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 = makeSequence(sourceDb);
191 dna.setDescription(description);
192 DBRefEntry retrievedref = new DBRefEntry(sourceDb,
193 getSequenceVersion(), accession);
194 dna.addDBRef(retrievedref);
195 // add map to indicate the sequence is a valid coordinate frame for the
197 retrievedref.setMap(new Mapping(null, new int[] { 1, dna.getLength() },
198 new int[] { 1, dna.getLength() }, 1, 1));
202 * transform EMBL Database refs to canonical form
206 for (DBRefEntry dbref : dbRefs)
208 dbref.setSource(DBRefUtils.getCanonicalName(dbref.getSource()));
213 SequenceIdMatcher matcher = new SequenceIdMatcher(peptides);
216 for (EmblFeature feature : features)
218 if (FeatureProperties.isCodingFeature(sourceDb, feature.getName()))
220 parseCodingFeature(feature, sourceDb, dna, peptides, matcher);
223 } catch (Exception e)
225 System.err.println("EMBL Record Features parsing error!");
227 .println("Please report the following to help@jalview.org :");
228 System.err.println("EMBL Record " + accession);
229 System.err.println("Resulted in exception: " + e.getMessage());
230 e.printStackTrace(System.err);
240 SequenceI makeSequence(String sourceDb)
242 SequenceI dna = new Sequence(sourceDb + "|" + accession,
243 sequence.getSequence());
248 * Extracts coding region and product from a CDS feature and properly decorate
249 * it with annotations.
254 * source database for the EMBLXML
256 * parent dna sequence for this record
258 * list of protein product sequences for Embl entry
260 * helper to match xrefs in already retrieved sequences
262 void parseCodingFeature(EmblFeature feature, String sourceDb,
263 SequenceI dna, List<SequenceI> peptides, SequenceIdMatcher matcher)
265 boolean isEmblCdna = sourceDb.equals(DBRefSource.EMBLCDS);
267 int[] exons = getCdsRanges(feature);
269 String translation = null;
270 String proteinName = "";
271 String proteinId = null;
272 Map<String, String> vals = new Hashtable<String, String>();
275 * codon_start 1/2/3 in EMBL corresponds to phase 0/1/2 in CDS
276 * (phase is required for CDS features in GFF3 format)
281 * parse qualifiers, saving protein translation, protein id,
282 * codon start position, product (name), and 'other values'
284 if (feature.getQualifiers() != null)
286 for (Qualifier q : feature.getQualifiers())
288 String qname = q.getName();
289 if (qname.equals("translation"))
291 // remove all spaces (precompiled String.replaceAll(" ", ""))
292 translation = SPACE_PATTERN.matcher(q.getValues()[0]).replaceAll("");
294 else if (qname.equals("protein_id"))
296 proteinId = q.getValues()[0].trim();
298 else if (qname.equals("codon_start"))
302 codonStart = Integer.parseInt(q.getValues()[0].trim());
303 } catch (NumberFormatException e)
305 System.err.println("Invalid codon_start in XML for "
306 + accession + ": " + e.getMessage());
309 else if (qname.equals("product"))
311 // sometimes name is returned e.g. for V00488
312 proteinName = q.getValues()[0].trim();
316 // throw anything else into the additional properties hash
317 String[] qvals = q.getValues();
320 String commaSeparated = StringUtils.arrayToSeparatorList(qvals,
322 vals.put(qname, commaSeparated);
328 DBRefEntry proteinToEmblProteinRef = null;
329 exons = MappingUtils.removeStartPositions(codonStart - 1, exons);
331 SequenceI product = null;
332 Mapping dnaToProteinMapping = null;
333 if (translation != null && proteinName != null && proteinId != null)
335 int translationLength = translation.length();
338 * look for product in peptides list, if not found, add it
340 product = matcher.findIdMatch(proteinId);
343 product = new Sequence(proteinId, translation, 1, translationLength);
344 product.setDescription(((proteinName.length() == 0) ? "Protein Product from "
