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 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 StringBuffer out; // output buffer
90 public StockholmFile()
95 * Creates a new StockholmFile object for output.
97 public StockholmFile(AlignmentI al)
102 public StockholmFile(String inFile, DataSourceType type)
108 public StockholmFile(FileParse source) throws IOException
114 public void initData()
120 * Parse a file in Stockholm format into Jalview's data model using VARNA
122 * @throws IOException
123 * If there is an error with the input file
125 public void parse_with_VARNA(java.io.File inFile) throws IOException
127 FileReader fr = null;
128 fr = new FileReader(inFile);
130 BufferedReader r = new BufferedReader(fr);
131 List<RNA> result = null;
134 result = RNAFactory.loadSecStrStockholm(r);
135 } catch (ExceptionUnmatchedClosingParentheses umcp)
137 errormessage = "Unmatched parentheses in annotation. Aborting ("
138 + umcp.getMessage() + ")";
139 throw new IOException(umcp);
141 // DEBUG System.out.println("this is the secondary scructure:"
143 SequenceI[] seqs = new SequenceI[result.size()];
145 for (int i = 0; i < result.size(); i++)
147 // DEBUG System.err.println("Processing i'th sequence in Stockholm file")
148 RNA current = result.get(i);
150 String seq = current.getSeq();
151 String rna = current.getStructDBN(true);
152 // DEBUG System.out.println(seq);
153 // DEBUG System.err.println(rna);
155 int end = seq.length() - 1;
156 id = safeName(getDataName());
157 seqs[i] = new Sequence(id, seq, begin, end);
158 String[] annot = new String[rna.length()];
159 Annotation[] ann = new Annotation[rna.length()];
160 for (int j = 0; j < rna.length(); j++)
162 annot[j] = rna.substring(j, j + 1);
166 for (int k = 0; k < rna.length(); k++)
168 ann[k] = new Annotation(annot[k], "", Rna.getRNASecStrucState(
169 annot[k]).charAt(0), 0f);
172 AlignmentAnnotation align = new AlignmentAnnotation("Sec. str.",
173 current.getID(), ann);
175 seqs[i].addAlignmentAnnotation(align);
176 seqs[i].setRNA(result.get(i));
177 this.annotations.addElement(align);
184 * Parse a file in Stockholm format into Jalview's data model. The file has to
185 * be passed at construction time
187 * @throws IOException
188 * If there is an error with the input file
191 public void parse() throws IOException
193 StringBuffer treeString = new StringBuffer();
194 String treeName = null;
195 // --------------- Variable Definitions -------------------
199 Hashtable seqAnn = new Hashtable(); // Sequence related annotations
200 LinkedHashMap<String, String> seqs = new LinkedHashMap<String, String>();
201 Regex p, r, rend, s, x;
202 // Temporary line for processing RNA annotation
203 // String RNAannot = "";
205 // ------------------ Parsing File ----------------------
206 // First, we have to check that this file has STOCKHOLM format, i.e. the
207 // first line must match
209 r = new Regex("# STOCKHOLM ([\\d\\.]+)");
210 if (!r.search(nextLine()))
212 throw new IOException(
214 .getString("exception.stockholm_invalid_format"));
218 version = r.stringMatched(1);
220 // logger.debug("Stockholm version: " + version);
223 // We define some Regexes here that will be used regularily later
224 rend = new Regex("^\\s*\\/\\/"); // Find the end of an alignment
225 p = new Regex("(\\S+)\\/(\\d+)\\-(\\d+)"); // split sequence id in
227 s = new Regex("(\\S+)\\s+(\\S*)\\s+(.*)"); // Parses annotation subtype
228 r = new Regex("#=(G[FSRC]?)\\s+(.*)"); // Finds any annotation line
229 x = new Regex("(\\S+)\\s+(\\S+)"); // split id from sequence
