Clover icon

jalviewX

  1. Project Clover database Wed Oct 31 2018 15:13:58 GMT
  2. Package jalview.ext.ensembl

File EnsemblGene.java

 

Coverage histogram

../../../img/srcFileCovDistChart3.png
47% of files have more coverage

Code metrics

56
160
29
1
703
409
60
0.38
5.52
29
2.07

Classes

Class Line # Actions
EnsemblGene 52 160 60 177
0.2775510327.8%
 

Contributing tests

This file is covered by 85 tests. .

Source view

1    /*
2    * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
3    * Copyright (C) $$Year-Rel$$ The Jalview Authors
4    *
5    * This file is part of Jalview.
6    *
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.
11    *
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.
16    *
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.
20    */
21    package jalview.ext.ensembl;
22   
23    import jalview.api.FeatureColourI;
24    import jalview.api.FeatureSettingsModelI;
25    import jalview.datamodel.AlignmentI;
26    import jalview.datamodel.DBRefEntry;
27    import jalview.datamodel.GeneLociI;
28    import jalview.datamodel.Sequence;
29    import jalview.datamodel.SequenceFeature;
30    import jalview.datamodel.SequenceI;
31    import jalview.datamodel.features.SequenceFeatures;
32    import jalview.io.gff.SequenceOntologyFactory;
33    import jalview.io.gff.SequenceOntologyI;
34    import jalview.schemes.FeatureColour;
35    import jalview.schemes.FeatureSettingsAdapter;
36    import jalview.util.MapList;
37   
38    import java.awt.Color;
39    import java.io.UnsupportedEncodingException;
40    import java.net.URLDecoder;
41    import java.util.ArrayList;
42    import java.util.Arrays;
43    import java.util.List;
44   
45    import com.stevesoft.pat.Regex;
46   
47    /**
48    * A class that fetches genomic sequence and all transcripts for an Ensembl gene
49    *
50    * @author gmcarstairs
51    */
 
52    public class EnsemblGene extends EnsemblSeqProxy
53    {
54    private static final String GENE_PREFIX = "gene:";
55   
56    /*
57    * accepts anything as we will attempt lookup of gene or
58    * transcript id or gene name
59    */
60    private static final Regex ACCESSION_REGEX = new Regex(".*");
61   
62    private static final EnsemblFeatureType[] FEATURES_TO_FETCH = {
63    EnsemblFeatureType.gene, EnsemblFeatureType.transcript,
64    EnsemblFeatureType.exon, EnsemblFeatureType.cds,
65    EnsemblFeatureType.variation };
66   
67    /**
68    * Default constructor (to use rest.ensembl.org)
69    */
 
70  18 toggle public EnsemblGene()
71    {
72  18 super();
73    }
74   
75    /**
76    * Constructor given the target domain to fetch data from
77    *
78    * @param d
79    */
 
80  0 toggle public EnsemblGene(String d)
81    {
82  0 super(d);
83    }
84   
 
85  1108 toggle @Override
86    public String getDbName()
87    {
88  1108 return "ENSEMBL";
89    }
90   
 
91  0 toggle @Override
92    protected EnsemblFeatureType[] getFeaturesToFetch()
93    {
94  0 return FEATURES_TO_FETCH;
95    }
96   
 
97  0 toggle @Override
98    protected EnsemblSeqType getSourceEnsemblType()
99    {
100  0 return EnsemblSeqType.GENOMIC;
101    }
102   
 
