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(MessageManager
222 .getString("exception.stockholm_invalid_format") +" ("+r+")");
226 version = r.stringMatched(1);
228 // logger.debug("Stockholm version: " + version);
231 // We define some Regexes here that will be used regularily later
232 rend = new Regex("^\\s*\\/\\/"); // Find the end of an alignment
233 p = new Regex("(\\S+)\\/(\\d+)\\-(\\d+)"); // split sequence id in
235 s = new Regex("(\\S+)\\s+(\\S*)\\s+(.*)"); // Parses annotation subtype
236 r = new Regex("#=(G[FSRC]?)\\s+(.*)"); // Finds any annotation line
237 x = new Regex("(\\S+)\\s+(\\S+)"); // split id from sequence
239 // Convert all bracket types to parentheses (necessary for passing to VARNA)
240 Regex openparen = new Regex("(<|\\[)", "(");
241 Regex closeparen = new Regex("(>|\\])", ")");
243 // // Detect if file is RNA by looking for bracket types
244 // Regex detectbrackets = new Regex("(<|>|\\[|\\]|\\(|\\))");
251 openparen.optimize();
252 closeparen.optimize();
254 while ((line = nextLine()) != null)
256 if (line.length() == 0)
260 if (rend.search(line))
262 // End of the alignment, pass stuff back
263 this.noSeqs = seqs.size();
265 String dbsource = null;
266 Regex pf = new Regex("PF[0-9]{5}(.*)"); // Finds AC for Pfam
267 Regex rf = new Regex("RF[0-9]{5}(.*)"); // Finds AC for Rfam
268 if (getAlignmentProperty("AC") != null)
270 String dbType = getAlignmentProperty("AC").toString();
271 if (pf.search(dbType))
273 // PFAM Alignment - so references are typically from Uniprot
276 else if (rf.search(dbType))
281 // logger.debug("Number of sequences: " + this.noSeqs);
282 for (Map.Entry<String, String> skey : seqs.entrySet())
284 // logger.debug("Processing sequence " + acc);
285 String acc = skey.getKey();
286 String seq = skey.getValue();
287 if (maxLength < seq.length())
289 maxLength = seq.length();
295 * Retrieve hash of annotations for this accession Associate
296 * Annotation with accession
298 Hashtable accAnnotations = null;
300 if (seqAnn != null && seqAnn.containsKey(acc))
302 accAnnotations = (Hashtable) seqAnn.remove(acc);
303 // TODO: add structures to sequence
306 // Split accession in id and from/to
309 sid = p.stringMatched(1);
310 start = Integer.parseInt(p.stringMatched(2));
311 end = Integer.parseInt(p.stringMatched(3));
313 // logger.debug(sid + ", " + start + ", " + end);
315 Sequence seqO = new Sequence(sid, seq, start, end);
316 // Add Description (if any)
317 if (accAnnotations != null && accAnnotations.containsKey("DE"))
319 String desc = (String) accAnnotations.get("DE");
320 seqO.setDescription((desc == null) ? "" : desc);
322 // Add DB References (if any)
323 if (accAnnotations != null && accAnnotations.containsKey("DR"))
325 String dbr = (String) accAnnotations.get("DR");
326 if (dbr != null && dbr.indexOf(";") > -1)
328 String src = dbr.substring(0, dbr.indexOf(";"));
329 String acn = dbr.substring(dbr.indexOf(";") + 1);
330 jalview.util.DBRefUtils.parseToDbRef(seqO, src, "0", acn);
334 if (accAnnotations != null && accAnnotations.containsKey("AC"))
336 String dbr = (String) accAnnotations.get("AC");
339 // we could get very clever here - but for now - just try to
340 // guess accession type from type of sequence, source of alignment
344 guessDatabaseFor(seqO, dbr, dbsource);
346 // else - do what ? add the data anyway and prompt the user to
347 // specify what references these are ?
350 Hashtable features = null;
351 // We need to adjust the positions of all features to account for gaps
354 features = (Hashtable) accAnnotations.remove("features");
355 } catch (java.lang.NullPointerException e)
357 // loggerwarn("Getting Features for " + acc + ": " +
361 // if we have features
362 if (features != null)
364 int posmap[] = seqO.findPositionMap();
365 Enumeration i = features.keys();
366 while (i.hasMoreElements())
368 // TODO: parse out secondary structure annotation as annotation
370 // TODO: parse out scores as annotation row
371 // TODO: map coding region to core jalview feature types
372 String type = i.nextElement().toString();
373 Hashtable content = (Hashtable) features.remove(type);
375 // add alignment annotation for this feature
376 String key = type2id(type);
379 * have we added annotation rows for this type ?