347 peptides.add(product);
348 matcher.add(product);
351 // we have everything - create the mapping and perhaps the protein
353 if (exons == null || exons.length == 0)
356 * workaround until we handle dna location for CDS sequence
357 * e.g. location="X53828.1:60..1058" correctly
360 .println("Implementation Notice: EMBLCDS records not properly supported yet - Making up the CDNA region of this sequence... may be incorrect ("
361 + sourceDb + ":" + getAccession() + ")");
362 if (translationLength * 3 == (1 - codonStart + dna.getSequence().length))
365 .println("Not allowing for additional stop codon at end of cDNA fragment... !");
366 // this might occur for CDS sequences where no features are marked
367 exons = new int[] { dna.getStart() + (codonStart - 1),
369 dnaToProteinMapping = new Mapping(product, exons, new int[] { 1,
370 translationLength }, 3, 1);
372 if ((translationLength + 1) * 3 == (1 - codonStart + dna
373 .getSequence().length))
376 .println("Allowing for additional stop codon at end of cDNA fragment... will probably cause an error in VAMSAs!");
377 exons = new int[] { dna.getStart() + (codonStart - 1),
379 dnaToProteinMapping = new Mapping(product, exons, new int[] { 1,
380 translationLength }, 3, 1);
385 // Trim the exon mapping if necessary - the given product may only be a
386 // fragment of a larger protein. (EMBL:AY043181 is an example)
390 // TODO: Add a DbRef back to the parent EMBL sequence with the exon
392 // if given a dataset reference, search dataset for parent EMBL
393 // sequence if it exists and set its map
394 // make a new feature annotating the coding contig
398 // final product length truncation check
399 int[] cdsRanges = adjustForProteinLength(translationLength, exons);
400 dnaToProteinMapping = new Mapping(product, cdsRanges, new int[] {
401 1, translationLength }, 3, 1);
405 * make xref with mapping from protein to EMBL dna
407 DBRefEntry proteinToEmblRef = new DBRefEntry(DBRefSource.EMBL,
408 getSequenceVersion(), proteinId, new Mapping(
409 dnaToProteinMapping.getMap().getInverse()));
410 product.addDBRef(proteinToEmblRef);
413 * make xref from protein to EMBLCDS; we assume here that the
414 * CDS sequence version is same as dna sequence (?!)
416 MapList proteinToCdsMapList = new MapList(new int[] { 1,
417 translationLength }, new int[] { 1 + (codonStart - 1),
418 (codonStart - 1) + 3 * translationLength }, 1, 3);
419 DBRefEntry proteinToEmblCdsRef = new DBRefEntry(
420 DBRefSource.EMBLCDS, getSequenceVersion(), proteinId,
421 new Mapping(proteinToCdsMapList));
422 product.addDBRef(proteinToEmblCdsRef);
425 * make 'direct' xref from protein to EMBLCDSPROTEIN
427 proteinToEmblProteinRef = new DBRefEntry(proteinToEmblCdsRef);
428 proteinToEmblProteinRef.setSource(DBRefSource.EMBLCDSProduct);
429 proteinToEmblProteinRef.setMap(null);
430 product.addDBRef(proteinToEmblProteinRef);
436 * add cds features to dna sequence
438 for (int xint = 0; exons != null && xint < exons.length; xint += 2)
440 SequenceFeature sf = makeCdsFeature(exons, xint, proteinName,
441 proteinId, vals, codonStart);
442 sf.setType(feature.getName()); // "CDS"
443 sf.setEnaLocation(feature.getLocation());
444 sf.setFeatureGroup(sourceDb);
445 dna.addSequenceFeature(sf);
450 * add feature dbRefs to sequence, and mappings for Uniprot xrefs
452 boolean hasUniprotDbref = false;
453 if (feature.dbRefs != null)
455 boolean mappingUsed = false;
456 for (DBRefEntry ref : feature.dbRefs)
459 * ensure UniProtKB/Swiss-Prot converted to UNIPROT
461 String source = DBRefUtils.getCanonicalName(ref.getSource());
462 ref.setSource(source);
463 DBRefEntry proteinDbRef = new DBRefEntry(ref.getSource(), ref.getVersion(), ref
465 if (source.equals(DBRefSource.UNIPROT))
