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;
33 import java.util.regex.Matcher;
34 import java.util.regex.Pattern;
37 * A file parse for T-Coffee score ascii format. This file contains the
38 * alignment consensus for each resude in any sequence.
40 * This file is procuded by <code>t_coffee</code> providing the option
41 * <code>-output=score_ascii </code> to the program command line
43 * An example file is the following
46 * T-COFFEE, Version_9.02.r1228 (2012-02-16 18:15:12 - Revision 1228 - Build 336)
63 * 1PHT 999999999999999999999999998762112222543211112134
64 * 1BB9 99999999999999999999999999987-------4322----2234
65 * 1UHC 99999999999999999999999999987-------5321----2246
66 * 1YCS 99999999999999999999999999986-------4321----1-35
67 * 1OOT 999999999999999999999999999861-------3------1135
68 * 1ABO 99999999999999999999999999986-------422-------34
69 * 1FYN 99999999999999999999999999985-------32--------35
70 * 1QCF 99999999999999999999999999974-------2---------24
71 * cons 999999999999999999999999999851000110321100001134
74 * 1PHT ----------5666642367889999999999889
75 * 1BB9 1111111111676653-355679999999999889
76 * 1UHC ----------788774--66789999999999889
77 * 1YCS ----------78777--356789999999999889
78 * 1OOT ----------78877--356789999999997-67
79 * 1ABO ----------687774--56779999999999889
80 * 1FYN ----------6888842356789999999999889
81 * 1QCF ----------6878742356789999999999889
82 * cons 00100000006877641356789999999999889
86 * @author Paolo Di Tommaso
89 public class TCoffeeScoreFile extends AlignFile
92 public TCoffeeScoreFile(String inFile, String type) throws IOException
98 public TCoffeeScoreFile(FileParse source) throws IOException
103 /** The {@link Header} structure holder */
107 * Holds the consensues values for each sequences. It uses a LinkedHashMap to
108 * maintaint the insertion order.
110 LinkedHashMap<String, StringBuilder> scores;
115 * Parse the provided reader for the T-Coffee scores file format
118 * public static TCoffeeScoreFile load(Reader reader) {
120 * try { BufferedReader in = (BufferedReader) (reader instanceof
121 * BufferedReader ? reader : new BufferedReader(reader));
122 * TCoffeeScoreFile result = new TCoffeeScoreFile();
123 * result.doParsing(in); return result.header != null &&
124 * result.scores != null ? result : null; } catch( Exception e) {
125 * throw new RuntimeException(e); } }
129 * @return The 'height' of the score matrix i.e. the numbers of score rows
130 * that should matches the number of sequences in the alignment
132 public int getHeight()
134 // the last entry will always be the 'global' alingment consensus scores, so
136 // from the 'height' count to make this value compatible with the number of
137 // sequences in the MSA
138 return scores != null && scores.size() > 0 ? scores.size() - 1 : 0;
142 * @return The 'width' of the score matrix i.e. the number of columns. Since
143 * the score value are supposed to be calculated for an 'aligned' MSA,
144 * all the entries have to have the same width.
146 public int getWidth()
148 return fWidth != null ? fWidth : 0;
152 * Get the string of score values for the specified seqeunce ID.
156 * @return The scores as a string of values e.g. {@code 99999987-------432}.
157 * It return an empty string when the specified ID is missing.
