datamodel/compbio/data/sequence/SequenceUtil.java

   1 /* Copyright (c) 2011 Peter Troshin\r
   2  *  \r
   3  *  JAva Bioinformatics Analysis Web Services (JABAWS) @version: 2.0     \r
   4  * \r
   5  *  This library is free software; you can redistribute it and/or modify it under the terms of the\r
   6  *  Apache License version 2 as published by the Apache Software Foundation\r
   7  * \r
   8  *  This library is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without\r
   9  *  even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the Apache \r
  10  *  License for more details.\r
  11  * \r
  12  *  A copy of the license is in apache_license.txt. It is also available here:\r
  13  * @see: http://www.apache.org/licenses/LICENSE-2.0.txt\r
  14  * \r
  15  * Any republication or derived work distributed in source code form\r
  16  * must include this copyright and license notice.\r
  17  */\r
  18 \r
  19 package compbio.data.sequence;\r
  20 \r
  21 import java.io.BufferedReader;\r
  22 import java.io.BufferedWriter;\r
  23 import java.io.Closeable;\r
  24 import java.io.File;\r
  25 import java.io.FileInputStream;\r
  26 import java.io.FileNotFoundException;\r
  27 import java.io.IOException;\r
  28 import java.io.InputStream;\r
  29 import java.io.InputStreamReader;\r
  30 import java.io.OutputStream;\r
  31 import java.io.OutputStreamWriter;\r
  32 import java.util.ArrayList;\r
  33 import java.util.Arrays;\r
  34 import java.util.HashMap;\r
  35 import java.util.HashSet;\r
  36 import java.util.List;\r
  37 import java.util.Map;\r
  38 import java.util.Scanner;\r
  39 import java.util.Set;\r
  40 import java.util.TreeMap;\r
  41 import java.util.TreeSet;\r
  42 import java.util.logging.Level;\r
  43 import java.util.regex.Matcher;\r
  44 import java.util.regex.Pattern;\r
  45 \r
  46 import compbio.util.Util;\r
  47 \r
  48 /**\r
  49  * Utility class for operations on sequences\r
  50  * \r
  51  * @author Peter Troshin\r
  52  * @since 1.0\r
  53  * @version 2.0 June 2011\r
  54  */\r
  55 public final class SequenceUtil {\r
  56 \r
  57         /**\r
  58          * A whitespace character: [\t\n\x0B\f\r]\r
  59          */\r
  60         public static final Pattern WHITE_SPACE = Pattern.compile("\\s");\r
  61 \r
  62         /**\r
  63          * A digit\r
  64          */\r
  65         public static final Pattern DIGIT = Pattern.compile("\\d");\r
  66 \r
  67         /**\r
  68          * Non word\r
  69          */\r
  70         public static final Pattern NONWORD = Pattern.compile("\\W");\r
  71 \r
  72         /**\r
  73          * Valid Amino acids\r
  74          */\r
  75         public static final Pattern AA = Pattern.compile("[ARNDCQEGHILKMFPSTWYV]+",\r
  76                         Pattern.CASE_INSENSITIVE);\r
  77 \r
  78         /**\r
  79          * inversion of AA pattern\r
  80          */\r
  81         public static final Pattern NON_AA = Pattern.compile(\r
  82                         "[^ARNDCQEGHILKMFPSTWYV]+", Pattern.CASE_INSENSITIVE);\r
  83 \r
  84         /**\r
  85          * Same as AA pattern but with two additional letters - XU\r
  86          */\r
  87         public static final Pattern AMBIGUOUS_AA = Pattern.compile(\r
  88                         "[ARNDCQEGHILKMFPSTWYVXU]+", Pattern.CASE_INSENSITIVE);\r
  89 \r
  90         /**\r
  91          * Nucleotides a, t, g, c, u\r
  92          */\r
  93         public static final Pattern NUCLEOTIDE = Pattern.compile("[AGTCU]+",\r
  94                         Pattern.CASE_INSENSITIVE);\r
  95 \r
  96         /**\r
  97          * Ambiguous nucleotide\r
  98          */\r
  99         public static final Pattern AMBIGUOUS_NUCLEOTIDE = Pattern.compile(\r
 100                         "[AGTCRYMKSWHBVDNU]+", Pattern.CASE_INSENSITIVE); // see IUPAC\r
 101         /**\r
 102          * Non nucleotide\r
 103          */\r
 104         public static final Pattern NON_NUCLEOTIDE = Pattern.compile("[^AGTCU]+",\r
 105                         Pattern.CASE_INSENSITIVE);\r
 106 \r
 107         private SequenceUtil() {\r
 108         } // utility class, no instantiation\r
 109 \r
 110         /**\r
 111          * @return true is the sequence contains only letters a,c, t, g, u\r
 112          */\r
 113         public static boolean isNucleotideSequence(final FastaSequence s) {\r
 114                 return SequenceUtil.isNonAmbNucleotideSequence(s.getSequence());\r
 115         }\r
 116 \r
 117         /**\r
 118          * Ambiguous DNA chars : AGTCRYMKSWHBVDN // differs from protein in only one\r
 119          * (!) - B char\r
 120          */\r
 121         public static boolean isNonAmbNucleotideSequence(String sequence) {\r
 122                 sequence = SequenceUtil.cleanSequence(sequence);\r
 123                 if (SequenceUtil.DIGIT.matcher(sequence).find()) {\r
 124                         return false;\r
 125                 }\r
 126                 if (SequenceUtil.NON_NUCLEOTIDE.matcher(sequence).find()) {\r
