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.HashMap;
36 import java.util.HashSet;
37 import java.util.Hashtable;
38 import java.util.Iterator;
39 import java.util.List;
42 import java.util.Vector;
45 * Data structure to hold and manipulate a multiple sequence alignment
51 public class Alignment implements AlignmentI, AutoCloseable
53 private Alignment dataset;
55 private List<SequenceI> sequences;
57 protected List<SequenceGroup> groups;
59 protected char gapCharacter = '-';
61 private boolean nucleotide = true;
63 public boolean hasRNAStructure = false;
65 public AlignmentAnnotation[] annotations;
67 HiddenSequences hiddenSequences;
69 HiddenColumns hiddenCols;
71 public Hashtable alignmentProperties;
73 private List<AlignedCodonFrame> codonFrameList;
75 private void initAlignment(SequenceI[] seqs)
77 groups = Collections.synchronizedList(new ArrayList<SequenceGroup>());
78 hiddenSequences = new HiddenSequences(this);
79 hiddenCols = new HiddenColumns();
80 codonFrameList = new ArrayList<>();
82 nucleotide = Comparison.isNucleotide(seqs);
84 sequences = Collections.synchronizedList(new ArrayList<SequenceI>());
86 for (int i = 0; i < seqs.length; i++)
88 sequences.add(seqs[i]);
94 * Make a 'copy' alignment - sequences have new copies of features and
95 * annotations, but share the original dataset sequences.
97 public Alignment(AlignmentI al)
99 SequenceI[] seqs = al.getSequencesArray();
100 for (int i = 0; i < seqs.length; i++)
102 seqs[i] = new Sequence(seqs[i]);
108 * Share the same dataset sequence mappings (if any).
110 if (dataset == null && al.getDataset() == null)
112 this.setCodonFrames(al.getCodonFrames());
117 * Make an alignment from an array of Sequences.
121 public Alignment(SequenceI[] seqs)
127 * Make a new alignment from an array of SeqCigars
132 public Alignment(SeqCigar[] alseqs)
134 SequenceI[] seqs = SeqCigar.createAlignmentSequences(alseqs,
135 gapCharacter, new HiddenColumns(), null);
140 * Make a new alignment from an CigarArray JBPNote - can only do this when
141 * compactAlignment does not contain hidden regions. JBPNote - must also check
142 * that compactAlignment resolves to a set of SeqCigars - or construct them
145 * @param compactAlignment
148 public static AlignmentI createAlignment(CigarArray compactAlignment)
150 throw new Error(MessageManager
151 .getString("error.alignment_cigararray_not_implemented"));
152 // this(compactAlignment.refCigars);
156 public List<SequenceI> getSequences()
162 public List<SequenceI> getSequences(
163 Map<SequenceI, SequenceCollectionI> hiddenReps)
165 // TODO: in jalview 2.8 we don't do anything with hiddenreps - fix design to
171 public SequenceI[] getSequencesArray()
173 if (sequences == null)
177 synchronized (sequences)
179 return sequences.toArray(new SequenceI[sequences.size()]);
184 * Returns a map of lists of sequences keyed by sequence name.
189 public Map<String, List<SequenceI>> getSequencesByName()
191 return AlignmentUtils.getSequencesByName(this);
195 public SequenceI getSequenceAt(int i)
197 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()
308 if (getDataset() != null)
312 getDataset().removeAlignmentRef();
313 } catch (Throwable e)
323 * Defensively nulls out references in case this object is not garbage
326 void nullReferences()
332 hiddenSequences = null;
336 * decrement the alignmentRefs counter by one and null references if it goes
341 private void removeAlignmentRef() throws Throwable
343 if (--alignmentRefs == 0)
350 public void deleteSequence(SequenceI s)
352 synchronized (sequences)
354 deleteSequence(findIndex(s));
359 public void deleteSequence(int i)
361 synchronized (sequences)
363 if (i > -1 && i < getHeight())
366 hiddenSequences.adjustHeightSequenceDeleted(i);
372 public void deleteHiddenSequence(int i)
374 synchronized (sequences)
376 if (i > -1 && i < getHeight())
386 * @see jalview.datamodel.AlignmentI#findGroup(jalview.datamodel.SequenceI)
389 public SequenceGroup findGroup(SequenceI seq, int position)
391 synchronized (groups)
393 for (SequenceGroup sg : groups)
395 if (sg.getSequences(null).contains(seq))
397 if (position >= sg.getStartRes() && position <= sg.getEndRes())
411 * jalview.datamodel.AlignmentI#findAllGroups(jalview.datamodel.SequenceI)
414 public SequenceGroup[] findAllGroups(SequenceI s)
