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
105 public void initData()
111 * Parse a file in Stockholm format into Jalview's data model using VARNA
113 * @throws IOException
114 * If there is an error with the input file
116 public void parse_with_VARNA(java.io.File inFile) throws IOException
118 FileReader fr = null;
119 fr = new FileReader(inFile);
121 BufferedReader r = new BufferedReader(fr);
125 result = RNAFactory.loadSecStrStockholm(r);
126 } catch (ExceptionUnmatchedClosingParentheses umcp)
128 errormessage = "Unmatched parentheses in annotation. Aborting ("
129 + umcp.getMessage() + ")";
130 throw new IOException(umcp);
132 // DEBUG System.out.println("this is the secondary scructure:"
134 SequenceI[] seqs = new SequenceI[result.size()];
136 for (int i = 0; i < result.size(); i++)
138 // DEBUG System.err.println("Processing i'th sequence in Stockholm file")
139 RNA current = result.get(i);
141 String seq = current.getSeq();
142 String rna = current.getStructDBN(true);
143 // DEBUG System.out.println(seq);
144 // DEBUG System.err.println(rna);
146 int end = seq.length() - 1;
147 id = safeName(getDataName());
148 seqs[i] = new Sequence(id, seq, begin, end);
149 String[] annot = new String[rna.length()];
150 Annotation[] ann = new Annotation[rna.length()];
151 for (int j = 0; j < rna.length(); j++)
153 annot[j] = rna.substring(j, j + 1);
157 for (int k = 0; k < rna.length(); k++)
159 ann[k] = new Annotation(annot[k], "",
160 jalview.schemes.ResidueProperties.getRNASecStrucState(
161 annot[k]).charAt(0), 0f);
164 AlignmentAnnotation align = new AlignmentAnnotation("Sec. str.",
165 current.getID(), ann);
167 seqs[i].addAlignmentAnnotation(align);
168 seqs[i].setRNA(result.get(i));
169 this.annotations.addElement(align);
176 * Parse a file in Stockholm format into Jalview's data model. The file has to
177 * be passed at construction time
179 * @throws IOException
180 * If there is an error with the input file
183 public void parse() throws IOException
185 StringBuffer treeString = new StringBuffer();
186 String treeName = null;
187 // --------------- Variable Definitions -------------------
191 Hashtable seqAnn = new Hashtable(); // Sequence related annotations
192 LinkedHashMap<String, String> seqs = new LinkedHashMap<String, String>();
193 Regex p, r, rend, s, x;
194 // Temporary line for processing RNA annotation
195 // String RNAannot = "";
197 // ------------------ Parsing File ----------------------
198 // First, we have to check that this file has STOCKHOLM format, i.e. the
199 // first line must match
201 r = new Regex("# STOCKHOLM ([\\d\\.]+)");
202 if (!r.search(nextLine()))
204 throw new IOException(
206 .getString("exception.stockholm_invalid_format"));
210 version = r.stringMatched(1);
212 // logger.debug("Stockholm version: " + version);
215 // We define some Regexes here that will be used regularily later
216 rend = new Regex("^\\s*\\/\\/"); // Find the end of an alignment
217 p = new Regex("(\\S+)\\/(\\d+)\\-(\\d+)"); // split sequence id in
219 s = new Regex("(\\S+)\\s+(\\S*)\\s+(.*)"); // Parses annotation subtype
220 r = new Regex("#=(G[FSRC]?)\\s+(.*)"); // Finds any annotation line
221 x = new Regex("(\\S+)\\s+(\\S+)"); // split id from sequence
223 // Convert all bracket types to parentheses (necessary for passing to VARNA)
224 Regex openparen = new Regex("(<|\\[)", "(");
225 Regex closeparen = new Regex("(>|\\])", ")");
227 // Detect if file is RNA by looking for bracket types
228 Regex detectbrackets = new Regex("(<|>|\\[|\\]|\\(|\\))");
235 openparen.optimize();
236 closeparen.optimize();
238 while ((line = nextLine()) != null)
240 if (line.length() == 0)
244 if (rend.search(line))
246 // End of the alignment, pass stuff back
247 this.noSeqs = seqs.size();
249 String seqdb, dbsource = null;
250 Regex pf = new Regex("PF[0-9]{5}(.*)"); // Finds AC for Pfam
251 Regex rf = new Regex("RF[0-9]{5}(.*)"); // Finds AC for Rfam
252 if (getAlignmentProperty("AC") != null)
254 String dbType = getAlignmentProperty("AC").toString();
255 if (pf.search(dbType))
257 // PFAM Alignment - so references are typically from Uniprot
260 else if (rf.search(dbType))
265 // logger.debug("Number of sequences: " + this.noSeqs);
266 for (Map.Entry<String, String> skey : seqs.entrySet())
268 // logger.debug("Processing sequence " + acc);
269 String acc = skey.getKey();
270 String seq = skey.getValue();
271 if (maxLength < seq.length())
273 maxLength = seq.length();
279 * Retrieve hash of annotations for this accession Associate
280 * Annotation with accession
282 Hashtable accAnnotations = null;
284 if (seqAnn != null && seqAnn.containsKey(acc))
286 accAnnotations = (Hashtable) seqAnn.remove(acc);
287 // TODO: add structures to sequence
