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.datamodel.AlignmentAnnotation;
27 import jalview.datamodel.AlignmentI;
28 import jalview.datamodel.Annotation;
29 import jalview.datamodel.DBRefEntry;
30 import jalview.datamodel.Mapping;
31 import jalview.datamodel.Sequence;
32 import jalview.datamodel.SequenceFeature;
33 import jalview.datamodel.SequenceI;
34 import jalview.util.Format;
35 import jalview.util.MessageManager;
37 import java.io.BufferedReader;
38 import java.io.FileReader;
39 import java.io.IOException;
40 import java.util.ArrayList;
41 import java.util.Enumeration;
42 import java.util.Hashtable;
43 import java.util.LinkedHashMap;
44 import java.util.List;
46 import java.util.StringTokenizer;
47 import java.util.Vector;
49 import com.stevesoft.pat.Regex;
51 import fr.orsay.lri.varna.exceptions.ExceptionUnmatchedClosingParentheses;
52 import fr.orsay.lri.varna.factories.RNAFactory;
53 import fr.orsay.lri.varna.models.rna.RNA;
55 // import org.apache.log4j.*;
58 * This class is supposed to parse a Stockholm format file into Jalview There
59 * are TODOs in this class: we do not know what the database source and version
60 * is for the file when parsing the #GS= AC tag which associates accessions with
61 * sequences. Database references are also not parsed correctly: a separate
62 * reference string parser must be added to parse the database reference form
63 * into Jalview's local representation.
65 * @author bsb at sanger.ac.uk
66 * @author Natasha Shersnev (Dundee, UK) (Stockholm file writer)
67 * @author Lauren Lui (UCSC, USA) (RNA secondary structure annotation import as
69 * @author Anne Menard (Paris, FR) (VARNA parsing of Stockholm file data)
70 * @version 0.3 + jalview mods
73 public class StockholmFile extends AlignFile
75 // static Logger logger = Logger.getLogger("jalview.io.StockholmFile");
76 protected ArrayList<RNA> result;
78 StringBuffer out; // output buffer
82 public StockholmFile()
87 * Creates a new StockholmFile object for output.
89 public StockholmFile(AlignmentI al)
94 public StockholmFile(String inFile, String type) throws IOException
99 public StockholmFile(FileParse source) throws IOException
104 public void initData()
110 * Parse a file in Stockholm format into Jalview's data model using VARNA
112 * @throws IOException
113 * If there is an error with the input file
115 public void parse_with_VARNA(java.io.File inFile) throws IOException
117 FileReader fr = null;
118 fr = new FileReader(inFile);
120 BufferedReader r = new BufferedReader(fr);
124 result = RNAFactory.loadSecStrStockholm(r);
125 } catch (ExceptionUnmatchedClosingParentheses umcp)
127 errormessage = "Unmatched parentheses in annotation. Aborting ("
128 + umcp.getMessage() + ")";
129 throw new IOException(umcp);
131 // DEBUG System.out.println("this is the secondary scructure:"
133 SequenceI[] seqs = new SequenceI[result.size()];
135 for (int i = 0; i < result.size(); i++)
137 // DEBUG System.err.println("Processing i'th sequence in Stockholm file")
138 RNA current = result.get(i);
140 String seq = current.getSeq();
141 String rna = current.getStructDBN(true);
142 // DEBUG System.out.println(seq);
143 // DEBUG System.err.println(rna);
145 int end = seq.length() - 1;
146 id = safeName(getDataName());
147 seqs[i] = new Sequence(id, seq, begin, end);
148 String[] annot = new String[rna.length()];
149 Annotation[] ann = new Annotation[rna.length()];
150 for (int j = 0; j < rna.length(); j++)
152 annot[j] = rna.substring(j, j + 1);
156 for (int k = 0; k < rna.length(); k++)
158 ann[k] = new Annotation(annot[k], "",
159 jalview.schemes.ResidueProperties.getRNASecStrucState(
160 annot[k]).charAt(0), 0f);
163 AlignmentAnnotation align = new AlignmentAnnotation("Sec. str.",
164 current.getID(), ann);
166 seqs[i].addAlignmentAnnotation(align);
167 seqs[i].setRNA(result.get(i));
168 this.annotations.addElement(align);
175 * Parse a file in Stockholm format into Jalview's data model. The file has to
176 * be passed at construction time
178 * @throws IOException
179 * If there is an error with the input file
