2 * Jalview - A Sequence Alignment Editor and Viewer (Version 2.8.1)
3 * Copyright (C) 2014 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 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/>.
17 * The Jalview Authors are detailed in the 'AUTHORS' file.
21 import jalview.analysis.SequenceIdMatcher;
22 import jalview.datamodel.AlignmentAnnotation;
23 import jalview.datamodel.AlignmentI;
24 import jalview.datamodel.Annotation;
25 import jalview.datamodel.SequenceI;
27 import java.awt.Color;
28 import java.io.IOException;
29 import java.util.ArrayList;
30 import java.util.HashMap;
31 import java.util.LinkedHashMap;
32 import java.util.List;
34 import java.util.regex.Matcher;
35 import java.util.regex.Pattern;
38 * A file parse for T-Coffee score ascii format. This file contains the
39 * alignment consensus for each resude in any sequence.
41 * This file is procuded by <code>t_coffee</code> providing the option
42 * <code>-output=score_ascii </code> to the program command line
44 * An example file is the following
47 * T-COFFEE, Version_9.02.r1228 (2012-02-16 18:15:12 - Revision 1228 - Build 336)
64 * 1PHT 999999999999999999999999998762112222543211112134
65 * 1BB9 99999999999999999999999999987-------4322----2234
66 * 1UHC 99999999999999999999999999987-------5321----2246
67 * 1YCS 99999999999999999999999999986-------4321----1-35
68 * 1OOT 999999999999999999999999999861-------3------1135
69 * 1ABO 99999999999999999999999999986-------422-------34
70 * 1FYN 99999999999999999999999999985-------32--------35
71 * 1QCF 99999999999999999999999999974-------2---------24
72 * cons 999999999999999999999999999851000110321100001134
75 * 1PHT ----------5666642367889999999999889
76 * 1BB9 1111111111676653-355679999999999889
77 * 1UHC ----------788774--66789999999999889
78 * 1YCS ----------78777--356789999999999889
79 * 1OOT ----------78877--356789999999997-67
80 * 1ABO ----------687774--56779999999999889
81 * 1FYN ----------6888842356789999999999889
82 * 1QCF ----------6878742356789999999999889
83 * cons 00100000006877641356789999999999889
87 * @author Paolo Di Tommaso
90 public class TCoffeeScoreFile extends AlignFile
93 public TCoffeeScoreFile(String inFile, String type) throws IOException
99 public TCoffeeScoreFile(FileParse source) throws IOException
104 /** The {@link Header} structure holder */
108 * Holds the consensues values for each sequences. It uses a LinkedHashMap to
109 * maintaint the insertion order.
111 LinkedHashMap<String, StringBuilder> scores;
116 * Parse the provided reader for the T-Coffee scores file format
119 * public static TCoffeeScoreFile load(Reader reader) {
121 * try { BufferedReader in = (BufferedReader) (reader instanceof
122 * BufferedReader ? reader : new BufferedReader(reader));
123 * TCoffeeScoreFile result = new TCoffeeScoreFile();
124 * result.doParsing(in); return result.header != null &&
125 * result.scores != null ? result : null; } catch( Exception e) {
126 * throw new RuntimeException(e); } }
130 * @return The 'height' of the score matrix i.e. the numbers of score rows
131 * that should matches the number of sequences in the alignment
133 public int getHeight()
135 // the last entry will always be the 'global' alingment consensus scores, so
137 // from the 'height' count to make this value compatible with the number of
138 // sequences in the MSA
139 return scores != null && scores.size() > 0 ? scores.size() - 1 : 0;
143 * @return The 'width' of the score matrix i.e. the number of columns. Since
144 * the score value are supposed to be calculated for an 'aligned' MSA,
145 * all the entries have to have the same width.
147 public int getWidth()
149 return fWidth != null ? fWidth : 0;
153 * Get the string of score values for the specified seqeunce ID.
157 * @return The scores as a string of values e.g. {@code 99999987-------432}.
158 * It return an empty string when the specified ID is missing.
