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.LinkedIdentityHashSet;
28 import jalview.util.MessageManager;
30 import java.util.ArrayList;
31 import java.util.Arrays;
32 import java.util.BitSet;
33 import java.util.Collections;
34 import java.util.Enumeration;
35 import java.util.HashSet;
36 import java.util.Hashtable;
37 import java.util.Iterator;
38 import java.util.List;
41 import java.util.Vector;
44 * Data structure to hold and manipulate a multiple sequence alignment
50 public class Alignment implements AlignmentI
52 private Alignment dataset;
54 private List<SequenceI> sequences;
56 protected List<SequenceGroup> groups;
58 protected char gapCharacter = '-';
60 private boolean nucleotide = true;
62 public boolean hasRNAStructure = false;
64 public AlignmentAnnotation[] annotations;
66 HiddenSequences hiddenSequences;
68 HiddenColumns hiddenCols;
70 public Hashtable alignmentProperties;
72 private List<AlignedCodonFrame> codonFrameList;
74 private void initAlignment(SequenceI[] seqs)
76 groups = Collections.synchronizedList(new ArrayList<SequenceGroup>());
77 hiddenSequences = new HiddenSequences(this);
78 hiddenCols = new HiddenColumns();
79 codonFrameList = new ArrayList<>();
81 nucleotide = Comparison.isNucleotide(seqs);
83 sequences = Collections.synchronizedList(new ArrayList<SequenceI>());
85 for (int i = 0; i < seqs.length; i++)
87 sequences.add(seqs[i]);
93 * Make a 'copy' alignment - sequences have new copies of features and
94 * annotations, but share the original dataset sequences.
96 public Alignment(AlignmentI al)
98 SequenceI[] seqs = al.getSequencesArray();
99 for (int i = 0; i < seqs.length; i++)
101 seqs[i] = new Sequence(seqs[i]);
107 * Share the same dataset sequence mappings (if any).
109 if (dataset == null && al.getDataset() == null)
111 this.setCodonFrames(al.getCodonFrames());
116 * Make an alignment from an array of Sequences.
120 public Alignment(SequenceI[] seqs)
126 * Make a new alignment from an array of SeqCigars
131 public Alignment(SeqCigar[] alseqs)
133 SequenceI[] seqs = SeqCigar.createAlignmentSequences(alseqs,
134 gapCharacter, new HiddenColumns(), null);
139 * Make a new alignment from an CigarArray JBPNote - can only do this when
140 * compactAlignment does not contain hidden regions. JBPNote - must also check
141 * that compactAlignment resolves to a set of SeqCigars - or construct them
144 * @param compactAlignment
147 public static AlignmentI createAlignment(CigarArray compactAlignment)
149 throw new Error(MessageManager
150 .getString("error.alignment_cigararray_not_implemented"));
151 // this(compactAlignment.refCigars);
155 public List<SequenceI> getSequences()
161 public List<SequenceI> getSequences(
162 Map<SequenceI, SequenceCollectionI> hiddenReps)
164 // TODO: in jalview 2.8 we don't do anything with hiddenreps - fix design to
170 public SequenceI[] getSequencesArray()
172 if (sequences == null)
176 synchronized (sequences)
178 return sequences.toArray(new SequenceI[sequences.size()]);
183 * Returns a map of lists of sequences keyed by sequence name.
188 public Map<String, List<SequenceI>> getSequencesByName()
190 return AlignmentUtils.getSequencesByName(this);
194 public SequenceI getSequenceAt(int i)
196 synchronized (sequences)
199 if (i > -1 && i < sequences.size())
201 return sequences.get(i);
209 public SequenceI getSequenceAtAbsoluteIndex(int i)
211 SequenceI seq = null;
212 if (getHiddenSequences().getSize() > 0)
214 seq = getHiddenSequences().getHiddenSequence(i);
217 // didn't find the sequence in the hidden sequences, get it from the
219 int index = getHiddenSequences().findIndexWithoutHiddenSeqs(i);
220 seq = getSequenceAt(index);
225 seq = getSequenceAt(i);
231 * Adds a sequence to the alignment. Recalculates maxLength and size. Note
232 * this currently does not recalculate whether or not the alignment is
233 * nucleotide, so mixed alignments may have undefined behaviour.
238 public void addSequence(SequenceI snew)
243 // maintain dataset integrity
244 SequenceI dsseq = snew.getDatasetSequence();
247 // derive new sequence
248 SequenceI adding = snew.deriveSequence();
250 dsseq = snew.getDatasetSequence();
252 if (getDataset().findIndex(dsseq) == -1)
254 getDataset().addSequence(dsseq);
258 if (sequences == null)
260 initAlignment(new SequenceI[] { snew });
264 synchronized (sequences)
269 if (hiddenSequences != null)
271 hiddenSequences.adjustHeightSequenceAdded();
276 public SequenceI replaceSequenceAt(int i, SequenceI snew)
278 synchronized (sequences)
280 if (sequences.size() > i)
282 return sequences.set(i, snew);
288 hiddenSequences.adjustHeightSequenceAdded();
297 * @return DOCUMENT ME!
