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.AlignmentUtils;
24 import jalview.datamodel.AlignedCodonFrame.SequenceToSequenceMapping;
25 import jalview.io.FastaFile;
26 import jalview.util.Comparison;
27 import jalview.util.MessageManager;
29 import java.util.ArrayList;
30 import java.util.Collections;
31 import java.util.Enumeration;
32 import java.util.HashSet;
33 import java.util.Hashtable;
34 import java.util.List;
37 import java.util.Vector;
40 * Data structure to hold and manipulate a multiple sequence alignment
46 public class Alignment implements AlignmentI
48 private Alignment dataset;
50 protected List<SequenceI> sequences;
52 protected List<SequenceGroup> groups;
54 protected char gapCharacter = '-';
56 protected int type = NUCLEOTIDE;
58 public static final int PROTEIN = 0;
60 public static final int NUCLEOTIDE = 1;
62 public boolean hasRNAStructure = false;
64 public AlignmentAnnotation[] annotations;
66 HiddenSequences hiddenSequences;
68 public Hashtable alignmentProperties;
70 private List<AlignedCodonFrame> codonFrameList;
72 private void initAlignment(SequenceI[] seqs)
74 groups = Collections.synchronizedList(new ArrayList<SequenceGroup>());
75 hiddenSequences = new HiddenSequences(this);
76 codonFrameList = new ArrayList<AlignedCodonFrame>();
78 if (Comparison.isNucleotide(seqs))
87 sequences = Collections.synchronizedList(new ArrayList<SequenceI>());
89 for (int i = 0; i < seqs.length; i++)
91 sequences.add(seqs[i]);
97 * Make a 'copy' alignment - sequences have new copies of features and
98 * annotations, but share the original dataset sequences.
100 public Alignment(AlignmentI al)
102 SequenceI[] seqs = al.getSequencesArray();
103 for (int i = 0; i < seqs.length; i++)
105 seqs[i] = new Sequence(seqs[i]);
111 * Share the same dataset sequence mappings (if any).
113 if (dataset == null && al.getDataset() == null)
115 this.setCodonFrames(al.getCodonFrames());
120 * Make an alignment from an array of Sequences.
124 public Alignment(SequenceI[] seqs)
130 * Make a new alignment from an array of SeqCigars
135 public Alignment(SeqCigar[] alseqs)
137 SequenceI[] seqs = SeqCigar.createAlignmentSequences(alseqs,
138 gapCharacter, new ColumnSelection(), null);
143 * Make a new alignment from an CigarArray JBPNote - can only do this when
144 * compactAlignment does not contain hidden regions. JBPNote - must also check
145 * that compactAlignment resolves to a set of SeqCigars - or construct them
148 * @param compactAlignment
151 public static AlignmentI createAlignment(CigarArray compactAlignment)
155 .getString("error.alignment_cigararray_not_implemented"));
156 // this(compactAlignment.refCigars);
160 public List<SequenceI> getSequences()
166 public List<SequenceI> getSequences(
167 Map<SequenceI, SequenceCollectionI> hiddenReps)
169 // TODO: in jalview 2.8 we don't do anything with hiddenreps - fix design to
175 public SequenceI[] getSequencesArray()
177 if (sequences == null)
181 synchronized (sequences)
183 return sequences.toArray(new SequenceI[sequences.size()]);
188 * Returns a map of lists of sequences keyed by sequence name.
193 public Map<String, List<SequenceI>> getSequencesByName()
195 return AlignmentUtils.getSequencesByName(this);
204 * @return DOCUMENT ME!
207 public SequenceI getSequenceAt(int i)
209 synchronized (sequences)
211 if (i > -1 && i < sequences.size())
213 return sequences.get(i);
220 * Adds a sequence to the alignment. Recalculates maxLength and size.
225 public void addSequence(SequenceI snew)
229 // maintain dataset integrity
230 if (snew.getDatasetSequence() != null)
232 getDataset().addSequence(snew.getDatasetSequence());
236 // derive new sequence
237 SequenceI adding = snew.deriveSequence();
238 getDataset().addSequence(adding.getDatasetSequence());
242 if (sequences == null)
244 initAlignment(new SequenceI[] { snew });
248 synchronized (sequences)
253 if (hiddenSequences != null)
255 hiddenSequences.adjustHeightSequenceAdded();
260 public SequenceI replaceSequenceAt(int i, SequenceI snew)
262 synchronized (sequences)
264 if (sequences.size() > i)
266 return sequences.set(i, snew);
272 hiddenSequences.adjustHeightSequenceAdded();
281 * @return DOCUMENT ME!