103  0 toggle @Override
104    protected String getObjectType()
105    {
106  0 return OBJECT_TYPE_GENE;
107    }
108   
109    /**
110    * Returns an alignment containing the gene(s) for the given gene or
111    * transcript identifier, or external identifier (e.g. Uniprot id). If given a
112    * gene name or external identifier, returns any related gene sequences found
113    * for model organisms. If only a single gene is queried for, then its
114    * transcripts are also retrieved and added to the alignment. <br>
115    * Method:
116    * <ul>
117    * <li>resolves a transcript identifier by looking up its parent gene id</li>
118    * <li>resolves an external identifier by looking up xref-ed gene ids</li>
119    * <li>fetches the gene sequence</li>
120    * <li>fetches features on the sequence</li>
121    * <li>identifies "transcript" features whose Parent is the requested
122    * gene</li>
123    * <li>fetches the transcript sequence for each transcript</li>
124    * <li>makes a mapping from the gene to each transcript</li>
125    * <li>copies features from gene to transcript sequences</li>
126    * <li>fetches the protein sequence for each transcript, maps and saves it as
127    * a cross-reference</li>
128    * <li>aligns each transcript against the gene sequence based on the position
129    * mappings</li>
130    * </ul>
131    *
132    * @param query
133    * a single gene or transcript identifier or gene name
134    * @return an alignment containing a gene, and possibly transcripts, or null
135    */
 
136  0 toggle @Override
137    public AlignmentI getSequenceRecords(String query) throws Exception
138    {
139    /*
140    * convert to a non-duplicated list of gene identifiers
141    */
142  0 List<String> geneIds = getGeneIds(query);
143   
144  0 AlignmentI al = null;
145  0 for (String geneId : geneIds)
146    {
147    /*
148    * fetch the gene sequence(s) with features and xrefs
149    */
150  0 AlignmentI geneAlignment = super.getSequenceRecords(geneId);
151  0 if (geneAlignment == null)
152    {
153  0 continue;
154    }
155   
156  0 if (geneAlignment.getHeight() == 1)
157    {
158    // ensure id has 'correct' case for the Ensembl identifier
159  0 geneId = geneAlignment.getSequenceAt(0).getName();
160   
161  0 findGeneLoci(geneAlignment.getSequenceAt(0), geneId);
162   
163  0 getTranscripts(geneAlignment, geneId);
164    }
165  0 if (al == null)
166    {
167  0 al = geneAlignment;
168    }
169    else
170    {
171  0 al.append(geneAlignment);
172    }
173    }
174  0 return al;
175    }
176   
177    /**
178    * Calls the /lookup/id REST service, parses the response for gene
179    * coordinates, and if successful, adds these to the sequence. If this fails,
180    * fall back on trying to parse the sequence description in case it is in
181    * Ensembl-gene format e.g. chromosome:GRCh38:17:45051610:45109016:1.
182    *
183    * @param seq
184    * @param geneId
185    */
 
186  0 toggle void findGeneLoci(SequenceI seq, String geneId)
187    {
188  0 GeneLociI geneLoci = new EnsemblLookup(getDomain()).getGeneLoci(geneId);
189  0 if (geneLoci != null)
190    {
191  0 seq.setGeneLoci(geneLoci.getSpeciesId(), geneLoci.getAssemblyId(),
192    geneLoci.getChromosomeId(), geneLoci.getMap());
193    }
194    else
195    {
196  0 parseChromosomeLocations(seq);
197    }
198    }
199   
200    /**
201    * Parses and saves fields of an Ensembl-style description e.g.
202    * chromosome:GRCh38:17:45051610:45109016:1
203    *
204    * @param seq
205    */
 
206  0 toggle boolean parseChromosomeLocations(SequenceI seq)
207    {
208  0 String description = seq.getDescription();
209  0 if (description == null)
210    {
211  0 return false;
212    }
213  0 String[] tokens = description.split(":");
214  0 if (tokens.length == 6 && tokens[0].startsWith(DBRefEntry.CHROMOSOME))
215    {
216  0 String ref = tokens[1];
217  0 String chrom = tokens[2];
218  0 try
219    {
220  0 int chStart = Integer.parseInt(tokens[3]);
221  0 int chEnd = Integer.parseInt(tokens[4]);
222  0 boolean forwardStrand = "1".equals(tokens[5]);
223  0 String species = ""; // not known here
224  0 int[] from = new int[] { seq.getStart(), seq.getEnd() };
225  0 int[] to = new int[] { forwardStrand ? chStart : chEnd,
226  0 forwardStrand ? chEnd : chStart };
227  0 MapList map = new MapList(from, to, 1, 1);
228  0 seq.setGeneLoci(species, ref, chrom, map);
229  0 return true;
230    } catch (NumberFormatException e)
231    {
232  0 System.err.println("Bad integers in description " + description);
233    }
234    }
235  0 return false;
236    }
237   
238    /**
239    * Converts a query, which may contain one or more gene, transcript, or
240    * external (to Ensembl) identifiers, into a non-redundant list of gene
241    * identifiers.
242    *
243    * @param accessions
244    * @return
245    */
 