300 public List<SequenceGroup> getGroups()
306 public void finalize() throws Throwable
308 if (getDataset() != null)
310 getDataset().removeAlignmentRef();
318 * Defensively nulls out references in case this object is not garbage
321 void nullReferences()
327 hiddenSequences = null;
331 * decrement the alignmentRefs counter by one and null references if it goes
336 private void removeAlignmentRef() throws Throwable
338 if (--alignmentRefs == 0)
345 public void deleteSequence(SequenceI s)
347 synchronized (sequences)
349 deleteSequence(findIndex(s));
354 public void deleteSequence(int i)
356 synchronized (sequences)
358 if (i > -1 && i < getHeight())
361 hiddenSequences.adjustHeightSequenceDeleted(i);
367 public void deleteHiddenSequence(int i)
369 synchronized (sequences)
371 if (i > -1 && i < getHeight())
381 * @see jalview.datamodel.AlignmentI#findGroup(jalview.datamodel.SequenceI)
384 public SequenceGroup findGroup(SequenceI seq, int position)
386 synchronized (groups)
388 for (SequenceGroup sg : groups)
390 if (sg.getSequences(null).contains(seq))
392 if (position >= sg.getStartRes() && position <= sg.getEndRes())
406 * jalview.datamodel.AlignmentI#findAllGroups(jalview.datamodel.SequenceI)
409 public SequenceGroup[] findAllGroups(SequenceI s)
411 ArrayList<SequenceGroup> temp = new ArrayList<>();
413 synchronized (groups)
415 int gSize = groups.size();
416 for (int i = 0; i < gSize; i++)
418 SequenceGroup sg = groups.get(i);
419 if (sg == null || sg.getSequences() == null)
421 this.deleteGroup(sg);
426 if (sg.getSequences().contains(s))
432 SequenceGroup[] ret = new SequenceGroup[temp.size()];
433 return temp.toArray(ret);
438 public void addGroup(SequenceGroup sg)
440 synchronized (groups)
442 if (!groups.contains(sg))
444 if (hiddenSequences.getSize() > 0)
446 int i, iSize = sg.getSize();
447 for (i = 0; i < iSize; i++)
449 if (!sequences.contains(sg.getSequenceAt(i)))
451 sg.deleteSequence(sg.getSequenceAt(i), false);
457 if (sg.getSize() < 1)
462 sg.setContext(this, true);
469 * remove any annotation that references gp
472 * (if null, removes all group associated annotation)
474 private void removeAnnotationForGroup(SequenceGroup gp)
476 if (annotations == null || annotations.length == 0)
480 // remove annotation very quickly
481 AlignmentAnnotation[] t,
482 todelete = new AlignmentAnnotation[annotations.length],
483 tokeep = new AlignmentAnnotation[annotations.length];
487 for (i = 0, p = 0, k = 0; i < annotations.length; i++)
489 if (annotations[i].groupRef != null)
491 todelete[p++] = annotations[i];
495 tokeep[k++] = annotations[i];
501 for (i = 0, p = 0, k = 0; i < annotations.length; i++)
503 if (annotations[i].groupRef == gp)
505 todelete[p++] = annotations[i];
509 tokeep[k++] = annotations[i];
515 // clear out the group associated annotation.
