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 jalview.analysis.Rna;
27 import jalview.datamodel.AlignmentAnnotation;
28 import jalview.datamodel.AlignmentI;
29 import jalview.datamodel.Annotation;
30 import jalview.datamodel.DBRefEntry;
31 import jalview.datamodel.Mapping;
32 import jalview.datamodel.Sequence;
33 import jalview.datamodel.SequenceFeature;
34 import jalview.datamodel.SequenceI;
35 import jalview.schemes.ResidueProperties;
36 import jalview.util.Comparison;
37 import jalview.util.Format;
38 import jalview.util.MessageManager;
40 import java.io.BufferedReader;
41 import java.io.FileReader;
42 import java.io.IOException;
43 import java.util.ArrayList;
44 import java.util.Enumeration;
45 import java.util.Hashtable;
46 import java.util.LinkedHashMap;
47 import java.util.List;
49 import java.util.Vector;
51 import com.stevesoft.pat.Regex;
53 import fr.orsay.lri.varna.exceptions.ExceptionUnmatchedClosingParentheses;
54 import fr.orsay.lri.varna.factories.RNAFactory;
55 import fr.orsay.lri.varna.models.rna.RNA;
57 // import org.apache.log4j.*;
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 Regex OPEN_PAREN = new Regex("(<|\\[)", "(");
79 private static final Regex CLOSE_PAREN = new Regex("(>|\\])", ")");
81 public static final Regex DETECT_BRACKETS = new Regex(
82 "(<|>|\\[|\\]|\\(|\\)|\\{|\\})");
84 StringBuffer out; // output buffer
88 public StockholmFile()
93 * Creates a new StockholmFile object for output.
95 public StockholmFile(AlignmentI al)
100 public StockholmFile(String inFile, DataSourceType type)
106 public StockholmFile(FileParse source) throws IOException
112 public void initData()
118 * Parse a file in Stockholm format into Jalview's data model using VARNA
120 * @throws IOException
121 * If there is an error with the input file
123 public void parse_with_VARNA(java.io.File inFile) throws IOException
125 FileReader fr = null;
126 fr = new FileReader(inFile);
128 BufferedReader r = new BufferedReader(fr);
129 List<RNA> result = null;
132 result = RNAFactory.loadSecStrStockholm(r);
133 } catch (ExceptionUnmatchedClosingParentheses umcp)
135 errormessage = "Unmatched parentheses in annotation. Aborting ("
136 + umcp.getMessage() + ")";
137 throw new IOException(umcp);
139 // DEBUG System.out.println("this is the secondary scructure:"
141 SequenceI[] seqs = new SequenceI[result.size()];
143 for (int i = 0; i < result.size(); i++)
145 // DEBUG System.err.println("Processing i'th sequence in Stockholm file")
146 RNA current = result.get(i);
148 String seq = current.getSeq();
149 String rna = current.getStructDBN(true);
150 // DEBUG System.out.println(seq);
151 // DEBUG System.err.println(rna);
153 int end = seq.length() - 1;
154 id = safeName(getDataName());
155 seqs[i] = new Sequence(id, seq, begin, end);
156 String[] annot = new String[rna.length()];
157 Annotation[] ann = new Annotation[rna.length()];
158 for (int j = 0; j < rna.length(); j++)
160 annot[j] = rna.substring(j, j + 1);
164 for (int k = 0; k < rna.length(); k++)
166 ann[k] = new Annotation(annot[k], "",
167 Rna.getRNASecStrucState(annot[k]).charAt(0), 0f);
170 AlignmentAnnotation align = new AlignmentAnnotation("Sec. str.",
171 current.getID(), ann);
173 seqs[i].addAlignmentAnnotation(align);
174 seqs[i].setRNA(result.get(i));
175 this.annotations.addElement(align);
182 * Parse a file in Stockholm format into Jalview's data model. The file has to
183 * be passed at construction time
185 * @throws IOException
186 * If there is an error with the input file
189 public void parse() throws IOException
191 StringBuffer treeString = new StringBuffer();
192 String treeName = null;
193 // --------------- Variable Definitions -------------------
197 Hashtable seqAnn = new Hashtable(); // Sequence related annotations