231 // Convert all bracket types to parentheses (necessary for passing to VARNA)
232 Regex openparen = new Regex("(<|\\[)", "(");
233 Regex closeparen = new Regex("(>|\\])", ")");
235 // Detect if file is RNA by looking for bracket types
236 Regex detectbrackets = new Regex("(<|>|\\[|\\]|\\(|\\))");
243 openparen.optimize();
244 closeparen.optimize();
246 while ((line = nextLine()) != null)
248 if (line.length() == 0)
252 if (rend.search(line))
254 // End of the alignment, pass stuff back
255 this.noSeqs = seqs.size();
257 String seqdb, dbsource = null;
258 Regex pf = new Regex("PF[0-9]{5}(.*)"); // Finds AC for Pfam
259 Regex rf = new Regex("RF[0-9]{5}(.*)"); // Finds AC for Rfam
260 if (getAlignmentProperty("AC") != null)
262 String dbType = getAlignmentProperty("AC").toString();
263 if (pf.search(dbType))
265 // PFAM Alignment - so references are typically from Uniprot
268 else if (rf.search(dbType))
273 // logger.debug("Number of sequences: " + this.noSeqs);
274 for (Map.Entry<String, String> skey : seqs.entrySet())
276 // logger.debug("Processing sequence " + acc);
277 String acc = skey.getKey();
278 String seq = skey.getValue();
279 if (maxLength < seq.length())
281 maxLength = seq.length();
287 * Retrieve hash of annotations for this accession Associate
288 * Annotation with accession
290 Hashtable accAnnotations = null;
292 if (seqAnn != null && seqAnn.containsKey(acc))
294 accAnnotations = (Hashtable) seqAnn.remove(acc);
295 // TODO: add structures to sequence
298 // Split accession in id and from/to
301 sid = p.stringMatched(1);
302 start = Integer.parseInt(p.stringMatched(2));
303 end = Integer.parseInt(p.stringMatched(3));
305 // logger.debug(sid + ", " + start + ", " + end);
307 Sequence seqO = new Sequence(sid, seq, start, end);
308 // Add Description (if any)
309 if (accAnnotations != null && accAnnotations.containsKey("DE"))
311 String desc = (String) accAnnotations.get("DE");
312 seqO.setDescription((desc == null) ? "" : desc);
314 // Add DB References (if any)
315 if (accAnnotations != null && accAnnotations.containsKey("DR"))
317 String dbr = (String) accAnnotations.get("DR");
318 if (dbr != null && dbr.indexOf(";") > -1)
320 String src = dbr.substring(0, dbr.indexOf(";"));
321 String acn = dbr.substring(dbr.indexOf(";") + 1);
322 jalview.util.DBRefUtils.parseToDbRef(seqO, src, "0", acn);
326 if (accAnnotations != null && accAnnotations.containsKey("AC"))
328 if (dbsource != null)
330 String dbr = (String) accAnnotations.get("AC");
333 // we could get very clever here - but for now - just try to
334 // guess accession type from source of alignment plus structure
336 guessDatabaseFor(seqO, dbr, dbsource);
340 // else - do what ? add the data anyway and prompt the user to
341 // specify what references these are ?
344 Hashtable features = null;
345 // We need to adjust the positions of all features to account for gaps
348 features = (Hashtable) accAnnotations.remove("features");
349 } catch (java.lang.NullPointerException e)
351 // loggerwarn("Getting Features for " + acc + ": " +
355 // if we have features
356 if (features != null)
358 int posmap[] = seqO.findPositionMap();
359 Enumeration i = features.keys();
360 while (i.hasMoreElements())
362 // TODO: parse out secondary structure annotation as annotation
364 // TODO: parse out scores as annotation row
365 // TODO: map coding region to core jalview feature types
366 String type = i.nextElement().toString();
367 Hashtable content = (Hashtable) features.remove(type);
369 // add alignment annotation for this feature
370 String key = type2id(type);
373 * have we added annotation rows for this type ?