416 ArrayList<SequenceGroup> temp = new ArrayList<>();
418 synchronized (groups)
420 int gSize = groups.size();
421 for (int i = 0; i < gSize; i++)
423 SequenceGroup sg = groups.get(i);
424 if (sg == null || sg.getSequences() == null)
426 this.deleteGroup(sg);
431 if (sg.getSequences().contains(s))
437 SequenceGroup[] ret = new SequenceGroup[temp.size()];
438 return temp.toArray(ret);
443 public void addGroup(SequenceGroup sg)
445 synchronized (groups)
447 if (!groups.contains(sg))
449 if (hiddenSequences.getSize() > 0)
451 int i, iSize = sg.getSize();
452 for (i = 0; i < iSize; i++)
454 if (!sequences.contains(sg.getSequenceAt(i)))
456 sg.deleteSequence(sg.getSequenceAt(i), false);
462 if (sg.getSize() < 1)
467 sg.setContext(this, true);
474 * remove any annotation that references gp
477 * (if null, removes all group associated annotation)
479 private void removeAnnotationForGroup(SequenceGroup gp)
481 if (annotations == null || annotations.length == 0)
485 // remove annotation very quickly
486 AlignmentAnnotation[] t,
487 todelete = new AlignmentAnnotation[annotations.length],
488 tokeep = new AlignmentAnnotation[annotations.length];
492 for (i = 0, p = 0, k = 0; i < annotations.length; i++)
494 if (annotations[i].groupRef != null)
496 todelete[p++] = annotations[i];
500 tokeep[k++] = annotations[i];
506 for (i = 0, p = 0, k = 0; i < annotations.length; i++)
508 if (annotations[i].groupRef == gp)
510 todelete[p++] = annotations[i];
514 tokeep[k++] = annotations[i];
520 // clear out the group associated annotation.
521 for (i = 0; i < p; i++)
523 unhookAnnotation(todelete[i]);
526 t = new AlignmentAnnotation[k];
527 for (i = 0; i < k; i++)
536 public void deleteAllGroups()
538 synchronized (groups)
540 if (annotations != null)
542 removeAnnotationForGroup(null);
544 for (SequenceGroup sg : groups)
546 sg.setContext(null, false);
554 public void deleteGroup(SequenceGroup g)
556 synchronized (groups)
558 if (groups.contains(g))
560 removeAnnotationForGroup(g);
562 g.setContext(null, false);
569 public SequenceI findName(String name)
571 return findName(name, false);
577 * @see jalview.datamodel.AlignmentI#findName(java.lang.String, boolean)
580 public SequenceI findName(String token, boolean b)
582 return findName(null, token, b);
588 * @see jalview.datamodel.AlignmentI#findName(SequenceI, java.lang.String,
592 public SequenceI findName(SequenceI startAfter, String token, boolean b)
597 String sqname = null;
598 if (startAfter != null)
600 // try to find the sequence in the alignment
601 boolean matched = false;
602 while (i < sequences.size())
604 if (getSequenceAt(i++) == startAfter)
615 while (i < sequences.size())
617 sq = getSequenceAt(i);
618 sqname = sq.getName();
619 if (sqname.equals(token) // exact match
620 || (b && // allow imperfect matches - case varies
621 (sqname.equalsIgnoreCase(token))))
623 return getSequenceAt(i);
633 public SequenceI[] findSequenceMatch(String name)
635 Vector matches = new Vector();
638 while (i < sequences.size())
640 if (getSequenceAt(i).getName().equals(name))
642 matches.addElement(getSequenceAt(i));
647 SequenceI[] result = new SequenceI[matches.size()];
648 for (i = 0; i < result.length; i++)
650 result[i] = (SequenceI) matches.elementAt(i);
660 * @see jalview.datamodel.AlignmentI#findIndex(jalview.datamodel.SequenceI)
663 public int findIndex(SequenceI s)
667 while (i < sequences.size())
669 if (s == getSequenceAt(i))
684 * jalview.datamodel.AlignmentI#findIndex(jalview.datamodel.SearchResults)
687 public int findIndex(SearchResultsI results)
691 while (i < sequences.size())
693 if (results.involvesSequence(getSequenceAt(i)))
703 public int getHeight()
705 return sequences.size();
709 public int getAbsoluteHeight()
711 return sequences.size() + getHiddenSequences().getSize();
715 public int getWidth()
719 for (int i = 0; i < sequences.size(); i++)
721 maxLength = Math.max(maxLength, getSequenceAt(i).getLength());
727 public int getVisibleWidth()
730 if (hiddenCols != null)
732 w -= hiddenCols.getSize();
744 public void setGapCharacter(char gc)
747 synchronized (sequences)
749 for (SequenceI seq : sequences)
751 seq.setSequence(seq.getSequenceAsString().replace('.', gc)
752 .replace('-', gc).replace(' ', gc));
760 * @return DOCUMENT ME!
