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.
23 import jalview.analysis.AlignmentUtils;
24 import jalview.analysis.CrossRef;
25 import jalview.api.AlignmentViewPanel;
26 import jalview.api.FeatureSettingsModelI;
27 import jalview.bin.Cache;
28 import jalview.bin.Console;
29 import jalview.datamodel.Alignment;
30 import jalview.datamodel.AlignmentI;
31 import jalview.datamodel.DBRefEntry;
32 import jalview.datamodel.DBRefSource;
33 import jalview.datamodel.GeneLociI;
34 import jalview.datamodel.SequenceI;
35 import jalview.ext.ensembl.EnsemblInfo;
36 import jalview.ext.ensembl.EnsemblMap;
37 import jalview.io.gff.SequenceOntologyI;
38 import jalview.structure.StructureSelectionManager;
39 import jalview.util.DBRefUtils;
40 import jalview.util.MapList;
41 import jalview.util.MappingUtils;
42 import jalview.util.MessageManager;
43 import jalview.viewmodel.seqfeatures.FeatureRendererModel;
44 import jalview.ws.SequenceFetcher;
46 import java.util.ArrayList;
47 import java.util.HashMap;
48 import java.util.List;
53 * Factory constructor and runnable for discovering and displaying
54 * cross-references for a set of aligned sequences
59 public class CrossRefAction implements Runnable
61 private AlignFrame alignFrame;
63 private SequenceI[] sel;
65 private final boolean _odna;
67 private String source;
69 List<AlignmentViewPanel> xrefViews = new ArrayList<>();
71 List<AlignmentViewPanel> getXrefViews()
79 final long sttime = System.currentTimeMillis();
80 alignFrame.setProgressBar(MessageManager.formatMessage(
81 "status.searching_for_sequences_from", new Object[]
85 AlignmentI alignment = alignFrame.getViewport().getAlignment();
86 AlignmentI dataset = alignment.getDataset() == null ? alignment
87 : alignment.getDataset();
88 boolean dna = alignment.isNucleotide();
92 .println("Conflict: showProducts for alignment originally "
93 + "thought to be " + (_odna ? "DNA" : "Protein")
94 + " now searching for " + (dna ? "DNA" : "Protein")
97 AlignmentI xrefs = new CrossRef(sel, dataset)
98 .findXrefSequences(source, dna);
105 * try to look up chromosomal coordinates for nucleotide
106 * sequences (if not already retrieved)
108 findGeneLoci(xrefs.getSequences());
111 * get display scheme (if any) to apply to features
113 FeatureSettingsModelI featureColourScheme = new SequenceFetcher()
114 .getFeatureColourScheme(source);
116 if (dna && AlignmentUtils.looksLikeEnsembl(alignment))
118 // override default featureColourScheme so products have Ensembl variant colours
119 featureColourScheme = new SequenceFetcher()
120 .getFeatureColourScheme(DBRefSource.ENSEMBL);
123 AlignmentI xrefsAlignment = makeCrossReferencesAlignment(dataset,
127 xrefsAlignment = AlignmentUtils.makeCdsAlignment(
128 xrefsAlignment.getSequencesArray(), dataset, sel);
129 xrefsAlignment.alignAs(alignment);
133 * If we are opening a splitframe, make a copy of this alignment (sharing the same dataset
134 * sequences). If we are DNA, drop introns and update mappings
136 AlignmentI copyAlignment = null;
138 if (Cache.getDefault(Preferences.ENABLE_SPLIT_FRAME, true))
140 copyAlignment = copyAlignmentForSplitFrame(alignment, dataset, dna,
141 xrefs, xrefsAlignment);
142 if (copyAlignment == null)
149 * build AlignFrame(s) according to available alignment data
151 AlignFrame newFrame = new AlignFrame(xrefsAlignment,
152 AlignFrame.DEFAULT_WIDTH, AlignFrame.DEFAULT_HEIGHT);
153 if (Cache.getDefault("HIDE_INTRONS", true))
