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 char UNDERSCORE = '_';
81 private static final Regex OPEN_PAREN = new Regex("(<|\\[)", "(");
83 private static final Regex CLOSE_PAREN = new Regex("(>|\\])", ")");
85 // private static final Regex OPEN_PAREN = new Regex("(<|\\[)", "(");
86 // private static final Regex CLOSE_PAREN = new Regex("(>|\\])", ")");
88 public static final Regex DETECT_BRACKETS = new Regex(
89 "(<|>|\\[|\\]|\\(|\\)|\\{|\\})");
91 // WUSS extended symbols. Avoid ambiguity with protein SS annotations by using NOT_RNASS first.
92 public static final String RNASS_BRACKETS = "<>[](){}AaBbCcDdEeFfGgHhIiJjKkLlMmNnOoPpQqRrSsTtUuVvWwXxYyZz";
94 // use the following regex to decide an annotations (whole) line is NOT an RNA
95 // SS (it contains only E,H,e,h and other non-brace/non-alpha chars)
96 private static final Regex NOT_RNASS = new Regex(
97 "^[^<>[\\](){}A-DF-Za-df-z]*$");
99 StringBuffer out; // output buffer
101 private AlignmentI al;
103 public StockholmFile()
108 * Creates a new StockholmFile object for output
110 public StockholmFile(AlignmentI al)
115 public StockholmFile(String inFile, DataSourceType type)
121 public StockholmFile(FileParse source) throws IOException
127 public void initData()
133 * Parse a file in Stockholm format into Jalview's data model using VARNA
135 * @throws IOException
136 * If there is an error with the input file
138 public void parse_with_VARNA(java.io.File inFile) throws IOException
140 FileReader fr = null;
141 fr = new FileReader(inFile);
143 BufferedReader r = new BufferedReader(fr);
144 List<RNA> result = null;
147 result = RNAFactory.loadSecStrStockholm(r);
148 } catch (ExceptionUnmatchedClosingParentheses umcp)
150 errormessage = "Unmatched parentheses in annotation. Aborting ("
151 + umcp.getMessage() + ")";
152 throw new IOException(umcp);
154 // DEBUG System.out.println("this is the secondary scructure:"
156 SequenceI[] seqs = new SequenceI[result.size()];
158 for (int i = 0; i < result.size(); i++)
160 // DEBUG System.err.println("Processing i'th sequence in Stockholm file")
161 RNA current = result.get(i);
163 String seq = current.getSeq();
164 String rna = current.getStructDBN(true);
165 // DEBUG System.out.println(seq);
166 // DEBUG System.err.println(rna);
168 int end = seq.length() - 1;
169 id = safeName(getDataName());
170 seqs[i] = new Sequence(id, seq, begin, end);
171 String[] annot = new String[rna.length()];
172 Annotation[] ann = new Annotation[rna.length()];
173 for (int j = 0; j < rna.length(); j++)
175 annot[j] = rna.substring(j, j + 1);
179 for (int k = 0; k < rna.length(); k++)
181 ann[k] = new Annotation(annot[k], "",
182 Rna.getRNASecStrucState(annot[k]).charAt(0), 0f);
185 AlignmentAnnotation align = new AlignmentAnnotation("Sec. str.",
186 current.getID(), ann);
188 seqs[i].addAlignmentAnnotation(align);
189 seqs[i].setRNA(result.get(i));
190 this.annotations.addElement(align);
197 * Parse a file in Stockholm format into Jalview's data model. The file has to
198 * be passed at construction time
200 * @throws IOException
201 * If there is an error with the input file
204 public void parse() throws IOException
206 StringBuffer treeString = new StringBuffer();
207 String treeName = null;
208 // --------------- Variable Definitions -------------------
212 Hashtable seqAnn = new Hashtable(); // Sequence related annotations
213 LinkedHashMap<String, String> seqs = new LinkedHashMap<>();
214 Regex p, r, rend, s, x;
215 // Temporary line for processing RNA annotation
216 // String RNAannot = "";
218 // ------------------ Parsing File ----------------------