516 for (i = 0; i < p; i++)
518 unhookAnnotation(todelete[i]);
521 t = new AlignmentAnnotation[k];
522 for (i = 0; i < k; i++)
531 public void deleteAllGroups()
533 synchronized (groups)
535 if (annotations != null)
537 removeAnnotationForGroup(null);
539 for (SequenceGroup sg : groups)
541 sg.setContext(null, false);
549 public void deleteGroup(SequenceGroup g)
551 synchronized (groups)
553 if (groups.contains(g))
555 removeAnnotationForGroup(g);
557 g.setContext(null, false);
564 public SequenceI findName(String name)
566 return findName(name, false);
572 * @see jalview.datamodel.AlignmentI#findName(java.lang.String, boolean)
575 public SequenceI findName(String token, boolean b)
577 return findName(null, token, b);
583 * @see jalview.datamodel.AlignmentI#findName(SequenceI, java.lang.String,
587 public SequenceI findName(SequenceI startAfter, String token, boolean b)
592 String sqname = null;
593 int nseq = sequences.size();
594 if (startAfter != null)
596 // try to find the sequence in the alignment
597 boolean matched = false;
600 if (getSequenceAt(i++) == startAfter)
613 sq = getSequenceAt(i);
614 sqname = sq.getName();
615 if (sqname.equals(token) // exact match
616 || (b && // allow imperfect matches - case varies
617 (sqname.equalsIgnoreCase(token))))
619 return getSequenceAt(i);
629 public SequenceI[] findSequenceMatch(String name)
631 Vector matches = new Vector();
634 while (i < sequences.size())
636 if (getSequenceAt(i).getName().equals(name))
638 matches.addElement(getSequenceAt(i));
643 SequenceI[] result = new SequenceI[matches.size()];
644 for (i = 0; i < result.length; i++)
646 result[i] = (SequenceI) matches.elementAt(i);
656 * @see jalview.datamodel.AlignmentI#findIndex(jalview.datamodel.SequenceI)
659 public int findIndex(SequenceI s)
663 while (i < sequences.size())
665 if (s == getSequenceAt(i))
680 * jalview.datamodel.AlignmentI#findIndex(jalview.datamodel.SearchResults)
683 public int findIndex(SearchResultsI results)
687 while (i < sequences.size())
689 if (results.involvesSequence(getSequenceAt(i)))
699 public int getHeight()
701 return sequences.size();
705 public int getAbsoluteHeight()
707 return sequences.size() + getHiddenSequences().getSize();
711 public int getWidth()
715 for (int i = 0; i < sequences.size(); i++)
717 maxLength = Math.max(maxLength, getSequenceAt(i).getLength());
729 public void setGapCharacter(char gc)
732 synchronized (sequences)
734 for (SequenceI seq : sequences)
736 seq.setSequence(seq.getSequenceAsString().replace('.', gc)
737 .replace('-', gc).replace(' ', gc));
745 * @return DOCUMENT ME!
748 public char getGapCharacter()
756 * @see jalview.datamodel.AlignmentI#isAligned()
759 public boolean isAligned()
761 return isAligned(false);
767 * @see jalview.datamodel.AlignmentI#isAligned(boolean)
770 public boolean isAligned(boolean includeHidden)
772 int width = getWidth();
773 if (hiddenSequences == null || hiddenSequences.getSize() == 0)
775 includeHidden = true; // no hidden sequences to check against.
777 for (int i = 0; i < sequences.size(); i++)
779 if (includeHidden || !hiddenSequences.isHidden(getSequenceAt(i)))
781 if (getSequenceAt(i).getLength() != width)
792 public boolean isHidden(int alignmentIndex)
794 return (getHiddenSequences().getHiddenSequence(alignmentIndex) != null);
798 * Delete all annotations, including auto-calculated if the flag is set true.
799 * Returns true if at least one annotation was deleted, else false.
801 * @param includingAutoCalculated
805 public boolean deleteAllAnnotations(boolean includingAutoCalculated)
807 boolean result = false;
808 for (AlignmentAnnotation alan : getAlignmentAnnotation())
810 if (!alan.autoCalculated || includingAutoCalculated)
812 deleteAnnotation(alan);
822 * @seejalview.datamodel.AlignmentI#deleteAnnotation(jalview.datamodel.
823 * AlignmentAnnotation)
826 public boolean deleteAnnotation(AlignmentAnnotation aa)
828 return deleteAnnotation(aa, true);
832 public boolean deleteAnnotation(AlignmentAnnotation aa, boolean unhook)
836 if (annotations != null)
838 aSize = annotations.length;
846 AlignmentAnnotation[] temp = new AlignmentAnnotation[aSize - 1];
848 boolean swap = false;
851 for (int i = 0; i < aSize; i++)
853 if (annotations[i] == aa)
858 if (tIndex < temp.length)
860 temp[tIndex++] = annotations[i];
869 unhookAnnotation(aa);
876 * remove any object references associated with this annotation
880 private void unhookAnnotation(AlignmentAnnotation aa)
882 if (aa.sequenceRef != null)
884 aa.sequenceRef.removeAlignmentAnnotation(aa);
886 if (aa.groupRef != null)
888 // probably need to do more here in the future (post 2.5.0)
896 * @seejalview.datamodel.AlignmentI#addAnnotation(jalview.datamodel.
897 * AlignmentAnnotation)
900 public void addAnnotation(AlignmentAnnotation aa)
902 addAnnotation(aa, -1);
908 * @seejalview.datamodel.AlignmentI#addAnnotation(jalview.datamodel.
