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.ext.ensembl;
23 import jalview.api.FeatureColourI;
24 import jalview.api.FeatureSettingsModelI;
25 import jalview.datamodel.AlignmentI;
26 import jalview.datamodel.Sequence;
27 import jalview.datamodel.SequenceFeature;
28 import jalview.datamodel.SequenceI;
29 import jalview.datamodel.features.SequenceFeatures;
30 import jalview.io.gff.SequenceOntologyFactory;
31 import jalview.io.gff.SequenceOntologyI;
32 import jalview.schemes.FeatureColour;
33 import jalview.schemes.FeatureSettingsAdapter;
34 import jalview.util.MapList;
36 import java.awt.Color;
37 import java.io.UnsupportedEncodingException;
38 import java.net.URLDecoder;
39 import java.util.ArrayList;
40 import java.util.Arrays;
41 import java.util.List;
43 import com.stevesoft.pat.Regex;
46 * A class that fetches genomic sequence and all transcripts for an Ensembl gene
50 public class EnsemblGene extends EnsemblSeqProxy
52 private static final String GENE_PREFIX = "gene:";
55 * accepts anything as we will attempt lookup of gene or
56 * transcript id or gene name
58 private static final Regex ACCESSION_REGEX = new Regex(".*");
60 private static final EnsemblFeatureType[] FEATURES_TO_FETCH = {
61 EnsemblFeatureType.gene, EnsemblFeatureType.transcript,
62 EnsemblFeatureType.exon, EnsemblFeatureType.cds,
63 EnsemblFeatureType.variation };
66 * Default constructor (to use rest.ensembl.org)
74 * Constructor given the target domain to fetch data from
78 public EnsemblGene(String d)
84 public String getDbName()
90 protected EnsemblFeatureType[] getFeaturesToFetch()
92 return FEATURES_TO_FETCH;
96 protected EnsemblSeqType getSourceEnsemblType()
98 return EnsemblSeqType.GENOMIC;
102 * Returns an alignment containing the gene(s) for the given gene or
103 * transcript identifier, or external identifier (e.g. Uniprot id). If given a
104 * gene name or external identifier, returns any related gene sequences found
105 * for model organisms. If only a single gene is queried for, then its
106 * transcripts are also retrieved and added to the alignment. <br>
109 * <li>resolves a transcript identifier by looking up its parent gene id</li>
110 * <li>resolves an external identifier by looking up xref-ed gene ids</li>
111 * <li>fetches the gene sequence</li>
112 * <li>fetches features on the sequence</li>
113 * <li>identifies "transcript" features whose Parent is the requested gene</li>
114 * <li>fetches the transcript sequence for each transcript</li>
115 * <li>makes a mapping from the gene to each transcript</li>
116 * <li>copies features from gene to transcript sequences</li>
117 * <li>fetches the protein sequence for each transcript, maps and saves it as
118 * a cross-reference</li>
119 * <li>aligns each transcript against the gene sequence based on the position
124 * a single gene or transcript identifier or gene name
125 * @return an alignment containing a gene, and possibly transcripts, or null
128 public AlignmentI getSequenceRecords(String query) throws Exception
131 * convert to a non-duplicated list of gene identifiers
133 List<String> geneIds = getGeneIds(query);
135 AlignmentI al = null;
136 for (String geneId : geneIds)
139 * fetch the gene sequence(s) with features and xrefs
141 AlignmentI geneAlignment = super.getSequenceRecords(geneId);
142 if (geneAlignment == null)
146 if (geneAlignment.getHeight() == 1)
148 getTranscripts(geneAlignment, geneId);
156 al.append(geneAlignment);
163 * Converts a query, which may contain one or more gene or transcript
164 * identifiers, into a non-redundant list of gene identifiers.
169 List<String> getGeneIds(String accessions)
171 List<String> geneIds = new ArrayList<String>();
173 for (String acc : accessions.split(getAccessionSeparator()))
175 if (isGeneIdentifier(acc))
177 if (!geneIds.contains(acc))
184 * if given a transcript id, look up its gene parent
186 else if (isTranscriptIdentifier(acc))
188 String geneId = new EnsemblLookup(getDomain()).getParent(acc);
189 if (geneId != null && !geneIds.contains(geneId))
194 else if (isProteinIdentifier(acc))
196 String tscriptId = new EnsemblLookup(getDomain()).getParent(acc);
197 if (tscriptId != null)
199 String geneId = new EnsemblLookup(getDomain())
200 .getParent(tscriptId);
202 if (geneId != null && !geneIds.contains(geneId))
207 // NOTE - acc is lost if it resembles an ENS.+ ID but isn't actually
208 // resolving to one... e.g. ENSMICP00000009241
211 * if given a gene or other external name, lookup and fetch
212 * the corresponding gene for all model organisms
216 List<String> ids = new EnsemblSymbol(getDomain(), getDbSource(),
217 getDbVersion()).getIds(acc);
218 for (String geneId : ids)
220 if (!geneIds.contains(geneId))
231 * Attempts to get Ensembl stable identifiers for model organisms for a gene
232 * name by calling the xrefs symbol REST service to resolve the gene name.