284 public List<SequenceGroup> getGroups()
290 public void finalize()
292 if (getDataset() != null)
294 getDataset().removeAlignmentRef();
301 hiddenSequences = null;
305 * decrement the alignmentRefs counter by one and call finalize if it goes to
308 private void removeAlignmentRef()
310 if (--alignmentRefs == 0)
323 public void deleteSequence(SequenceI s)
325 deleteSequence(findIndex(s));
335 public void deleteSequence(int i)
337 if (i > -1 && i < getHeight())
339 synchronized (sequences)
342 hiddenSequences.adjustHeightSequenceDeleted(i);
350 * @see jalview.datamodel.AlignmentI#findGroup(jalview.datamodel.SequenceI)
353 public SequenceGroup findGroup(SequenceI s)
355 synchronized (groups)
357 for (int i = 0; i < this.groups.size(); i++)
359 SequenceGroup sg = groups.get(i);
361 if (sg.getSequences(null).contains(s))
374 * jalview.datamodel.AlignmentI#findAllGroups(jalview.datamodel.SequenceI)
377 public SequenceGroup[] findAllGroups(SequenceI s)
379 ArrayList<SequenceGroup> temp = new ArrayList<SequenceGroup>();
381 synchronized (groups)
383 int gSize = groups.size();
384 for (int i = 0; i < gSize; i++)
386 SequenceGroup sg = groups.get(i);
387 if (sg == null || sg.getSequences() == null)
389 this.deleteGroup(sg);
394 if (sg.getSequences().contains(s))
400 SequenceGroup[] ret = new SequenceGroup[temp.size()];
401 return temp.toArray(ret);
406 public void addGroup(SequenceGroup sg)
408 synchronized (groups)
410 if (!groups.contains(sg))
412 if (hiddenSequences.getSize() > 0)
414 int i, iSize = sg.getSize();
415 for (i = 0; i < iSize; i++)
417 if (!sequences.contains(sg.getSequenceAt(i)))
419 sg.deleteSequence(sg.getSequenceAt(i), false);
425 if (sg.getSize() < 1)
437 * remove any annotation that references gp
440 * (if null, removes all group associated annotation)
442 private void removeAnnotationForGroup(SequenceGroup gp)
444 if (annotations == null || annotations.length == 0)
448 // remove annotation very quickly
449 AlignmentAnnotation[] t, todelete = new AlignmentAnnotation[annotations.length], tokeep = new AlignmentAnnotation[annotations.length];
453 for (i = 0, p = 0, k = 0; i < annotations.length; i++)
455 if (annotations[i].groupRef != null)
457 todelete[p++] = annotations[i];
461 tokeep[k++] = annotations[i];
467 for (i = 0, p = 0, k = 0; i < annotations.length; i++)
469 if (annotations[i].groupRef == gp)
471 todelete[p++] = annotations[i];
475 tokeep[k++] = annotations[i];
481 // clear out the group associated annotation.
482 for (i = 0; i < p; i++)
484 unhookAnnotation(todelete[i]);
487 t = new AlignmentAnnotation[k];
488 for (i = 0; i < k; i++)
497 public void deleteAllGroups()
499 synchronized (groups)
501 if (annotations != null)
503 removeAnnotationForGroup(null);
505 for (SequenceGroup sg : groups)
515 public void deleteGroup(SequenceGroup g)
517 synchronized (groups)
519 if (groups.contains(g))
521 removeAnnotationForGroup(g);
530 public SequenceI findName(String name)
532 return findName(name, false);
538 * @see jalview.datamodel.AlignmentI#findName(java.lang.String, boolean)
541 public SequenceI findName(String token, boolean b)
543 return findName(null, token, b);
549 * @see jalview.datamodel.AlignmentI#findName(SequenceI, java.lang.String,
553 public SequenceI findName(SequenceI startAfter, String token, boolean b)
558 String sqname = null;
559 if (startAfter != null)
561 // try to find the sequence in the alignment
562 boolean matched = false;
563 while (i < sequences.size())
565 if (getSequenceAt(i++) == startAfter)
576 while (i < sequences.size())
578 sq = getSequenceAt(i);
579 sqname = sq.getName();
580 if (sqname.equals(token) // exact match
581 || (b && // allow imperfect matches - case varies
582 (sqname.equalsIgnoreCase(token))))
584 return getSequenceAt(i);
594 public SequenceI[] findSequenceMatch(String name)
596 Vector matches = new Vector();
599 while (i < sequences.size())
601 if (getSequenceAt(i).getName().equals(name))
603 matches.addElement(getSequenceAt(i));
608 SequenceI[] result = new SequenceI[matches.size()];
609 for (i = 0; i < result.length; i++)
611 result[i] = (SequenceI) matches.elementAt(i);
621 * @see jalview.datamodel.AlignmentI#findIndex(jalview.datamodel.SequenceI)
624 public int findIndex(SequenceI s)
628 while (i < sequences.size())
630 if (s == getSequenceAt(i))
645 * jalview.datamodel.AlignmentI#findIndex(jalview.datamodel.SearchResults)
648 public int findIndex(SearchResults results)
652 while (i < sequences.size())
654 if (results.involvesSequence(getSequenceAt(i)))
666 * @return DOCUMENT ME!