181 public void parse() throws IOException
183 StringBuffer treeString = new StringBuffer();
184 String treeName = null;
185 // --------------- Variable Definitions -------------------
189 Hashtable seqAnn = new Hashtable(); // Sequence related annotations
190 LinkedHashMap<String, String> seqs = new LinkedHashMap<String, String>();
191 Regex p, r, rend, s, x;
192 // Temporary line for processing RNA annotation
193 // String RNAannot = "";
195 // ------------------ Parsing File ----------------------
196 // First, we have to check that this file has STOCKHOLM format, i.e. the
197 // first line must match
199 r = new Regex("# STOCKHOLM ([\\d\\.]+)");
200 if (!r.search(nextLine()))
202 throw new IOException(
204 .getString("exception.stockholm_invalid_format"));
208 version = r.stringMatched(1);
210 // logger.debug("Stockholm version: " + version);
213 // We define some Regexes here that will be used regularily later
214 rend = new Regex("^\\s*\\/\\/"); // Find the end of an alignment
215 p = new Regex("(\\S+)\\/(\\d+)\\-(\\d+)"); // split sequence id in
217 s = new Regex("(\\S+)\\s+(\\S*)\\s+(.*)"); // Parses annotation subtype
218 r = new Regex("#=(G[FSRC]?)\\s+(.*)"); // Finds any annotation line
219 x = new Regex("(\\S+)\\s+(\\S+)"); // split id from sequence
221 // Convert all bracket types to parentheses (necessary for passing to VARNA)
222 Regex openparen = new Regex("(<|\\[)", "(");
223 Regex closeparen = new Regex("(>|\\])", ")");
225 // Detect if file is RNA by looking for bracket types
226 Regex detectbrackets = new Regex("(<|>|\\[|\\]|\\(|\\))");
233 openparen.optimize();
234 closeparen.optimize();
236 while ((line = nextLine()) != null)
238 if (line.length() == 0)
242 if (rend.search(line))
244 // End of the alignment, pass stuff back
245 this.noSeqs = seqs.size();
247 String seqdb, dbsource = null;
248 Regex pf = new Regex("PF[0-9]{5}(.*)"); // Finds AC for Pfam
249 Regex rf = new Regex("RF[0-9]{5}(.*)"); // Finds AC for Rfam
250 if (getAlignmentProperty("AC") != null)
252 String dbType = getAlignmentProperty("AC").toString();
253 if (pf.search(dbType))
255 // PFAM Alignment - so references are typically from Uniprot
258 else if (rf.search(dbType))
263 // logger.debug("Number of sequences: " + this.noSeqs);
264 for (Map.Entry<String, String> skey : seqs.entrySet())
266 // logger.debug("Processing sequence " + acc);
267 String acc = skey.getKey();
268 String seq = skey.getValue();
269 if (maxLength < seq.length())
271 maxLength = seq.length();
277 * Retrieve hash of annotations for this accession Associate
278 * Annotation with accession
280 Hashtable accAnnotations = null;
282 if (seqAnn != null && seqAnn.containsKey(acc))
284 accAnnotations = (Hashtable) seqAnn.remove(acc);
285 // TODO: add structures to sequence
288 // Split accession in id and from/to
291 sid = p.stringMatched(1);
292 start = Integer.parseInt(p.stringMatched(2));
293 end = Integer.parseInt(p.stringMatched(3));
295 // logger.debug(sid + ", " + start + ", " + end);
297 Sequence seqO = new Sequence(sid, seq, start, end);
298 // Add Description (if any)
299 if (accAnnotations != null && accAnnotations.containsKey("DE"))
301 String desc = (String) accAnnotations.get("DE");
302 seqO.setDescription((desc == null) ? "" : desc);
304 // Add DB References (if any)
305 if (accAnnotations != null && accAnnotations.containsKey("DR"))
307 String dbr = (String) accAnnotations.get("DR");
308 if (dbr != null && dbr.indexOf(";") > -1)
310 String src = dbr.substring(0, dbr.indexOf(";"));
311 String acn = dbr.substring(dbr.indexOf(";") + 1);
312 jalview.util.DBRefUtils.parseToDbRef(seqO, src, "0", acn);
316 if (accAnnotations != null && accAnnotations.containsKey("AC"))
318 if (dbsource != null)
320 String dbr = (String) accAnnotations.get("AC");
323 // we could get very clever here - but for now - just try to
324 // guess accession type from source of alignment plus structure
326 guessDatabaseFor(seqO, dbr, dbsource);
330 // else - do what ? add the data anyway and prompt the user to
331 // specify what references these are ?