763 public char getGapCharacter()
771 * @see jalview.datamodel.AlignmentI#isAligned()
774 public boolean isAligned()
776 return isAligned(false);
782 * @see jalview.datamodel.AlignmentI#isAligned(boolean)
785 public boolean isAligned(boolean includeHidden)
787 int width = getWidth();
788 if (hiddenSequences == null || hiddenSequences.getSize() == 0)
790 includeHidden = true; // no hidden sequences to check against.
792 for (int i = 0; i < sequences.size(); i++)
794 if (includeHidden || !hiddenSequences.isHidden(getSequenceAt(i)))
796 if (getSequenceAt(i).getLength() != width)
807 public boolean isHidden(int alignmentIndex)
809 return (getHiddenSequences().getHiddenSequence(alignmentIndex) != null);
813 * Delete all annotations, including auto-calculated if the flag is set true.
814 * Returns true if at least one annotation was deleted, else false.
816 * @param includingAutoCalculated
820 public boolean deleteAllAnnotations(boolean includingAutoCalculated)
822 boolean result = false;
823 for (AlignmentAnnotation alan : getAlignmentAnnotation())
825 if (!alan.autoCalculated || includingAutoCalculated)
827 deleteAnnotation(alan);
837 * @seejalview.datamodel.AlignmentI#deleteAnnotation(jalview.datamodel.
838 * AlignmentAnnotation)
841 public boolean deleteAnnotation(AlignmentAnnotation aa)
843 return deleteAnnotation(aa, true);
847 public boolean deleteAnnotation(AlignmentAnnotation aa, boolean unhook)
851 if (annotations != null)
853 aSize = annotations.length;
861 AlignmentAnnotation[] temp = new AlignmentAnnotation[aSize - 1];
863 boolean swap = false;
866 for (int i = 0; i < aSize; i++)
868 if (annotations[i] == aa)
873 if (tIndex < temp.length)
875 temp[tIndex++] = annotations[i];
884 unhookAnnotation(aa);
891 * remove any object references associated with this annotation
895 private void unhookAnnotation(AlignmentAnnotation aa)
897 if (aa.sequenceRef != null)
899 aa.sequenceRef.removeAlignmentAnnotation(aa);
901 if (aa.groupRef != null)
903 // probably need to do more here in the future (post 2.5.0)
911 * @seejalview.datamodel.AlignmentI#addAnnotation(jalview.datamodel.
912 * AlignmentAnnotation)
915 public void addAnnotation(AlignmentAnnotation aa)
917 addAnnotation(aa, -1);
923 * @seejalview.datamodel.AlignmentI#addAnnotation(jalview.datamodel.
924 * AlignmentAnnotation, int)
927 public void addAnnotation(AlignmentAnnotation aa, int pos)
929 if (aa.getRNAStruc() != null)
931 hasRNAStructure = true;
935 if (annotations != null)
937 aSize = annotations.length + 1;
940 AlignmentAnnotation[] temp = new AlignmentAnnotation[aSize];
942 if (pos == -1 || pos >= aSize)
944 temp[aSize - 1] = aa;
953 for (i = 0; i < (aSize - 1); i++, p++)
961 temp[p] = annotations[i];
970 public void setAnnotationIndex(AlignmentAnnotation aa, int index)
972 if (aa == null || annotations == null || annotations.length - 1 < index)
977 int aSize = annotations.length;
978 AlignmentAnnotation[] temp = new AlignmentAnnotation[aSize];
982 for (int i = 0; i < aSize; i++)
991 temp[i] = annotations[i];
995 temp[i] = annotations[i - 1];
1004 * returns all annotation on the alignment
1006 public AlignmentAnnotation[] getAlignmentAnnotation()
1012 public boolean isNucleotide()
1018 public boolean hasRNAStructure()
1020 // TODO can it happen that structure is removed from alignment?