155 newFrame.hideFeatureColumns(SequenceOntologyI.EXON, false);
157 String newtitle = String.format("%s %s %s",
158 dna ? MessageManager.getString("label.proteins")
159 : MessageManager.getString("label.nucleotides"),
160 MessageManager.getString("label.for"), alignFrame.getTitle());
161 newFrame.setTitle(newtitle);
163 if (copyAlignment == null)
166 * split frame display is turned off in preferences file
168 Desktop.addInternalFrame(newFrame, newtitle,
169 AlignFrame.DEFAULT_WIDTH, AlignFrame.DEFAULT_HEIGHT);
170 xrefViews.add(newFrame.alignPanel);
171 return; // via finally clause
174 AlignFrame copyThis = new AlignFrame(copyAlignment,
175 AlignFrame.DEFAULT_WIDTH, AlignFrame.DEFAULT_HEIGHT);
176 copyThis.setTitle(alignFrame.getTitle());
178 boolean showSequenceFeatures = alignFrame.getViewport()
179 .isShowSequenceFeatures();
180 newFrame.setShowSeqFeatures(showSequenceFeatures);
181 copyThis.setShowSeqFeatures(showSequenceFeatures);
182 FeatureRendererModel myFeatureStyling = alignFrame.alignPanel
183 .getSeqPanel().seqCanvas.getFeatureRenderer();
186 * copy feature rendering settings to split frame
188 FeatureRendererModel fr1 = newFrame.alignPanel.getSeqPanel().seqCanvas
189 .getFeatureRenderer();
190 fr1.transferSettings(myFeatureStyling);
191 fr1.findAllFeatures(true);
192 FeatureRendererModel fr2 = copyThis.alignPanel.getSeqPanel().seqCanvas
193 .getFeatureRenderer();
194 fr2.transferSettings(myFeatureStyling);
195 fr2.findAllFeatures(true);
198 * apply 'database source' feature configuration
199 * if any - first to the new splitframe view about to be displayed
202 newFrame.getViewport().applyFeaturesStyle(featureColourScheme);
203 copyThis.getViewport().applyFeaturesStyle(featureColourScheme);
206 * and for JAL-3330 also to original alignFrame view(s)
207 * this currently trashes any original settings.
209 for (AlignmentViewPanel origpanel: alignFrame.getAlignPanels()) {
210 origpanel.getAlignViewport()
211 .mergeFeaturesStyle(featureColourScheme);
214 SplitFrame sf = new SplitFrame(dna ? copyThis : newFrame,
215 dna ? newFrame : copyThis);
217 newFrame.setVisible(true);
218 copyThis.setVisible(true);
219 String linkedTitle = MessageManager
220 .getString("label.linked_view_title");
221 Desktop.addInternalFrame(sf, linkedTitle, -1, -1);
222 sf.adjustInitialLayout();
224 // finally add the top, then bottom frame to the view list
225 xrefViews.add(dna ? copyThis.alignPanel : newFrame.alignPanel);
226 xrefViews.add(!dna ? copyThis.alignPanel : newFrame.alignPanel);
228 } catch (OutOfMemoryError e)
230 new OOMWarning("whilst fetching crossreferences", e);
231 } catch (Throwable e)
233 Console.error("Error when finding crossreferences", e);
236 alignFrame.setProgressBar(MessageManager.formatMessage(
237 "status.finished_searching_for_sequences_from", new Object[]
238 { source }), sttime);
243 * Tries to add chromosomal coordinates to any nucleotide sequence which does
244 * not already have them. Coordinates are retrieved from Ensembl given an
245 * Ensembl identifier, either on the sequence itself or on a peptide sequence
246 * it has a reference to.
250 * - fetch EMBLCDS cross-references for Uniprot entry P30419
251 * - the EMBL sequences do not have xrefs to Ensembl
252 * - the Uniprot entry has xrefs to
253 * ENSP00000258960, ENSP00000468424, ENST00000258960, ENST00000592782
254 * - either of the transcript ids can be used to retrieve gene loci e.g.