669 public int getHeight()
671 return sequences.size();
677 * @return DOCUMENT ME!
680 public int getWidth()
684 for (int i = 0; i < sequences.size(); i++)
686 if (getSequenceAt(i).getLength() > maxLength)
688 maxLength = getSequenceAt(i).getLength();
702 public void setGapCharacter(char gc)
705 synchronized (sequences)
707 for (SequenceI seq : sequences)
709 seq.setSequence(seq.getSequenceAsString().replace('.', gc)
710 .replace('-', gc).replace(' ', gc));
718 * @return DOCUMENT ME!
721 public char getGapCharacter()
729 * @see jalview.datamodel.AlignmentI#isAligned()
732 public boolean isAligned()
734 return isAligned(false);
740 * @see jalview.datamodel.AlignmentI#isAligned(boolean)
743 public boolean isAligned(boolean includeHidden)
745 int width = getWidth();
746 if (hiddenSequences == null || hiddenSequences.getSize() == 0)
748 includeHidden = true; // no hidden sequences to check against.
750 for (int i = 0; i < sequences.size(); i++)
752 if (includeHidden || !hiddenSequences.isHidden(getSequenceAt(i)))
754 if (getSequenceAt(i).getLength() != width)
765 * Delete all annotations, including auto-calculated if the flag is set true.
766 * Returns true if at least one annotation was deleted, else false.
768 * @param includingAutoCalculated
772 public boolean deleteAllAnnotations(boolean includingAutoCalculated)
774 boolean result = false;
775 for (AlignmentAnnotation alan : getAlignmentAnnotation())
777 if (!alan.autoCalculated || includingAutoCalculated)
779 deleteAnnotation(alan);
789 * @seejalview.datamodel.AlignmentI#deleteAnnotation(jalview.datamodel.
790 * AlignmentAnnotation)
793 public boolean deleteAnnotation(AlignmentAnnotation aa)
795 return deleteAnnotation(aa, true);
799 public boolean deleteAnnotation(AlignmentAnnotation aa, boolean unhook)
803 if (annotations != null)
805 aSize = annotations.length;
813 AlignmentAnnotation[] temp = new AlignmentAnnotation[aSize - 1];
815 boolean swap = false;
818 for (int i = 0; i < aSize; i++)
820 if (annotations[i] == aa)
825 if (tIndex < temp.length)
827 temp[tIndex++] = annotations[i];
836 unhookAnnotation(aa);
843 * remove any object references associated with this annotation
847 private void unhookAnnotation(AlignmentAnnotation aa)
849 if (aa.sequenceRef != null)
851 aa.sequenceRef.removeAlignmentAnnotation(aa);
853 if (aa.groupRef != null)
855 // probably need to do more here in the future (post 2.5.0)
863 * @seejalview.datamodel.AlignmentI#addAnnotation(jalview.datamodel.
864 * AlignmentAnnotation)
867 public void addAnnotation(AlignmentAnnotation aa)
869 addAnnotation(aa, -1);
875 * @seejalview.datamodel.AlignmentI#addAnnotation(jalview.datamodel.
