2 * Jalview - A Sequence Alignment Editor and Viewer (Development Version 2.4.1)
3 * Copyright (C) 2009 AM Waterhouse, J Procter, G Barton, M Clamp, S Searle
5 * This program is free software; you can redistribute it and/or
6 * modify it under the terms of the GNU General Public License
7 * as published by the Free Software Foundation; either version 2
8 * of the License, or (at your option) any later version.
10 * This program is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 * GNU General Public License for more details.
15 * You should have received a copy of the GNU General Public License
16 * along with this program; if not, write to the Free Software
17 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA
19 package jalview.datamodel.xdb.embl;
21 import jalview.datamodel.DBRefEntry;
22 import jalview.datamodel.DBRefSource;
23 import jalview.datamodel.FeatureProperties;
24 import jalview.datamodel.Mapping;
25 import jalview.datamodel.Sequence;
26 import jalview.datamodel.SequenceFeature;
27 import jalview.datamodel.SequenceI;
29 import java.util.Enumeration;
30 import java.util.Hashtable;
31 import java.util.Iterator;
32 import java.util.Vector;
34 public class EmblEntry
58 EmblSequence sequence;
61 * @return the accession
63 public String getAccession()
70 * the accession to set
72 public void setAccession(String accession)
74 this.accession = accession;
80 public Vector getDbRefs()
89 public void setDbRefs(Vector dbRefs)
97 public String getDesc()
106 public void setDesc(String desc)
112 * @return the features
114 public Vector getFeatures()
121 * the features to set
123 public void setFeatures(Vector features)
125 this.features = features;
129 * @return the keywords
131 public Vector getKeywords()
138 * the keywords to set
140 public void setKeywords(Vector keywords)
142 this.keywords = keywords;
146 * @return the lastUpdated
148 public String getLastUpdated()
155 * the lastUpdated to set
157 public void setLastUpdated(String lastUpdated)
159 this.lastUpdated = lastUpdated;
165 public Vector getRefs()
174 public void setRefs(Vector refs)
180 * @return the releaseCreated
182 public String getRCreated()
188 * @param releaseCreated
189 * the releaseCreated to set
191 public void setRcreated(String releaseCreated)
193 this.rCreated = releaseCreated;
197 * @return the releaseLastUpdated
199 public String getRLastUpdated()
205 * @param releaseLastUpdated
206 * the releaseLastUpdated to set
208 public void setRLastUpdated(String releaseLastUpdated)
210 this.rLastUpdated = releaseLastUpdated;
214 * @return the sequence
216 public EmblSequence getSequence()
223 * the sequence to set
225 public void setSequence(EmblSequence sequence)
227 this.sequence = sequence;
231 * @return the taxDivision
233 public String getTaxDivision()
240 * the taxDivision to set
242 public void setTaxDivision(String taxDivision)
244 this.taxDivision = taxDivision;
248 * @return the version
250 public String getVersion()
259 public void setVersion(String version)
261 this.version = version;
265 * EMBL Feature support is limited. The text below is included for the benefit
266 * of any developer working on improving EMBL feature import in Jalview.
267 * Extract from EMBL feature specification see
268 * http://www.embl-ebi.ac.uk/embl/Documentation
269 * /FT_definitions/feature_table.html 3.5 Location 3.5.1 Purpose
271 * The location indicates the region of the presented sequence which
272 * corresponds to a feature.
274 * 3.5.2 Format and conventions The location contains at least one sequence
275 * location descriptor and may contain one or more operators with one or more
276 * sequence location descriptors. Base numbers refer to the numbering in the
277 * entry. This numbering designates the first base (5' end) of the presented
278 * sequence as base 1. Base locations beyond the range of the presented
279 * sequence may not be used in location descriptors, the only exception being
280 * location in a remote entry (see 3.5.2.1, e).
282 * Location operators and descriptors are discussed in more detail below.
284 * 3.5.2.1 Location descriptors
286 * The location descriptor can be one of the following: (a) a single base
287 * number (b) a site between two indicated adjoining bases (c) a single base
288 * chosen from within a specified range of bases (not allowed for new entries)
289 * (d) the base numbers delimiting a sequence span (e) a remote entry
290 * identifier followed by a local location descriptor (i.e., a-d)
292 * A site between two adjoining nucleotides, such as endonucleolytic cleavage
293 * site, is indicated by listing the two points separated by a carat (^). The
294 * permitted formats for this descriptor are n^n+1 (for example 55^56), or,
295 * for circular molecules, n^1, where "n" is the full length of the molecule,
296 * ie 1000^1 for circular molecule with length 1000.