381 boolean annotsAdded = false;
384 if (accAnnotations != null
385 && accAnnotations.containsKey(key))
387 Vector vv = (Vector) accAnnotations.get(key);
388 for (int ii = 0; ii < vv.size(); ii++)
391 AlignmentAnnotation an = (AlignmentAnnotation) vv
393 seqO.addAlignmentAnnotation(an);
399 Enumeration j = content.keys();
400 while (j.hasMoreElements())
402 String desc = j.nextElement().toString();
403 if (ANNOTATION.equals(desc) && annotsAdded)
405 // don't add features if we already added an annotation row
408 String ns = content.get(desc).toString();
409 char[] byChar = ns.toCharArray();
410 for (int k = 0; k < byChar.length; k++)
413 if (!(c == ' ' || c == '_' || c == '-' || c == '.')) // PFAM
420 int new_pos = posmap[k]; // look up nearest seqeunce
421 // position to this column
422 SequenceFeature feat = new SequenceFeature(type, desc,
423 new_pos, new_pos, null);
425 seqO.addSequenceFeature(feat);
435 // logger.debug("Adding seq " + acc + " from " + start + " to " + end
437 this.seqs.addElement(seqO);
439 return; // finished parsing this segment of source
441 else if (!r.search(line))
443 // System.err.println("Found sequence line: " + line);
445 // Split sequence in sequence and accession parts
448 // logger.error("Could not parse sequence line: " + line);
449 throw new IOException(MessageManager.formatMessage(
450 "exception.couldnt_parse_sequence_line", new String[]
453 String ns = seqs.get(x.stringMatched(1));
458 ns += x.stringMatched(2);
460 seqs.put(x.stringMatched(1), ns);
464 String annType = r.stringMatched(1);
465 String annContent = r.stringMatched(2);
467 // System.err.println("type:" + annType + " content: " + annContent);
469 if (annType.equals("GF"))
472 * Generic per-File annotation, free text Magic features: #=GF NH
473 * <tree in New Hampshire eXtended format> #=GF TN <Unique identifier
474 * for the next tree> Pfam descriptions: 7. DESCRIPTION OF FIELDS
476 * Compulsory fields: ------------------
478 * AC Accession number: Accession number in form PFxxxxx.version or
479 * PBxxxxxx. ID Identification: One word name for family. DE
480 * Definition: Short description of family. AU Author: Authors of the
481 * entry. SE Source of seed: The source suggesting the seed members
482 * belong to one family. GA Gathering method: Search threshold to
483 * build the full alignment. TC Trusted Cutoff: Lowest sequence score
484 * and domain score of match in the full alignment. NC Noise Cutoff:
485 * Highest sequence score and domain score of match not in full
486 * alignment. TP Type: Type of family -- presently Family, Domain,
487 * Motif or Repeat. SQ Sequence: Number of sequences in alignment. AM
488 * Alignment Method The order ls and fs hits are aligned to the model
489 * to build the full align. // End of alignment.
491 * Optional fields: ----------------
493 * DC Database Comment: Comment about database reference. DR Database
494 * Reference: Reference to external database. RC Reference Comment:
495 * Comment about literature reference. RN Reference Number: Reference
496 * Number. RM Reference Medline: Eight digit medline UI number. RT
497 * Reference Title: Reference Title. RA Reference Author: Reference
498 * Author RL Reference Location: Journal location. PI Previous
499 * identifier: Record of all previous ID lines. KW Keywords: Keywords.
500 * CC Comment: Comments. NE Pfam accession: Indicates a nested domain.
501 * NL Location: Location of nested domains - sequence ID, start and
504 * Obsolete fields: ----------- AL Alignment method of seed: The
505 * method used to align the seed members.
507 // Let's save the annotations, maybe we'll be able to do something
508 // with them later...
