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 import jalview.analysis.SequenceIdMatcher;
24 import jalview.datamodel.AlignmentAnnotation;
25 import jalview.datamodel.AlignmentI;
26 import jalview.datamodel.Annotation;
27 import jalview.datamodel.SequenceI;
29 import java.awt.Color;
30 import java.io.IOException;
31 import java.util.ArrayList;
32 import java.util.HashMap;
33 import java.util.LinkedHashMap;
34 import java.util.List;
36 import java.util.regex.Matcher;
37 import java.util.regex.Pattern;
40 * A file parse for T-Coffee score ascii format. This file contains the
41 * alignment consensus for each resude in any sequence.
43 * This file is procuded by <code>t_coffee</code> providing the option
44 * <code>-output=score_ascii </code> to the program command line
46 * An example file is the following
49 * T-COFFEE, Version_9.02.r1228 (2012-02-16 18:15:12 - Revision 1228 - Build 336)
66 * 1PHT 999999999999999999999999998762112222543211112134
67 * 1BB9 99999999999999999999999999987-------4322----2234
68 * 1UHC 99999999999999999999999999987-------5321----2246
69 * 1YCS 99999999999999999999999999986-------4321----1-35
70 * 1OOT 999999999999999999999999999861-------3------1135
71 * 1ABO 99999999999999999999999999986-------422-------34
72 * 1FYN 99999999999999999999999999985-------32--------35
73 * 1QCF 99999999999999999999999999974-------2---------24
74 * cons 999999999999999999999999999851000110321100001134
77 * 1PHT ----------5666642367889999999999889
78 * 1BB9 1111111111676653-355679999999999889
79 * 1UHC ----------788774--66789999999999889
80 * 1YCS ----------78777--356789999999999889
81 * 1OOT ----------78877--356789999999997-67
82 * 1ABO ----------687774--56779999999999889
83 * 1FYN ----------6888842356789999999999889
84 * 1QCF ----------6878742356789999999999889
85 * cons 00100000006877641356789999999999889
89 * @author Paolo Di Tommaso
92 public class TCoffeeScoreFile extends AlignFile
94 public TCoffeeScoreFile(String inFile, String type) throws IOException
100 public TCoffeeScoreFile(FileParse source) throws IOException
105 /** The {@link Header} structure holder */
109 * Holds the consensues values for each sequences. It uses a LinkedHashMap to
110 * maintaint the insertion order.
112 LinkedHashMap<String, StringBuilder> scores;
117 * Parse the provided reader for the T-Coffee scores file format
120 * public static TCoffeeScoreFile load(Reader reader) {
122 * try { BufferedReader in = (BufferedReader) (reader instanceof
123 * BufferedReader ? reader : new BufferedReader(reader));
124 * TCoffeeScoreFile result = new TCoffeeScoreFile();
125 * result.doParsing(in); return result.header != null &&
126 * result.scores != null ? result : null; } catch( Exception e) {
127 * throw new RuntimeException(e); } }
131 * @return The 'height' of the score matrix i.e. the numbers of score rows
132 * that should matches the number of sequences in the alignment
134 public int getHeight()
136 // the last entry will always be the 'global' alingment consensus scores, so
138 // from the 'height' count to make this value compatible with the number of
139 // sequences in the MSA
140 return scores != null && scores.size() > 0 ? scores.size() - 1 : 0;
144 * @return The 'width' of the score matrix i.e. the number of columns. Since
145 * the score value are supposed to be calculated for an 'aligned' MSA,
146 * all the entries have to have the same width.
148 public int getWidth()
150 return fWidth != null ? fWidth : 0;
154 * Get the string of score values for the specified seqeunce ID.
158 * @return The scores as a string of values e.g. {@code 99999987-------432}.
159 * It return an empty string when the specified ID is missing.