909 * AlignmentAnnotation, int)
912 public void addAnnotation(AlignmentAnnotation aa, int pos)
914 if (aa.getRNAStruc() != null)
916 hasRNAStructure = true;
920 if (annotations != null)
922 aSize = annotations.length + 1;
925 AlignmentAnnotation[] temp = new AlignmentAnnotation[aSize];
927 if (pos == -1 || pos >= aSize)
929 temp[aSize - 1] = aa;
938 for (i = 0; i < (aSize - 1); i++, p++)
946 temp[p] = annotations[i];
955 public void setAnnotationIndex(AlignmentAnnotation aa, int index)
957 if (aa == null || annotations == null || annotations.length - 1 < index)
962 int aSize = annotations.length;
963 AlignmentAnnotation[] temp = new AlignmentAnnotation[aSize];
967 for (int i = 0; i < aSize; i++)
976 temp[i] = annotations[i];
980 temp[i] = annotations[i - 1];
989 * returns all annotation on the alignment
991 public AlignmentAnnotation[] getAlignmentAnnotation()
997 public boolean isNucleotide()
1003 public boolean hasRNAStructure()
1005 // TODO can it happen that structure is removed from alignment?
1006 return hasRNAStructure;
1010 public void setDataset(AlignmentI data)
1012 if (dataset == null && data == null)
1014 createDatasetAlignment();
1016 else if (dataset == null && data != null)
1020 throw new IllegalArgumentException("Circular dataset reference");
1022 if (!(data instanceof Alignment))
1025 "Implementation Error: jalview.datamodel.Alignment does not yet support other implementations of AlignmentI as its dataset reference");
1027 dataset = (Alignment) data;
1028 for (int i = 0; i < getHeight(); i++)
1030 SequenceI currentSeq = getSequenceAt(i);
1031 SequenceI dsq = currentSeq.getDatasetSequence();
1034 dsq = currentSeq.createDatasetSequence();
1035 dataset.addSequence(dsq);
1039 while (dsq.getDatasetSequence() != null)
1041 dsq = dsq.getDatasetSequence();
1043 if (dataset.findIndex(dsq) == -1)
1045 dataset.addSequence(dsq);
1050 dataset.addAlignmentRef();
1054 * add dataset sequences to seq for currentSeq and any sequences it references
1056 private void resolveAndAddDatasetSeq(SequenceI currentSeq,
1057 Set<SequenceI> seqs, boolean createDatasetSequence)
1059 SequenceI alignedSeq = currentSeq;
1060 if (currentSeq.getDatasetSequence() != null)
1062 currentSeq = currentSeq.getDatasetSequence();
1066 if (createDatasetSequence)
1068 currentSeq = currentSeq.createDatasetSequence();
1072 List<SequenceI> toProcess = new ArrayList<>();
1073 toProcess.add(currentSeq);
1074 while (toProcess.size() > 0)
1077 SequenceI curDs = toProcess.remove(0);
1079 if (!seqs.add(curDs))
1083 // iterate over database references, making sure we add forward referenced
1085 if (curDs.getDBRefs() != null)
1087 for (DBRefEntry dbr : curDs.getDBRefs())
1089 if (dbr.getMap() != null && dbr.getMap().getTo() != null)
1091 if (dbr.getMap().getTo() == alignedSeq)
1094 * update mapping to be to the newly created dataset sequence
1096 dbr.getMap().setTo(currentSeq);
1098 if (dbr.getMap().getTo().getDatasetSequence() != null)
1100 throw new Error("Implementation error: Map.getTo() for dbref "
1101 + dbr + " from " + curDs.getName()
1102 + " is not a dataset sequence.");
1104 // we recurse to add all forward references to dataset sequences via
1106 toProcess.add(dbr.getMap().getTo());
1114 * Creates a new dataset for this alignment. Can only be done once - if
1115 * dataset is not null this will not be performed.
1117 public void createDatasetAlignment()
1119 if (dataset != null)
1123 // try to avoid using SequenceI.equals at this stage, it will be expensive
1124 Set<SequenceI> seqs = new LinkedIdentityHashSet<>();
1126 for (int i = 0; i < getHeight(); i++)
1128 SequenceI currentSeq = getSequenceAt(i);
1129 resolveAndAddDatasetSeq(currentSeq, seqs, true);
1132 // verify all mappings are in dataset
1133 for (AlignedCodonFrame cf : codonFrameList)
1135 for (SequenceToSequenceMapping ssm : cf.getMappings())
1137 if (!seqs.contains(ssm.getFromSeq()))
1139 resolveAndAddDatasetSeq(ssm.getFromSeq(), seqs, false);
1141 if (!seqs.contains(ssm.getMapping().getTo()))
1143 resolveAndAddDatasetSeq(ssm.getMapping().getTo(), seqs, false);
1147 // finally construct dataset
1148 dataset = new Alignment(seqs.toArray(new SequenceI[seqs.size()]));
1149 // move mappings to the dataset alignment
1150 dataset.codonFrameList = this.codonFrameList;
1151 this.codonFrameList = null;
1155 * reference count for number of alignments referencing this one.