290 // Split accession in id and from/to
293 sid = p.stringMatched(1);
294 start = Integer.parseInt(p.stringMatched(2));
295 end = Integer.parseInt(p.stringMatched(3));
297 // logger.debug(sid + ", " + start + ", " + end);
299 Sequence seqO = new Sequence(sid, seq, start, end);
300 // Add Description (if any)
301 if (accAnnotations != null && accAnnotations.containsKey("DE"))
303 String desc = (String) accAnnotations.get("DE");
304 seqO.setDescription((desc == null) ? "" : desc);
306 // Add DB References (if any)
307 if (accAnnotations != null && accAnnotations.containsKey("DR"))
309 String dbr = (String) accAnnotations.get("DR");
310 if (dbr != null && dbr.indexOf(";") > -1)
312 String src = dbr.substring(0, dbr.indexOf(";"));
313 String acn = dbr.substring(dbr.indexOf(";") + 1);
314 jalview.util.DBRefUtils.parseToDbRef(seqO, src, "0", acn);
318 if (accAnnotations != null && accAnnotations.containsKey("AC"))
320 if (dbsource != null)
322 String dbr = (String) accAnnotations.get("AC");
325 // we could get very clever here - but for now - just try to
326 // guess accession type from source of alignment plus structure
328 guessDatabaseFor(seqO, dbr, dbsource);
332 // else - do what ? add the data anyway and prompt the user to
333 // specify what references these are ?
336 Hashtable features = null;
337 // We need to adjust the positions of all features to account for gaps
340 features = (Hashtable) accAnnotations.remove("features");
341 } catch (java.lang.NullPointerException e)
343 // loggerwarn("Getting Features for " + acc + ": " +
347 // if we have features
348 if (features != null)
350 int posmap[] = seqO.findPositionMap();
351 Enumeration i = features.keys();
352 while (i.hasMoreElements())
354 // TODO: parse out secondary structure annotation as annotation
356 // TODO: parse out scores as annotation row
357 // TODO: map coding region to core jalview feature types
358 String type = i.nextElement().toString();
359 Hashtable content = (Hashtable) features.remove(type);
361 // add alignment annotation for this feature
362 String key = type2id(type);
365 if (accAnnotations != null
366 && accAnnotations.containsKey(key))
368 Vector vv = (Vector) accAnnotations.get(key);
369 for (int ii = 0; ii < vv.size(); ii++)
371 AlignmentAnnotation an = (AlignmentAnnotation) vv
373 seqO.addAlignmentAnnotation(an);
379 Enumeration j = content.keys();
380 while (j.hasMoreElements())
382 String desc = j.nextElement().toString();
383 String ns = content.get(desc).toString();
384 char[] byChar = ns.toCharArray();
385 for (int k = 0; k < byChar.length; k++)
388 if (!(c == ' ' || c == '_' || c == '-' || c == '.')) // PFAM
395 int new_pos = posmap[k]; // look up nearest seqeunce
396 // position to this column
397 SequenceFeature feat = new SequenceFeature(type, desc,
398 new_pos, new_pos, 0f, null);
400 seqO.addSequenceFeature(feat);
410 // logger.debug("Adding seq " + acc + " from " + start + " to " + end
412 this.seqs.addElement(seqO);
414 return; // finished parsing this segment of source
416 else if (!r.search(line))
418 // System.err.println("Found sequence line: " + line);
420 // Split sequence in sequence and accession parts
423 // logger.error("Could not parse sequence line: " + line);
424 throw new IOException(MessageManager.formatMessage(
425 "exception.couldnt_parse_sequence_line",
426 new String[] { line }));
428 String ns = seqs.get(x.stringMatched(1));
433 ns += x.stringMatched(2);
435 seqs.put(x.stringMatched(1), ns);
439 String annType = r.stringMatched(1);
440 String annContent = r.stringMatched(2);
442 // System.err.println("type:" + annType + " content: " + annContent);
444 if (annType.equals("GF"))
447 * Generic per-File annotation, free text Magic features: #=GF NH
448 * <tree in New Hampshire eXtended format> #=GF TN <Unique identifier
449 * for the next tree> Pfam descriptions: 7. DESCRIPTION OF FIELDS
451 * Compulsory fields: ------------------
453 * AC Accession number: Accession number in form PFxxxxx.version or
454 * PBxxxxxx. ID Identification: One word name for family. DE
455 * Definition: Short description of family. AU Author: Authors of the
456 * entry. SE Source of seed: The source suggesting the seed members
457 * belong to one family. GA Gathering method: Search threshold to
458 * build the full alignment. TC Trusted Cutoff: Lowest sequence score
459 * and domain score of match in the full alignment. NC Noise Cutoff:
460 * Highest sequence score and domain score of match not in full
461 * alignment. TP Type: Type of family -- presently Family, Domain,
462 * Motif or Repeat. SQ Sequence: Number of sequences in alignment. AM
463 * Alignment Method The order ls and fs hits are aligned to the model
464 * to build the full align. // End of alignment.
