2 * Jalview - A Sequence Alignment Editor and Viewer (Version 2.6)
3 * Copyright (C) 2010 J Procter, AM Waterhouse, G Barton, M Clamp, S Searle
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 of the License, or (at your option) any later version.
11 * Jalview is distributed in the hope that it will be useful, but
12 * WITHOUT ANY WARRANTY; without even the implied warranty
13 * of MERCHANTABILITY or FITNESS FOR A PARTICULAR
14 * PURPOSE. See the GNU General Public License for more details.
16 * You should have received a copy of the GNU General Public License along with Jalview. If not, see <http://www.gnu.org/licenses/>.
18 package jalview.datamodel.xdb.embl;
20 import jalview.datamodel.DBRefEntry;
21 import jalview.datamodel.DBRefSource;
22 import jalview.datamodel.FeatureProperties;
23 import jalview.datamodel.Mapping;
24 import jalview.datamodel.Sequence;
25 import jalview.datamodel.SequenceFeature;
26 import jalview.datamodel.SequenceI;
28 import java.util.Enumeration;
29 import java.util.Hashtable;
30 import java.util.Iterator;
31 import java.util.Vector;
33 public class EmblEntry
57 EmblSequence sequence;
60 * @return the accession
62 public String getAccession()
69 * the accession to set
71 public void setAccession(String accession)
73 this.accession = accession;
79 public Vector getDbRefs()
88 public void setDbRefs(Vector dbRefs)
96 public String getDesc()
105 public void setDesc(String desc)
111 * @return the features
113 public Vector getFeatures()
120 * the features to set
122 public void setFeatures(Vector features)
124 this.features = features;
128 * @return the keywords
130 public Vector getKeywords()
137 * the keywords to set
139 public void setKeywords(Vector keywords)
141 this.keywords = keywords;
145 * @return the lastUpdated
147 public String getLastUpdated()
154 * the lastUpdated to set
156 public void setLastUpdated(String lastUpdated)
158 this.lastUpdated = lastUpdated;
164 public Vector getRefs()
173 public void setRefs(Vector refs)
179 * @return the releaseCreated
181 public String getRCreated()
187 * @param releaseCreated
188 * the releaseCreated to set
190 public void setRcreated(String releaseCreated)
192 this.rCreated = releaseCreated;
196 * @return the releaseLastUpdated
198 public String getRLastUpdated()
204 * @param releaseLastUpdated
205 * the releaseLastUpdated to set
207 public void setRLastUpdated(String releaseLastUpdated)
209 this.rLastUpdated = releaseLastUpdated;
213 * @return the sequence
215 public EmblSequence getSequence()
222 * the sequence to set
224 public void setSequence(EmblSequence sequence)
226 this.sequence = sequence;
230 * @return the taxDivision
232 public String getTaxDivision()
239 * the taxDivision to set
241 public void setTaxDivision(String taxDivision)
243 this.taxDivision = taxDivision;
247 * @return the version
249 public String getVersion()
258 public void setVersion(String version)
260 this.version = version;
264 * EMBL Feature support is limited. The text below is included for the benefit
265 * of any developer working on improving EMBL feature import in Jalview.
266 * Extract from EMBL feature specification see
267 * http://www.embl-ebi.ac.uk/embl/Documentation
268 * /FT_definitions/feature_table.html 3.5 Location 3.5.1 Purpose
270 * The location indicates the region of the presented sequence which
271 * corresponds to a feature.
273 * 3.5.2 Format and conventions The location contains at least one sequence
274 * location descriptor and may contain one or more operators with one or more
275 * sequence location descriptors. Base numbers refer to the numbering in the
276 * entry. This numbering designates the first base (5' end) of the presented
277 * sequence as base 1. Base locations beyond the range of the presented
278 * sequence may not be used in location descriptors, the only exception being
279 * location in a remote entry (see 3.5.2.1, e).
281 * Location operators and descriptors are discussed in more detail below.
