2 * Jalview - A Sequence Alignment Editor and Viewer (Version 2.8.0b1)
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 {
91 public TCoffeeScoreFile(String inFile, String type) throws IOException
97 public TCoffeeScoreFile(FileParse source) throws IOException
102 /** The {@link Header} structure holder */
106 * Holds the consensues values for each sequences. It uses a LinkedHashMap to
107 * maintaint the insertion order.
109 LinkedHashMap<String, StringBuilder> scores;
114 * Parse the provided reader for the T-Coffee scores file format
117 * public static TCoffeeScoreFile load(Reader reader) {
119 * try { BufferedReader in = (BufferedReader) (reader instanceof
120 * BufferedReader ? reader : new BufferedReader(reader));
121 * TCoffeeScoreFile result = new TCoffeeScoreFile();
122 * result.doParsing(in); return result.header != null &&
123 * result.scores != null ? result : null; } catch( Exception e) {
124 * throw new RuntimeException(e); } }
128 * @return The 'height' of the score matrix i.e. the numbers of score rows
129 * that should matches the number of sequences in the alignment
131 public int getHeight()
133 // the last entry will always be the 'global' alingment consensus scores, so
135 // from the 'height' count to make this value compatible with the number of
136 // sequences in the MSA
137 return scores != null && scores.size() > 0 ? scores.size() - 1 : 0;
141 * @return The 'width' of the score matrix i.e. the number of columns. Since
142 * the score value are supposed to be calculated for an 'aligned' MSA,
143 * all the entries have to have the same width.
145 public int getWidth()
147 return fWidth != null ? fWidth : 0;
151 * Get the string of score values for the specified seqeunce ID.
155 * @return The scores as a string of values e.g. {@code 99999987-------432}.
156 * It return an empty string when the specified ID is missing.
158 public String getScoresFor(String id)
160 return scores != null && scores.containsKey(id) ? scores.get(id)
165 * @return The list of score string as a {@link List} object, in the same
166 * ordeer of the insertion i.e. in the MSA
168 public List<String> getScoresList()
174 List<String> result = new ArrayList<String>(scores.size());
175 for (Map.Entry<String, StringBuilder> it : scores.entrySet())
177 result.add(it.getValue().toString());
184 * @return The parsed score values a matrix of bytes
186 public byte[][] getScoresArray()
192 byte[][] result = new byte[scores.size()][];
195 for (Map.Entry<String, StringBuilder> it : scores.entrySet())
197 String line = it.getValue().toString();
198 byte[] seqValues = new byte[line.length()];
199 for (int j = 0, c = line.length(); j < c; j++)
202 byte val = (byte) (line.charAt(j) - '0');
204 seqValues[j] = (val >= 0 && val <= 9) ? val : -1;
207 result[rowCount++] = seqValues;
213 public void parse() throws IOException
218 header = readHeader(this);
225 scores = new LinkedHashMap<String, StringBuilder>();
228 * initilize the structure
230 for (Map.Entry<String, Integer> entry : header.scores.entrySet())
232 scores.put(entry.getKey(), new StringBuilder());
236 * go with the reading
239 while ((block = readBlock(this, header.scores.size())) != null)
243 * append sequences read in the block
245 for (Map.Entry<String, String> entry : block.items.entrySet())
247 StringBuilder scoreStringBuilder = scores.get(entry.getKey());
248 if (scoreStringBuilder == null)
251 errormessage = String
252 .format("Invalid T-Coffee score file: Sequence ID '%s' is not declared in header section",
257 scoreStringBuilder.append(entry.getValue());
262 * verify that all rows have the same width
264 for (StringBuilder str : scores.values())
268 fWidth = str.length();
270 else if (fWidth != str.length())
273 errormessage = "Invalid T-Coffee score file: All the score sequences must have the same length";
281 static int parseInt(String str)
285 return Integer.parseInt(str);
286 } catch (NumberFormatException e)
288 // TODO report a warning ?
