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.datamodel.xdb.embl;
23 import jalview.analysis.SequenceIdMatcher;
24 import jalview.bin.Cache;
25 import jalview.datamodel.DBRefEntry;
26 import jalview.datamodel.DBRefSource;
27 import jalview.datamodel.FeatureProperties;
28 import jalview.datamodel.Mapping;
29 import jalview.datamodel.Sequence;
30 import jalview.datamodel.SequenceFeature;
31 import jalview.datamodel.SequenceI;
32 import jalview.util.DBRefUtils;
33 import jalview.util.DnaUtils;
34 import jalview.util.MapList;
35 import jalview.util.MappingUtils;
36 import jalview.util.StringUtils;
38 import java.text.ParseException;
39 import java.util.Arrays;
40 import java.util.Hashtable;
41 import java.util.List;
43 import java.util.Map.Entry;
44 import java.util.Vector;
45 import java.util.regex.Pattern;
48 * Data model for one entry returned from an EMBL query, as marshalled by a
52 * http://www.ebi.ac.uk/Tools/dbfetch/dbfetch?db=ena_sequence&id=J03321
55 * @see embl_mapping.xml
57 public class EmblEntry
59 private static final Pattern SPACE_PATTERN = Pattern.compile(" ");
65 String sequenceVersion;
73 String sequenceLength;
75 String taxonomicDivision;
79 String firstPublicDate;
81 String firstPublicRelease;
83 String lastUpdatedDate;
85 String lastUpdatedRelease;
87 Vector<String> keywords;
89 Vector<DBRefEntry> dbRefs;
91 Vector<EmblFeature> features;
93 EmblSequence sequence;
96 * @return the accession
98 public String getAccession()
105 * the accession to set
107 public void setAccession(String accession)
109 this.accession = accession;
115 public Vector<DBRefEntry> getDbRefs()
124 public void setDbRefs(Vector<DBRefEntry> dbRefs)
126 this.dbRefs = dbRefs;
130 * @return the features
132 public Vector<EmblFeature> getFeatures()
139 * the features to set
141 public void setFeatures(Vector<EmblFeature> features)
143 this.features = features;
147 * @return the keywords
149 public Vector<String> getKeywords()
156 * the keywords to set
158 public void setKeywords(Vector<String> keywords)
160 this.keywords = keywords;
164 * @return the sequence
166 public EmblSequence getSequence()
173 * the sequence to set
175 public void setSequence(EmblSequence sequence)
177 this.sequence = sequence;
181 * Recover annotated sequences from EMBL file
185 * a list of protein products found so far (to add to)
186 * @return dna dataset sequence with DBRefs and features
188 public SequenceI getSequence(String sourceDb, List<SequenceI> peptides)
190 SequenceI dna = new Sequence(sourceDb + "|" + accession,
191 sequence.getSequence());
192 dna.setDescription(description);
193 DBRefEntry retrievedref = new DBRefEntry(sourceDb,
194 getSequenceVersion(), accession);
195 dna.addDBRef(retrievedref);
196 // add map to indicate the sequence is a valid coordinate frame for the
198 retrievedref.setMap(new Mapping(null, new int[] { 1, dna.getLength() },
199 new int[] { 1, dna.getLength() }, 1, 1));
203 * transform EMBL Database refs to canonical form
207 for (DBRefEntry dbref : dbRefs)
209 dbref.setSource(DBRefUtils.getCanonicalName(dbref.getSource()));
214 SequenceIdMatcher matcher = new SequenceIdMatcher(peptides);
217 for (EmblFeature feature : features)
219 if (FeatureProperties.isCodingFeature(sourceDb, feature.getName()))
221 parseCodingFeature(feature, sourceDb, dna, peptides, matcher);
224 } catch (Exception e)
226 System.err.println("EMBL Record Features parsing error!");
228 .println("Please report the following to help@jalview.org :");
229 System.err.println("EMBL Record " + accession);
230 System.err.println("Resulted in exception: " + e.getMessage());
231 e.printStackTrace(System.err);
238 * Extracts coding region and product from a CDS feature and properly decorate
239 * it with annotations.