298 * A single base chosen from a range of bases is indicated by the first base
299 * number and the last base number of the range separated by a single period
300 * (e.g., '12.21' indicates a single base taken from between the indicated
301 * points). From October 2006 the usage of this descriptor is restricted : it
302 * is illegal to use "a single base from a range" (c) either on its own or in
303 * combination with the "sequence span" (d) descriptor for newly created
304 * entries. The existing entries where such descriptors exist are going to be
307 * Sequence spans are indicated by the starting base number and the ending
308 * base number separated by two periods (e.g., '34..456'). The '<' and '>'
309 * symbols may be used with the starting and ending base numbers to indicate
310 * that an end point is beyond the specified base number. The starting and
311 * ending base positions can be represented as distinct base numbers
312 * ('34..456') or a site between two indicated adjoining bases.
314 * A location in a remote entry (not the entry to which the feature table
315 * belongs) can be specified by giving the accession-number and sequence
316 * version of the remote entry, followed by a colon ":", followed by a
317 * location descriptor which applies to that entry's sequence (i.e.
318 * J12345.1:1..15, see also examples below)
322 * The location operator is a prefix that specifies what must be done to the
323 * indicated sequence to find or construct the location corresponding to the
324 * feature. A list of operators is given below with their definitions and most
327 * complement(location) Find the complement of the presented sequence in the
328 * span specified by " location" (i.e., read the complement of the presented
329 * strand in its 5'-to-3' direction)
331 * join(location,location, ... location) The indicated elements should be
332 * joined (placed end-to-end) to form one contiguous sequence
334 * order(location,location, ... location) The elements can be found in the
335 * specified order (5' to 3' direction), but nothing is implied about the
336 * reasonableness about joining them
338 * Note : location operator "complement" can be used in combination with
339 * either " join" or "order" within the same location; combinations of "join"
340 * and "order" within the same location (nested operators) are illegal.
344 * 3.5.3 Location examples
346 * The following is a list of common location descriptors with their meanings:
348 * Location Description
350 * 467 Points to a single base in the presented sequence
352 * 340..565 Points to a continuous range of bases bounded by and including the
353 * starting and ending bases
355 * <345..500 Indicates that the exact lower boundary point of a feature is
356 * unknown. The location begins at some base previous to the first base
357 * specified (which need not be contained in the presented sequence) and
358 * continues to and includes the ending base
360 * <1..888 The feature starts before the first sequenced base and continues to
361 * and includes base 888
363 * 1..>888 The feature starts at the first sequenced base and continues beyond
366 * 102.110 Indicates that the exact location is unknown but that it is one of
367 * the bases between bases 102 and 110, inclusive
369 * 123^124 Points to a site between bases 123 and 124
371 * join(12..78,134..202) Regions 12 to 78 and 134 to 202 should be joined to
372 * form one contiguous sequence
375 * complement(34..126) Start at the base complementary to 126 and finish at
376 * the base complementary to base 34 (the feature is on the strand
377 * complementary to the presented strand)
380 * complement(join(2691..4571,4918..5163)) Joins regions 2691 to 4571 and 4918
381 * to 5163, then complements the joined segments (the feature is on the strand
382 * complementary to the presented strand)
384 * join(complement(4918..5163),complement(2691..4571)) Complements regions
385 * 4918 to 5163 and 2691 to 4571, then joins the complemented segments (the
386 * feature is on the strand complementary to the presented strand)
388 * J00194.1:100..202 Points to bases 100 to 202, inclusive, in the entry (in
389 * this database) with primary accession number 'J00194'
391 * join(1..100,J00194.1:100..202) Joins region 1..100 of the existing entry
392 * with the region 100..202 of remote entry J00194
395 * Recover annotated sequences from EMBL file
398 * don't return nucleic acid sequences
402 * don't return any translated protein sequences marked in features
403 * @return dataset sequences with DBRefs and features - DNA always comes first
405 public jalview.datamodel.SequenceI[] getSequences(boolean noNa,
406 boolean noPeptide, String sourceDb)
407 { // TODO: ensure emblEntry.getSequences behaves correctly for returning all
408 // cases of noNa and noPeptide
409 Vector seqs = new Vector();
413 // In theory we still need to create this if noNa is set to avoid a null
415 dna = new Sequence(sourceDb + "|" + accession, sequence.getSequence());
416 dna.setDescription(desc);
417 DBRefEntry retrievedref = new DBRefEntry(sourceDb, version, accession);
418 dna.addDBRef(retrievedref);
419 // add map to indicate the sequence is a valid coordinate frame for the
421 retrievedref.setMap(new Mapping(null, new int[]
422 { 1, dna.getLength() }, new int[]
423 { 1, dna.getLength() }, 1, 1));
424 // TODO: transform EMBL Database refs to canonical form
426 for (Iterator i = dbRefs.iterator(); i.hasNext(); dna
427 .addDBRef((DBRefEntry) i.next()))
432 for (Iterator i = features.iterator(); i.hasNext();)
434 EmblFeature feature = (EmblFeature) i.next();
437 if (feature.dbRefs != null && feature.dbRefs.size() > 0)
439 for (Iterator dbr = feature.dbRefs.iterator(); dbr.hasNext(); dna
440 .addDBRef((DBRefEntry) dbr.next()))
444 if (FeatureProperties.isCodingFeature(sourceDb, feature.getName()))
446 parseCodingFeature(feature, sourceDb, seqs, dna, noPeptide);
450 // General feature type.