246  0 toggle List<String> getGeneIds(String accessions)
247    {
248  0 List<String> geneIds = new ArrayList<>();
249   
250  0 for (String acc : accessions.split(getAccessionSeparator()))
251    {
252    /*
253    * First try lookup as an Ensembl (gene or transcript) identifier
254    */
255  0 String geneId = new EnsemblLookup(getDomain()).getGeneId(acc);
256  0 if (geneId != null)
257    {
258  0 if (!geneIds.contains(geneId))
259    {
260  0 geneIds.add(geneId);
261    }
262    }
263    else
264    {
265    /*
266    * if given a gene or other external name, lookup and fetch
267    * the corresponding gene for all model organisms
268    */
269  0 List<String> ids = new EnsemblSymbol(getDomain(), getDbSource(),
270    getDbVersion()).getGeneIds(acc);
271  0 for (String id : ids)
272    {
273  0 if (!geneIds.contains(id))
274    {
275  0 geneIds.add(id);
276    }
277    }
278    }
279    }
280  0 return geneIds;
281    }
282   
283    /**
284    * Constructs all transcripts for the gene, as identified by "transcript"
285    * features whose Parent is the requested gene. The coding transcript
286    * sequences (i.e. with introns omitted) are added to the alignment.
287    *
288    * @param al
289    * @param accId
290    * @throws Exception
291    */
 
292  0 toggle protected void getTranscripts(AlignmentI al, String accId)
293    throws Exception
294    {
295  0 SequenceI gene = al.getSequenceAt(0);
296  0 List<SequenceFeature> transcriptFeatures = getTranscriptFeatures(accId,
297    gene);
298   
299  0 for (SequenceFeature transcriptFeature : transcriptFeatures)
300    {
301  0 makeTranscript(transcriptFeature, al, gene);
302    }
303   
304  0 clearGeneFeatures(gene);
305    }
306   
307    /**
308    * Remove unwanted features (transcript, exon, CDS) from the gene sequence
309    * after we have used them to derive transcripts and transfer features
310    *
311    * @param gene
312    */
 
313  0 toggle protected void clearGeneFeatures(SequenceI gene)
314    {
315    /*
316    * Note we include NMD_transcript_variant here because it behaves like
317    * 'transcript' in Ensembl, although strictly speaking it is not
318    * (it is a sub-type of sequence_variant)
319    */
320  0 String[] soTerms = new String[] {
321    SequenceOntologyI.NMD_TRANSCRIPT_VARIANT,
322    SequenceOntologyI.TRANSCRIPT, SequenceOntologyI.EXON,
323    SequenceOntologyI.CDS };
324  0 List<SequenceFeature> sfs = gene.getFeatures().getFeaturesByOntology(
325    soTerms);
326  0 for (SequenceFeature sf : sfs)
327    {
328  0 gene.deleteFeature(sf);
329    }
330    }
331   
332    /**
333    * Constructs a spliced transcript sequence by finding 'exon' features for the
334    * given id (or failing that 'CDS'). Copies features on to the new sequence.
335    * 'Aligns' the new sequence against the gene sequence by padding with gaps,
336    * and adds it to the alignment.
337    *
338    * @param transcriptFeature
339    * @param al
340    * the alignment to which to add the new sequence
341    * @param gene
342    * the parent gene sequence, with features
343    * @return
344    */
 
