Clover icon

Coverage Report

  1. Project Clover database Mon Sep 2 2024 17:57:51 BST
  2. Package jalview.ext.ensembl

File EnsemblGene.java

 

Coverage histogram

../../../img/srcFileCovDistChart3.png
52% of files have more coverage

Code metrics

56
161
28
1
695
409
59
0.37
5.75
28
2.11

Classes

Class Line # Actions
EnsemblGene 52 161 59
0.2938775429.4%
 

Contributing tests

This file is covered by 206 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.GeneLociI;
27    import jalview.datamodel.Sequence;
28    import jalview.datamodel.SequenceFeature;
29    import jalview.datamodel.SequenceI;
30    import jalview.datamodel.features.SequenceFeatures;
31    import jalview.io.gff.SequenceOntologyFactory;
32    import jalview.io.gff.SequenceOntologyI;
33    import jalview.schemes.FeatureColour;
34    import jalview.schemes.FeatureSettingsAdapter;
35    import jalview.util.MapList;
36    import jalview.util.Platform;
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    /*
55    * accepts anything as we will attempt lookup of gene or
56    * transcript id or gene name
57    */
58    private static final Regex ACCESSION_REGEX = new Regex(".*");
59   
60    private static final EnsemblFeatureType[] FEATURES_TO_FETCH = {
61    EnsemblFeatureType.gene, EnsemblFeatureType.transcript,
62    EnsemblFeatureType.exon, EnsemblFeatureType.cds,
63    EnsemblFeatureType.variation };
64   
65    private static final String CHROMOSOME = "chromosome";
66   
67    /**
68    * Default constructor (to use rest.ensembl.org)
69    */
 
70  22 toggle public EnsemblGene()
71    {
72  22 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  3480 toggle @Override
86    public String getDbName()
87    {
88  3480 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  0 AlignmentI al = null;
144  0 for (String geneId : geneIds)
145    {
146    /*
147    * fetch the gene sequence(s) with features and xrefs
148    */
149  0 AlignmentI geneAlignment = super.getSequenceRecords(geneId);
150  0 if (geneAlignment == null)
151    {
152  0 continue;
153    }
154   
155  0 if (geneAlignment.getHeight() == 1)
156    {
157    // ensure id has 'correct' case for the Ensembl identifier
158  0 geneId = geneAlignment.getSequenceAt(0).getName();
159  0 findGeneLoci(geneAlignment.getSequenceAt(0), geneId);
160  0 getTranscripts(geneAlignment, geneId);
161    }
162  0 if (al == null)
163    {
164  0 al = geneAlignment;
165    }
166    else
167    {
168  0 al.append(geneAlignment);
169    }
170    }
171  0 return al;
172    }
173   
174    /**
175    * Calls the /lookup/id REST service, parses the response for gene
176    * coordinates, and if successful, adds these to the sequence. If this fails,
177    * fall back on trying to parse the sequence description in case it is in
178    * Ensembl-gene format e.g. chromosome:GRCh38:17:45051610:45109016:1.
179    *
180    * @param seq
181    * @param geneId
182    */
 
183  0 toggle void findGeneLoci(SequenceI seq, String geneId)
184    {
185  0 GeneLociI geneLoci = new EnsemblLookup(getDomain()).getGeneLoci(geneId);
186  0 if (geneLoci != null)
187    {
188  0 seq.setGeneLoci(geneLoci.getSpeciesId(), geneLoci.getAssemblyId(),
189    geneLoci.getChromosomeId(), geneLoci.getMapping());
190    }
191    else
192    {
193  0 parseChromosomeLocations(seq);
194    }
195    }
196   
197    /**
198    * Parses and saves fields of an Ensembl-style description e.g.
199    * chromosome:GRCh38:17:45051610:45109016:1
200    *
201    * @param seq
202    */
 