375 boolean annotsAdded = false;
378 if (accAnnotations != null
379 && accAnnotations.containsKey(key))
381 Vector vv = (Vector) accAnnotations.get(key);
382 for (int ii = 0; ii < vv.size(); ii++)
385 AlignmentAnnotation an = (AlignmentAnnotation) vv
387 seqO.addAlignmentAnnotation(an);
393 Enumeration j = content.keys();
394 while (j.hasMoreElements())
396 String desc = j.nextElement().toString();
397 if (ANNOTATION.equals(desc) && annotsAdded)
399 // don't add features if we already added an annotation row
402 String ns = content.get(desc).toString();
403 char[] byChar = ns.toCharArray();
404 for (int k = 0; k < byChar.length; k++)
407 if (!(c == ' ' || c == '_' || c == '-' || c == '.')) // PFAM
414 int new_pos = posmap[k]; // look up nearest seqeunce
415 // position to this column
416 SequenceFeature feat = new SequenceFeature(type, desc,
417 new_pos, new_pos, null);
419 seqO.addSequenceFeature(feat);
429 // logger.debug("Adding seq " + acc + " from " + start + " to " + end
431 this.seqs.addElement(seqO);
433 return; // finished parsing this segment of source
435 else if (!r.search(line))
437 // System.err.println("Found sequence line: " + line);
439 // Split sequence in sequence and accession parts
442 // logger.error("Could not parse sequence line: " + line);
443 throw new IOException(MessageManager.formatMessage(
444 "exception.couldnt_parse_sequence_line",
445 new String[] { line }));
447 String ns = seqs.get(x.stringMatched(1));
452 ns += x.stringMatched(2);
454 seqs.put(x.stringMatched(1), ns);
458 String annType = r.stringMatched(1);
459 String annContent = r.stringMatched(2);
461 // System.err.println("type:" + annType + " content: " + annContent);
463 if (annType.equals("GF"))
466 * Generic per-File annotation, free text Magic features: #=GF NH
467 * <tree in New Hampshire eXtended format> #=GF TN <Unique identifier
468 * for the next tree> Pfam descriptions: 7. DESCRIPTION OF FIELDS
470 * Compulsory fields: ------------------
472 * AC Accession number: Accession number in form PFxxxxx.version or
473 * PBxxxxxx. ID Identification: One word name for family. DE
474 * Definition: Short description of family. AU Author: Authors of the
475 * entry. SE Source of seed: The source suggesting the seed members
476 * belong to one family. GA Gathering method: Search threshold to
477 * build the full alignment. TC Trusted Cutoff: Lowest sequence score
478 * and domain score of match in the full alignment. NC Noise Cutoff:
479 * Highest sequence score and domain score of match not in full
480 * alignment. TP Type: Type of family -- presently Family, Domain,
481 * Motif or Repeat. SQ Sequence: Number of sequences in alignment. AM
482 * Alignment Method The order ls and fs hits are aligned to the model
483 * to build the full align. // End of alignment.
485 * Optional fields: ----------------
487 * DC Database Comment: Comment about database reference. DR Database
488 * Reference: Reference to external database. RC Reference Comment:
489 * Comment about literature reference. RN Reference Number: Reference
490 * Number. RM Reference Medline: Eight digit medline UI number. RT
491 * Reference Title: Reference Title. RA Reference Author: Reference
492 * Author RL Reference Location: Journal location. PI Previous
493 * identifier: Record of all previous ID lines. KW Keywords: Keywords.
494 * CC Comment: Comments. NE Pfam accession: Indicates a nested domain.
495 * NL Location: Location of nested domains - sequence ID, start and
498 * Obsolete fields: ----------- AL Alignment method of seed: The
499 * method used to align the seed members.
501 // Let's save the annotations, maybe we'll be able to do something
502 // with them later...
503 Regex an = new Regex("(\\w+)\\s*(.*)");
504 if (an.search(annContent))
506 if (an.stringMatched(1).equals("NH"))
508 treeString.append(an.stringMatched(2));
510 else if (an.stringMatched(1).equals("TN"))
512 if (treeString.length() > 0)
514 if (treeName == null)
516 treeName = "Tree " + (getTreeCount() + 1);
518 addNewickTree(treeName, treeString.toString());
520 treeName = an.stringMatched(2);
521 treeString = new StringBuffer();
523 setAlignmentProperty(an.stringMatched(1), an.stringMatched(2));
526 else if (annType.equals("GS"))
528 // Generic per-Sequence annotation, free text
530 * Pfam uses these features: Feature Description ---------------------
531 * ----------- AC <accession> ACcession number DE <freetext>
532 * DEscription DR <db>; <accession>; Database Reference OS <organism>
533 * OrganiSm (species) OC <clade> Organism Classification (clade, etc.)