334 Hashtable features = null;
335 // We need to adjust the positions of all features to account for gaps
338 features = (Hashtable) accAnnotations.remove("features");
339 } catch (java.lang.NullPointerException e)
341 // loggerwarn("Getting Features for " + acc + ": " +
345 // if we have features
346 if (features != null)
348 int posmap[] = seqO.findPositionMap();
349 Enumeration i = features.keys();
350 while (i.hasMoreElements())
352 // TODO: parse out secondary structure annotation as annotation
354 // TODO: parse out scores as annotation row
355 // TODO: map coding region to core jalview feature types
356 String type = i.nextElement().toString();
357 Hashtable content = (Hashtable) features.remove(type);
359 // add alignment annotation for this feature
360 String key = type2id(type);
363 if (accAnnotations != null
364 && accAnnotations.containsKey(key))
366 Vector vv = (Vector) accAnnotations.get(key);
367 for (int ii = 0; ii < vv.size(); ii++)
369 AlignmentAnnotation an = (AlignmentAnnotation) vv
371 seqO.addAlignmentAnnotation(an);
377 Enumeration j = content.keys();
378 while (j.hasMoreElements())
380 String desc = j.nextElement().toString();
381 String ns = content.get(desc).toString();
382 char[] byChar = ns.toCharArray();
383 for (int k = 0; k < byChar.length; k++)
386 if (!(c == ' ' || c == '_' || c == '-' || c == '.')) // PFAM
393 int new_pos = posmap[k]; // look up nearest seqeunce
394 // position to this column
395 SequenceFeature feat = new SequenceFeature(type, desc,
396 new_pos, new_pos, 0f, null);
398 seqO.addSequenceFeature(feat);
408 // logger.debug("Adding seq " + acc + " from " + start + " to " + end
410 this.seqs.addElement(seqO);
412 return; // finished parsing this segment of source
414 else if (!r.search(line))
416 // System.err.println("Found sequence line: " + line);
418 // Split sequence in sequence and accession parts
421 // logger.error("Could not parse sequence line: " + line);
422 throw new IOException(MessageManager.formatMessage(
423 "exception.couldnt_parse_sequence_line",
424 new String[] { line }));
426 String ns = seqs.get(x.stringMatched(1));
431 ns += x.stringMatched(2);
433 seqs.put(x.stringMatched(1), ns);
437 String annType = r.stringMatched(1);
438 String annContent = r.stringMatched(2);
440 // System.err.println("type:" + annType + " content: " + annContent);
442 if (annType.equals("GF"))
445 * Generic per-File annotation, free text Magic features: #=GF NH
446 * <tree in New Hampshire eXtended format> #=GF TN <Unique identifier
447 * for the next tree> Pfam descriptions: 7. DESCRIPTION OF FIELDS
449 * Compulsory fields: ------------------
451 * AC Accession number: Accession number in form PFxxxxx.version or
452 * PBxxxxxx. ID Identification: One word name for family. DE
453 * Definition: Short description of family. AU Author: Authors of the
454 * entry. SE Source of seed: The source suggesting the seed members
455 * belong to one family. GA Gathering method: Search threshold to
456 * build the full alignment. TC Trusted Cutoff: Lowest sequence score
457 * and domain score of match in the full alignment. NC Noise Cutoff:
458 * Highest sequence score and domain score of match not in full
459 * alignment. TP Type: Type of family -- presently Family, Domain,
460 * Motif or Repeat. SQ Sequence: Number of sequences in alignment. AM
461 * Alignment Method The order ls and fs hits are aligned to the model
462 * to build the full align. // End of alignment.