466 * Optional fields: ----------------
468 * DC Database Comment: Comment about database reference. DR Database
469 * Reference: Reference to external database. RC Reference Comment:
470 * Comment about literature reference. RN Reference Number: Reference
471 * Number. RM Reference Medline: Eight digit medline UI number. RT
472 * Reference Title: Reference Title. RA Reference Author: Reference
473 * Author RL Reference Location: Journal location. PI Previous
474 * identifier: Record of all previous ID lines. KW Keywords: Keywords.
475 * CC Comment: Comments. NE Pfam accession: Indicates a nested domain.
476 * NL Location: Location of nested domains - sequence ID, start and
479 * Obsolete fields: ----------- AL Alignment method of seed: The
480 * method used to align the seed members.
482 // Let's save the annotations, maybe we'll be able to do something
483 // with them later...
484 Regex an = new Regex("(\\w+)\\s*(.*)");
485 if (an.search(annContent))
487 if (an.stringMatched(1).equals("NH"))
489 treeString.append(an.stringMatched(2));
491 else if (an.stringMatched(1).equals("TN"))
493 if (treeString.length() > 0)
495 if (treeName == null)
497 treeName = "Tree " + (getTreeCount() + 1);
499 addNewickTree(treeName, treeString.toString());
501 treeName = an.stringMatched(2);
502 treeString = new StringBuffer();
504 setAlignmentProperty(an.stringMatched(1), an.stringMatched(2));
507 else if (annType.equals("GS"))
509 // Generic per-Sequence annotation, free text
511 * Pfam uses these features: Feature Description ---------------------
512 * ----------- AC <accession> ACcession number DE <freetext>
513 * DEscription DR <db>; <accession>; Database Reference OS <organism>
514 * OrganiSm (species) OC <clade> Organism Classification (clade, etc.)
515 * LO <look> Look (Color, etc.)
517 if (s.search(annContent))
519 String acc = s.stringMatched(1);
520 String type = s.stringMatched(2);
521 String content = s.stringMatched(3);
522 // TODO: store DR in a vector.
523 // TODO: store AC according to generic file db annotation.
525 if (seqAnn.containsKey(acc))
527 ann = (Hashtable) seqAnn.get(acc);
531 ann = new Hashtable();
533 ann.put(type, content);
534 seqAnn.put(acc, ann);
538 // throw new IOException("Error parsing " + line);
539 System.err.println(">> missing annotation: " + line);
542 else if (annType.equals("GC"))
544 // Generic per-Column annotation, exactly 1 char per column
545 // always need a label.
546 if (x.search(annContent))
548 // parse out and create alignment annotation directly.