244 * source database for the EMBLXML
246 * parent dna sequence for this record
248 * list of protein product sequences for Embl entry
250 * helper to match xrefs in already retrieved sequences
252 void parseCodingFeature(EmblFeature feature, String sourceDb,
253 SequenceI dna, List<SequenceI> peptides, SequenceIdMatcher matcher)
255 boolean isEmblCdna = sourceDb.equals(DBRefSource.EMBLCDS);
257 int[] exons = getCdsRanges(feature);
259 String translation = null;
260 String proteinName = "";
261 String proteinId = null;
262 Map<String, String> vals = new Hashtable<String, String>();
265 * codon_start 1/2/3 in EMBL corresponds to phase 0/1/2 in CDS
266 * (phase is required for CDS features in GFF3 format)
271 * parse qualifiers, saving protein translation, protein id,
272 * codon start position, product (name), and 'other values'
274 if (feature.getQualifiers() != null)
276 for (Qualifier q : feature.getQualifiers())
278 String qname = q.getName();
279 if (qname.equals("translation"))
281 // remove all spaces (precompiled String.replaceAll(" ", ""))
282 translation = SPACE_PATTERN.matcher(q.getValues()[0]).replaceAll("");
284 else if (qname.equals("protein_id"))
286 proteinId = q.getValues()[0].trim();
288 else if (qname.equals("codon_start"))
292 codonStart = Integer.parseInt(q.getValues()[0].trim());
293 } catch (NumberFormatException e)
295 System.err.println("Invalid codon_start in XML for "
296 + accession + ": " + e.getMessage());
299 else if (qname.equals("product"))
301 // sometimes name is returned e.g. for V00488
302 proteinName = q.getValues()[0].trim();
306 // throw anything else into the additional properties hash
307 String[] qvals = q.getValues();
310 String commaSeparated = StringUtils.arrayToSeparatorList(qvals,
312 vals.put(qname, commaSeparated);
318 DBRefEntry proteinToEmblProteinRef = null;
319 exons = MappingUtils.removeStartPositions(codonStart - 1, exons);
321 SequenceI product = null;
322 Mapping dnaToProteinMapping = null;
323 if (translation != null && proteinName != null && proteinId != null)
326 * look for product in peptides list, if not found, add it
328 product = matcher.findIdMatch(proteinId);
331 product = new Sequence(proteinId, translation, 1, translation.length());
332 product.setDescription(((proteinName.length() == 0) ? "Protein Product from "
335 peptides.add(product);
336 matcher.add(product);
339 // we have everything - create the mapping and perhaps the protein
341 if (exons == null || exons.length == 0)
344 .println("Implementation Notice: EMBLCDS records not properly supported yet - Making up the CDNA region of this sequence... may be incorrect ("
345 + sourceDb + ":" + getAccession() + ")");
346 if (translation.length() * 3 == (1 - codonStart + dna.getSequence().length))
349 .println("Not allowing for additional stop codon at end of cDNA fragment... !");
350 // this might occur for CDS sequences where no features are
352 exons = new int[] { dna.getStart() + (codonStart - 1),
354 dnaToProteinMapping = new Mapping(product, exons, new int[] { 1,
355 translation.length() },
358 if ((translation.length() + 1) * 3 == (1 - codonStart + dna.getSequence().length))
361 .println("Allowing for additional stop codon at end of cDNA fragment... will probably cause an error in VAMSAs!");
362 exons = new int[] { dna.getStart() + (codonStart - 1),
364 dnaToProteinMapping = new Mapping(product, exons, new int[] { 1,
365 translation.length() },
371 // Trim the exon mapping if necessary - the given product may only be a
372 // fragment of a larger protein. (EMBL:AY043181 is an example)
376 // TODO: Add a DbRef back to the parent EMBL sequence with the exon
378 // if given a dataset reference, search dataset for parent EMBL
379 // sequence if it exists and set its map
380 // make a new feature annotating the coding contig
384 // final product length truncation check
385 int[] cdsRanges = adjustForProteinLength(translation.length(), exons);
386 dnaToProteinMapping = new Mapping(product, cdsRanges, new int[] { 1,
387 translation.length() }, 3, 1);
391 * make xref from protein to EMBLCDS; we assume here that the
392 * CDS sequence version is same as dna sequence (?!)