283 * 3.5.2.1 Location descriptors
285 * The location descriptor can be one of the following: (a) a single base
286 * number (b) a site between two indicated adjoining bases (c) a single base
287 * chosen from within a specified range of bases (not allowed for new entries)
288 * (d) the base numbers delimiting a sequence span (e) a remote entry
289 * identifier followed by a local location descriptor (i.e., a-d)
291 * A site between two adjoining nucleotides, such as endonucleolytic cleavage
292 * site, is indicated by listing the two points separated by a carat (^). The
293 * permitted formats for this descriptor are n^n+1 (for example 55^56), or,
294 * for circular molecules, n^1, where "n" is the full length of the molecule,
295 * ie 1000^1 for circular molecule with length 1000.
297 * A single base chosen from a range of bases is indicated by the first base
298 * number and the last base number of the range separated by a single period
299 * (e.g., '12.21' indicates a single base taken from between the indicated
300 * points). From October 2006 the usage of this descriptor is restricted : it
301 * is illegal to use "a single base from a range" (c) either on its own or in
302 * combination with the "sequence span" (d) descriptor for newly created
303 * entries. The existing entries where such descriptors exist are going to be
306 * Sequence spans are indicated by the starting base number and the ending
307 * base number separated by two periods (e.g., '34..456'). The '<' and '>'
308 * symbols may be used with the starting and ending base numbers to indicate
309 * that an end point is beyond the specified base number. The starting and
310 * ending base positions can be represented as distinct base numbers
311 * ('34..456') or a site between two indicated adjoining bases.
313 * A location in a remote entry (not the entry to which the feature table
314 * belongs) can be specified by giving the accession-number and sequence
315 * version of the remote entry, followed by a colon ":", followed by a
316 * location descriptor which applies to that entry's sequence (i.e.
317 * J12345.1:1..15, see also examples below)
321 * The location operator is a prefix that specifies what must be done to the
322 * indicated sequence to find or construct the location corresponding to the
323 * feature. A list of operators is given below with their definitions and most
326 * complement(location) Find the complement of the presented sequence in the
327 * span specified by " location" (i.e., read the complement of the presented
328 * strand in its 5'-to-3' direction)
330 * join(location,location, ... location) The indicated elements should be
331 * joined (placed end-to-end) to form one contiguous sequence
333 * order(location,location, ... location) The elements can be found in the
334 * specified order (5' to 3' direction), but nothing is implied about the
335 * reasonableness about joining them
337 * Note : location operator "complement" can be used in combination with
338 * either " join" or "order" within the same location; combinations of "join"
339 * and "order" within the same location (nested operators) are illegal.
343 * 3.5.3 Location examples
345 * The following is a list of common location descriptors with their meanings:
347 * Location Description
349 * 467 Points to a single base in the presented sequence
351 * 340..565 Points to a continuous range of bases bounded by and including the
352 * starting and ending bases
354 * <345..500 Indicates that the exact lower boundary point of a feature is
355 * unknown. The location begins at some base previous to the first base
356 * specified (which need not be contained in the presented sequence) and
357 * continues to and includes the ending base
359 * <1..888 The feature starts before the first sequenced base and continues to
360 * and includes base 888