534 * LO <look> Look (Color, etc.)
536 if (s.search(annContent))
538 String acc = s.stringMatched(1);
539 String type = s.stringMatched(2);
540 String content = s.stringMatched(3);
541 // TODO: store DR in a vector.
542 // TODO: store AC according to generic file db annotation.
544 if (seqAnn.containsKey(acc))
546 ann = (Hashtable) seqAnn.get(acc);
550 ann = new Hashtable();
552 ann.put(type, content);
553 seqAnn.put(acc, ann);
557 // throw new IOException("Error parsing " + line);
558 System.err.println(">> missing annotation: " + line);
561 else if (annType.equals("GC"))
563 // Generic per-Column annotation, exactly 1 char per column
564 // always need a label.
565 if (x.search(annContent))
567 // parse out and create alignment annotation directly.
568 parseAnnotationRow(annotations, x.stringMatched(1),
572 else if (annType.equals("GR"))
574 // Generic per-Sequence AND per-Column markup, exactly 1 char per
577 * Feature Description Markup letters ------- -----------
578 * -------------- SS Secondary Structure [HGIEBTSCX] SA Surface
579 * Accessibility [0-9X] (0=0%-10%; ...; 9=90%-100%) TM TransMembrane
580 * [Mio] PP Posterior Probability [0-9*] (0=0.00-0.05; 1=0.05-0.15;
581 * *=0.95-1.00) LI LIgand binding [*] AS Active Site [*] IN INtron (in
584 if (s.search(annContent))
586 String acc = s.stringMatched(1);
587 String type = s.stringMatched(2);
588 String oseq = s.stringMatched(3);
590 * copy of annotation field that may be processed into whitespace chunks
592 String seq = new String(oseq);
595 // Get an object with all the annotations for this sequence
596 if (seqAnn.containsKey(acc))
598 // logger.debug("Found annotations for " + acc);
599 ann = (Hashtable) seqAnn.get(acc);
603 // logger.debug("Creating new annotations holder for " + acc);
604 ann = new Hashtable();
605 seqAnn.put(acc, ann);
608 // // start of block for appending annotation lines for wrapped
610 // TODO test structure, call parseAnnotationRow with vector from
611 // hashtable for specific sequence
614 // Get an object with all the content for an annotation
615 if (ann.containsKey("features"))
617 // logger.debug("Found features for " + acc);
618 features = (Hashtable) ann.get("features");
622 // logger.debug("Creating new features holder for " + acc);
623 features = new Hashtable();
624 ann.put("features", features);
628 if (features.containsKey(this.id2type(type)))
630 // logger.debug("Found content for " + this.id2type(type));
631 content = (Hashtable) features.get(this.id2type(type));
635 // logger.debug("Creating new content holder for " +
636 // this.id2type(type));
637 content = new Hashtable();
638 features.put(this.id2type(type), content);
640 String ns = (String) content.get(ANNOTATION);
646 // finally, append the annotation line
648 content.put(ANNOTATION, ns);
649 // // end of wrapped annotation block.
650 // // Now a new row is created with the current set of data
653 if (seqAnn.containsKey(acc))
655 strucAnn = (Hashtable) seqAnn.get(acc);
659 strucAnn = new Hashtable();
662 Vector<AlignmentAnnotation> newStruc = new Vector<AlignmentAnnotation>();
663 parseAnnotationRow(newStruc, type, ns);
664 for (AlignmentAnnotation alan : newStruc)
666 alan.visible = false;
668 // new annotation overwrites any existing annotation...