294 * Reaad the header section in the T-Coffee score file format
298 * @return The parser {@link Header} instance
299 * @throws RuntimeException
300 * when the header is not in the expected format
302 static Header readHeader(FileParse reader) throws IOException
305 Header result = null;
308 result = new Header();
309 result.head = reader.nextLine();
313 while ((line = reader.nextLine()) != null)
315 if (line.startsWith("SCORE="))
317 result.score = parseInt(line.substring(6).trim());
322 if ((line = reader.nextLine()) == null || !"*".equals(line.trim()))
325 "Invalid T-COFFEE score format (NO BAD/AVG/GOOD header)");
328 if ((line = reader.nextLine()) == null
329 || !"BAD AVG GOOD".equals(line.trim()))
332 "Invalid T-COFFEE score format (NO BAD/AVG/GOOD header)");
335 if ((line = reader.nextLine()) == null || !"*".equals(line.trim()))
338 "Invalid T-COFFEE score format (NO BAD/AVG/GOOD header)");
343 * now are expected a list if sequences ID up to the first blank line
345 while ((line = reader.nextLine()) != null)
352 int p = line.indexOf(":");
355 // TODO report a warning
359 String id = line.substring(0, p).trim();
360 int val = parseInt(line.substring(p + 1).trim());
363 // TODO report warning
367 result.scores.put(id, val);
372 error(reader, "T-COFFEE score file had no per-sequence scores");
375 } catch (IOException e)
377 error(reader, "Unexpected problem parsing T-Coffee score ascii file");
384 private static void error(FileParse reader, String errm)
387 if (reader.errormessage == null)
389 reader.errormessage = errm;
393 reader.errormessage += "\n" + errm;
397 static Pattern SCORES_WITH_RESIDUE_NUMS = Pattern.compile("^\\d+\\s([^\\s]+)\\s+\\d+$");
400 * Read a scores block ihe provided stream.
403 * The stream to parse
405 * The expected number of the sequence to be read
406 * @return The {@link Block} instance read or {link null} null if the end of
408 * @throws IOException
409 * Something went wrong on the 'wire'
411 static Block readBlock(FileParse reader, int size) throws IOException
413 Block result = new Block(size);
417 * read blank lines (eventually)
419 while ((line = reader.nextLine()) != null && "".equals(line.trim()))
421 // consume blank lines
430 * read the scores block
434 if ("".equals(line.trim()))
440 // split the line on the first blank
441 // the first part have to contain the sequence id
442 // the remaining part are the scores values
443 int p = line.indexOf(" ");
446 if (reader.warningMessage == null)
448 reader.warningMessage = "";
450 reader.warningMessage += "Possible parsing error - expected to find a space in line: '"
455 String id = line.substring(0, p).trim();
456 String val = line.substring(p + 1).trim();
458 Matcher m = SCORES_WITH_RESIDUE_NUMS.matcher(val);
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 String info = String.format("align w: %s, h: %s; score: w: %s; h: %s ", al.getWidth(), al.getHeight(), getWidth(), getHeight() );
552 warningMessage = "Alignment shape does not match T-Coffee score file shape -- " + info;
555 boolean added = false;
557 SequenceIdMatcher sidmatcher = new SequenceIdMatcher(
558 al.getSequencesArray());
559 byte[][] scoreMatrix = getScoresArray();
560 // for 2.8 - we locate any existing TCoffee annotation and remove it first
561 // before adding this.
562 for (Map.Entry<String, StringBuilder> id : scores.entrySet())
564 byte[] srow = scoreMatrix[i];
568 s = sidmatcher.findIdMatch(id.getKey());
572 s = al.getSequenceAt(i);
575 if (s == null && i != scores.size() && !id.getKey().equals("cons"))
577 System.err.println("No "
578 + (matchids ? "match " : " sequences left ")
579 + " for TCoffee score set : " + id.getKey());
582 int jSize = al.getWidth() < srow.length ? al.getWidth() : srow.length;
583 Annotation[] annotations = new Annotation[al.getWidth()];
584 for (int j = 0; j < jSize; j++)
587 if (s != null && jalview.util.Comparison.isGap(s.getCharAt(j)))
589 annotations[j] = null;
593 .println("Warning: non-zero value for positional T-COFFEE score for gap at "
594 + j + " in sequence " + s.getName());
599 annotations[j] = new Annotation(s == null ? "" + val : null,
600 s == null ? "" + val : null, '\0', val * 1f, val >= 0
601 && val < colors.length ? colors[val]
605 // this will overwrite any existing t-coffee scores for the alignment
606 AlignmentAnnotation aa = al.findOrCreateAnnotation(TCOFFEE_SCORE,
607 TCOFFEE_SCORE, false, s, null);
610 aa.label = "T-COFFEE";
611 aa.description = "" + id.getKey();
612 aa.annotations = annotations;
614 aa.belowAlignment = false;
615 aa.setScore(header.getScoreFor(id.getKey()));
616 aa.createSequenceMapping(s, s.getStart(), true);
617 s.addAlignmentAnnotation(aa);
618 aa.adjustForAlignment();
622 aa.graph = AlignmentAnnotation.NO_GRAPH;
623 aa.label = "T-COFFEE";
624 aa.description = "TCoffee column reliability score";
625 aa.annotations = annotations;
626 aa.belowAlignment = true;
628 aa.setScore(header.getScoreAvg());
630 aa.showAllColLabels = true;
631 aa.validateRangeAndDisplay();
639 public String print()
641 // TODO Auto-generated method stub