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.
23 import jalview.datamodel.DBRefEntry;
24 import jalview.datamodel.DBRefSource;
25 import jalview.datamodel.PDBEntry;
26 import jalview.datamodel.SequenceI;
28 import java.util.ArrayList;
29 import java.util.Arrays;
30 import java.util.HashMap;
31 import java.util.HashSet;
32 import java.util.List;
36 import com.stevesoft.pat.Regex;
39 * Utilities for handling DBRef objects and their collections.
41 public class DBRefUtils
44 * lookup from lower-case form of a name to its canonical (standardised) form
46 private static Map<String, String> canonicalSourceNameLookup = new HashMap<String, String>();
48 private static Map<String, String> dasCoordinateSystemsLookup = new HashMap<String, String>();
52 // TODO load these from a resource file?
53 canonicalSourceNameLookup.put("uniprotkb/swiss-prot",
55 canonicalSourceNameLookup.put("uniprotkb/trembl", DBRefSource.UNIPROT);
57 // Ensembl values for dbname in xref REST service:
58 canonicalSourceNameLookup.put("uniprot/sptrembl", DBRefSource.UNIPROT);
59 canonicalSourceNameLookup.put("uniprot/swissprot", DBRefSource.UNIPROT);
61 canonicalSourceNameLookup.put("pdb", DBRefSource.PDB);
62 canonicalSourceNameLookup.put("ensembl", DBRefSource.ENSEMBL);
63 // Ensembl Gn and Tr are for Ensembl genomic and transcript IDs as served
65 canonicalSourceNameLookup.put("ensembl-tr", DBRefSource.ENSEMBL);
66 canonicalSourceNameLookup.put("ensembl-gn", DBRefSource.ENSEMBL);
68 // Make sure we have lowercase entries for all canonical string lookups
69 Set<String> keys = canonicalSourceNameLookup.keySet();
72 canonicalSourceNameLookup.put(k.toLowerCase(),
73 canonicalSourceNameLookup.get(k));
76 dasCoordinateSystemsLookup.put("pdbresnum", DBRefSource.PDB);
77 dasCoordinateSystemsLookup.put("uniprot", DBRefSource.UNIPROT);
78 dasCoordinateSystemsLookup.put("embl", DBRefSource.EMBL);
79 // dasCoordinateSystemsLookup.put("embl", DBRefSource.EMBLCDS);
83 * Returns those DBRefEntry objects whose source identifier (once converted to
84 * Jalview's canonical form) is in the list of sources to search for. Returns
85 * null if no matches found.
88 * DBRefEntry objects to search
90 * array of sources to select
93 public static DBRefEntry[] selectRefs(DBRefEntry[] dbrefs,
96 if (dbrefs == null || sources == null)
100 HashSet<String> srcs = new HashSet<String>();
101 for (String src : sources)
103 srcs.add(src.toUpperCase());
106 List<DBRefEntry> res = new ArrayList<DBRefEntry>();
107 for (DBRefEntry dbr : dbrefs)
109 String source = getCanonicalName(dbr.getSource());
110 if (srcs.contains(source.toUpperCase()))
118 DBRefEntry[] reply = new DBRefEntry[res.size()];
119 return res.toArray(reply);
125 * isDasCoordinateSystem
131 * @return boolean true if Source DBRefEntry is compatible with DAS
132 * CoordinateSystem name
135 public static boolean isDasCoordinateSystem(String string,
136 DBRefEntry dBRefEntry)
138 if (string == null || dBRefEntry == null)
142 String coordsys = dasCoordinateSystemsLookup.get(string.toLowerCase());
143 return coordsys == null ? false
144 : coordsys.equals(dBRefEntry.getSource());
148 * look up source in an internal list of database reference sources and return
149 * the canonical jalview name for the source, or the original string if it has
153 * @return canonical jalview source (one of jalview.datamodel.DBRefSource.*)
156 public static String getCanonicalName(String source)
162 String canonical = canonicalSourceNameLookup.get(source.toLowerCase());
163 return canonical == null ? source : canonical;
167 * Returns a (possibly empty) list of those references that match the given
168 * entry. Currently uses a comparator which matches if
170 * <li>database sources are the same</li>
171 * <li>accession ids are the same</li>
172 * <li>both have no mapping, or the mappings are the same</li>
176 * Set of references to search
181 public static List<DBRefEntry> searchRefs(DBRefEntry[] ref,
184 return searchRefs(ref, entry,
185 matchDbAndIdAndEitherMapOrEquivalentMapList);
189 * Returns a list of those references that match the given accession id
191 * <li>database sources are the same</li>
192 * <li>accession ids are the same</li>
193 * <li>both have no mapping, or the mappings are the same</li>
197 * Set of references to search
199 * accession id to match
202 public static List<DBRefEntry> searchRefs(DBRefEntry[] refs, String accId)
204 return searchRefs(refs, new DBRefEntry("", "", accId), matchId);
208 * Returns a (possibly empty) list of those references that match the given
209 * entry, according to the given comparator.
