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));
303 public void setSequenceFeatures(SequenceFeature[] features)
305 sequenceFeatures = features;
309 public synchronized void addSequenceFeature(SequenceFeature sf)
311 // TODO add to dataset sequence instead if there is one?
312 if (sequenceFeatures == null)
314 sequenceFeatures = new SequenceFeature[0];
317 for (int i = 0; i < sequenceFeatures.length; i++)
319 if (sequenceFeatures[i].equals(sf))
325 SequenceFeature[] temp = new SequenceFeature[sequenceFeatures.length + 1];
326 System.arraycopy(sequenceFeatures, 0, temp, 0, sequenceFeatures.length);
327 temp[sequenceFeatures.length] = sf;
329 sequenceFeatures = temp;
333 public void deleteFeature(SequenceFeature sf)
335 if (sequenceFeatures == null)
341 for (index = 0; index < sequenceFeatures.length; index++)
343 if (sequenceFeatures[index].equals(sf))
349 if (index == sequenceFeatures.length)
354 int sfLength = sequenceFeatures.length;
357 sequenceFeatures = null;
361 SequenceFeature[] temp = new SequenceFeature[sfLength - 1];
362 System.arraycopy(sequenceFeatures, 0, temp, 0, index);
364 if (index < sfLength)
366 System.arraycopy(sequenceFeatures, index + 1, temp, index,
367 sequenceFeatures.length - index - 1);
370 sequenceFeatures = temp;
375 * Returns the sequence features (if any), looking first on the sequence, then
376 * on its dataset sequence, and so on until a non-null value is found (or
377 * none). This supports retrieval of sequence features stored on the sequence
378 * (as in the applet) or on the dataset sequence (as in the Desktop version).
383 public SequenceFeature[] getSequenceFeatures()
385 SequenceFeature[] features = sequenceFeatures;
387 SequenceI seq = this;
388 int count = 0; // failsafe against loop in sequence.datasetsequence...
389 while (features == null && seq.getDatasetSequence() != null
392 seq = seq.getDatasetSequence();
393 features = ((Sequence) seq).sequenceFeatures;
399 public void addPDBId(PDBEntry entry)
403 pdbIds = new Vector<PDBEntry>();
405 if (pdbIds.contains(entry))
407 updatePDBEntry(pdbIds.get(pdbIds.indexOf(entry)), entry);
411 pdbIds.addElement(entry);
415 private static void updatePDBEntry(PDBEntry oldEntry, PDBEntry newEntry)
417 if (newEntry.getFile() != null)
419 oldEntry.setFile(newEntry.getFile());
430 public void setPDBId(Vector<PDBEntry> id)
438 * @return DOCUMENT ME!
441 public Vector<PDBEntry> getAllPDBEntries()
443 return pdbIds == null ? new Vector<PDBEntry>() : pdbIds;
449 * @return DOCUMENT ME!
452 public String getDisplayId(boolean jvsuffix)
454 StringBuffer result = new StringBuffer(name);
457 result.append("/" + start + "-" + end);
460 return result.toString();
470 public void setName(String name)
479 * @return DOCUMENT ME!
482 public String getName()
494 public void setStart(int start)
502 * @return DOCUMENT ME!
505 public int getStart()
517 public void setEnd(int end)
525 * @return DOCUMENT ME!
536 * @return DOCUMENT ME!
539 public int getLength()
541 return this.sequence.length;
551 public void setSequence(String seq)
553 this.sequence = seq.toCharArray();
558 public String getSequenceAsString()
560 return new String(sequence);
564 public String getSequenceAsString(int start, int end)
566 return new String(getSequence(start, end));
570 public char[] getSequence()
578 * @see jalview.datamodel.SequenceI#getSequence(int, int)
581 public char[] getSequence(int start, int end)
587 // JBPNote - left to user to pad the result here (TODO:Decide on this
589 if (start >= sequence.length)
594 if (end >= sequence.length)
596 end = sequence.length;
599 char[] reply = new char[end - start];
600 System.arraycopy(sequence, start, reply, 0, end - start);
606 public SequenceI getSubSequence(int start, int end)
612 char[] seq = getSequence(start, end);
617 int nstart = findPosition(start);
618 int nend = findPosition(end) - 1;
619 // JBPNote - this is an incomplete copy.
620 SequenceI nseq = new Sequence(this.getName(), seq, nstart, nend);
621 nseq.setDescription(description);
622 if (datasetSequence != null)
624 nseq.setDatasetSequence(datasetSequence);
628 nseq.setDatasetSequence(this);
634 * Returns the character of the aligned sequence at the given position (base
635 * zero), or space if the position is not within the sequence's bounds
640 public char getCharAt(int i)
642 if (i >= 0 && i < sequence.length)
659 public void setDescription(String desc)
661 this.description = desc;
667 * @return DOCUMENT ME!