1021 return hasRNAStructure;
1025 public void setDataset(AlignmentI data)
1027 if (dataset == null && data == null)
1029 createDatasetAlignment();
1031 else if (dataset == null && data != null)
1035 throw new IllegalArgumentException("Circular dataset reference");
1037 if (!(data instanceof Alignment))
1040 "Implementation Error: jalview.datamodel.Alignment does not yet support other implementations of AlignmentI as its dataset reference");
1042 dataset = (Alignment) data;
1043 for (int i = 0; i < getHeight(); i++)
1045 SequenceI currentSeq = getSequenceAt(i);
1046 SequenceI dsq = currentSeq.getDatasetSequence();
1049 dsq = currentSeq.createDatasetSequence();
1050 dataset.addSequence(dsq);
1054 while (dsq.getDatasetSequence() != null)
1056 dsq = dsq.getDatasetSequence();
1058 if (dataset.findIndex(dsq) == -1)
1060 dataset.addSequence(dsq);
1065 dataset.addAlignmentRef();
1069 * add dataset sequences to seq for currentSeq and any sequences it references
1071 private void resolveAndAddDatasetSeq(SequenceI currentSeq,
1072 Set<SequenceI> seqs, boolean createDatasetSequence)
1074 SequenceI alignedSeq = currentSeq;
1075 if (currentSeq.getDatasetSequence() != null)
1077 currentSeq = currentSeq.getDatasetSequence();
1081 if (createDatasetSequence)
1083 currentSeq = currentSeq.createDatasetSequence();
1087 List<SequenceI> toProcess = new ArrayList<>();
1088 toProcess.add(currentSeq);
1089 while (toProcess.size() > 0)
1092 SequenceI curDs = toProcess.remove(0);
1094 if (!seqs.add(curDs))
1098 // iterate over database references, making sure we add forward referenced
1100 if (curDs.getDBRefs() != null)
1102 for (DBRefEntry dbr : curDs.getDBRefs())
1104 if (dbr.getMap() != null && dbr.getMap().getTo() != null)
1106 if (dbr.getMap().getTo() == alignedSeq)
1109 * update mapping to be to the newly created dataset sequence
1111 dbr.getMap().setTo(currentSeq);
1113 if (dbr.getMap().getTo().getDatasetSequence() != null)
1115 throw new Error("Implementation error: Map.getTo() for dbref "
1116 + dbr + " from " + curDs.getName()
1117 + " is not a dataset sequence.");
1119 // we recurse to add all forward references to dataset sequences via
1121 toProcess.add(dbr.getMap().getTo());
1129 * Creates a new dataset for this alignment. Can only be done once - if
1130 * dataset is not null this will not be performed.
1132 public void createDatasetAlignment()
1134 if (dataset != null)
1138 // try to avoid using SequenceI.equals at this stage, it will be expensive
1139 Set<SequenceI> seqs = new LinkedIdentityHashSet<>();
1141 for (int i = 0; i < getHeight(); i++)
1143 SequenceI currentSeq = getSequenceAt(i);
1144 resolveAndAddDatasetSeq(currentSeq, seqs, true);
1147 // verify all mappings are in dataset
1148 for (AlignedCodonFrame cf : codonFrameList)
1150 for (SequenceToSequenceMapping ssm : cf.getMappings())
1152 if (!seqs.contains(ssm.getFromSeq()))
1154 resolveAndAddDatasetSeq(ssm.getFromSeq(), seqs, false);
1156 if (!seqs.contains(ssm.getMapping().getTo()))
1158 resolveAndAddDatasetSeq(ssm.getMapping().getTo(), seqs, false);
1162 // finally construct dataset
1163 dataset = new Alignment(seqs.toArray(new SequenceI[seqs.size()]));
1164 // move mappings to the dataset alignment
1165 dataset.codonFrameList = this.codonFrameList;
1166 this.codonFrameList = null;
1170 * reference count for number of alignments referencing this one.
1172 int alignmentRefs = 0;
1175 * increase reference count to this alignment.