212 * an array of database references to search
214 * an entry to compare against
218 static List<DBRefEntry> searchRefs(DBRefEntry[] refs, DBRefEntry entry,
219 DbRefComp comparator)
221 List<DBRefEntry> rfs = new ArrayList<DBRefEntry>();
222 if (refs == null || entry == null)
226 for (int i = 0; i < refs.length; i++)
228 if (comparator.matches(entry, refs[i]))
238 public boolean matches(DBRefEntry refa, DBRefEntry refb);
242 * match on all non-null fields in refa
244 // TODO unused - remove?
245 public static DbRefComp matchNonNullonA = new DbRefComp()
248 public boolean matches(DBRefEntry refa, DBRefEntry refb)
250 if (refa.getSource() == null
251 || DBRefUtils.getCanonicalName(refb.getSource()).equals(
252 DBRefUtils.getCanonicalName(refa.getSource())))
254 if (refa.getVersion() == null
255 || refb.getVersion().equals(refa.getVersion()))
257 if (refa.getAccessionId() == null
258 || refb.getAccessionId().equals(refa.getAccessionId()))
260 if (refa.getMap() == null || (refb.getMap() != null
261 && refb.getMap().equals(refa.getMap())))
273 * either field is null or field matches for all of source, version, accession
276 // TODO unused - remove?
277 public static DbRefComp matchEitherNonNull = new DbRefComp()
280 public boolean matches(DBRefEntry refa, DBRefEntry refb)
282 if (nullOrEqualSource(refa.getSource(), refb.getSource())
283 && nullOrEqual(refa.getVersion(), refb.getVersion())
284 && nullOrEqual(refa.getAccessionId(), refb.getAccessionId())
285 && nullOrEqual(refa.getMap(), refb.getMap()))
294 * accession ID and DB must be identical. Version is ignored. Map is either
295 * not defined or is a match (or is compatible?)
297 // TODO unused - remove?
298 public static DbRefComp matchDbAndIdAndEitherMap = new DbRefComp()
301 public boolean matches(DBRefEntry refa, DBRefEntry refb)
303 if (refa.getSource() != null && refb.getSource() != null
304 && DBRefUtils.getCanonicalName(refb.getSource()).equals(
305 DBRefUtils.getCanonicalName(refa.getSource())))
307 // We dont care about version
308 if (refa.getAccessionId() != null && refb.getAccessionId() != null
309 // FIXME should be && not || here?
310 || refb.getAccessionId().equals(refa.getAccessionId()))
312 if ((refa.getMap() == null || refb.getMap() == null)
313 || (refa.getMap() != null && refb.getMap() != null
314 && refb.getMap().equals(refa.getMap())))
325 * accession ID and DB must be identical. Version is ignored. No map on either
326 * or map but no maplist on either or maplist of map on a is the complement of
327 * maplist of map on b.
329 // TODO unused - remove?
330 public static DbRefComp matchDbAndIdAndComplementaryMapList = new DbRefComp()
333 public boolean matches(DBRefEntry refa, DBRefEntry refb)
335 if (refa.getSource() != null && refb.getSource() != null
336 && DBRefUtils.getCanonicalName(refb.getSource()).equals(
337 DBRefUtils.getCanonicalName(refa.getSource())))
339 // We dont care about version
340 if (refa.getAccessionId() != null && refb.getAccessionId() != null
341 || refb.getAccessionId().equals(refa.getAccessionId()))
343 if ((refa.getMap() == null && refb.getMap() == null)
344 || (refa.getMap() != null && refb.getMap() != null))
346 if ((refb.getMap().getMap() == null
347 && refa.getMap().getMap() == null)
348 || (refb.getMap().getMap() != null
349 && refa.getMap().getMap() != null
350 && refb.getMap().getMap().getInverse()
351 .equals(refa.getMap().getMap())))
363 * accession ID and DB must be identical. Version is ignored. No map on both
364 * or or map but no maplist on either or maplist of map on a is equivalent to
365 * the maplist of map on b.
