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.ws.sifts;
23 import java.util.Locale;
26 import java.io.FileInputStream;
27 import java.io.FileOutputStream;
28 import java.io.IOException;
29 import java.io.InputStream;
30 import java.io.PrintStream;
32 import java.net.URLConnection;
33 import java.nio.file.Files;
34 import java.nio.file.Path;
35 import java.nio.file.attribute.BasicFileAttributes;
36 import java.util.ArrayList;
37 import java.util.Arrays;
38 import java.util.Collection;
39 import java.util.Collections;
40 import java.util.Date;
41 import java.util.HashMap;
42 import java.util.HashSet;
43 import java.util.List;
46 import java.util.TreeMap;
47 import java.util.zip.GZIPInputStream;
49 import javax.xml.bind.JAXBContext;
50 import javax.xml.bind.JAXBElement;
51 import javax.xml.bind.Unmarshaller;
52 import javax.xml.stream.XMLInputFactory;
53 import javax.xml.stream.XMLStreamReader;
55 import jalview.analysis.AlignSeq;
56 import jalview.analysis.scoremodels.ScoreMatrix;
57 import jalview.analysis.scoremodels.ScoreModels;
58 import jalview.api.DBRefEntryI;
59 import jalview.api.SiftsClientI;
60 import jalview.bin.Console;
61 import jalview.datamodel.DBRefEntry;
62 import jalview.datamodel.DBRefSource;
63 import jalview.datamodel.SequenceI;
64 import jalview.io.BackupFiles;
65 import jalview.io.StructureFile;
66 import jalview.schemes.ResidueProperties;
67 import jalview.structure.StructureMapping;
68 import jalview.util.Comparison;
69 import jalview.util.DBRefUtils;
70 import jalview.util.Format;
71 import jalview.util.Platform;
72 import jalview.xml.binding.sifts.Entry;
73 import jalview.xml.binding.sifts.Entry.Entity;
74 import jalview.xml.binding.sifts.Entry.Entity.Segment;
75 import jalview.xml.binding.sifts.Entry.Entity.Segment.ListMapRegion.MapRegion;
76 import jalview.xml.binding.sifts.Entry.Entity.Segment.ListResidue.Residue;
77 import jalview.xml.binding.sifts.Entry.Entity.Segment.ListResidue.Residue.CrossRefDb;
78 import jalview.xml.binding.sifts.Entry.Entity.Segment.ListResidue.Residue.ResidueDetail;
80 import mc_view.PDBChain;
82 public class SiftsClient implements SiftsClientI
85 * for use in mocking out file fetch for tests only
86 * - reset to null after testing!
88 private static File mockSiftsFile;
90 private Entry siftsEntry;
92 private StructureFile pdb;
96 private String structId;
98 private CoordinateSys seqCoordSys = CoordinateSys.UNIPROT;
101 * PDB sequence position to sequence coordinate mapping as derived from SIFTS
102 * record for the identified SeqCoordSys Used for lift-over from sequence
103 * derived from PDB (with first extracted PDBRESNUM as 'start' to the sequence
104 * being annotated with PDB data
106 private jalview.datamodel.Mapping seqFromPdbMapping;
108 private static final int BUFFER_SIZE = 4096;
110 public static final int UNASSIGNED = Integer.MIN_VALUE;
112 private static final int PDB_RES_POS = 0;
114 private static final int PDB_ATOM_POS = 1;
116 private static final int PDBE_POS = 2;
118 private static final String NOT_OBSERVED = "Not_Observed";
120 private static final String SIFTS_SPLIT_FTP_BASE_URL = "https://ftp.ebi.ac.uk/pub/databases/msd/sifts/split_xml/";
122 private final static String NEWLINE = System.lineSeparator();
124 private String curSourceDBRef;
126 private HashSet<String> curDBRefAccessionIdsString;
128 private enum CoordinateSys
130 UNIPROT("UniProt"), PDB("PDBresnum"), PDBe("PDBe");
134 private CoordinateSys(String name)
139 public String getName()
145 private enum ResidueDetailType
147 NAME_SEC_STRUCTURE("nameSecondaryStructure"),
148 CODE_SEC_STRUCTURE("codeSecondaryStructure"), ANNOTATION("Annotation");
152 private ResidueDetailType(String code)
