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.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.Enumeration;
32 import java.util.Hashtable;
33 import java.util.Iterator;
34 import java.util.Vector;
36 public class EmblEntry
60 EmblSequence sequence;
63 * @return the accession
65 public String getAccession()
72 * the accession to set
74 public void setAccession(String accession)
76 this.accession = accession;
82 public Vector getDbRefs()
91 public void setDbRefs(Vector dbRefs)
99 public String getDesc()
108 public void setDesc(String desc)
114 * @return the features
116 public Vector getFeatures()
123 * the features to set
125 public void setFeatures(Vector features)
127 this.features = features;
131 * @return the keywords
133 public Vector getKeywords()
140 * the keywords to set
142 public void setKeywords(Vector keywords)
144 this.keywords = keywords;
148 * @return the lastUpdated
150 public String getLastUpdated()
157 * the lastUpdated to set
159 public void setLastUpdated(String lastUpdated)
161 this.lastUpdated = lastUpdated;
167 public Vector getRefs()
176 public void setRefs(Vector refs)
182 * @return the releaseCreated
184 public String getRCreated()
190 * @param releaseCreated
191 * the releaseCreated to set
193 public void setRcreated(String releaseCreated)
195 this.rCreated = releaseCreated;
199 * @return the releaseLastUpdated
201 public String getRLastUpdated()
207 * @param releaseLastUpdated
208 * the releaseLastUpdated to set
210 public void setRLastUpdated(String releaseLastUpdated)
212 this.rLastUpdated = releaseLastUpdated;
216 * @return the sequence
218 public EmblSequence getSequence()
225 * the sequence to set
227 public void setSequence(EmblSequence sequence)
229 this.sequence = sequence;
233 * @return the taxDivision
235 public String getTaxDivision()
242 * the taxDivision to set
244 public void setTaxDivision(String taxDivision)
246 this.taxDivision = taxDivision;
250 * @return the version
252 public String getVersion()
261 public void setVersion(String version)
263 this.version = version;
267 * EMBL Feature support is limited. The text below is included for the benefit
268 * of any developer working on improving EMBL feature import in Jalview.
269 * Extract from EMBL feature specification see
270 * http://www.embl-ebi.ac.uk/embl/Documentation
271 * /FT_definitions/feature_table.html 3.5 Location 3.5.1 Purpose
273 * The location indicates the region of the presented sequence which
274 * corresponds to a feature.
276 * 3.5.2 Format and conventions The location contains at least one sequence
277 * location descriptor and may contain one or more operators with one or more
278 * sequence location descriptors. Base numbers refer to the numbering in the
279 * entry. This numbering designates the first base (5' end) of the presented
280 * sequence as base 1. Base locations beyond the range of the presented
281 * sequence may not be used in location descriptors, the only exception being
282 * location in a remote entry (see 3.5.2.1, e).
284 * Location operators and descriptors are discussed in more detail below.
286 * 3.5.2.1 Location descriptors
288 * The location descriptor can be one of the following: (a) a single base
289 * number (b) a site between two indicated adjoining bases (c) a single base
290 * chosen from within a specified range of bases (not allowed for new entries)
291 * (d) the base numbers delimiting a sequence span (e) a remote entry
292 * identifier followed by a local location descriptor (i.e., a-d)
294 * A site between two adjoining nucleotides, such as endonucleolytic cleavage
295 * site, is indicated by listing the two points separated by a carat (^). The
296 * permitted formats for this descriptor are n^n+1 (for example 55^56), or,
297 * for circular molecules, n^1, where "n" is the full length of the molecule,
298 * ie 1000^1 for circular molecule with length 1000.
