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.
23 * JalviewX / Vamsas Project
24 * JPred.seq.concise reader
28 import jalview.datamodel.Alignment;
29 import jalview.datamodel.AlignmentAnnotation;
30 import jalview.datamodel.Sequence;
31 import jalview.datamodel.SequenceI;
32 import jalview.util.MessageManager;
34 import java.io.IOException;
35 import java.util.Hashtable;
36 import java.util.Locale;
37 import java.util.StringTokenizer;
38 import java.util.Vector;
41 * Parser for the JPred/JNet concise format. This is a series of CSV lines, each
42 * line is either a sequence (QUERY), a sequence profile (align;), or jnet
43 * prediction annotation (anything else). Automagic translation happens for
44 * annotation called 'JNETPRED' (translated to Secondary Structure Prediction),
45 * or 'JNETCONF' (translates to 'Prediction Confidence'). Numeric scores are
46 * differentiated from symbolic by being parseable into a float vector. They are
47 * put in Scores. Symscores gets the others. JNetAnnotationMaker translates the
48 * data parsed by this object into annotation on an alignment. It is
49 * automatically called but can be used to transfer the annotation onto a
50 * sequence in another alignment (and insert gaps where necessary)
55 public class JPredFile extends AlignFile
61 Hashtable Scores; // Hash of names and score vectors
63 Hashtable Symscores; // indexes of symbol annotation properties in sequenceI
67 private int QuerySeqPosition;
70 * Creates a new JPredFile object.
80 public JPredFile(String inFile, DataSourceType sourceType)
83 super(inFile, sourceType);
86 public JPredFile(FileParse source) throws IOException
94 * @param QuerySeqPosition
97 public void setQuerySeqPosition(int QuerySeqPosition)
99 this.QuerySeqPosition = QuerySeqPosition;
105 * @return DOCUMENT ME!
107 public int getQuerySeqPosition()
109 return QuerySeqPosition;
115 * @return DOCUMENT ME!
117 public Hashtable getScores()
125 * @return DOCUMENT ME!
127 public Hashtable getSymscores()
136 public void initData()
139 Scores = new Hashtable();
145 * parse a JPred concise file into a sequence-alignment like object.
148 public void parse() throws IOException
150 // JBPNote log.System.out.println("all read in ");
152 QuerySeqPosition = -1;
155 Vector seq_entries = new Vector();
156 Vector ids = new Vector();
157 Hashtable Symscores = new Hashtable();
159 while ((line = nextLine()) != null)
161 // Concise format allows no comments or non comma-formatted data
162 StringTokenizer str = new StringTokenizer(line, ":");
165 if (!str.hasMoreTokens())
170 id = str.nextToken();
172 String seqsym = str.nextToken();
173 StringTokenizer symbols = new StringTokenizer(seqsym, ",");
175 // decide if we have more than just alphanumeric symbols
176 int numSymbols = symbols.countTokens();
183 if (seqsym.length() != (2 * numSymbols))
185 // Set of scalars for some property
186 if (Scores.containsKey(id))
190 while (Scores.containsKey(id + "_" + i))
198 Vector scores = new Vector();
200 // Typecheck from first entry
202 String ascore = "dead";
206 // store elements as floats...
207 while (symbols.hasMoreTokens())
209 ascore = symbols.nextToken();
211 Float score = Float.valueOf(ascore);
212 scores.addElement(score);
215 Scores.put(id, scores);
216 } catch (Exception e)
218 // or just keep them as strings
221 for (int j = 0; j < i; j++)
223 scores.setElementAt(((Float) scores.elementAt(j)).toString(),
227 scores.addElement(ascore);
229 while (symbols.hasMoreTokens())
231 ascore = symbols.nextToken();
232 scores.addElement(ascore);
235 Scores.put(id, scores);
238 else if (id.equals("jnetconf"))
240 // log.debug System.out.println("here");
241 id = "Prediction Confidence";
242 this.conf = new Vector(numSymbols);
244 for (int i = 0; i < numSymbols; i++)
246 conf.setElementAt(symbols.nextToken(), i);
251 // Sequence or a prediction string (rendered as sequence)
252 StringBuffer newseq = new StringBuffer();
254 for (int i = 0; i < numSymbols; i++)
256 newseq.append(symbols.nextToken());
259 if (id.indexOf(";") > -1)
261 seq_entries.addElement(newseq);
264 String name = id.substring(id.indexOf(";") + 1);
