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 import jalview.analysis.SequenceIdMatcher;
21 import jalview.datamodel.AlignmentAnnotation;
22 import jalview.datamodel.AlignmentI;
23 import jalview.datamodel.Annotation;
24 import jalview.datamodel.SequenceI;
26 import java.awt.Color;
27 import java.io.IOException;
28 import java.util.ArrayList;
29 import java.util.HashMap;
30 import java.util.LinkedHashMap;
31 import java.util.List;
34 import javax.xml.parsers.ParserConfigurationException;
36 import org.xml.sax.SAXException;
38 import fr.orsay.lri.varna.exceptions.ExceptionFileFormatOrSyntax;
39 import fr.orsay.lri.varna.exceptions.ExceptionLoadingFailed;
40 import fr.orsay.lri.varna.exceptions.ExceptionPermissionDenied;
41 import fr.orsay.lri.varna.exceptions.ExceptionUnmatchedClosingParentheses;
44 * A file parse for T-Coffee score ascii format. This file contains the
45 * alignment consensus for each resude in any sequence.
47 * This file is procuded by <code>t_coffee</code> providing the option
48 * <code>-output=score_ascii </code> to the program command line
50 * An example file is the following
53 * T-COFFEE, Version_9.02.r1228 (2012-02-16 18:15:12 - Revision 1228 - Build 336)
70 * 1PHT 999999999999999999999999998762112222543211112134
71 * 1BB9 99999999999999999999999999987-------4322----2234
72 * 1UHC 99999999999999999999999999987-------5321----2246
73 * 1YCS 99999999999999999999999999986-------4321----1-35
74 * 1OOT 999999999999999999999999999861-------3------1135
75 * 1ABO 99999999999999999999999999986-------422-------34
76 * 1FYN 99999999999999999999999999985-------32--------35
77 * 1QCF 99999999999999999999999999974-------2---------24
78 * cons 999999999999999999999999999851000110321100001134
81 * 1PHT ----------5666642367889999999999889
82 * 1BB9 1111111111676653-355679999999999889
83 * 1UHC ----------788774--66789999999999889
84 * 1YCS ----------78777--356789999999999889
85 * 1OOT ----------78877--356789999999997-67
86 * 1ABO ----------687774--56779999999999889
87 * 1FYN ----------6888842356789999999999889
88 * 1QCF ----------6878742356789999999999889
89 * cons 00100000006877641356789999999999889
93 * @author Paolo Di Tommaso
96 public class TCoffeeScoreFile extends AlignFile {
98 public TCoffeeScoreFile(String inFile, String type) throws IOException, ExceptionFileFormatOrSyntax, ParserConfigurationException, SAXException, ExceptionPermissionDenied, ExceptionLoadingFailed, InterruptedException, ExceptionUnmatchedClosingParentheses
104 public TCoffeeScoreFile(FileParse source) throws IOException, ExceptionFileFormatOrSyntax, ParserConfigurationException, SAXException, ExceptionPermissionDenied, ExceptionLoadingFailed, InterruptedException, ExceptionUnmatchedClosingParentheses
109 /** The {@link Header} structure holder */
113 * Holds the consensues values for each sequences. It uses a LinkedHashMap to
114 * maintaint the insertion order.
116 LinkedHashMap<String, StringBuilder> scores;
121 * Parse the provided reader for the T-Coffee scores file format
124 * public static TCoffeeScoreFile load(Reader reader) {
126 * try { BufferedReader in = (BufferedReader) (reader instanceof
127 * BufferedReader ? reader : new BufferedReader(reader));
128 * TCoffeeScoreFile result = new TCoffeeScoreFile();
129 * result.doParsing(in); return result.header != null &&
130 * result.scores != null ? result : null; } catch( Exception e) {
131 * throw new RuntimeException(e); } }
135 * @return The 'height' of the score matrix i.e. the numbers of score rows
136 * that should matches the number of sequences in the alignment
138 public int getHeight()
140 // the last entry will always be the 'global' alingment consensus scores, so
142 // from the 'height' count to make this value compatible with the number of
143 // sequences in the MSA
144 return scores != null && scores.size() > 0 ? scores.size() - 1 : 0;
148 * @return The 'width' of the score matrix i.e. the number of columns. Since
149 * teh score value are supposd to be calculated for an 'aligned' MSA,
150 * all the entries have to have the same width.
152 public int getWidth()
154 return fWidth != null ? fWidth : 0;
158 * Get the string of score values for the specified seqeunce ID.