464 * Optional fields: ----------------
466 * DC Database Comment: Comment about database reference. DR Database
467 * Reference: Reference to external database. RC Reference Comment:
468 * Comment about literature reference. RN Reference Number: Reference
469 * Number. RM Reference Medline: Eight digit medline UI number. RT
470 * Reference Title: Reference Title. RA Reference Author: Reference
471 * Author RL Reference Location: Journal location. PI Previous
472 * identifier: Record of all previous ID lines. KW Keywords: Keywords.
473 * CC Comment: Comments. NE Pfam accession: Indicates a nested domain.
474 * NL Location: Location of nested domains - sequence ID, start and
477 * Obsolete fields: ----------- AL Alignment method of seed: The
478 * method used to align the seed members.
480 // Let's save the annotations, maybe we'll be able to do something
481 // with them later...
482 Regex an = new Regex("(\\w+)\\s*(.*)");
483 if (an.search(annContent))
485 if (an.stringMatched(1).equals("NH"))
487 treeString.append(an.stringMatched(2));
489 else if (an.stringMatched(1).equals("TN"))
491 if (treeString.length() > 0)
493 if (treeName == null)
495 treeName = "Tree " + (getTreeCount() + 1);
497 addNewickTree(treeName, treeString.toString());
499 treeName = an.stringMatched(2);
500 treeString = new StringBuffer();
502 setAlignmentProperty(an.stringMatched(1), an.stringMatched(2));
505 else if (annType.equals("GS"))
507 // Generic per-Sequence annotation, free text
509 * Pfam uses these features: Feature Description ---------------------
510 * ----------- AC <accession> ACcession number DE <freetext>
511 * DEscription DR <db>; <accession>; Database Reference OS <organism>
512 * OrganiSm (species) OC <clade> Organism Classification (clade, etc.)
513 * LO <look> Look (Color, etc.)
515 if (s.search(annContent))
517 String acc = s.stringMatched(1);
518 String type = s.stringMatched(2);
519 String content = s.stringMatched(3);
520 // TODO: store DR in a vector.
521 // TODO: store AC according to generic file db annotation.
523 if (seqAnn.containsKey(acc))
525 ann = (Hashtable) seqAnn.get(acc);
529 ann = new Hashtable();
531 ann.put(type, content);
532 seqAnn.put(acc, ann);
536 // throw new IOException(MessageManager.formatMessage(
537 // "exception.error_parsing_line", new String[] { line }));
538 System.err.println(">> missing annotation: " + line);
541 else if (annType.equals("GC"))
543 // Generic per-Column annotation, exactly 1 char per column
544 // always need a label.
545 if (x.search(annContent))
547 // parse out and create alignment annotation directly.
548 parseAnnotationRow(annotations, x.stringMatched(1),
552 else if (annType.equals("GR"))
554 // Generic per-Sequence AND per-Column markup, exactly 1 char per
557 * Feature Description Markup letters ------- -----------
558 * -------------- SS Secondary Structure [HGIEBTSCX] SA Surface
559 * Accessibility [0-9X] (0=0%-10%; ...; 9=90%-100%) TM TransMembrane
560 * [Mio] PP Posterior Probability [0-9*] (0=0.00-0.05; 1=0.05-0.15;
561 * *=0.95-1.00) LI LIgand binding [*] AS Active Site [*] IN INtron (in
564 if (s.search(annContent))
566 String acc = s.stringMatched(1);
567 String type = s.stringMatched(2);
568 String seq = new String(s.stringMatched(3));
569 String description = null;
570 // Check for additional information about the current annotation
571 // We use a simple string tokenizer here for speed
572 StringTokenizer sep = new StringTokenizer(seq, " \t");
573 description = sep.nextToken();
574 if (sep.hasMoreTokens())
576 seq = sep.nextToken();
581 description = new String();
583 // sequence id with from-to fields
586 // Get an object with all the annotations for this sequence
587 if (seqAnn.containsKey(acc))
589 // logger.debug("Found annotations for " + acc);
590 ann = (Hashtable) seqAnn.get(acc);
594 // logger.debug("Creating new annotations holder for " + acc);
595 ann = new Hashtable();
596 seqAnn.put(acc, ann);
598 // TODO test structure, call parseAnnotationRow with vector from
599 // hashtable for specific sequence
601 // Get an object with all the content for an annotation
602 if (ann.containsKey("features"))