 127                         return false;\r
 128                         /*\r
 129                          * System.out.format("I found the text starting at " +\r
 130                          * "index %d and ending at index %d.%n", nonDNAmatcher .start(),\r
 131                          * nonDNAmatcher.end());\r
 132                          */\r
 133                 }\r
 134                 final Matcher DNAmatcher = SequenceUtil.NUCLEOTIDE.matcher(sequence);\r
 135                 return DNAmatcher.find();\r
 136         }\r
 137 \r
 138         /**\r
 139          * Removes all whitespace chars in the sequence string\r
 140          * \r
 141          * @param sequence\r
 142          * @return cleaned up sequence\r
 143          */\r
 144         public static String cleanSequence(String sequence) {\r
 145                 assert sequence != null;\r
 146                 final Matcher m = SequenceUtil.WHITE_SPACE.matcher(sequence);\r
 147                 sequence = m.replaceAll("").toUpperCase();\r
 148                 return sequence;\r
 149         }\r
 150 \r
 151         /**\r
 152          * Removes all special characters and digits as well as whitespace chars\r
 153          * from the sequence\r
 154          * \r
 155          * @param sequence\r
 156          * @return cleaned up sequence\r
 157          */\r
 158         public static String deepCleanSequence(String sequence) {\r
 159                 sequence = SequenceUtil.cleanSequence(sequence);\r
 160                 sequence = SequenceUtil.DIGIT.matcher(sequence).replaceAll("");\r
 161                 sequence = SequenceUtil.NONWORD.matcher(sequence).replaceAll("");\r
 162                 final Pattern othernonSeqChars = Pattern.compile("[_-]+");\r
 163                 sequence = othernonSeqChars.matcher(sequence).replaceAll("");\r
 164                 return sequence;\r
 165         }\r
 166 \r
 167         /**\r
 168          * Remove all non AA chars from the sequence\r
 169          * \r
 170          * @param sequence\r
 171          *            the sequence to clean\r
 172          * @return cleaned sequence\r
 173          */\r
 174         public static String cleanProteinSequence(String sequence) {\r
 175                 return SequenceUtil.NON_AA.matcher(sequence).replaceAll("");\r
 176         }\r
 177 \r
 178         /**\r
 179          * @param sequence\r
 180          * @return true is the sequence is a protein sequence, false overwise\r
 181          */\r
 182         public static boolean isProteinSequence(String sequence) {\r
 183                 sequence = SequenceUtil.cleanSequence(sequence);\r
 184                 if (SequenceUtil.isNonAmbNucleotideSequence(sequence)) {\r
 185                         return false;\r
 186                 }\r
 187                 if (SequenceUtil.DIGIT.matcher(sequence).find()) {\r
 188                         return false;\r
 189                 }\r
 190                 if (SequenceUtil.NON_AA.matcher(sequence).find()) {\r
 191                         return false;\r
 192                 }\r
 193                 final Matcher protmatcher = SequenceUtil.AA.matcher(sequence);\r
 194                 return protmatcher.find();\r
 195         }\r
 196 \r
 197         /**\r
 198          * Check whether the sequence confirms to amboguous protein sequence\r
 199          * \r
 200          * @param sequence\r
 201          * @return return true only if the sequence if ambiguous protein sequence\r
 202          *         Return false otherwise. e.g. if the sequence is non-ambiguous\r
 203          *         protein or DNA\r
 204          */\r
 205         public static boolean isAmbiguosProtein(String sequence) {\r
 206                 sequence = SequenceUtil.cleanSequence(sequence);\r
 207                 if (SequenceUtil.isNonAmbNucleotideSequence(sequence)) {\r
 208                         return false;\r
 209                 }\r
 210                 if (SequenceUtil.DIGIT.matcher(sequence).find()) {\r
 211                         return false;\r
 212                 }\r
 213                 if (SequenceUtil.NON_AA.matcher(sequence).find()) {\r
 214                         return false;\r
 215                 }\r
 216                 if (SequenceUtil.AA.matcher(sequence).find()) {\r
 217                         return false;\r
 218                 }\r
 219                 final Matcher amb_prot = SequenceUtil.AMBIGUOUS_AA.matcher(sequence);\r
 220                 return amb_prot.find();\r
 221         }\r
 222 \r
 223         /**\r
 224          * Writes list of FastaSequeces into the outstream formatting the sequence\r
 225          * so that it contains width chars on each line\r
 226          * \r
 227          * @param outstream\r
 228          * @param sequences\r
 229          * @param width\r
 230          *            - the maximum number of characters to write in one line\r
 231          * @throws IOException\r
 232          */\r
 233         public static void writeFasta(final OutputStream outstream,\r