453 if (feature.dbRefs != null && feature.dbRefs.size() > 0)
455 for (Iterator dbr = feature.dbRefs.iterator(); dbr.hasNext(); dna
456 .addDBRef((DBRefEntry) dbr.next()))
462 } catch (Exception e)
464 System.err.println("EMBL Record Features parsing error!");
466 .println("Please report the following to help@jalview.org :");
467 System.err.println("EMBL Record " + accession);
468 System.err.println("Resulted in exception: " + e.getMessage());
469 e.printStackTrace(System.err);
471 if (!noNa && dna != null)
475 SequenceI[] sqs = new SequenceI[seqs.size()];
476 for (int i = 0, j = seqs.size(); i < j; i++)
478 sqs[i] = (SequenceI) seqs.elementAt(i);
485 * attempt to extract coding region and product from a feature and properly
486 * decorate it with annotations.
491 * source database for the EMBLXML
493 * place where sequences go
495 * parent dna sequence for this record
497 * flag for generation of Peptide sequence objects
499 private void parseCodingFeature(EmblFeature feature, String sourceDb,
500 Vector seqs, Sequence dna, boolean noPeptide)
502 boolean isEmblCdna = sourceDb.equals(DBRefSource.EMBLCDS);
503 // extract coding region(s)
504 jalview.datamodel.Mapping map = null;
506 if (feature.locations != null && feature.locations.size() > 0)
508 for (Enumeration locs = feature.locations.elements(); locs
511 EmblFeatureLocations loc = (EmblFeatureLocations) locs
513 int[] se = loc.getElementRanges(accession);
520 int[] t = new int[exon.length + se.length];
521 System.arraycopy(exon, 0, t, 0, exon.length);
522 System.arraycopy(se, 0, t, exon.length, se.length);
528 String prname = new String();
530 Hashtable vals = new Hashtable();
533 if (feature.getQualifiers() != null
534 && feature.getQualifiers().size() > 0)
536 for (Iterator quals = feature.getQualifiers().iterator(); quals
539 Qualifier q = (Qualifier) quals.next();
540 if (q.getName().equals("translation"))
542 StringBuffer prsq = new StringBuffer(q.getValues()[0]);
543 int p = prsq.indexOf(" ");
546 prsq.deleteCharAt(p);
547 p = prsq.indexOf(" ", p);
549 prseq = prsq.toString();
553 else if (q.getName().equals("protein_id"))
555 prid = q.getValues()[0];
557 else if (q.getName().equals("codon_start"))
559 prstart = Integer.parseInt(q.getValues()[0]);
561 else if (q.getName().equals("product"))
563 prname = q.getValues()[0];
567 // throw anything else into the additional properties hash
568 String[] s = q.getValues();
569 StringBuffer sb = new StringBuffer();
572 for (int i = 0; i < s.length; i++)
578 vals.put(q.getName(), sb.toString());
582 Sequence product = null;
583 exon = adjustForPrStart(prstart, exon);
585 if (prseq != null && prname != null && prid != null)
588 product = new Sequence(prid, prseq, 1, prseq.length());
590 .setDescription(((prname.length() == 0) ? "Protein Product from "
595 // Protein is also added to vector of sequences returned
598 // we have everything - create the mapping and perhaps the protein
600 if (exon == null || exon.length == 0)
603 .println("Implementation Notice: EMBLCDS records not properly supported yet - Making up the CDNA region of this sequence... may be incorrect ("
604 + sourceDb + ":" + getAccession() + ")");
605 if (prseq.length() * 3 == (1-prstart + dna.getSequence().length))
608 .println("Not allowing for additional stop codon at end of cDNA fragment... !");
609 // this might occur for CDS sequences where no features are
612 { dna.getStart() + (prstart - 1), dna.getEnd() };
613 map = new jalview.datamodel.Mapping(product, exon, new int[]
614 { 1, prseq.length() }, 3, 1);
616 if ((prseq.length() + 1) * 3 == (1-prstart + dna.getSequence().length))
619 .println("Allowing for additional stop codon at end of cDNA fragment... will probably cause an error in VAMSAs!");
621 { dna.getStart() + (prstart - 1), dna.getEnd() - 3 };
622 map = new jalview.datamodel.Mapping(product, exon, new int[]
623 { 1, prseq.length() }, 3, 1);