362 * 1..>888 The feature starts at the first sequenced base and continues beyond
365 * 102.110 Indicates that the exact location is unknown but that it is one of
366 * the bases between bases 102 and 110, inclusive
368 * 123^124 Points to a site between bases 123 and 124
370 * join(12..78,134..202) Regions 12 to 78 and 134 to 202 should be joined to
371 * form one contiguous sequence
374 * complement(34..126) Start at the base complementary to 126 and finish at
375 * the base complementary to base 34 (the feature is on the strand
376 * complementary to the presented strand)
379 * complement(join(2691..4571,4918..5163)) Joins regions 2691 to 4571 and 4918
380 * to 5163, then complements the joined segments (the feature is on the strand
381 * complementary to the presented strand)
383 * join(complement(4918..5163),complement(2691..4571)) Complements regions
384 * 4918 to 5163 and 2691 to 4571, then joins the complemented segments (the
385 * feature is on the strand complementary to the presented strand)
387 * J00194.1:100..202 Points to bases 100 to 202, inclusive, in the entry (in
388 * this database) with primary accession number 'J00194'
390 * join(1..100,J00194.1:100..202) Joins region 1..100 of the existing entry
391 * with the region 100..202 of remote entry J00194
394 * Recover annotated sequences from EMBL file
397 * don't return nucleic acid sequences
401 * don't return any translated protein sequences marked in features
402 * @return dataset sequences with DBRefs and features - DNA always comes first
404 public jalview.datamodel.SequenceI[] getSequences(boolean noNa,
405 boolean noPeptide, String sourceDb)
406 { // TODO: ensure emblEntry.getSequences behaves correctly for returning all
407 // cases of noNa and noPeptide
408 Vector seqs = new Vector();
412 // In theory we still need to create this if noNa is set to avoid a null
414 dna = new Sequence(sourceDb + "|" + accession, sequence.getSequence());
415 dna.setDescription(desc);
416 DBRefEntry retrievedref = new DBRefEntry(sourceDb, version, accession);
417 dna.addDBRef(retrievedref);
418 // add map to indicate the sequence is a valid coordinate frame for the
420 retrievedref.setMap(new Mapping(null, new int[]
421 { 1, dna.getLength() }, new int[]
422 { 1, dna.getLength() }, 1, 1));
423 // TODO: transform EMBL Database refs to canonical form
425 for (Iterator i = dbRefs.iterator(); i.hasNext(); dna
426 .addDBRef((DBRefEntry) i.next()))
431 for (Iterator i = features.iterator(); i.hasNext();)
433 EmblFeature feature = (EmblFeature) i.next();
436 if (feature.dbRefs != null && feature.dbRefs.size() > 0)
438 for (Iterator dbr = feature.dbRefs.iterator(); dbr.hasNext(); dna
439 .addDBRef((DBRefEntry) dbr.next()))
443 if (FeatureProperties.isCodingFeature(sourceDb, feature.getName()))
445 parseCodingFeature(feature, sourceDb, seqs, dna, noPeptide);
449 // General feature type.
452 if (feature.dbRefs != null && feature.dbRefs.size() > 0)
454 for (Iterator dbr = feature.dbRefs.iterator(); dbr.hasNext(); dna
455 .addDBRef((DBRefEntry) dbr.next()))
461 } catch (Exception e)
463 System.err.println("EMBL Record Features parsing error!");
465 .println("Please report the following to help@jalview.org :");
466 System.err.println("EMBL Record " + accession);
467 System.err.println("Resulted in exception: " + e.getMessage());
468 e.printStackTrace(System.err);
470 if (!noNa && dna != null)
474 SequenceI[] sqs = new SequenceI[seqs.size()];
475 for (int i = 0, j = seqs.size(); i < j; i++)
477 sqs[i] = (SequenceI) seqs.elementAt(i);
484 * attempt to extract coding region and product from a feature and properly
485 * decorate it with annotations.
490 * source database for the EMBLXML
492 * place where sequences go
494 * parent dna sequence for this record
496 * flag for generation of Peptide sequence objects