300 * A single base chosen from a range of bases is indicated by the first base
301 * number and the last base number of the range separated by a single period
302 * (e.g., '12.21' indicates a single base taken from between the indicated
303 * points). From October 2006 the usage of this descriptor is restricted : it
304 * is illegal to use "a single base from a range" (c) either on its own or in
305 * combination with the "sequence span" (d) descriptor for newly created
306 * entries. The existing entries where such descriptors exist are going to be
309 * Sequence spans are indicated by the starting base number and the ending
310 * base number separated by two periods (e.g., '34..456'). The '<' and '>'
311 * symbols may be used with the starting and ending base numbers to indicate
312 * that an end point is beyond the specified base number. The starting and
313 * ending base positions can be represented as distinct base numbers
314 * ('34..456') or a site between two indicated adjoining bases.
316 * A location in a remote entry (not the entry to which the feature table
317 * belongs) can be specified by giving the accession-number and sequence
318 * version of the remote entry, followed by a colon ":", followed by a
319 * location descriptor which applies to that entry's sequence (i.e.
320 * J12345.1:1..15, see also examples below)
324 * The location operator is a prefix that specifies what must be done to the
325 * indicated sequence to find or construct the location corresponding to the
326 * feature. A list of operators is given below with their definitions and most
329 * complement(location) Find the complement of the presented sequence in the
330 * span specified by " location" (i.e., read the complement of the presented
331 * strand in its 5'-to-3' direction)
333 * join(location,location, ... location) The indicated elements should be
334 * joined (placed end-to-end) to form one contiguous sequence
336 * order(location,location, ... location) The elements can be found in the
337 * specified order (5' to 3' direction), but nothing is implied about the
338 * reasonableness about joining them
340 * Note : location operator "complement" can be used in combination with
341 * either " join" or "order" within the same location; combinations of "join"
342 * and "order" within the same location (nested operators) are illegal.
346 * 3.5.3 Location examples
348 * The following is a list of common location descriptors with their meanings:
350 * Location Description
352 * 467 Points to a single base in the presented sequence
354 * 340..565 Points to a continuous range of bases bounded by and including the
355 * starting and ending bases
357 * <345..500 Indicates that the exact lower boundary point of a feature is
358 * unknown. The location begins at some base previous to the first base
359 * specified (which need not be contained in the presented sequence) and
360 * continues to and includes the ending base
362 * <1..888 The feature starts before the first sequenced base and continues to
363 * and includes base 888
365 * 1..>888 The feature starts at the first sequenced base and continues beyond
368 * 102.110 Indicates that the exact location is unknown but that it is one of
369 * the bases between bases 102 and 110, inclusive
371 * 123^124 Points to a site between bases 123 and 124
373 * join(12..78,134..202) Regions 12 to 78 and 134 to 202 should be joined to
374 * form one contiguous sequence
377 * complement(34..126) Start at the base complementary to 126 and finish at
378 * the base complementary to base 34 (the feature is on the strand