345  0 toggle SequenceI makeTranscript(SequenceFeature transcriptFeature, AlignmentI al,
346    SequenceI gene)
347    {
348  0 String accId = getTranscriptId(transcriptFeature);
349  0 if (accId == null)
350    {
351  0 return null;
352    }
353   
354    /*
355    * NB we are mapping from gene sequence (not genome), so do not
356    * need to check for reverse strand (gene and transcript sequences
357    * are in forward sense)
358    */
359   
360    /*
361    * make a gene-length sequence filled with gaps
362    * we will fill in the bases for transcript regions
363    */
364  0 char[] seqChars = new char[gene.getLength()];
365  0 Arrays.fill(seqChars, al.getGapCharacter());
366   
367    /*
368    * look for exon features of the transcript, failing that for CDS
369    * (for example ENSG00000124610 has 1 CDS but no exon features)
370    */
371  0 String parentId = "transcript:" + accId;
372  0 List<SequenceFeature> splices = findFeatures(gene,
373    SequenceOntologyI.EXON, parentId);
374  0 if (splices.isEmpty())
375    {
376  0 splices = findFeatures(gene, SequenceOntologyI.CDS, parentId);
377    }
378  0 SequenceFeatures.sortFeatures(splices, true);
379   
380  0 int transcriptLength = 0;
381  0 final char[] geneChars = gene.getSequence();
382  0 int offset = gene.getStart(); // to convert to 0-based positions
383  0 List<int[]> mappedFrom = new ArrayList<>();
384   
385  0 for (SequenceFeature sf : splices)
386    {
387  0 int start = sf.getBegin() - offset;
388  0 int end = sf.getEnd() - offset;
389  0 int spliceLength = end - start + 1;
390  0 System.arraycopy(geneChars, start, seqChars, start, spliceLength);
391  0 transcriptLength += spliceLength;
392  0 mappedFrom.add(new int[] { sf.getBegin(), sf.getEnd() });
393    }
394   
395  0 Sequence transcript = new Sequence(accId, seqChars, 1,
396    transcriptLength);
397   
398    /*
399    * Ensembl has gene name as transcript Name
400    * EnsemblGenomes doesn't, but has a url-encoded description field
401    */
402  0 String description = (String) transcriptFeature.getValue(NAME);
403  0 if (description == null)
404    {
405  0 description = (String) transcriptFeature.getValue(DESCRIPTION);
406    }
407  0 if (description != null)
408    {
409  0 try
410    {
411  0 transcript.setDescription(URLDecoder.decode(description, "UTF-8"));
412    } catch (UnsupportedEncodingException e)
413    {
414  0 e.printStackTrace(); // as if
415    }
416    }
417  0 transcript.createDatasetSequence();
418   
419  0 al.addSequence(transcript);
420   
421    /*
422    * transfer features to the new sequence; we use EnsemblCdna to do this,
423    * to filter out unwanted features types (see method retainFeature)
424    */
425  0 List<int[]> mapTo = new ArrayList<>();
426  0 mapTo.add(new int[] { 1, transcriptLength });
427  0 MapList mapping = new MapList(mappedFrom, mapTo, 1, 1);
428  0 EnsemblCdna cdna = new EnsemblCdna(getDomain());
429  0 cdna.transferFeatures(gene.getFeatures().getPositionalFeatures(),
430    transcript.getDatasetSequence(), mapping, parentId);
431   
432  0 mapTranscriptToChromosome(transcript, gene, mapping);
433   
434    /*
435    * fetch and save cross-references
436    */
437  0 cdna.getCrossReferences(transcript);
438   
439    /*
440    * and finally fetch the protein product and save as a cross-reference
441    */
442  0 cdna.addProteinProduct(transcript);
443   
444  0 return transcript;
445    }
446   
447    /**
448    * If the gene has a mapping to chromosome coordinates, derive the transcript
449    * chromosome regions and save on the transcript sequence
450    *
451    * @param transcript
452    * @param gene
453    * @param mapping
454    * the mapping from gene to transcript positions
455    */
 