219 // First, we have to check that this file has STOCKHOLM format, i.e. the
220 // first line must match
222 r = new Regex("# STOCKHOLM ([\\d\\.]+)");
223 if (!r.search(nextLine()))
225 throw new IOException(MessageManager
226 .getString("exception.stockholm_invalid_format"));
230 version = r.stringMatched(1);
232 // logger.debug("Stockholm version: " + version);
235 // We define some Regexes here that will be used regularly later
236 rend = new Regex("^\\s*\\/\\/"); // Find the end of an alignment
237 p = new Regex("(\\S+)\\/(\\d+)\\-(\\d+)"); // split sequence id in
239 s = new Regex("(\\S+)\\s+(\\S*)\\s+(.*)"); // Parses annotation subtype
240 r = new Regex("#=(G[FSRC]?)\\s+(.*)"); // Finds any annotation line
241 x = new Regex("(\\S+)\\s+(\\S+)"); // split id from sequence
243 // Convert all bracket types to parentheses (necessary for passing to VARNA)
244 Regex openparen = new Regex("(<|\\[)", "(");
245 Regex closeparen = new Regex("(>|\\])", ")");
247 // Detect if file is RNA by looking for bracket types
248 Regex detectbrackets = new Regex("(<|>|\\[|\\]|\\(|\\))");
255 openparen.optimize();
256 closeparen.optimize();
258 while ((line = nextLine()) != null)
260 if (line.length() == 0)
264 if (rend.search(line))
266 // End of the alignment, pass stuff back
267 this.noSeqs = seqs.size();
269 String seqdb, dbsource = null;
270 Regex pf = new Regex("PF[0-9]{5}(.*)"); // Finds AC for Pfam
271 Regex rf = new Regex("RF[0-9]{5}(.*)"); // Finds AC for Rfam
272 if (getAlignmentProperty("AC") != null)
274 String dbType = getAlignmentProperty("AC").toString();
275 if (pf.search(dbType))
277 // PFAM Alignment - so references are typically from Uniprot
280 else if (rf.search(dbType))
285 // logger.debug("Number of sequences: " + this.noSeqs);
286 for (Map.Entry<String, String> skey : seqs.entrySet())
288 // logger.debug("Processing sequence " + acc);
289 String acc = skey.getKey();
290 String seq = skey.getValue();
291 if (maxLength < seq.length())
293 maxLength = seq.length();
299 * Retrieve hash of annotations for this accession Associate
300 * Annotation with accession
302 Hashtable accAnnotations = null;
304 if (seqAnn != null && seqAnn.containsKey(acc))
306 accAnnotations = (Hashtable) seqAnn.remove(acc);
307 // TODO: add structures to sequence
310 // Split accession in id and from/to
313 sid = p.stringMatched(1);
314 start = Integer.parseInt(p.stringMatched(2));
315 end = Integer.parseInt(p.stringMatched(3));
317 // logger.debug(sid + ", " + start + ", " + end);
319 Sequence seqO = new Sequence(sid, seq, start, end);
320 // Add Description (if any)
321 if (accAnnotations != null && accAnnotations.containsKey("DE"))
323 String desc = (String) accAnnotations.get("DE");
324 seqO.setDescription((desc == null) ? "" : desc);
326 // Add DB References (if any)
327 if (accAnnotations != null && accAnnotations.containsKey("DR"))
329 String dbr = (String) accAnnotations.get("DR");
330 if (dbr != null && dbr.indexOf(";") > -1)
332 String src = dbr.substring(0, dbr.indexOf(";"));
333 String acn = dbr.substring(dbr.indexOf(";") + 1);
334 jalview.util.DBRefUtils.parseToDbRef(seqO, src, "0", acn);
338 if (accAnnotations != null && accAnnotations.containsKey("AC"))
340 if (dbsource != null)
342 String dbr = (String) accAnnotations.get("AC");
345 // we could get very clever here - but for now - just try to
346 // guess accession type from source of alignment plus structure
348 guessDatabaseFor(seqO, dbr, dbsource);
352 // else - do what ? add the data anyway and prompt the user to
353 // specify what references these are ?