379 * complementary to the presented strand)
382 * complement(join(2691..4571,4918..5163)) Joins regions 2691 to 4571 and 4918
383 * to 5163, then complements the joined segments (the feature is on the strand
384 * complementary to the presented strand)
386 * join(complement(4918..5163),complement(2691..4571)) Complements regions
387 * 4918 to 5163 and 2691 to 4571, then joins the complemented segments (the
388 * feature is on the strand complementary to the presented strand)
390 * J00194.1:100..202 Points to bases 100 to 202, inclusive, in the entry (in
391 * this database) with primary accession number 'J00194'
393 * join(1..100,J00194.1:100..202) Joins region 1..100 of the existing entry
394 * with the region 100..202 of remote entry J00194
397 * Recover annotated sequences from EMBL file
400 * don't return nucleic acid sequences
404 * don't return any translated protein sequences marked in features
405 * @return dataset sequences with DBRefs and features - DNA always comes first
407 public jalview.datamodel.SequenceI[] getSequences(boolean noNa,
408 boolean noPeptide, String sourceDb)
409 { // TODO: ensure emblEntry.getSequences behaves correctly for returning all
410 // cases of noNa and noPeptide
411 Vector seqs = new Vector();
415 // In theory we still need to create this if noNa is set to avoid a null
417 dna = new Sequence(sourceDb + "|" + accession, sequence.getSequence());
418 dna.setDescription(desc);
419 DBRefEntry retrievedref = new DBRefEntry(sourceDb, version, accession);
420 dna.addDBRef(retrievedref);
421 // add map to indicate the sequence is a valid coordinate frame for the
423 retrievedref.setMap(new Mapping(null, new int[]
424 { 1, dna.getLength() }, new int[]
425 { 1, dna.getLength() }, 1, 1));
426 // TODO: transform EMBL Database refs to canonical form
429 for (Iterator i = dbRefs.iterator(); i.hasNext(); dna
430 .addDBRef((DBRefEntry) i.next()))
438 for (Iterator i = features.iterator(); i.hasNext();)
440 EmblFeature feature = (EmblFeature) i.next();
443 if (feature.dbRefs != null && feature.dbRefs.size() > 0)
445 for (Iterator dbr = feature.dbRefs.iterator(); dbr.hasNext(); dna
446 .addDBRef((DBRefEntry) dbr.next()))
452 if (FeatureProperties.isCodingFeature(sourceDb, feature.getName()))
454 parseCodingFeature(feature, sourceDb, seqs, dna, noPeptide);
458 // General feature type.
461 if (feature.dbRefs != null && feature.dbRefs.size() > 0)
463 for (Iterator dbr = feature.dbRefs.iterator(); dbr.hasNext(); dna
464 .addDBRef((DBRefEntry) dbr.next()))
472 } catch (Exception e)
474 System.err.println("EMBL Record Features parsing error!");
476 .println("Please report the following to help@jalview.org :");
477 System.err.println("EMBL Record " + accession);
478 System.err.println("Resulted in exception: " + e.getMessage());
479 e.printStackTrace(System.err);
481 if (!noNa && dna != null)
485 SequenceI[] sqs = new SequenceI[seqs.size()];
486 for (int i = 0, j = seqs.size(); i < j; i++)
488 sqs[i] = (SequenceI) seqs.elementAt(i);
495 * attempt to extract coding region and product from a feature and properly
496 * decorate it with annotations.
501 * source database for the EMBLXML
503 * place where sequences go
505 * parent dna sequence for this record
507 * flag for generation of Peptide sequence objects
509 private void parseCodingFeature(EmblFeature feature, String sourceDb,
510 Vector seqs, Sequence dna, boolean noPeptide)
512 boolean isEmblCdna = sourceDb.equals(DBRefSource.EMBLCDS);
513 // extract coding region(s)
514 jalview.datamodel.Mapping map = null;