1157 int alignmentRefs = 0;
1160 * increase reference count to this alignment.
1162 private void addAlignmentRef()
1168 public Alignment getDataset()
1174 public boolean padGaps()
1176 boolean modified = false;
1178 // Remove excess gaps from the end of alignment
1182 for (int i = 0; i < sequences.size(); i++)
1184 current = getSequenceAt(i);
1185 for (int j = current.getLength(); j > maxLength; j--)
1188 && !jalview.util.Comparison.isGap(current.getCharAt(j)))
1199 for (int i = 0; i < sequences.size(); i++)
1201 current = getSequenceAt(i);
1202 cLength = current.getLength();
1204 if (cLength < maxLength)
1206 current.insertCharAt(cLength, maxLength - cLength, gapCharacter);
1209 else if (current.getLength() > maxLength)
1211 current.deleteChars(maxLength, current.getLength());
1218 * Justify the sequences to the left or right by deleting and inserting gaps
1219 * before the initial residue or after the terminal residue
1222 * true if alignment padded to right, false to justify to left
1223 * @return true if alignment was changed
1226 public boolean justify(boolean right)
1228 boolean modified = false;
1230 // Remove excess gaps from the end of alignment
1232 int ends[] = new int[sequences.size() * 2];
1234 for (int i = 0; i < sequences.size(); i++)
1236 current = getSequenceAt(i);
1237 // This should really be a sequence method
1238 ends[i * 2] = current.findIndex(current.getStart());
1239 ends[i * 2 + 1] = current
1240 .findIndex(current.getStart() + current.getLength());
1241 boolean hitres = false;
1242 for (int j = 0, rs = 0, ssiz = current.getLength(); j < ssiz; j++)
1244 if (!jalview.util.Comparison.isGap(current.getCharAt(j)))
1253 ends[i * 2 + 1] = j;
1254 if (j - ends[i * 2] > maxLength)
1256 maxLength = j - ends[i * 2];
1264 // now edit the flanking gaps to justify to either left or right
1265 int cLength, extent, diff;
1266 for (int i = 0; i < sequences.size(); i++)
1268 current = getSequenceAt(i);
1270 cLength = 1 + ends[i * 2 + 1] - ends[i * 2];
1271 diff = maxLength - cLength; // number of gaps to indent
1272 extent = current.getLength();
1276 if (extent > ends[i * 2 + 1])
1278 current.deleteChars(ends[i * 2 + 1] + 1, extent);
1281 if (ends[i * 2] > diff)
1283 current.deleteChars(0, ends[i * 2] - diff);
1288 if (ends[i * 2] < diff)
1290 current.insertCharAt(0, diff - ends[i * 2], gapCharacter);
1298 if (ends[i * 2] > 0)
1300 current.deleteChars(0, ends[i * 2]);
1302 ends[i * 2 + 1] -= ends[i * 2];
1303 extent -= ends[i * 2];
1305 if (extent > maxLength)
1307 current.deleteChars(maxLength + 1, extent);
1312 if (extent < maxLength)
1314 current.insertCharAt(extent, maxLength - extent, gapCharacter);
1324 public HiddenSequences getHiddenSequences()
1326 return hiddenSequences;
1330 public HiddenColumns getHiddenColumns()
1336 public CigarArray getCompactAlignment()
1338 synchronized (sequences)
1340 SeqCigar alseqs[] = new SeqCigar[sequences.size()];
1342 for (SequenceI seq : sequences)
1344 alseqs[i++] = new SeqCigar(seq);
1346 CigarArray cal = new CigarArray(alseqs);
1347 cal.addOperation(CigarArray.M, getWidth());
1353 public void setProperty(Object key, Object value)
1355 if (alignmentProperties == null)
1357 alignmentProperties = new Hashtable();
1360 alignmentProperties.put(key, value);
1364 public Object getProperty(Object key)
1366 if (alignmentProperties != null)
1368 return alignmentProperties.get(key);
1377 public Hashtable getProperties()
1379 return alignmentProperties;
1383 * Adds the given mapping to the stored set. Note this may be held on the
1384 * dataset alignment.
1387 public void addCodonFrame(AlignedCodonFrame codons)
1389 List<AlignedCodonFrame> acfs = getCodonFrames();
1390 if (codons != null && acfs != null && !acfs.contains(codons))
1400 * jalview.datamodel.AlignmentI#getCodonFrame(jalview.datamodel.SequenceI)
1403 public List<AlignedCodonFrame> getCodonFrame(SequenceI seq)
1409 List<AlignedCodonFrame> cframes = new ArrayList<>();
1410 for (AlignedCodonFrame acf : getCodonFrames())
1412 if (acf.involvesSequence(seq))
1421 * Sets the codon frame mappings (replacing any existing mappings). Note the
1422 * mappings are set on the dataset alignment instead if there is one.