394 MapList proteinToCdsMapList = new MapList(new int[] { 1,
395 translation.length() }, new int[] { 1 + (codonStart - 1),
396 (codonStart - 1) + 3 * translation.length() }, 1, 3);
397 DBRefEntry proteinToEmblCdsRef = new DBRefEntry(
398 DBRefSource.EMBLCDS, getSequenceVersion(), proteinId,
399 new Mapping(proteinToCdsMapList));
400 product.addDBRef(proteinToEmblCdsRef);
403 * make xref from protein to EMBLCDSPROTEIN
405 proteinToEmblProteinRef = new DBRefEntry(proteinToEmblCdsRef);
406 proteinToEmblProteinRef.setSource(DBRefSource.EMBLCDSProduct);
407 product.addDBRef(proteinToEmblProteinRef);
413 * add cds features to dna sequence
415 for (int xint = 0; exons != null && xint < exons.length; xint += 2)
417 SequenceFeature sf = makeCdsFeature(exons, xint, proteinName, proteinId, vals,
419 sf.setType(feature.getName()); // "CDS"
420 sf.setEnaLocation(feature.getLocation());
421 sf.setFeatureGroup(sourceDb);
422 dna.addSequenceFeature(sf);
427 * add feature dbRefs to sequence, and mappings for Uniprot xrefs
429 boolean hasUniprotDbref = false;
430 if (feature.dbRefs != null)
432 boolean mappingUsed = false;
433 for (DBRefEntry ref : feature.dbRefs)
436 * ensure UniProtKB/Swiss-Prot converted to UNIPROT
438 String source = DBRefUtils.getCanonicalName(ref.getSource());
439 ref.setSource(source);
440 DBRefEntry proteinToDnaRef = new DBRefEntry(ref.getSource(), ref.getVersion(), ref
442 if (source.equals(DBRefSource.UNIPROT))
444 String proteinSeqName = DBRefSource.UNIPROT + "|"
445 + ref.getAccessionId();
446 if (dnaToProteinMapping != null && dnaToProteinMapping.getTo() != null)
451 * two or more Uniprot xrefs for the same CDS -
452 * each needs a distinct Mapping (as to a different sequence)
454 dnaToProteinMapping = new Mapping(dnaToProteinMapping);
459 * try to locate the protein mapped to (possibly by a
460 * previous CDS feature); if not found, construct it from
461 * the EMBL translation
463 SequenceI proteinSeq = matcher.findIdMatch(proteinSeqName);
464 if (proteinSeq == null)
466 proteinSeq = new Sequence(proteinSeqName,
467 product.getSequenceAsString());
468 matcher.add(proteinSeq);
469 peptides.add(proteinSeq);
471 dnaToProteinMapping.setTo(proteinSeq);
472 proteinSeq.addDBRef(proteinToDnaRef);
473 ref.setMap(dnaToProteinMapping);
475 hasUniprotDbref = true;
480 * copy feature dbref to our protein product
482 DBRefEntry pref = proteinToDnaRef;
483 pref.setMap(null); // reference is direct
484 product.addDBRef(pref);
485 // Add converse mapping reference
486 if (dnaToProteinMapping != null)
488 Mapping pmap = new Mapping(dna, dnaToProteinMapping.getMap()
490 pref = new DBRefEntry(sourceDb, getSequenceVersion(),
491 this.getAccession());
493 if (dnaToProteinMapping.getTo() != null)
495 dnaToProteinMapping.getTo().addDBRef(pref);
504 * if we have a product (translation) but no explicit Uniprot dbref
505 * (example: EMBL AAFI02000057 protein_id EAL65544.1)
506 * then construct mappings to an assumed EMBLCDSPROTEIN accession
508 if (!hasUniprotDbref && product != null)
510 if (proteinToEmblProteinRef == null)
512 // assuming CDSPROTEIN sequence version = dna version (?!)
513 proteinToEmblProteinRef = new DBRefEntry(
514 DBRefSource.EMBLCDSProduct, getSequenceVersion(), proteinId);
516 product.addDBRef(proteinToEmblProteinRef);
518 if (dnaToProteinMapping != null
519 && dnaToProteinMapping.getTo() != null)
521 DBRefEntry dnaToEmblProteinRef = new DBRefEntry(
522 DBRefSource.EMBLCDSProduct, getSequenceVersion(), proteinId);
523 dnaToEmblProteinRef.setMap(dnaToProteinMapping);
524 dna.addDBRef(dnaToEmblProteinRef);
530 * Helper method to construct a SequenceFeature for one cds range
533 * array of cds [start, end, ...] positions
534 * @param exonStartIndex
535 * offset into the exons array
537 * @param proteinAccessionId
539 * map of 'miscellaneous values' for feature
541 * codon start position for CDS (1/2/3, normally 1)
544 protected SequenceFeature makeCdsFeature(int[] exons, int exonStartIndex,
545 String proteinName, String proteinAccessionId,
546 Map<String, String> vals, int codonStart)
548 int exonNumber = exonStartIndex / 2 + 1;
549 SequenceFeature sf = new SequenceFeature();
550 sf.setBegin(Math.min(exons[exonStartIndex], exons[exonStartIndex + 1]));
551 sf.setEnd(Math.max(exons[exonStartIndex], exons[exonStartIndex + 1]));
552 sf.setDescription(String.format("Exon %d for protein '%s' EMBLCDS:%s",
553 exonNumber, proteinName, proteinAccessionId));
554 sf.setPhase(String.valueOf(codonStart - 1));
555 sf.setStrand(exons[exonStartIndex] <= exons[exonStartIndex + 1] ? "+"
557 sf.setValue(FeatureProperties.EXONPOS, exonNumber);
558 sf.setValue(FeatureProperties.EXONPRODUCT, proteinName);
561 StringBuilder sb = new StringBuilder();
562 boolean first = true;
563 for (Entry<String, String> val : vals.entrySet())
569 sb.append(val.getKey()).append("=").append(val.getValue());
571 sf.setValue(val.getKey(), val.getValue());
573 sf.setAttributes(sb.toString());
579 * Returns the CDS positions as a single array of [start, end, start, end...]
