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.
22 * This extension was written by Benjamin Schuster-Boeckler at sanger.ac.uk
26 import java.io.BufferedReader;
27 import java.io.FileReader;
28 import java.io.IOException;
29 import java.util.ArrayList;
30 import java.util.Enumeration;
31 import java.util.Hashtable;
32 import java.util.LinkedHashMap;
33 import java.util.List;
34 import java.util.Locale;
36 import java.util.Vector;
38 import com.stevesoft.pat.Regex;
40 import fr.orsay.lri.varna.exceptions.ExceptionUnmatchedClosingParentheses;
41 import fr.orsay.lri.varna.factories.RNAFactory;
42 import fr.orsay.lri.varna.models.rna.RNA;
43 import jalview.analysis.Rna;
44 import jalview.datamodel.AlignmentAnnotation;
45 import jalview.datamodel.AlignmentI;
46 import jalview.datamodel.Annotation;
47 import jalview.datamodel.DBRefEntry;
48 import jalview.datamodel.DBRefSource;
49 import jalview.datamodel.Mapping;
50 import jalview.datamodel.Sequence;
51 import jalview.datamodel.SequenceFeature;
52 import jalview.datamodel.SequenceI;
53 import jalview.schemes.ResidueProperties;
54 import jalview.util.Comparison;
55 import jalview.util.DBRefUtils;
56 import jalview.util.Format;
57 import jalview.util.MessageManager;
60 * This class is supposed to parse a Stockholm format file into Jalview There
61 * are TODOs in this class: we do not know what the database source and version
62 * is for the file when parsing the #GS= AC tag which associates accessions with
63 * sequences. Database references are also not parsed correctly: a separate
64 * reference string parser must be added to parse the database reference form
65 * into Jalview's local representation.
67 * @author bsb at sanger.ac.uk
68 * @author Natasha Shersnev (Dundee, UK) (Stockholm file writer)
69 * @author Lauren Lui (UCSC, USA) (RNA secondary structure annotation import as
71 * @author Anne Menard (Paris, FR) (VARNA parsing of Stockholm file data)
72 * @version 0.3 + jalview mods
75 public class StockholmFile extends AlignFile
77 private static final String ANNOTATION = "annotation";
79 // private static final Regex OPEN_PAREN = new Regex("(<|\\[)", "(");
81 // private static final Regex CLOSE_PAREN = new Regex("(>|\\])", ")");
83 public static final Regex DETECT_BRACKETS = new Regex(
84 "(<|>|\\[|\\]|\\(|\\)|\\{|\\})");
86 // WUSS extended symbols. Avoid ambiguity with protein SS annotations by using
88 public static final String RNASS_BRACKETS = "<>[](){}AaBbCcDdEeFfGgHhIiJjKkLlMmNnOoPpQqRrSsTtUuVvWwXxYyZz";
90 // use the following regex to decide an annotations (whole) line is NOT an RNA
91 // SS (it contains only E,H,e,h and other non-brace/non-alpha chars)
92 private static final Regex NOT_RNASS = new Regex(
93 "^[^<>[\\](){}ADFJ-RUVWYZadfj-ruvwyz]*$");
95 StringBuffer out; // output buffer
99 public StockholmFile()
104 * Creates a new StockholmFile object for output.