198 LinkedHashMap<String, String> seqs = new LinkedHashMap<String, String>();
199 Regex p, r, rend, s, x;
200 // Temporary line for processing RNA annotation
201 // String RNAannot = "";
203 // ------------------ Parsing File ----------------------
204 // First, we have to check that this file has STOCKHOLM format, i.e. the
205 // first line must match
207 r = new Regex("# STOCKHOLM ([\\d\\.]+)");
208 if (!r.search(nextLine()))
210 throw new IOException(MessageManager
211 .getString("exception.stockholm_invalid_format"));
215 version = r.stringMatched(1);
217 // logger.debug("Stockholm version: " + version);
220 // We define some Regexes here that will be used regularily later
221 rend = new Regex("^\\s*\\/\\/"); // Find the end of an alignment
222 p = new Regex("(\\S+)\\/(\\d+)\\-(\\d+)"); // split sequence id in
224 s = new Regex("(\\S+)\\s+(\\S*)\\s+(.*)"); // Parses annotation subtype
225 r = new Regex("#=(G[FSRC]?)\\s+(.*)"); // Finds any annotation line
226 x = new Regex("(\\S+)\\s+(\\S+)"); // split id from sequence
228 // Convert all bracket types to parentheses (necessary for passing to VARNA)
229 Regex openparen = new Regex("(<|\\[)", "(");
230 Regex closeparen = new Regex("(>|\\])", ")");
232 // Detect if file is RNA by looking for bracket types
233 Regex detectbrackets = new Regex("(<|>|\\[|\\]|\\(|\\))");
240 openparen.optimize();
241 closeparen.optimize();
243 while ((line = nextLine()) != null)
245 if (line.length() == 0)
249 if (rend.search(line))
251 // End of the alignment, pass stuff back
252 this.noSeqs = seqs.size();
254 String seqdb, dbsource = null;
255 Regex pf = new Regex("PF[0-9]{5}(.*)"); // Finds AC for Pfam
256 Regex rf = new Regex("RF[0-9]{5}(.*)"); // Finds AC for Rfam
257 if (getAlignmentProperty("AC") != null)
259 String dbType = getAlignmentProperty("AC").toString();
260 if (pf.search(dbType))
262 // PFAM Alignment - so references are typically from Uniprot
265 else if (rf.search(dbType))
270 // logger.debug("Number of sequences: " + this.noSeqs);
271 for (Map.Entry<String, String> skey : seqs.entrySet())
273 // logger.debug("Processing sequence " + acc);
274 String acc = skey.getKey();
275 String seq = skey.getValue();
276 if (maxLength < seq.length())
278 maxLength = seq.length();
284 * Retrieve hash of annotations for this accession Associate
285 * Annotation with accession
287 Hashtable accAnnotations = null;
289 if (seqAnn != null && seqAnn.containsKey(acc))
291 accAnnotations = (Hashtable) seqAnn.remove(acc);
292 // TODO: add structures to sequence
295 // Split accession in id and from/to
298 sid = p.stringMatched(1);
299 start = Integer.parseInt(p.stringMatched(2));
300 end = Integer.parseInt(p.stringMatched(3));
302 // logger.debug(sid + ", " + start + ", " + end);
304 Sequence seqO = new Sequence(sid, seq, start, end);
305 // Add Description (if any)
306 if (accAnnotations != null && accAnnotations.containsKey("DE"))
308 String desc = (String) accAnnotations.get("DE");
309 seqO.setDescription((desc == null) ? "" : desc);
311 // Add DB References (if any)
312 if (accAnnotations != null && accAnnotations.containsKey("DR"))
314 String dbr = (String) accAnnotations.get("DR");
315 if (dbr != null && dbr.indexOf(";") > -1)
317 String src = dbr.substring(0, dbr.indexOf(";"));
318 String acn = dbr.substring(dbr.indexOf(";") + 1);
319 jalview.util.DBRefUtils.parseToDbRef(seqO, src, "0", acn);
323 if (accAnnotations != null && accAnnotations.containsKey("AC"))
325 if (dbsource != null)
327 String dbr = (String) accAnnotations.get("AC");
330 // we could get very clever here - but for now - just try to
331 // guess accession type from source of alignment plus structure
333 guessDatabaseFor(seqO, dbr, dbsource);
337 // else - do what ? add the data anyway and prompt the user to
338 // specify what references these are ?