159 public String getScoresFor(String id)
161 return scores != null && scores.containsKey(id) ? scores.get(id)
166 * @return The list of score string as a {@link List} object, in the same
167 * ordeer of the insertion i.e. in the MSA
169 public List<String> getScoresList()
175 List<String> result = new ArrayList<String>(scores.size());
176 for (Map.Entry<String, StringBuilder> it : scores.entrySet())
178 result.add(it.getValue().toString());
185 * @return The parsed score values a matrix of bytes
187 public byte[][] getScoresArray()
193 byte[][] result = new byte[scores.size()][];
196 for (Map.Entry<String, StringBuilder> it : scores.entrySet())
198 String line = it.getValue().toString();
199 byte[] seqValues = new byte[line.length()];
200 for (int j = 0, c = line.length(); j < c; j++)
203 byte val = (byte) (line.charAt(j) - '0');
205 seqValues[j] = (val >= 0 && val <= 9) ? val : -1;
208 result[rowCount++] = seqValues;
214 public void parse() throws IOException
219 header = readHeader(this);
226 scores = new LinkedHashMap<String, StringBuilder>();
229 * initilize the structure
231 for (Map.Entry<String, Integer> entry : header.scores.entrySet())
233 scores.put(entry.getKey(), new StringBuilder());
237 * go with the reading
240 while ((block = readBlock(this, header.scores.size())) != null)
244 * append sequences read in the block
246 for (Map.Entry<String, String> entry : block.items.entrySet())
248 StringBuilder scoreStringBuilder = scores.get(entry.getKey());
249 if (scoreStringBuilder == null)
252 errormessage = String
253 .format("Invalid T-Coffee score file: Sequence ID '%s' is not declared in header section",
258 scoreStringBuilder.append(entry.getValue());
263 * verify that all rows have the same width
265 for (StringBuilder str : scores.values())
269 fWidth = str.length();
271 else if (fWidth != str.length())
274 errormessage = "Invalid T-Coffee score file: All the score sequences must have the same length";
282 static int parseInt(String str)
286 return Integer.parseInt(str);
287 } catch (NumberFormatException e)
289 // TODO report a warning ?
295 * Reaad the header section in the T-Coffee score file format
299 * @return The parser {@link Header} instance
300 * @throws RuntimeException
301 * when the header is not in the expected format
303 static Header readHeader(FileParse reader) throws IOException
306 Header result = null;
309 result = new Header();
310 result.head = reader.nextLine();
314 while ((line = reader.nextLine()) != null)
316 if (line.startsWith("SCORE="))
318 result.score = parseInt(line.substring(6).trim());
323 if ((line = reader.nextLine()) == null || !"*".equals(line.trim()))
326 "Invalid T-COFFEE score format (NO BAD/AVG/GOOD header)");
329 if ((line = reader.nextLine()) == null
330 || !"BAD AVG GOOD".equals(line.trim()))
333 "Invalid T-COFFEE score format (NO BAD/AVG/GOOD header)");
336 if ((line = reader.nextLine()) == null || !"*".equals(line.trim()))
339 "Invalid T-COFFEE score format (NO BAD/AVG/GOOD header)");
344 * now are expected a list if sequences ID up to the first blank line
346 while ((line = reader.nextLine()) != null)
353 int p = line.indexOf(":");
356 // TODO report a warning
360 String id = line.substring(0, p).trim();
361 int val = parseInt(line.substring(p + 1).trim());
364 // TODO report warning
368 result.scores.put(id, val);
373 error(reader, "T-COFFEE score file had no per-sequence scores");
376 } catch (IOException e)
378 error(reader, "Unexpected problem parsing T-Coffee score ascii file");
385 private static void error(FileParse reader, String errm)
388 if (reader.errormessage == null)
390 reader.errormessage = errm;
394 reader.errormessage += "\n" + errm;
398 static Pattern SCORES_WITH_RESIDUE_NUMS = Pattern.compile("^\\d+\\s([^\\s]+)\\s+\\d+$");
401 * Read a scores block ihe provided stream.