255 * http://rest.ensembl.org/map/cds/ENST00000592782/1..100000
256 * Example (invertebrate):
257 * - fetch EMBLCDS cross-references for Uniprot entry Q43517 (FER1_SOLLC)
258 * - the Uniprot entry has an xref to ENSEMBLPLANTS Solyc10g044520.1.1
259 * - can retrieve gene loci with
260 * http://rest.ensemblgenomes.org/map/cds/Solyc10g044520.1.1/1..100000
265 public static void findGeneLoci(List<SequenceI> sequences)
267 Map<DBRefEntry, GeneLociI> retrievedLoci = new HashMap<>();
268 for (SequenceI seq : sequences)
270 findGeneLoci(seq, retrievedLoci);
275 * Tres to find chromosomal coordinates for the sequence, by searching its
276 * direct and indirect cross-references for Ensembl. If the loci have already
277 * been retrieved, just reads them out of the map of retrievedLoci; this is
278 * the case of an alternative transcript for the same protein. Otherwise calls
279 * a REST service to retrieve the loci, and if successful, adds them to the
280 * sequence and to the retrievedLoci.
283 * @param retrievedLoci
285 static void findGeneLoci(SequenceI seq,
286 Map<DBRefEntry, GeneLociI> retrievedLoci)
289 * don't replace any existing chromosomal coordinates
291 if (seq == null || seq.isProtein() || seq.getGeneLoci() != null
292 || seq.getDBRefs() == null)
297 Set<String> ensemblDivisions = new EnsemblInfo().getDivisions();
300 * first look for direct dbrefs from sequence to Ensembl
302 String[] divisionsArray = ensemblDivisions
303 .toArray(new String[ensemblDivisions.size()]);
304 List<DBRefEntry> seqRefs = seq.getDBRefs();
305 List<DBRefEntry> directEnsemblRefs = DBRefUtils.selectRefs(seqRefs,
307 if (directEnsemblRefs != null)
309 for (DBRefEntry ensemblRef : directEnsemblRefs)
311 if (fetchGeneLoci(seq, ensemblRef, retrievedLoci))
319 * else look for indirect dbrefs from sequence to Ensembl
321 for (DBRefEntry dbref : seq.getDBRefs())
323 if (dbref.getMap() != null && dbref.getMap().getTo() != null)
325 List<DBRefEntry> dbrefs = dbref.getMap().getTo().getDBRefs();
326 List<DBRefEntry> indirectEnsemblRefs = DBRefUtils.selectRefs(dbrefs,
328 if (indirectEnsemblRefs != null)
330 for (DBRefEntry ensemblRef : indirectEnsemblRefs)
332 if (fetchGeneLoci(seq, ensemblRef, retrievedLoci))
343 * Retrieves chromosomal coordinates for the Ensembl (or EnsemblGenomes)
344 * identifier in dbref. If successful, and the sequence length matches gene
345 * loci length, then add it to the sequence, and to the retrievedLoci map.
346 * Answers true if successful, else false.
350 * @param retrievedLoci
353 static boolean fetchGeneLoci(SequenceI seq, DBRefEntry dbref,
354 Map<DBRefEntry, GeneLociI> retrievedLoci)
356 String accession = dbref.getAccessionId();
357 String division = dbref.getSource();
360 * hack: ignore cross-references to Ensembl protein ids
361 * (or use map/translation perhaps?)
362 * todo: is there an equivalent in EnsemblGenomes?