1424 * @see jalview.datamodel.AlignmentI#setCodonFrames()
1427 public void setCodonFrames(List<AlignedCodonFrame> acfs)
1429 if (dataset != null)
1431 dataset.setCodonFrames(acfs);
1435 this.codonFrameList = acfs;
1440 * Returns the set of codon frame mappings. Any changes to the returned set
1441 * will affect the alignment. The mappings are held on (and read from) the
1442 * dataset alignment if there is one.
1444 * @see jalview.datamodel.AlignmentI#getCodonFrames()
1447 public List<AlignedCodonFrame> getCodonFrames()
1449 // TODO: Fix this method to fix failing AlignedCodonFrame tests
1450 // this behaviour is currently incorrect. method should return codon frames
1451 // for just the alignment,
1452 // selected from dataset
1453 return dataset != null ? dataset.getCodonFrames() : codonFrameList;
1457 * Removes the given mapping from the stored set. Note that the mappings are
1458 * held on the dataset alignment if there is one.
1461 public boolean removeCodonFrame(AlignedCodonFrame codons)
1463 List<AlignedCodonFrame> acfs = getCodonFrames();
1464 if (codons == null || acfs == null)
1468 return acfs.remove(codons);
1472 public void append(AlignmentI toappend)
1474 // TODO JAL-1270 needs test coverage
1475 // currently tested for use in jalview.gui.SequenceFetcher
1476 char oldc = toappend.getGapCharacter();
1477 boolean samegap = oldc == getGapCharacter();
1478 boolean hashidden = toappend.getHiddenSequences() != null
1479 && toappend.getHiddenSequences().hiddenSequences != null;
1480 // get all sequences including any hidden ones
1481 List<SequenceI> sqs = (hashidden)
1482 ? toappend.getHiddenSequences().getFullAlignment()
1484 : toappend.getSequences();
1487 // avoid self append deadlock by
1488 List<SequenceI> toappendsq = new ArrayList<>();
1491 for (SequenceI addedsq : sqs)
1495 addedsq.replace(oldc, gapCharacter);
1497 toappendsq.add(addedsq);
1500 for (SequenceI addedsq : toappendsq)
1502 addSequence(addedsq);
1505 AlignmentAnnotation[] alan = toappend.getAlignmentAnnotation();
1506 for (int a = 0; alan != null && a < alan.length; a++)
1508 addAnnotation(alan[a]);
1512 getCodonFrames().addAll(toappend.getCodonFrames());
1514 List<SequenceGroup> sg = toappend.getGroups();
1517 for (SequenceGroup _sg : sg)
1522 if (toappend.getHiddenSequences() != null)
1524 HiddenSequences hs = toappend.getHiddenSequences();
1525 if (hiddenSequences == null)
1527 hiddenSequences = new HiddenSequences(this);
1529 if (hs.hiddenSequences != null)
1531 for (int s = 0; s < hs.hiddenSequences.length; s++)
1533 // hide the newly appended sequence in the alignment
1534 if (hs.hiddenSequences[s] != null)
1536 hiddenSequences.hideSequence(hs.hiddenSequences[s]);
1541 if (toappend.getProperties() != null)
1543 // we really can't do very much here - just try to concatenate strings
1544 // where property collisions occur.