266 while (ids.lastIndexOf(name) > -1)
268 name = id.substring(id.indexOf(";") + 1) + "_" + ++i;
271 if (QuerySeqPosition == -1)
273 QuerySeqPosition = ids.size();
275 ids.addElement(name);
280 if (id.equals("JNETPRED"))
282 id = "Predicted Secondary Structure";
285 seq_entries.addElement(newseq.toString());
287 Symscores.put(id, Integer.valueOf(ids.size() - 1));
292 * leave it to the parser user to actually check this. if (noSeqs < 1) {
293 * throw new IOException( "JpredFile Parser: No sequence in the
297 maxLength = seq_entries.elementAt(0).toString().length();
299 for (int i = 0; i < ids.size(); i++)
301 // Add all sequence like objects
302 Sequence newSeq = new Sequence(ids.elementAt(i).toString(),
303 seq_entries.elementAt(i).toString(), 1,
304 seq_entries.elementAt(i).toString().length());
306 if (maxLength != seq_entries.elementAt(i).toString().length())
308 throw new IOException(MessageManager.formatMessage(
309 "exception.jpredconcide_entry_has_unexpected_number_of_columns",
311 { ids.elementAt(i).toString() }));
314 if ((newSeq.getName().startsWith("QUERY")
315 || newSeq.getName().startsWith("align;"))
316 && (QuerySeqPosition == -1))
318 QuerySeqPosition = seqs.size();
321 seqs.addElement(newSeq);
323 if (seqs.size() > 0 && QuerySeqPosition > -1)
325 // try to make annotation for a prediction only input (default if no
326 // alignment is given and prediction contains a QUERY or align;sequence_id
328 Alignment tal = new Alignment(this.getSeqsAsArray());
331 JnetAnnotationMaker.add_annotation(this, tal, QuerySeqPosition,
333 } catch (Exception e)
336 IOException ex = new IOException(MessageManager.formatMessage(
337 "exception.couldnt_parse_concise_annotation_for_prediction",
339 { e.getMessage() }));
340 e.printStackTrace(); // java 1.1 does not have :
341 // ex.setStackTrace(e.getStackTrace());
344 this.annotations = new Vector();
345 AlignmentAnnotation[] aan = tal.getAlignmentAnnotation();
346 for (int aai = 0; aan != null && aai < aan.length; aai++)
348 annotations.addElement(aan[aai]);
359 public String print(SequenceI[] sqs, boolean jvsuffix)
361 return "Not Supported";
370 public static void main(String[] args)
374 JPredFile jpred = new JPredFile(args[0], DataSourceType.FILE);
376 for (int i = 0; i < jpred.seqs.size(); i++)
378 System.out.println(((Sequence) jpred.seqs.elementAt(i)).getName()
380 + ((Sequence) jpred.seqs.elementAt(i)).getSequenceAsString()
383 } catch (java.io.IOException e)
385 System.err.println("Exception " + e);
386 // e.printStackTrace(); not java 1.1 compatible!
390 Vector annotSeqs = null;
395 public void removeNonSequences()
397 if (annotSeqs != null)
401 annotSeqs = new Vector();
402 Vector newseqs = new Vector();
405 for (; i < QuerySeqPosition; i++)
407 annotSeqs.addElement(seqs.elementAt(i));
409 // check that no stray annotations have been added at the end.
411 SequenceI sq = seqs.elementAt(j - 1);
412 if (sq.getName().toUpperCase(Locale.ROOT).startsWith("JPRED"))
414 annotSeqs.addElement(sq);
415 seqs.removeElementAt(--j);
420 newseqs.addElement(seqs.elementAt(i));
423 seqs.removeAllElements();
429 * StringBuffer out = new StringBuffer();
431 * out.append("START PRED\n"); for (int i = 0; i < s[0].sequence.length(); i++)
432 * { out.append(s[0].sequence.substring(i, i + 1) + " ");
433 * out.append(s[1].sequence.substring(i, i + 1) + " ");
434 * out.append(s[1].score[0].elementAt(i) + " ");
435 * out.append(s[1].score[1].elementAt(i) + " ");
436 * out.append(s[1].score[2].elementAt(i) + " ");
437 * out.append(s[1].score[3].elementAt(i) + " ");
439 * out.append("\n"); } out.append("END PRED\n"); return out.toString(); }
441 * public static void main(String[] args) { try { BLCFile blc = new
442 * BLCFile(args[0], "File"); DrawableSequence[] s = new
443 * DrawableSequence[blc.seqs.size()]; for (int i = 0; i < blc.seqs.size(); i++)
444 * { s[i] = new DrawableSequence( (Sequence) blc.seqs.elementAt(i)); } String
445 * out = BLCFile.print(s);
447 * AlignFrame af = new AlignFrame(null, s); af.resize(700, 500); af.show();
448 * System.out.println(out); } catch (java.io.IOException e) {
449 * System.out.println("Exception " + e); } } }