2 * Jalview - A Sequence Alignment Editor and Viewer (Version 2.8)
3 * Copyright (C) 2012 J Procter, AM Waterhouse, LM Lui, J Engelhardt, G Barton, M Clamp, S Searle
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 of the License, or (at your option) any later version.
11 * Jalview is distributed in the hope that it will be useful, but
12 * WITHOUT ANY WARRANTY; without even the implied warranty
13 * of MERCHANTABILITY or FITNESS FOR A PARTICULAR
14 * PURPOSE. See the GNU General Public License for more details.
16 * You should have received a copy of the GNU General Public License along with Jalview. If not, see <http://www.gnu.org/licenses/>.
20 * JalviewX / Vamsas Project
21 * JPred.seq.concise reader
28 import javax.xml.parsers.ParserConfigurationException;
30 import org.xml.sax.SAXException;
32 import fr.orsay.lri.varna.exceptions.ExceptionFileFormatOrSyntax;
33 import fr.orsay.lri.varna.exceptions.ExceptionLoadingFailed;
34 import fr.orsay.lri.varna.exceptions.ExceptionPermissionDenied;
35 import fr.orsay.lri.varna.exceptions.ExceptionUnmatchedClosingParentheses;
37 import jalview.datamodel.*;
40 * Parser for the JPred/JNet concise format. This is a series of CSV lines, each
41 * line is either a sequence (QUERY), a sequence profile (align;), or jnet
42 * prediction annotation (anything else). Automagic translation happens for
43 * annotation called 'JNETPRED' (translated to Secondary Structure Prediction),
44 * or 'JNETCONF' (translates to 'Prediction Confidence'). Numeric scores are
45 * differentiated from symbolic by being parseable into a float vector. They are
46 * put in Scores. Symscores gets the others. JNetAnnotationMaker translates the
47 * data parsed by this object into annotation on an alignment. It is
48 * automatically called but can be used to transfer the annotation onto a
49 * sequence in another alignment (and insert gaps where necessary)
54 public class JPredFile extends AlignFile
60 Hashtable Scores; // Hash of names and score vectors
62 Hashtable Symscores; // indexes of symbol annotation properties in sequenceI
66 private int QuerySeqPosition;
69 * Creates a new JPredFile object.
77 public JPredFile(String inFile, String type) throws Exception
82 public JPredFile(FileParse source) throws Exception
90 * @param QuerySeqPosition
93 public void setQuerySeqPosition(int QuerySeqPosition)
95 this.QuerySeqPosition = QuerySeqPosition;
101 * @return DOCUMENT ME!
103 public int getQuerySeqPosition()
105 return QuerySeqPosition;
111 * @return DOCUMENT ME!
113 public Hashtable getScores()
121 * @return DOCUMENT ME!
123 public Hashtable getSymscores()
131 public void initData()
134 Scores = new Hashtable();
140 * parse a JPred concise file into a sequence-alignment like object.
142 public void parse() throws IOException
144 // JBPNote log.System.out.println("all read in ");
146 QuerySeqPosition = -1;
149 Vector seq_entries = new Vector();
150 Vector ids = new Vector();
151 Hashtable Symscores = new Hashtable();
153 while ((line = nextLine()) != null)
155 // Concise format allows no comments or non comma-formatted data
156 StringTokenizer str = new StringTokenizer(line, ":");
159 if (!str.hasMoreTokens())
164 id = str.nextToken();
166 String seqsym = str.nextToken();
167 StringTokenizer symbols = new StringTokenizer(seqsym, ",");
169 // decide if we have more than just alphanumeric symbols
170 int numSymbols = symbols.countTokens();
177 if (seqsym.length() != (2 * numSymbols))
179 // Set of scalars for some property
180 if (Scores.containsKey(id))
184 while (Scores.containsKey(id + "_" + i))
192 Vector scores = new Vector();
194 // Typecheck from first entry
196 String ascore = "dead";
200 // store elements as floats...