356 Hashtable features = null;
357 // We need to adjust the positions of all features to account for gaps
360 features = (Hashtable) accAnnotations.remove("features");
361 } catch (java.lang.NullPointerException e)
363 // loggerwarn("Getting Features for " + acc + ": " +
367 // if we have features
368 if (features != null)
370 int posmap[] = seqO.findPositionMap();
371 Enumeration i = features.keys();
372 while (i.hasMoreElements())
374 // TODO: parse out secondary structure annotation as annotation
376 // TODO: parse out scores as annotation row
377 // TODO: map coding region to core jalview feature types
378 String type = i.nextElement().toString();
379 Hashtable content = (Hashtable) features.remove(type);
381 // add alignment annotation for this feature
382 String key = type2id(type);
385 * have we added annotation rows for this type ?
387 boolean annotsAdded = false;
390 if (accAnnotations != null
391 && accAnnotations.containsKey(key))
393 Vector vv = (Vector) accAnnotations.get(key);
394 for (int ii = 0; ii < vv.size(); ii++)
397 AlignmentAnnotation an = (AlignmentAnnotation) vv
399 seqO.addAlignmentAnnotation(an);
405 Enumeration j = content.keys();
406 while (j.hasMoreElements())
408 String desc = j.nextElement().toString();
409 if (ANNOTATION.equals(desc) && annotsAdded)
411 // don't add features if we already added an annotation row
414 String ns = content.get(desc).toString();
415 char[] byChar = ns.toCharArray();
416 for (int k = 0; k < byChar.length; k++)
419 if (!(c == ' ' || c == '_' || c == '-' || c == '.')) // PFAM
426 int new_pos = posmap[k]; // look up nearest seqeunce
427 // position to this column
428 SequenceFeature feat = new SequenceFeature(type, desc,
429 new_pos, new_pos, null);
431 seqO.addSequenceFeature(feat);
441 // logger.debug("Adding seq " + acc + " from " + start + " to " + end
443 this.seqs.addElement(seqO);
445 return; // finished parsing this segment of source
447 else if (!r.search(line))
449 // System.err.println("Found sequence line: " + line);
451 // Split sequence in sequence and accession parts
454 // logger.error("Could not parse sequence line: " + line);
455 throw new IOException(MessageManager.formatMessage(
456 "exception.couldnt_parse_sequence_line", new String[]
459 String ns = seqs.get(x.stringMatched(1));
464 ns += x.stringMatched(2);
466 seqs.put(x.stringMatched(1), ns);
470 String annType = r.stringMatched(1);
471 String annContent = r.stringMatched(2);
473 // System.err.println("type:" + annType + " content: " + annContent);
475 if (annType.equals("GF"))
478 * Generic per-File annotation, free text Magic features: #=GF NH
479 * <tree in New Hampshire eXtended format> #=GF TN <Unique identifier
480 * for the next tree> Pfam descriptions: 7. DESCRIPTION OF FIELDS
482 * Compulsory fields: ------------------
484 * AC Accession number: Accession number in form PFxxxxx.version or
485 * PBxxxxxx. ID Identification: One word name for family. DE
486 * Definition: Short description of family. AU Author: Authors of the
487 * entry. SE Source of seed: The source suggesting the seed members
488 * belong to one family. GA Gathering method: Search threshold to
489 * build the full alignment. TC Trusted Cutoff: Lowest sequence score
490 * and domain score of match in the full alignment. NC Noise Cutoff:
491 * Highest sequence score and domain score of match not in full
492 * alignment. TP Type: Type of family -- presently Family, Domain,
493 * Motif or Repeat. SQ Sequence: Number of sequences in alignment. AM
494 * Alignment Method The order ls and fs hits are aligned to the model
495 * to build the full align. // End of alignment.
497 * Optional fields: ----------------
499 * DC Database Comment: Comment about database reference. DR Database
500 * Reference: Reference to external database. RC Reference Comment:
501 * Comment about literature reference. RN Reference Number: Reference
502 * Number. RM Reference Medline: Eight digit medline UI number. RT
503 * Reference Title: Reference Title. RA Reference Author: Reference
504 * Author RL Reference Location: Journal location. PI Previous
505 * identifier: Record of all previous ID lines. KW Keywords: Keywords.
506 * CC Comment: Comments. NE Pfam accession: Indicates a nested domain.
