2 * Jalview - A Sequence Alignment Editor and Viewer (Version 2.9.0b2)
3 * Copyright (C) 2015 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.datamodel.DBRefEntry;
24 import jalview.datamodel.DBRefSource;
25 import jalview.datamodel.FeatureProperties;
26 import jalview.datamodel.Mapping;
27 import jalview.datamodel.Sequence;
28 import jalview.datamodel.SequenceFeature;
29 import jalview.datamodel.SequenceI;
31 import java.util.Hashtable;
32 import java.util.Map.Entry;
33 import java.util.Vector;
36 * Data model for one entry returned from an EMBL query, as marshalled by a
39 * For example: http://www.ebi.ac.uk/Tools/dbfetch/dbfetch/embl/x53828/emblxml
41 * @see embl_mapping.xml
43 public class EmblEntry
59 Vector<String> keywords;
61 Vector<DBRefEntry> dbRefs;
63 Vector<EmblFeature> features;
65 EmblSequence sequence;
68 * @return the accession
70 public String getAccession()
77 * the accession to set
79 public void setAccession(String accession)
81 this.accession = accession;
87 public Vector<DBRefEntry> getDbRefs()
96 public void setDbRefs(Vector<DBRefEntry> dbRefs)
104 public String getDesc()
113 public void setDesc(String desc)
119 * @return the features
121 public Vector<EmblFeature> getFeatures()
128 * the features to set
130 public void setFeatures(Vector<EmblFeature> features)
132 this.features = features;
136 * @return the keywords
138 public Vector<String> getKeywords()
145 * the keywords to set
147 public void setKeywords(Vector<String> keywords)
149 this.keywords = keywords;
153 * @return the lastUpdated
155 public String getLastUpdated()
162 * the lastUpdated to set
164 public void setLastUpdated(String lastUpdated)
166 this.lastUpdated = lastUpdated;
170 * @return the releaseCreated
172 public String getRCreated()
178 * @param releaseCreated
179 * the releaseCreated to set
181 public void setRCreated(String releaseCreated)
183 this.rCreated = releaseCreated;
187 * @return the releaseLastUpdated
189 public String getRLastUpdated()
195 * @param releaseLastUpdated
196 * the releaseLastUpdated to set
198 public void setRLastUpdated(String releaseLastUpdated)
200 this.rLastUpdated = releaseLastUpdated;
204 * @return the sequence
206 public EmblSequence getSequence()
213 * the sequence to set
215 public void setSequence(EmblSequence sequence)
217 this.sequence = sequence;
221 * @return the taxDivision
223 public String getTaxDivision()
230 * the taxDivision to set
232 public void setTaxDivision(String taxDivision)
234 this.taxDivision = taxDivision;
238 * @return the version
240 public String getVersion()
249 public void setVersion(String version)
251 this.version = version;
255 * EMBL Feature support is limited. The text below is included for the benefit
256 * of any developer working on improving EMBL feature import in Jalview.
257 * Extract from EMBL feature specification see
258 * http://www.embl-ebi.ac.uk/embl/Documentation
259 * /FT_definitions/feature_table.html 3.5 Location 3.5.1 Purpose
261 * The location indicates the region of the presented sequence which
262 * corresponds to a feature.
264 * 3.5.2 Format and conventions The location contains at least one sequence
265 * location descriptor and may contain one or more operators with one or more
266 * sequence location descriptors. Base numbers refer to the numbering in the
267 * entry. This numbering designates the first base (5' end) of the presented
268 * sequence as base 1. Base locations beyond the range of the presented
269 * sequence may not be used in location descriptors, the only exception being
270 * location in a remote entry (see 3.5.2.1, e).
272 * Location operators and descriptors are discussed in more detail below.
274 * 3.5.2.1 Location descriptors
276 * The location descriptor can be one of the following: (a) a single base
277 * number (b) a site between two indicated adjoining bases (c) a single base
278 * chosen from within a specified range of bases (not allowed for new entries)
279 * (d) the base numbers delimiting a sequence span (e) a remote entry
280 * identifier followed by a local location descriptor (i.e., a-d)
282 * A site between two adjoining nucleotides, such as endonucleolytic cleavage
283 * site, is indicated by listing the two points separated by a carat (^). The
284 * permitted formats for this descriptor are n^n+1 (for example 55^56), or,
285 * for circular molecules, n^1, where "n" is the full length of the molecule,
286 * ie 1000^1 for circular molecule with length 1000.