157 public String getCode()
164 * Fetch SIFTs file for the given PDBfile and construct an instance of
168 * @throws SiftsException
170 public SiftsClient(StructureFile pdb) throws SiftsException
173 this.pdbId = pdb.getId();
174 File siftsFile = getSiftsFile(pdbId);
175 siftsEntry = parseSIFTs(siftsFile);
179 * Parse the given SIFTs File and return a JAXB POJO of parsed data
182 * - the GZipped SIFTs XML file to parse
185 * if a problem occurs while parsing the SIFTs XML
187 private Entry parseSIFTs(File siftFile) throws SiftsException
189 try (InputStream in = new FileInputStream(siftFile);
190 GZIPInputStream gzis = new GZIPInputStream(in);)
192 // jalview.bin.Console.outPrintln("File : " + siftFile.getAbsolutePath());
193 JAXBContext jc = JAXBContext.newInstance("jalview.xml.binding.sifts");
194 XMLStreamReader streamReader = XMLInputFactory.newInstance()
195 .createXMLStreamReader(gzis);
196 Unmarshaller um = jc.createUnmarshaller();
197 JAXBElement<Entry> jbe = um.unmarshal(streamReader, Entry.class);
198 return jbe.getValue();
199 } catch (Exception e)
202 throw new SiftsException(e.getMessage());
207 * Get a SIFTs XML file for a given PDB Id from Cache or download from FTP
208 * repository if not found in cache
211 * @return SIFTs XML file
212 * @throws SiftsException
214 public static File getSiftsFile(String pdbId) throws SiftsException
217 * return mocked file if it has been set
219 if (mockSiftsFile != null)
221 return mockSiftsFile;
224 String siftsFileName = SiftsSettings.getSiftDownloadDirectory()
225 + pdbId.toLowerCase(Locale.ROOT) + ".xml.gz";
226 File siftsFile = new File(siftsFileName);
227 if (siftsFile.exists())
229 // The line below is required for unit testing... don't comment it out!!!
230 jalview.bin.Console.outPrintln(">>> SIFTS File already downloaded for " + pdbId);
232 if (isFileOlderThanThreshold(siftsFile,
233 SiftsSettings.getCacheThresholdInDays()))
235 File oldSiftsFile = new File(siftsFileName + "_old");
236 BackupFiles.moveFileToFile(siftsFile, oldSiftsFile);
239 siftsFile = downloadSiftsFile(pdbId.toLowerCase(Locale.ROOT));
240 oldSiftsFile.delete();
242 } catch (IOException e)
245 BackupFiles.moveFileToFile(oldSiftsFile, siftsFile);
246 return new File(siftsFileName);
256 siftsFile = downloadSiftsFile(pdbId.toLowerCase(Locale.ROOT));
257 } catch (IOException e)
259 throw new SiftsException(e.getMessage());
265 * This method enables checking if a cached file has exceeded a certain
271 * the threshold in days
274 public static boolean isFileOlderThanThreshold(File file, int noOfDays)
276 Path filePath = file.toPath();
277 BasicFileAttributes attr;
281 attr = Files.readAttributes(filePath, BasicFileAttributes.class);
282 diffInDays = (int) ((new Date().getTime()
283 - attr.lastModifiedTime().toMillis())
284 / (1000 * 60 * 60 * 24));
285 // jalview.bin.Console.outPrintln("Diff in days : " + diffInDays);
286 } catch (IOException e)
290 return noOfDays <= diffInDays;
294 * Download a SIFTs XML file for a given PDB Id from an FTP repository
297 * @return downloaded SIFTs XML file
298 * @throws SiftsException
299 * @throws IOException
301 public static File downloadSiftsFile(String pdbId)
302 throws SiftsException, IOException
304 if (pdbId.contains(".cif"))
306 pdbId = pdbId.replace(".cif", "");
308 String siftFile = pdbId + ".xml.gz";
309 String siftsFileFTPURL = getDownloadUrlFor(siftFile);
312 * Download the file from URL to either
313 * Java: directory of cached downloaded SIFTS files
314 * Javascript: temporary 'file' (in-memory cache)
316 File downloadTo = null;
319 downloadTo = File.createTempFile(siftFile, ".xml.gz");
323 downloadTo = new File(
324 SiftsSettings.getSiftDownloadDirectory() + siftFile);