367 // TODO unused - remove?
368 public static DbRefComp matchDbAndIdAndEquivalentMapList = new DbRefComp()
371 public boolean matches(DBRefEntry refa, DBRefEntry refb)
373 if (refa.getSource() != null && refb.getSource() != null
374 && DBRefUtils.getCanonicalName(refb.getSource()).equals(
375 DBRefUtils.getCanonicalName(refa.getSource())))
377 // We dont care about version
378 // if ((refa.getVersion()==null || refb.getVersion()==null)
379 // || refb.getVersion().equals(refa.getVersion()))
381 if (refa.getAccessionId() != null && refb.getAccessionId() != null
382 || refb.getAccessionId().equals(refa.getAccessionId()))
384 if (refa.getMap() == null && refb.getMap() == null)
388 if (refa.getMap() != null && refb.getMap() != null
389 && ((refb.getMap().getMap() == null
390 && refa.getMap().getMap() == null)
391 || (refb.getMap().getMap() != null
392 && refa.getMap().getMap() != null
393 && refb.getMap().getMap()
394 .equals(refa.getMap().getMap()))))
405 * accession ID and DB must be identical, or null on a. Version is ignored. No
406 * map on either or map but no maplist on either or maplist of map on a is
407 * equivalent to the maplist of map on b.
409 public static DbRefComp matchDbAndIdAndEitherMapOrEquivalentMapList = new DbRefComp()
412 public boolean matches(DBRefEntry refa, DBRefEntry refb)
414 if (refa.getSource() != null && refb.getSource() != null
415 && DBRefUtils.getCanonicalName(refb.getSource()).equals(
416 DBRefUtils.getCanonicalName(refa.getSource())))
418 // We dont care about version
420 if (refa.getAccessionId() == null
421 || refa.getAccessionId().equals(refb.getAccessionId()))
423 if (refa.getMap() == null || refb.getMap() == null)
427 if ((refa.getMap() != null && refb.getMap() != null)
428 && (refb.getMap().getMap() == null
429 && refa.getMap().getMap() == null)
430 || (refb.getMap().getMap() != null
431 && refa.getMap().getMap() != null
432 && (refb.getMap().getMap()
433 .equals(refa.getMap().getMap()))))
444 * accession ID only must be identical.
446 public static DbRefComp matchId = new DbRefComp()
449 public boolean matches(DBRefEntry refa, DBRefEntry refb)
451 if (refa.getAccessionId() != null && refb.getAccessionId() != null
452 && refb.getAccessionId().equals(refa.getAccessionId()))
461 * Parses a DBRefEntry and adds it to the sequence, also a PDBEntry if the
464 * Used by file parsers to generate DBRefs from annotation within file (eg
471 * where to annotate with reference
472 * @return parsed version of entry that was added to seq (if any)
474 public static DBRefEntry parseToDbRef(SequenceI seq, String dbname,
475 String version, String acn)
477 DBRefEntry ref = null;
480 String locsrc = DBRefUtils.getCanonicalName(dbname);
481 if (locsrc.equals(DBRefSource.PDB))
484 * Check for PFAM style stockhom PDB accession id citation e.g.
487 Regex r = new com.stevesoft.pat.Regex(
488 "([0-9][0-9A-Za-z]{3})\\s*(.?)\\s*;\\s*([0-9]+)-([0-9]+)");
489 if (r.search(acn.trim()))
491 String pdbid = r.stringMatched(1);
492 String chaincode = r.stringMatched(2);
493 if (chaincode == null)
497 // String mapstart = r.stringMatched(3);
498 // String mapend = r.stringMatched(4);
499 if (chaincode.equals(" "))
503 // construct pdb ref.