1545 Enumeration key = toappend.getProperties().keys();
1546 while (key.hasMoreElements())
1548 Object k = key.nextElement();
1549 Object ourval = this.getProperty(k);
1550 Object toapprop = toappend.getProperty(k);
1553 if (ourval.getClass().equals(toapprop.getClass())
1554 && !ourval.equals(toapprop))
1556 if (ourval instanceof String)
1560 ((String) ourval) + "; " + ((String) toapprop));
1564 if (ourval instanceof Vector)
1567 Enumeration theirv = ((Vector) toapprop).elements();
1568 while (theirv.hasMoreElements())
1570 ((Vector) ourval).addElement(theirv);
1578 // just add new property directly
1579 setProperty(k, toapprop);
1587 public AlignmentAnnotation findOrCreateAnnotation(String name,
1588 String calcId, boolean autoCalc, SequenceI seqRef,
1589 SequenceGroup groupRef)
1591 if (annotations != null)
1593 for (AlignmentAnnotation annot : getAlignmentAnnotation())
1595 if (annot.autoCalculated == autoCalc && (name.equals(annot.label))
1596 && (calcId == null || annot.getCalcId().equals(calcId))
1597 && annot.sequenceRef == seqRef
1598 && annot.groupRef == groupRef)
1604 AlignmentAnnotation annot = new AlignmentAnnotation(name, name,
1605 new Annotation[1], 0f, 0f, AlignmentAnnotation.BAR_GRAPH);
1606 annot.hasText = false;
1609 annot.setCalcId(new String(calcId));
1611 annot.autoCalculated = autoCalc;
1614 annot.setSequenceRef(seqRef);
1616 annot.groupRef = groupRef;
1617 addAnnotation(annot);
1623 public Iterable<AlignmentAnnotation> findAnnotation(String calcId)
1625 AlignmentAnnotation[] alignmentAnnotation = getAlignmentAnnotation();
1626 if (alignmentAnnotation != null)
1628 return AlignmentAnnotation.findAnnotation(
1629 Arrays.asList(getAlignmentAnnotation()), calcId);
1631 return Arrays.asList(new AlignmentAnnotation[] {});
1635 public Iterable<AlignmentAnnotation> findAnnotations(SequenceI seq,
1636 String calcId, String label)
1638 return AlignmentAnnotation.findAnnotations(
1639 Arrays.asList(getAlignmentAnnotation()), seq, calcId, label);
1643 public void moveSelectedSequencesByOne(SequenceGroup sg,
1644 Map<SequenceI, SequenceCollectionI> map, boolean up)
1646 synchronized (sequences)
1651 for (int i = 1, iSize = sequences.size(); i < iSize; i++)
1653 SequenceI seq = sequences.get(i);
1654 if (!sg.getSequences(map).contains(seq))
1659 SequenceI temp = sequences.get(i - 1);
1660 if (sg.getSequences(null).contains(temp))
1665 sequences.set(i, temp);
1666 sequences.set(i - 1, seq);
1671 for (int i = sequences.size() - 2; i > -1; i--)
1673 SequenceI seq = sequences.get(i);
1674 if (!sg.getSequences(map).contains(seq))
1679 SequenceI temp = sequences.get(i + 1);
1680 if (sg.getSequences(map).contains(temp))
1685 sequences.set(i, temp);
1686 sequences.set(i + 1, seq);
1694 public void validateAnnotation(AlignmentAnnotation alignmentAnnotation)
1696 alignmentAnnotation.validateRangeAndDisplay();
1697 if (isNucleotide() && alignmentAnnotation.isValidStruc())
1699 hasRNAStructure = true;
1703 private SequenceI seqrep = null;
1707 * @return the representative sequence for this group
1710 public SequenceI getSeqrep()
1716 * set the representative sequence for this group. Note - this affects the
1717 * interpretation of the Hidereps attribute.
1720 * the seqrep to set (null means no sequence representative)
1723 public void setSeqrep(SequenceI seqrep)
1725 this.seqrep = seqrep;
1730 * @return true if group has a sequence representative
1733 public boolean hasSeqrep()
1735 return seqrep != null;
1739 public int getEndRes()
1741 return getWidth() - 1;
1745 public int getStartRes()
1751 * In the case of AlignmentI - returns the dataset for the alignment, if set
1754 * @see jalview.datamodel.AnnotatedCollectionI#getContext()
1757 public AnnotatedCollectionI getContext()
1763 * Align this alignment like the given (mapped) one.
1766 public int alignAs(AlignmentI al)
1769 * Currently retains unmapped gaps (in introns), regaps mapped regions
1772 return alignAs(al, false, true);
1776 * Align this alignment 'the same as' the given one. Mapped sequences only are
1777 * realigned. If both of the same type (nucleotide/protein) then align both
1778 * identically. If this is nucleotide and the other is protein, make 3 gaps
1779 * for each gap in the protein sequences. If this is protein and the other is
1780 * nucleotide, insert a gap for each 3 gaps (or part thereof) between
1781 * nucleotide bases. If this is protein and the other is nucleotide, gaps
1782 * protein to match the relative ordering of codons in the nucleotide.
1784 * Parameters control whether gaps in exon (mapped) and intron (unmapped)
1785 * regions are preserved. Gaps that connect introns to exons are treated
1786 * conservatively, i.e. only preserved if both intron and exon gaps are
1787 * preserved. TODO: check caveats below where the implementation fails
1790 * - must have same dataset, and sequences in al must have equivalent
1791 * dataset sequence and start/end bounds under given mapping
1792 * @param preserveMappedGaps
1793 * if true, gaps within and between mapped codons are preserved
1794 * @param preserveUnmappedGaps
1795 * if true, gaps within and between unmapped codons are preserved
1798 public int alignAs(AlignmentI al, boolean preserveMappedGaps,
1799 boolean preserveUnmappedGaps)
1801 // TODO should this method signature be the one in the interface?
1802 // JBPComment - yes - neither flag is used, so should be deleted.