456  0 toggle protected void mapTranscriptToChromosome(SequenceI transcript,
457    SequenceI gene, MapList mapping)
458    {
459  0 GeneLociI loci = gene.getGeneLoci();
460  0 if (loci == null)
461    {
462  0 return;
463    }
464   
465  0 MapList geneMapping = loci.getMap();
466   
467  0 List<int[]> exons = mapping.getFromRanges();
468  0 List<int[]> transcriptLoci = new ArrayList<>();
469   
470  0 for (int[] exon : exons)
471    {
472  0 transcriptLoci.add(geneMapping.locateInTo(exon[0], exon[1]));
473    }
474   
475  0 List<int[]> transcriptRange = Arrays.asList(new int[] {
476    transcript.getStart(), transcript.getEnd() });
477  0 MapList mapList = new MapList(transcriptRange, transcriptLoci, 1, 1);
478   
479  0 transcript.setGeneLoci(loci.getSpeciesId(), loci.getAssemblyId(),
480    loci.getChromosomeId(), mapList);
481    }
482   
483    /**
484    * Returns the 'transcript_id' property of the sequence feature (or null)
485    *
486    * @param feature
487    * @return
488    */
 
489  0 toggle protected String getTranscriptId(SequenceFeature feature)
490    {
491  0 return (String) feature.getValue("transcript_id");
492    }
493   
494    /**
495    * Returns a list of the transcript features on the sequence whose Parent is
496    * the gene for the accession id.
497    * <p>
498    * Transcript features are those of type "transcript", or any of its sub-types
499    * in the Sequence Ontology e.g. "mRNA", "processed_transcript". We also
500    * include "NMD_transcript_variant", because this type behaves like a
501    * transcript identifier in Ensembl, although strictly speaking it is not in
502    * the SO.
503    *
504    * @param accId
505    * @param geneSequence
506    * @return
507    */
 
508  1 toggle protected List<SequenceFeature> getTranscriptFeatures(String accId,
509    SequenceI geneSequence)
510    {
511  1 List<SequenceFeature> transcriptFeatures = new ArrayList<>();
512   
513  1 String parentIdentifier = GENE_PREFIX + accId;
514   
515  1 List<SequenceFeature> sfs = geneSequence.getFeatures()
516    .getFeaturesByOntology(SequenceOntologyI.TRANSCRIPT);
517  1 sfs.addAll(geneSequence.getFeatures().getPositionalFeatures(
518    SequenceOntologyI.NMD_TRANSCRIPT_VARIANT));
519   
520  1 for (SequenceFeature sf : sfs)
521    {
522  4 String parent = (String) sf.getValue(PARENT);
523  4 if (parentIdentifier.equalsIgnoreCase(parent))
524    {
525  3 transcriptFeatures.add(sf);
526    }
527    }
528   
529  1 return transcriptFeatures;
530    }
531   
 
532  0 toggle @Override
533    public String getDescription()
534    {
535  0 return "Fetches all transcripts and variant features for a gene or transcript";
536    }
537   
538    /**
539    * Default test query is a gene id (can also enter a transcript id)
540    */
 
541  0 toggle @Override
542    public String getTestQuery()
543    {
544  0 return "ENSG00000157764"; // BRAF, 5 transcripts, reverse strand
545    // ENSG00000090266 // NDUFB2, 15 transcripts, forward strand
546    // ENSG00000101812 // H2BFM histone, 3 transcripts, forward strand
547    // ENSG00000123569 // H2BFWT histone, 2 transcripts, reverse strand
548    }
549   
550    /**
551    * Answers true for a feature of type 'gene' (or a sub-type of gene in the
552    * Sequence Ontology), whose ID is the accession we are retrieving
553    */
 
554  9 toggle @Override
555    protected boolean identifiesSequence(SequenceFeature sf, String accId)
556    {
557  9 if (SequenceOntologyFactory.getInstance().isA(sf.getType(),
558    SequenceOntologyI.GENE))
559    {
560    // NB features as gff use 'ID'; rest services return as 'id'
561  7 String id = (String) sf.getValue("ID");
562  7 if ((GENE_PREFIX + accId).equalsIgnoreCase(id))
563    {
564  5 return true;
565    }
566    }
567  4 return false;
568    }
569   
570    /**
571    * Answers true unless feature type is 'gene', or 'transcript' with a parent
572    * which is a different gene. We need the gene features to identify the range,
573    * but it is redundant information on the gene sequence. Checking the parent
574    * allows us to drop transcript features which belong to different
575    * (overlapping) genes.
576    */
 