604 // logger.debug("Found features for " + acc);
605 features = (Hashtable) ann.get("features");
609 // logger.debug("Creating new features holder for " + acc);
610 features = new Hashtable();
611 ann.put("features", features);
615 if (features.containsKey(this.id2type(type)))
617 // logger.debug("Found content for " + this.id2type(type));
618 content = (Hashtable) features.get(this.id2type(type));
622 // logger.debug("Creating new content holder for " +
623 // this.id2type(type));
624 content = new Hashtable();
625 features.put(this.id2type(type), content);
627 String ns = (String) content.get(description);
633 content.put(description, ns);
635 // if(type.equals("SS")){
637 if (seqAnn.containsKey(acc))
639 strucAnn = (Hashtable) seqAnn.get(acc);
643 strucAnn = new Hashtable();
646 Vector<AlignmentAnnotation> newStruc = new Vector<AlignmentAnnotation>();
647 parseAnnotationRow(newStruc, type, ns);
648 for (AlignmentAnnotation alan : newStruc)
650 alan.visible = false;
652 // annotations.addAll(newStruc);
653 strucAnn.put(type, newStruc);
654 seqAnn.put(acc, strucAnn);
660 .println("Warning - couldn't parse sequence annotation row line:\n"
662 // throw new IOException("Error parsing " + line);
667 throw new IOException(MessageManager.formatMessage(
668 "exception.unknown_annotation_detected", new String[] {
669 annType, annContent }));
673 if (treeString.length() > 0)
675 if (treeName == null)
677 treeName = "Tree " + (1 + getTreeCount());
679 addNewickTree(treeName, treeString.toString());
684 * Demangle an accession string and guess the originating sequence database
685 * for a given sequence
688 * sequence to be annotated
690 * Accession string for sequence
692 * source database for alignment (PFAM or RFAM)
694 private void guessDatabaseFor(Sequence seqO, String dbr, String dbsource)
696 DBRefEntry dbrf = null;
697 List<DBRefEntry> dbrs = new ArrayList<DBRefEntry>();
698 String seqdb = "Unknown", sdbac = "" + dbr;
699 int st = -1, en = -1, p;
700 if ((st = sdbac.indexOf("/")) > -1)
702 String num, range = sdbac.substring(st + 1);
703 sdbac = sdbac.substring(0, st);
704 if ((p = range.indexOf("-")) > -1)
707 if (p < range.length())
709 num = range.substring(p).trim();
712 en = Integer.parseInt(num);
713 } catch (NumberFormatException x)
715 // could warn here that index is invalid
724 num = range.substring(0, p).trim();
727 st = Integer.parseInt(num);
728 } catch (NumberFormatException x)
730 // could warn here that index is invalid
734 if (dbsource.equals("PFAM"))
737 if (sdbac.indexOf(".") > -1)
739 // strip of last subdomain
740 sdbac = sdbac.substring(0, sdbac.indexOf("."));
741 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, seqdb, dbsource,
748 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, dbsource, dbsource,
757 seqdb = "EMBL"; // total guess - could be ENA, or something else these
759 if (sdbac.indexOf(".") > -1)
761 // strip off last subdomain
762 sdbac = sdbac.substring(0, sdbac.indexOf("."));
763 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, seqdb, dbsource,
771 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, dbsource, dbsource,
778 if (st != -1 && en != -1)
780 for (DBRefEntry d : dbrs)
782 jalview.util.MapList mp = new jalview.util.MapList(new int[] {
783 seqO.getStart(), seqO.getEnd() }, new int[] { st, en }, 1, 1);
784 jalview.datamodel.Mapping mping = new Mapping(mp);
790 protected static AlignmentAnnotation parseAnnotationRow(
791 Vector annotation, String label, String annots)
793 String convert1, convert2 = null;
795 // Convert all bracket types to parentheses
796 Regex openparen = new Regex("(<|\\[)", "(");
797 Regex closeparen = new Regex("(>|\\])", ")");
799 // Detect if file is RNA by looking for bracket types
800 Regex detectbrackets = new Regex("(<|>|\\[|\\]|\\(|\\))");
802 convert1 = openparen.replaceAll(annots);
803 convert2 = closeparen.replaceAll(convert1);
807 if (label.contains("_cons"))
809 type = (label.indexOf("_cons") == label.length() - 5) ? label
810 .substring(0, label.length() - 5) : label;
813 type = id2type(type);
814 if (type.equals("secondary structure"))
818 // decide on secondary structure or not.