1803 boolean thisIsNucleotide = this.isNucleotide();
1804 boolean thatIsProtein = !al.isNucleotide();
1805 if (!thatIsProtein && !thisIsNucleotide)
1807 return AlignmentUtils.alignProteinAsDna(this, al);
1809 else if (thatIsProtein && thisIsNucleotide)
1811 return AlignmentUtils.alignCdsAsProtein(this, al);
1813 return AlignmentUtils.alignAs(this, al);
1817 * Returns the alignment in Fasta format. Behaviour of this method is not
1818 * guaranteed between versions.
1821 public String toString()
1823 return new FastaFile().print(getSequencesArray(), true);
1827 * Returns the set of distinct sequence names. No ordering is guaranteed.
1830 public Set<String> getSequenceNames()
1832 Set<String> names = new HashSet<>();
1833 for (SequenceI seq : getSequences())
1835 names.add(seq.getName());
1841 public boolean hasValidSequence()
1843 boolean hasValidSeq = false;
1844 for (SequenceI seq : getSequences())
1846 if ((seq.getEnd() - seq.getStart()) > 0)
1856 * Update any mappings to 'virtual' sequences to compatible real ones, if
1857 * present in the added sequences. Returns a count of mappings updated.
1863 public int realiseMappings(List<SequenceI> seqs)
1866 for (SequenceI seq : seqs)
1868 for (AlignedCodonFrame mapping : getCodonFrames())
1870 count += mapping.realiseWith(seq);
1877 * Returns the first AlignedCodonFrame that has a mapping between the given
1885 public AlignedCodonFrame getMapping(SequenceI mapFrom, SequenceI mapTo)
1887 for (AlignedCodonFrame acf : getCodonFrames())
1889 if (acf.getAaForDnaSeq(mapFrom) == mapTo)
1898 public void setHiddenColumns(HiddenColumns cols)
1904 public void setupJPredAlignment()
1906 SequenceI repseq = getSequenceAt(0);
1908 HiddenColumns cs = new HiddenColumns();
1909 cs.hideList(repseq.getInsertions());
1910 setHiddenColumns(cs);
1914 public HiddenColumns propagateInsertions(SequenceI profileseq,
1915 AlignmentView input)
1919 char gc = getGapCharacter();
1920 Object[] alandhidden = input.getAlignmentAndHiddenColumns(gc);
1921 HiddenColumns nview = (HiddenColumns) alandhidden[1];
1922 SequenceI origseq = ((SequenceI[]) alandhidden[0])[profsqpos];
1923 return propagateInsertions(profileseq, origseq, nview);
1929 * sequence in al which corresponds to origseq
1931 * alignment which is to have gaps inserted into it
1933 * sequence corresponding to profileseq which defines gap map for
1936 private HiddenColumns propagateInsertions(SequenceI profileseq,
1937 SequenceI origseq, HiddenColumns hc)
1939 // take the set of hidden columns, and the set of gaps in origseq,
1940 // and remove all the hidden gaps from hiddenColumns
1942 // first get the gaps as a Bitset
1943 // then calculate hidden ^ not(gap)
1944 BitSet gaps = origseq.gapBitset();
1947 // for each sequence in the alignment, except the profile sequence,
1948 // insert gaps corresponding to each hidden region but where each hidden
1949 // column region is shifted backwards by the number of preceding visible
1950 // gaps update hidden columns at the same time
1951 HiddenColumns newhidden = new HiddenColumns();
1953 int numGapsBefore = 0;
1954 int gapPosition = 0;
1955 Iterator<int[]> it = hc.iterator();
1956 while (it.hasNext())
1958 int[] region = it.next();
1960 // get region coordinates accounting for gaps
1961 // we can rely on gaps not being *in* hidden regions because we already
1963 while (gapPosition < region[0])
1966 if (gaps.get(gapPosition))
1972 int left = region[0] - numGapsBefore;
1973 int right = region[1] - numGapsBefore;
1975 newhidden.hideColumns(left, right);
1976 padGaps(left, right, profileseq);
1982 * Pad gaps in all sequences in alignment except profileseq
1985 * position of first gap to insert
1987 * position of last gap to insert
1989 * sequence not to pad
1991 private void padGaps(int left, int right, SequenceI profileseq)
1993 char gc = getGapCharacter();
1995 // make a string with number of gaps = length of hidden region
1996 StringBuilder sb = new StringBuilder();
1997 for (int g = 0; g < right - left + 1; g++)
2002 // loop over the sequences and pad with gaps where required
2003 for (int s = 0, ns = getHeight(); s < ns; s++)
2005 SequenceI sqobj = getSequenceAt(s);
2006 if ((sqobj != profileseq) && (sqobj.getLength() >= left))
2008 String sq = sqobj.getSequenceAsString();
2010 sq.substring(0, left) + sb.toString() + sq.substring(left));