161 public String getScoresFor(String id)
163 return scores != null && scores.containsKey(id) ? scores.get(id)
168 * @return The list of score string as a {@link List} object, in the same
169 * ordeer of the insertion i.e. in the MSA
171 public List<String> getScoresList()
177 List<String> result = new ArrayList<String>(scores.size());
178 for (Map.Entry<String, StringBuilder> it : scores.entrySet())
180 result.add(it.getValue().toString());
187 * @return The parsed score values a matrix of bytes
189 public byte[][] getScoresArray()
195 byte[][] result = new byte[scores.size()][];
198 for (Map.Entry<String, StringBuilder> it : scores.entrySet())
200 String line = it.getValue().toString();
201 byte[] seqValues = new byte[line.length()];
202 for (int j = 0, c = line.length(); j < c; j++)
205 byte val = (byte) (line.charAt(j) - '0');
207 seqValues[j] = (val >= 0 && val <= 9) ? val : -1;
210 result[rowCount++] = seqValues;
216 public void parse() throws IOException
221 header = readHeader(this);
228 scores = new LinkedHashMap<String, StringBuilder>();
231 * initilize the structure
233 for (Map.Entry<String, Integer> entry : header.scores.entrySet())
235 scores.put(entry.getKey(), new StringBuilder());
239 * go with the reading
242 while ((block = readBlock(this, header.scores.size())) != null)
246 * append sequences read in the block
248 for (Map.Entry<String, String> entry : block.items.entrySet())
250 StringBuilder scoreStringBuilder = scores.get(entry.getKey());
251 if (scoreStringBuilder == null)
254 errormessage = String
255 .format("Invalid T-Coffee score file: Sequence ID '%s' is not declared in header section",
260 scoreStringBuilder.append(entry.getValue());
265 * verify that all rows have the same width
267 for (StringBuilder str : scores.values())
271 fWidth = str.length();
273 else if (fWidth != str.length())
276 errormessage = "Invalid T-Coffee score file: All the score sequences must have the same length";
284 static int parseInt(String str)
288 return Integer.parseInt(str);
289 } catch (NumberFormatException e)
291 // TODO report a warning ?
297 * Reaad the header section in the T-Coffee score file format
301 * @return The parser {@link Header} instance
302 * @throws RuntimeException
303 * when the header is not in the expected format
305 static Header readHeader(FileParse reader) throws IOException
308 Header result = null;
311 result = new Header();
312 result.head = reader.nextLine();
316 while ((line = reader.nextLine()) != null)
318 if (line.startsWith("SCORE="))
320 result.score = parseInt(line.substring(6).trim());
325 if ((line = reader.nextLine()) == null || !"*".equals(line.trim()))
328 "Invalid T-COFFEE score format (NO BAD/AVG/GOOD header)");
331 if ((line = reader.nextLine()) == null
332 || !"BAD AVG GOOD".equals(line.trim()))
335 "Invalid T-COFFEE score format (NO BAD/AVG/GOOD header)");
338 if ((line = reader.nextLine()) == null || !"*".equals(line.trim()))
341 "Invalid T-COFFEE score format (NO BAD/AVG/GOOD header)");
346 * now are expected a list if sequences ID up to the first blank line
348 while ((line = reader.nextLine()) != null)
355 int p = line.indexOf(":");
358 // TODO report a warning
362 String id = line.substring(0, p).trim();
363 int val = parseInt(line.substring(p + 1).trim());
366 // TODO report warning
370 result.scores.put(id, val);
375 error(reader, "T-COFFEE score file had no per-sequence scores");
378 } catch (IOException e)
380 error(reader, "Unexpected problem parsing T-Coffee score ascii file");
387 private static void error(FileParse reader, String errm)
390 if (reader.errormessage == null)
392 reader.errormessage = errm;
396 reader.errormessage += "\n" + errm;
400 static Pattern SCORES_WITH_RESIDUE_NUMS = Pattern
401 .compile("^\\d+\\s([^\\s]+)\\s+\\d+$");
404 * Read a scores block ihe provided stream.