203  0 toggle boolean parseChromosomeLocations(SequenceI seq)
204    {
205  0 String description = seq.getDescription();
206  0 if (description == null)
207    {
208  0 return false;
209    }
210  0 String[] tokens = description.split(":");
211  0 if (tokens.length == 6 && tokens[0].startsWith(CHROMOSOME))
212    {
213  0 String ref = tokens[1];
214  0 String chrom = tokens[2];
215  0 try
216    {
217  0 int chStart = Integer.parseInt(tokens[3]);
218  0 int chEnd = Integer.parseInt(tokens[4]);
219  0 boolean forwardStrand = "1".equals(tokens[5]);
220  0 String species = ""; // not known here
221  0 int[] from = new int[] { seq.getStart(), seq.getEnd() };
222  0 int[] to = new int[] { forwardStrand ? chStart : chEnd,
223  0 forwardStrand ? chEnd : chStart };
224  0 MapList map = new MapList(from, to, 1, 1);
225  0 seq.setGeneLoci(species, ref, chrom, map);
226  0 return true;
227    } catch (NumberFormatException e)
228    {
229  0 jalview.bin.Console
230    .errPrintln("Bad integers in description " + description);
231    }
232    }
233  0 return false;
234    }
235   
236    /**
237    * Converts a query, which may contain one or more gene, transcript, or
238    * external (to Ensembl) identifiers, into a non-redundant list of gene
239    * identifiers.
240    *
241    * @param accessions
242    * @return
243    */
 
244  0 toggle List<String> getGeneIds(String accessions)
245    {
246  0 List<String> geneIds = new ArrayList<>();
247   
248  0 for (String acc : accessions.split(getAccessionSeparator()))
249    {
250    /*
251    * First try lookup as an Ensembl (gene or transcript) identifier
252    */
253  0 String geneId = new EnsemblLookup(getDomain()).getGeneId(acc);
254  0 if (geneId != null)
255    {
256  0 if (!geneIds.contains(geneId))
257    {
258  0 geneIds.add(geneId);
259    }
260    }
261    else
262    {
263    /*
264    * if given a gene or other external name, lookup and fetch
265    * the corresponding gene for all model organisms
266    */
267  0 List<String> ids = new EnsemblSymbol(getDomain(), getDbSource(),
268    getDbVersion()).getGeneIds(acc);
269  0 for (String id : ids)
270    {
271  0 if (!geneIds.contains(id))
272    {
273  0 geneIds.add(id);
274    }
275    }
276    }
277    }
278  0 return geneIds;
279    }
280   
281    /**
282    * Constructs all transcripts for the gene, as identified by "transcript"
283    * features whose Parent is the requested gene. The coding transcript
284    * sequences (i.e. with introns omitted) are added to the alignment.
285    *
286    * @param al
287    * @param accId
288    * @throws Exception
289    */
 
290  0 toggle protected void getTranscripts(AlignmentI al, String accId)
291    throws Exception
292    {
293  0 SequenceI gene = al.getSequenceAt(0);
294  0 List<SequenceFeature> transcriptFeatures = getTranscriptFeatures(accId,
295    gene);
296   
297  0 for (SequenceFeature transcriptFeature : transcriptFeatures)
298    {
299  0 makeTranscript(transcriptFeature, al, gene);
300    }
301   
302  0 clearGeneFeatures(gene);
303    }
304   
305    /**
306    * Remove unwanted features (transcript, exon, CDS) from the gene sequence
307    * after we have used them to derive transcripts and transfer features
308    *
309    * @param gene
310    */
 
311  0 toggle protected void clearGeneFeatures(SequenceI gene)
312    {
313    /*
314    * Note we include NMD_transcript_variant here because it behaves like
315    * 'transcript' in Ensembl, although strictly speaking it is not
316    * (it is a sub-type of sequence_variant)
317    */
318  0 String[] soTerms = new String[] {
319    SequenceOntologyI.NMD_TRANSCRIPT_VARIANT,
320    SequenceOntologyI.TRANSCRIPT, SequenceOntologyI.EXON,
321    SequenceOntologyI.CDS };
322  0 List<SequenceFeature> sfs = gene.getFeatures()
323    .getFeaturesByOntology(soTerms);
324  0 for (SequenceFeature sf : sfs)
325    {
326  0 gene.deleteFeature(sf);
327    }
328    }
329   
330    /**
331    * Constructs a spliced transcript sequence by finding 'exon' features for the
332    * given id (or failing that 'CDS'). Copies features on to the new sequence.
333    * 'Aligns' the new sequence against the gene sequence by padding with gaps,
334    * and adds it to the alignment.
335    *
336    * @param transcriptFeature
337    * @param al
338    * the alignment to which to add the new sequence
339    * @param gene
340    * the parent gene sequence, with features
341    * @return
342    */
 