288 * A single base chosen from a range of bases is indicated by the first base
289 * number and the last base number of the range separated by a single period
290 * (e.g., '12.21' indicates a single base taken from between the indicated
291 * points). From October 2006 the usage of this descriptor is restricted : it
292 * is illegal to use "a single base from a range" (c) either on its own or in
293 * combination with the "sequence span" (d) descriptor for newly created
294 * entries. The existing entries where such descriptors exist are going to be
297 * Sequence spans are indicated by the starting base number and the ending
298 * base number separated by two periods (e.g., '34..456'). The '<' and '>'
299 * symbols may be used with the starting and ending base numbers to indicate
300 * that an end point is beyond the specified base number. The starting and
301 * ending base positions can be represented as distinct base numbers
302 * ('34..456') or a site between two indicated adjoining bases.
304 * A location in a remote entry (not the entry to which the feature table
305 * belongs) can be specified by giving the accession-number and sequence
306 * version of the remote entry, followed by a colon ":", followed by a
307 * location descriptor which applies to that entry's sequence (i.e.
308 * J12345.1:1..15, see also examples below)
312 * The location operator is a prefix that specifies what must be done to the
313 * indicated sequence to find or construct the location corresponding to the
314 * feature. A list of operators is given below with their definitions and most
317 * complement(location) Find the complement of the presented sequence in the
318 * span specified by " location" (i.e., read the complement of the presented
319 * strand in its 5'-to-3' direction)
321 * join(location,location, ... location) The indicated elements should be
322 * joined (placed end-to-end) to form one contiguous sequence
324 * order(location,location, ... location) The elements can be found in the
325 * specified order (5' to 3' direction), but nothing is implied about the
326 * reasonableness about joining them
328 * Note : location operator "complement" can be used in combination with
329 * either " join" or "order" within the same location; combinations of "join"
330 * and "order" within the same location (nested operators) are illegal.
334 * 3.5.3 Location examples
336 * The following is a list of common location descriptors with their meanings:
338 * Location Description
340 * 467 Points to a single base in the presented sequence
342 * 340..565 Points to a continuous range of bases bounded by and including the
343 * starting and ending bases
345 * <345..500 Indicates that the exact lower boundary point of a feature is
346 * unknown. The location begins at some base previous to the first base
347 * specified (which need not be contained in the presented sequence) and
348 * continues to and includes the ending base
350 * <1..888 The feature starts before the first sequenced base and continues to
351 * and includes base 888
353 * 1..>888 The feature starts at the first sequenced base and continues beyond
356 * 102.110 Indicates that the exact location is unknown but that it is one of
357 * the bases between bases 102 and 110, inclusive
359 * 123^124 Points to a site between bases 123 and 124
361 * join(12..78,134..202) Regions 12 to 78 and 134 to 202 should be joined to
362 * form one contiguous sequence
365 * complement(34..126) Start at the base complementary to 126 and finish at
366 * the base complementary to base 34 (the feature is on the strand
367 * complementary to the presented strand)
370 * complement(join(2691..4571,4918..5163)) Joins regions 2691 to 4571 and 4918
371 * to 5163, then complements the joined segments (the feature is on the strand
372 * complementary to the presented strand)
374 * join(complement(4918..5163),complement(2691..4571)) Complements regions
375 * 4918 to 5163 and 2691 to 4571, then joins the complemented segments (the