404 * The stream to parse
406 * The expected number of the sequence to be read
407 * @return The {@link Block} instance read or {link null} null if the end of
409 * @throws IOException
410 * Something went wrong on the 'wire'
412 static Block readBlock(FileParse reader, int size) throws IOException
414 Block result = new Block(size);
418 * read blank lines (eventually)
420 while ((line = reader.nextLine()) != null && "".equals(line.trim()))
422 // consume blank lines
431 * read the scores block
435 if ("".equals(line.trim()))
441 // split the line on the first blank
442 // the first part have to contain the sequence id
443 // the remaining part are the scores values
444 int p = line.indexOf(" ");
447 if (reader.warningMessage == null)
449 reader.warningMessage = "";
451 reader.warningMessage += "Possible parsing error - expected to find a space in line: '"
456 String id = line.substring(0, p).trim();
457 String val = line.substring(p + 1).trim();
459 Matcher m = SCORES_WITH_RESIDUE_NUMS.matcher(val);
464 result.items.put(id, val);
466 } while ((line = reader.nextLine()) != null);
472 * The score file header
480 LinkedHashMap<String, Integer> scores = new LinkedHashMap<String, Integer>();
482 public int getScoreAvg()
487 public int getScoreFor(String ID)
490 return scores.containsKey(ID) ? scores.get(ID) : -1;
496 * Hold a single block values block in the score file
502 Map<String, String> items;
504 public Block(int size)
507 this.items = new HashMap<String, String>(size);
510 String getScoresFor(String id)
512 return items.get(id);
515 String getConsensus()
517 return items.get("cons");
522 * TCOFFEE score colourscheme
524 static final Color[] colors =
525 { new Color(102, 102, 255), // #6666FF
526 new Color(0, 255, 0), // #00FF00
527 new Color(102, 255, 0), // #66FF00
528 new Color(204, 255, 0), // #CCFF00
529 new Color(255, 255, 0), // #FFFF00
530 new Color(255, 204, 0), // #FFCC00
531 new Color(255, 153, 0), // #FF9900
532 new Color(255, 102, 0), // #FF6600
533 new Color(255, 51, 0), // #FF3300
534 new Color(255, 34, 0) // #FF2000
537 public final static String TCOFFEE_SCORE = "TCoffeeScore";
540 * generate annotation for this TCoffee score set on the given alignment
543 * alignment to annotate
545 * if true, annotate sequences based on matching sequence names
546 * @return true if alignment annotation was modified, false otherwise.
548 public boolean annotateAlignment(AlignmentI al, boolean matchids)
550 if (al.getHeight() != getHeight() || al.getWidth() != getWidth())
552 String info = String.format("align w: %s, h: %s; score: w: %s; h: %s ", al.getWidth(), al.getHeight(), getWidth(), getHeight() );
553 warningMessage = "Alignment shape does not match T-Coffee score file shape -- " + info;
556 boolean added = false;
558 SequenceIdMatcher sidmatcher = new SequenceIdMatcher(
559 al.getSequencesArray());
560 byte[][] scoreMatrix = getScoresArray();
561 // for 2.8 - we locate any existing TCoffee annotation and remove it first
562 // before adding this.
563 for (Map.Entry<String, StringBuilder> id : scores.entrySet())
565 byte[] srow = scoreMatrix[i];
569 s = sidmatcher.findIdMatch(id.getKey());
573 s = al.getSequenceAt(i);
576 if (s == null && i != scores.size() && !id.getKey().equals("cons"))
578 System.err.println("No "
579 + (matchids ? "match " : " sequences left ")
580 + " for TCoffee score set : " + id.getKey());
583 int jSize = al.getWidth() < srow.length ? al.getWidth() : srow.length;
584 Annotation[] annotations = new Annotation[al.getWidth()];
585 for (int j = 0; j < jSize; j++)
588 if (s != null && jalview.util.Comparison.isGap(s.getCharAt(j)))
590 annotations[j] = null;
594 .println("Warning: non-zero value for positional T-COFFEE score for gap at "
595 + j + " in sequence " + s.getName());
600 annotations[j] = new Annotation(s == null ? "" + val : null,
601 s == null ? "" + val : null, '\0', val * 1f, val >= 0
602 && val < colors.length ? colors[val]
606 // this will overwrite any existing t-coffee scores for the alignment
607 AlignmentAnnotation aa = al.findOrCreateAnnotation(TCOFFEE_SCORE,
608 TCOFFEE_SCORE, false, s, null);
611 aa.label = "T-COFFEE";
612 aa.description = "" + id.getKey();
613 aa.annotations = annotations;
615 aa.belowAlignment = false;
616 aa.setScore(header.getScoreFor(id.getKey()));
617 aa.createSequenceMapping(s, s.getStart(), true);
618 s.addAlignmentAnnotation(aa);
619 aa.adjustForAlignment();
623 aa.graph = AlignmentAnnotation.NO_GRAPH;
624 aa.label = "T-COFFEE";
625 aa.description = "TCoffee column reliability score";
626 aa.annotations = annotations;
627 aa.belowAlignment = true;
629 aa.setScore(header.getScoreAvg());
631 aa.showAllColLabels = true;
632 aa.validateRangeAndDisplay();
640 public String print()
642 // TODO Auto-generated method stub