343  0 toggle SequenceI makeTranscript(SequenceFeature transcriptFeature, AlignmentI al,
344    SequenceI gene)
345    {
346  0 String accId = getTranscriptId(transcriptFeature);
347  0 if (accId == null)
348    {
349  0 return null;
350    }
351   
352    /*
353    * NB we are mapping from gene sequence (not genome), so do not
354    * need to check for reverse strand (gene and transcript sequences
355    * are in forward sense)
356    */
357   
358    /*
359    * make a gene-length sequence filled with gaps
360    * we will fill in the bases for transcript regions
361    */
362  0 char[] seqChars = new char[gene.getLength()];
363  0 Arrays.fill(seqChars, al.getGapCharacter());
364   
365    /*
366    * look for exon features of the transcript, failing that for CDS
367    * (for example ENSG00000124610 has 1 CDS but no exon features)
368    */
369  0 String parentId = accId;
370  0 List<SequenceFeature> splices = findFeatures(gene,
371    SequenceOntologyI.EXON, parentId);
372  0 if (splices.isEmpty())
373    {
374  0 splices = findFeatures(gene, SequenceOntologyI.CDS, parentId);
375    }
376  0 SequenceFeatures.sortFeatures(splices, true);
377   
378  0 int transcriptLength = 0;
379  0 final char[] geneChars = gene.getSequence();
380  0 int offset = gene.getStart(); // to convert to 0-based positions
381  0 List<int[]> mappedFrom = new ArrayList<>();
382   
383  0 for (SequenceFeature sf : splices)
384    {
385  0 int start = sf.getBegin() - offset;
386  0 int end = sf.getEnd() - offset;
387  0 int spliceLength = end - start + 1;
388  0 System.arraycopy(geneChars, start, seqChars, start, spliceLength);
389  0 transcriptLength += spliceLength;
390  0 mappedFrom.add(new int[] { sf.getBegin(), sf.getEnd() });
391    }
392   
393  0 Sequence transcript = new Sequence(accId, seqChars, 1,
394    transcriptLength);
395   
396    /*
397    * Ensembl has gene name as transcript Name
398    * EnsemblGenomes doesn't, but has a url-encoded description field
399    */
400  0 String description = transcriptFeature.getDescription();
401  0 if (description == null)
402    {
403  0 description = (String) transcriptFeature.getValue(DESCRIPTION);
404    }
405  0 if (description != null)
406    {
407  0 try
408    {
409  0 transcript.setDescription(URLDecoder.decode(description, "UTF-8"));
410    } catch (UnsupportedEncodingException e)
411    {
412  0 e.printStackTrace(); // as if
413    }
414    }
415  0 transcript.createDatasetSequence();
416   
417  0 al.addSequence(transcript);
418   
419    /*
420    * transfer features to the new sequence; we use EnsemblCdna to do this,
421    * to filter out unwanted features types (see method retainFeature)
422    */
423  0 List<int[]> mapTo = new ArrayList<>();
424  0 mapTo.add(new int[] { 1, transcriptLength });
425  0 MapList mapping = new MapList(mappedFrom, mapTo, 1, 1);
426  0 EnsemblCdna cdna = new EnsemblCdna(getDomain());
427  0 cdna.transferFeatures(gene.getFeatures().getPositionalFeatures(),
428    transcript.getDatasetSequence(), mapping, parentId);
429   
430  0 mapTranscriptToChromosome(transcript, gene, mapping);
431   
432    /*
433    * fetch and save cross-references
434    */
435  0 cdna.getCrossReferences(transcript);
436   
437    /*
438    * and finally fetch the protein product and save as a cross-reference
439    */
440  0 cdna.addProteinProduct(transcript);
441   
442  0 return transcript;
443    }
444   
445    /**
446    * If the gene has a mapping to chromosome coordinates, derive the transcript
447    * chromosome regions and save on the transcript sequence
448    *
449    * @param transcript
450    * @param gene
451    * @param mapping
452    * the mapping from gene to transcript positions
453    */
 