364 if (accession.startsWith("ENSP"))
368 EnsemblMap mapper = new EnsemblMap();
371 * try CDS mapping first
373 GeneLociI geneLoci = mapper.getCdsMapping(division, accession, 1,
375 if (geneLoci != null)
377 MapList map = geneLoci.getMapping();
378 int mappedFromLength = MappingUtils.getLength(map.getFromRanges());
379 if (mappedFromLength == seq.getLength())
381 seq.setGeneLoci(geneLoci.getSpeciesId(), geneLoci.getAssemblyId(),
382 geneLoci.getChromosomeId(), map);
383 retrievedLoci.put(dbref, geneLoci);
389 * else try CDNA mapping
391 geneLoci = mapper.getCdnaMapping(division, accession, 1,
393 if (geneLoci != null)
395 MapList map = geneLoci.getMapping();
396 int mappedFromLength = MappingUtils.getLength(map.getFromRanges());
397 if (mappedFromLength == seq.getLength())
399 seq.setGeneLoci(geneLoci.getSpeciesId(), geneLoci.getAssemblyId(),
400 geneLoci.getChromosomeId(), map);
401 retrievedLoci.put(dbref, geneLoci);
414 * @param xrefsAlignment
417 protected AlignmentI copyAlignmentForSplitFrame(AlignmentI alignment,
418 AlignmentI dataset, boolean dna, AlignmentI xrefs,
419 AlignmentI xrefsAlignment)
421 AlignmentI copyAlignment;
422 boolean copyAlignmentIsAligned = false;
425 copyAlignment = AlignmentUtils.makeCdsAlignment(sel, dataset,
426 xrefsAlignment.getSequencesArray());
427 if (copyAlignment.getHeight() == 0)
429 JvOptionPane.showMessageDialog(alignFrame,
430 MessageManager.getString("label.cant_map_cds"),
431 MessageManager.getString("label.operation_failed"),
432 JvOptionPane.OK_OPTION);
433 System.err.println("Failed to make CDS alignment");
438 * pending getting Embl transcripts to 'align',
439 * we are only doing this for Ensembl
441 // TODO proper criteria for 'can align as cdna'
442 if (DBRefSource.ENSEMBL.equalsIgnoreCase(source)
443 || AlignmentUtils.looksLikeEnsembl(alignment))
445 copyAlignment.alignAs(alignment);
446 copyAlignmentIsAligned = true;
451 copyAlignment = AlignmentUtils.makeCopyAlignment(sel,
452 xrefs.getSequencesArray(), dataset);
455 .setGapCharacter(alignFrame.viewport.getGapCharacter());
457 StructureSelectionManager ssm = StructureSelectionManager
458 .getStructureSelectionManager(Desktop.instance);
461 * register any new mappings for sequence mouseover etc
462 * (will not duplicate any previously registered mappings)
464 ssm.registerMappings(dataset.getCodonFrames());
466 if (copyAlignment.getHeight() <= 0)
469 "No Sequences generated for xRef type " + source);
474 * align protein to dna
476 if (dna && copyAlignmentIsAligned)
478 xrefsAlignment.alignAs(copyAlignment);
483 * align cdna to protein - currently only if
484 * fetching and aligning Ensembl transcripts!
486 // TODO: generalise for other sources of locus/transcript/cds data
487 if (dna && DBRefSource.ENSEMBL.equalsIgnoreCase(source))
489 copyAlignment.alignAs(xrefsAlignment);
493 return copyAlignment;
497 * Makes an alignment containing the given sequences, and adds them to the
498 * given dataset, which is also set as the dataset for the new alignment
500 * TODO: refactor to DatasetI method
506 protected AlignmentI makeCrossReferencesAlignment(AlignmentI dataset,
509 SequenceI[] sprods = new SequenceI[seqs.getHeight()];
510 for (int s = 0; s < sprods.length; s++)
512 sprods[s] = (seqs.getSequenceAt(s)).deriveSequence();
513 if (dataset.getSequences() == null || !dataset.getSequences()
514 .contains(sprods[s].getDatasetSequence()))
516 dataset.addSequence(sprods[s].getDatasetSequence());
518 sprods[s].updatePDBIds();
520 Alignment al = new Alignment(sprods);
521 al.setDataset(dataset);
533 CrossRefAction(AlignFrame af, SequenceI[] seqs, boolean fromDna,
536 this.alignFrame = af;
538 this._odna = fromDna;
539 this.source = dbSource;
542 public static CrossRefAction getHandlerFor(final SequenceI[] sel,
543 final boolean fromDna, final String source,
544 final AlignFrame alignFrame)
546 return new CrossRefAction(alignFrame, sel, fromDna, source);