504 ref = new DBRefEntry(locsrc, version, pdbid + chaincode);
505 PDBEntry pdbr = new PDBEntry();
507 pdbr.setType(PDBEntry.Type.PDB);
508 pdbr.setChainCode(chaincode);
513 System.err.println("Malformed PDB DR line:" + acn);
519 ref = new DBRefEntry(locsrc, version, acn);
530 * Returns true if either object is null, or they are equal
536 public static boolean nullOrEqual(Object o1, Object o2)
538 if (o1 == null || o2 == null)
542 return o1.equals(o2);
546 * canonicalise source string before comparing. null is always wildcard
549 * - null or source string to compare
551 * - null or source string to compare
552 * @return true if either o1 or o2 are null, or o1 equals o2 under
553 * DBRefUtils.getCanonicalName
554 * (o1).equals(DBRefUtils.getCanonicalName(o2))
556 public static boolean nullOrEqualSource(String o1, String o2)
558 if (o1 == null || o2 == null)
562 return DBRefUtils.getCanonicalName(o1)
563 .equals(DBRefUtils.getCanonicalName(o2));
567 * Selects just the DNA or protein references from a set of references
570 * if true, select references to 'standard' DNA databases, else to
571 * 'standard' peptide databases
573 * a set of references to select from
576 public static DBRefEntry[] selectDbRefs(boolean selectDna,
579 return selectRefs(refs,
580 selectDna ? DBRefSource.DNACODINGDBS : DBRefSource.PROTEINDBS);
581 // could attempt to find other cross
582 // refs here - ie PDB xrefs
583 // (not dna, not protein seq)
587 * Returns the (possibly empty) list of those supplied dbrefs which have the
588 * specified source database, with a case-insensitive match of source name
594 public static List<DBRefEntry> searchRefsForSource(DBRefEntry[] dbRefs,
597 List<DBRefEntry> matches = new ArrayList<DBRefEntry>();
598 if (dbRefs != null && source != null)
600 for (DBRefEntry dbref : dbRefs)
602 if (source.equalsIgnoreCase(dbref.getSource()))
612 * promote direct database references to primary for nucleotide or protein
613 * sequences if they have an appropriate primary ref
617 * <th>Primary DB</th>
618 * <th>Direct which will be promoted</th>
639 public static void ensurePrimaries(SequenceI sequence)
641 List<DBRefEntry> pr = sequence.getPrimaryDBRefs();
647 List<DBRefEntry> selfs = new ArrayList<DBRefEntry>();
649 DBRefEntry[] selfArray = selectDbRefs(!sequence.isProtein(),
650 sequence.getDBRefs());
651 if (selfArray == null || selfArray.length == 0)
656 selfs.addAll(Arrays.asList(selfArray));
659 // filter non-primary refs
660 for (DBRefEntry p : pr)
662 while (selfs.contains(p))
667 List<DBRefEntry> toPromote = new ArrayList<DBRefEntry>();
669 for (DBRefEntry p : pr)
671 List<String> promType = new ArrayList<String>();
672 if (sequence.isProtein())
674 switch (getCanonicalName(p.getSource()))
676 case DBRefSource.UNIPROT:
677 // case DBRefSource.UNIPROTKB:
678 // case DBRefSource.UP_NAME:
679 // search for and promote ensembl
680 promType.add(DBRefSource.ENSEMBL);
682 case DBRefSource.ENSEMBL:
683 // search for and promote Uniprot
684 promType.add(DBRefSource.UNIPROT);
690 // TODO: promote transcript refs
693 // collate candidates and promote them
694 DBRefEntry[] candidates = selectRefs(selfs.toArray(new DBRefEntry[0]),
695 promType.toArray(new String[0]));
696 if (candidates != null)
698 for (DBRefEntry cand : candidates)
702 if (cand.getMap().getTo() != null
703 && cand.getMap().getTo() != sequence)
705 // can't promote refs with mappings to other sequences
708 if (cand.getMap().getMap().getFromLowest() != sequence
710 && cand.getMap().getMap().getFromHighest() != sequence
713 // can't promote refs with mappings from a region of this sequence
719 cand.setVersion(p.getVersion() + " (promoted)");
722 if (!cand.isPrimaryCandidate())
725 "Warning: Couldn't promote dbref " + cand.toString()
726 + " for sequence " + sequence.toString());