106 public StockholmFile(AlignmentI al)
111 public StockholmFile(String inFile, DataSourceType type)
117 public StockholmFile(FileParse source) throws IOException
123 public void initData()
129 * Parse a file in Stockholm format into Jalview's data model using VARNA
131 * @throws IOException
132 * If there is an error with the input file
134 public void parse_with_VARNA(java.io.File inFile) throws IOException
136 FileReader fr = null;
137 fr = new FileReader(inFile);
139 BufferedReader r = new BufferedReader(fr);
140 List<RNA> result = null;
143 result = RNAFactory.loadSecStrStockholm(r);
144 } catch (ExceptionUnmatchedClosingParentheses umcp)
146 errormessage = "Unmatched parentheses in annotation. Aborting ("
147 + umcp.getMessage() + ")";
148 throw new IOException(umcp);
150 // DEBUG System.out.println("this is the secondary scructure:"
152 SequenceI[] seqs = new SequenceI[result.size()];
154 for (int i = 0; i < result.size(); i++)
156 // DEBUG System.err.println("Processing i'th sequence in Stockholm file")
157 RNA current = result.get(i);
159 String seq = current.getSeq();
160 String rna = current.getStructDBN(true);
161 // DEBUG System.out.println(seq);
162 // DEBUG System.err.println(rna);
164 int end = seq.length() - 1;
165 id = safeName(getDataName());
166 seqs[i] = new Sequence(id, seq, begin, end);
167 String[] annot = new String[rna.length()];
168 Annotation[] ann = new Annotation[rna.length()];
169 for (int j = 0; j < rna.length(); j++)
171 annot[j] = rna.substring(j, j + 1);
175 for (int k = 0; k < rna.length(); k++)
177 ann[k] = new Annotation(annot[k], "",
178 Rna.getRNASecStrucState(annot[k]).charAt(0), 0f);
181 AlignmentAnnotation align = new AlignmentAnnotation("Sec. str.",
182 current.getID(), ann);
184 seqs[i].addAlignmentAnnotation(align);
185 seqs[i].setRNA(result.get(i));
186 this.annotations.addElement(align);
193 * Parse a file in Stockholm format into Jalview's data model. The file has to
194 * be passed at construction time
196 * @throws IOException
197 * If there is an error with the input file
200 public void parse() throws IOException
202 StringBuffer treeString = new StringBuffer();
203 String treeName = null;
204 // --------------- Variable Definitions -------------------
208 Hashtable seqAnn = new Hashtable(); // Sequence related annotations
209 LinkedHashMap<String, String> seqs = new LinkedHashMap<>();
210 Regex p, r, rend, s, x;
211 // Temporary line for processing RNA annotation
212 // String RNAannot = "";
214 // ------------------ Parsing File ----------------------
215 // First, we have to check that this file has STOCKHOLM format, i.e. the
216 // first line must match
218 r = new Regex("# STOCKHOLM ([\\d\\.]+)");
219 if (!r.search(nextLine()))
221 throw new IOException(
222 MessageManager.getString("exception.stockholm_invalid_format")
227 version = r.stringMatched(1);
229 // logger.debug("Stockholm version: " + version);
232 // We define some Regexes here that will be used regularily later
233 rend = new Regex("^\\s*\\/\\/"); // Find the end of an alignment
234 p = new Regex("(\\S+)\\/(\\d+)\\-(\\d+)"); // split sequence id in
236 s = new Regex("(\\S+)\\s+(\\S*)\\s+(.*)"); // Parses annotation subtype
237 r = new Regex("#=(G[FSRC]?)\\s+(.*)"); // Finds any annotation line
238 x = new Regex("(\\S+)\\s+(\\S+)"); // split id from sequence
240 // Convert all bracket types to parentheses (necessary for passing to VARNA)
241 Regex openparen = new Regex("(<|\\[)", "(");
242 Regex closeparen = new Regex("(>|\\])", ")");
244 // // Detect if file is RNA by looking for bracket types
245 // Regex detectbrackets = new Regex("(<|>|\\[|\\]|\\(|\\))");
252 openparen.optimize();
253 closeparen.optimize();
255 while ((line = nextLine()) != null)
257 if (line.length() == 0)
261 if (rend.search(line))
263 // End of the alignment, pass stuff back
264 this.noSeqs = seqs.size();
266 String dbsource = null;
267 Regex pf = new Regex("PF[0-9]{5}(.*)"); // Finds AC for Pfam
268 Regex rf = new Regex("RF[0-9]{5}(.*)"); // Finds AC for Rfam
