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.util.MapList;
24 import jalview.util.MappingUtils;
26 import java.util.AbstractList;
27 import java.util.ArrayList;
28 import java.util.List;
31 * Stores mapping between the columns of a protein alignment and a DNA alignment
32 * and a list of individual codon to amino acid mappings between sequences.
34 public class AlignedCodonFrame
38 * Data bean to hold mappings from one sequence to another
40 public class SequenceToSequenceMapping
42 private SequenceI fromSeq;
44 private Mapping mapping;
46 SequenceToSequenceMapping(SequenceI from, Mapping map)
53 * Readable representation for debugging only, not guaranteed not to change
56 public String toString()
58 return String.format("From %s %s", fromSeq.getName(),
63 * Returns a hashCode derived from the hashcodes of the mappings and fromSeq
65 * @see SequenceToSequenceMapping#hashCode()
70 return (fromSeq == null ? 0 : fromSeq.hashCode() * 31)
75 * Answers true if the objects hold the same mapping between the same two
81 public boolean equals(Object obj)
83 if (!(obj instanceof SequenceToSequenceMapping))
87 SequenceToSequenceMapping that = (SequenceToSequenceMapping) obj;
88 if (this.mapping == null)
90 return that.mapping == null;
92 // TODO: can simplify by asserting fromSeq is a dataset sequence
93 return (this.fromSeq == that.fromSeq
94 || (this.fromSeq != null && that.fromSeq != null
95 && this.fromSeq.getDatasetSequence() != null
96 && this.fromSeq.getDatasetSequence() == that.fromSeq
97 .getDatasetSequence()))
98 && this.mapping.equals(that.mapping);
101 public SequenceI getFromSeq()
106 public Mapping getMapping()
112 private List<SequenceToSequenceMapping> mappings;
117 public AlignedCodonFrame()
119 mappings = new ArrayList<>();
123 * Adds a mapping between the dataset sequences for the associated dna and
124 * protein sequence objects
130 public void addMap(SequenceI dnaseq, SequenceI aaseq, MapList map)
132 addMap(dnaseq, aaseq, map, null);
136 * Adds a mapping between the dataset sequences for the associated dna and
137 * protein sequence objects
144 public void addMap(SequenceI dnaseq, SequenceI aaseq, MapList map,
147 // JBPNote DEBUG! THIS !
148 // dnaseq.transferAnnotation(aaseq, mp);
149 // aaseq.transferAnnotation(dnaseq, new Mapping(map.getInverse()));
151 SequenceI fromSeq = (dnaseq.getDatasetSequence() == null) ? dnaseq
152 : dnaseq.getDatasetSequence();
153 SequenceI toSeq = (aaseq.getDatasetSequence() == null) ? aaseq
154 : aaseq.getDatasetSequence();
157 * if we already hold a mapping between these sequences, just add to it
158 * note that 'adding' a duplicate map does nothing; this protects against
159 * creating duplicate mappings in AlignedCodonFrame
161 for (SequenceToSequenceMapping ssm : mappings)
163 if (ssm.fromSeq == fromSeq && ssm.mapping.to == toSeq)
165 ssm.mapping.map.addMapList(map);
171 * otherwise, add a new sequence mapping
173 Mapping mp = new Mapping(toSeq, map);
174 mp.setMappedFromId(mapFromId);
175 mappings.add(new SequenceToSequenceMapping(fromSeq, mp));
178 public SequenceI[] getdnaSeqs()
180 // TODO return a list instead?
182 List<SequenceI> seqs = new ArrayList<>();
183 for (SequenceToSequenceMapping ssm : mappings)
185 seqs.add(ssm.fromSeq);
187 return seqs.toArray(new SequenceI[seqs.size()]);
190 public SequenceI[] getAaSeqs()
192 // TODO not used - remove?