580 * positions. If on the reverse strand, these will be in descending order.
585 protected int[] getCdsRanges(EmblFeature feature)
587 if (feature.location == null)
594 List<int[]> ranges = DnaUtils.parseLocation(feature.location);
595 return listToArray(ranges);
596 } catch (ParseException e)
598 Cache.log.warn(String.format(
599 "Not parsing inexact CDS location %s in ENA %s",
600 feature.location, this.accession));
606 * Converts a list of [start, end] ranges to a single array of [start, end,
612 int[] listToArray(List<int[]> ranges)
614 int[] result = new int[ranges.size() * 2];
616 for (int[] range : ranges)
618 result[i++] = range[0];
619 result[i++] = range[1];
625 * Truncates (if necessary) the exon intervals to match 3 times the length of
626 * the protein; also accepts 3 bases longer (for stop codon not included in
629 * @param proteinLength
631 * an array of [start, end, start, end...] intervals
632 * @return the same array (if unchanged) or a truncated copy
634 static int[] adjustForProteinLength(int proteinLength, int[] exon)
636 if (proteinLength <= 0 || exon == null)
640 int expectedCdsLength = proteinLength * 3;
641 int exonLength = MappingUtils.getLength(Arrays.asList(exon));
644 * if exon length matches protein, or is shorter, or longer by the
645 * length of a stop codon (3 bases), then leave it unchanged
647 if (expectedCdsLength >= exonLength
648 || expectedCdsLength == exonLength - 3)
656 origxon = new int[exon.length];
657 System.arraycopy(exon, 0, origxon, 0, exon.length);
659 for (int x = 0; x < exon.length; x += 2)
661 cdspos += Math.abs(exon[x + 1] - exon[x]) + 1;
662 if (expectedCdsLength <= cdspos)
664 // advanced beyond last codon.
666 if (expectedCdsLength != cdspos)
669 // .println("Truncating final exon interval on region by "
670 // + (cdspos - cdslength));
674 * shrink the final exon - reduce end position if forward
675 * strand, increase it if reverse
677 if (exon[x + 1] >= exon[x])
679 endxon = exon[x + 1] - cdspos + expectedCdsLength;
683 endxon = exon[x + 1] + cdspos - expectedCdsLength;
691 // and trim the exon interval set if necessary
692 int[] nxon = new int[sxpos + 2];
693 System.arraycopy(exon, 0, nxon, 0, sxpos + 2);
694 nxon[sxpos + 1] = endxon; // update the end boundary for the new exon
701 public String getSequenceVersion()
703 return sequenceVersion;
706 public void setSequenceVersion(String sequenceVersion)
708 this.sequenceVersion = sequenceVersion;
711 public String getSequenceLength()
713 return sequenceLength;
716 public void setSequenceLength(String sequenceLength)
718 this.sequenceLength = sequenceLength;
721 public String getEntryVersion()
726 public void setEntryVersion(String entryVersion)
728 this.entryVersion = entryVersion;
731 public String getMoleculeType()
736 public void setMoleculeType(String moleculeType)
738 this.moleculeType = moleculeType;
741 public String getTopology()
746 public void setTopology(String topology)
748 this.topology = topology;
751 public String getTaxonomicDivision()
753 return taxonomicDivision;
756 public void setTaxonomicDivision(String taxonomicDivision)
758 this.taxonomicDivision = taxonomicDivision;
761 public String getDescription()
766 public void setDescription(String description)
768 this.description = description;
771 public String getFirstPublicDate()
773 return firstPublicDate;
776 public void setFirstPublicDate(String firstPublicDate)
778 this.firstPublicDate = firstPublicDate;
781 public String getFirstPublicRelease()
783 return firstPublicRelease;
786 public void setFirstPublicRelease(String firstPublicRelease)
788 this.firstPublicRelease = firstPublicRelease;
791 public String getLastUpdatedDate()
793 return lastUpdatedDate;
796 public void setLastUpdatedDate(String lastUpdatedDate)
798 this.lastUpdatedDate = lastUpdatedDate;
801 public String getLastUpdatedRelease()
803 return lastUpdatedRelease;
806 public void setLastUpdatedRelease(String lastUpdatedRelease)
808 this.lastUpdatedRelease = lastUpdatedRelease;
811 public String getDataClass()
816 public void setDataClass(String dataClass)
818 this.dataClass = dataClass;