876 * AlignmentAnnotation, int)
879 public void addAnnotation(AlignmentAnnotation aa, int pos)
881 if (aa.getRNAStruc() != null)
883 hasRNAStructure = true;
887 if (annotations != null)
889 aSize = annotations.length + 1;
892 AlignmentAnnotation[] temp = new AlignmentAnnotation[aSize];
894 if (pos == -1 || pos >= aSize)
896 temp[aSize - 1] = aa;
905 for (i = 0; i < (aSize - 1); i++, p++)
913 temp[p] = annotations[i];
922 public void setAnnotationIndex(AlignmentAnnotation aa, int index)
924 if (aa == null || annotations == null || annotations.length - 1 < index)
929 int aSize = annotations.length;
930 AlignmentAnnotation[] temp = new AlignmentAnnotation[aSize];
934 for (int i = 0; i < aSize; i++)
943 temp[i] = annotations[i];
947 temp[i] = annotations[i - 1];
956 * returns all annotation on the alignment
958 public AlignmentAnnotation[] getAlignmentAnnotation()
964 public void setNucleotide(boolean b)
977 public boolean isNucleotide()
979 if (type == NUCLEOTIDE)
990 public boolean hasRNAStructure()
992 // TODO can it happen that structure is removed from alignment?
993 return hasRNAStructure;
997 public void setDataset(AlignmentI data)
999 if (dataset == null && data == null)
1001 createDatasetAlignment();
1003 else if (dataset == null && data != null)
1005 if (!(data instanceof Alignment))
1008 "Implementation Error: jalview.datamodel.Alignment does not yet support other implementations of AlignmentI as its dataset reference");
1010 dataset = (Alignment) data;
1011 for (int i = 0; i < getHeight(); i++)
1013 SequenceI currentSeq = getSequenceAt(i);
1014 SequenceI dsq = currentSeq.getDatasetSequence();
1017 dsq = currentSeq.createDatasetSequence();
1018 dataset.addSequence(dsq);
1022 while (dsq.getDatasetSequence() != null)
1024 dsq = dsq.getDatasetSequence();
1026 if (dataset.findIndex(dsq) == -1)
1028 dataset.addSequence(dsq);
1033 dataset.addAlignmentRef();
1037 * add dataset sequences to seq for currentSeq and any sequences it references
1039 private void resolveAndAddDatasetSeq(SequenceI currentSeq,
1040 Set<SequenceI> seqs, boolean createDatasetSequence)
1042 if (currentSeq.getDatasetSequence() != null)
1044 currentSeq = currentSeq.getDatasetSequence();
1048 if (createDatasetSequence)
1050 currentSeq = currentSeq.createDatasetSequence();
1053 if (seqs.contains(currentSeq))
1057 List<SequenceI> toProcess = new ArrayList<SequenceI>();
1058 toProcess.add(currentSeq);
1059 while (toProcess.size() > 0)
1062 SequenceI curDs = toProcess.remove(0);
1063 if (seqs.contains(curDs))
1068 // iterate over database references, making sure we add forward referenced
1070 if (curDs.getDBRefs() != null)
1072 for (DBRefEntry dbr : curDs.getDBRefs())
1074 if (dbr.getMap() != null && dbr.getMap().getTo() != null)
1076 if (dbr.getMap().getTo().getDatasetSequence() != null)
1078 throw new Error("Implementation error: Map.getTo() for dbref"
1079 + dbr + " is not a dataset sequence.");
1080 // TODO: if this happens, could also rewrite the reference to
1081 // point to new dataset sequence
1083 // we recurse to add all forward references to dataset sequences via
1085 toProcess.add(dbr.getMap().getTo());
1093 * Creates a new dataset for this alignment. Can only be done once - if
1094 * dataset is not null this will not be performed.
1096 public void createDatasetAlignment()
1098 if (dataset != null)
1102 // try to avoid using SequenceI.equals at this stage, it will be expensive
1103 Set<SequenceI> seqs = new jalview.util.LinkedIdentityHashSet<SequenceI>();
1105 for (int i = 0; i < getHeight(); i++)
1107 SequenceI currentSeq = getSequenceAt(i);
1108 resolveAndAddDatasetSeq(currentSeq, seqs, true);
1111 // verify all mappings are in dataset
1112 for (AlignedCodonFrame cf : codonFrameList)
1114 for (SequenceToSequenceMapping ssm : cf.getMappings())
1116 if (!seqs.contains(ssm.getFromSeq()))
1118 resolveAndAddDatasetSeq(ssm.getFromSeq(), seqs, false);
1120 if (!seqs.contains(ssm.getMapping().getTo()))
1122 resolveAndAddDatasetSeq(ssm.getMapping().getTo(), seqs, false);
1126 // finally construct dataset
1127 dataset = new Alignment(seqs.toArray(new SequenceI[seqs.size()]));
1128 // move mappings to the dataset alignment
1129 dataset.codonFrameList = this.codonFrameList;
1130 this.codonFrameList = null;
1134 * reference count for number of alignments referencing this one.