407 * The stream to parse
409 * The expected number of the sequence to be read
410 * @return The {@link Block} instance read or {link null} null if the end of
412 * @throws IOException
413 * Something went wrong on the 'wire'
415 static Block readBlock(FileParse reader, int size) throws IOException
417 Block result = new Block(size);
421 * read blank lines (eventually)
423 while ((line = reader.nextLine()) != null && "".equals(line.trim()))
425 // consume blank lines
434 * read the scores block
438 if ("".equals(line.trim()))
444 // split the line on the first blank
445 // the first part have to contain the sequence id
446 // the remaining part are the scores values
447 int p = line.indexOf(" ");
450 if (reader.warningMessage == null)
452 reader.warningMessage = "";
454 reader.warningMessage += "Possible parsing error - expected to find a space in line: '"
459 String id = line.substring(0, p).trim();
460 String val = line.substring(p + 1).trim();
462 Matcher m = SCORES_WITH_RESIDUE_NUMS.matcher(val);
468 result.items.put(id, val);
470 } while ((line = reader.nextLine()) != null);
476 * The score file header
484 LinkedHashMap<String, Integer> scores = new LinkedHashMap<String, Integer>();
486 public int getScoreAvg()
491 public int getScoreFor(String ID)
494 return scores.containsKey(ID) ? scores.get(ID) : -1;
500 * Hold a single block values block in the score file
506 Map<String, String> items;
508 public Block(int size)
511 this.items = new HashMap<String, String>(size);
514 String getScoresFor(String id)
516 return items.get(id);
519 String getConsensus()
521 return items.get("cons");
526 * TCOFFEE score colourscheme
528 static final Color[] colors =
529 { new Color(102, 102, 255), // #6666FF
530 new Color(0, 255, 0), // #00FF00
531 new Color(102, 255, 0), // #66FF00
532 new Color(204, 255, 0), // #CCFF00
533 new Color(255, 255, 0), // #FFFF00
534 new Color(255, 204, 0), // #FFCC00
535 new Color(255, 153, 0), // #FF9900
536 new Color(255, 102, 0), // #FF6600
537 new Color(255, 51, 0), // #FF3300
538 new Color(255, 34, 0) // #FF2000
541 public final static String TCOFFEE_SCORE = "TCoffeeScore";
544 * generate annotation for this TCoffee score set on the given alignment
547 * alignment to annotate
549 * if true, annotate sequences based on matching sequence names
550 * @return true if alignment annotation was modified, false otherwise.
552 public boolean annotateAlignment(AlignmentI al, boolean matchids)
554 if (al.getHeight() != getHeight() || al.getWidth() != getWidth())
556 String info = String.format(
557 "align w: %s, h: %s; score: w: %s; h: %s ", al.getWidth(),
558 al.getHeight(), getWidth(), getHeight());
559 warningMessage = "Alignment shape does not match T-Coffee score file shape -- "
563 boolean added = false;
565 SequenceIdMatcher sidmatcher = new SequenceIdMatcher(
566 al.getSequencesArray());
567 byte[][] scoreMatrix = getScoresArray();
568 // for 2.8 - we locate any existing TCoffee annotation and remove it first
569 // before adding this.
570 for (Map.Entry<String, StringBuilder> id : scores.entrySet())
572 byte[] srow = scoreMatrix[i];
576 s = sidmatcher.findIdMatch(id.getKey());
580 s = al.getSequenceAt(i);
583 if (s == null && i != scores.size() && !id.getKey().equals("cons"))
585 System.err.println("No "
586 + (matchids ? "match " : " sequences left ")
587 + " for TCoffee score set : " + id.getKey());
590 int jSize = al.getWidth() < srow.length ? al.getWidth() : srow.length;
591 Annotation[] annotations = new Annotation[al.getWidth()];
592 for (int j = 0; j < jSize; j++)
595 if (s != null && jalview.util.Comparison.isGap(s.getCharAt(j)))
597 annotations[j] = null;
601 .println("Warning: non-zero value for positional T-COFFEE score for gap at "
602 + j + " in sequence " + s.getName());
607 annotations[j] = new Annotation(s == null ? "" + val : null,
608 s == null ? "" + val : null, '\0', val * 1f, val >= 0
609 && val < colors.length ? colors[val]
613 // this will overwrite any existing t-coffee scores for the alignment
614 AlignmentAnnotation aa = al.findOrCreateAnnotation(TCOFFEE_SCORE,
615 TCOFFEE_SCORE, false, s, null);
618 aa.label = "T-COFFEE";
619 aa.description = "" + id.getKey();
620 aa.annotations = annotations;
622 aa.belowAlignment = false;
623 aa.setScore(header.getScoreFor(id.getKey()));
624 aa.createSequenceMapping(s, s.getStart(), true);
625 s.addAlignmentAnnotation(aa);
626 aa.adjustForAlignment();
630 aa.graph = AlignmentAnnotation.NO_GRAPH;
631 aa.label = "T-COFFEE";
632 aa.description = "TCoffee column reliability score";
633 aa.annotations = annotations;
634 aa.belowAlignment = true;
636 aa.setScore(header.getScoreAvg());
638 aa.showAllColLabels = true;
639 aa.validateRangeAndDisplay();
647 public String print()
649 // TODO Auto-generated method stub