819 Annotation[] els = new Annotation[annots.length()];
820 for (int i = 0; i < annots.length(); i++)
822 String pos = annots.substring(i, i + 1);
824 ann = new Annotation(pos, "", ' ', 0f); // 0f is 'valid' null - will not
828 // if (" .-_".indexOf(pos) == -1)
830 if (detectbrackets.search(pos))
832 ann.secondaryStructure = jalview.schemes.ResidueProperties
833 .getRNASecStrucState(pos).charAt(0);
837 ann.secondaryStructure = jalview.schemes.ResidueProperties
838 .getDssp3state(pos).charAt(0);
841 if (ann.secondaryStructure == pos.charAt(0))
843 ann.displayCharacter = ""; // null; // " ";
847 ann.displayCharacter = " " + ann.displayCharacter;
855 AlignmentAnnotation annot = null;
856 Enumeration e = annotation.elements();
857 while (e.hasMoreElements())
859 annot = (AlignmentAnnotation) e.nextElement();
860 if (annot.label.equals(type))
868 annot = new AlignmentAnnotation(type, type, els);
869 annotation.addElement(annot);
873 Annotation[] anns = new Annotation[annot.annotations.length
875 System.arraycopy(annot.annotations, 0, anns, 0,
876 annot.annotations.length);
877 System.arraycopy(els, 0, anns, annot.annotations.length, els.length);
878 annot.annotations = anns;
879 // System.out.println("else: ");
884 public String print(SequenceI[] s)
886 // find max length of id
890 Hashtable dataRef = null;
891 while ((in < s.length) && (s[in] != null))
893 String tmp = printId(s[in]);
894 if (s[in].getSequence().length > max)
896 max = s[in].getSequence().length;
899 if (tmp.length() > maxid)
901 maxid = tmp.length();
903 if (s[in].getDBRefs() != null)
905 for (int idb = 0; idb < s[in].getDBRefs().length; idb++)
909 dataRef = new Hashtable();
912 String datAs1 = s[in].getDBRefs()[idb].getSource().toString()
914 + s[in].getDBRefs()[idb].getAccessionId().toString();
915 dataRef.put(tmp, datAs1);
923 // output database type
924 if (al.getProperties() != null)
926 if (!al.getProperties().isEmpty())
928 Enumeration key = al.getProperties().keys();
929 Enumeration val = al.getProperties().elements();
930 while (key.hasMoreElements())
932 out.append("#=GF " + key.nextElement() + " " + val.nextElement());
938 // output database accessions
941 Enumeration en = dataRef.keys();
942 while (en.hasMoreElements())
944 Object idd = en.nextElement();
945 String type = (String) dataRef.remove(idd);
946 out.append(new Format("%-" + (maxid - 2) + "s").form("#=GS "
947 + idd.toString() + " "));
948 if (type.contains("PFAM") || type.contains("RFAM"))
951 out.append(" AC " + type.substring(type.indexOf(";") + 1));
955 out.append(" DR " + type + " ");
961 // output annotations
962 while (i < s.length && s[i] != null)
964 if (s[i].getDatasetSequence() != null)
966 SequenceI ds = s[i].getDatasetSequence();
967 AlignmentAnnotation[] alAnot;
970 alAnot = s[i].getAnnotation();
974 for (int j = 0; j < alAnot.length; j++)
976 if (ds.getSequenceFeatures() != null)
978 feature = ds.getSequenceFeatures()[0].type;
980 // ?bug - feature may still have previous loop value
981 String key = type2id(feature);
988 // out.append("#=GR ");
989 out.append(new Format("%-" + maxid + "s").form("#=GR "
990 + printId(s[i]) + " " + key + " "));
991 ann = alAnot[j].annotations;
992 boolean isrna = alAnot[j].isValidStruc();
994 for (int k = 0; k < ann.length; k++)
996 seq += outputCharacter(key, k, isrna, ann, s[i]);