193 List<SequenceI> seqs = new ArrayList<>();
194 for (SequenceToSequenceMapping ssm : mappings)
196 seqs.add(ssm.mapping.to);
198 return seqs.toArray(new SequenceI[seqs.size()]);
201 public MapList[] getdnaToProt()
203 List<MapList> maps = new ArrayList<>();
204 for (SequenceToSequenceMapping ssm : mappings)
206 maps.add(ssm.mapping.map);
208 return maps.toArray(new MapList[maps.size()]);
211 public Mapping[] getProtMappings()
213 List<Mapping> maps = new ArrayList<>();
214 for (SequenceToSequenceMapping ssm : mappings)
216 maps.add(ssm.mapping);
218 return maps.toArray(new Mapping[maps.size()]);
222 * Returns the first mapping found which is to or from the given sequence, or
228 public Mapping getMappingForSequence(SequenceI seq, boolean cdsOnly)
230 SequenceI seqDs = seq.getDatasetSequence();
231 seqDs = seqDs != null ? seqDs : seq;
233 for (SequenceToSequenceMapping ssm : mappings)
235 if (ssm.fromSeq == seqDs || ssm.mapping.to == seqDs)
237 if (!cdsOnly || ssm.fromSeq.getName().startsWith("CDS")
238 || ssm.mapping.to.getName().startsWith("CDS"))
248 * Return the corresponding aligned or dataset aa sequence for given dna
249 * sequence, null if not found.
254 public SequenceI getAaForDnaSeq(SequenceI dnaSeqRef)
256 SequenceI dnads = dnaSeqRef.getDatasetSequence();
257 for (SequenceToSequenceMapping ssm : mappings)
259 if (ssm.fromSeq == dnaSeqRef || ssm.fromSeq == dnads)
261 return ssm.mapping.to;
270 * @return null or corresponding aaSeq entry for dnaSeq entry
272 public SequenceI getDnaForAaSeq(SequenceI aaSeqRef)
274 SequenceI aads = aaSeqRef.getDatasetSequence();
275 for (SequenceToSequenceMapping ssm : mappings)
277 if (ssm.mapping.to == aaSeqRef || ssm.mapping.to == aads)
286 * test to see if codon frame involves seq in any way
289 * a nucleotide or protein sequence
290 * @return true if a mapping exists to or from this sequence to any translated
293 public boolean involvesSequence(SequenceI seq)
295 return getAaForDnaSeq(seq) != null || getDnaForAaSeq(seq) != null;
299 * Add search results for regions in other sequences that translate or are
300 * translated from a particular position in seq
306 * where highlighted regions go
308 public void markMappedRegion(SequenceI seq, int index,
309 SearchResultsI results)
312 SequenceI ds = seq.getDatasetSequence();
313 for (SequenceToSequenceMapping ssm : mappings)
315 if (ssm.fromSeq == seq || ssm.fromSeq == ds)
317 codon = ssm.mapping.map.locateInTo(index, index);
320 for (int i = 0; i < codon.length; i += 2)
322 results.addResult(ssm.mapping.to, codon[i], codon[i + 1]);
326 else if (ssm.mapping.to == seq || ssm.mapping.to == ds)
329 codon = ssm.mapping.map.locateInFrom(index, index);
332 for (int i = 0; i < codon.length; i += 2)
334 results.addResult(ssm.fromSeq, codon[i], codon[i + 1]);
343 * Returns the DNA codon positions (base 1) for the given position (base 1) in
344 * a mapped protein sequence, or null if no mapping is found.
346 * Intended for use in aligning cDNA to match aligned protein. Only the first
347 * mapping found is returned, so not suitable for use if multiple protein
348 * sequences are mapped to the same cDNA (but aligning cDNA as protein is
349 * ill-defined for this case anyway).
352 * the DNA dataset sequence
354 * residue position (base 1) in a protein sequence
357 public int[] getDnaPosition(SequenceI seq, int aaPos)
360 * Adapted from markMappedRegion().
364 for (SequenceToSequenceMapping ssm : mappings)
366 if (ssm.fromSeq == seq)
368 ml = getdnaToProt()[i];
373 return ml == null ? null : ml.locateInFrom(aaPos, aaPos);
377 * Convenience method to return the first aligned sequence in the given
378 * alignment whose dataset has a mapping with the given (aligned or dataset)
386 public SequenceI findAlignedSequence(SequenceI seq, AlignmentI al)
389 * Search mapped protein ('to') sequences first.
391 for (SequenceToSequenceMapping ssm : mappings)
393 if (ssm.fromSeq == seq || ssm.fromSeq == seq.getDatasetSequence())
395 for (SequenceI sourceAligned : al.getSequences())
397 if (ssm.mapping.to == sourceAligned.getDatasetSequence()
398 || ssm.mapping.to == sourceAligned)
400 return sourceAligned;
407 * Then try mapped dna sequences.