341 Hashtable features = null;
342 // We need to adjust the positions of all features to account for gaps
345 features = (Hashtable) accAnnotations.remove("features");
346 } catch (java.lang.NullPointerException e)
348 // loggerwarn("Getting Features for " + acc + ": " +
352 // if we have features
353 if (features != null)
355 int posmap[] = seqO.findPositionMap();
356 Enumeration i = features.keys();
357 while (i.hasMoreElements())
359 // TODO: parse out secondary structure annotation as annotation
361 // TODO: parse out scores as annotation row
362 // TODO: map coding region to core jalview feature types
363 String type = i.nextElement().toString();
364 Hashtable content = (Hashtable) features.remove(type);
366 // add alignment annotation for this feature
367 String key = type2id(type);
370 * have we added annotation rows for this type ?
372 boolean annotsAdded = false;
375 if (accAnnotations != null
376 && accAnnotations.containsKey(key))
378 Vector vv = (Vector) accAnnotations.get(key);
379 for (int ii = 0; ii < vv.size(); ii++)
382 AlignmentAnnotation an = (AlignmentAnnotation) vv
384 seqO.addAlignmentAnnotation(an);
390 Enumeration j = content.keys();
391 while (j.hasMoreElements())
393 String desc = j.nextElement().toString();
394 if ("annotations".equals(desc) && annotsAdded)
396 // don't add features if we already added an annotation row
399 String ns = content.get(desc).toString();
400 char[] byChar = ns.toCharArray();
401 for (int k = 0; k < byChar.length; k++)
404 if (!(c == ' ' || c == '_' || c == '-' || c == '.')) // PFAM
411 int new_pos = posmap[k]; // look up nearest seqeunce
412 // position to this column
413 SequenceFeature feat = new SequenceFeature(type, desc,
414 new_pos, new_pos, 0f, null);
416 seqO.addSequenceFeature(feat);
426 // logger.debug("Adding seq " + acc + " from " + start + " to " + end
428 this.seqs.addElement(seqO);
430 return; // finished parsing this segment of source
432 else if (!r.search(line))
434 // System.err.println("Found sequence line: " + line);
436 // Split sequence in sequence and accession parts
439 // logger.error("Could not parse sequence line: " + line);
440 throw new IOException(MessageManager.formatMessage(
441 "exception.couldnt_parse_sequence_line", new String[]
444 String ns = seqs.get(x.stringMatched(1));
449 ns += x.stringMatched(2);
451 seqs.put(x.stringMatched(1), ns);
455 String annType = r.stringMatched(1);
456 String annContent = r.stringMatched(2);
458 // System.err.println("type:" + annType + " content: " + annContent);
460 if (annType.equals("GF"))
463 * Generic per-File annotation, free text Magic features: #=GF NH
464 * <tree in New Hampshire eXtended format> #=GF TN <Unique identifier
465 * for the next tree> Pfam descriptions: 7. DESCRIPTION OF FIELDS
467 * Compulsory fields: ------------------
469 * AC Accession number: Accession number in form PFxxxxx.version or
470 * PBxxxxxx. ID Identification: One word name for family. DE
471 * Definition: Short description of family. AU Author: Authors of the
472 * entry. SE Source of seed: The source suggesting the seed members
473 * belong to one family. GA Gathering method: Search threshold to
474 * build the full alignment. TC Trusted Cutoff: Lowest sequence score
475 * and domain score of match in the full alignment. NC Noise Cutoff:
476 * Highest sequence score and domain score of match not in full
477 * alignment. TP Type: Type of family -- presently Family, Domain,
478 * Motif or Repeat. SQ Sequence: Number of sequences in alignment. AM
479 * Alignment Method The order ls and fs hits are aligned to the model
480 * to build the full align. // End of alignment.
