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.datamodel.Alignment;
29 import jalview.datamodel.AlignmentI;
30 import jalview.datamodel.DBRefEntry;
31 import jalview.datamodel.DBRefSource;
32 import jalview.datamodel.GeneLociI;
33 import jalview.datamodel.SequenceI;
34 import jalview.ext.ensembl.EnsemblInfo;
35 import jalview.ext.ensembl.EnsemblMap;
36 import jalview.io.gff.SequenceOntologyI;
37 import jalview.structure.StructureSelectionManager;
38 import jalview.util.DBRefUtils;
39 import jalview.util.MapList;
40 import jalview.util.MappingUtils;
41 import jalview.util.MessageManager;
42 import jalview.viewmodel.seqfeatures.FeatureRendererModel;
43 import jalview.ws.SequenceFetcher;
45 import java.util.ArrayList;
46 import java.util.HashMap;
47 import java.util.List;
52 * Factory constructor and runnable for discovering and displaying
53 * cross-references for a set of aligned sequences
58 public class CrossRefAction implements Runnable
60 private AlignFrame alignFrame;
62 private SequenceI[] sel;
64 private final boolean _odna;
66 private String source;
68 List<AlignmentViewPanel> xrefViews = new ArrayList<>();
70 List<AlignmentViewPanel> getXrefViews()
78 final long sttime = System.currentTimeMillis();
79 alignFrame.setProgressBar(MessageManager.formatMessage(
80 "status.searching_for_sequences_from", new Object[]
84 AlignmentI alignment = alignFrame.getViewport().getAlignment();
85 AlignmentI dataset = alignment.getDataset() == null ? alignment
86 : alignment.getDataset();
87 boolean dna = alignment.isNucleotide();
91 .println("Conflict: showProducts for alignment originally "
92 + "thought to be " + (_odna ? "DNA" : "Protein")
93 + " now searching for " + (dna ? "DNA" : "Protein")
96 AlignmentI xrefs = new CrossRef(sel, dataset)
97 .findXrefSequences(source, dna);
104 * try to look up chromosomal coordinates for nucleotide
105 * sequences (if not already retrieved)
107 findGeneLoci(xrefs.getSequences());
110 * get display scheme (if any) to apply to features
112 FeatureSettingsModelI featureColourScheme = new SequenceFetcher()
113 .getFeatureColourScheme(source);
115 AlignmentI xrefsAlignment = makeCrossReferencesAlignment(dataset,
119 xrefsAlignment = AlignmentUtils.makeCdsAlignment(
120 xrefsAlignment.getSequencesArray(), dataset, sel);
121 xrefsAlignment.alignAs(alignment);
125 * If we are opening a splitframe, make a copy of this alignment (sharing the same dataset
126 * sequences). If we are DNA, drop introns and update mappings
128 AlignmentI copyAlignment = null;
130 if (Cache.getDefault(Preferences.ENABLE_SPLIT_FRAME, true))
132 copyAlignment = copyAlignmentForSplitFrame(alignment, dataset, dna,
133 xrefs, xrefsAlignment);
134 if (copyAlignment == null)
141 * build AlignFrame(s) according to available alignment data
143 AlignFrame newFrame = new AlignFrame(xrefsAlignment,
144 AlignFrame.DEFAULT_WIDTH, AlignFrame.DEFAULT_HEIGHT);
145 if (Cache.getDefault("HIDE_INTRONS", true))
147 newFrame.hideFeatureColumns(SequenceOntologyI.EXON, false);
149 String newtitle = String.format("%s %s %s",
150 dna ? MessageManager.getString("label.proteins")
151 : MessageManager.getString("label.nucleotides"),
152 MessageManager.getString("label.for"), alignFrame.getTitle());
153 newFrame.setTitle(newtitle);
155 if (copyAlignment == null)
158 * split frame display is turned off in preferences file
160 Desktop.addInternalFrame(newFrame, newtitle,
161 AlignFrame.DEFAULT_WIDTH, AlignFrame.DEFAULT_HEIGHT);
162 xrefViews.add(newFrame.alignPanel);
163 return; // via finally clause
166 AlignFrame copyThis = new AlignFrame(copyAlignment,
167 AlignFrame.DEFAULT_WIDTH, AlignFrame.DEFAULT_HEIGHT);
168 copyThis.setTitle(alignFrame.getTitle());
170 boolean showSequenceFeatures = alignFrame.getViewport()
171 .isShowSequenceFeatures();
172 newFrame.setShowSeqFeatures(showSequenceFeatures);
173 copyThis.setShowSeqFeatures(showSequenceFeatures);
174 FeatureRendererModel myFeatureStyling = alignFrame.alignPanel
175 .getSeqPanel().seqCanvas.getFeatureRenderer();
178 * copy feature rendering settings to split frame
180 FeatureRendererModel fr1 = newFrame.alignPanel.getSeqPanel().seqCanvas
181 .getFeatureRenderer();
182 fr1.transferSettings(myFeatureStyling);
183 fr1.findAllFeatures(true);
184 FeatureRendererModel fr2 = copyThis.alignPanel.getSeqPanel().seqCanvas