507 * NL Location: Location of nested domains - sequence ID, start and
510 * Obsolete fields: ----------- AL Alignment method of seed: The
511 * method used to align the seed members.
513 // Let's save the annotations, maybe we'll be able to do something
514 // with them later...
515 Regex an = new Regex("(\\w+)\\s*(.*)");
516 if (an.search(annContent))
518 if (an.stringMatched(1).equals("NH"))
520 treeString.append(an.stringMatched(2));
522 else if (an.stringMatched(1).equals("TN"))
524 if (treeString.length() > 0)
526 if (treeName == null)
528 treeName = "Tree " + (getTreeCount() + 1);
530 addNewickTree(treeName, treeString.toString());
532 treeName = an.stringMatched(2);
533 treeString = new StringBuffer();
535 setAlignmentProperty(an.stringMatched(1), an.stringMatched(2));
538 else if (annType.equals("GS"))
540 // Generic per-Sequence annotation, free text
542 * Pfam uses these features: Feature Description ---------------------
543 * ----------- AC <accession> ACcession number DE <freetext>
544 * DEscription DR <db>; <accession>; Database Reference OS <organism>
545 * OrganiSm (species) OC <clade> Organism Classification (clade, etc.)
546 * LO <look> Look (Color, etc.)
548 if (s.search(annContent))
550 String acc = s.stringMatched(1);
551 String type = s.stringMatched(2);
552 String content = s.stringMatched(3);
553 // TODO: store DR in a vector.
554 // TODO: store AC according to generic file db annotation.
556 if (seqAnn.containsKey(acc))
558 ann = (Hashtable) seqAnn.get(acc);
562 ann = new Hashtable();
564 ann.put(type, content);
565 seqAnn.put(acc, ann);
569 // throw new IOException("Error parsing " + line);
570 System.err.println(">> missing annotation: " + line);
573 else if (annType.equals("GC"))
575 // Generic per-Column annotation, exactly 1 char per column
576 // always need a label.
577 if (x.search(annContent))
579 // parse out and create alignment annotation directly.
580 parseAnnotationRow(annotations, x.stringMatched(1),
584 else if (annType.equals("GR"))
586 // Generic per-Sequence AND per-Column markup, exactly 1 char per
589 * Feature Description Markup letters ------- -----------
590 * -------------- SS Secondary Structure [HGIEBTSCX] SA Surface
591 * Accessibility [0-9X] (0=0%-10%; ...; 9=90%-100%) TM TransMembrane
592 * [Mio] PP Posterior Probability [0-9*] (0=0.00-0.05; 1=0.05-0.15;
593 * *=0.95-1.00) LI LIgand binding [*] AS Active Site [*] IN INtron (in
596 if (s.search(annContent))
598 String acc = s.stringMatched(1);
599 String type = s.stringMatched(2);
600 String oseq = s.stringMatched(3);
602 * copy of annotation field that may be processed into whitespace chunks
604 String seq = new String(oseq);
607 // Get an object with all the annotations for this sequence
608 if (seqAnn.containsKey(acc))
610 // logger.debug("Found annotations for " + acc);
611 ann = (Hashtable) seqAnn.get(acc);
615 // logger.debug("Creating new annotations holder for " + acc);
616 ann = new Hashtable();
617 seqAnn.put(acc, ann);
620 // // start of block for appending annotation lines for wrapped
622 // TODO test structure, call parseAnnotationRow with vector from
623 // hashtable for specific sequence
626 // Get an object with all the content for an annotation
627 if (ann.containsKey("features"))
629 // logger.debug("Found features for " + acc);
630 features = (Hashtable) ann.get("features");
634 // logger.debug("Creating new features holder for " + acc);
635 features = new Hashtable();
636 ann.put("features", features);
640 if (features.containsKey(this.id2type(type)))
642 // logger.debug("Found content for " + this.id2type(type));
643 content = (Hashtable) features
644 .get(this.id2type(type));
648 // logger.debug("Creating new content holder for " +
649 // this.id2type(type));
650 content = new Hashtable();
651 features.put(id2type(type), content);
653 String ns = (String) content.get(ANNOTATION);
659 // finally, append the annotation line
661 content.put(ANNOTATION, ns);
662 // // end of wrapped annotation block.