1177 private void addAlignmentRef()
1183 public Alignment getDataset()
1189 public boolean padGaps()
1191 boolean modified = false;
1193 // Remove excess gaps from the end of alignment
1197 for (int i = 0; i < sequences.size(); i++)
1199 current = getSequenceAt(i);
1200 for (int j = current.getLength(); j > maxLength; j--)
1203 && !jalview.util.Comparison.isGap(current.getCharAt(j)))
1214 for (int i = 0; i < sequences.size(); i++)
1216 current = getSequenceAt(i);
1217 cLength = current.getLength();
1219 if (cLength < maxLength)
1221 current.insertCharAt(cLength, maxLength - cLength, gapCharacter);
1224 else if (current.getLength() > maxLength)
1226 current.deleteChars(maxLength, current.getLength());
1233 * Justify the sequences to the left or right by deleting and inserting gaps
1234 * before the initial residue or after the terminal residue
1237 * true if alignment padded to right, false to justify to left
1238 * @return true if alignment was changed
1241 public boolean justify(boolean right)
1243 boolean modified = false;
1245 // Remove excess gaps from the end of alignment
1247 int ends[] = new int[sequences.size() * 2];
1249 for (int i = 0; i < sequences.size(); i++)
1251 current = getSequenceAt(i);
1252 // This should really be a sequence method
1253 ends[i * 2] = current.findIndex(current.getStart());
1254 ends[i * 2 + 1] = current
1255 .findIndex(current.getStart() + current.getLength());
1256 boolean hitres = false;
1257 for (int j = 0, rs = 0, ssiz = current.getLength(); j < ssiz; j++)
1259 if (!jalview.util.Comparison.isGap(current.getCharAt(j)))
1268 ends[i * 2 + 1] = j;
1269 if (j - ends[i * 2] > maxLength)
1271 maxLength = j - ends[i * 2];
1279 // now edit the flanking gaps to justify to either left or right
1280 int cLength, extent, diff;
1281 for (int i = 0; i < sequences.size(); i++)
1283 current = getSequenceAt(i);
1285 cLength = 1 + ends[i * 2 + 1] - ends[i * 2];
1286 diff = maxLength - cLength; // number of gaps to indent
1287 extent = current.getLength();
1291 if (extent > ends[i * 2 + 1])
1293 current.deleteChars(ends[i * 2 + 1] + 1, extent);
1296 if (ends[i * 2] > diff)
1298 current.deleteChars(0, ends[i * 2] - diff);
1303 if (ends[i * 2] < diff)
1305 current.insertCharAt(0, diff - ends[i * 2], gapCharacter);
1313 if (ends[i * 2] > 0)
1315 current.deleteChars(0, ends[i * 2]);
1317 ends[i * 2 + 1] -= ends[i * 2];
1318 extent -= ends[i * 2];
1320 if (extent > maxLength)
1322 current.deleteChars(maxLength + 1, extent);
1327 if (extent < maxLength)
1329 current.insertCharAt(extent, maxLength - extent, gapCharacter);
1339 public HiddenSequences getHiddenSequences()
1341 return hiddenSequences;
1345 public HiddenColumns getHiddenColumns()
1351 public CigarArray getCompactAlignment()
1353 synchronized (sequences)
1355 SeqCigar alseqs[] = new SeqCigar[sequences.size()];
1357 for (SequenceI seq : sequences)
1359 alseqs[i++] = new SeqCigar(seq);
1361 CigarArray cal = new CigarArray(alseqs);
1362 cal.addOperation(CigarArray.M, getWidth());
1368 public void setProperty(Object key, Object value)
1370 if (alignmentProperties == null)
1372 alignmentProperties = new Hashtable();
1375 alignmentProperties.put(key, value);
1379 public Object getProperty(Object key)
1381 if (alignmentProperties != null)
1383 return alignmentProperties.get(key);
1392 public Hashtable getProperties()
1394 return alignmentProperties;
1398 * Adds the given mapping to the stored set. Note this may be held on the
1399 * dataset alignment.
1402 public void addCodonFrame(AlignedCodonFrame codons)
1404 List<AlignedCodonFrame> acfs = getCodonFrames();
1405 if (codons != null && acfs != null && !acfs.contains(codons))
1415 * jalview.datamodel.AlignmentI#getCodonFrame(jalview.datamodel.SequenceI)
1418 public List<AlignedCodonFrame> getCodonFrame(SequenceI seq)
1424 List<AlignedCodonFrame> cframes = new ArrayList<>();
1425 for (AlignedCodonFrame acf : getCodonFrames())
1427 if (acf.involvesSequence(seq))
1436 * Sets the codon frame mappings (replacing any existing mappings). Note the
1437 * mappings are set on the dataset alignment instead if there is one.
1439 * @see jalview.datamodel.AlignmentI#setCodonFrames()
1442 public void setCodonFrames(List<AlignedCodonFrame> acfs)
1444 if (dataset != null)
1446 dataset.setCodonFrames(acfs);
1450 this.codonFrameList = acfs;
1455 * Returns the set of codon frame mappings. Any changes to the returned set
1456 * will affect the alignment. The mappings are held on (and read from) the
1457 * dataset alignment if there is one.
1459 * @see jalview.datamodel.AlignmentI#getCodonFrames()
1462 public List<AlignedCodonFrame> getCodonFrames()
1464 // TODO: Fix this method to fix failing AlignedCodonFrame tests
1465 // this behaviour is currently incorrect. method should return codon frames
1466 // for just the alignment,
1467 // selected from dataset
1468 return dataset != null ? dataset.getCodonFrames() : codonFrameList;
1472 * Removes the given mapping from the stored set. Note that the mappings are
1473 * held on the dataset alignment if there is one.