670 public String getDescription()
672 return this.description;
678 * @see jalview.datamodel.SequenceI#findIndex(int)
681 public int findIndex(int pos)
683 // returns the alignment position for a residue
686 // Rely on end being at least as long as the length of the sequence.
687 while ((i < sequence.length) && (j <= end) && (j <= pos))
689 if (!jalview.util.Comparison.isGap(sequence[i]))
697 if ((j == end) && (j < pos))
708 public int findPosition(int i)
712 int seqlen = sequence.length;
713 while ((j < i) && (j < seqlen))
715 if (!jalview.util.Comparison.isGap(sequence[j]))
727 * Returns an int array where indices correspond to each residue in the
728 * sequence and the element value gives its position in the alignment
730 * @return int[SequenceI.getEnd()-SequenceI.getStart()+1] or null if no
731 * residues in SequenceI object
734 public int[] gapMap()
736 String seq = jalview.analysis.AlignSeq.extractGaps(
737 jalview.util.Comparison.GapChars, new String(sequence));
738 int[] map = new int[seq.length()];
742 while (j < sequence.length)
744 if (!jalview.util.Comparison.isGap(sequence[j]))
756 public int[] findPositionMap()
758 int map[] = new int[sequence.length];
761 int seqlen = sequence.length;
765 if (!jalview.util.Comparison.isGap(sequence[j]))
776 public List<int[]> getInsertions()
778 ArrayList<int[]> map = new ArrayList<int[]>();
779 int lastj = -1, j = 0;
781 int seqlen = sequence.length;
784 if (jalview.util.Comparison.isGap(sequence[j]))
795 map.add(new int[] { lastj, j - 1 });
803 map.add(new int[] { lastj, j - 1 });
810 public void deleteChars(int i, int j)
812 int newstart = start, newend = end;
813 if (i >= sequence.length || i < 0)
818 char[] tmp = StringUtils.deleteChars(sequence, i, j);
819 boolean createNewDs = false;
820 // TODO: take a (second look) at the dataset creation validation method for
821 // the very large sequence case
822 int eindex = -1, sindex = -1;
823 boolean ecalc = false, scalc = false;
824 for (int s = i; s < j; s++)
826 if (jalview.schemes.ResidueProperties.aaIndex[sequence[s]] != 23)
836 sindex = findIndex(start) - 1;
841 // delete characters including start of sequence
842 newstart = findPosition(j);
843 break; // don't need to search for any more residue characters.
847 // delete characters after start.
850 eindex = findIndex(end) - 1;
855 // delete characters at end of sequence
856 newend = findPosition(i - 1);
857 break; // don't need to search for any more residue characters.
862 newend--; // decrease end position by one for the deleted residue
863 // and search further
869 // deletion occured in the middle of the sequence
870 if (createNewDs && this.datasetSequence != null)
872 // construct a new sequence
873 Sequence ds = new Sequence(datasetSequence);
874 // TODO: remove any non-inheritable properties ?
875 // TODO: create a sequence mapping (since there is a relation here ?)
876 ds.deleteChars(i, j);
877 datasetSequence = ds;
885 public void insertCharAt(int i, int length, char c)
887 char[] tmp = new char[sequence.length + length];
889 if (i >= sequence.length)
891 System.arraycopy(sequence, 0, tmp, 0, sequence.length);
896 System.arraycopy(sequence, 0, tmp, 0, i);
906 if (i < sequence.length)
908 System.arraycopy(sequence, i, tmp, index, sequence.length - i);
915 public void insertCharAt(int i, char c)
917 insertCharAt(i, 1, c);
921 public String getVamsasId()
927 public void setVamsasId(String id)
933 public void setDBRefs(DBRefEntry[] dbref)
939 public DBRefEntry[] getDBRefs()
941 if (dbrefs == null && datasetSequence != null
942 && this != datasetSequence)
944 return datasetSequence.getDBRefs();
950 public void addDBRef(DBRefEntry entry)
952 // TODO add to dataset sequence instead if there is one?
955 dbrefs = new DBRefEntry[0];
958 int i, iSize = dbrefs.length;
960 for (i = 0; i < iSize; i++)
962 if (dbrefs[i].equalRef(entry))
964 if (entry.getMap() != null)
966 if (dbrefs[i].getMap() == null)
968 // overwrite with 'superior' entry that contains a mapping.