1004 out.append(new Format("%-" + maxid + "s").form(printId(s[i]) + " "));
1005 out.append(s[i].getSequenceAsString());
1006 out.append(newline);
1010 // alignment annotation
1011 AlignmentAnnotation aa;
1012 if (al.getAlignmentAnnotation() != null)
1014 for (int ia = 0; ia < al.getAlignmentAnnotation().length; ia++)
1016 aa = al.getAlignmentAnnotation()[ia];
1017 if (aa.autoCalculated || !aa.visible || aa.sequenceRef != null)
1024 if (aa.label.equals("seq"))
1030 key = type2id(aa.label.toLowerCase());
1037 label = key + "_cons";
1044 label = label.replace(" ", "_");
1046 out.append(new Format("%-" + maxid + "s").form("#=GC " + label
1048 boolean isrna = aa.isValidStruc();
1049 for (int j = 0; j < aa.annotations.length; j++)
1051 seq += outputCharacter(key, j, isrna, aa.annotations, null);
1054 out.append(newline);
1057 return out.toString();
1061 * add an annotation character to the output row
1070 private char outputCharacter(String key, int k, boolean isrna,
1071 Annotation[] ann, SequenceI sequenceI)
1074 Annotation annot = ann[k];
1075 String ch = (annot == null) ? ((sequenceI == null) ? "-" : Character
1076 .toString(sequenceI.getCharAt(k))) : annot.displayCharacter;
1077 if (key != null && key.equals("SS"))
1081 // sensible gap character if one is available or make one up
1082 return sequenceI == null ? '-' : sequenceI.getCharAt(k);
1086 // valid secondary structure AND no alternative label (e.g. ' B')
1087 if (annot.secondaryStructure > ' ' && ch.length() < 2)
1089 return annot.secondaryStructure;
1094 if (ch.length() == 0)
1098 else if (ch.length() == 1)
1102 else if (ch.length() > 1)
1109 public String print()
1111 out = new StringBuffer();
1112 out.append("# STOCKHOLM 1.0");
1113 out.append(newline);
1114 print(getSeqsAsArray());
1117 out.append(newline);
1118 return out.toString();
1121 private static Hashtable typeIds = null;
1124 if (typeIds == null)
1126 typeIds = new Hashtable();
1127 typeIds.put("SS", "secondary structure");
1128 typeIds.put("SA", "surface accessibility");
1129 typeIds.put("TM", "transmembrane");
1130 typeIds.put("PP", "posterior probability");
1131 typeIds.put("LI", "ligand binding");
1132 typeIds.put("AS", "active site");
1133 typeIds.put("IN", "intron");
1134 typeIds.put("IR", "interacting residue");
1135 typeIds.put("AC", "accession");
1136 typeIds.put("OS", "organism");
1137 typeIds.put("CL", "class");
1138 typeIds.put("DE", "description");
1139 typeIds.put("DR", "reference");
1140 typeIds.put("LO", "look");
1141 typeIds.put("RF", "reference positions");
1146 protected static String id2type(String id)
1148 if (typeIds.containsKey(id))
1150 return (String) typeIds.get(id);
1152 System.err.println("Warning : Unknown Stockholm annotation type code "
1157 protected static String type2id(String type)
1160 Enumeration e = typeIds.keys();
1161 while (e.hasMoreElements())
1163 Object ll = e.nextElement();
1164 if (typeIds.get(ll).toString().equals(type))
1174 System.err.println("Warning : Unknown Stockholm annotation type: "
1180 * make a friendly ID string.
1183 * @return truncated dataName to after last '/'
1185 private String safeName(String dataName)
1188 while ((b = dataName.indexOf("/")) > -1 && b < dataName.length())
1190 dataName = dataName.substring(b + 1).trim();
1193 int e = (dataName.length() - dataName.indexOf(".")) + 1;
1194 dataName = dataName.substring(1, e).trim();