549 parseAnnotationRow(annotations, x.stringMatched(1),
553 else if (annType.equals("GR"))
555 // Generic per-Sequence AND per-Column markup, exactly 1 char per
558 * Feature Description Markup letters ------- -----------
559 * -------------- SS Secondary Structure [HGIEBTSCX] SA Surface
560 * Accessibility [0-9X] (0=0%-10%; ...; 9=90%-100%) TM TransMembrane
561 * [Mio] PP Posterior Probability [0-9*] (0=0.00-0.05; 1=0.05-0.15;
562 * *=0.95-1.00) LI LIgand binding [*] AS Active Site [*] IN INtron (in
565 if (s.search(annContent))
567 String acc = s.stringMatched(1);
568 String type = s.stringMatched(2);
569 String seq = new String(s.stringMatched(3));
570 String description = null;
571 // Check for additional information about the current annotation
572 // We use a simple string tokenizer here for speed
573 StringTokenizer sep = new StringTokenizer(seq, " \t");
574 description = sep.nextToken();
575 if (sep.hasMoreTokens())
577 seq = sep.nextToken();
582 description = new String();
584 // sequence id with from-to fields
587 // Get an object with all the annotations for this sequence
588 if (seqAnn.containsKey(acc))
590 // logger.debug("Found annotations for " + acc);
591 ann = (Hashtable) seqAnn.get(acc);
595 // logger.debug("Creating new annotations holder for " + acc);
596 ann = new Hashtable();
597 seqAnn.put(acc, ann);
599 // TODO test structure, call parseAnnotationRow with vector from
600 // hashtable for specific sequence
602 // Get an object with all the content for an annotation
603 if (ann.containsKey("features"))
605 // logger.debug("Found features for " + acc);
606 features = (Hashtable) ann.get("features");
610 // logger.debug("Creating new features holder for " + acc);
611 features = new Hashtable();
612 ann.put("features", features);
616 if (features.containsKey(this.id2type(type)))
618 // logger.debug("Found content for " + this.id2type(type));
619 content = (Hashtable) features.get(this.id2type(type));
623 // logger.debug("Creating new content holder for " +
624 // this.id2type(type));
625 content = new Hashtable();
626 features.put(this.id2type(type), content);
628 String ns = (String) content.get(description);
634 content.put(description, ns);
636 // if(type.equals("SS")){
638 if (seqAnn.containsKey(acc))
640 strucAnn = (Hashtable) seqAnn.get(acc);
644 strucAnn = new Hashtable();
647 Vector<AlignmentAnnotation> newStruc = new Vector<AlignmentAnnotation>();
648 parseAnnotationRow(newStruc, type, ns);
649 for (AlignmentAnnotation alan : newStruc)
651 alan.visible = false;
653 // annotations.addAll(newStruc);
654 strucAnn.put(type, newStruc);
655 seqAnn.put(acc, strucAnn);
661 .println("Warning - couldn't parse sequence annotation row line:\n"
663 // throw new IOException("Error parsing " + line);
668 throw new IOException(MessageManager.formatMessage(
669 "exception.unknown_annotation_detected", new String[] {
670 annType, annContent }));
674 if (treeString.length() > 0)
676 if (treeName == null)
678 treeName = "Tree " + (1 + getTreeCount());
680 addNewickTree(treeName, treeString.toString());
685 * Demangle an accession string and guess the originating sequence database
686 * for a given sequence
689 * sequence to be annotated
691 * Accession string for sequence
693 * source database for alignment (PFAM or RFAM)
695 private void guessDatabaseFor(Sequence seqO, String dbr, String dbsource)
697 DBRefEntry dbrf = null;
698 List<DBRefEntry> dbrs = new ArrayList<DBRefEntry>();
699 String seqdb = "Unknown", sdbac = "" + dbr;
700 int st = -1, en = -1, p;
701 if ((st = sdbac.indexOf("/")) > -1)
703 String num, range = sdbac.substring(st + 1);
704 sdbac = sdbac.substring(0, st);
705 if ((p = range.indexOf("-")) > -1)
708 if (p < range.length())
710 num = range.substring(p).trim();
713 en = Integer.parseInt(num);
714 } catch (NumberFormatException x)
716 // could warn here that index is invalid
725 num = range.substring(0, p).trim();
728 st = Integer.parseInt(num);
729 } catch (NumberFormatException x)
731 // could warn here that index is invalid
735 if (dbsource.equals("PFAM"))
738 if (sdbac.indexOf(".") > -1)
740 // strip of last subdomain
741 sdbac = sdbac.substring(0, sdbac.indexOf("."));
742 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, seqdb, dbsource,
749 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, dbsource, dbsource,
758 seqdb = "EMBL"; // total guess - could be ENA, or something else these
760 if (sdbac.indexOf(".") > -1)
762 // strip off last subdomain
763 sdbac = sdbac.substring(0, sdbac.indexOf("."));
764 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, seqdb, dbsource,
772 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, dbsource, dbsource,
779 if (st != -1 && en != -1)
781 for (DBRefEntry d : dbrs)
783 jalview.util.MapList mp = new jalview.util.MapList(new int[] {
784 seqO.getStart(), seqO.getEnd() }, new int[] { st, en }, 1, 1);
785 jalview.datamodel.Mapping mping = new Mapping(mp);
791 protected static AlignmentAnnotation parseAnnotationRow(
792 Vector annotation, String label, String annots)
794 String convert1, convert2 = null;
796 // Convert all bracket types to parentheses
797 Regex openparen = new Regex("(<|\\[)", "(");
798 Regex closeparen = new Regex("(>|\\])", ")");
800 // Detect if file is RNA by looking for bracket types
801 Regex detectbrackets = new Regex("(<|>|\\[|\\]|\\(|\\))");
803 convert1 = openparen.replaceAll(annots);
804 convert2 = closeparen.replaceAll(convert1);
808 if (label.contains("_cons"))
810 type = (label.indexOf("_cons") == label.length() - 5) ? label
811 .substring(0, label.length() - 5) : label;
814 type = id2type(type);
815 if (type.equals("secondary structure"))
819 // decide on secondary structure or not.