663 // // Now a new row is created with the current set of data
666 if (seqAnn.containsKey(acc))
668 strucAnn = (Hashtable) seqAnn.get(acc);
672 strucAnn = new Hashtable();
675 Vector<AlignmentAnnotation> newStruc = new Vector<>();
676 parseAnnotationRow(newStruc, type, ns);
677 for (AlignmentAnnotation alan : newStruc)
679 alan.visible = false;
681 // new annotation overwrites any existing annotation...
683 strucAnn.put(type, newStruc);
684 seqAnn.put(acc, strucAnn);
690 "Warning - couldn't parse sequence annotation row line:\n"
692 // throw new IOException("Error parsing " + line);
697 throw new IOException(MessageManager.formatMessage(
698 "exception.unknown_annotation_detected", new String[]
699 { annType, annContent }));
703 if (treeString.length() > 0)
705 if (treeName == null)
707 treeName = "Tree " + (1 + getTreeCount());
709 addNewickTree(treeName, treeString.toString());
714 * Demangle an accession string and guess the originating sequence database
715 * for a given sequence
718 * sequence to be annotated
720 * Accession string for sequence
722 * source database for alignment (PFAM or RFAM)
724 private void guessDatabaseFor(Sequence seqO, String dbr, String dbsource)
726 DBRefEntry dbrf = null;
727 List<DBRefEntry> dbrs = new ArrayList<>();
728 String seqdb = "Unknown", sdbac = "" + dbr;
729 int st = -1, en = -1, p;
730 if ((st = sdbac.indexOf("/")) > -1)
732 String num, range = sdbac.substring(st + 1);
733 sdbac = sdbac.substring(0, st);
734 if ((p = range.indexOf("-")) > -1)
737 if (p < range.length())
739 num = range.substring(p).trim();
742 en = Integer.parseInt(num);
743 } catch (NumberFormatException x)
745 // could warn here that index is invalid
754 num = range.substring(0, p).trim();
757 st = Integer.parseInt(num);
758 } catch (NumberFormatException x)
760 // could warn here that index is invalid
764 if (dbsource.equals("PFAM"))
767 if (sdbac.indexOf(".") > -1)
769 // strip of last subdomain
770 sdbac = sdbac.substring(0, sdbac.indexOf("."));
771 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, seqdb, dbsource,
778 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, dbsource, dbsource,
787 seqdb = "EMBL"; // total guess - could be ENA, or something else these
789 if (sdbac.indexOf(".") > -1)
791 // strip off last subdomain
792 sdbac = sdbac.substring(0, sdbac.indexOf("."));
793 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, seqdb, dbsource,
801 dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, dbsource, dbsource,
808 if (st != -1 && en != -1)
810 for (DBRefEntry d : dbrs)
812 jalview.util.MapList mp = new jalview.util.MapList(
814 { seqO.getStart(), seqO.getEnd() }, new int[] { st, en }, 1,
816 jalview.datamodel.Mapping mping = new Mapping(mp);
822 protected static AlignmentAnnotation parseAnnotationRow(
823 Vector<AlignmentAnnotation> annotation, String label,
826 String convert1, convert2 = null;
827 // String convert1 = OPEN_PAREN.replaceAll(annots);
828 // String convert2 = CLOSE_PAREN.replaceAll(convert1);
829 // annots = convert2;
832 if (label.contains("_cons"))
834 type = (label.indexOf("_cons") == label.length() - 5)
835 ? label.substring(0, label.length() - 5)
838 boolean ss = false, posterior = false;
839 type = id2type(type);
841 boolean isrnass = false;
843 if (type.equalsIgnoreCase("secondary structure"))
846 isrnass = !NOT_RNASS.search(annots); // sorry about the double negative
847 // here (it's easier for dealing with
848 // other non-alpha-non-brace chars)
850 if (type.equalsIgnoreCase("posterior probability"))
854 // decide on secondary structure or not.