482 * Optional fields: ----------------
484 * DC Database Comment: Comment about database reference. DR Database
485 * Reference: Reference to external database. RC Reference Comment:
486 * Comment about literature reference. RN Reference Number: Reference
487 * Number. RM Reference Medline: Eight digit medline UI number. RT
488 * Reference Title: Reference Title. RA Reference Author: Reference
489 * Author RL Reference Location: Journal location. PI Previous
490 * identifier: Record of all previous ID lines. KW Keywords: Keywords.
491 * CC Comment: Comments. NE Pfam accession: Indicates a nested domain.
492 * NL Location: Location of nested domains - sequence ID, start and
495 * Obsolete fields: ----------- AL Alignment method of seed: The
496 * method used to align the seed members.
498 // Let's save the annotations, maybe we'll be able to do something
499 // with them later...
500 Regex an = new Regex("(\\w+)\\s*(.*)");
501 if (an.search(annContent))
503 if (an.stringMatched(1).equals("NH"))
505 treeString.append(an.stringMatched(2));
507 else if (an.stringMatched(1).equals("TN"))
509 if (treeString.length() > 0)
511 if (treeName == null)
513 treeName = "Tree " + (getTreeCount() + 1);
515 addNewickTree(treeName, treeString.toString());
517 treeName = an.stringMatched(2);
518 treeString = new StringBuffer();
520 setAlignmentProperty(an.stringMatched(1), an.stringMatched(2));
523 else if (annType.equals("GS"))
525 // Generic per-Sequence annotation, free text
527 * Pfam uses these features: Feature Description ---------------------
528 * ----------- AC <accession> ACcession number DE <freetext>
529 * DEscription DR <db>; <accession>; Database Reference OS <organism>
530 * OrganiSm (species) OC <clade> Organism Classification (clade, etc.)
531 * LO <look> Look (Color, etc.)
533 if (s.search(annContent))
535 String acc = s.stringMatched(1);
536 String type = s.stringMatched(2);
537 String content = s.stringMatched(3);
538 // TODO: store DR in a vector.
539 // TODO: store AC according to generic file db annotation.
541 if (seqAnn.containsKey(acc))
543 ann = (Hashtable) seqAnn.get(acc);
547 ann = new Hashtable();
549 ann.put(type, content);
550 seqAnn.put(acc, ann);
554 // throw new IOException("Error parsing " + line);
555 System.err.println(">> missing annotation: " + line);
558 else if (annType.equals("GC"))
560 // Generic per-Column annotation, exactly 1 char per column
561 // always need a label.
562 if (x.search(annContent))
564 // parse out and create alignment annotation directly.
565 parseAnnotationRow(annotations, x.stringMatched(1),
569 else if (annType.equals("GR"))
571 // Generic per-Sequence AND per-Column markup, exactly 1 char per
574 * Feature Description Markup letters ------- -----------
575 * -------------- SS Secondary Structure [HGIEBTSCX] SA Surface
576 * Accessibility [0-9X] (0=0%-10%; ...; 9=90%-100%) TM TransMembrane
577 * [Mio] PP Posterior Probability [0-9*] (0=0.00-0.05; 1=0.05-0.15;
578 * *=0.95-1.00) LI LIgand binding [*] AS Active Site [*] IN INtron (in
581 if (s.search(annContent))
583 String acc = s.stringMatched(1);
584 String type = s.stringMatched(2);
585 String oseq = s.stringMatched(3);
587 * copy of annotation field that may be processed into whitespace chunks
589 String seq = new String(oseq);
592 // Get an object with all the annotations for this sequence
593 if (seqAnn.containsKey(acc))
595 // logger.debug("Found annotations for " + acc);
596 ann = (Hashtable) seqAnn.get(acc);
600 // logger.debug("Creating new annotations holder for " + acc);
601 ann = new Hashtable();
602 seqAnn.put(acc, ann);
605 // // start of block for appending annotation lines for wrapped
607 // TODO test structure, call parseAnnotationRow with vector from
608 // hashtable for specific sequence
611 // Get an object with all the content for an annotation
612 if (ann.containsKey("features"))
614 // logger.debug("Found features for " + acc);
615 features = (Hashtable) ann.get("features");
619 // logger.debug("Creating new features holder for " + acc);
620 features = new Hashtable();
621 ann.put("features", features);
625 if (features.containsKey(this.id2type(type)))
627 // logger.debug("Found content for " + this.id2type(type));
628 content = (Hashtable) features.get(this.id2type(type));
632 // logger.debug("Creating new content holder for " +
633 // this.id2type(type));
634 content = new Hashtable();
635 features.put(this.id2type(type), content);
637 String ns = (String) content.get("annotation");
643 // finally, append the annotation line
645 content.put("annotation", ns);
646 // // end of wrapped annotation block.