1136 int alignmentRefs = 0;
1139 * increase reference count to this alignment.
1141 private void addAlignmentRef()
1147 public Alignment getDataset()
1153 public boolean padGaps()
1155 boolean modified = false;
1157 // Remove excess gaps from the end of alignment
1161 for (int i = 0; i < sequences.size(); i++)
1163 current = getSequenceAt(i);
1164 for (int j = current.getLength(); j > maxLength; j--)
1167 && !jalview.util.Comparison.isGap(current.getCharAt(j)))
1178 for (int i = 0; i < sequences.size(); i++)
1180 current = getSequenceAt(i);
1181 cLength = current.getLength();
1183 if (cLength < maxLength)
1185 current.insertCharAt(cLength, maxLength - cLength, gapCharacter);
1188 else if (current.getLength() > maxLength)
1190 current.deleteChars(maxLength, current.getLength());
1197 * Justify the sequences to the left or right by deleting and inserting gaps
1198 * before the initial residue or after the terminal residue
1201 * true if alignment padded to right, false to justify to left
1202 * @return true if alignment was changed
1205 public boolean justify(boolean right)
1207 boolean modified = false;
1209 // Remove excess gaps from the end of alignment
1211 int ends[] = new int[sequences.size() * 2];
1213 for (int i = 0; i < sequences.size(); i++)
1215 current = getSequenceAt(i);
1216 // This should really be a sequence method
1217 ends[i * 2] = current.findIndex(current.getStart());
1218 ends[i * 2 + 1] = current.findIndex(current.getStart()
1219 + current.getLength());
1220 boolean hitres = false;
1221 for (int j = 0, rs = 0, ssiz = current.getLength(); j < ssiz; j++)
1223 if (!jalview.util.Comparison.isGap(current.getCharAt(j)))
1232 ends[i * 2 + 1] = j;
1233 if (j - ends[i * 2] > maxLength)
1235 maxLength = j - ends[i * 2];
1243 // now edit the flanking gaps to justify to either left or right
1244 int cLength, extent, diff;
1245 for (int i = 0; i < sequences.size(); i++)
1247 current = getSequenceAt(i);
1249 cLength = 1 + ends[i * 2 + 1] - ends[i * 2];
1250 diff = maxLength - cLength; // number of gaps to indent
1251 extent = current.getLength();
1255 if (extent > ends[i * 2 + 1])
1257 current.deleteChars(ends[i * 2 + 1] + 1, extent);
1260 if (ends[i * 2] > diff)
1262 current.deleteChars(0, ends[i * 2] - diff);
1267 if (ends[i * 2] < diff)
1269 current.insertCharAt(0, diff - ends[i * 2], gapCharacter);
1277 if (ends[i * 2] > 0)
1279 current.deleteChars(0, ends[i * 2]);
1281 ends[i * 2 + 1] -= ends[i * 2];
1282 extent -= ends[i * 2];
1284 if (extent > maxLength)
1286 current.deleteChars(maxLength + 1, extent);
1291 if (extent < maxLength)
1293 current.insertCharAt(extent, maxLength - extent, gapCharacter);
1303 public HiddenSequences getHiddenSequences()
1305 return hiddenSequences;
1309 public CigarArray getCompactAlignment()
1311 synchronized (sequences)
1313 SeqCigar alseqs[] = new SeqCigar[sequences.size()];
1315 for (SequenceI seq : sequences)
1317 alseqs[i++] = new SeqCigar(seq);
1319 CigarArray cal = new CigarArray(alseqs);
1320 cal.addOperation(CigarArray.M, getWidth());
1326 public void setProperty(Object key, Object value)
1328 if (alignmentProperties == null)
1330 alignmentProperties = new Hashtable();
1333 alignmentProperties.put(key, value);
1337 public Object getProperty(Object key)
1339 if (alignmentProperties != null)
1341 return alignmentProperties.get(key);
1350 public Hashtable getProperties()
1352 return alignmentProperties;
1356 * Adds the given mapping to the stored set. Note this may be held on the
1357 * dataset alignment.