467 String proteinSeqName = DBRefSource.UNIPROT + "|"
468 + ref.getAccessionId();
469 if (dnaToProteinMapping != null && dnaToProteinMapping.getTo() != null)
474 * two or more Uniprot xrefs for the same CDS -
475 * each needs a distinct Mapping (as to a different sequence)
477 dnaToProteinMapping = new Mapping(dnaToProteinMapping);
482 * try to locate the protein mapped to (possibly by a
483 * previous CDS feature); if not found, construct it from
484 * the EMBL translation
486 SequenceI proteinSeq = matcher.findIdMatch(proteinSeqName);
487 if (proteinSeq == null)
489 proteinSeq = new Sequence(proteinSeqName,
490 product.getSequenceAsString());
491 matcher.add(proteinSeq);
492 peptides.add(proteinSeq);
494 dnaToProteinMapping.setTo(proteinSeq);
495 dnaToProteinMapping.setMappedFromId(proteinId);
496 proteinSeq.addDBRef(proteinDbRef);
497 ref.setMap(dnaToProteinMapping);
499 hasUniprotDbref = true;
504 * copy feature dbref to our protein product
506 DBRefEntry pref = proteinDbRef;
507 pref.setMap(null); // reference is direct
508 product.addDBRef(pref);
509 // Add converse mapping reference
510 if (dnaToProteinMapping != null)
512 Mapping pmap = new Mapping(dna, dnaToProteinMapping.getMap()
514 pref = new DBRefEntry(sourceDb, getSequenceVersion(),
515 this.getAccession());
517 if (dnaToProteinMapping.getTo() != null)
519 dnaToProteinMapping.getTo().addDBRef(pref);
528 * if we have a product (translation) but no explicit Uniprot dbref
529 * (example: EMBL AAFI02000057 protein_id EAL65544.1)
530 * then construct mappings to an assumed EMBLCDSPROTEIN accession
532 if (!hasUniprotDbref && product != null)
534 if (proteinToEmblProteinRef == null)
536 // assuming CDSPROTEIN sequence version = dna version (?!)
537 proteinToEmblProteinRef = new DBRefEntry(
538 DBRefSource.EMBLCDSProduct, getSequenceVersion(), proteinId);
540 product.addDBRef(proteinToEmblProteinRef);
542 if (dnaToProteinMapping != null
543 && dnaToProteinMapping.getTo() != null)
545 DBRefEntry dnaToEmblProteinRef = new DBRefEntry(
546 DBRefSource.EMBLCDSProduct, getSequenceVersion(), proteinId);
547 dnaToEmblProteinRef.setMap(dnaToProteinMapping);
548 dnaToProteinMapping.setMappedFromId(proteinId);
549 dna.addDBRef(dnaToEmblProteinRef);
555 * Helper method to construct a SequenceFeature for one cds range
558 * array of cds [start, end, ...] positions
559 * @param exonStartIndex
560 * offset into the exons array
562 * @param proteinAccessionId
564 * map of 'miscellaneous values' for feature
566 * codon start position for CDS (1/2/3, normally 1)
569 protected SequenceFeature makeCdsFeature(int[] exons, int exonStartIndex,
570 String proteinName, String proteinAccessionId,
571 Map<String, String> vals, int codonStart)
573 int exonNumber = exonStartIndex / 2 + 1;
574 SequenceFeature sf = new SequenceFeature();
575 sf.setBegin(Math.min(exons[exonStartIndex], exons[exonStartIndex + 1]));
576 sf.setEnd(Math.max(exons[exonStartIndex], exons[exonStartIndex + 1]));
577 sf.setDescription(String.format("Exon %d for protein '%s' EMBLCDS:%s",
578 exonNumber, proteinName, proteinAccessionId));
579 sf.setPhase(String.valueOf(codonStart - 1));
580 sf.setStrand(exons[exonStartIndex] <= exons[exonStartIndex + 1] ? "+"
582 sf.setValue(FeatureProperties.EXONPOS, exonNumber);
583 sf.setValue(FeatureProperties.EXONPRODUCT, proteinName);
586 StringBuilder sb = new StringBuilder();
587 boolean first = true;
588 for (Entry<String, String> val : vals.entrySet())
594 sb.append(val.getKey()).append("=").append(val.getValue());
596 sf.setValue(val.getKey(), val.getValue());
598 sf.setAttributes(sb.toString());
604 * Returns the CDS positions as a single array of [start, end, start, end...]
605 * positions. If on the reverse strand, these will be in descending order.