454  0 toggle protected void mapTranscriptToChromosome(SequenceI transcript,
455    SequenceI gene, MapList mapping)
456    {
457  0 GeneLociI loci = gene.getGeneLoci();
458  0 if (loci == null)
459    {
460  0 return;
461    }
462   
463  0 MapList geneMapping = loci.getMapping();
464   
465  0 List<int[]> exons = mapping.getFromRanges();
466  0 List<int[]> transcriptLoci = new ArrayList<>();
467   
468  0 for (int[] exon : exons)
469    {
470  0 transcriptLoci.add(geneMapping.locateInTo(exon[0], exon[1]));
471    }
472   
473  0 List<int[]> transcriptRange = Arrays
474    .asList(new int[]
475    { transcript.getStart(), transcript.getEnd() });
476  0 MapList mapList = new MapList(transcriptRange, transcriptLoci, 1, 1);
477   
478  0 transcript.setGeneLoci(loci.getSpeciesId(), loci.getAssemblyId(),
479    loci.getChromosomeId(), mapList);
480    }
481   
482    /**
483    * Returns the 'transcript_id' property of the sequence feature (or null)
484    *
485    * @param feature
486    * @return
487    */
 
488  0 toggle protected String getTranscriptId(SequenceFeature feature)
489    {
490  0 return (String) feature.getValue(JSON_ID);
491    }
492   
493    /**
494    * Returns a list of the transcript features on the sequence whose Parent is
495    * the gene for the accession id.
496    * <p>
497    * Transcript features are those of type "transcript", or any of its sub-types
498    * in the Sequence Ontology e.g. "mRNA", "processed_transcript". We also
499    * include "NMD_transcript_variant", because this type behaves like a
500    * transcript identifier in Ensembl, although strictly speaking it is not in
501    * the SO.
502    *
503    * @param accId
504    * @param geneSequence
505    * @return
506    */
 
507  1 toggle protected List<SequenceFeature> getTranscriptFeatures(String accId,
508    SequenceI geneSequence)
509    {
510  1 List<SequenceFeature> transcriptFeatures = new ArrayList<>();
511   
512  1 String parentIdentifier = accId;
513   
514  1 List<SequenceFeature> sfs = geneSequence.getFeatures()
515    .getFeaturesByOntology(SequenceOntologyI.TRANSCRIPT);
516  1 sfs.addAll(geneSequence.getFeatures().getPositionalFeatures(
517    SequenceOntologyI.NMD_TRANSCRIPT_VARIANT));
518   
519  1 for (SequenceFeature sf : sfs)
520    {
521  4 String parent = (String) sf.getValue(PARENT);
522  4 if (parentIdentifier.equalsIgnoreCase(parent))
523    {
524  3 transcriptFeatures.add(sf);
525    }
526    }
527   
528  1 return transcriptFeatures;
529    }
530   
 
531  0 toggle @Override
532    public String getDescription()
533    {
534  0 return "Fetches all transcripts and variant features for a gene or transcript";
535    }
536   
537    /**
538    * Default test query is a gene id (can also enter a transcript id)
539    */
 
540  0 toggle @Override
541    public String getTestQuery()
542    {
543  0 return Platform.isJS() ? "ENSG00000123569" : "ENSG00000157764";
544    // ENSG00000123569 // H2BFWT histone, 2 transcripts, reverse strand
545    // ENSG00000157764 // BRAF, 5 transcripts, reverse strand
546    // ENSG00000090266 // NDUFB2, 15 transcripts, forward strand
547    // ENSG00000101812 // H2BFM histone, 3 transcripts, forward strand
548    }
549   
550    /**
551    * Answers a list of sequence features (if any) whose type is 'gene' (or a
552    * subtype of gene in the Sequence Ontology), and whose ID is the accession we
553    * are retrieving
554    */
 