647 // // Now a new row is created with the current set of data
650 if (seqAnn.containsKey(acc))
652 strucAnn = (Hashtable) seqAnn.get(acc);
656 strucAnn = new Hashtable();
659 Vector<AlignmentAnnotation> newStruc = new Vector<AlignmentAnnotation>();
660 parseAnnotationRow(newStruc, type, ns);
661 for (AlignmentAnnotation alan : newStruc)
663 alan.visible = false;
665 // new annotation overwrites any existing annotation...
667 strucAnn.put(type, newStruc);
668 seqAnn.put(acc, strucAnn);
674 "Warning - couldn't parse sequence annotation row line:\n"
676 // throw new IOException("Error parsing " + line);
681 throw new IOException(MessageManager.formatMessage(
682 "exception.unknown_annotation_detected", new String[]
683 { annType, annContent }));
687 if (treeString.length() > 0)
689 if (treeName == null)
691 treeName = "Tree " + (1 + getTreeCount());
693 addNewickTree(treeName, treeString.toString());
698 * Demangle an accession string and guess the originating sequence database
699 * for a given sequence
702 * sequence to be annotated
704 * Accession string for sequence
706 * source database for alignment (PFAM or RFAM)
708 private void guessDatabaseFor(Sequence seqO, String dbr, String dbsource)
710 DBRefEntry dbrf = null;
711 List<DBRefEntry> dbrs = new ArrayList<DBRefEntry>();
712 String seqdb = "Unknown", sdbac = "" + dbr;
713 int st = -1, en = -1, p;
714 if ((st = sdbac.indexOf("/")) > -1)
716 String num, range = sdbac.substring(st + 1);
717 sdbac = sdbac.substring(0, st);
718 if ((p = range.indexOf("-")) > -1)
721 if (p < range.length())
723 num = range.substring(p).trim();
726 en = Integer.parseInt(num);
727 } catch (NumberFormatException x)
729 // could warn here that index is invalid
738 num = range.substring(0, p).trim();
741 st = Integer.parseInt(num);
742 } catch (NumberFormatException x)
744 // could warn here that index is invalid
748 if (dbsource.equals("PFAM"))
751 if (sdbac.indexOf(".") > -1)
753 // strip of last subdomain
754 sdbac = sdbac.substring(0, sdbac.indexOf("."));
755 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, seqdb, dbsource,
762 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, dbsource, dbsource,
771 seqdb = "EMBL"; // total guess - could be ENA, or something else these
773 if (sdbac.indexOf(".") > -1)
775 // strip off last subdomain
776 sdbac = sdbac.substring(0, sdbac.indexOf("."));
777 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, seqdb, dbsource,
785 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, dbsource, dbsource,
792 if (st != -1 && en != -1)
794 for (DBRefEntry d : dbrs)
796 jalview.util.MapList mp = new jalview.util.MapList(
798 { seqO.getStart(), seqO.getEnd() }, new int[] { st, en }, 1,
800 jalview.datamodel.Mapping mping = new Mapping(mp);
806 protected static AlignmentAnnotation parseAnnotationRow(
807 Vector<AlignmentAnnotation> annotation, String label,
810 String convert1, convert2 = null;
812 // convert1 = OPEN_PAREN.replaceAll(annots);
813 // convert2 = CLOSE_PAREN.replaceAll(convert1);
814 // annots = convert2;
817 if (label.contains("_cons"))
819 type = (label.indexOf("_cons") == label.length() - 5)
820 ? label.substring(0, label.length() - 5)
823 boolean ss = false, posterior = false;
824 type = id2type(type);
825 if (type.equalsIgnoreCase("secondary structure"))
829 if (type.equalsIgnoreCase("posterior probability"))
833 // decide on secondary structure or not.