325 File siftsDownloadDir = new File(
326 SiftsSettings.getSiftDownloadDirectory());
327 if (!siftsDownloadDir.exists())
329 siftsDownloadDir.mkdirs();
333 // jalview.bin.Console.outPrintln(">> Download ftp url : " + siftsFileFTPURL);
334 // long now = System.currentTimeMillis();
335 URL url = new URL(siftsFileFTPURL);
336 URLConnection conn = url.openConnection();
337 InputStream inputStream = conn.getInputStream();
338 FileOutputStream outputStream = new FileOutputStream(downloadTo);
339 byte[] buffer = new byte[BUFFER_SIZE];
341 while ((bytesRead = inputStream.read(buffer)) != -1)
343 outputStream.write(buffer, 0, bytesRead);
345 outputStream.close();
347 // jalview.bin.Console.outPrintln(">>> File downloaded : " + downloadedSiftsFile
348 // + " took " + (System.currentTimeMillis() - now) + "ms");
352 public static String getDownloadUrlFor(String siftFile)
354 String durl = SIFTS_SPLIT_FTP_BASE_URL+siftFile.substring(1, 3)+"/"+siftFile;
355 Console.trace("SIFTS URL for "+siftFile+" is "+durl);
361 * Delete the SIFTs file for the given PDB Id in the local SIFTs download
365 * @return true if the file was deleted or doesn't exist
367 public static boolean deleteSiftsFileByPDBId(String pdbId)
369 File siftsFile = new File(SiftsSettings.getSiftDownloadDirectory()
370 + pdbId.toLowerCase(Locale.ROOT) + ".xml.gz");
371 if (siftsFile.exists())
373 return siftsFile.delete();
379 * Get a valid SIFTs DBRef for the given sequence current SIFTs entry
382 * - the target sequence for the operation
383 * @return a valid DBRefEntry that is SIFTs compatible
385 * if no valid source DBRefEntry was found for the given sequences
387 public DBRefEntryI getValidSourceDBRef(SequenceI seq)
388 throws SiftsException
390 List<DBRefEntry> dbRefs = seq.getPrimaryDBRefs();
391 if (dbRefs == null || dbRefs.size() < 1)
393 throw new SiftsException(
394 "Source DBRef could not be determined. DBRefs might not have been retrieved.");
397 for (DBRefEntry dbRef : dbRefs)
399 if (dbRef == null || dbRef.getAccessionId() == null
400 || dbRef.getSource() == null)
404 String canonicalSource = DBRefUtils
405 .getCanonicalName(dbRef.getSource());
406 if (isValidDBRefEntry(dbRef)
407 && (canonicalSource.equalsIgnoreCase(DBRefSource.UNIPROT)
408 || canonicalSource.equalsIgnoreCase(DBRefSource.PDB)))
413 throw new SiftsException("Could not get source DB Ref");
417 * Check that the DBRef Entry is properly populated and is available in this
418 * SiftClient instance
421 * - DBRefEntry to validate
422 * @return true validation is successful otherwise false is returned.
424 boolean isValidDBRefEntry(DBRefEntryI entry)
426 return entry != null && entry.getAccessionId() != null
427 && isFoundInSiftsEntry(entry.getAccessionId());
431 public HashSet<String> getAllMappingAccession()
433 HashSet<String> accessions = new HashSet<String>();
434 List<Entity> entities = siftsEntry.getEntity();
435 for (Entity entity : entities)
437 List<Segment> segments = entity.getSegment();
438 for (Segment segment : segments)
440 List<MapRegion> mapRegions = segment.getListMapRegion()
442 for (MapRegion mapRegion : mapRegions)
444 accessions.add(mapRegion.getDb().getDbAccessionId()
445 .toLowerCase(Locale.ROOT));
453 public StructureMapping getSiftsStructureMapping(SequenceI seq,
454 String pdbFile, String chain) throws SiftsException
456 SequenceI aseq = seq;
457 while (seq.getDatasetSequence() != null)
459 seq = seq.getDatasetSequence();
461 structId = (chain == null) ? pdbId : pdbId + "|" + chain;
462 jalview.bin.Console.outPrintln("Getting SIFTS mapping for " + structId + ": seq "
465 final StringBuilder mappingDetails = new StringBuilder(128);
466 PrintStream ps = new PrintStream(System.out)