376 * feature is on the strand complementary to the presented strand)
378 * J00194.1:100..202 Points to bases 100 to 202, inclusive, in the entry (in
379 * this database) with primary accession number 'J00194'
381 * join(1..100,J00194.1:100..202) Joins region 1..100 of the existing entry
382 * with the region 100..202 of remote entry J00194
385 * Recover annotated sequences from EMBL file
388 * don't return nucleic acid sequences
392 * don't return any translated protein sequences marked in features
393 * @return dataset sequences with DBRefs and features - DNA always comes first
395 public jalview.datamodel.SequenceI[] getSequences(boolean noNa,
396 boolean noPeptide, String sourceDb)
397 { // TODO: ensure emblEntry.getSequences behaves correctly for returning all
398 // cases of noNa and noPeptide
399 Vector<SequenceI> seqs = new Vector<SequenceI>();
403 // In theory we still need to create this if noNa is set to avoid a null
405 dna = new Sequence(sourceDb + "|" + accession, sequence.getSequence());
406 dna.setDescription(desc);
407 DBRefEntry retrievedref = new DBRefEntry(sourceDb, version, accession);
408 dna.addDBRef(retrievedref);
409 // add map to indicate the sequence is a valid coordinate frame for the
411 retrievedref.setMap(new Mapping(null,
412 new int[] { 1, dna.getLength() }, new int[] { 1,
413 dna.getLength() }, 1, 1));
414 // TODO: transform EMBL Database refs to canonical form
417 for (DBRefEntry dbref : dbRefs)
425 for (EmblFeature feature : features)
429 if (feature.dbRefs != null)
431 for (DBRefEntry dbref : feature.dbRefs)
437 if (FeatureProperties.isCodingFeature(sourceDb, feature.getName()))
439 parseCodingFeature(feature, sourceDb, seqs, dna, noPeptide);
443 // General feature type.
444 // TODO this is just duplicated code ??
447 if (feature.dbRefs != null)
449 for (DBRefEntry dbref : feature.dbRefs)
457 } catch (Exception e)
459 System.err.println("EMBL Record Features parsing error!");
461 .println("Please report the following to help@jalview.org :");
462 System.err.println("EMBL Record " + accession);
463 System.err.println("Resulted in exception: " + e.getMessage());
464 e.printStackTrace(System.err);
466 if (!noNa && dna != null)
470 SequenceI[] sqs = new SequenceI[seqs.size()];
471 for (int i = 0, j = seqs.size(); i < j; i++)
473 sqs[i] = seqs.elementAt(i);
480 * attempt to extract coding region and product from a feature and properly
481 * decorate it with annotations.
486 * source database for the EMBLXML
488 * place where sequences go
490 * parent dna sequence for this record
492 * flag for generation of Peptide sequence objects
494 private void parseCodingFeature(EmblFeature feature, String sourceDb,
495 Vector<SequenceI> seqs, Sequence dna, boolean noPeptide)
497 boolean isEmblCdna = sourceDb.equals(DBRefSource.EMBLCDS);
498 // extract coding region(s)
499 jalview.datamodel.Mapping map = null;
501 if (feature.locations != null)
503 for (EmblFeatureLocations loc : feature.locations)
505 int[] se = loc.getElementRanges(accession);
512 int[] t = new int[exon.length + se.length];
513 System.arraycopy(exon, 0, t, 0, exon.length);
514 System.arraycopy(se, 0, t, exon.length, se.length);
520 String prname = new String();
522 Hashtable<String, String> vals = new Hashtable<String, String>();
525 if (feature.getQualifiers() != null)
527 for (Qualifier q : feature.getQualifiers())
529 String qname = q.getName();
530 if (qname.equals("translation"))
532 StringBuilder prsq = new StringBuilder(q.getValues()[0]);
533 int p = prsq.indexOf(" ");
536 prsq.deleteCharAt(p);
537 p = prsq.indexOf(" ", p);
539 prseq = prsq.toString();
543 else if (qname.equals("protein_id"))
545 prid = q.getValues()[0];
547 else if (qname.equals("codon_start"))
549 prstart = Integer.parseInt(q.getValues()[0]);
551 else if (qname.equals("product"))
553 prname = q.getValues()[0];
557 // throw anything else into the additional properties hash
558 String[] s = q.getValues();
559 StringBuilder sb = new StringBuilder();
562 for (int i = 0; i < s.length; i++)
568 vals.put(qname, sb.toString());