162 * @return The scores as a string of values e.g. {@code 99999987-------432}.
163 * It return an empty string when the specified ID is missing.
165 public String getScoresFor(String id)
167 return scores != null && scores.containsKey(id) ? scores.get(id)
172 * @return The list of score string as a {@link List} object, in the same
173 * ordeer of the insertion i.e. in the MSA
175 public List<String> getScoresList()
181 List<String> result = new ArrayList<String>(scores.size());
182 for (Map.Entry<String, StringBuilder> it : scores.entrySet())
184 result.add(it.getValue().toString());
191 * @return The parsed score values a matrix of bytes
193 public byte[][] getScoresArray()
199 byte[][] result = new byte[scores.size()][];
202 for (Map.Entry<String, StringBuilder> it : scores.entrySet())
204 String line = it.getValue().toString();
205 byte[] seqValues = new byte[line.length()];
206 for (int j = 0, c = line.length(); j < c; j++)
209 byte val = (byte) (line.charAt(j) - '0');
211 seqValues[j] = (val >= 0 && val <= 9) ? val : -1;
214 result[rowCount++] = seqValues;
220 public void parse() throws IOException
225 header = readHeader(this);
232 scores = new LinkedHashMap<String, StringBuilder>();
235 * initilize the structure
237 for (Map.Entry<String, Integer> entry : header.scores.entrySet())
239 scores.put(entry.getKey(), new StringBuilder());
243 * go with the reading
246 while ((block = readBlock(this, header.scores.size())) != null)
250 * append sequences read in the block
252 for (Map.Entry<String, String> entry : block.items.entrySet())
254 StringBuilder scoreStringBuilder = scores.get(entry.getKey());
255 if (scoreStringBuilder == null)
258 errormessage = String
259 .format("Invalid T-Coffee score file: Sequence ID '%s' is not declared in header section",
264 scoreStringBuilder.append(entry.getValue());
269 * verify that all rows have the same width
271 for (StringBuilder str : scores.values())
275 fWidth = str.length();
277 else if (fWidth != str.length())
280 errormessage = "Invalid T-Coffee score file: All the score sequences must have the same length";
288 static int parseInt(String str)
292 return Integer.parseInt(str);
293 } catch (NumberFormatException e)
295 // TODO report a warning ?
301 * Reaad the header section in the T-Coffee score file format
305 * @return The parser {@link Header} instance
306 * @throws RuntimeException
307 * when the header is not in the expected format
309 static Header readHeader(FileParse reader) throws IOException
312 Header result = null;
315 result = new Header();
316 result.head = reader.nextLine();
320 while ((line = reader.nextLine()) != null)
322 if (line.startsWith("SCORE="))
324 result.score = parseInt(line.substring(6).trim());
329 if ((line = reader.nextLine()) == null || !"*".equals(line.trim()))
332 "Invalid T-COFFEE score format (NO BAD/AVG/GOOD header)");
335 if ((line = reader.nextLine()) == null
336 || !"BAD AVG GOOD".equals(line.trim()))
339 "Invalid T-COFFEE score format (NO BAD/AVG/GOOD header)");
342 if ((line = reader.nextLine()) == null || !"*".equals(line.trim()))
345 "Invalid T-COFFEE score format (NO BAD/AVG/GOOD header)");
350 * now are expected a list if sequences ID up to the first blank line
352 while ((line = reader.nextLine()) != null)
359 int p = line.indexOf(":");
362 // TODO report a warning
366 String id = line.substring(0, p).trim();
367 int val = parseInt(line.substring(p + 1).trim());
370 // TODO report warning
374 result.scores.put(id, val);
379 error(reader, "T-COFFEE score file had no per-sequence scores");
382 } catch (IOException e)
384 error(reader, "Unexpected problem parsing T-Coffee score ascii file");
391 private static void error(FileParse reader, String errm)
394 if (reader.errormessage == null)
396 reader.errormessage = errm;
400 reader.errormessage += "\n" + errm;
405 * Read a scores block ihe provided stream.