820 Annotation[] els = new Annotation[annots.length()];
821 for (int i = 0; i < annots.length(); i++)
823 String pos = annots.substring(i, i + 1);
825 ann = new Annotation(pos, "", ' ', 0f); // 0f is 'valid' null - will not
829 // if (" .-_".indexOf(pos) == -1)
831 if (detectbrackets.search(pos))
833 ann.secondaryStructure = jalview.schemes.ResidueProperties
834 .getRNASecStrucState(pos).charAt(0);
838 ann.secondaryStructure = jalview.schemes.ResidueProperties
839 .getDssp3state(pos).charAt(0);
842 if (ann.secondaryStructure == pos.charAt(0))
844 ann.displayCharacter = ""; // null; // " ";
848 ann.displayCharacter = " " + ann.displayCharacter;
856 AlignmentAnnotation annot = null;
857 Enumeration e = annotation.elements();
858 while (e.hasMoreElements())
860 annot = (AlignmentAnnotation) e.nextElement();
861 if (annot.label.equals(type))
869 annot = new AlignmentAnnotation(type, type, els);
870 annotation.addElement(annot);
874 Annotation[] anns = new Annotation[annot.annotations.length
876 System.arraycopy(annot.annotations, 0, anns, 0,
877 annot.annotations.length);
878 System.arraycopy(els, 0, anns, annot.annotations.length, els.length);
879 annot.annotations = anns;
880 // System.out.println("else: ");
885 public String print(SequenceI[] s)
887 // find max length of id
891 Hashtable dataRef = null;
892 while ((in < s.length) && (s[in] != null))
894 String tmp = printId(s[in]);
895 if (s[in].getSequence().length > max)
897 max = s[in].getSequence().length;
900 if (tmp.length() > maxid)
902 maxid = tmp.length();
904 if (s[in].getDBRefs() != null)
906 for (int idb = 0; idb < s[in].getDBRefs().length; idb++)
910 dataRef = new Hashtable();
913 String datAs1 = s[in].getDBRefs()[idb].getSource().toString()
915 + s[in].getDBRefs()[idb].getAccessionId().toString();
916 dataRef.put(tmp, datAs1);
924 // output database type
925 if (al.getProperties() != null)
927 if (!al.getProperties().isEmpty())
929 Enumeration key = al.getProperties().keys();
930 Enumeration val = al.getProperties().elements();
931 while (key.hasMoreElements())
933 out.append("#=GF " + key.nextElement() + " " + val.nextElement());
939 // output database accessions
942 Enumeration en = dataRef.keys();
943 while (en.hasMoreElements())
945 Object idd = en.nextElement();
946 String type = (String) dataRef.remove(idd);
947 out.append(new Format("%-" + (maxid - 2) + "s").form("#=GS "
948 + idd.toString() + " "));
949 if (type.contains("PFAM") || type.contains("RFAM"))
952 out.append(" AC " + type.substring(type.indexOf(";") + 1));
956 out.append(" DR " + type + " ");
962 // output annotations
963 while (i < s.length && s[i] != null)
965 if (s[i].getDatasetSequence() != null)
967 SequenceI ds = s[i].getDatasetSequence();
968 AlignmentAnnotation[] alAnot;
971 alAnot = s[i].getAnnotation();
975 for (int j = 0; j < alAnot.length; j++)
977 if (ds.getSequenceFeatures() != null)
979 feature = ds.getSequenceFeatures()[0].type;
981 // ?bug - feature may still have previous loop value
982 String key = type2id(feature);
989 // out.append("#=GR ");
990 out.append(new Format("%-" + maxid + "s").form("#=GR "
991 + printId(s[i]) + " " + key + " "));
992 ann = alAnot[j].annotations;
993 boolean isrna = alAnot[j].isValidStruc();
995 for (int k = 0; k < ann.length; k++)
997 seq += outputCharacter(key, k, isrna, ann, s[i]);
1000 out.append(newline);