160 public String getScoresFor(String id)
162 return scores != null && scores.containsKey(id) ? scores.get(id)
167 * @return The list of score string as a {@link List} object, in the same
168 * ordeer of the insertion i.e. in the MSA
170 public List<String> getScoresList()
176 List<String> result = new ArrayList<String>(scores.size());
177 for (Map.Entry<String, StringBuilder> it : scores.entrySet())
179 result.add(it.getValue().toString());
186 * @return The parsed score values a matrix of bytes
188 public byte[][] getScoresArray()
194 byte[][] result = new byte[scores.size()][];
197 for (Map.Entry<String, StringBuilder> it : scores.entrySet())
199 String line = it.getValue().toString();
200 byte[] seqValues = new byte[line.length()];
201 for (int j = 0, c = line.length(); j < c; j++)
204 byte val = (byte) (line.charAt(j) - '0');
206 seqValues[j] = (val >= 0 && val <= 9) ? val : -1;
209 result[rowCount++] = seqValues;
215 public void parse() throws IOException
220 header = readHeader(this);
227 scores = new LinkedHashMap<String, StringBuilder>();
230 * initilize the structure
232 for (Map.Entry<String, Integer> entry : header.scores.entrySet())
234 scores.put(entry.getKey(), new StringBuilder());
238 * go with the reading
241 while ((block = readBlock(this, header.scores.size())) != null)
245 * append sequences read in the block
247 for (Map.Entry<String, String> entry : block.items.entrySet())
249 StringBuilder scoreStringBuilder = scores.get(entry.getKey());
250 if (scoreStringBuilder == null)
253 errormessage = String
254 .format("Invalid T-Coffee score file: Sequence ID '%s' is not declared in header section",
259 scoreStringBuilder.append(entry.getValue());
264 * verify that all rows have the same width
266 for (StringBuilder str : scores.values())
270 fWidth = str.length();
272 else if (fWidth != str.length())
275 errormessage = "Invalid T-Coffee score file: All the score sequences must have the same length";
283 static int parseInt(String str)
287 return Integer.parseInt(str);
288 } catch (NumberFormatException e)
290 // TODO report a warning ?
296 * Reaad the header section in the T-Coffee score file format
300 * @return The parser {@link Header} instance
301 * @throws RuntimeException
302 * when the header is not in the expected format
304 static Header readHeader(FileParse reader) throws IOException
307 Header result = null;
310 result = new Header();
311 result.head = reader.nextLine();
315 while ((line = reader.nextLine()) != null)
317 if (line.startsWith("SCORE="))
319 result.score = parseInt(line.substring(6).trim());
324 if ((line = reader.nextLine()) == null || !"*".equals(line.trim()))
327 "Invalid T-COFFEE score format (NO BAD/AVG/GOOD header)");
330 if ((line = reader.nextLine()) == null
331 || !"BAD AVG GOOD".equals(line.trim()))
334 "Invalid T-COFFEE score format (NO BAD/AVG/GOOD header)");
337 if ((line = reader.nextLine()) == null || !"*".equals(line.trim()))
340 "Invalid T-COFFEE score format (NO BAD/AVG/GOOD header)");
345 * now are expected a list if sequences ID up to the first blank line
347 while ((line = reader.nextLine()) != null)
354 int p = line.indexOf(":");
357 // TODO report a warning
361 String id = line.substring(0, p).trim();
362 int val = parseInt(line.substring(p + 1).trim());
365 // TODO report warning
369 result.scores.put(id, val);
374 error(reader, "T-COFFEE score file had no per-sequence scores");
377 } catch (IOException e)
379 error(reader, "Unexpected problem parsing T-Coffee score ascii file");
386 private static void error(FileParse reader, String errm)
389 if (reader.errormessage == null)
391 reader.errormessage = errm;
395 reader.errormessage += "\n" + errm;
399 static Pattern SCORES_WITH_RESIDUE_NUMS = Pattern.compile("^\\d+\\s([^\\s]+)\\s+\\d+$");
402 * Read a scores block ihe provided stream.