577  6 toggle @Override
578    protected boolean retainFeature(SequenceFeature sf, String accessionId)
579    {
580  6 SequenceOntologyI so = SequenceOntologyFactory.getInstance();
581  6 String type = sf.getType();
582  6 if (so.isA(type, SequenceOntologyI.GENE))
583    {
584  1 return false;
585    }
586  5 if (isTranscript(type))
587    {
588  4 String parent = (String) sf.getValue(PARENT);
589  4 if (!(GENE_PREFIX + accessionId).equalsIgnoreCase(parent))
590    {
591  1 return false;
592    }
593    }
594  4 return true;
595    }
596   
597    /**
598    * Answers false. This allows an optimisation - a single 'gene' feature is all
599    * that is needed to identify the positions of the gene on the genomic
600    * sequence.
601    */
 
602  2 toggle @Override
603    protected boolean isSpliceable()
604    {
605  2 return false;
606    }
607   
608    /**
609    * Override to do nothing as Ensembl doesn't return a protein sequence for a
610    * gene identifier
611    */
 
612  0 toggle @Override
613    protected void addProteinProduct(SequenceI querySeq)
614    {
615    }
616   
 
617  0 toggle @Override
618    public Regex getAccessionValidator()
619    {
620  0 return ACCESSION_REGEX;
621    }
622   
623    /**
624    * Returns a descriptor for suitable feature display settings with
625    * <ul>
626    * <li>only exon or sequence_variant features (or their subtypes in the
627    * Sequence Ontology) visible</li>
628    * <li>variant features coloured red</li>
629    * <li>exon features coloured by label (exon name)</li>
630    * <li>variants displayed above (on top of) exons</li>
631    * </ul>
632    */
 
633  1 toggle @Override
634    public FeatureSettingsModelI getFeatureColourScheme()
635    {
636  1 return new FeatureSettingsAdapter()
637    {
638    SequenceOntologyI so = SequenceOntologyFactory.getInstance();
639   
 
640  5 toggle @Override
641    public boolean isFeatureDisplayed(String type)
642    {
643  5 return (so.isA(type, SequenceOntologyI.EXON)
644    || so.isA(type, SequenceOntologyI.SEQUENCE_VARIANT));
645    }
646   
 
647  4 toggle @Override
648    public FeatureColourI getFeatureColour(String type)
649    {
650  4 if (so.isA(type, SequenceOntologyI.EXON))
651    {
652  2 return new FeatureColour()
653    {
 
654  2 toggle @Override
655    public boolean isColourByLabel()
656    {
657  2 return true;
658    }
659    };
660    }
661  2 if (so.isA(type, SequenceOntologyI.SEQUENCE_VARIANT))
662    {
663  2 return new FeatureColour()
664    {
665   
 
666  2 toggle @Override
667    public Color getColour()
668    {
669  2 return Color.RED;
670    }
671    };
672    }
673  0 return null;
674    }
675   
676    /**
677    * order to render sequence_variant after exon after the rest
678    */
 
679  4 toggle @Override
680    public int compare(String feature1, String feature2)
681    {
682  4 if (so.isA(feature1, SequenceOntologyI.SEQUENCE_VARIANT))
683    {
684  2 return +1;
685    }
686  2 if (so.isA(feature2, SequenceOntologyI.SEQUENCE_VARIANT))
687    {
688  2 return -1;
689    }
690  0 if (so.isA(feature1, SequenceOntologyI.EXON))
691    {
692  0 return +1;
693    }
694  0 if (so.isA(feature2, SequenceOntologyI.EXON))
695    {
696  0 return -1;
697    }
698  0 return 0;
699    }
700    };
701    }
702   
703    }
  • Report generated by OpenClover v 4.2.1 on Wed Oct 31 2018 15:15:54 GMT using coverage data from Wed Oct 31 2018 15:15:47 GMT.
  • Clover free edition.