409 for (SequenceToSequenceMapping ssm : mappings)
411 if (ssm.mapping.to == seq
412 || ssm.mapping.to == seq.getDatasetSequence())
414 for (SequenceI sourceAligned : al.getSequences())
416 if (ssm.fromSeq == sourceAligned.getDatasetSequence())
418 return sourceAligned;
428 * Returns the region in the target sequence's dataset that is mapped to the
429 * given position (base 1) in the query sequence's dataset. The region is a
430 * set of start/end position pairs.
437 public int[] getMappedRegion(SequenceI target, SequenceI query,
440 SequenceI targetDs = target.getDatasetSequence() == null ? target
441 : target.getDatasetSequence();
442 SequenceI queryDs = query.getDatasetSequence() == null ? query
443 : query.getDatasetSequence();
444 if (targetDs == null || queryDs == null /*|| dnaToProt == null*/)
448 for (SequenceToSequenceMapping ssm : mappings)
451 * try mapping from target to query
453 if (ssm.fromSeq == targetDs && ssm.mapping.to == queryDs)
455 int[] codon = ssm.mapping.map.locateInFrom(queryPos, queryPos);
462 * else try mapping from query to target
464 else if (ssm.fromSeq == queryDs && ssm.mapping.to == targetDs)
466 int[] codon = ssm.mapping.map.locateInTo(queryPos, queryPos);
477 * Returns the mapped DNA codons for the given position in a protein sequence,
478 * or null if no mapping is found. Returns a list of (e.g.) ['g', 'c', 't']
479 * codons. There may be more than one codon mapped to the protein if (for
480 * example), there are mappings to cDNA variants.
483 * the peptide dataset sequence
485 * residue position (base 1) in the peptide sequence
488 public List<char[]> getMappedCodons(SequenceI protein, int aaPos)
491 SequenceI dnaSeq = null;
492 List<char[]> result = new ArrayList<>();
494 for (SequenceToSequenceMapping ssm : mappings)
496 if (ssm.mapping.to == protein
497 && ssm.mapping.getMap().getFromRatio() == 3)
499 ml = ssm.mapping.map;
500 dnaSeq = ssm.fromSeq;
502 int[] codonPos = ml.locateInFrom(aaPos, aaPos);
503 if (codonPos == null)
509 * Read off the mapped nucleotides (converting to position base 0)
511 codonPos = MappingUtils.flattenRanges(codonPos);
512 int start = dnaSeq.getStart();
513 char c1 = dnaSeq.getCharAt(codonPos[0] - start);
514 char c2 = dnaSeq.getCharAt(codonPos[1] - start);
515 char c3 = dnaSeq.getCharAt(codonPos[2] - start);
516 result.add(new char[] { c1, c2, c3 });
519 return result.isEmpty() ? null : result;
523 * Returns any mappings found which are from the given sequence, and to
524 * distinct sequences.
529 public List<Mapping> getMappingsFromSequence(SequenceI seq)
531 List<Mapping> result = new ArrayList<>();
532 List<SequenceI> related = new ArrayList<>();
533 SequenceI seqDs = seq.getDatasetSequence();
534 seqDs = seqDs != null ? seqDs : seq;
536 for (SequenceToSequenceMapping ssm : mappings)
538 final Mapping mapping = ssm.mapping;
539 if (ssm.fromSeq == seqDs)
541 if (!related.contains(mapping.to))
544 related.add(mapping.to);
552 * Test whether the given sequence is substitutable for one or more dummy
553 * sequences in this mapping
559 public boolean isRealisableWith(SequenceI seq)
561 return realiseWith(seq, false) > 0;
565 * Replace any matchable mapped dummy sequences with the given real one.
566 * Returns the count of sequence mappings instantiated.
571 public int realiseWith(SequenceI seq)
573 return realiseWith(seq, true);
577 * Returns the number of mapped dummy sequences that could be replaced with
578 * the given real sequence.