498 private void parseCodingFeature(EmblFeature feature, String sourceDb,
499 Vector seqs, Sequence dna, boolean noPeptide)
501 boolean isEmblCdna = sourceDb.equals(DBRefSource.EMBLCDS);
502 // extract coding region(s)
503 jalview.datamodel.Mapping map = null;
505 if (feature.locations != null && feature.locations.size() > 0)
507 for (Enumeration locs = feature.locations.elements(); locs
510 EmblFeatureLocations loc = (EmblFeatureLocations) locs
512 int[] se = loc.getElementRanges(accession);
519 int[] t = new int[exon.length + se.length];
520 System.arraycopy(exon, 0, t, 0, exon.length);
521 System.arraycopy(se, 0, t, exon.length, se.length);
527 String prname = new String();
529 Hashtable vals = new Hashtable();
532 if (feature.getQualifiers() != null
533 && feature.getQualifiers().size() > 0)
535 for (Iterator quals = feature.getQualifiers().iterator(); quals
538 Qualifier q = (Qualifier) quals.next();
539 if (q.getName().equals("translation"))
541 StringBuffer prsq = new StringBuffer(q.getValues()[0]);
542 int p = prsq.indexOf(" ");
545 prsq.deleteCharAt(p);
546 p = prsq.indexOf(" ", p);
548 prseq = prsq.toString();
552 else if (q.getName().equals("protein_id"))
554 prid = q.getValues()[0];
556 else if (q.getName().equals("codon_start"))
558 prstart = Integer.parseInt(q.getValues()[0]);
560 else if (q.getName().equals("product"))
562 prname = q.getValues()[0];
566 // throw anything else into the additional properties hash
567 String[] s = q.getValues();
568 StringBuffer sb = new StringBuffer();
571 for (int i = 0; i < s.length; i++)
577 vals.put(q.getName(), sb.toString());
581 Sequence product = null;
582 exon = adjustForPrStart(prstart, exon);
584 if (prseq != null && prname != null && prid != null)
587 product = new Sequence(prid, prseq, 1, prseq.length());
588 product.setDescription(((prname.length() == 0) ? "Protein Product from "
593 // Protein is also added to vector of sequences returned
596 // we have everything - create the mapping and perhaps the protein
598 if (exon == null || exon.length == 0)
601 .println("Implementation Notice: EMBLCDS records not properly supported yet - Making up the CDNA region of this sequence... may be incorrect ("
602 + sourceDb + ":" + getAccession() + ")");
603 if (prseq.length() * 3 == (1 - prstart + dna.getSequence().length))
606 .println("Not allowing for additional stop codon at end of cDNA fragment... !");
607 // this might occur for CDS sequences where no features are
610 { dna.getStart() + (prstart - 1), dna.getEnd() };
611 map = new jalview.datamodel.Mapping(product, exon, new int[]
612 { 1, prseq.length() }, 3, 1);
614 if ((prseq.length() + 1) * 3 == (1 - prstart + dna.getSequence().length))
617 .println("Allowing for additional stop codon at end of cDNA fragment... will probably cause an error in VAMSAs!");
619 { dna.getStart() + (prstart - 1), dna.getEnd() - 3 };
620 map = new jalview.datamodel.Mapping(product, exon, new int[]
621 { 1, prseq.length() }, 3, 1);
626 // Trim the exon mapping if necessary - the given product may only be a
627 // fragment of a larger protein. (EMBL:AY043181 is an example)
631 // TODO: Add a DbRef back to the parent EMBL sequence with the exon
633 // if given a dataset reference, search dataset for parent EMBL
634 // sequence if it exists and set its map
635 // make a new feature annotating the coding contig
639 // final product length trunctation check
641 map = new jalview.datamodel.Mapping(product,
642 adjustForProteinLength(prseq.length(), exon), new int[]
643 { 1, prseq.length() }, 3, 1);
644 // reconstruct the EMBLCDS entry
645 // TODO: this is only necessary when there codon annotation is
646 // complete (I think JBPNote)
647 DBRefEntry pcdnaref = new DBRefEntry();
648 pcdnaref.setAccessionId(prid);
649 pcdnaref.setSource(DBRefSource.EMBLCDS);
650 pcdnaref.setVersion(getVersion()); // same as parent EMBL version.