1360 public void addCodonFrame(AlignedCodonFrame codons)
1362 List<AlignedCodonFrame> acfs = getCodonFrames();
1363 if (codons != null && acfs != null && !acfs.contains(codons))
1373 * jalview.datamodel.AlignmentI#getCodonFrame(jalview.datamodel.SequenceI)
1376 public List<AlignedCodonFrame> getCodonFrame(SequenceI seq)
1382 List<AlignedCodonFrame> cframes = new ArrayList<AlignedCodonFrame>();
1383 for (AlignedCodonFrame acf : getCodonFrames())
1385 if (acf.involvesSequence(seq))
1394 * Sets the codon frame mappings (replacing any existing mappings). Note the
1395 * mappings are set on the dataset alignment instead if there is one.
1397 * @see jalview.datamodel.AlignmentI#setCodonFrames()
1400 public void setCodonFrames(List<AlignedCodonFrame> acfs)
1402 if (dataset != null)
1404 dataset.setCodonFrames(acfs);
1408 this.codonFrameList = acfs;
1413 * Returns the set of codon frame mappings. Any changes to the returned set
1414 * will affect the alignment. The mappings are held on (and read from) the
1415 * dataset alignment if there is one.
1417 * @see jalview.datamodel.AlignmentI#getCodonFrames()
1420 public List<AlignedCodonFrame> getCodonFrames()
1422 // TODO: Fix this method to fix failing AlignedCodonFrame tests
1423 // this behaviour is currently incorrect. method should return codon frames
1424 // for just the alignment,
1425 // selected from dataset
1426 return dataset != null ? dataset.getCodonFrames() : codonFrameList;
1430 * Removes the given mapping from the stored set. Note that the mappings are
1431 * held on the dataset alignment if there is one.
1434 public boolean removeCodonFrame(AlignedCodonFrame codons)
1436 List<AlignedCodonFrame> acfs = getCodonFrames();
1437 if (codons == null || acfs == null)
1441 return acfs.remove(codons);
1445 public void append(AlignmentI toappend)
1447 // TODO JAL-1270 needs test coverage
1448 // currently tested for use in jalview.gui.SequenceFetcher
1449 boolean samegap = toappend.getGapCharacter() == getGapCharacter();
1450 char oldc = toappend.getGapCharacter();
1451 boolean hashidden = toappend.getHiddenSequences() != null
1452 && toappend.getHiddenSequences().hiddenSequences != null;
1453 // get all sequences including any hidden ones
1454 List<SequenceI> sqs = (hashidden) ? toappend.getHiddenSequences()
1455 .getFullAlignment().getSequences() : toappend.getSequences();
1458 // avoid self append deadlock by
1459 List<SequenceI> toappendsq = new ArrayList<SequenceI>();
1462 for (SequenceI addedsq : sqs)
1466 char[] oldseq = addedsq.getSequence();
1467 for (int c = 0; c < oldseq.length; c++)
1469 if (oldseq[c] == oldc)
1471 oldseq[c] = gapCharacter;
1475 toappendsq.add(addedsq);
1478 for (SequenceI addedsq : toappendsq)
1480 addSequence(addedsq);
1483 AlignmentAnnotation[] alan = toappend.getAlignmentAnnotation();
1484 for (int a = 0; alan != null && a < alan.length; a++)
1486 addAnnotation(alan[a]);
1490 getCodonFrames().addAll(toappend.getCodonFrames());
1492 List<SequenceGroup> sg = toappend.getGroups();
1495 for (SequenceGroup _sg : sg)
1500 if (toappend.getHiddenSequences() != null)
1502 HiddenSequences hs = toappend.getHiddenSequences();
1503 if (hiddenSequences == null)
1505 hiddenSequences = new HiddenSequences(this);
1507 if (hs.hiddenSequences != null)
1509 for (int s = 0; s < hs.hiddenSequences.length; s++)
1511 // hide the newly appended sequence in the alignment
1512 if (hs.hiddenSequences[s] != null)
1514 hiddenSequences.hideSequence(hs.hiddenSequences[s]);
1519 if (toappend.getProperties() != null)
1521 // we really can't do very much here - just try to concatenate strings
1522 // where property collisions occur.
