getVamsasId()
931 public void setVamsasId(String id)
937 public void setDBRefs(DBRefEntry[] dbref)
943 public DBRefEntry[] getDBRefs()
945 if (dbrefs == null && datasetSequence != null
946 && this != datasetSequence)
948 return datasetSequence.getDBRefs();
954 public void addDBRef(DBRefEntry entry)
956 // TODO add to dataset sequence instead if there is one?
959 dbrefs = new DBRefEntry[0];
962 int i, iSize = dbrefs.length;
964 for (i = 0; i < iSize; i++)
966 if (dbrefs[i].equalRef(entry))
968 if (entry.getMap() != null)
970 if (dbrefs[i].getMap() == null)
972 // overwrite with 'superior' entry that contains a mapping.
980 DBRefEntry[] temp = new DBRefEntry[iSize + 1];
981 System.arraycopy(dbrefs, 0, temp, 0, iSize);
982 temp[temp.length - 1] = entry;
988 public void setDatasetSequence(SequenceI seq)
990 // TODO check for circular reference before setting?
991 datasetSequence = seq;
995 public SequenceI getDatasetSequence()
997 return datasetSequence;
1001 public AlignmentAnnotation[] getAnnotation()
1003 return annotation == null ? null : annotation
1004 .toArray(new AlignmentAnnotation[annotation.size()]);
1008 public boolean hasAnnotation(AlignmentAnnotation ann)
1010 return annotation == null ? false : annotation.contains(ann);
1014 public void addAlignmentAnnotation(AlignmentAnnotation annotation)
1016 if (this.annotation == null)
1018 this.annotation = new Vector<AlignmentAnnotation>();
1020 if (!this.annotation.contains(annotation))
1022 this.annotation.addElement(annotation);
1024 annotation.setSequenceRef(this);
1028 public void removeAlignmentAnnotation(AlignmentAnnotation annotation)
1030 if (this.annotation != null)
1032 this.annotation.removeElement(annotation);
1033 if (this.annotation.size() == 0)
1035 this.annotation = null;
1041 * test if this is a valid candidate for another sequence's dataset sequence.
1044 private boolean isValidDatasetSequence()
1046 if (datasetSequence != null)
1050 for (int i = 0; i < sequence.length; i++)
1052 if (jalview.util.Comparison.isGap(sequence[i]))
1061 public SequenceI deriveSequence()
1064 if (datasetSequence == null)
1066 if (isValidDatasetSequence())
1068 // Use this as dataset sequence
1069 seq = new Sequence(getName(), "", 1, -1);
1070 seq.setDatasetSequence(this);
1071 seq.initSeqFrom(this, getAnnotation());
1076 // Create a new, valid dataset sequence
1077 createDatasetSequence();
1080 return new Sequence(this);
1086 * @see jalview.datamodel.SequenceI#createDatasetSequence()
1089 public SequenceI createDatasetSequence()
1091 if (datasetSequence == null)
1093 Sequence dsseq = new Sequence(getName(), AlignSeq.extractGaps(
1094 jalview.util.Comparison.GapChars, getSequenceAsString()),
1095 getStart(), getEnd());
1097 datasetSequence = dsseq;
1099 dsseq.setDescription(description);
1100 // move features and database references onto dataset sequence
1101 dsseq.sequenceFeatures = sequenceFeatures;
1102 sequenceFeatures=null;
1103 dsseq.dbrefs = dbrefs;
1105 // TODO: search and replace any references to this sequence with
1106 // references to the dataset sequence in Mappings on dbref
1107 dsseq.pdbIds = pdbIds;
1109 datasetSequence.updatePDBIds();
1110 if (annotation != null)
1112 // annotation is cloned rather than moved, to preserve what's currently
1114 for (AlignmentAnnotation aa : annotation)
1116 AlignmentAnnotation _aa = new AlignmentAnnotation(aa);
1117 _aa.sequenceRef = datasetSequence;
1118 _aa.adjustForAlignment(); // uses annotation's own record of
1119 // sequence-column mapping
1120 datasetSequence.addAlignmentAnnotation(_aa);
1124 return datasetSequence;
1131 * jalview.datamodel.SequenceI#setAlignmentAnnotation(AlignmmentAnnotation[]
1135 public void setAlignmentAnnotation(AlignmentAnnotation[] annotations)
1137 if (annotation != null)
1139 annotation.removeAllElements();
1141 if (annotations != null)
1143 for (int i = 0; i < annotations.length; i++)
1145 if (annotations[i] != null)
1147 addAlignmentAnnotation(annotations[i]);
1154 public AlignmentAnnotation[] getAnnotation(String label)
1156 if (annotation == null || annotation.size() == 0)
1161 Vector subset = new Vector();