516 if (feature.locations != null && feature.locations.size() > 0)
518 for (Enumeration locs = feature.locations.elements(); locs
521 EmblFeatureLocations loc = (EmblFeatureLocations) locs
523 int[] se = loc.getElementRanges(accession);
530 int[] t = new int[exon.length + se.length];
531 System.arraycopy(exon, 0, t, 0, exon.length);
532 System.arraycopy(se, 0, t, exon.length, se.length);
538 String prname = new String();
540 Hashtable vals = new Hashtable();
543 if (feature.getQualifiers() != null
544 && feature.getQualifiers().size() > 0)
546 for (Iterator quals = feature.getQualifiers().iterator(); quals
549 Qualifier q = (Qualifier) quals.next();
550 if (q.getName().equals("translation"))
552 StringBuffer prsq = new StringBuffer(q.getValues()[0]);
553 int p = prsq.indexOf(" ");
556 prsq.deleteCharAt(p);
557 p = prsq.indexOf(" ", p);
559 prseq = prsq.toString();
563 else if (q.getName().equals("protein_id"))
565 prid = q.getValues()[0];
567 else if (q.getName().equals("codon_start"))
569 prstart = Integer.parseInt(q.getValues()[0]);
571 else if (q.getName().equals("product"))
573 prname = q.getValues()[0];
577 // throw anything else into the additional properties hash
578 String[] s = q.getValues();
579 StringBuffer sb = new StringBuffer();
582 for (int i = 0; i < s.length; i++)
588 vals.put(q.getName(), sb.toString());
592 Sequence product = null;
593 exon = adjustForPrStart(prstart, exon);
595 if (prseq != null && prname != null && prid != null)
598 product = new Sequence(prid, prseq, 1, prseq.length());
599 product.setDescription(((prname.length() == 0) ? "Protein Product from "
604 // Protein is also added to vector of sequences returned
607 // we have everything - create the mapping and perhaps the protein
609 if (exon == null || exon.length == 0)
612 .println("Implementation Notice: EMBLCDS records not properly supported yet - Making up the CDNA region of this sequence... may be incorrect ("
613 + sourceDb + ":" + getAccession() + ")");
614 if (prseq.length() * 3 == (1 - prstart + dna.getSequence().length))
617 .println("Not allowing for additional stop codon at end of cDNA fragment... !");
618 // this might occur for CDS sequences where no features are
621 { dna.getStart() + (prstart - 1), dna.getEnd() };
622 map = new jalview.datamodel.Mapping(product, exon, new int[]
623 { 1, prseq.length() }, 3, 1);
625 if ((prseq.length() + 1) * 3 == (1 - prstart + dna.getSequence().length))
628 .println("Allowing for additional stop codon at end of cDNA fragment... will probably cause an error in VAMSAs!");
630 { dna.getStart() + (prstart - 1), dna.getEnd() - 3 };
631 map = new jalview.datamodel.Mapping(product, exon, new int[]
632 { 1, prseq.length() }, 3, 1);
637 // Trim the exon mapping if necessary - the given product may only be a
638 // fragment of a larger protein. (EMBL:AY043181 is an example)
642 // TODO: Add a DbRef back to the parent EMBL sequence with the exon
644 // if given a dataset reference, search dataset for parent EMBL
645 // sequence if it exists and set its map
646 // make a new feature annotating the coding contig
650 // final product length trunctation check
652 map = new jalview.datamodel.Mapping(product,
653 adjustForProteinLength(prseq.length(), exon), new int[]
654 { 1, prseq.length() }, 3, 1);
655 // reconstruct the EMBLCDS entry
656 // TODO: this is only necessary when there codon annotation is
657 // complete (I think JBPNote)
658 DBRefEntry pcdnaref = new DBRefEntry();
659 pcdnaref.setAccessionId(prid);
660 pcdnaref.setSource(DBRefSource.EMBLCDS);
661 pcdnaref.setVersion(getVersion()); // same as parent EMBL version.