1523 Enumeration key = toappend.getProperties().keys();
1524 while (key.hasMoreElements())
1526 Object k = key.nextElement();
1527 Object ourval = this.getProperty(k);
1528 Object toapprop = toappend.getProperty(k);
1531 if (ourval.getClass().equals(toapprop.getClass())
1532 && !ourval.equals(toapprop))
1534 if (ourval instanceof String)
1537 this.setProperty(k, ((String) ourval) + "; "
1538 + ((String) toapprop));
1542 if (ourval instanceof Vector)
1545 Enumeration theirv = ((Vector) toapprop).elements();
1546 while (theirv.hasMoreElements())
1548 ((Vector) ourval).addElement(theirv);
1556 // just add new property directly
1557 setProperty(k, toapprop);
1565 public AlignmentAnnotation findOrCreateAnnotation(String name,
1566 String calcId, boolean autoCalc, SequenceI seqRef,
1567 SequenceGroup groupRef)
1569 assert (name != null);
1570 if (annotations != null)
1572 for (AlignmentAnnotation annot : getAlignmentAnnotation())
1574 if (annot.autoCalculated == autoCalc && (name.equals(annot.label))
1575 && (calcId == null || annot.getCalcId().equals(calcId))
1576 && annot.sequenceRef == seqRef
1577 && annot.groupRef == groupRef)
1583 AlignmentAnnotation annot = new AlignmentAnnotation(name, name,
1584 new Annotation[1], 0f, 0f, AlignmentAnnotation.BAR_GRAPH);
1585 annot.hasText = false;
1586 annot.setCalcId(new String(calcId));
1587 annot.autoCalculated = autoCalc;
1590 annot.setSequenceRef(seqRef);
1592 annot.groupRef = groupRef;
1593 addAnnotation(annot);
1599 public Iterable<AlignmentAnnotation> findAnnotation(String calcId)
1601 ArrayList<AlignmentAnnotation> aa = new ArrayList<AlignmentAnnotation>();
1602 for (AlignmentAnnotation a : getAlignmentAnnotation())
1604 if (a.getCalcId() == calcId
1605 || (a.getCalcId() != null && calcId != null && a.getCalcId()
1615 * Returns an iterable collection of any annotations that match on given
1616 * sequence ref, calcId and label (ignoring null values).
1619 public Iterable<AlignmentAnnotation> findAnnotations(SequenceI seq,
1620 String calcId, String label)
1622 ArrayList<AlignmentAnnotation> aa = new ArrayList<AlignmentAnnotation>();
1623 for (AlignmentAnnotation ann : getAlignmentAnnotation())
1625 if (ann.getCalcId() != null && ann.getCalcId().equals(calcId)
1626 && ann.sequenceRef != null && ann.sequenceRef == seq
1627 && ann.label != null && ann.label.equals(label))
1636 public void moveSelectedSequencesByOne(SequenceGroup sg,
1637 Map<SequenceI, SequenceCollectionI> map, boolean up)
1639 synchronized (sequences)
1644 for (int i = 1, iSize = sequences.size(); i < iSize; i++)
1646 SequenceI seq = sequences.get(i);
1647 if (!sg.getSequences(map).contains(seq))
1652 SequenceI temp = sequences.get(i - 1);
1653 if (sg.getSequences(null).contains(temp))
1658 sequences.set(i, temp);
1659 sequences.set(i - 1, seq);
1664 for (int i = sequences.size() - 2; i > -1; i--)
1666 SequenceI seq = sequences.get(i);
1667 if (!sg.getSequences(map).contains(seq))
1672 SequenceI temp = sequences.get(i + 1);
1673 if (sg.getSequences(map).contains(temp))
1678 sequences.set(i, temp);
1679 sequences.set(i + 1, seq);
1687 public void validateAnnotation(AlignmentAnnotation alignmentAnnotation)
1689 alignmentAnnotation.validateRangeAndDisplay();
1690 if (isNucleotide() && alignmentAnnotation.isValidStruc())
1692 hasRNAStructure = true;
1696 private SequenceI seqrep = null;
1700 * @return the representative sequence for this group
1703 public SequenceI getSeqrep()
1709 * set the representative sequence for this group. Note - this affects the
1710 * interpretation of the Hidereps attribute.
1713 * the seqrep to set (null means no sequence representative)
1716 public void setSeqrep(SequenceI seqrep)
1718 this.seqrep = seqrep;
1723 * @return true if group has a sequence representative
1726 public boolean hasSeqrep()
1728 return seqrep != null;
1732 public int getEndRes()
1734 return getWidth() - 1;
1738 public int getStartRes()
1744 * In the case of AlignmentI - returns the dataset for the alignment, if set
1747 * @see jalview.datamodel.AnnotatedCollectionI#getContext()
1750 public AnnotatedCollectionI getContext()
1756 * Align this alignment like the given (mapped) one.