976 DBRefEntry[] temp = new DBRefEntry[iSize + 1];
977 System.arraycopy(dbrefs, 0, temp, 0, iSize);
978 temp[temp.length - 1] = entry;
984 public void setDatasetSequence(SequenceI seq)
986 // TODO check for circular reference before setting?
987 datasetSequence = seq;
991 public SequenceI getDatasetSequence()
993 return datasetSequence;
997 public AlignmentAnnotation[] getAnnotation()
999 return annotation == null ? null : annotation
1000 .toArray(new AlignmentAnnotation[annotation.size()]);
1004 public boolean hasAnnotation(AlignmentAnnotation ann)
1006 return annotation == null ? false : annotation.contains(ann);
1010 public void addAlignmentAnnotation(AlignmentAnnotation annotation)
1012 if (this.annotation == null)
1014 this.annotation = new Vector<AlignmentAnnotation>();
1016 if (!this.annotation.contains(annotation))
1018 this.annotation.addElement(annotation);
1020 annotation.setSequenceRef(this);
1024 public void removeAlignmentAnnotation(AlignmentAnnotation annotation)
1026 if (this.annotation != null)
1028 this.annotation.removeElement(annotation);
1029 if (this.annotation.size() == 0)
1031 this.annotation = null;
1037 * test if this is a valid candidate for another sequence's dataset sequence.
1040 private boolean isValidDatasetSequence()
1042 if (datasetSequence != null)
1046 for (int i = 0; i < sequence.length; i++)
1048 if (jalview.util.Comparison.isGap(sequence[i]))
1057 public SequenceI deriveSequence()
1059 SequenceI seq = new Sequence(this);
1060 if (datasetSequence != null)
1062 // duplicate current sequence with same dataset
1063 seq.setDatasetSequence(datasetSequence);
1067 if (isValidDatasetSequence())
1069 // Use this as dataset sequence
1070 seq.setDatasetSequence(this);
1074 // Create a new, valid dataset sequence
1076 ds.setSequence(AlignSeq.extractGaps(
1077 jalview.util.Comparison.GapChars, new String(sequence)));
1078 setDatasetSequence(ds);
1079 ds.setSequenceFeatures(getSequenceFeatures());
1080 seq = this; // and return this sequence as the derived sequence.
1089 * @see jalview.datamodel.SequenceI#createDatasetSequence()
1092 public SequenceI createDatasetSequence()
1094 if (datasetSequence == null)
1096 Sequence dsseq = new Sequence(getName(), AlignSeq.extractGaps(
1097 jalview.util.Comparison.GapChars, getSequenceAsString()),
1098 getStart(), getEnd());
1100 datasetSequence = dsseq;
1102 dsseq.setDescription(description);
1103 // move features and database references onto dataset sequence
1104 dsseq.sequenceFeatures = sequenceFeatures;
1105 sequenceFeatures=null;
1106 dsseq.dbrefs = dbrefs;
1108 // TODO: search and replace any references to this sequence with
1109 // references to the dataset sequence in Mappings on dbref
1110 dsseq.pdbIds = pdbIds;
1112 datasetSequence.updatePDBIds();
1113 if (annotation != null)
1115 // annotation is cloned rather than moved, to preserve what's currently
1117 for (AlignmentAnnotation aa : annotation)
1119 AlignmentAnnotation _aa = new AlignmentAnnotation(aa);
1120 _aa.sequenceRef = datasetSequence;
1121 _aa.adjustForAlignment(); // uses annotation's own record of
1122 // sequence-column mapping
1123 datasetSequence.addAlignmentAnnotation(_aa);
1127 return datasetSequence;
1134 * jalview.datamodel.SequenceI#setAlignmentAnnotation(AlignmmentAnnotation[]
1138 public void setAlignmentAnnotation(AlignmentAnnotation[] annotations)
1140 if (annotation != null)
1142 annotation.removeAllElements();
1144 if (annotations != null)
1146 for (int i = 0; i < annotations.length; i++)
1148 if (annotations[i] != null)
1150 addAlignmentAnnotation(annotations[i]);
1157 public AlignmentAnnotation[] getAnnotation(String label)
1159 if (annotation == null || annotation.size() == 0)
1164 Vector subset = new Vector();
1165 Enumeration e = annotation.elements();
1166 while (e.hasMoreElements())
1168 AlignmentAnnotation ann = (AlignmentAnnotation) e.nextElement();
1169 if (ann.label != null && ann.label.equals(label))