269 if (getAlignmentProperty("AC") != null)
271 String dbType = getAlignmentProperty("AC").toString();
272 if (pf.search(dbType))
274 // PFAM Alignment - so references are typically from Uniprot
277 else if (rf.search(dbType))
282 // logger.debug("Number of sequences: " + this.noSeqs);
283 for (Map.Entry<String, String> skey : seqs.entrySet())
285 // logger.debug("Processing sequence " + acc);
286 String acc = skey.getKey();
287 String seq = skey.getValue();
288 if (maxLength < seq.length())
290 maxLength = seq.length();
296 * Retrieve hash of annotations for this accession Associate
297 * Annotation with accession
299 Hashtable accAnnotations = null;
301 if (seqAnn != null && seqAnn.containsKey(acc))
303 accAnnotations = (Hashtable) seqAnn.remove(acc);
304 // TODO: add structures to sequence
307 // Split accession in id and from/to
310 sid = p.stringMatched(1);
311 start = Integer.parseInt(p.stringMatched(2));
312 end = Integer.parseInt(p.stringMatched(3));
314 // logger.debug(sid + ", " + start + ", " + end);
316 Sequence seqO = new Sequence(sid, seq, start, end);
317 // Add Description (if any)
318 if (accAnnotations != null && accAnnotations.containsKey("DE"))
320 String desc = (String) accAnnotations.get("DE");
321 seqO.setDescription((desc == null) ? "" : desc);
323 // Add DB References (if any)
324 if (accAnnotations != null && accAnnotations.containsKey("DR"))
326 String dbr = (String) accAnnotations.get("DR");
327 if (dbr != null && dbr.indexOf(";") > -1)
329 String src = dbr.substring(0, dbr.indexOf(";"));
330 String acn = dbr.substring(dbr.indexOf(";") + 1);
331 jalview.util.DBRefUtils.parseToDbRef(seqO, src, "0", acn);
335 if (accAnnotations != null && accAnnotations.containsKey("AC"))
337 String dbr = (String) accAnnotations.get("AC");
340 // we could get very clever here - but for now - just try to
341 // guess accession type from type of sequence, source of alignment
345 guessDatabaseFor(seqO, dbr, dbsource);
347 // else - do what ? add the data anyway and prompt the user to
348 // specify what references these are ?
351 Hashtable features = null;
352 // We need to adjust the positions of all features to account for gaps
355 features = (Hashtable) accAnnotations.remove("features");
356 } catch (java.lang.NullPointerException e)
358 // loggerwarn("Getting Features for " + acc + ": " +
362 // if we have features
363 if (features != null)
365 int posmap[] = seqO.findPositionMap();
366 Enumeration i = features.keys();
367 while (i.hasMoreElements())
369 // TODO: parse out secondary structure annotation as annotation
371 // TODO: parse out scores as annotation row
372 // TODO: map coding region to core jalview feature types
373 String type = i.nextElement().toString();
374 Hashtable content = (Hashtable) features.remove(type);
376 // add alignment annotation for this feature
377 String key = type2id(type);
380 * have we added annotation rows for this type ?
382 boolean annotsAdded = false;
385 if (accAnnotations != null
386 && accAnnotations.containsKey(key))
388 Vector vv = (Vector) accAnnotations.get(key);
389 for (int ii = 0; ii < vv.size(); ii++)
392 AlignmentAnnotation an = (AlignmentAnnotation) vv
394 seqO.addAlignmentAnnotation(an);
400 Enumeration j = content.keys();
401 while (j.hasMoreElements())
403 String desc = j.nextElement().toString();
404 if (ANNOTATION.equals(desc) && annotsAdded)
406 // don't add features if we already added an annotation row
409 String ns = content.get(desc).toString();
410 char[] byChar = ns.toCharArray();
411 for (int k = 0; k < byChar.length; k++)
414 if (!(c == ' ' || c == '_' || c == '-' || c == '.')) // PFAM
421 int new_pos = posmap[k]; // look up nearest seqeunce
422 // position to this column
423 SequenceFeature feat = new SequenceFeature(type, desc,
424 new_pos, new_pos, null);
426 seqO.addSequenceFeature(feat);
436 // logger.debug("Adding seq " + acc + " from " + start + " to " + end
438 this.seqs.addElement(seqO);