834 Annotation[] els = new Annotation[annots.length()];
835 for (int i = 0; i < annots.length(); i++)
837 String pos = annots.substring(i, i + 1);
839 ann = new Annotation(pos, "", ' ', 0f); // 0f is 'valid' null - will not
843 // if (" .-_".indexOf(pos) == -1)
845 if (DETECT_BRACKETS.search(pos))
847 ann.secondaryStructure = Rna.getRNASecStrucState(pos).charAt(0);
848 ann.displayCharacter = "" + pos.charAt(0);
852 ann.secondaryStructure = ResidueProperties.getDssp3state(pos)
855 if (ann.secondaryStructure == pos.charAt(0))
857 ann.displayCharacter = ""; // null; // " ";
861 ann.displayCharacter = " " + ann.displayCharacter;
867 if (posterior && !ann.isWhitespace()
868 && !Comparison.isGap(pos.charAt(0)))
871 // symbol encodes values - 0..*==0..10
872 if (pos.charAt(0) == '*')
878 val = pos.charAt(0) - '0';
889 AlignmentAnnotation annot = null;
890 Enumeration<AlignmentAnnotation> e = annotation.elements();
891 while (e.hasMoreElements())
893 annot = e.nextElement();
894 if (annot.label.equals(type))
902 annot = new AlignmentAnnotation(type, type, els);
903 annotation.addElement(annot);
907 Annotation[] anns = new Annotation[annot.annotations.length
909 System.arraycopy(annot.annotations, 0, anns, 0,
910 annot.annotations.length);
911 System.arraycopy(els, 0, anns, annot.annotations.length, els.length);
912 annot.annotations = anns;
913 // System.out.println("else: ");
919 public String print(SequenceI[] s, boolean jvSuffix)
921 out = new StringBuffer();
922 out.append("# STOCKHOLM 1.0");
925 // find max length of id
929 Hashtable dataRef = null;
930 while ((in < s.length) && (s[in] != null))
932 String tmp = printId(s[in], jvSuffix);
933 if (s[in].getSequence().length > max)
935 max = s[in].getSequence().length;
938 if (tmp.length() > maxid)
940 maxid = tmp.length();
942 if (s[in].getDBRefs() != null)
944 for (int idb = 0; idb < s[in].getDBRefs().length; idb++)
948 dataRef = new Hashtable();
951 String datAs1 = s[in].getDBRefs()[idb].getSource().toString()
953 + s[in].getDBRefs()[idb].getAccessionId().toString();
954 dataRef.put(tmp, datAs1);
962 // output database type
963 if (al.getProperties() != null)
965 if (!al.getProperties().isEmpty())
967 Enumeration key = al.getProperties().keys();
968 Enumeration val = al.getProperties().elements();
969 while (key.hasMoreElements())
971 out.append("#=GF " + key.nextElement() + " " + val.nextElement());
977 // output database accessions
980 Enumeration en = dataRef.keys();
981 while (en.hasMoreElements())
983 Object idd = en.nextElement();
984 String type = (String) dataRef.remove(idd);
985 out.append(new Format("%-" + (maxid - 2) + "s")
986 .form("#=GS " + idd.toString() + " "));
987 if (type.contains("PFAM") || type.contains("RFAM"))
990 out.append(" AC " + type.substring(type.indexOf(";") + 1));
994 out.append(" DR " + type + " ");
1000 // output annotations
1001 while (i < s.length && s[i] != null)
1003 AlignmentAnnotation[] alAnot = s[i].getAnnotation();
1007 for (int j = 0; j < alAnot.length; j++)
1010 String key = type2id(alAnot[j].label);
1011 boolean isrna = alAnot[j].isValidStruc();
1015 // hardwire to secondary structure if there is RNA secondary
1016 // structure on the annotation
1025 // out.append("#=GR ");
1026 out.append(new Format("%-" + maxid + "s").form(
1027 "#=GR " + printId(s[i], jvSuffix) + " " + key + " "));
1028 ann = alAnot[j].annotations;
1030 for (int k = 0; k < ann.length; k++)