1171 subset.addElement(ann);
1174 if (subset.size() == 0)
1178 AlignmentAnnotation[] anns = new AlignmentAnnotation[subset.size()];
1180 e = subset.elements();
1181 while (e.hasMoreElements())
1183 anns[i++] = (AlignmentAnnotation) e.nextElement();
1185 subset.removeAllElements();
1190 public boolean updatePDBIds()
1192 if (datasetSequence != null)
1194 // TODO: could merge DBRefs
1195 return datasetSequence.updatePDBIds();
1197 if (dbrefs == null || dbrefs.length == 0)
1201 Vector newpdb = new Vector();
1202 for (int i = 0; i < dbrefs.length; i++)
1204 if (DBRefSource.PDB.equals(dbrefs[i].getSource()))
1206 PDBEntry pdbe = new PDBEntry();
1207 pdbe.setId(dbrefs[i].getAccessionId());
1208 if (pdbIds == null || pdbIds.size() == 0)
1210 newpdb.addElement(pdbe);
1214 Enumeration en = pdbIds.elements();
1215 boolean matched = false;
1216 while (!matched && en.hasMoreElements())
1218 PDBEntry anentry = (PDBEntry) en.nextElement();
1219 if (anentry.getId().equals(pdbe.getId()))
1226 newpdb.addElement(pdbe);
1231 if (newpdb.size() > 0)
1233 Enumeration en = newpdb.elements();
1234 while (en.hasMoreElements())
1236 addPDBId((PDBEntry) en.nextElement());
1244 public void transferAnnotation(SequenceI entry, Mapping mp)
1246 if (datasetSequence != null)
1248 datasetSequence.transferAnnotation(entry, mp);
1251 if (entry.getDatasetSequence() != null)
1253 transferAnnotation(entry.getDatasetSequence(), mp);
1256 // transfer any new features from entry onto sequence
1257 if (entry.getSequenceFeatures() != null)
1260 SequenceFeature[] sfs = entry.getSequenceFeatures();
1261 for (int si = 0; si < sfs.length; si++)
1263 SequenceFeature sf[] = (mp != null) ? mp.locateFeature(sfs[si])
1264 : new SequenceFeature[] { new SequenceFeature(sfs[si]) };
1265 if (sf != null && sf.length > 0)
1267 for (int sfi = 0; sfi < sf.length; sfi++)
1269 addSequenceFeature(sf[sfi]);
1275 // transfer PDB entries
1276 if (entry.getAllPDBEntries() != null)
1278 Enumeration e = entry.getAllPDBEntries().elements();
1279 while (e.hasMoreElements())
1281 PDBEntry pdb = (PDBEntry) e.nextElement();
1285 // transfer database references
1286 DBRefEntry[] entryRefs = entry.getDBRefs();
1287 if (entryRefs != null)
1289 for (int r = 0; r < entryRefs.length; r++)
1291 DBRefEntry newref = new DBRefEntry(entryRefs[r]);
1292 if (newref.getMap() != null && mp != null)
1294 // remap ref using our local mapping
1296 // we also assume all version string setting is done by dbSourceProxy
1298 * if (!newref.getSource().equalsIgnoreCase(dbSource)) {
1299 * newref.setSource(dbSource); }
1307 * @return The index (zero-based) on this sequence in the MSA. It returns
1308 * {@code -1} if this information is not available.
1311 public int getIndex()
1317 * Defines the position of this sequence in the MSA. Use the value {@code -1}
1318 * if this information is undefined.
1321 * position for this sequence. This value is zero-based (zero for
1322 * this first sequence)
1325 public void setIndex(int value)
1331 public void setRNA(RNA r)
1343 public List<AlignmentAnnotation> getAlignmentAnnotations(String calcId,
1346 List<AlignmentAnnotation> result = new ArrayList<AlignmentAnnotation>();
1347 if (this.annotation != null)
1349 for (AlignmentAnnotation ann : annotation)
1351 if (ann.calcId != null && ann.calcId.equals(calcId)
1352 && ann.label != null && ann.label.equals(label))
1362 public String toString()
1364 return getDisplayId(false);
1368 public PDBEntry getPDBEntry(String pdbIdStr)
1370 if (getDatasetSequence() == null
1371 || getDatasetSequence().getAllPDBEntries() == null)
1375 List<PDBEntry> entries = getDatasetSequence().getAllPDBEntries();
1376 for (PDBEntry entry : entries)
1378 if (entry.getId().equalsIgnoreCase(pdbIdStr))
1387 public void setSourceDBRef(DBRefEntryI dbRef)
1389 this.sourceDBRef = dbRef;
1393 public DBRefEntryI getSourceDBRef()
1395 return this.sourceDBRef;