440 return; // finished parsing this segment of source
442 else if (!r.search(line))
444 // System.err.println("Found sequence line: " + line);
446 // Split sequence in sequence and accession parts
449 // logger.error("Could not parse sequence line: " + line);
450 throw new IOException(MessageManager.formatMessage(
451 "exception.couldnt_parse_sequence_line", new String[]
454 String ns = seqs.get(x.stringMatched(1));
459 ns += x.stringMatched(2);
461 seqs.put(x.stringMatched(1), ns);
465 String annType = r.stringMatched(1);
466 String annContent = r.stringMatched(2);
468 // System.err.println("type:" + annType + " content: " + annContent);
470 if (annType.equals("GF"))
473 * Generic per-File annotation, free text Magic features: #=GF NH
474 * <tree in New Hampshire eXtended format> #=GF TN <Unique identifier
475 * for the next tree> Pfam descriptions: 7. DESCRIPTION OF FIELDS
477 * Compulsory fields: ------------------
479 * AC Accession number: Accession number in form PFxxxxx.version or
480 * PBxxxxxx. ID Identification: One word name for family. DE
481 * Definition: Short description of family. AU Author: Authors of the
482 * entry. SE Source of seed: The source suggesting the seed members
483 * belong to one family. GA Gathering method: Search threshold to
484 * build the full alignment. TC Trusted Cutoff: Lowest sequence score
485 * and domain score of match in the full alignment. NC Noise Cutoff:
486 * Highest sequence score and domain score of match not in full
487 * alignment. TP Type: Type of family -- presently Family, Domain,
488 * Motif or Repeat. SQ Sequence: Number of sequences in alignment. AM
489 * Alignment Method The order ls and fs hits are aligned to the model
490 * to build the full align. // End of alignment.
492 * Optional fields: ----------------
494 * DC Database Comment: Comment about database reference. DR Database
495 * Reference: Reference to external database. RC Reference Comment:
496 * Comment about literature reference. RN Reference Number: Reference
497 * Number. RM Reference Medline: Eight digit medline UI number. RT
498 * Reference Title: Reference Title. RA Reference Author: Reference
499 * Author RL Reference Location: Journal location. PI Previous
500 * identifier: Record of all previous ID lines. KW Keywords: Keywords.
501 * CC Comment: Comments. NE Pfam accession: Indicates a nested domain.
502 * NL Location: Location of nested domains - sequence ID, start and
505 * Obsolete fields: ----------- AL Alignment method of seed: The
506 * method used to align the seed members.
508 // Let's save the annotations, maybe we'll be able to do something
509 // with them later...
510 Regex an = new Regex("(\\w+)\\s*(.*)");
511 if (an.search(annContent))
513 if (an.stringMatched(1).equals("NH"))
515 treeString.append(an.stringMatched(2));
517 else if (an.stringMatched(1).equals("TN"))
519 if (treeString.length() > 0)
521 if (treeName == null)
523 treeName = "Tree " + (getTreeCount() + 1);
525 addNewickTree(treeName, treeString.toString());
527 treeName = an.stringMatched(2);
528 treeString = new StringBuffer();
530 // TODO: JAL-3532 - this is where GF comments and database
531 // references are lost
532 // suggest overriding this method for Stockholm files to catch and
534 // process CC, DR etc into multivalued properties
535 setAlignmentProperty(an.stringMatched(1), an.stringMatched(2));
538 else if (annType.equals("GS"))
540 // Generic per-Sequence annotation, free text
542 * Pfam uses these features: Feature Description ---------------------
543 * ----------- AC <accession> ACcession number DE <freetext>
544 * DEscription DR <db>; <accession>; Database Reference OS <organism>
545 * OrganiSm (species) OC <clade> Organism Classification (clade, etc.)
546 * LO <look> Look (Color, etc.)
548 if (s.search(annContent))
550 String acc = s.stringMatched(1);
551 String type = s.stringMatched(2);
552 String content = s.stringMatched(3);
553 // TODO: store DR in a vector.
554 // TODO: store AC according to generic file db annotation.