670 strucAnn.put(type, newStruc);
671 seqAnn.put(acc, strucAnn);
677 .println("Warning - couldn't parse sequence annotation row line:\n"
679 // throw new IOException("Error parsing " + line);
684 throw new IOException(MessageManager.formatMessage(
685 "exception.unknown_annotation_detected", new String[] {
686 annType, annContent }));
690 if (treeString.length() > 0)
692 if (treeName == null)
694 treeName = "Tree " + (1 + getTreeCount());
696 addNewickTree(treeName, treeString.toString());
701 * Demangle an accession string and guess the originating sequence database
702 * for a given sequence
705 * sequence to be annotated
707 * Accession string for sequence
709 * source database for alignment (PFAM or RFAM)
711 private void guessDatabaseFor(Sequence seqO, String dbr, String dbsource)
713 DBRefEntry dbrf = null;
714 List<DBRefEntry> dbrs = new ArrayList<DBRefEntry>();
715 String seqdb = "Unknown", sdbac = "" + dbr;
716 int st = -1, en = -1, p;
717 if ((st = sdbac.indexOf("/")) > -1)
719 String num, range = sdbac.substring(st + 1);
720 sdbac = sdbac.substring(0, st);
721 if ((p = range.indexOf("-")) > -1)
724 if (p < range.length())
726 num = range.substring(p).trim();
729 en = Integer.parseInt(num);
730 } catch (NumberFormatException x)
732 // could warn here that index is invalid
741 num = range.substring(0, p).trim();
744 st = Integer.parseInt(num);
745 } catch (NumberFormatException x)
747 // could warn here that index is invalid
751 if (dbsource.equals("PFAM"))
754 if (sdbac.indexOf(".") > -1)
756 // strip of last subdomain
757 sdbac = sdbac.substring(0, sdbac.indexOf("."));
758 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, seqdb, dbsource,
765 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, dbsource, dbsource,
774 seqdb = "EMBL"; // total guess - could be ENA, or something else these
776 if (sdbac.indexOf(".") > -1)
778 // strip off last subdomain
779 sdbac = sdbac.substring(0, sdbac.indexOf("."));
780 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, seqdb, dbsource,
788 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, dbsource, dbsource,
795 if (st != -1 && en != -1)
797 for (DBRefEntry d : dbrs)
799 jalview.util.MapList mp = new jalview.util.MapList(new int[] {
800 seqO.getStart(), seqO.getEnd() }, new int[] { st, en }, 1, 1);
801 jalview.datamodel.Mapping mping = new Mapping(mp);
807 protected static AlignmentAnnotation parseAnnotationRow(
808 Vector<AlignmentAnnotation> annotation, String label,
811 String convert1, convert2 = null;
813 // convert1 = OPEN_PAREN.replaceAll(annots);
814 // convert2 = CLOSE_PAREN.replaceAll(convert1);
815 // annots = convert2;
818 if (label.contains("_cons"))
820 type = (label.indexOf("_cons") == label.length() - 5) ? label
821 .substring(0, label.length() - 5) : label;
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 max = Math.max(max, s[in].getLength());
935 if (tmp.length() > maxid)
937 maxid = tmp.length();
939 if (s[in].getDBRefs() != null)
941 for (int idb = 0; idb < s[in].getDBRefs().length; idb++)
945 dataRef = new Hashtable();
948 String datAs1 = s[in].getDBRefs()[idb].getSource().toString()
950 + s[in].getDBRefs()[idb].getAccessionId().toString();
951 dataRef.put(tmp, datAs1);
959 // output database type
960 if (al.getProperties() != null)
962 if (!al.getProperties().isEmpty())
964 Enumeration key = al.getProperties().keys();
965 Enumeration val = al.getProperties().elements();
966 while (key.hasMoreElements())
968 out.append("#=GF " + key.nextElement() + " " + val.nextElement());
974 // output database accessions
977 Enumeration en = dataRef.keys();
978 while (en.hasMoreElements())
980 Object idd = en.nextElement();
981 String type = (String) dataRef.remove(idd);
982 out.append(new Format("%-" + (maxid - 2) + "s").form("#=GS "
983 + idd.toString() + " "));
984 if (type.contains("PFAM") || type.contains("RFAM"))
987 out.append(" AC " + type.substring(type.indexOf(";") + 1));
991 out.append(" DR " + type + " ");
997 // output annotations