651 jalview.util.MapList mp = new jalview.util.MapList(new int[]
652 { 1, prseq.length() }, new int[]
653 { 1 + (prstart - 1), (prstart - 1) + 3 * prseq.length() }, 1, 3);
654 // { 1 + (prstart - 1) * 3,
655 // 1 + (prstart - 1) * 3 + prseq.length() * 3 - 1 }, new int[]
656 // { 1prstart, prstart + prseq.length() - 1 }, 3, 1);
657 pcdnaref.setMap(new Mapping(mp));
659 product.addDBRef(pcdnaref);
663 // add cds feature to dna seq - this may include the stop codon
664 for (int xint = 0; exon != null && xint < exon.length; xint += 2)
666 SequenceFeature sf = new SequenceFeature();
667 sf.setBegin(exon[xint]);
668 sf.setEnd(exon[xint + 1]);
669 sf.setType(feature.getName());
670 sf.setFeatureGroup(sourceDb);
671 sf.setDescription("Exon " + (1 + (int) (xint / 2))
672 + " for protein '" + prname + "' EMBLCDS:" + prid);
673 sf.setValue(FeatureProperties.EXONPOS, new Integer(1 + xint));
674 sf.setValue(FeatureProperties.EXONPRODUCT, prname);
675 if (vals != null && vals.size() > 0)
677 Enumeration kv = vals.elements();
678 while (kv.hasMoreElements())
680 Object key = kv.nextElement();
682 sf.setValue(key.toString(), vals.get(key));
685 dna.addSequenceFeature(sf);
688 // add dbRefs to sequence
689 if (feature.dbRefs != null && feature.dbRefs.size() > 0)
691 for (Iterator dbr = feature.dbRefs.iterator(); dbr.hasNext();)
693 DBRefEntry ref = (DBRefEntry) dbr.next();
694 ref.setSource(jalview.util.DBRefUtils.getCanonicalName(ref
696 // Hard code the kind of protein product accessions that EMBL cite
697 if (ref.getSource().equals(jalview.datamodel.DBRefSource.UNIPROT))
700 if (map != null && map.getTo() != null)
702 map.getTo().addDBRef(
703 new DBRefEntry(ref.getSource(), ref.getVersion(), ref
704 .getAccessionId())); // don't copy map over.
705 if (map.getTo().getName().indexOf(prid) == 0)
708 jalview.datamodel.DBRefSource.UNIPROT + "|"
709 + ref.getAccessionId());
715 DBRefEntry pref = new DBRefEntry(ref.getSource(),
716 ref.getVersion(), ref.getAccessionId());
717 pref.setMap(null); // reference is direct
718 product.addDBRef(pref);
719 // Add converse mapping reference
722 Mapping pmap = new Mapping(dna, map.getMap().getInverse());
723 pref = new DBRefEntry(sourceDb, getVersion(),
724 this.getAccession());
726 if (map.getTo() != null)
728 map.getTo().addDBRef(pref);
737 private int[] adjustForPrStart(int prstart, int[] exon)
740 int origxon[], sxpos = -1;
741 int sxstart, sxstop; // unnecessary variables used for debugging
742 // first adjust range for codon start attribute
745 origxon = new int[exon.length];
746 System.arraycopy(exon, 0, origxon, 0, exon.length);
748 for (int x = 0; x < exon.length && sxpos == -1; x += 2)
750 cdspos += exon[x + 1] - exon[x] + 1;
751 if (prstart <= cdspos)
755 sxstop = exon[x + 1];
756 // and adjust start boundary of first exon.
757 exon[x] = exon[x + 1] - cdspos + prstart;
764 int[] nxon = new int[exon.length - sxpos];
765 System.arraycopy(exon, sxpos, nxon, 0, exon.length - sxpos);
773 * truncate the last exon interval to the prlength'th codon
779 private int[] adjustForProteinLength(int prlength, int[] exon)
782 int origxon[], sxpos = -1, endxon = 0, cdslength = prlength * 3;
783 int sxstart, sxstop; // unnecessary variables used for debugging
784 // first adjust range for codon start attribute
785 if (prlength >= 1 && exon != null)
787 origxon = new int[exon.length];
788 System.arraycopy(exon, 0, origxon, 0, exon.length);
790 for (int x = 0; x < exon.length && sxpos == -1; x += 2)
792 cdspos += exon[x + 1] - exon[x] + 1;
793 if (cdslength <= cdspos)
795 // advanced beyond last codon.
798 sxstop = exon[x + 1];
799 if (cdslength != cdspos)
802 .println("Truncating final exon interval on region by "
803 + (cdspos - cdslength));
805 // locate the new end boundary of final exon as endxon
806 endxon = exon[x + 1] - cdspos + cdslength;
813 // and trim the exon interval set if necessary
814 int[] nxon = new int[sxpos + 2];
815 System.arraycopy(exon, 0, nxon, 0, sxpos + 2);
816 nxon[sxpos + 1] = endxon; // update the end boundary for the new exon