556 if (seqAnn.containsKey(acc))
558 ann = (Hashtable) seqAnn.get(acc);
562 ann = new Hashtable();
564 ann.put(type, content);
565 seqAnn.put(acc, ann);
569 // throw new IOException("Error parsing " + line);
570 System.err.println(">> missing annotation: " + line);
573 else if (annType.equals("GC"))
575 // Generic per-Column annotation, exactly 1 char per column
576 // always need a label.
577 if (x.search(annContent))
579 // parse out and create alignment annotation directly.
580 parseAnnotationRow(annotations, x.stringMatched(1),
584 else if (annType.equals("GR"))
586 // Generic per-Sequence AND per-Column markup, exactly 1 char per
589 * Feature Description Markup letters ------- -----------
590 * -------------- SS Secondary Structure [HGIEBTSCX] SA Surface
591 * Accessibility [0-9X] (0=0%-10%; ...; 9=90%-100%) TM TransMembrane
592 * [Mio] PP Posterior Probability [0-9*] (0=0.00-0.05; 1=0.05-0.15;
593 * *=0.95-1.00) LI LIgand binding [*] AS Active Site [*] IN INtron (in
596 if (s.search(annContent))
598 String acc = s.stringMatched(1);
599 String type = s.stringMatched(2);
600 String oseq = s.stringMatched(3);
602 * copy of annotation field that may be processed into whitespace chunks
604 String seq = new String(oseq);
607 // Get an object with all the annotations for this sequence
608 if (seqAnn.containsKey(acc))
610 // logger.debug("Found annotations for " + acc);
611 ann = (Hashtable) seqAnn.get(acc);
615 // logger.debug("Creating new annotations holder for " + acc);
616 ann = new Hashtable();
617 seqAnn.put(acc, ann);
620 // // start of block for appending annotation lines for wrapped
622 // TODO test structure, call parseAnnotationRow with vector from
623 // hashtable for specific sequence
626 // Get an object with all the content for an annotation
627 if (ann.containsKey("features"))
629 // logger.debug("Found features for " + acc);
630 features = (Hashtable) ann.get("features");
634 // logger.debug("Creating new features holder for " + acc);
635 features = new Hashtable();
636 ann.put("features", features);
640 if (features.containsKey(this.id2type(type)))
642 // logger.debug("Found content for " + this.id2type(type));
643 content = (Hashtable) features.get(this.id2type(type));
647 // logger.debug("Creating new content holder for " +
648 // this.id2type(type));
649 content = new Hashtable();
650 features.put(this.id2type(type), content);
652 String ns = (String) content.get(ANNOTATION);
658 // finally, append the annotation line
660 content.put(ANNOTATION, ns);
661 // // end of wrapped annotation block.
662 // // Now a new row is created with the current set of data
665 if (seqAnn.containsKey(acc))
667 strucAnn = (Hashtable) seqAnn.get(acc);
671 strucAnn = new Hashtable();
674 Vector<AlignmentAnnotation> newStruc = new Vector<>();
675 parseAnnotationRow(newStruc, type, ns);
676 for (AlignmentAnnotation alan : newStruc)
678 alan.visible = false;
680 // new annotation overwrites any existing annotation...