1476 public boolean removeCodonFrame(AlignedCodonFrame codons)
1478 List<AlignedCodonFrame> acfs = getCodonFrames();
1479 if (codons == null || acfs == null)
1483 return acfs.remove(codons);
1487 public void append(AlignmentI toappend)
1489 // TODO JAL-1270 needs test coverage
1490 // currently tested for use in jalview.gui.SequenceFetcher
1491 char oldc = toappend.getGapCharacter();
1492 boolean samegap = oldc == getGapCharacter();
1493 boolean hashidden = toappend.getHiddenSequences() != null
1494 && toappend.getHiddenSequences().hiddenSequences != null;
1495 // get all sequences including any hidden ones
1496 List<SequenceI> sqs = (hashidden)
1497 ? toappend.getHiddenSequences().getFullAlignment()
1499 : toappend.getSequences();
1502 // avoid self append deadlock by
1503 List<SequenceI> toappendsq = new ArrayList<>();
1506 for (SequenceI addedsq : sqs)
1510 addedsq.replace(oldc, gapCharacter);
1512 toappendsq.add(addedsq);
1515 for (SequenceI addedsq : toappendsq)
1517 addSequence(addedsq);
1520 AlignmentAnnotation[] alan = toappend.getAlignmentAnnotation();
1521 for (int a = 0; alan != null && a < alan.length; a++)
1523 addAnnotation(alan[a]);
1527 getCodonFrames().addAll(toappend.getCodonFrames());
1529 List<SequenceGroup> sg = toappend.getGroups();
1532 for (SequenceGroup _sg : sg)
1537 if (toappend.getHiddenSequences() != null)
1539 HiddenSequences hs = toappend.getHiddenSequences();
1540 if (hiddenSequences == null)
1542 hiddenSequences = new HiddenSequences(this);
1544 if (hs.hiddenSequences != null)
1546 for (int s = 0; s < hs.hiddenSequences.length; s++)
1548 // hide the newly appended sequence in the alignment
1549 if (hs.hiddenSequences[s] != null)
1551 hiddenSequences.hideSequence(hs.hiddenSequences[s]);
1556 if (toappend.getProperties() != null)
1558 // we really can't do very much here - just try to concatenate strings
1559 // where property collisions occur.
1560 Enumeration key = toappend.getProperties().keys();
1561 while (key.hasMoreElements())
1563 Object k = key.nextElement();
1564 Object ourval = this.getProperty(k);
1565 Object toapprop = toappend.getProperty(k);
1568 if (ourval.getClass().equals(toapprop.getClass())
1569 && !ourval.equals(toapprop))
1571 if (ourval instanceof String)
1575 ((String) ourval) + "; " + ((String) toapprop));
1579 if (ourval instanceof Vector)
1582 Enumeration theirv = ((Vector) toapprop).elements();
1583 while (theirv.hasMoreElements())
1585 ((Vector) ourval).addElement(theirv);
1593 // just add new property directly
1594 setProperty(k, toapprop);
1602 public AlignmentAnnotation findOrCreateAnnotation(String name,
1603 String calcId, boolean autoCalc, SequenceI seqRef,
1604 SequenceGroup groupRef)
1606 if (annotations != null)
1608 for (AlignmentAnnotation annot : getAlignmentAnnotation())
1610 if (annot.autoCalculated == autoCalc && (name.equals(annot.label))
1611 && (calcId == null || annot.getCalcId().equals(calcId))
1612 && annot.sequenceRef == seqRef
1613 && annot.groupRef == groupRef)
1619 AlignmentAnnotation annot = new AlignmentAnnotation(name, name,
1620 new Annotation[1], 0f, 0f, AlignmentAnnotation.BAR_GRAPH);
1621 annot.hasText = false;
1624 annot.setCalcId(new String(calcId));
1626 annot.autoCalculated = autoCalc;
1629 annot.setSequenceRef(seqRef);
1631 annot.groupRef = groupRef;
1632 addAnnotation(annot);
1638 public Iterable<AlignmentAnnotation> findAnnotation(String calcId)
1640 AlignmentAnnotation[] alignmentAnnotation = getAlignmentAnnotation();
1641 if (alignmentAnnotation != null)
1643 return AlignmentAnnotation.findAnnotation(
1644 Arrays.asList(getAlignmentAnnotation()), calcId);
1646 return Arrays.asList(new AlignmentAnnotation[] {});
1650 public Iterable<AlignmentAnnotation> findAnnotations(SequenceI seq,
1651 String calcId, String label)
1653 return AlignmentAnnotation.findAnnotations(
1654 Arrays.asList(getAlignmentAnnotation()), seq, calcId, label);
1658 public void moveSelectedSequencesByOne(SequenceGroup sg,
1659 Map<SequenceI, SequenceCollectionI> map, boolean up)
1661 synchronized (sequences)
1666 for (int i = 1, iSize = sequences.size(); i < iSize; i++)
1668 SequenceI seq = sequences.get(i);
1669 if (!sg.getSequences(map).contains(seq))
1674 SequenceI temp = sequences.get(i - 1);
1675 if (sg.getSequences(null).contains(temp))
1680 sequences.set(i, temp);
1681 sequences.set(i - 1, seq);
1686 for (int i = sequences.size() - 2; i > -1; i--)
1688 SequenceI seq = sequences.get(i);
1689 if (!sg.getSequences(map).contains(seq))
1694 SequenceI temp = sequences.get(i + 1);
1695 if (sg.getSequences(map).contains(temp))
1700 sequences.set(i, temp);
1701 sequences.set(i + 1, seq);
1709 public void validateAnnotation(AlignmentAnnotation alignmentAnnotation)
1711 alignmentAnnotation.validateRangeAndDisplay();
1712 if (isNucleotide() && alignmentAnnotation.isValidStruc())
1714 hasRNAStructure = true;
1718 private SequenceI seqrep = null;
1722 * @return the representative sequence for this group
1725 public SequenceI getSeqrep()
1731 * set the representative sequence for this group. Note - this affects the
1732 * interpretation of the Hidereps attribute.