509 Regex an = new Regex("(\\w+)\\s*(.*)");
510 if (an.search(annContent))
512 if (an.stringMatched(1).equals("NH"))
514 treeString.append(an.stringMatched(2));
516 else if (an.stringMatched(1).equals("TN"))
518 if (treeString.length() > 0)
520 if (treeName == null)
522 treeName = "Tree " + (getTreeCount() + 1);
524 addNewickTree(treeName, treeString.toString());
526 treeName = an.stringMatched(2);
527 treeString = new StringBuffer();
529 // TODO: JAL-3532 - this is where GF comments and database
530 // references are lost
531 // suggest overriding this method for Stockholm files to catch and
533 // process CC, DR etc into multivalued properties
534 setAlignmentProperty(an.stringMatched(1), an.stringMatched(2));
537 else if (annType.equals("GS"))
539 // Generic per-Sequence annotation, free text
541 * Pfam uses these features: Feature Description ---------------------
542 * ----------- AC <accession> ACcession number DE <freetext>
543 * DEscription DR <db>; <accession>; Database Reference OS <organism>
544 * OrganiSm (species) OC <clade> Organism Classification (clade, etc.)
545 * LO <look> Look (Color, etc.)
547 if (s.search(annContent))
549 String acc = s.stringMatched(1);
550 String type = s.stringMatched(2);
551 String content = s.stringMatched(3);
552 // TODO: store DR in a vector.
553 // TODO: store AC according to generic file db annotation.
555 if (seqAnn.containsKey(acc))
557 ann = (Hashtable) seqAnn.get(acc);
561 ann = new Hashtable();
563 ann.put(type, content);
564 seqAnn.put(acc, ann);
568 // throw new IOException("Error parsing " + line);
569 System.err.println(">> missing annotation: " + line);
572 else if (annType.equals("GC"))
574 // Generic per-Column annotation, exactly 1 char per column
575 // always need a label.
576 if (x.search(annContent))
578 // parse out and create alignment annotation directly.
579 parseAnnotationRow(annotations, x.stringMatched(1),
583 else if (annType.equals("GR"))
585 // Generic per-Sequence AND per-Column markup, exactly 1 char per
588 * Feature Description Markup letters ------- -----------
589 * -------------- SS Secondary Structure [HGIEBTSCX] SA Surface
590 * Accessibility [0-9X] (0=0%-10%; ...; 9=90%-100%) TM TransMembrane
591 * [Mio] PP Posterior Probability [0-9*] (0=0.00-0.05; 1=0.05-0.15;
592 * *=0.95-1.00) LI LIgand binding [*] AS Active Site [*] IN INtron (in
595 if (s.search(annContent))
597 String acc = s.stringMatched(1);
598 String type = s.stringMatched(2);
599 String oseq = s.stringMatched(3);
601 * copy of annotation field that may be processed into whitespace chunks
603 String seq = new String(oseq);
606 // Get an object with all the annotations for this sequence
607 if (seqAnn.containsKey(acc))
609 // logger.debug("Found annotations for " + acc);
610 ann = (Hashtable) seqAnn.get(acc);
614 // logger.debug("Creating new annotations holder for " + acc);
615 ann = new Hashtable();
616 seqAnn.put(acc, ann);
619 // // start of block for appending annotation lines for wrapped
621 // TODO test structure, call parseAnnotationRow with vector from
622 // hashtable for specific sequence
625 // Get an object with all the content for an annotation
626 if (ann.containsKey("features"))
628 // logger.debug("Found features for " + acc);
629 features = (Hashtable) ann.get("features");
633 // logger.debug("Creating new features holder for " + acc);
634 features = new Hashtable();
635 ann.put("features", features);
639 if (features.containsKey(this.id2type(type)))
641 // logger.debug("Found content for " + this.id2type(type));
642 content = (Hashtable) features.get(this.id2type(type));
646 // logger.debug("Creating new content holder for " +
647 // this.id2type(type));
648 content = new Hashtable();
649 features.put(this.id2type(type), content);
651 String ns = (String) content.get(ANNOTATION);
657 // finally, append the annotation line
659 content.put(ANNOTATION, ns);
660 // // end of wrapped annotation block.
661 // // Now a new row is created with the current set of data
664 if (seqAnn.containsKey(acc))
666 strucAnn = (Hashtable) seqAnn.get(acc);
670 strucAnn = new Hashtable();
673 Vector<AlignmentAnnotation> newStruc = new Vector<>();
674 parseAnnotationRow(newStruc, type, ns);
675 for (AlignmentAnnotation alan : newStruc)
677 alan.visible = false;
679 // new annotation overwrites any existing annotation...