237 protected String getGeneIdentifiersForName(String query)
239 List<String> ids = new EnsemblSymbol(getDomain(), getDbSource(),
240 getDbVersion()).getIds(query);
243 for (String id : ids)
245 if (isGeneIdentifier(id))
255 * Constructs all transcripts for the gene, as identified by "transcript"
256 * features whose Parent is the requested gene. The coding transcript
257 * sequences (i.e. with introns omitted) are added to the alignment.
263 protected void getTranscripts(AlignmentI al, String accId)
266 SequenceI gene = al.getSequenceAt(0);
267 List<SequenceFeature> transcriptFeatures = getTranscriptFeatures(accId,
270 for (SequenceFeature transcriptFeature : transcriptFeatures)
272 makeTranscript(transcriptFeature, al, gene);
275 clearGeneFeatures(gene);
279 * Remove unwanted features (transcript, exon, CDS) from the gene sequence
280 * after we have used them to derive transcripts and transfer features
284 protected void clearGeneFeatures(SequenceI gene)
287 * Note we include NMD_transcript_variant here because it behaves like
288 * 'transcript' in Ensembl, although strictly speaking it is not
289 * (it is a sub-type of sequence_variant)
291 String[] soTerms = new String[] {
292 SequenceOntologyI.NMD_TRANSCRIPT_VARIANT,
293 SequenceOntologyI.TRANSCRIPT, SequenceOntologyI.EXON,
294 SequenceOntologyI.CDS };
295 List<SequenceFeature> sfs = gene.getFeatures().getFeaturesByOntology(
297 for (SequenceFeature sf : sfs)
299 gene.deleteFeature(sf);
304 * Constructs a spliced transcript sequence by finding 'exon' features for the
305 * given id (or failing that 'CDS'). Copies features on to the new sequence.
306 * 'Aligns' the new sequence against the gene sequence by padding with gaps,
307 * and adds it to the alignment.
309 * @param transcriptFeature
311 * the alignment to which to add the new sequence
313 * the parent gene sequence, with features
316 SequenceI makeTranscript(SequenceFeature transcriptFeature,
317 AlignmentI al, SequenceI gene)
319 String accId = getTranscriptId(transcriptFeature);
326 * NB we are mapping from gene sequence (not genome), so do not
327 * need to check for reverse strand (gene and transcript sequences
328 * are in forward sense)
332 * make a gene-length sequence filled with gaps
333 * we will fill in the bases for transcript regions
335 char[] seqChars = new char[gene.getLength()];
336 Arrays.fill(seqChars, al.getGapCharacter());
339 * look for exon features of the transcript, failing that for CDS
340 * (for example ENSG00000124610 has 1 CDS but no exon features)
342 String parentId = "transcript:" + accId;
343 List<SequenceFeature> splices = findFeatures(gene,
344 SequenceOntologyI.EXON, parentId);
345 if (splices.isEmpty())
347 splices = findFeatures(gene, SequenceOntologyI.CDS, parentId);
349 SequenceFeatures.sortFeatures(splices, true);
351 int transcriptLength = 0;
352 final char[] geneChars = gene.getSequence();
353 int offset = gene.getStart(); // to convert to 0-based positions
354 List<int[]> mappedFrom = new ArrayList<int[]>();
356 for (SequenceFeature sf : splices)
358 int start = sf.getBegin() - offset;
359 int end = sf.getEnd() - offset;
360 int spliceLength = end - start + 1;
361 System.arraycopy(geneChars, start, seqChars, start, spliceLength);
362 transcriptLength += spliceLength;
363 mappedFrom.add(new int[] { sf.getBegin(), sf.getEnd() });
366 Sequence transcript = new Sequence(accId, seqChars, 1, transcriptLength);
369 * Ensembl has gene name as transcript Name
370 * EnsemblGenomes doesn't, but has a url-encoded description field
372 String description = (String) transcriptFeature.getValue(NAME);
373 if (description == null)
375 description = (String) transcriptFeature.getValue(DESCRIPTION);
377 if (description != null)
381 transcript.setDescription(URLDecoder.decode(description, "UTF-8"));
382 } catch (UnsupportedEncodingException e)
384 e.printStackTrace(); // as if
387 transcript.createDatasetSequence();
389 al.addSequence(transcript);
392 * transfer features to the new sequence; we use EnsemblCdna to do this,
393 * to filter out unwanted features types (see method retainFeature)
395 List<int[]> mapTo = new ArrayList<int[]>();
396 mapTo.add(new int[] { 1, transcriptLength });
397 MapList mapping = new MapList(mappedFrom, mapTo, 1, 1);
398 EnsemblCdna cdna = new EnsemblCdna(getDomain());
399 cdna.transferFeatures(gene.getFeatures().getPositionalFeatures(),
400 transcript.getDatasetSequence(), mapping, parentId);
403 * fetch and save cross-references
405 cdna.getCrossReferences(transcript);
408 * and finally fetch the protein product and save as a cross-reference
410 cdna.addProteinProduct(transcript);
416 * Returns the 'transcript_id' property of the sequence feature (or null)
421 protected String getTranscriptId(SequenceFeature feature)
423 return (String) feature.getValue("transcript_id");
427 * Returns a list of the transcript features on the sequence whose Parent is
428 * the gene for the accession id.