583 * if true, performs replacements, else only counts
586 protected int realiseWith(SequenceI seq, boolean doUpdate)
588 SequenceI ds = seq.getDatasetSequence() != null
589 ? seq.getDatasetSequence()
594 * check for replaceable DNA ('map from') sequences
596 for (SequenceToSequenceMapping ssm : mappings)
598 SequenceI dna = ssm.fromSeq;
599 if (dna instanceof SequenceDummy
600 && dna.getName().equals(ds.getName()))
602 Mapping mapping = ssm.mapping;
603 int mapStart = mapping.getMap().getFromLowest();
604 int mapEnd = mapping.getMap().getFromHighest();
605 boolean mappable = couldRealiseSequence(dna, ds, mapStart, mapEnd);
611 // TODO: new method ? ds.realise(dna);
612 // might want to copy database refs as well
613 ds.setSequenceFeatures(dna.getSequenceFeatures());
616 System.out.println("Realised mapped sequence " + ds.getName());
622 * check for replaceable protein ('map to') sequences
624 Mapping mapping = ssm.mapping;
625 SequenceI prot = mapping.getTo();
626 int mapStart = mapping.getMap().getToLowest();
627 int mapEnd = mapping.getMap().getToHighest();
628 boolean mappable = couldRealiseSequence(prot, ds, mapStart, mapEnd);
634 // TODO: new method ? ds.realise(dna);
635 // might want to copy database refs as well
636 ds.setSequenceFeatures(dna.getSequenceFeatures());
637 ssm.mapping.setTo(ds);
645 * Helper method to test whether a 'real' sequence could replace a 'dummy'
646 * sequence in the map. The criteria are that they have the same name, and
647 * that the mapped region overlaps the candidate sequence.
655 protected static boolean couldRealiseSequence(SequenceI existing,
656 SequenceI replacement, int mapStart, int mapEnd)
658 if (existing instanceof SequenceDummy
659 && !(replacement instanceof SequenceDummy)
660 && existing.getName().equals(replacement.getName()))
662 int start = replacement.getStart();
663 int end = replacement.getEnd();
664 boolean mappingOverlapsSequence = (mapStart >= start
665 && mapStart <= end) || (mapEnd >= start && mapEnd <= end);
666 if (mappingOverlapsSequence)
675 * Change any mapping to the given sequence to be to its dataset sequence
676 * instead. For use when mappings are created before their referenced
677 * sequences are instantiated, for example when parsing GFF data.
681 public void updateToDataset(SequenceI seq)
683 if (seq == null || seq.getDatasetSequence() == null)
687 SequenceI ds = seq.getDatasetSequence();
689 for (SequenceToSequenceMapping ssm : mappings)
694 if (ssm.fromSeq == seq)
702 if (ssm.mapping.to == seq)
710 * Answers true if this object contains no mappings
714 public boolean isEmpty()
716 return mappings.isEmpty();
720 * Method for debug / inspection purposes only, may change in future
723 public String toString()
725 return mappings == null ? "null" : mappings.toString();
729 * Returns the first mapping found that is between 'fromSeq' and 'toSeq', or
733 * aligned or dataset sequence
735 * aligned or dataset sequence
738 public Mapping getMappingBetween(SequenceI fromSeq, SequenceI toSeq)
740 SequenceI dssFrom = fromSeq.getDatasetSequence() == null ? fromSeq
741 : fromSeq.getDatasetSequence();
742 SequenceI dssTo = toSeq.getDatasetSequence() == null ? toSeq
743 : toSeq.getDatasetSequence();
745 for (SequenceToSequenceMapping mapping : mappings)
747 SequenceI from = mapping.fromSeq;
748 SequenceI to = mapping.mapping.to;
749 if ((from == dssFrom && to == dssTo)
750 || (from == dssTo && to == dssFrom))
752 return mapping.mapping;
759 * Returns a hashcode derived from the list of sequence mappings
761 * @see SequenceToSequenceMapping#hashCode()
762 * @see AbstractList#hashCode()
765 public int hashCode()
767 return this.mappings.hashCode();
771 * Two AlignedCodonFrame objects are equal if they hold the same ordered list
774 * @see SequenceToSequenceMapping#
777 public boolean equals(Object obj)
779 if (!(obj instanceof AlignedCodonFrame))
783 return this.mappings.equals(((AlignedCodonFrame) obj).mappings);
786 public List<SequenceToSequenceMapping> getMappings()