855 Annotation[] els = new Annotation[annots.length()];
856 for (int i = 0; i < annots.length(); i++)
858 String pos = annots.substring(i, i + 1);
859 if (UNDERSCORE == pos.charAt(0))
864 ann = new Annotation(pos, "", ' ', 0f); // 0f is 'valid' null - will not
868 // if (" .-_".indexOf(pos) == -1)
870 if (isrnass && RNASS_BRACKETS.indexOf(pos) >= 0)
872 ann.secondaryStructure = Rna.getRNASecStrucState(pos).charAt(0);
873 ann.displayCharacter = "" + pos.charAt(0);
877 ann.secondaryStructure = ResidueProperties.getDssp3state(pos)
880 if (ann.secondaryStructure == pos.charAt(0))
882 ann.displayCharacter = ""; // null; // " ";
886 ann.displayCharacter = " " + ann.displayCharacter;
892 if (posterior && !ann.isWhitespace()
893 && !Comparison.isGap(pos.charAt(0)))
896 // symbol encodes values - 0..*==0..10
897 if (pos.charAt(0) == '*')
903 val = pos.charAt(0) - '0';
914 AlignmentAnnotation annot = null;
915 Enumeration<AlignmentAnnotation> e = annotation.elements();
916 while (e.hasMoreElements())
918 annot = e.nextElement();
919 if (annot.label.equals(type))
927 annot = new AlignmentAnnotation(type, type, els);
928 annotation.addElement(annot);
932 Annotation[] anns = new Annotation[annot.annotations.length
934 System.arraycopy(annot.annotations, 0, anns, 0,
935 annot.annotations.length);
936 System.arraycopy(els, 0, anns, annot.annotations.length, els.length);
937 annot.annotations = anns;
938 // System.out.println("else: ");
944 public String print(SequenceI[] s, boolean jvSuffix)
946 out = new StringBuffer();
947 out.append("# STOCKHOLM 1.0");
950 // find max length of id
954 Hashtable dataRef = null;
955 while ((in < s.length) && (s[in] != null))
957 String tmp = printId(s[in], jvSuffix);
958 max = Math.max(max, s[in].getLength());
960 if (tmp.length() > maxid)
962 maxid = tmp.length();
964 if (s[in].getDBRefs() != null)
966 for (int idb = 0; idb < s[in].getDBRefs().length; idb++)
970 dataRef = new Hashtable();
973 String datAs1 = s[in].getDBRefs()[idb].getSource().toString()
975 + s[in].getDBRefs()[idb].getAccessionId().toString();
976 dataRef.put(tmp, datAs1);
984 // output database type
985 if (al.getProperties() != null)
987 if (!al.getProperties().isEmpty())
989 Enumeration key = al.getProperties().keys();
990 Enumeration val = al.getProperties().elements();
991 while (key.hasMoreElements())
993 out.append("#=GF " + key.nextElement() + " " + val.nextElement());
999 // output database accessions
1000 if (dataRef != null)
1002 Enumeration en = dataRef.keys();
1003 while (en.hasMoreElements())
1005 Object idd = en.nextElement();
1006 String type = (String) dataRef.remove(idd);
1007 out.append(new Format("%-" + (maxid - 2) + "s")
1008 .form("#=GS " + idd.toString() + " "));
1009 if (type.contains("PFAM") || type.contains("RFAM"))
1012 out.append(" AC " + type.substring(type.indexOf(";") + 1));
1016 out.append(" DR " + type + " ");
1018 out.append(newline);
1022 // output annotations
1023 while (i < s.length && s[i] != null)
1025 AlignmentAnnotation[] alAnot = s[i].getAnnotation();
1030 for (int j = 0; j < alAnot.length; j++)
1032 if (alAnot[j].annotations != null)
1034 String key = type2id(alAnot[j].label);
1035 boolean isrna = alAnot[j].isValidStruc();
1039 // hardwire to secondary structure if there is RNA secondary
1040 // structure on the annotation
1049 // out.append("#=GR ");
1050 out.append(new Format("%-" + maxid + "s").form(
1051 "#=GR " + printId(s[i], jvSuffix) + " " + key + " "));
1052 ann = alAnot[j].annotations;
1054 for (int k = 0; k < ann.length; k++)
1056 seq += outputCharacter(key, k, isrna, ann, s[i]);
1059 out.append(newline);