681 strucAnn.put(type, newStruc);
682 seqAnn.put(acc, strucAnn);
688 "Warning - couldn't parse sequence annotation row line:\n"
690 // throw new IOException("Error parsing " + line);
695 throw new IOException(MessageManager.formatMessage(
696 "exception.unknown_annotation_detected", new String[]
697 { annType, annContent }));
701 if (treeString.length() > 0)
703 if (treeName == null)
705 treeName = "Tree " + (1 + getTreeCount());
707 addNewickTree(treeName, treeString.toString());
712 * Demangle an accession string and guess the originating sequence database
713 * for a given sequence
716 * sequence to be annotated
718 * Accession string for sequence
720 * source database for alignment (PFAM or RFAM)
722 private void guessDatabaseFor(Sequence seqO, String dbr, String dbsource)
724 DBRefEntry dbrf = null;
725 List<DBRefEntry> dbrs = new ArrayList<>();
726 String seqdb = "Unknown", sdbac = "" + dbr;
727 int st = -1, en = -1, p;
728 if ((st = sdbac.indexOf("/")) > -1)
730 String num, range = sdbac.substring(st + 1);
731 sdbac = sdbac.substring(0, st);
732 if ((p = range.indexOf("-")) > -1)
735 if (p < range.length())
737 num = range.substring(p).trim();
740 en = Integer.parseInt(num);
741 } catch (NumberFormatException x)
743 // could warn here that index is invalid
752 num = range.substring(0, p).trim();
755 st = Integer.parseInt(num);
756 } catch (NumberFormatException x)
758 // could warn here that index is invalid
762 if (dbsource == null)
764 // make up an origin based on whether the sequence looks like it is
767 dbsource = (seqO.isProtein()) ? "PFAM" : "RFAM";
769 if (dbsource.equals("PFAM"))
772 if (sdbac.indexOf(".") > -1)
774 // strip of last subdomain
775 sdbac = sdbac.substring(0, sdbac.indexOf("."));
776 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, seqdb, dbsource,
783 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, dbsource, dbsource,
792 seqdb = "EMBL"; // total guess - could be ENA, or something else these
794 if (sdbac.indexOf(".") > -1)
796 // strip off last subdomain
797 sdbac = sdbac.substring(0, sdbac.indexOf("."));
798 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, seqdb, dbsource,
806 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, dbsource, dbsource,
813 if (st != -1 && en != -1)
815 for (DBRefEntry d : dbrs)
817 jalview.util.MapList mp = new jalview.util.MapList(
819 { seqO.getStart(), seqO.getEnd() }, new int[] { st, en }, 1,
821 jalview.datamodel.Mapping mping = new Mapping(mp);
827 protected static AlignmentAnnotation parseAnnotationRow(
828 Vector<AlignmentAnnotation> annotation, String label,
831 String convert1, convert2 = null;
833 // convert1 = OPEN_PAREN.replaceAll(annots);
834 // convert2 = CLOSE_PAREN.replaceAll(convert1);
835 // annots = convert2;
838 if (label.contains("_cons"))
840 type = (label.indexOf("_cons") == label.length() - 5)
841 ? label.substring(0, label.length() - 5)
844 boolean ss = false, posterior = false;
845 type = id2type(type);
847 boolean isrnass = false;
848 if (type.equalsIgnoreCase("secondary structure"))
851 isrnass = !NOT_RNASS.search(annots); // sorry about the double negative
852 // here (it's easier for dealing with
853 // other non-alpha-non-brace chars)
855 if (type.equalsIgnoreCase("posterior probability"))
859 // decide on secondary structure or not.