572 Sequence product = null;
573 DBRefEntry protEMBLCDS = null;
574 exon = adjustForPrStart(prstart, exon);
575 boolean noProteinDbref = true;
577 if (prseq != null && prname != null && prid != null)
580 product = new Sequence(prid, prseq, 1, prseq.length());
581 product.setDescription(((prname.length() == 0) ? "Protein Product from "
586 // Protein is also added to vector of sequences returned
589 // we have everything - create the mapping and perhaps the protein
591 if (exon == null || exon.length == 0)
594 .println("Implementation Notice: EMBLCDS records not properly supported yet - Making up the CDNA region of this sequence... may be incorrect ("
595 + sourceDb + ":" + getAccession() + ")");
596 if (prseq.length() * 3 == (1 - prstart + dna.getSequence().length))
599 .println("Not allowing for additional stop codon at end of cDNA fragment... !");
600 // this might occur for CDS sequences where no features are
602 exon = new int[] { dna.getStart() + (prstart - 1), dna.getEnd() };
603 map = new jalview.datamodel.Mapping(product, exon, new int[] { 1,
604 prseq.length() }, 3, 1);
606 if ((prseq.length() + 1) * 3 == (1 - prstart + dna.getSequence().length))
609 .println("Allowing for additional stop codon at end of cDNA fragment... will probably cause an error in VAMSAs!");
610 exon = new int[] { dna.getStart() + (prstart - 1),
612 map = new jalview.datamodel.Mapping(product, exon, new int[] { 1,
613 prseq.length() }, 3, 1);
618 // Trim the exon mapping if necessary - the given product may only be a
619 // fragment of a larger protein. (EMBL:AY043181 is an example)
623 // TODO: Add a DbRef back to the parent EMBL sequence with the exon
625 // if given a dataset reference, search dataset for parent EMBL
626 // sequence if it exists and set its map
627 // make a new feature annotating the coding contig
631 // final product length trunctation check
633 map = new jalview.datamodel.Mapping(product,
634 adjustForProteinLength(prseq.length(), exon), new int[] {
635 1, prseq.length() }, 3, 1);
636 // reconstruct the EMBLCDS entry
637 // TODO: this is only necessary when there codon annotation is
638 // complete (I think JBPNote)
639 DBRefEntry pcdnaref = new DBRefEntry();
640 pcdnaref.setAccessionId(prid);
641 pcdnaref.setSource(DBRefSource.EMBLCDS);
642 pcdnaref.setVersion(getVersion()); // same as parent EMBL version.
643 jalview.util.MapList mp = new jalview.util.MapList(new int[] { 1,
644 prseq.length() }, new int[] { 1 + (prstart - 1),
645 (prstart - 1) + 3 * prseq.length() }, 1, 3);
646 // { 1 + (prstart - 1) * 3,
647 // 1 + (prstart - 1) * 3 + prseq.length() * 3 - 1 }, new int[]
648 // { 1prstart, prstart + prseq.length() - 1 }, 3, 1);
649 pcdnaref.setMap(new Mapping(mp));
652 product.addDBRef(pcdnaref);
653 protEMBLCDS = new DBRefEntry(pcdnaref);
654 protEMBLCDS.setSource(DBRefSource.EMBLCDSProduct);
655 product.addDBRef(protEMBLCDS);
661 // add cds feature to dna seq - this may include the stop codon
662 for (int xint = 0; exon != null && xint < exon.length; xint += 2)
664 SequenceFeature sf = new SequenceFeature();
665 sf.setBegin(exon[xint]);
666 sf.setEnd(exon[xint + 1]);
667 sf.setType(feature.getName());
668 sf.setFeatureGroup(sourceDb);
669 sf.setDescription("Exon " + (1 + xint / 2) + " for protein '"
670 + prname + "' EMBLCDS:" + prid);
671 sf.setValue(FeatureProperties.EXONPOS, new Integer(1 + xint));
672 sf.setValue(FeatureProperties.EXONPRODUCT, prname);
675 for (Entry<String, String> val : vals.entrySet())
677 sf.setValue(val.getKey(), val.getValue());
680 dna.addSequenceFeature(sf);
683 // add dbRefs to sequence
684 if (feature.dbRefs != null)
686 for (DBRefEntry ref : feature.dbRefs)
688 ref.setSource(jalview.util.DBRefUtils.getCanonicalName(ref
690 // Hard code the kind of protein product accessions that EMBL cite
691 if (ref.getSource().equals(jalview.datamodel.DBRefSource.UNIPROT))
694 if (map != null && map.getTo() != null)
696 map.getTo().addDBRef(
697 new DBRefEntry(ref.getSource(), ref.getVersion(), ref
698 .getAccessionId())); // don't copy map over.