1065 out.append(new Format("%-" + maxid + "s")
1066 .form(printId(s[i], jvSuffix) + " "));
1067 out.append(s[i].getSequenceAsString());
1068 out.append(newline);
1072 // alignment annotation
1073 AlignmentAnnotation aa;
1074 if (al.getAlignmentAnnotation() != null)
1076 for (int ia = 0; ia < al.getAlignmentAnnotation().length; ia++)
1078 aa = al.getAlignmentAnnotation()[ia];
1079 if (aa.autoCalculated || !aa.visible || aa.sequenceRef != null)
1086 if (aa.label.equals("seq"))
1092 key = type2id(aa.label.toLowerCase());
1099 label = key + "_cons";
1106 label = label.replace(" ", "_");
1109 new Format("%-" + maxid + "s").form("#=GC " + label + " "));
1110 boolean isrna = aa.isValidStruc();
1111 for (int j = 0; j < aa.annotations.length; j++)
1113 seq += outputCharacter(key, j, isrna, aa.annotations, null);
1116 out.append(newline);
1121 out.append(newline);
1123 return out.toString();
1128 * add an annotation character to the output row
1137 private char outputCharacter(String key, int k, boolean isrna,
1138 Annotation[] ann, SequenceI sequenceI)
1141 Annotation annot = ann[k];
1142 String ch = (annot == null)
1143 ? ((sequenceI == null) ? "-"
1144 : Character.toString(sequenceI.getCharAt(k)))
1145 : (annot.displayCharacter == null
1146 ? String.valueOf(annot.secondaryStructure)
1147 : annot.displayCharacter);
1152 if (key != null && key.equals("SS"))
1154 char ssannotchar = ' ';
1155 boolean charset = false;
1158 // sensible gap character
1164 // valid secondary structure AND no alternative label (e.g. ' B')
1165 if (annot.secondaryStructure > ' ' && ch.length() < 2)
1167 ssannotchar = annot.secondaryStructure;
1173 return (ssannotchar == ' ' && isrna) ? '.' : ssannotchar;
1177 if (ch.length() == 0)
1181 else if (ch.length() == 1)
1185 else if (ch.length() > 1)
1190 return (seq == ' ' && key != null && key.equals("SS") && isrna) ? '.'
1195 * make a friendly ID string.
1198 * @return truncated dataName to after last '/'
1200 private String safeName(String dataName)
1203 while ((b = dataName.indexOf("/")) > -1 && b < dataName.length())
1205 dataName = dataName.substring(b + 1).trim();
1208 int e = (dataName.length() - dataName.indexOf(".")) + 1;
1209 dataName = dataName.substring(1, e).trim();
1214 public String print()
1216 out = new StringBuffer();
1217 out.append("# STOCKHOLM 1.0");
1218 out.append(newline);
1219 print(getSeqsAsArray(), false);
1222 out.append(newline);
1223 return out.toString();
1226 private static Hashtable typeIds = null;
1230 if (typeIds == null)
1232 typeIds = new Hashtable();
1233 typeIds.put("SS", "Secondary Structure");
1234 typeIds.put("SA", "Surface Accessibility");
1235 typeIds.put("TM", "transmembrane");
1236 typeIds.put("PP", "Posterior Probability");
1237 typeIds.put("LI", "ligand binding");
1238 typeIds.put("AS", "active site");
1239 typeIds.put("IN", "intron");
1240 typeIds.put("IR", "interacting residue");
1241 typeIds.put("AC", "accession");
1242 typeIds.put("OS", "organism");
1243 typeIds.put("CL", "class");
1244 typeIds.put("DE", "description");
1245 typeIds.put("DR", "reference");
1246 typeIds.put("LO", "look");
1247 typeIds.put("RF", "Reference Positions");
1252 protected static String id2type(String id)
1254 if (typeIds.containsKey(id))
1256 return (String) typeIds.get(id);
1259 "Warning : Unknown Stockholm annotation type code " + id);
1263 protected static String type2id(String type)
1266 Enumeration e = typeIds.keys();
1267 while (e.hasMoreElements())
1269 Object ll = e.nextElement();
1270 if (typeIds.get(ll).toString().equalsIgnoreCase(type))
1281 "Warning : Unknown Stockholm annotation type: " + type);