469 public void print(String x)
471 mappingDetails.append(x);
475 public void println()
477 mappingDetails.append(NEWLINE);
480 HashMap<Integer, int[]> mapping = getGreedyMapping(chain, seq, ps);
482 String mappingOutput = mappingDetails.toString();
483 StructureMapping siftsMapping = new StructureMapping(aseq, pdbFile,
484 pdbId, chain, mapping, mappingOutput, seqFromPdbMapping);
490 public HashMap<Integer, int[]> getGreedyMapping(String entityId,
491 SequenceI seq, java.io.PrintStream os) throws SiftsException
493 List<Integer> omitNonObserved = new ArrayList<>();
494 int nonObservedShiftIndex = 0, pdbeNonObserved = 0;
495 // jalview.bin.Console.outPrintln("Generating mappings for : " + entityId);
496 Entity entity = null;
497 entity = getEntityById(entityId);
498 String originalSeq = AlignSeq.extractGaps(
499 jalview.util.Comparison.GapChars, seq.getSequenceAsString());
500 HashMap<Integer, int[]> mapping = new HashMap<Integer, int[]>();
501 DBRefEntryI sourceDBRef;
502 sourceDBRef = getValidSourceDBRef(seq);
503 // TODO ensure sequence start/end is in the same coordinate system and
504 // consistent with the choosen sourceDBRef
506 // set sequence coordinate system - default value is UniProt
507 if (sourceDBRef.getSource().equalsIgnoreCase(DBRefSource.PDB))
509 seqCoordSys = CoordinateSys.PDB;
512 HashSet<String> dbRefAccessionIdsString = new HashSet<String>();
513 for (DBRefEntry dbref : seq.getDBRefs())
515 dbRefAccessionIdsString
516 .add(dbref.getAccessionId().toLowerCase(Locale.ROOT));
518 dbRefAccessionIdsString
519 .add(sourceDBRef.getAccessionId().toLowerCase(Locale.ROOT));
521 curDBRefAccessionIdsString = dbRefAccessionIdsString;
522 curSourceDBRef = sourceDBRef.getAccessionId();
524 TreeMap<Integer, String> resNumMap = new TreeMap<Integer, String>();
525 List<Segment> segments = entity.getSegment();
526 SegmentHelperPojo shp = new SegmentHelperPojo(seq, mapping, resNumMap,
527 omitNonObserved, nonObservedShiftIndex, pdbeNonObserved);
528 processSegments(segments, shp);
531 populateAtomPositions(entityId, mapping);
532 } catch (Exception e)
536 if (seqCoordSys == CoordinateSys.UNIPROT)
538 padWithGaps(resNumMap, omitNonObserved);
540 int seqStart = UNASSIGNED;
541 int seqEnd = UNASSIGNED;
542 int pdbStart = UNASSIGNED;
543 int pdbEnd = UNASSIGNED;
545 if (mapping.isEmpty())
547 throw new SiftsException("SIFTS mapping failed for "+entityId+" and "+seq.getName());
549 // also construct a mapping object between the seq-coord sys and the PDB
552 Integer[] keys = mapping.keySet().toArray(new Integer[0]);
555 seqEnd = keys[keys.length - 1];
556 List<int[]> from = new ArrayList<>(), to = new ArrayList<>();
557 int[] _cfrom = null, _cto = null;
558 String matchedSeq = originalSeq;
559 if (seqStart != UNASSIGNED) // fixme! seqStart can map to -1 for a pdb
560 // sequence that starts <-1
562 for (int seqps : keys)
564 int pdbpos = mapping.get(seqps)[PDBE_POS];
565 if (pdbpos == UNASSIGNED)
567 // not correct - pdbpos might be -1, but leave it for now
570 if (_cfrom == null || seqps != _cfrom[1] + 1)
572 _cfrom = new int[] { seqps, seqps };
574 _cto = null; // discontinuity
580 if (_cto == null || pdbpos != 1 + _cto[1])
582 _cto = new int[] { pdbpos, pdbpos };
590 _cfrom = new int[from.size() * 2];
591 _cto = new int[to.size() * 2];
593 for (int[] range : from)
595 _cfrom[p++] = range[0];
596 _cfrom[p++] = range[1];
600 for (int[] range : to)
602 _cto[p++] = range[0];
603 _cto[p++] = range[1];
607 seqFromPdbMapping = new jalview.datamodel.Mapping(null, _cto, _cfrom,
609 pdbStart = mapping.get(seqStart)[PDB_RES_POS];
610 pdbEnd = mapping.get(seqEnd)[PDB_RES_POS];
611 int orignalSeqStart = seq.getStart();
612 if (orignalSeqStart >= 1)