699 if (map.getTo().getName().indexOf(prid) == 0)
702 jalview.datamodel.DBRefSource.UNIPROT + "|"
703 + ref.getAccessionId());
706 noProteinDbref = false;
710 DBRefEntry pref = new DBRefEntry(ref.getSource(),
711 ref.getVersion(), ref.getAccessionId());
712 pref.setMap(null); // reference is direct
713 product.addDBRef(pref);
714 // Add converse mapping reference
717 Mapping pmap = new Mapping(dna, map.getMap().getInverse());
718 pref = new DBRefEntry(sourceDb, getVersion(),
719 this.getAccession());
721 if (map.getTo() != null)
723 map.getTo().addDBRef(pref);
729 if (noProteinDbref && product != null)
731 // add protein coding reference to dna sequence so xref matches
732 if (protEMBLCDS == null)
734 protEMBLCDS = new DBRefEntry();
735 protEMBLCDS.setAccessionId(prid);
736 protEMBLCDS.setSource(DBRefSource.EMBLCDSProduct);
737 protEMBLCDS.setVersion(getVersion());
739 .setMap(new Mapping(product, map.getMap().getInverse()));
741 product.addDBRef(protEMBLCDS);
743 // Add converse mapping reference
746 Mapping pmap = new Mapping(product, protEMBLCDS.getMap().getMap()
748 DBRefEntry ncMap = new DBRefEntry(protEMBLCDS);
750 if (map.getTo() != null)
759 private int[] adjustForPrStart(int prstart, int[] exon)
762 int origxon[], sxpos = -1;
763 int sxstart, sxstop; // unnecessary variables used for debugging
764 // first adjust range for codon start attribute
767 origxon = new int[exon.length];
768 System.arraycopy(exon, 0, origxon, 0, exon.length);
770 for (int x = 0; x < exon.length && sxpos == -1; x += 2)
772 cdspos += exon[x + 1] - exon[x] + 1;
773 if (prstart <= cdspos)
777 sxstop = exon[x + 1];
778 // and adjust start boundary of first exon.
779 exon[x] = exon[x + 1] - cdspos + prstart;
786 int[] nxon = new int[exon.length - sxpos];
787 System.arraycopy(exon, sxpos, nxon, 0, exon.length - sxpos);
795 * truncate the last exon interval to the prlength'th codon
801 private int[] adjustForProteinLength(int prlength, int[] exon)
804 int origxon[], sxpos = -1, endxon = 0, cdslength = prlength * 3;
805 int sxstart, sxstop; // unnecessary variables used for debugging
806 // first adjust range for codon start attribute
807 if (prlength >= 1 && exon != null)
809 origxon = new int[exon.length];
810 System.arraycopy(exon, 0, origxon, 0, exon.length);
812 for (int x = 0; x < exon.length && sxpos == -1; x += 2)
814 cdspos += exon[x + 1] - exon[x] + 1;
815 if (cdslength <= cdspos)
817 // advanced beyond last codon.
820 sxstop = exon[x + 1];
821 if (cdslength != cdspos)
824 .println("Truncating final exon interval on region by "
825 + (cdspos - cdslength));
827 // locate the new end boundary of final exon as endxon
828 endxon = exon[x + 1] - cdspos + cdslength;
835 // and trim the exon interval set if necessary
836 int[] nxon = new int[sxpos + 2];
837 System.arraycopy(exon, 0, nxon, 0, sxpos + 2);
838 nxon[sxpos + 1] = endxon; // update the end boundary for the new exon