1005 out.append(new Format("%-" + maxid + "s").form(printId(s[i]) + " "));
1006 out.append(s[i].getSequenceAsString());
1007 out.append(newline);
1011 // alignment annotation
1012 AlignmentAnnotation aa;
1013 if (al.getAlignmentAnnotation() != null)
1015 for (int ia = 0; ia < al.getAlignmentAnnotation().length; ia++)
1017 aa = al.getAlignmentAnnotation()[ia];
1018 if (aa.autoCalculated || !aa.visible || aa.sequenceRef != null)
1025 if (aa.label.equals("seq"))
1031 key = type2id(aa.label.toLowerCase());
1038 label = key + "_cons";
1045 label = label.replace(" ", "_");
1047 out.append(new Format("%-" + maxid + "s").form("#=GC " + label
1049 boolean isrna = aa.isValidStruc();
1050 for (int j = 0; j < aa.annotations.length; j++)
1052 seq += outputCharacter(key, j, isrna, aa.annotations, null);
1055 out.append(newline);
1058 return out.toString();
1062 * add an annotation character to the output row
1071 private char outputCharacter(String key, int k, boolean isrna,
1072 Annotation[] ann, SequenceI sequenceI)
1075 Annotation annot = ann[k];
1076 String ch = (annot == null) ? ((sequenceI == null) ? "-" : Character
1077 .toString(sequenceI.getCharAt(k))) : annot.displayCharacter;
1078 if (key != null && key.equals("SS"))
1082 // sensible gap character if one is available or make one up
1083 return sequenceI == null ? '-' : sequenceI.getCharAt(k);
1087 // valid secondary structure AND no alternative label (e.g. ' B')
1088 if (annot.secondaryStructure > ' ' && ch.length() < 2)
1090 return annot.secondaryStructure;
1095 if (ch.length() == 0)
1099 else if (ch.length() == 1)
1103 else if (ch.length() > 1)
1111 public String print()
1113 out = new StringBuffer();
1114 out.append("# STOCKHOLM 1.0");
1115 out.append(newline);
1116 print(getSeqsAsArray());
1119 out.append(newline);
1120 return out.toString();
1123 private static Hashtable typeIds = null;
1126 if (typeIds == null)
1128 typeIds = new Hashtable();
1129 typeIds.put("SS", "secondary structure");
1130 typeIds.put("SA", "surface accessibility");
1131 typeIds.put("TM", "transmembrane");
1132 typeIds.put("PP", "posterior probability");
1133 typeIds.put("LI", "ligand binding");
1134 typeIds.put("AS", "active site");
1135 typeIds.put("IN", "intron");
1136 typeIds.put("IR", "interacting residue");
1137 typeIds.put("AC", "accession");
1138 typeIds.put("OS", "organism");
1139 typeIds.put("CL", "class");
1140 typeIds.put("DE", "description");
1141 typeIds.put("DR", "reference");
1142 typeIds.put("LO", "look");
1143 typeIds.put("RF", "reference positions");
1148 protected static String id2type(String id)
1150 if (typeIds.containsKey(id))
1152 return (String) typeIds.get(id);
1154 System.err.println("Warning : Unknown Stockholm annotation type code "
1159 protected static String type2id(String type)
1162 Enumeration e = typeIds.keys();
1163 while (e.hasMoreElements())
1165 Object ll = e.nextElement();
1166 if (typeIds.get(ll).toString().equals(type))
1176 System.err.println("Warning : Unknown Stockholm annotation type: "
1182 * make a friendly ID string.
1185 * @return truncated dataName to after last '/'
1187 private String safeName(String dataName)
1190 while ((b = dataName.indexOf("/")) > -1 && b < dataName.length())
1192 dataName = dataName.substring(b + 1).trim();
1195 int e = (dataName.length() - dataName.indexOf(".")) + 1;
1196 dataName = dataName.substring(1, e).trim();