185 .getFeatureRenderer();
186 fr2.transferSettings(myFeatureStyling);
187 fr2.findAllFeatures(true);
190 * apply 'database source' feature configuration
193 // TODO is this the feature colouring for the original
194 // alignment or the fetched xrefs? either could be Ensembl
195 newFrame.getViewport().applyFeaturesStyle(featureColourScheme);
196 copyThis.getViewport().applyFeaturesStyle(featureColourScheme);
198 SplitFrame sf = new SplitFrame(dna ? copyThis : newFrame,
199 dna ? newFrame : copyThis);
200 newFrame.setVisible(true);
201 copyThis.setVisible(true);
202 String linkedTitle = MessageManager
203 .getString("label.linked_view_title");
204 Desktop.addInternalFrame(sf, linkedTitle, -1, -1);
205 sf.adjustInitialLayout();
207 // finally add the top, then bottom frame to the view list
208 xrefViews.add(dna ? copyThis.alignPanel : newFrame.alignPanel);
209 xrefViews.add(!dna ? copyThis.alignPanel : newFrame.alignPanel);
211 } catch (OutOfMemoryError e)
213 new OOMWarning("whilst fetching crossreferences", e);
214 } catch (Throwable e)
216 Cache.log.error("Error when finding crossreferences", e);
219 alignFrame.setProgressBar(MessageManager.formatMessage(
220 "status.finished_searching_for_sequences_from", new Object[]
221 { source }), sttime);
226 * Tries to add chromosomal coordinates to any nucleotide sequence which does
227 * not already have them. Coordinates are retrieved from Ensembl given an
228 * Ensembl identifier, either on the sequence itself or on a peptide sequence
229 * it has a reference to.
233 * - fetch EMBLCDS cross-references for Uniprot entry P30419
234 * - the EMBL sequences do not have xrefs to Ensembl
235 * - the Uniprot entry has xrefs to
236 * ENSP00000258960, ENSP00000468424, ENST00000258960, ENST00000592782
237 * - either of the transcript ids can be used to retrieve gene loci e.g.
238 * http://rest.ensembl.org/map/cds/ENST00000592782/1..100000
239 * Example (invertebrate):
240 * - fetch EMBLCDS cross-references for Uniprot entry Q43517 (FER1_SOLLC)
241 * - the Uniprot entry has an xref to ENSEMBLPLANTS Solyc10g044520.1.1
242 * - can retrieve gene loci with
243 * http://rest.ensemblgenomes.org/map/cds/Solyc10g044520.1.1/1..100000
248 public static void findGeneLoci(List<SequenceI> sequences)
250 Map<DBRefEntry, GeneLociI> retrievedLoci = new HashMap<>();
251 for (SequenceI seq : sequences)
253 findGeneLoci(seq, retrievedLoci);
258 * Tres to find chromosomal coordinates for the sequence, by searching its
259 * direct and indirect cross-references for Ensembl. If the loci have already
260 * been retrieved, just reads them out of the map of retrievedLoci; this is
261 * the case of an alternative transcript for the same protein. Otherwise calls
262 * a REST service to retrieve the loci, and if successful, adds them to the
263 * sequence and to the retrievedLoci.
266 * @param retrievedLoci
268 static void findGeneLoci(SequenceI seq,
269 Map<DBRefEntry, GeneLociI> retrievedLoci)
272 * don't replace any existing chromosomal coordinates
274 if (seq == null || seq.isProtein() || seq.getGeneLoci() != null
275 || seq.getDBRefs() == null)
280 Set<String> ensemblDivisions = new EnsemblInfo().getDivisions();
283 * first look for direct dbrefs from sequence to Ensembl
285 String[] divisionsArray = ensemblDivisions
286 .toArray(new String[ensemblDivisions.size()]);
287 DBRefEntry[] seqRefs = seq.getDBRefs();
288 DBRefEntry[] directEnsemblRefs = DBRefUtils.selectRefs(seqRefs,
290 if (directEnsemblRefs != null)
292 for (DBRefEntry ensemblRef : directEnsemblRefs)
294 if (fetchGeneLoci(seq, ensemblRef, retrievedLoci))
302 * else look for indirect dbrefs from sequence to Ensembl
304 for (DBRefEntry dbref : seq.getDBRefs())
306 if (dbref.getMap() != null && dbref.getMap().getTo() != null)
308 DBRefEntry[] dbrefs = dbref.getMap().getTo().getDBRefs();
309 DBRefEntry[] indirectEnsemblRefs = DBRefUtils.selectRefs(dbrefs,
311 if (indirectEnsemblRefs != null)
313 for (DBRefEntry ensemblRef : indirectEnsemblRefs)
315 if (fetchGeneLoci(seq, ensemblRef, retrievedLoci))
326 * Retrieves chromosomal coordinates for the Ensembl (or EnsemblGenomes)
327 * identifier in dbref. If successful, and the sequence length matches gene
328 * loci length, then add it to the sequence, and to the retrievedLoci map.
329 * Answers true if successful, else false.