610 protected int[] getCdsRanges(EmblFeature feature)
612 if (feature.location == null)
619 List<int[]> ranges = DnaUtils.parseLocation(feature.location);
620 return listToArray(ranges);
621 } catch (ParseException e)
623 Cache.log.warn(String.format(
624 "Not parsing inexact CDS location %s in ENA %s",
625 feature.location, this.accession));
631 * Converts a list of [start, end] ranges to a single array of [start, end,
637 int[] listToArray(List<int[]> ranges)
639 int[] result = new int[ranges.size() * 2];
641 for (int[] range : ranges)
643 result[i++] = range[0];
644 result[i++] = range[1];
650 * Truncates (if necessary) the exon intervals to match 3 times the length of
651 * the protein; also accepts 3 bases longer (for stop codon not included in
654 * @param proteinLength
656 * an array of [start, end, start, end...] intervals
657 * @return the same array (if unchanged) or a truncated copy
659 static int[] adjustForProteinLength(int proteinLength, int[] exon)
661 if (proteinLength <= 0 || exon == null)
665 int expectedCdsLength = proteinLength * 3;
666 int exonLength = MappingUtils.getLength(Arrays.asList(exon));
669 * if exon length matches protein, or is shorter, or longer by the
670 * length of a stop codon (3 bases), then leave it unchanged
672 if (expectedCdsLength >= exonLength
673 || expectedCdsLength == exonLength - 3)
681 origxon = new int[exon.length];
682 System.arraycopy(exon, 0, origxon, 0, exon.length);
684 for (int x = 0; x < exon.length; x += 2)
686 cdspos += Math.abs(exon[x + 1] - exon[x]) + 1;
687 if (expectedCdsLength <= cdspos)
689 // advanced beyond last codon.
691 if (expectedCdsLength != cdspos)
694 // .println("Truncating final exon interval on region by "
695 // + (cdspos - cdslength));
699 * shrink the final exon - reduce end position if forward
700 * strand, increase it if reverse
702 if (exon[x + 1] >= exon[x])
704 endxon = exon[x + 1] - cdspos + expectedCdsLength;
708 endxon = exon[x + 1] + cdspos - expectedCdsLength;
716 // and trim the exon interval set if necessary
717 int[] nxon = new int[sxpos + 2];
718 System.arraycopy(exon, 0, nxon, 0, sxpos + 2);
719 nxon[sxpos + 1] = endxon; // update the end boundary for the new exon
726 public String getSequenceVersion()
728 return sequenceVersion;
731 public void setSequenceVersion(String sequenceVersion)
733 this.sequenceVersion = sequenceVersion;
736 public String getSequenceLength()
738 return sequenceLength;
741 public void setSequenceLength(String sequenceLength)
743 this.sequenceLength = sequenceLength;
746 public String getEntryVersion()
751 public void setEntryVersion(String entryVersion)
753 this.entryVersion = entryVersion;
756 public String getMoleculeType()
761 public void setMoleculeType(String moleculeType)
763 this.moleculeType = moleculeType;
766 public String getTopology()
771 public void setTopology(String topology)
773 this.topology = topology;
776 public String getTaxonomicDivision()
778 return taxonomicDivision;
781 public void setTaxonomicDivision(String taxonomicDivision)
783 this.taxonomicDivision = taxonomicDivision;
786 public String getDescription()
791 public void setDescription(String description)
793 this.description = description;
796 public String getFirstPublicDate()
798 return firstPublicDate;
801 public void setFirstPublicDate(String firstPublicDate)
803 this.firstPublicDate = firstPublicDate;
806 public String getFirstPublicRelease()
808 return firstPublicRelease;
811 public void setFirstPublicRelease(String firstPublicRelease)
813 this.firstPublicRelease = firstPublicRelease;
816 public String getLastUpdatedDate()
818 return lastUpdatedDate;
821 public void setLastUpdatedDate(String lastUpdatedDate)
823 this.lastUpdatedDate = lastUpdatedDate;
826 public String getLastUpdatedRelease()
828 return lastUpdatedRelease;
831 public void setLastUpdatedRelease(String lastUpdatedRelease)
833 this.lastUpdatedRelease = lastUpdatedRelease;
836 public String getDataClass()
841 public void setDataClass(String dataClass)
843 this.dataClass = dataClass;