998 while (i < s.length && s[i] != null)
1000 AlignmentAnnotation[] alAnot = s[i].getAnnotation();
1004 for (int j = 0; j < alAnot.length; j++)
1007 String key = type2id(alAnot[j].label);
1008 boolean isrna = alAnot[j].isValidStruc();
1012 // hardwire to secondary structure if there is RNA secondary
1013 // structure on the annotation
1022 // out.append("#=GR ");
1023 out.append(new Format("%-" + maxid + "s").form("#=GR "
1024 + printId(s[i], jvSuffix) + " " + key + " "));
1025 ann = alAnot[j].annotations;
1027 for (int k = 0; k < ann.length; k++)
1029 seq += outputCharacter(key, k, isrna, ann, s[i]);
1032 out.append(newline);
1036 out.append(new Format("%-" + maxid + "s")
1037 .form(printId(s[i], jvSuffix) + " "));
1038 out.append(s[i].getSequenceAsString());
1039 out.append(newline);
1043 // alignment annotation
1044 AlignmentAnnotation aa;
1045 if (al.getAlignmentAnnotation() != null)
1047 for (int ia = 0; ia < al.getAlignmentAnnotation().length; ia++)
1049 aa = al.getAlignmentAnnotation()[ia];
1050 if (aa.autoCalculated || !aa.visible || aa.sequenceRef != null)
1057 if (aa.label.equals("seq"))
1063 key = type2id(aa.label.toLowerCase());
1070 label = key + "_cons";
1077 label = label.replace(" ", "_");
1079 out.append(new Format("%-" + maxid + "s").form("#=GC " + label
1081 boolean isrna = aa.isValidStruc();
1082 for (int j = 0; j < aa.annotations.length; j++)
1084 seq += outputCharacter(key, j, isrna, aa.annotations, null);
1087 out.append(newline);
1092 out.append(newline);
1094 return out.toString();
1098 * add an annotation character to the output row
1107 private char outputCharacter(String key, int k, boolean isrna,
1108 Annotation[] ann, SequenceI sequenceI)
1111 Annotation annot = ann[k];
1112 String ch = (annot == null) ? ((sequenceI == null) ? "-" : Character
1113 .toString(sequenceI.getCharAt(k))) : annot.displayCharacter;
1114 if (key != null && key.equals("SS"))
1118 // sensible gap character
1123 // valid secondary structure AND no alternative label (e.g. ' B')
1124 if (annot.secondaryStructure > ' ' && ch.length() < 2)
1126 return annot.secondaryStructure;
1131 if (ch.length() == 0)
1135 else if (ch.length() == 1)
1139 else if (ch.length() > 1)
1146 public String print()
1148 out = new StringBuffer();
1149 out.append("# STOCKHOLM 1.0");
1150 out.append(newline);
1151 print(getSeqsAsArray(), false);
1154 out.append(newline);
1155 return out.toString();
1158 private static Hashtable typeIds = null;
1162 if (typeIds == null)
1164 typeIds = new Hashtable();
1165 typeIds.put("SS", "Secondary Structure");
1166 typeIds.put("SA", "Surface Accessibility");
1167 typeIds.put("TM", "transmembrane");
1168 typeIds.put("PP", "Posterior Probability");
1169 typeIds.put("LI", "ligand binding");
1170 typeIds.put("AS", "active site");
1171 typeIds.put("IN", "intron");
1172 typeIds.put("IR", "interacting residue");
1173 typeIds.put("AC", "accession");
1174 typeIds.put("OS", "organism");
1175 typeIds.put("CL", "class");
1176 typeIds.put("DE", "description");
1177 typeIds.put("DR", "reference");
1178 typeIds.put("LO", "look");
1179 typeIds.put("RF", "Reference Positions");
1184 protected static String id2type(String id)
1186 if (typeIds.containsKey(id))
1188 return (String) typeIds.get(id);
1190 System.err.println("Warning : Unknown Stockholm annotation type code "
1195 protected static String type2id(String type)
1198 Enumeration e = typeIds.keys();
1199 while (e.hasMoreElements())
1201 Object ll = e.nextElement();
1202 if (typeIds.get(ll).toString().equalsIgnoreCase(type))
1212 System.err.println("Warning : Unknown Stockholm annotation type: "
1218 * make a friendly ID string.
1221 * @return truncated dataName to after last '/'
1223 private String safeName(String dataName)
1226 while ((b = dataName.indexOf("/")) > -1 && b < dataName.length())
1228 dataName = dataName.substring(b + 1).trim();
1231 int e = (dataName.length() - dataName.indexOf(".")) + 1;
1232 dataName = dataName.substring(1, e).trim();