333 * @param retrievedLoci
336 static boolean fetchGeneLoci(SequenceI seq, DBRefEntry dbref,
337 Map<DBRefEntry, GeneLociI> retrievedLoci)
339 String accession = dbref.getAccessionId();
340 String division = dbref.getSource();
343 * hack: ignore cross-references to Ensembl protein ids
344 * (or use map/translation perhaps?)
345 * todo: is there an equivalent in EnsemblGenomes?
347 if (accession.startsWith("ENSP"))
351 EnsemblMap mapper = new EnsemblMap();
354 * try CDS mapping first
356 GeneLociI geneLoci = mapper.getCdsMapping(division, accession, 1,
358 if (geneLoci != null)
360 MapList map = geneLoci.getMap();
361 int mappedFromLength = MappingUtils.getLength(map.getFromRanges());
362 if (mappedFromLength == seq.getLength())
364 seq.setGeneLoci(geneLoci.getSpeciesId(), geneLoci.getAssemblyId(),
365 geneLoci.getChromosomeId(), geneLoci.getMap());
366 retrievedLoci.put(dbref, geneLoci);
372 * else try CDNA mapping
374 geneLoci = mapper.getCdnaMapping(division, accession, 1,
376 if (geneLoci != null)
378 MapList map = geneLoci.getMap();
379 int mappedFromLength = MappingUtils.getLength(map.getFromRanges());
380 if (mappedFromLength == seq.getLength())
382 seq.setGeneLoci(geneLoci.getSpeciesId(), geneLoci.getAssemblyId(),
383 geneLoci.getChromosomeId(), geneLoci.getMap());
384 retrievedLoci.put(dbref, geneLoci);
397 * @param xrefsAlignment
400 protected AlignmentI copyAlignmentForSplitFrame(AlignmentI alignment,
401 AlignmentI dataset, boolean dna, AlignmentI xrefs,
402 AlignmentI xrefsAlignment)
404 AlignmentI copyAlignment;
405 boolean copyAlignmentIsAligned = false;
408 copyAlignment = AlignmentUtils.makeCdsAlignment(sel, dataset,
409 xrefsAlignment.getSequencesArray());
410 if (copyAlignment.getHeight() == 0)
412 JvOptionPane.showMessageDialog(alignFrame,
413 MessageManager.getString("label.cant_map_cds"),
414 MessageManager.getString("label.operation_failed"),
415 JvOptionPane.OK_OPTION);
416 System.err.println("Failed to make CDS alignment");
421 * pending getting Embl transcripts to 'align',
422 * we are only doing this for Ensembl
424 // TODO proper criteria for 'can align as cdna'
425 if (DBRefSource.ENSEMBL.equalsIgnoreCase(source)
426 || AlignmentUtils.looksLikeEnsembl(alignment))
428 copyAlignment.alignAs(alignment);
429 copyAlignmentIsAligned = true;
434 copyAlignment = AlignmentUtils.makeCopyAlignment(sel,
435 xrefs.getSequencesArray(), dataset);
438 .setGapCharacter(alignFrame.viewport.getGapCharacter());
440 StructureSelectionManager ssm = StructureSelectionManager
441 .getStructureSelectionManager(Desktop.instance);
444 * register any new mappings for sequence mouseover etc
445 * (will not duplicate any previously registered mappings)
447 ssm.registerMappings(dataset.getCodonFrames());
449 if (copyAlignment.getHeight() <= 0)
452 "No Sequences generated for xRef type " + source);
457 * align protein to dna
459 if (dna && copyAlignmentIsAligned)
461 xrefsAlignment.alignAs(copyAlignment);
466 * align cdna to protein - currently only if
467 * fetching and aligning Ensembl transcripts!
469 // TODO: generalise for other sources of locus/transcript/cds data
470 if (dna && DBRefSource.ENSEMBL.equalsIgnoreCase(source))
472 copyAlignment.alignAs(xrefsAlignment);
476 return copyAlignment;
480 * Makes an alignment containing the given sequences, and adds them to the
481 * given dataset, which is also set as the dataset for the new alignment
483 * TODO: refactor to DatasetI method
489 protected AlignmentI makeCrossReferencesAlignment(AlignmentI dataset,
492 SequenceI[] sprods = new SequenceI[seqs.getHeight()];
493 for (int s = 0; s < sprods.length; s++)
495 sprods[s] = (seqs.getSequenceAt(s)).deriveSequence();
496 if (dataset.getSequences() == null || !dataset.getSequences()
497 .contains(sprods[s].getDatasetSequence()))
499 dataset.addSequence(sprods[s].getDatasetSequence());
501 sprods[s].updatePDBIds();
503 Alignment al = new Alignment(sprods);
504 al.setDataset(dataset);
516 CrossRefAction(AlignFrame af, SequenceI[] seqs, boolean fromDna,
519 this.alignFrame = af;
521 this._odna = fromDna;
522 this.source = dbSource;
525 public static CrossRefAction getHandlerFor(final SequenceI[] sel,
526 final boolean fromDna, final String source,
527 final AlignFrame alignFrame)
529 return new CrossRefAction(alignFrame, sel, fromDna, source);