555  3 toggle @Override
556    protected List<SequenceFeature> getIdentifyingFeatures(SequenceI seq,
557    String accId)
558    {
559  3 List<SequenceFeature> result = new ArrayList<>();
560  3 List<SequenceFeature> sfs = seq.getFeatures()
561    .getFeaturesByOntology(SequenceOntologyI.GENE);
562  3 for (SequenceFeature sf : sfs)
563    {
564  6 String id = (String) sf.getValue(JSON_ID);
565  6 if (accId.equalsIgnoreCase(id))
566    {
567  4 result.add(sf);
568    }
569    }
570  3 return result;
571    }
572   
573    /**
574    * Answers true unless feature type is 'gene', or 'transcript' with a parent
575    * which is a different gene. We need the gene features to identify the range,
576    * but it is redundant information on the gene sequence. Checking the parent
577    * allows us to drop transcript features which belong to different
578    * (overlapping) genes.
579    */
 
580  6 toggle @Override
581    protected boolean retainFeature(SequenceFeature sf, String accessionId)
582    {
583  6 SequenceOntologyI so = SequenceOntologyFactory.getInstance();
584  6 String type = sf.getType();
585  6 if (so.isA(type, SequenceOntologyI.GENE))
586    {
587  1 return false;
588    }
589  5 if (isTranscript(type))
590    {
591  4 String parent = (String) sf.getValue(PARENT);
592  4 if (!accessionId.equalsIgnoreCase(parent))
593    {
594  1 return false;
595    }
596    }
597  4 return true;
598    }
599   
600    /**
601    * Override to do nothing as Ensembl doesn't return a protein sequence for a
602    * gene identifier
603    */
 
604  0 toggle @Override
605    protected void addProteinProduct(SequenceI querySeq)
606    {
607    }
608   
 
609  0 toggle @Override
610    public Regex getAccessionValidator()
611    {
612  0 return ACCESSION_REGEX;
613    }
614   
615    /**
616    * Returns a descriptor for suitable feature display settings with
617    * <ul>
618    * <li>only exon or sequence_variant features (or their subtypes in the
619    * Sequence Ontology) visible</li>
620    * <li>variant features coloured red</li>
621    * <li>exon features coloured by label (exon name)</li>
622    * <li>variants displayed above (on top of) exons</li>
623    * </ul>
624    */
 
625  3 toggle @Override
626    public FeatureSettingsModelI getFeatureColourScheme()
627    {
628  3 return new FeatureSettingsAdapter()
629    {
630    SequenceOntologyI so = SequenceOntologyFactory.getInstance();
631   
 
632  20 toggle @Override
633    public boolean isFeatureHidden(String type)
634    {
635  20 return (!so.isA(type, SequenceOntologyI.EXON)
636    && !so.isA(type, SequenceOntologyI.SEQUENCE_VARIANT));
637    }
638   
 
639  22 toggle @Override
640    public FeatureColourI getFeatureColour(String type)
641    {
642  22 if (so.isA(type, SequenceOntologyI.EXON))
643    {
644  8 return new FeatureColour()
645    {
 
646  2 toggle @Override
647    public boolean isColourByLabel()
648    {
649  2 return true;
650    }
651    };
652    }
653  14 if (so.isA(type, SequenceOntologyI.SEQUENCE_VARIANT))
654    {
655  12 return new FeatureColour()
656    {
657   
 
658  2 toggle @Override
659    public Color getColour()
660    {
661  2 return Color.RED;
662    }
663    };
664    }
665  2 return null;
666    }
667   
668    /**
669    * order to render sequence_variant after exon after the rest
670    */
 
671  16 toggle @Override
672    public int compare(String feature1, String feature2)
673    {
674  16 if (so.isA(feature1, SequenceOntologyI.SEQUENCE_VARIANT))
675    {
676  12 return +1;
677    }
678  4 if (so.isA(feature2, SequenceOntologyI.SEQUENCE_VARIANT))
679    {
680  2 return -1;
681    }
682  2 if (so.isA(feature1, SequenceOntologyI.EXON))
683    {
684  2 return +1;
685    }
686  0 if (so.isA(feature2, SequenceOntologyI.EXON))
687    {
688  0 return -1;
689    }
690  0 return 0;
691    }
692    };
693    }
694   
695    }