1759 public int alignAs(AlignmentI al)
1762 * Currently retains unmapped gaps (in introns), regaps mapped regions
1765 return alignAs(al, false, true);
1769 * Align this alignment 'the same as' the given one. Mapped sequences only are
1770 * realigned. If both of the same type (nucleotide/protein) then align both
1771 * identically. If this is nucleotide and the other is protein, make 3 gaps
1772 * for each gap in the protein sequences. If this is protein and the other is
1773 * nucleotide, insert a gap for each 3 gaps (or part thereof) between
1774 * nucleotide bases. If this is protein and the other is nucleotide, gaps
1775 * protein to match the relative ordering of codons in the nucleotide.
1777 * Parameters control whether gaps in exon (mapped) and intron (unmapped)
1778 * regions are preserved. Gaps that connect introns to exons are treated
1779 * conservatively, i.e. only preserved if both intron and exon gaps are
1780 * preserved. TODO: check caveats below where the implementation fails
1783 * - must have same dataset, and sequences in al must have equivalent
1784 * dataset sequence and start/end bounds under given mapping
1785 * @param preserveMappedGaps
1786 * if true, gaps within and between mapped codons are preserved
1787 * @param preserveUnmappedGaps
1788 * if true, gaps within and between unmapped codons are preserved
1791 public int alignAs(AlignmentI al, boolean preserveMappedGaps,
1792 boolean preserveUnmappedGaps)
1794 // TODO should this method signature be the one in the interface?
1795 // JBPComment - yes - neither flag is used, so should be deleted.
1796 boolean thisIsNucleotide = this.isNucleotide();
1797 boolean thatIsProtein = !al.isNucleotide();
1798 if (!thatIsProtein && !thisIsNucleotide)
1800 return AlignmentUtils.alignProteinAsDna(this, al);
1802 else if (thatIsProtein && thisIsNucleotide)
1804 return AlignmentUtils.alignCdsAsProtein(this, al);
1806 return AlignmentUtils.alignAs(this, al);
1810 * Returns the alignment in Fasta format. Behaviour of this method is not
1811 * guaranteed between versions.
1814 public String toString()
1816 return new FastaFile().print(getSequencesArray());
1820 * Returns the set of distinct sequence names. No ordering is guaranteed.
1823 public Set<String> getSequenceNames()
1825 Set<String> names = new HashSet<String>();
1826 for (SequenceI seq : getSequences())
1828 names.add(seq.getName());
1834 public boolean hasValidSequence()
1836 boolean hasValidSeq = false;
1837 for (SequenceI seq : getSequences())
1839 if ((seq.getEnd() - seq.getStart()) > 0)
1849 * Update any mappings to 'virtual' sequences to compatible real ones, if
1850 * present in the added sequences. Returns a count of mappings updated.
1856 public int realiseMappings(List<SequenceI> seqs)
1859 for (SequenceI seq : seqs)
1861 for (AlignedCodonFrame mapping : getCodonFrames())
1863 count += mapping.realiseWith(seq);
1870 * Returns the first AlignedCodonFrame that has a mapping between the given
1878 public AlignedCodonFrame getMapping(SequenceI mapFrom, SequenceI mapTo)
1880 for (AlignedCodonFrame acf : getCodonFrames())
1882 if (acf.getAaForDnaSeq(mapFrom) == mapTo)
1891 public int[] getVisibleStartAndEndIndex(List<int[]> hiddenCols)
1893 int[] alignmentStartEnd = new int[] { 0, getWidth() - 1 };
1894 int startPos = alignmentStartEnd[0];
1895 int endPos = alignmentStartEnd[1];
1897 int[] lowestRange = new int[] { -1, -1 };
1898 int[] higestRange = new int[] { -1, -1 };
1900 for (int[] hiddenCol : hiddenCols)
1902 lowestRange = (hiddenCol[0] <= startPos) ? hiddenCol : lowestRange;
1903 higestRange = (hiddenCol[1] >= endPos) ? hiddenCol : higestRange;
1906 if (lowestRange[0] == -1 && lowestRange[1] == -1)
1908 startPos = alignmentStartEnd[0];
1912 startPos = lowestRange[1] + 1;
1915 if (higestRange[0] == -1 && higestRange[1] == -1)
1917 endPos = alignmentStartEnd[1];
1921 endPos = higestRange[0] - 1;
1923 return new int[] { startPos, endPos };