431 * @param geneSequence
434 protected List<SequenceFeature> getTranscriptFeatures(String accId,
435 SequenceI geneSequence)
437 List<SequenceFeature> transcriptFeatures = new ArrayList<SequenceFeature>();
439 String parentIdentifier = GENE_PREFIX + accId;
440 // todo optimise here by transcript type!
441 List<SequenceFeature> sfs = geneSequence.getFeatures()
442 .getPositionalFeatures();
444 for (SequenceFeature sf : sfs)
446 if (isTranscript(sf.getType()))
448 String parent = (String) sf.getValue(PARENT);
449 if (parentIdentifier.equals(parent))
451 transcriptFeatures.add(sf);
456 return transcriptFeatures;
460 public String getDescription()
462 return "Fetches all transcripts and variant features for a gene or transcript";
466 * Default test query is a gene id (can also enter a transcript id)
469 public String getTestQuery()
471 return "ENSG00000157764"; // BRAF, 5 transcripts, reverse strand
472 // ENSG00000090266 // NDUFB2, 15 transcripts, forward strand
473 // ENSG00000101812 // H2BFM histone, 3 transcripts, forward strand
474 // ENSG00000123569 // H2BFWT histone, 2 transcripts, reverse strand
478 * Answers true for a feature of type 'gene' (or a sub-type of gene in the
479 * Sequence Ontology), whose ID is the accession we are retrieving
482 protected boolean identifiesSequence(SequenceFeature sf, String accId)
484 if (SequenceOntologyFactory.getInstance().isA(sf.getType(),
485 SequenceOntologyI.GENE))
487 String id = (String) sf.getValue(ID);
488 if ((GENE_PREFIX + accId).equals(id))
497 * Answers true unless feature type is 'gene', or 'transcript' with a parent
498 * which is a different gene. We need the gene features to identify the range,
499 * but it is redundant information on the gene sequence. Checking the parent
500 * allows us to drop transcript features which belong to different
501 * (overlapping) genes.
504 protected boolean retainFeature(SequenceFeature sf, String accessionId)
506 SequenceOntologyI so = SequenceOntologyFactory.getInstance();
507 String type = sf.getType();
508 if (so.isA(type, SequenceOntologyI.GENE))
512 if (isTranscript(type))
514 String parent = (String) sf.getValue(PARENT);
515 if (!(GENE_PREFIX + accessionId).equals(parent))
524 * Answers false. This allows an optimisation - a single 'gene' feature is all
525 * that is needed to identify the positions of the gene on the genomic
529 protected boolean isSpliceable()
535 * Override to do nothing as Ensembl doesn't return a protein sequence for a
539 protected void addProteinProduct(SequenceI querySeq)
544 public Regex getAccessionValidator()
546 return ACCESSION_REGEX;
550 * Returns a descriptor for suitable feature display settings with
552 * <li>only exon or sequence_variant features (or their subtypes in the
553 * Sequence Ontology) visible</li>
554 * <li>variant features coloured red</li>
555 * <li>exon features coloured by label (exon name)</li>
556 * <li>variants displayed above (on top of) exons</li>
560 public FeatureSettingsModelI getFeatureColourScheme()
562 return new FeatureSettingsAdapter()
564 SequenceOntologyI so = SequenceOntologyFactory.getInstance();
567 public boolean isFeatureDisplayed(String type)
569 return (so.isA(type, SequenceOntologyI.EXON) || so.isA(type,
570 SequenceOntologyI.SEQUENCE_VARIANT));
574 public FeatureColourI getFeatureColour(String type)
576 if (so.isA(type, SequenceOntologyI.EXON))
578 return new FeatureColour()
581 public boolean isColourByLabel()
587 if (so.isA(type, SequenceOntologyI.SEQUENCE_VARIANT))
589 return new FeatureColour()
593 public Color getColour()
603 * order to render sequence_variant after exon after the rest
606 public int compare(String feature1, String feature2)
608 if (so.isA(feature1, SequenceOntologyI.SEQUENCE_VARIANT))
612 if (so.isA(feature2, SequenceOntologyI.SEQUENCE_VARIANT))
616 if (so.isA(feature1, SequenceOntologyI.EXON))
620 if (so.isA(feature2, SequenceOntologyI.EXON))