682 strucAnn.put(type, newStruc);
683 seqAnn.put(acc, strucAnn);
689 "Warning - couldn't parse sequence annotation row line:\n"
691 // throw new IOException("Error parsing " + line);
696 throw new IOException(MessageManager.formatMessage(
697 "exception.unknown_annotation_detected", new String[]
698 { annType, annContent }));
702 if (treeString.length() > 0)
704 if (treeName == null)
706 treeName = "Tree " + (1 + getTreeCount());
708 addNewickTree(treeName, treeString.toString());
713 * Demangle an accession string and guess the originating sequence database
714 * for a given sequence
717 * sequence to be annotated
719 * Accession string for sequence
721 * source database for alignment (PFAM or RFAM)
723 private void guessDatabaseFor(Sequence seqO, String dbr, String dbsource)
725 DBRefEntry dbrf = null;
726 List<DBRefEntry> dbrs = new ArrayList<>();
727 String seqdb = "Unknown", sdbac = "" + dbr;
728 int st = -1, en = -1, p;
729 if ((st = sdbac.indexOf("/")) > -1)
731 String num, range = sdbac.substring(st + 1);
732 sdbac = sdbac.substring(0, st);
733 if ((p = range.indexOf("-")) > -1)
736 if (p < range.length())
738 num = range.substring(p).trim();
741 en = Integer.parseInt(num);
742 } catch (NumberFormatException x)
744 // could warn here that index is invalid
753 num = range.substring(0, p).trim();
756 st = Integer.parseInt(num);
757 } catch (NumberFormatException x)
759 // could warn here that index is invalid
763 if (dbsource == null)
765 // make up an origin based on whether the sequence looks like it is
768 dbsource = (seqO.isProtein()) ? "PFAM" : "RFAM";
770 if (dbsource.equals("PFAM"))
773 if (sdbac.indexOf(".") > -1)
775 // strip of last subdomain
776 sdbac = sdbac.substring(0, sdbac.indexOf("."));
777 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, seqdb, dbsource,
784 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, dbsource, dbsource,
793 seqdb = "EMBL"; // total guess - could be ENA, or something else these
795 if (sdbac.indexOf(".") > -1)
797 // strip off last subdomain
798 sdbac = sdbac.substring(0, sdbac.indexOf("."));
799 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, seqdb, dbsource,
807 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, dbsource, dbsource,
814 if (st != -1 && en != -1)
816 for (DBRefEntry d : dbrs)
818 jalview.util.MapList mp = new jalview.util.MapList(
820 { seqO.getStart(), seqO.getEnd() }, new int[] { st, en }, 1,
822 jalview.datamodel.Mapping mping = new Mapping(mp);
828 protected static AlignmentAnnotation parseAnnotationRow(
829 Vector<AlignmentAnnotation> annotation, String label,
832 String convert1, convert2 = null;
834 // convert1 = OPEN_PAREN.replaceAll(annots);
835 // convert2 = CLOSE_PAREN.replaceAll(convert1);
836 // annots = convert2;
839 if (label.contains("_cons"))
841 type = (label.indexOf("_cons") == label.length() - 5)
842 ? label.substring(0, label.length() - 5)
845 boolean ss = false, posterior = false;
846 type = id2type(type);
848 boolean isrnass = false;
849 if (type.equalsIgnoreCase("secondary structure"))
852 isrnass = !NOT_RNASS.search(annots); // sorry about the double negative
853 // here (it's easier for dealing with
854 // other non-alpha-non-brace chars)
856 if (type.equalsIgnoreCase("posterior probability"))
860 // decide on secondary structure or not.
861 Annotation[] els = new Annotation[annots.length()];
862 for (int i = 0; i < annots.length(); i++)
864 String pos = annots.substring(i, i + 1);
866 ann = new Annotation(pos, "", ' ', 0f); // 0f is 'valid' null - will not
870 // if (" .-_".indexOf(pos) == -1)
872 if (isrnass && RNASS_BRACKETS.indexOf(pos) >= 0)
874 ann.secondaryStructure = Rna.getRNASecStrucState(pos).charAt(0);
875 ann.displayCharacter = "" + pos.charAt(0);
879 ann.secondaryStructure = ResidueProperties.getDssp3state(pos)