628 // Trim the exon mapping if necessary - the given product may only be a fragment of a larger protein. (EMBL:AY043181 is an example)
633 // TODO: Add a DbRef back to the parent EMBL sequence with the exon
635 // if given a dataset reference, search dataset for parent EMBL
636 // sequence if it exists and set its map
637 // make a new feature annotating the coding contig
641 // final product length trunctation check
643 map = new jalview.datamodel.Mapping(product, adjustForProteinLength(prseq.length(),exon), new int[]
644 { 1, prseq.length() }, 3, 1);
645 // reconstruct the EMBLCDS entry
646 // TODO: this is only necessary when there codon annotation is 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() },
655 (prstart - 1) + 3 * prseq.length() }, 1, 3);
656 // { 1 + (prstart - 1) * 3,
657 // 1 + (prstart - 1) * 3 + prseq.length() * 3 - 1 }, new int[]
658 // { 1prstart, prstart + prseq.length() - 1 }, 3, 1);
659 pcdnaref.setMap(new Mapping(mp));
661 product.addDBRef(pcdnaref);
665 // add cds feature to dna seq - this may include the stop codon
666 for (int xint = 0; exon != null && xint < exon.length; xint += 2)
668 SequenceFeature sf = new SequenceFeature();
669 sf.setBegin(exon[xint]);
670 sf.setEnd(exon[xint + 1]);
671 sf.setType(feature.getName());
672 sf.setFeatureGroup(sourceDb);
673 sf.setDescription("Exon " + (1 + (int) (xint / 2))
674 + " for protein '" + prname + "' EMBLCDS:" + prid);
675 sf.setValue(FeatureProperties.EXONPOS, new Integer(1 + xint));
676 sf.setValue(FeatureProperties.EXONPRODUCT, prname);
677 if (vals != null && vals.size() > 0)
679 Enumeration kv = vals.elements();
680 while (kv.hasMoreElements())
682 Object key = kv.nextElement();
684 sf.setValue(key.toString(), vals.get(key));
687 dna.addSequenceFeature(sf);
690 // add dbRefs to sequence
691 if (feature.dbRefs != null && feature.dbRefs.size() > 0)
693 for (Iterator dbr = feature.dbRefs.iterator(); dbr.hasNext();)
695 DBRefEntry ref = (DBRefEntry) dbr.next();
696 ref.setSource(jalview.util.DBRefUtils.getCanonicalName(ref
698 // Hard code the kind of protein product accessions that EMBL cite
699 if (ref.getSource().equals(jalview.datamodel.DBRefSource.UNIPROT))
702 if (map != null && map.getTo() != null)
704 map.getTo().addDBRef(
705 new DBRefEntry(ref.getSource(), ref.getVersion(), ref
706 .getAccessionId())); // don't copy map over.
707 if (map.getTo().getName().indexOf(prid) == 0)
710 jalview.datamodel.DBRefSource.UNIPROT + "|"
711 + ref.getAccessionId());
717 DBRefEntry pref = new DBRefEntry(ref.getSource(), ref
718 .getVersion(), ref.getAccessionId());
719 pref.setMap(null); // reference is direct
720 product.addDBRef(pref);
721 // Add converse mapping reference
724 Mapping pmap = new Mapping(dna, map.getMap().getInverse());
725 pref = new DBRefEntry(sourceDb, getVersion(), this
728 if (map.getTo() != null)
730 map.getTo().addDBRef(pref);
739 private int[] adjustForPrStart(int prstart, int[] exon)
742 int origxon[], sxpos = -1;
743 int sxstart, sxstop; // unnecessary variables used for debugging
744 // first adjust range for codon start attribute
747 origxon = new int[exon.length];
748 System.arraycopy(exon, 0, origxon, 0, exon.length);
750 for (int x = 0; x < exon.length && sxpos == -1; x += 2)
752 cdspos += exon[x + 1] - exon[x] + 1;
753 if (prstart <= cdspos)
757 sxstop = exon[x + 1];
758 // and adjust start boundary of first exon.
759 exon[x] = exon[x + 1] - cdspos + prstart;
766 int[] nxon = new int[exon.length - sxpos];
767 System.arraycopy(exon, sxpos, nxon, 0, exon.length - sxpos);
774 * 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) {
800 System.err.println("Truncating final exon interval on region by "+(cdspos-cdslength));
802 // locate the new end boundary of final exon as endxon
803 endxon = exon[x+1] - cdspos + cdslength;
810 // and trim the exon interval set if necessary
811 int[] nxon = new int[sxpos+2];
812 System.arraycopy(exon, 0, nxon, 0, sxpos+2);
813 nxon[sxpos+1] = endxon; // update the end boundary for the new exon set