1735 * the seqrep to set (null means no sequence representative)
1738 public void setSeqrep(SequenceI seqrep)
1740 this.seqrep = seqrep;
1745 * @return true if group has a sequence representative
1748 public boolean hasSeqrep()
1750 return seqrep != null;
1754 public int getEndRes()
1756 return getWidth() - 1;
1760 public int getStartRes()
1766 * In the case of AlignmentI - returns the dataset for the alignment, if set
1769 * @see jalview.datamodel.AnnotatedCollectionI#getContext()
1772 public AnnotatedCollectionI getContext()
1778 * Align this alignment like the given (mapped) one.
1781 public int alignAs(AlignmentI al)
1784 * Currently retains unmapped gaps (in introns), regaps mapped regions
1787 return alignAs(al, false, true);
1791 * Align this alignment 'the same as' the given one. Mapped sequences only are
1792 * realigned. If both of the same type (nucleotide/protein) then align both
1793 * identically. If this is nucleotide and the other is protein, make 3 gaps
1794 * for each gap in the protein sequences. If this is protein and the other is
1795 * nucleotide, insert a gap for each 3 gaps (or part thereof) between
1796 * nucleotide bases. If this is protein and the other is nucleotide, gaps
1797 * protein to match the relative ordering of codons in the nucleotide.
1799 * Parameters control whether gaps in exon (mapped) and intron (unmapped)
1800 * regions are preserved. Gaps that connect introns to exons are treated
1801 * conservatively, i.e. only preserved if both intron and exon gaps are
1802 * preserved. TODO: check caveats below where the implementation fails
1805 * - must have same dataset, and sequences in al must have equivalent
1806 * dataset sequence and start/end bounds under given mapping
1807 * @param preserveMappedGaps
1808 * if true, gaps within and between mapped codons are preserved
1809 * @param preserveUnmappedGaps
1810 * if true, gaps within and between unmapped codons are preserved
1813 public int alignAs(AlignmentI al, boolean preserveMappedGaps,
1814 boolean preserveUnmappedGaps)
1816 // TODO should this method signature be the one in the interface?
1817 // JBPComment - yes - neither flag is used, so should be deleted.
1818 boolean thisIsNucleotide = this.isNucleotide();
1819 boolean thatIsProtein = !al.isNucleotide();
1820 if (!thatIsProtein && !thisIsNucleotide)
1822 return AlignmentUtils.alignProteinAsDna(this, al);
1824 else if (thatIsProtein && thisIsNucleotide)
1826 return AlignmentUtils.alignCdsAsProtein(this, al);
1828 return AlignmentUtils.alignAs(this, al);
1832 * Returns the alignment in Fasta format. Behaviour of this method is not
1833 * guaranteed between versions.
1836 public String toString()
1838 return new FastaFile().print(getSequencesArray(), true);
1842 * Returns the set of distinct sequence names. No ordering is guaranteed.
1845 public Set<String> getSequenceNames()
1847 Set<String> names = new HashSet<>();
1848 for (SequenceI seq : getSequences())
1850 names.add(seq.getName());
1856 public boolean hasValidSequence()
1858 boolean hasValidSeq = false;
1859 for (SequenceI seq : getSequences())
1861 if ((seq.getEnd() - seq.getStart()) > 0)
1871 * Update any mappings to 'virtual' sequences to compatible real ones, if
1872 * present in the added sequences. Returns a count of mappings updated.