882 if (ann.secondaryStructure == pos.charAt(0))
884 ann.displayCharacter = ""; // null; // " ";
888 ann.displayCharacter = " " + ann.displayCharacter;
894 if (posterior && !ann.isWhitespace()
895 && !Comparison.isGap(pos.charAt(0)))
898 // symbol encodes values - 0..*==0..10
899 if (pos.charAt(0) == '*')
905 val = pos.charAt(0) - '0';
916 AlignmentAnnotation annot = null;
917 Enumeration<AlignmentAnnotation> e = annotation.elements();
918 while (e.hasMoreElements())
920 annot = e.nextElement();
921 if (annot.label.equals(type))
929 annot = new AlignmentAnnotation(type, type, els);
930 annotation.addElement(annot);
934 Annotation[] anns = new Annotation[annot.annotations.length
936 System.arraycopy(annot.annotations, 0, anns, 0,
937 annot.annotations.length);
938 System.arraycopy(els, 0, anns, annot.annotations.length, els.length);
939 annot.annotations = anns;
940 // System.out.println("else: ");
945 private String dbref_to_ac_record(DBRefEntry ref)
947 return ref.getSource().toString() + " ; "
948 + ref.getAccessionId().toString();
952 public String print(SequenceI[] s, boolean jvSuffix)
954 out = new StringBuffer();
955 out.append("# STOCKHOLM 1.0");
958 // find max length of id
964 Hashtable<String, String> dataRef = null;
965 boolean isAA = s[in].isProtein();
966 while ((in < slen) && ((seq = s[in]) != null))
968 String tmp = printId(seq, jvSuffix);
969 max = Math.max(max, seq.getLength());
971 if (tmp.length() > maxid)
973 maxid = tmp.length();
975 List<DBRefEntry> seqrefs = seq.getDBRefs();
977 if (seqrefs != null && (ndb = seqrefs.size()) > 0)
981 dataRef = new Hashtable<>();
983 List<DBRefEntry> primrefs = seq.getPrimaryDBRefs();
984 if (primrefs.size() >= 1)
986 dataRef.put(tmp, dbref_to_ac_record(primrefs.get(0)));
990 for (int idb = 0; idb < seq.getDBRefs().size(); idb++)
992 DBRefEntry dbref = seq.getDBRefs().get(idb);
993 dataRef.put(tmp, dbref_to_ac_record(dbref));
994 // if we put in a uniprot or EMBL record then we're done:
995 if (isAA && DBRefSource.UNIPROT
996 .equals(DBRefUtils.getCanonicalName(dbref.getSource())))
1000 if (!isAA && DBRefSource.EMBL
1001 .equals(DBRefUtils.getCanonicalName(dbref.getSource())))
1013 // output database type
1014 if (al.getProperties() != null)
1016 if (!al.getProperties().isEmpty())
1018 Enumeration key = al.getProperties().keys();
1019 Enumeration val = al.getProperties().elements();
1020 while (key.hasMoreElements())
1022 out.append("#=GF " + key.nextElement() + " " + val.nextElement());
1023 out.append(newline);
1028 // output database accessions
1029 if (dataRef != null)
1031 Enumeration<String> en = dataRef.keys();
1032 while (en.hasMoreElements())
1034 Object idd = en.nextElement();
1035 String type = dataRef.remove(idd);
1036 out.append(new Format("%-" + (maxid - 2) + "s")
1037 .form("#=GS " + idd.toString() + " "));
1038 if (isAA && type.contains("UNIPROT")
1039 || (!isAA && type.contains("EMBL")))
1042 out.append(" AC " + type.substring(type.indexOf(";") + 1));
1046 out.append(" DR " + type + " ");
1048 out.append(newline);
1052 // output description and annotations
1054 while (i < slen && (seq = s[i]) != null)
1056 if (seq.getDescription() != null)
1058 // out.append("#=GR ");
1059 out.append(new Format("%-" + maxid + "s").form("#=GS "
1060 + printId(seq, jvSuffix) + " DE " + seq.getDescription()));
1061 out.append(newline);
1064 AlignmentAnnotation[] alAnot = seq.getAnnotation();
1068 for (int j = 0, nj = alAnot.length; j < nj; j++)
1071 String key = type2id(alAnot[j].label);
1072 boolean isrna = alAnot[j].isValidStruc();
1076 // hardwire to secondary structure if there is RNA secondary
1077 // structure on the annotation
1086 // out.append("#=GR ");
1087 out.append(new Format("%-" + maxid + "s").form(