614 int subSeqStart = (seqStart >= orignalSeqStart)
615 ? seqStart - orignalSeqStart
617 int subSeqEnd = seqEnd - (orignalSeqStart - 1);
618 subSeqEnd = originalSeq.length() < subSeqEnd ? originalSeq.length()
620 matchedSeq = originalSeq.substring(subSeqStart, subSeqEnd);
624 matchedSeq = originalSeq.substring(1, originalSeq.length());
628 StringBuilder targetStrucSeqs = new StringBuilder();
629 for (String res : resNumMap.values())
631 targetStrucSeqs.append(res);
636 MappingOutputPojo mop = new MappingOutputPojo();
637 mop.setSeqStart(seqStart);
638 mop.setSeqEnd(seqEnd);
639 mop.setSeqName(seq.getName());
640 mop.setSeqResidue(matchedSeq);
642 mop.setStrStart(pdbStart);
643 mop.setStrEnd(pdbEnd);
644 mop.setStrName(structId);
645 mop.setStrResidue(targetStrucSeqs.toString());
648 os.print(getMappingOutput(mop).toString());
654 void processSegments(List<Segment> segments, SegmentHelperPojo shp)
656 SequenceI seq = shp.getSeq();
657 HashMap<Integer, int[]> mapping = shp.getMapping();
658 TreeMap<Integer, String> resNumMap = shp.getResNumMap();
659 List<Integer> omitNonObserved = shp.getOmitNonObserved();
660 int nonObservedShiftIndex = shp.getNonObservedShiftIndex();
661 int pdbeNonObservedCount = shp.getPdbeNonObserved();
662 int firstPDBResNum = UNASSIGNED;
663 for (Segment segment : segments)
665 // jalview.bin.Console.outPrintln("Mapping segments : " + segment.getSegId() + "\\"s
667 List<Residue> residues = segment.getListResidue().getResidue();
668 for (Residue residue : residues)
670 boolean isObserved = isResidueObserved(residue);
671 int pdbeIndex = getLeadingIntegerValue(residue.getDbResNum(),
673 int currSeqIndex = UNASSIGNED;
674 List<CrossRefDb> cRefDbs = residue.getCrossRefDb();
675 CrossRefDb pdbRefDb = null;
676 for (CrossRefDb cRefDb : cRefDbs)
678 if (cRefDb.getDbSource().equalsIgnoreCase(DBRefSource.PDB))
681 if (firstPDBResNum == UNASSIGNED)
683 firstPDBResNum = getLeadingIntegerValue(cRefDb.getDbResNum(),
690 // after we find the first observed residue we just increment
695 if (cRefDb.getDbCoordSys().equalsIgnoreCase(seqCoordSys.getName())
696 && isAccessionMatched(cRefDb.getDbAccessionId()))
698 currSeqIndex = getLeadingIntegerValue(cRefDb.getDbResNum(),
700 if (pdbRefDb != null)
702 break;// exit loop if pdb and uniprot are already found
708 ++pdbeNonObservedCount;
710 if (seqCoordSys == seqCoordSys.PDB) // FIXME: is seqCoordSys ever PDBe
713 // if the sequence has a primary reference to the PDB, then we are
714 // dealing with a sequence extracted directly from the PDB. In that
715 // case, numbering is PDBe - non-observed residues
716 currSeqIndex = seq.getStart() - 1 + pdbeIndex;
720 if (seqCoordSys != CoordinateSys.UNIPROT) // FIXME: PDB or PDBe only
723 // mapping to PDB or PDBe so we need to bookkeep for the
726 omitNonObserved.add(currSeqIndex);
727 ++nonObservedShiftIndex;
730 if (currSeqIndex == UNASSIGNED)
732 // change in logic - unobserved residues with no currSeqIndex
733 // corresponding are still counted in both nonObservedShiftIndex and
737 // if (currSeqIndex >= seq.getStart() && currSeqIndex <= seqlength) //
747 int resNum = (pdbRefDb == null)
748 ? getLeadingIntegerValue(residue.getDbResNum(),
750 : getLeadingIntegerValue(pdbRefDb.getDbResNum(),
755 char resCharCode = ResidueProperties
756 .getSingleCharacterCode(ResidueProperties
757 .getCanonicalAminoAcid(residue.getDbResName()));
758 resNumMap.put(currSeqIndex, String.valueOf(resCharCode));
760 int[] mappingcols = new int[] { Integer.valueOf(resNum),
761 UNASSIGNED, isObserved ? firstPDBResNum : UNASSIGNED };
763 mapping.put(currSeqIndex - nonObservedShiftIndex, mappingcols);
771 * Get the leading integer part of a string that begins with an integer.