860 Annotation[] els = new Annotation[annots.length()];
861 for (int i = 0; i < annots.length(); i++)
863 String pos = annots.substring(i, i + 1);
865 ann = new Annotation(pos, "", ' ', 0f); // 0f is 'valid' null - will not
869 // if (" .-_".indexOf(pos) == -1)
871 if (isrnass && RNASS_BRACKETS.indexOf(pos) >= 0)
873 ann.secondaryStructure = Rna.getRNASecStrucState(pos).charAt(0);
874 ann.displayCharacter = "" + pos.charAt(0);
878 ann.secondaryStructure = ResidueProperties.getDssp3state(pos)
881 if (ann.secondaryStructure == pos.charAt(0))
883 ann.displayCharacter = ""; // null; // " ";
887 ann.displayCharacter = " " + ann.displayCharacter;
893 if (posterior && !ann.isWhitespace()
894 && !Comparison.isGap(pos.charAt(0)))
897 // symbol encodes values - 0..*==0..10
898 if (pos.charAt(0) == '*')
904 val = pos.charAt(0) - '0';
915 AlignmentAnnotation annot = null;
916 Enumeration<AlignmentAnnotation> e = annotation.elements();
917 while (e.hasMoreElements())
919 annot = e.nextElement();
920 if (annot.label.equals(type))
928 annot = new AlignmentAnnotation(type, type, els);
929 annotation.addElement(annot);
933 Annotation[] anns = new Annotation[annot.annotations.length
935 System.arraycopy(annot.annotations, 0, anns, 0,
936 annot.annotations.length);
937 System.arraycopy(els, 0, anns, annot.annotations.length, els.length);
938 annot.annotations = anns;
939 // System.out.println("else: ");
944 private String dbref_to_ac_record(DBRefEntry ref)
946 return ref.getSource().toString() + " ; "
947 + ref.getAccessionId().toString();
951 public String print(SequenceI[] s, boolean jvSuffix)
953 out = new StringBuffer();
954 out.append("# STOCKHOLM 1.0");
957 // find max length of id
963 Hashtable<String, String> dataRef = null;
964 boolean isAA = s[in].isProtein();
965 while ((in < slen) && ((seq = s[in]) != null))
967 String tmp = printId(seq, jvSuffix);
968 max = Math.max(max, seq.getLength());
970 if (tmp.length() > maxid)
972 maxid = tmp.length();
974 List<DBRefEntry> seqrefs = seq.getDBRefs();
976 if (seqrefs != null && (ndb = seqrefs.size()) > 0)
980 dataRef = new Hashtable<>();
982 List<DBRefEntry> primrefs = seq.getPrimaryDBRefs();
983 if (primrefs.size() >= 1)
985 dataRef.put(tmp, dbref_to_ac_record(primrefs.get(0)));
989 for (int idb = 0; idb < seq.getDBRefs().size(); idb++)
991 DBRefEntry dbref = seq.getDBRefs().get(idb);
992 dataRef.put(tmp, dbref_to_ac_record(dbref));
993 // if we put in a uniprot or EMBL record then we're done:
994 if (isAA && DBRefSource.UNIPROT
995 .equals(DBRefUtils.getCanonicalName(dbref.getSource())))
999 if (!isAA && DBRefSource.EMBL
1000 .equals(DBRefUtils.getCanonicalName(dbref.getSource())))
1012 // output database type
1013 if (al.getProperties() != null)
1015 if (!al.getProperties().isEmpty())
1017 Enumeration key = al.getProperties().keys();
1018 Enumeration val = al.getProperties().elements();
1019 while (key.hasMoreElements())
1021 out.append("#=GF " + key.nextElement() + " " + val.nextElement());
1022 out.append(newline);
1027 // output database accessions
1028 if (dataRef != null)
1030 Enumeration<String> en = dataRef.keys();
1031 while (en.hasMoreElements())
1033 Object idd = en.nextElement();
1034 String type = dataRef.remove(idd);
1035 out.append(new Format("%-" + (maxid - 2) + "s")
1036 .form("#=GS " + idd.toString() + " "));
1037 if (isAA && type.contains("UNIPROT")
1038 || (!isAA && type.contains("EMBL")))
1041 out.append(" AC " + type.substring(type.indexOf(";") + 1));
1045 out.append(" DR " + type + " ");
1047 out.append(newline);
1051 // output annotations
1052 while (i < slen && (seq = s[i]) != null)
1054 AlignmentAnnotation[] alAnot = seq.getAnnotation();
1058 for (int j = 0, nj = alAnot.length; j < nj; j++)
1061 String key = type2id(alAnot[j].label);
1062 boolean isrna = alAnot[j].isValidStruc();
1066 // hardwire to secondary structure if there is RNA secondary
1067 // structure on the annotation