201 while (symbols.hasMoreTokens())
203 ascore = symbols.nextToken();
205 Float score = new Float(ascore);
206 scores.addElement((Object) score);
209 Scores.put(id, scores);
210 } catch (Exception e)
212 // or just keep them as strings
215 for (int j = 0; j < i; j++)
218 (Object) ((Float) scores.elementAt(j)).toString(), j);
221 scores.addElement((Object) ascore);
223 while (symbols.hasMoreTokens())
225 ascore = symbols.nextToken();
226 scores.addElement((Object) ascore);
229 Scores.put(id, scores);
232 else if (id.equals("jnetconf"))
234 // log.debug System.out.println("here");
235 id = "Prediction Confidence";
236 this.conf = new Vector(numSymbols);
238 for (int i = 0; i < numSymbols; i++)
240 conf.setElementAt(symbols.nextToken(), i);
245 // Sequence or a prediction string (rendered as sequence)
246 StringBuffer newseq = new StringBuffer();
248 for (int i = 0; i < numSymbols; i++)
250 newseq.append(symbols.nextToken());
253 if (id.indexOf(";") > -1)
255 seq_entries.addElement(newseq);
258 String name = id.substring(id.indexOf(";") + 1);
260 while (ids.lastIndexOf(name) > -1)
262 name = id.substring(id.indexOf(";") + 1) + "_" + ++i;
265 if (QuerySeqPosition == -1)
266 QuerySeqPosition = ids.size();
267 ids.addElement(name);
272 if (id.equals("JNETPRED"))
274 id = "Predicted Secondary Structure";
277 seq_entries.addElement(newseq.toString());
279 Symscores.put((Object) id, (Object) new Integer(ids.size() - 1));
284 * leave it to the parser user to actually check this. if (noSeqs < 1) {
285 * throw new IOException( "JpredFile Parser: No sequence in the
289 maxLength = seq_entries.elementAt(0).toString().length();
291 for (int i = 0; i < ids.size(); i++)
293 // Add all sequence like objects
294 Sequence newSeq = new Sequence(ids.elementAt(i).toString(),
295 seq_entries.elementAt(i).toString(), 1, seq_entries
296 .elementAt(i).toString().length());
298 if (maxLength != seq_entries.elementAt(i).toString().length())
300 throw new IOException("JPredConcise: Entry ("
301 + ids.elementAt(i).toString()
302 + ") has an unexpected number of columns");
305 if ((newSeq.getName().startsWith("QUERY") || newSeq.getName()
306 .startsWith("align;")) && (QuerySeqPosition == -1))
308 QuerySeqPosition = seqs.size();
311 seqs.addElement(newSeq);
313 if (seqs.size() > 0 && QuerySeqPosition > -1)
315 // try to make annotation for a prediction only input (default if no
316 // alignment is given and prediction contains a QUERY or align;sequence_id
318 Alignment tal = new Alignment(this.getSeqsAsArray());
321 JnetAnnotationMaker.add_annotation(this, tal, QuerySeqPosition,
323 } catch (Exception e)
326 IOException ex = new IOException(
327 "Couldn't parse concise annotation for prediction profile.\n"
329 e.printStackTrace(); // java 1.1 does not have :
330 // ex.setStackTrace(e.getStackTrace());
333 this.annotations = new Vector();
334 AlignmentAnnotation[] aan = tal.getAlignmentAnnotation();
335 for (int aai = 0; aan != null && aai < aan.length; aai++)
337 annotations.addElement(aan[aai]);
347 public String print()
349 return "Not Supported";
359 public static void main(String[] args) throws Exception
363 JPredFile blc = new JPredFile(args[0], "File");
365 for (int i = 0; i < blc.seqs.size(); i++)
367 System.out.println(((Sequence) blc.seqs.elementAt(i)).getName()
369 + ((Sequence) blc.seqs.elementAt(i)).getSequenceAsString()
372 } catch (java.io.IOException e)
374 System.err.println("Exception " + e);
375 // e.printStackTrace(); not java 1.1 compatible!
379 Vector annotSeqs = null;
384 public void removeNonSequences()
386 if (annotSeqs != null)
390 annotSeqs = new Vector();
391 Vector newseqs = new Vector();
394 for (; i < QuerySeqPosition; i++)
396 annotSeqs.addElement(seqs.elementAt(i));
398 // check that no stray annotations have been added at the end.
400 SequenceI sq = (SequenceI) seqs.elementAt(j - 1);
401 if (sq.getName().toUpperCase().startsWith("JPRED"))
403 annotSeqs.addElement(sq);
404 seqs.removeElementAt(--j);
409 newseqs.addElement(seqs.elementAt(i));
412 seqs.removeAllElements();
418 * StringBuffer out = new StringBuffer();
420 * out.append("START PRED\n"); for (int i = 0; i < s[0].sequence.length(); i++)
421 * { out.append(s[0].sequence.substring(i, i + 1) + " ");
422 * out.append(s[1].sequence.substring(i, i + 1) + " ");
423 * out.append(s[1].score[0].elementAt(i) + " ");
424 * out.append(s[1].score[1].elementAt(i) + " ");
425 * out.append(s[1].score[2].elementAt(i) + " ");
426 * out.append(s[1].score[3].elementAt(i) + " ");
428 * out.append("\n"); } out.append("END PRED\n"); return out.toString(); }
430 * public static void main(String[] args) { try { BLCFile blc = new
431 * BLCFile(args[0], "File"); DrawableSequence[] s = new
432 * DrawableSequence[blc.seqs.size()]; for (int i = 0; i < blc.seqs.size(); i++)
433 * { s[i] = new DrawableSequence( (Sequence) blc.seqs.elementAt(i)); } String
434 * out = BLCFile.print(s);
436 * AlignFrame af = new AlignFrame(null, s); af.resize(700, 500); af.show();
437 * System.out.println(out); } catch (java.io.IOException e) {
438 * System.out.println("Exception " + e); } } }