662 jalview.util.MapList mp = new jalview.util.MapList(new int[]
663 { 1, prseq.length() }, new int[]
664 { 1 + (prstart - 1), (prstart - 1) + 3 * prseq.length() }, 1, 3);
665 // { 1 + (prstart - 1) * 3,
666 // 1 + (prstart - 1) * 3 + prseq.length() * 3 - 1 }, new int[]
667 // { 1prstart, prstart + prseq.length() - 1 }, 3, 1);
668 pcdnaref.setMap(new Mapping(mp));
671 product.addDBRef(pcdnaref);
676 // add cds feature to dna seq - this may include the stop codon
677 for (int xint = 0; exon != null && xint < exon.length; xint += 2)
679 SequenceFeature sf = new SequenceFeature();
680 sf.setBegin(exon[xint]);
681 sf.setEnd(exon[xint + 1]);
682 sf.setType(feature.getName());
683 sf.setFeatureGroup(sourceDb);
684 sf.setDescription("Exon " + (1 + xint / 2)
685 + " for protein '" + prname + "' EMBLCDS:" + prid);
686 sf.setValue(FeatureProperties.EXONPOS, new Integer(1 + xint));
687 sf.setValue(FeatureProperties.EXONPRODUCT, prname);
688 if (vals != null && vals.size() > 0)
690 Enumeration kv = vals.keys();
691 while (kv.hasMoreElements())
693 Object key = kv.nextElement();
696 sf.setValue(key.toString(), vals.get(key));
700 dna.addSequenceFeature(sf);
703 // add dbRefs to sequence
704 if (feature.dbRefs != null && feature.dbRefs.size() > 0)
706 for (Iterator dbr = feature.dbRefs.iterator(); dbr.hasNext();)
708 DBRefEntry ref = (DBRefEntry) dbr.next();
709 ref.setSource(jalview.util.DBRefUtils.getCanonicalName(ref
711 // Hard code the kind of protein product accessions that EMBL cite
712 if (ref.getSource().equals(jalview.datamodel.DBRefSource.UNIPROT))
715 if (map != null && map.getTo() != null)
717 map.getTo().addDBRef(
718 new DBRefEntry(ref.getSource(), ref.getVersion(), ref
719 .getAccessionId())); // don't copy map over.
720 if (map.getTo().getName().indexOf(prid) == 0)
723 jalview.datamodel.DBRefSource.UNIPROT + "|"
724 + ref.getAccessionId());
730 DBRefEntry pref = new DBRefEntry(ref.getSource(),
731 ref.getVersion(), ref.getAccessionId());
732 pref.setMap(null); // reference is direct
733 product.addDBRef(pref);
734 // Add converse mapping reference
737 Mapping pmap = new Mapping(dna, map.getMap().getInverse());
738 pref = new DBRefEntry(sourceDb, getVersion(),
739 this.getAccession());
741 if (map.getTo() != null)
743 map.getTo().addDBRef(pref);
752 private int[] adjustForPrStart(int prstart, int[] exon)
755 int origxon[], sxpos = -1;
756 int sxstart, sxstop; // unnecessary variables used for debugging
757 // first adjust range for codon start attribute
760 origxon = new int[exon.length];
761 System.arraycopy(exon, 0, origxon, 0, exon.length);
763 for (int x = 0; x < exon.length && sxpos == -1; x += 2)
765 cdspos += exon[x + 1] - exon[x] + 1;
766 if (prstart <= cdspos)
770 sxstop = exon[x + 1];
771 // and adjust start boundary of first exon.
772 exon[x] = exon[x + 1] - cdspos + prstart;
779 int[] nxon = new int[exon.length - sxpos];
780 System.arraycopy(exon, sxpos, nxon, 0, exon.length - sxpos);
788 * truncate the last exon interval to the prlength'th codon
794 private int[] adjustForProteinLength(int prlength, int[] exon)
797 int origxon[], sxpos = -1, endxon = 0, cdslength = prlength * 3;
798 int sxstart, sxstop; // unnecessary variables used for debugging
799 // first adjust range for codon start attribute
800 if (prlength >= 1 && exon != null)
802 origxon = new int[exon.length];
803 System.arraycopy(exon, 0, origxon, 0, exon.length);
805 for (int x = 0; x < exon.length && sxpos == -1; x += 2)
807 cdspos += exon[x + 1] - exon[x] + 1;
808 if (cdslength <= cdspos)
810 // advanced beyond last codon.
813 sxstop = exon[x + 1];
814 if (cdslength != cdspos)
817 .println("Truncating final exon interval on region by "
818 + (cdspos - cdslength));
820 // locate the new end boundary of final exon as endxon
821 endxon = exon[x + 1] - cdspos + cdslength;
828 // and trim the exon interval set if necessary
829 int[] nxon = new int[sxpos + 2];
830 System.arraycopy(exon, 0, nxon, 0, sxpos + 2);
831 nxon[sxpos + 1] = endxon; // update the end boundary for the new exon