405 * The stream to parse
407 * The expected number of the sequence to be read
408 * @return The {@link Block} instance read or {link null} null if the end of
410 * @throws IOException
411 * Something went wrong on the 'wire'
413 static Block readBlock(FileParse reader, int size) throws IOException
415 Block result = new Block(size);
419 * read blank lines (eventually)
421 while ((line = reader.nextLine()) != null && "".equals(line.trim()))
423 // consume blank lines
432 * read the scores block
436 if ("".equals(line.trim()))
442 // split the line on the first blank
443 // the first part have to contain the sequence id
444 // the remaining part are the scores values
445 int p = line.indexOf(" ");
448 if (reader.warningMessage == null)
450 reader.warningMessage = "";
452 reader.warningMessage += "Possible parsing error - expected to find a space in line: '"
457 String id = line.substring(0, p).trim();
458 String val = line.substring(p + 1).trim();
460 Matcher m = SCORES_WITH_RESIDUE_NUMS.matcher(val);
465 result.items.put(id, val);
467 } while ((line = reader.nextLine()) != null);
473 * The score file header
481 LinkedHashMap<String, Integer> scores = new LinkedHashMap<String, Integer>();
483 public int getScoreAvg()
488 public int getScoreFor(String ID)
491 return scores.containsKey(ID) ? scores.get(ID) : -1;
497 * Hold a single block values block in the score file
503 Map<String, String> items;
505 public Block(int size)
508 this.items = new HashMap<String, String>(size);
511 String getScoresFor(String id)
513 return items.get(id);
516 String getConsensus()
518 return items.get("cons");
523 * TCOFFEE score colourscheme
525 static final Color[] colors =
526 { new Color(102, 102, 255), // #6666FF
527 new Color(0, 255, 0), // #00FF00
528 new Color(102, 255, 0), // #66FF00
529 new Color(204, 255, 0), // #CCFF00
530 new Color(255, 255, 0), // #FFFF00
531 new Color(255, 204, 0), // #FFCC00
532 new Color(255, 153, 0), // #FF9900
533 new Color(255, 102, 0), // #FF6600
534 new Color(255, 51, 0), // #FF3300
535 new Color(255, 34, 0) // #FF2000
538 public final static String TCOFFEE_SCORE = "TCoffeeScore";
541 * generate annotation for this TCoffee score set on the given alignment
544 * alignment to annotate
546 * if true, annotate sequences based on matching sequence names
547 * @return true if alignment annotation was modified, false otherwise.
549 public boolean annotateAlignment(AlignmentI al, boolean matchids)
551 if (al.getHeight() != getHeight() || al.getWidth() != getWidth())
553 String info = String.format("align w: %s, h: %s; score: w: %s; h: %s ", al.getWidth(), al.getHeight(), getWidth(), getHeight() );
554 warningMessage = "Alignment shape does not match T-Coffee score file shape -- " + info;
557 boolean added = false;
559 SequenceIdMatcher sidmatcher = new SequenceIdMatcher(
560 al.getSequencesArray());
561 byte[][] scoreMatrix = getScoresArray();
562 // for 2.8 - we locate any existing TCoffee annotation and remove it first
563 // before adding this.
564 for (Map.Entry<String, StringBuilder> id : scores.entrySet())
566 byte[] srow = scoreMatrix[i];
570 s = sidmatcher.findIdMatch(id.getKey());
574 s = al.getSequenceAt(i);
577 if (s == null && i != scores.size() && !id.getKey().equals("cons"))
579 System.err.println("No "
580 + (matchids ? "match " : " sequences left ")
581 + " for TCoffee score set : " + id.getKey());
584 int jSize = al.getWidth() < srow.length ? al.getWidth() : srow.length;
585 Annotation[] annotations = new Annotation[al.getWidth()];
586 for (int j = 0; j < jSize; j++)
589 if (s != null && jalview.util.Comparison.isGap(s.getCharAt(j)))
591 annotations[j] = null;
595 .println("Warning: non-zero value for positional T-COFFEE score for gap at "
596 + j + " in sequence " + s.getName());
601 annotations[j] = new Annotation(s == null ? "" + val : null,
602 s == null ? "" + val : null, '\0', val * 1f, val >= 0
603 && val < colors.length ? colors[val]
607 // this will overwrite any existing t-coffee scores for the alignment
608 AlignmentAnnotation aa = al.findOrCreateAnnotation(TCOFFEE_SCORE,
609 TCOFFEE_SCORE, false, s, null);
612 aa.label = "T-COFFEE";
613 aa.description = "" + id.getKey();
614 aa.annotations = annotations;
616 aa.belowAlignment = false;
617 aa.setScore(header.getScoreFor(id.getKey()));
618 aa.createSequenceMapping(s, s.getStart(), true);
619 s.addAlignmentAnnotation(aa);
620 aa.adjustForAlignment();
624 aa.graph = AlignmentAnnotation.NO_GRAPH;
625 aa.label = "T-COFFEE";
626 aa.description = "TCoffee column reliability score";
627 aa.annotations = annotations;
628 aa.belowAlignment = true;
630 aa.setScore(header.getScoreAvg());
632 aa.showAllColLabels = true;
633 aa.validateRangeAndDisplay();
641 public String print()
643 // TODO Auto-generated method stub