1032 seq += outputCharacter(key, k, isrna, ann, s[i]);
1035 out.append(newline);
1039 out.append(new Format("%-" + maxid + "s")
1040 .form(printId(s[i], jvSuffix) + " "));
1041 out.append(s[i].getSequenceAsString());
1042 out.append(newline);
1046 // alignment annotation
1047 AlignmentAnnotation aa;
1048 if (al.getAlignmentAnnotation() != null)
1050 for (int ia = 0; ia < al.getAlignmentAnnotation().length; ia++)
1052 aa = al.getAlignmentAnnotation()[ia];
1053 if (aa.autoCalculated || !aa.visible || aa.sequenceRef != null)
1060 if (aa.label.equals("seq"))
1066 key = type2id(aa.label.toLowerCase());
1073 label = key + "_cons";
1080 label = label.replace(" ", "_");
1083 new Format("%-" + maxid + "s").form("#=GC " + label + " "));
1084 boolean isrna = aa.isValidStruc();
1085 for (int j = 0; j < aa.annotations.length; j++)
1087 seq += outputCharacter(key, j, isrna, aa.annotations, null);
1090 out.append(newline);
1095 out.append(newline);
1097 return out.toString();
1101 * add an annotation character to the output row
1110 private char outputCharacter(String key, int k, boolean isrna,
1111 Annotation[] ann, SequenceI sequenceI)
1114 Annotation annot = ann[k];
1115 String ch = (annot == null)
1116 ? ((sequenceI == null) ? "-"
1117 : Character.toString(sequenceI.getCharAt(k)))
1118 : annot.displayCharacter;
1119 if (key != null && key.equals("SS"))
1123 // sensible gap character
1128 // valid secondary structure AND no alternative label (e.g. ' B')
1129 if (annot.secondaryStructure > ' ' && ch.length() < 2)
1131 return annot.secondaryStructure;
1136 if (ch.length() == 0)
1140 else if (ch.length() == 1)
1144 else if (ch.length() > 1)
1151 public String print()
1153 out = new StringBuffer();
1154 out.append("# STOCKHOLM 1.0");
1155 out.append(newline);
1156 print(getSeqsAsArray(), false);
1159 out.append(newline);
1160 return out.toString();
1163 private static Hashtable typeIds = null;
1167 if (typeIds == null)
1169 typeIds = new Hashtable();
1170 typeIds.put("SS", "Secondary Structure");
1171 typeIds.put("SA", "Surface Accessibility");
1172 typeIds.put("TM", "transmembrane");
1173 typeIds.put("PP", "Posterior Probability");
1174 typeIds.put("LI", "ligand binding");
1175 typeIds.put("AS", "active site");
1176 typeIds.put("IN", "intron");
1177 typeIds.put("IR", "interacting residue");
1178 typeIds.put("AC", "accession");
1179 typeIds.put("OS", "organism");
1180 typeIds.put("CL", "class");
1181 typeIds.put("DE", "description");
1182 typeIds.put("DR", "reference");
1183 typeIds.put("LO", "look");
1184 typeIds.put("RF", "Reference Positions");
1189 protected static String id2type(String id)
1191 if (typeIds.containsKey(id))
1193 return (String) typeIds.get(id);
1196 "Warning : Unknown Stockholm annotation type code " + id);
1200 protected static String type2id(String type)
1203 Enumeration e = typeIds.keys();
1204 while (e.hasMoreElements())
1206 Object ll = e.nextElement();
1207 if (typeIds.get(ll).toString().equalsIgnoreCase(type))
1218 "Warning : Unknown Stockholm annotation type: " + type);
1223 * make a friendly ID string.
1226 * @return truncated dataName to after last '/'
1228 private String safeName(String dataName)
1231 while ((b = dataName.indexOf("/")) > -1 && b < dataName.length())
1233 dataName = dataName.substring(b + 1).trim();
1236 int e = (dataName.length() - dataName.indexOf(".")) + 1;
1237 dataName = dataName.substring(1, e).trim();