774 * - the string input to process
776 * - value returned if unsuccessful
779 static int getLeadingIntegerValue(String input, int failValue)
785 String[] parts = input.split("(?=\\D)(?<=\\d)");
786 if (parts != null && parts.length > 0 && parts[0].matches("[0-9]+"))
788 return Integer.valueOf(parts[0]);
796 * Target chain to populate mapping of its atom positions.
798 * Two dimension array of residue index versus atom position
799 * @throws IllegalArgumentException
800 * Thrown if chainId or mapping is null
801 * @throws SiftsException
803 void populateAtomPositions(String chainId, Map<Integer, int[]> mapping)
804 throws IllegalArgumentException, SiftsException
808 PDBChain chain = pdb.findChain(chainId);
810 if (chain == null || mapping == null)
812 throw new IllegalArgumentException(
813 "Chain id or mapping must not be null.");
815 for (int[] map : mapping.values())
817 if (map[PDB_RES_POS] != UNASSIGNED)
819 map[PDB_ATOM_POS] = getAtomIndex(map[PDB_RES_POS], chain.atoms);
822 } catch (NullPointerException e)
824 throw new SiftsException(e.getMessage());
825 } catch (Exception e)
827 throw new SiftsException(e.getMessage());
833 * @param residueIndex
834 * The residue index used for the search
836 * A collection of Atom to search
837 * @return atom position for the given residue index
839 int getAtomIndex(int residueIndex, Collection<Atom> atoms)
843 throw new IllegalArgumentException(
844 "atoms collection must not be null!");
846 for (Atom atom : atoms)
848 if (atom.resNumber == residueIndex)
850 return atom.atomIndex;
857 * Checks if the residue instance is marked 'Not_observed' or not
862 private boolean isResidueObserved(Residue residue)
864 Set<String> annotations = getResidueAnnotaitons(residue,
865 ResidueDetailType.ANNOTATION);
866 if (annotations == null || annotations.isEmpty())
870 for (String annotation : annotations)
872 if (annotation.equalsIgnoreCase(NOT_OBSERVED))
881 * Get annotation String for a given residue and annotation type
887 private Set<String> getResidueAnnotaitons(Residue residue,
888 ResidueDetailType type)
890 HashSet<String> foundAnnotations = new HashSet<String>();
891 List<ResidueDetail> resDetails = residue.getResidueDetail();
892 for (ResidueDetail resDetail : resDetails)
894 if (resDetail.getProperty().equalsIgnoreCase(type.getCode()))
896 foundAnnotations.add(resDetail.getContent());
899 return foundAnnotations;
903 public boolean isAccessionMatched(String accession)
905 boolean isStrictMatch = true;
906 return isStrictMatch ? curSourceDBRef.equalsIgnoreCase(accession)
907 : curDBRefAccessionIdsString
908 .contains(accession.toLowerCase(Locale.ROOT));
911 private boolean isFoundInSiftsEntry(String accessionId)
913 Set<String> siftsDBRefs = getAllMappingAccession();
914 return accessionId != null
915 && siftsDBRefs.contains(accessionId.toLowerCase(Locale.ROOT));
919 * Pad omitted residue positions in PDB sequence with gaps
923 void padWithGaps(Map<Integer, String> resNumMap,
924 List<Integer> omitNonObserved)
926 if (resNumMap == null || resNumMap.isEmpty())
930 Integer[] keys = resNumMap.keySet().toArray(new Integer[0]);
931 // Arrays.sort(keys);
932 int firstIndex = keys[0];
933 int lastIndex = keys[keys.length - 1];
934 // jalview.bin.Console.outPrintln("Min value " + firstIndex);
935 // jalview.bin.Console.outPrintln("Max value " + lastIndex);
936 for (int x = firstIndex; x <= lastIndex; x++)
938 if (!resNumMap.containsKey(x) && !omitNonObserved.contains(x))
940 resNumMap.put(x, "-");
946 public Entity getEntityById(String id) throws SiftsException
948 // Determines an entity to process by performing a heuristic matching of all
949 // Entities with the given chainId and choosing the best matching Entity
950 Entity entity = getEntityByMostOptimalMatchedId(id);
955 throw new SiftsException("Entity " + id + " not found");
959 * This method was added because EntityId is NOT always equal to ChainId.