1878 public int realiseMappings(List<SequenceI> seqs)
1881 for (SequenceI seq : seqs)
1883 for (AlignedCodonFrame mapping : getCodonFrames())
1885 count += mapping.realiseWith(seq);
1892 * Returns the first AlignedCodonFrame that has a mapping between the given
1900 public AlignedCodonFrame getMapping(SequenceI mapFrom, SequenceI mapTo)
1902 for (AlignedCodonFrame acf : getCodonFrames())
1904 if (acf.getAaForDnaSeq(mapFrom) == mapTo)
1913 public boolean setHiddenColumns(HiddenColumns cols)
1915 boolean changed = cols == null ? hiddenCols != null
1916 : !cols.equals(hiddenCols);
1921 public void setupJPredAlignment()
1923 SequenceI repseq = getSequenceAt(0);
1925 HiddenColumns cs = new HiddenColumns();
1926 cs.hideList(repseq.getInsertions());
1927 setHiddenColumns(cs);
1931 public HiddenColumns propagateInsertions(SequenceI profileseq,
1932 AlignmentView input)
1936 char gc = getGapCharacter();
1937 Object[] alandhidden = input.getAlignmentAndHiddenColumns(gc);
1938 HiddenColumns nview = (HiddenColumns) alandhidden[1];
1939 SequenceI origseq = ((SequenceI[]) alandhidden[0])[profsqpos];
1940 return propagateInsertions(profileseq, origseq, nview);
1946 * sequence in al which corresponds to origseq
1948 * alignment which is to have gaps inserted into it
1950 * sequence corresponding to profileseq which defines gap map for
1953 private HiddenColumns propagateInsertions(SequenceI profileseq,
1954 SequenceI origseq, HiddenColumns hc)
1956 // take the set of hidden columns, and the set of gaps in origseq,
1957 // and remove all the hidden gaps from hiddenColumns
1959 // first get the gaps as a Bitset
1960 // then calculate hidden ^ not(gap)
1961 BitSet gaps = origseq.gapBitset();
1964 // for each sequence in the alignment, except the profile sequence,
1965 // insert gaps corresponding to each hidden region but where each hidden
1966 // column region is shifted backwards by the number of preceding visible
1967 // gaps update hidden columns at the same time
1968 HiddenColumns newhidden = new HiddenColumns();
1970 int numGapsBefore = 0;
1971 int gapPosition = 0;
1972 Iterator<int[]> it = hc.iterator();
1973 while (it.hasNext())
1975 int[] region = it.next();
1977 // get region coordinates accounting for gaps
1978 // we can rely on gaps not being *in* hidden regions because we already
1980 while (gapPosition < region[0])
1983 if (gaps.get(gapPosition))
1989 int left = region[0] - numGapsBefore;
1990 int right = region[1] - numGapsBefore;
1992 newhidden.hideColumns(left, right);
1993 padGaps(left, right, profileseq);
1999 * Pad gaps in all sequences in alignment except profileseq
2002 * position of first gap to insert
2004 * position of last gap to insert
2006 * sequence not to pad
2008 private void padGaps(int left, int right, SequenceI profileseq)
2010 char gc = getGapCharacter();
2012 // make a string with number of gaps = length of hidden region
2013 StringBuilder sb = new StringBuilder();
2014 for (int g = 0; g < right - left + 1; g++)
2019 // loop over the sequences and pad with gaps where required
2020 for (int s = 0, ns = getHeight(); s < ns; s++)
2022 SequenceI sqobj = getSequenceAt(s);
2023 if ((sqobj != profileseq) && (sqobj.getLength() >= left))
2025 String sq = sqobj.getSequenceAsString();
2027 sq.substring(0, left) + sb.toString() + sq.substring(left));
2032 Map<Object, ContactMatrixI> contactmaps = new HashMap<>();
2035 ContactListI getContactListFor(AlignmentAnnotation _aa, int column)
2037 ContactMatrixI cm = contactmaps.get(_aa.annotationId);
2042 return cm.getContactList(column);
2046 public AlignmentAnnotation addContactList(ContactMatrixI cm)
2048 Annotation _aa[] = new Annotation[getWidth()];
2049 Annotation dummy = new Annotation(0.0f);
2050 for (int i = 0; i < _aa.length; _aa[i++] = dummy)
2054 AlignmentAnnotation aa = new AlignmentAnnotation("Contact Matrix",
2055 "Contact Matrix", _aa);
2056 aa.graph = AlignmentAnnotation.CUSTOMRENDERER;
2057 aa.graphMin = cm.getMin();
2058 aa.graphMax = cm.getMax();
2059 aa.editable = false;
2060 // aa.autoCalculated = true;
2061 contactmaps.put(aa.annotationId, cm);