1088 "#=GR " + printId(seq, jvSuffix) + " " + key + " "));
1089 ann = alAnot[j].annotations;
1091 for (int k = 0, nk = ann.length; k < nk; k++)
1093 sseq += outputCharacter(key, k, isrna, ann, seq);
1096 out.append(newline);
1100 out.append(new Format("%-" + maxid + "s")
1101 .form(printId(seq, jvSuffix) + " "));
1102 out.append(seq.getSequenceAsString());
1103 out.append(newline);
1107 // alignment annotation
1108 AlignmentAnnotation aa;
1109 AlignmentAnnotation[] an = al.getAlignmentAnnotation();
1112 for (int ia = 0, na = an.length; ia < na; ia++)
1115 if (aa.autoCalculated || !aa.visible || aa.sequenceRef != null)
1122 if (aa.label.equals("seq"))
1128 key = type2id(aa.label.toLowerCase(Locale.ROOT));
1135 label = key + "_cons";
1142 label = label.replace(" ", "_");
1145 new Format("%-" + maxid + "s").form("#=GC " + label + " "));
1146 boolean isrna = aa.isValidStruc();
1147 for (int j = 0, nj = aa.annotations.length; j < nj; j++)
1149 sseq += outputCharacter(key, j, isrna, aa.annotations, null);
1152 out.append(newline);
1157 out.append(newline);
1159 return out.toString();
1163 * add an annotation character to the output row
1172 private char outputCharacter(String key, int k, boolean isrna,
1173 Annotation[] ann, SequenceI sequenceI)
1176 Annotation annot = ann[k];
1177 String ch = (annot == null)
1178 ? ((sequenceI == null) ? "-"
1179 : Character.toString(sequenceI.getCharAt(k)))
1180 : (annot.displayCharacter == null
1181 ? String.valueOf(annot.secondaryStructure)
1182 : annot.displayCharacter);
1187 if (key != null && key.equals("SS"))
1189 char ssannotchar = ' ';
1190 boolean charset = false;
1193 // sensible gap character
1199 // valid secondary structure AND no alternative label (e.g. ' B')
1200 if (annot.secondaryStructure > ' ' && ch.length() < 2)
1202 ssannotchar = annot.secondaryStructure;
1208 return (ssannotchar == ' ' && isrna) ? '.' : ssannotchar;
1212 if (ch.length() == 0)
1216 else if (ch.length() == 1)
1220 else if (ch.length() > 1)
1225 return (seq == ' ' && key != null && key.equals("SS") && isrna) ? '.'
1229 public String print()
1231 out = new StringBuffer();
1232 out.append("# STOCKHOLM 1.0");
1233 out.append(newline);
1234 print(getSeqsAsArray(), false);
1237 out.append(newline);
1238 return out.toString();
1241 private static Hashtable typeIds = null;
1245 if (typeIds == null)
1247 typeIds = new Hashtable();
1248 typeIds.put("SS", "Secondary Structure");
1249 typeIds.put("SA", "Surface Accessibility");
1250 typeIds.put("TM", "transmembrane");
1251 typeIds.put("PP", "Posterior Probability");
1252 typeIds.put("LI", "ligand binding");
1253 typeIds.put("AS", "active site");
1254 typeIds.put("IN", "intron");
1255 typeIds.put("IR", "interacting residue");
1256 typeIds.put("AC", "accession");
1257 typeIds.put("OS", "organism");
1258 typeIds.put("CL", "class");
1259 typeIds.put("DE", "description");
1260 typeIds.put("DR", "reference");
1261 typeIds.put("LO", "look");
1262 typeIds.put("RF", "Reference Positions");
1267 protected static String id2type(String id)
1269 if (typeIds.containsKey(id))
1271 return (String) typeIds.get(id);
1274 "Warning : Unknown Stockholm annotation type code " + id);
1278 protected static String type2id(String type)
1281 Enumeration e = typeIds.keys();
1282 while (e.hasMoreElements())
1284 Object ll = e.nextElement();
1285 if (typeIds.get(ll).toString().equalsIgnoreCase(type))
1296 "Warning : Unknown Stockholm annotation type: " + type);
1301 * make a friendly ID string.
1304 * @return truncated dataName to after last '/'
1306 private String safeName(String dataName)
1309 while ((b = dataName.indexOf("/")) > -1 && b < dataName.length())
1311 dataName = dataName.substring(b + 1).trim();
1314 int e = (dataName.length() - dataName.indexOf(".")) + 1;
1315 dataName = dataName.substring(1, e).trim();