960 * Hence, it provides the logic to greedily detect the "true" Entity for a
961 * given chainId where discrepancies exist.
966 public Entity getEntityByMostOptimalMatchedId(String chainId)
968 // jalview.bin.Console.outPrintln("---> advanced greedy entityId matching block
970 List<Entity> entities = siftsEntry.getEntity();
971 SiftsEntitySortPojo[] sPojo = new SiftsEntitySortPojo[entities.size()];
973 for (Entity entity : entities)
975 sPojo[count] = new SiftsEntitySortPojo();
976 sPojo[count].entityId = entity.getEntityId();
978 List<Segment> segments = entity.getSegment();
979 for (Segment segment : segments)
981 List<Residue> residues = segment.getListResidue().getResidue();
982 for (Residue residue : residues)
984 List<CrossRefDb> cRefDbs = residue.getCrossRefDb();
985 for (CrossRefDb cRefDb : cRefDbs)
987 if (!cRefDb.getDbSource().equalsIgnoreCase("PDB"))
991 ++sPojo[count].resCount;
992 if (cRefDb.getDbChainId().equalsIgnoreCase(chainId))
994 ++sPojo[count].chainIdFreq;
999 sPojo[count].pid = (100 * sPojo[count].chainIdFreq)
1000 / sPojo[count].resCount;
1003 Arrays.sort(sPojo, Collections.reverseOrder());
1004 // jalview.bin.Console.outPrintln("highest matched entity : " + sPojo[0].entityId);
1005 // jalview.bin.Console.outPrintln("highest matched pid : " + sPojo[0].pid);
1007 if (sPojo[0].entityId != null)
1009 if (sPojo[0].pid < 1)
1013 for (Entity entity : entities)
1015 if (!entity.getEntityId().equalsIgnoreCase(sPojo[0].entityId))
1025 private class SiftsEntitySortPojo
1026 implements Comparable<SiftsEntitySortPojo>
1028 public String entityId;
1030 public int chainIdFreq;
1034 public int resCount;
1037 public int compareTo(SiftsEntitySortPojo o)
1039 return this.pid - o.pid;
1043 private class SegmentHelperPojo
1045 private SequenceI seq;
1047 private HashMap<Integer, int[]> mapping;
1049 private TreeMap<Integer, String> resNumMap;
1051 private List<Integer> omitNonObserved;
1053 private int nonObservedShiftIndex;
1056 * count of number of 'not observed' positions in the PDB record's SEQRES
1057 * (total number of residues with coordinates == length(SEQRES) -
1060 private int pdbeNonObserved;
1062 public SegmentHelperPojo(SequenceI seq, HashMap<Integer, int[]> mapping,
1063 TreeMap<Integer, String> resNumMap,
1064 List<Integer> omitNonObserved, int nonObservedShiftIndex,
1065 int pdbeNonObserved)
1068 setMapping(mapping);
1069 setResNumMap(resNumMap);
1070 setOmitNonObserved(omitNonObserved);
1071 setNonObservedShiftIndex(nonObservedShiftIndex);
1072 setPdbeNonObserved(pdbeNonObserved);
1076 public void setPdbeNonObserved(int pdbeNonObserved2)
1078 this.pdbeNonObserved = pdbeNonObserved2;
1081 public int getPdbeNonObserved()
1083 return pdbeNonObserved;
1086 public SequenceI getSeq()
1091 public void setSeq(SequenceI seq)
1096 public HashMap<Integer, int[]> getMapping()
1101 public void setMapping(HashMap<Integer, int[]> mapping)
1103 this.mapping = mapping;
1106 public TreeMap<Integer, String> getResNumMap()
1111 public void setResNumMap(TreeMap<Integer, String> resNumMap)
1113 this.resNumMap = resNumMap;
1116 public List<Integer> getOmitNonObserved()
1118 return omitNonObserved;
1121 public void setOmitNonObserved(List<Integer> omitNonObserved)
1123 this.omitNonObserved = omitNonObserved;
1126 public int getNonObservedShiftIndex()
1128 return nonObservedShiftIndex;
1131 public void setNonObservedShiftIndex(int nonObservedShiftIndex)
1133 this.nonObservedShiftIndex = nonObservedShiftIndex;