1162 Enumeration e = annotation.elements();
1163 while (e.hasMoreElements())
1165 AlignmentAnnotation ann = (AlignmentAnnotation) e.nextElement();
1166 if (ann.label != null && ann.label.equals(label))
1168 subset.addElement(ann);
1171 if (subset.size() == 0)
1175 AlignmentAnnotation[] anns = new AlignmentAnnotation[subset.size()];
1177 e = subset.elements();
1178 while (e.hasMoreElements())
1180 anns[i++] = (AlignmentAnnotation) e.nextElement();
1182 subset.removeAllElements();
1187 public boolean updatePDBIds()
1189 if (datasetSequence != null)
1191 // TODO: could merge DBRefs
1192 return datasetSequence.updatePDBIds();
1194 if (dbrefs == null || dbrefs.length == 0)
1198 Vector newpdb = new Vector();
1199 for (int i = 0; i < dbrefs.length; i++)
1201 if (DBRefSource.PDB.equals(dbrefs[i].getSource()))
1203 PDBEntry pdbe = new PDBEntry();
1204 pdbe.setId(dbrefs[i].getAccessionId());
1205 if (pdbIds == null || pdbIds.size() == 0)
1207 newpdb.addElement(pdbe);
1211 Enumeration en = pdbIds.elements();
1212 boolean matched = false;
1213 while (!matched && en.hasMoreElements())
1215 PDBEntry anentry = (PDBEntry) en.nextElement();
1216 if (anentry.getId().equals(pdbe.getId()))
1223 newpdb.addElement(pdbe);
1228 if (newpdb.size() > 0)
1230 Enumeration en = newpdb.elements();
1231 while (en.hasMoreElements())
1233 addPDBId((PDBEntry) en.nextElement());
1241 public void transferAnnotation(SequenceI entry, Mapping mp)
1243 if (datasetSequence != null)
1245 datasetSequence.transferAnnotation(entry, mp);
1248 if (entry.getDatasetSequence() != null)
1250 transferAnnotation(entry.getDatasetSequence(), mp);
1253 // transfer any new features from entry onto sequence
1254 if (entry.getSequenceFeatures() != null)
1257 SequenceFeature[] sfs = entry.getSequenceFeatures();
1258 for (int si = 0; si < sfs.length; si++)
1260 SequenceFeature sf[] = (mp != null) ? mp.locateFeature(sfs[si])
1261 : new SequenceFeature[] { new SequenceFeature(sfs[si]) };
1262 if (sf != null && sf.length > 0)
1264 for (int sfi = 0; sfi < sf.length; sfi++)
1266 addSequenceFeature(sf[sfi]);
1272 // transfer PDB entries
1273 if (entry.getAllPDBEntries() != null)
1275 Enumeration e = entry.getAllPDBEntries().elements();
1276 while (e.hasMoreElements())
1278 PDBEntry pdb = (PDBEntry) e.nextElement();
1282 // transfer database references
1283 DBRefEntry[] entryRefs = entry.getDBRefs();
1284 if (entryRefs != null)
1286 for (int r = 0; r < entryRefs.length; r++)
1288 DBRefEntry newref = new DBRefEntry(entryRefs[r]);
1289 if (newref.getMap() != null && mp != null)
1291 // remap ref using our local mapping
1293 // we also assume all version string setting is done by dbSourceProxy
1295 * if (!newref.getSource().equalsIgnoreCase(dbSource)) {
1296 * newref.setSource(dbSource); }
1304 * @return The index (zero-based) on this sequence in the MSA. It returns
1305 * {@code -1} if this information is not available.
1308 public int getIndex()
1314 * Defines the position of this sequence in the MSA. Use the value {@code -1}
1315 * if this information is undefined.
1318 * position for this sequence. This value is zero-based (zero for
1319 * this first sequence)
1322 public void setIndex(int value)
1328 public void setRNA(RNA r)
1340 public List<AlignmentAnnotation> getAlignmentAnnotations(String calcId,
1343 List<AlignmentAnnotation> result = new ArrayList<AlignmentAnnotation>();
1344 if (this.annotation != null)
1346 for (AlignmentAnnotation ann : annotation)
1348 if (ann.calcId != null && ann.calcId.equals(calcId)
1349 && ann.label != null && ann.label.equals(label))
1359 public String toString()
1361 return getDisplayId(false);
1365 public PDBEntry getPDBEntry(String pdbIdStr)
1367 if (getDatasetSequence() == null
1368 || getDatasetSequence().getAllPDBEntries() == null)
1372 List<PDBEntry> entries = getDatasetSequence().getAllPDBEntries();
1373 for (PDBEntry entry : entries)
1375 if (entry.getId().equalsIgnoreCase(pdbIdStr))
1384 public void setSourceDBRef(DBRefEntryI dbRef)
1386 this.sourceDBRef = dbRef;
1390 public DBRefEntryI getSourceDBRef()
1392 return this.sourceDBRef;