408 * The stream to parse
410 * The expected number of the sequence to be read
411 * @return The {@link Block} instance read or {link null} null if the end of
413 * @throws IOException
414 * Something went wrong on the 'wire'
416 static Block readBlock(FileParse reader, int size) throws IOException
418 Block result = new Block(size);
422 * read blank lines (eventually)
424 while ((line = reader.nextLine()) != null && "".equals(line.trim()))
426 // consume blank lines
435 * read the scores block
439 if ("".equals(line.trim()))
445 // split the line on the first blank
446 // the first part have to contain the sequence id
447 // the remaining part are the scores values
448 int p = line.indexOf(" ");
451 if (reader.warningMessage == null)
453 reader.warningMessage = "";
455 reader.warningMessage += "Possible parsing error - expected to find a space in line: '"
460 String id = line.substring(0, p).trim();
461 String val = line.substring(p + 1).trim();
463 result.items.put(id, val);
465 } while ((line = reader.nextLine()) != null);
471 * The score file header
479 LinkedHashMap<String, Integer> scores = new LinkedHashMap<String, Integer>();
481 public int getScoreAvg()
486 public int getScoreFor(String ID)
489 return scores.containsKey(ID) ? scores.get(ID) : -1;
495 * Hold a single block values block in the score file
501 Map<String, String> items;
503 public Block(int size)
506 this.items = new HashMap<String, String>(size);
509 String getScoresFor(String id)
511 return items.get(id);
514 String getConsensus()
516 return items.get("cons");
521 * TCOFFEE score colourscheme
523 static final Color[] colors =
524 { new Color(102, 102, 255), // #6666FF
525 new Color(0, 255, 0), // #00FF00
526 new Color(102, 255, 0), // #66FF00
527 new Color(204, 255, 0), // #CCFF00
528 new Color(255, 255, 0), // #FFFF00
529 new Color(255, 204, 0), // #FFCC00
530 new Color(255, 153, 0), // #FF9900
531 new Color(255, 102, 0), // #FF6600
532 new Color(255, 51, 0), // #FF3300
533 new Color(255, 34, 0) // #FF2000
536 public final static String TCOFFEE_SCORE = "TCoffeeScore";
539 * generate annotation for this TCoffee score set on the given alignment
542 * alignment to annotate
544 * if true, annotate sequences based on matching sequence names
545 * @return true if alignment annotation was modified, false otherwise.
547 public boolean annotateAlignment(AlignmentI al, boolean matchids)
549 if (al.getHeight() != getHeight() || al.getWidth() != getWidth())
551 warningMessage = "Alignment shape does not match T-Coffee score file shape.";
554 boolean added = false;
556 SequenceIdMatcher sidmatcher = new SequenceIdMatcher(
557 al.getSequencesArray());
558 byte[][] scoreMatrix = getScoresArray();
559 // for 2.8 - we locate any existing TCoffee annotation and remove it first
560 // before adding this.
561 for (Map.Entry<String, StringBuilder> id : scores.entrySet())
563 byte[] srow = scoreMatrix[i];
567 s = sidmatcher.findIdMatch(id.getKey());
571 s = al.getSequenceAt(i);
574 if (s == null && i != scores.size() && !id.getKey().equals("cons"))
576 System.err.println("No "
577 + (matchids ? "match " : " sequences left ")
578 + " for TCoffee score set : " + id.getKey());
581 int jSize = al.getWidth() < srow.length ? al.getWidth() : srow.length;
582 Annotation[] annotations = new Annotation[al.getWidth()];
583 for (int j = 0; j < jSize; j++)
586 if (s != null && jalview.util.Comparison.isGap(s.getCharAt(j)))
588 annotations[j] = null;
592 .println("Warning: non-zero value for positional T-COFFEE score for gap at "
593 + j + " in sequence " + s.getName());
598 annotations[j] = new Annotation(s == null ? "" + val : null,
599 s == null ? "" + val : null, '\0', val * 1f, val >= 0
600 && val < colors.length ? colors[val]
604 // this will overwrite any existing t-coffee scores for the alignment
605 AlignmentAnnotation aa = al.findOrCreateAnnotation(TCOFFEE_SCORE,
606 TCOFFEE_SCORE, false, s, null);
609 aa.label = "T-COFFEE";
610 aa.description = "" + id.getKey();
611 aa.annotations = annotations;
613 aa.belowAlignment = false;
614 aa.setScore(header.getScoreFor(id.getKey()));
615 aa.createSequenceMapping(s, s.getStart(), true);
616 s.addAlignmentAnnotation(aa);
617 aa.adjustForAlignment();
621 aa.graph = AlignmentAnnotation.NO_GRAPH;
622 aa.label = "T-COFFEE";
623 aa.description = "TCoffee column reliability score";
624 aa.annotations = annotations;
625 aa.belowAlignment = true;
627 aa.setScore(header.getScoreAvg());
629 aa.showAllColLabels = true;
630 aa.validateRangeAndDisplay();
638 public String print()
640 // TODO Auto-generated method stub