1139 public StringBuilder getMappingOutput(MappingOutputPojo mp)
1140 throws SiftsException
1142 String seqRes = mp.getSeqResidue();
1143 String seqName = mp.getSeqName();
1144 int sStart = mp.getSeqStart();
1145 int sEnd = mp.getSeqEnd();
1147 String strRes = mp.getStrResidue();
1148 String strName = mp.getStrName();
1149 int pdbStart = mp.getStrStart();
1150 int pdbEnd = mp.getStrEnd();
1152 String type = mp.getType();
1154 int maxid = (seqName.length() >= strName.length()) ? seqName.length()
1156 int len = 72 - maxid - 1;
1158 int nochunks = ((seqRes.length()) / len)
1159 + ((seqRes.length()) % len > 0 ? 1 : 0);
1161 StringBuilder output = new StringBuilder(512);
1162 output.append(NEWLINE);
1163 output.append("Sequence \u27f7 Structure mapping details")
1165 output.append("Method: SIFTS");
1166 output.append(NEWLINE).append(NEWLINE);
1168 output.append(new Format("%" + maxid + "s").form(seqName));
1169 output.append(" : ");
1170 output.append(String.valueOf(sStart));
1171 output.append(" - ");
1172 output.append(String.valueOf(sEnd));
1173 output.append(" Maps to ");
1174 output.append(NEWLINE);
1175 output.append(new Format("%" + maxid + "s").form(structId));
1176 output.append(" : ");
1177 output.append(String.valueOf(pdbStart));
1178 output.append(" - ");
1179 output.append(String.valueOf(pdbEnd));
1180 output.append(NEWLINE).append(NEWLINE);
1182 ScoreMatrix pam250 = ScoreModels.getInstance().getPam250();
1183 int matchedSeqCount = 0;
1184 for (int j = 0; j < nochunks; j++)
1186 // Print the first aligned sequence
1187 output.append(new Format("%" + (maxid) + "s").form(seqName))
1190 for (int i = 0; i < len; i++)
1192 if ((i + (j * len)) < seqRes.length())
1194 output.append(seqRes.charAt(i + (j * len)));
1198 output.append(NEWLINE);
1199 output.append(new Format("%" + (maxid) + "s").form(" ")).append(" ");
1202 * Print out the match symbols:
1203 * | for exact match (ignoring case)
1204 * . if PAM250 score is positive
1207 for (int i = 0; i < len; i++)
1211 if ((i + (j * len)) < seqRes.length())
1213 char c1 = seqRes.charAt(i + (j * len));
1214 char c2 = strRes.charAt(i + (j * len));
1215 boolean sameChar = Comparison.isSameResidue(c1, c2, false);
1216 if (sameChar && !Comparison.isGap(c1))
1221 else if (type.equals("pep"))
1223 if (pam250.getPairwiseScore(c1, c2) > 0)
1237 } catch (IndexOutOfBoundsException e)
1242 // Now print the second aligned sequence
1243 output = output.append(NEWLINE);
1244 output = output.append(new Format("%" + (maxid) + "s").form(strName))
1246 for (int i = 0; i < len; i++)
1248 if ((i + (j * len)) < strRes.length())
1250 output.append(strRes.charAt(i + (j * len)));
1253 output.append(NEWLINE).append(NEWLINE);
1255 float pid = (float) matchedSeqCount / seqRes.length() * 100;
1256 if (pid < SiftsSettings.getFailSafePIDThreshold())
1258 throw new SiftsException(">>> Low PID detected for SIFTs mapping...");
1260 output.append("Length of alignment = " + seqRes.length())
1262 output.append(new Format("Percentage ID = %2.2f").form(pid));
1267 public int getEntityCount()
1269 return siftsEntry.getEntity().size();
1273 public String getDbAccessionId()
1275 return siftsEntry.getDbAccessionId();
1279 public String getDbCoordSys()
1281 return siftsEntry.getDbCoordSys();
1285 public String getDbSource()
1287 return siftsEntry.getDbSource();
1291 public String getDbVersion()
1293 return siftsEntry.getDbVersion();
1296 public static void setMockSiftsFile(File file)
1298 mockSiftsFile = file;