1076 // out.append("#=GR ");
1077 out.append(new Format("%-" + maxid + "s").form(
1078 "#=GR " + printId(seq, jvSuffix) + " " + key + " "));
1079 ann = alAnot[j].annotations;
1081 for (int k = 0, nk = ann.length; k < nk; k++)
1083 sseq += outputCharacter(key, k, isrna, ann, seq);
1086 out.append(newline);
1090 out.append(new Format("%-" + maxid + "s")
1091 .form(printId(seq, jvSuffix) + " "));
1092 out.append(seq.getSequenceAsString());
1093 out.append(newline);
1097 // alignment annotation
1098 AlignmentAnnotation aa;
1099 AlignmentAnnotation[] an = al.getAlignmentAnnotation();
1102 for (int ia = 0, na = an.length; ia < na; ia++)
1105 if (aa.autoCalculated || !aa.visible || aa.sequenceRef != null)
1112 if (aa.label.equals("seq"))
1118 key = type2id(aa.label.toLowerCase(Locale.ROOT));
1125 label = key + "_cons";
1132 label = label.replace(" ", "_");
1135 new Format("%-" + maxid + "s").form("#=GC " + label + " "));
1136 boolean isrna = aa.isValidStruc();
1137 for (int j = 0, nj = aa.annotations.length; j < nj; j++)
1139 sseq += outputCharacter(key, j, isrna, aa.annotations, null);
1142 out.append(newline);
1147 out.append(newline);
1149 return out.toString();
1153 * add an annotation character to the output row
1162 private char outputCharacter(String key, int k, boolean isrna,
1163 Annotation[] ann, SequenceI sequenceI)
1166 Annotation annot = ann[k];
1167 String ch = (annot == null)
1168 ? ((sequenceI == null) ? "-"
1169 : Character.toString(sequenceI.getCharAt(k)))
1170 : (annot.displayCharacter == null
1171 ? String.valueOf(annot.secondaryStructure)
1172 : annot.displayCharacter);
1177 if (key != null && key.equals("SS"))
1179 char ssannotchar = ' ';
1180 boolean charset = false;
1183 // sensible gap character
1189 // valid secondary structure AND no alternative label (e.g. ' B')
1190 if (annot.secondaryStructure > ' ' && ch.length() < 2)
1192 ssannotchar = annot.secondaryStructure;
1198 return (ssannotchar == ' ' && isrna) ? '.' : ssannotchar;
1202 if (ch.length() == 0)
1206 else if (ch.length() == 1)
1210 else if (ch.length() > 1)
1215 return (seq == ' ' && key != null && key.equals("SS") && isrna) ? '.'
1219 public String print()
1221 out = new StringBuffer();
1222 out.append("# STOCKHOLM 1.0");
1223 out.append(newline);
1224 print(getSeqsAsArray(), false);
1227 out.append(newline);
1228 return out.toString();
1231 private static Hashtable typeIds = null;
1235 if (typeIds == null)
1237 typeIds = new Hashtable();
1238 typeIds.put("SS", "Secondary Structure");
1239 typeIds.put("SA", "Surface Accessibility");
1240 typeIds.put("TM", "transmembrane");
1241 typeIds.put("PP", "Posterior Probability");
1242 typeIds.put("LI", "ligand binding");
1243 typeIds.put("AS", "active site");
1244 typeIds.put("IN", "intron");
1245 typeIds.put("IR", "interacting residue");
1246 typeIds.put("AC", "accession");
1247 typeIds.put("OS", "organism");
1248 typeIds.put("CL", "class");
1249 typeIds.put("DE", "description");
1250 typeIds.put("DR", "reference");
1251 typeIds.put("LO", "look");
1252 typeIds.put("RF", "Reference Positions");
1257 protected static String id2type(String id)
1259 if (typeIds.containsKey(id))
1261 return (String) typeIds.get(id);
1264 "Warning : Unknown Stockholm annotation type code " + id);
1268 protected static String type2id(String type)
1271 Enumeration e = typeIds.keys();
1272 while (e.hasMoreElements())
1274 Object ll = e.nextElement();
1275 if (typeIds.get(ll).toString().equalsIgnoreCase(type))
1286 "Warning : Unknown Stockholm annotation type: " + type);
1291 * make a friendly ID string.
1294 * @return truncated dataName to after last '/'
1296 private String safeName(String dataName)
1299 while ((b = dataName.indexOf("/")) > -1 && b < dataName.length())
1301 dataName = dataName.substring(b + 1).trim();
1304 int e = (dataName.length() - dataName.indexOf(".")) + 1;
1305 dataName = dataName.substring(1, e).trim();