 234                         final List<FastaSequence> sequences, final int width)\r
 235                         throws IOException {\r
 236                 writeFastaKeepTheStream(outstream, sequences, width);\r
 237                 outstream.close();\r
 238         }\r
 239 \r
 240         public static void writeFastaKeepTheStream(final OutputStream outstream,\r
 241                         final List<FastaSequence> sequences, final int width)\r
 242                         throws IOException {\r
 243                 final OutputStreamWriter writer = new OutputStreamWriter(outstream);\r
 244                 final BufferedWriter fastawriter = new BufferedWriter(writer);\r
 245                 for (final FastaSequence fs : sequences) {\r
 246                         fastawriter.write(">" + fs.getId() + "\n");\r
 247                         fastawriter.write(fs.getFormatedSequence(width));\r
 248                         fastawriter.write("\n");\r
 249                 }\r
 250                 fastawriter.flush();\r
 251                 writer.flush();\r
 252         }\r
 253 \r
 254         /**\r
 255          * Reads fasta sequences from inStream into the list of FastaSequence\r
 256          * objects\r
 257          * \r
 258          * @param inStream\r
 259          *            from\r
 260          * @return list of FastaSequence objects\r
 261          * @throws IOException\r
 262          */\r
 263         public static List<FastaSequence> readFasta(final InputStream inStream)\r
 264                         throws IOException {\r
 265                 final List<FastaSequence> seqs = new ArrayList<FastaSequence>();\r
 266                 FastaReader reader = new FastaReader(inStream);\r
 267                 while (reader.hasNext()) {\r
 268                         seqs.add(reader.next());\r
 269                 }\r
 270                 inStream.close();\r
 271                 return seqs;\r
 272         }\r
 273 \r
 274         /**\r
 275          * Writes FastaSequence in the file, each sequence will take one line only\r
 276          * \r
 277          * @param os\r
 278          * @param sequences\r
 279          * @throws IOException\r
 280          */\r
 281         public static void writeFasta(final OutputStream os,\r
 282                         final List<FastaSequence> sequences) throws IOException {\r
 283                 final OutputStreamWriter outWriter = new OutputStreamWriter(os);\r
 284                 final BufferedWriter fasta_out = new BufferedWriter(outWriter);\r
 285                 for (final FastaSequence fs : sequences) {\r
 286                         fasta_out.write(fs.getOnelineFasta());\r
 287                 }\r
 288                 fasta_out.close();\r
 289                 outWriter.close();\r
 290         }\r
 291 \r
 292         /**\r
 293          * Read IUPred output\r
 294          * \r
 295          * @param result\r
 296          * @return Map key->sequence name, value->Score\r
 297          * @throws IOException\r
 298          * @throws UnknownFileFormatException\r
 299          */\r
 300         public static Map<String, Score> readIUPred(final File result)\r
 301                         throws IOException, UnknownFileFormatException {\r
 302                 InputStream input = new FileInputStream(result);\r
 303                 Map<String, Score> sequences = readIUPred(input,\r
 304                                 IUPredResult.getType(result));\r
 305                 input.close();\r
 306                 return sequences;\r
 307         }\r
 308 \r
 309         // Check the type of the file e.g. long| short or domain\r
 310         // and read\r
 311         /**\r
 312          * ## Long Disorder\r
 313          * \r
 314          * # P53_HUMAN\r
 315          * \r
 316          * 1 M 0.9943\r
 317          * \r
 318          * 2 E 0.9917\r
 319          * \r
 320          * 3 E 0.9879\r
 321          * \r
 322          * (every line)\r
 323          * \r
 324          * @throws IOException\r
 325          * @throws UnknownFileFormatException\r
 326          * \r
 327          * \r
 328          */\r
 329         private static Map<String, Score> readIUPred(InputStream input,\r
 330                         IUPredResult type) throws IOException, UnknownFileFormatException {\r
 331 \r
 332                 Score score = null;\r
 333                 final Map<String, Score> seqs = new HashMap<String, Score>();\r
 334                 Scanner scan = new Scanner(input);\r
 335                 scan.useDelimiter("#");\r
 336                 while (scan.hasNext()) {\r
 337                         String nextEntry = scan.next();\r
 338                         Scanner entry = new Scanner(nextEntry);\r
 339                         String name = entry.nextLine().trim();\r
 340                         // inside entry:\r
 341                         if (IUPredResult.Glob == type) {\r
 342                                 // parse domains\r
 343                                 TreeSet<Range> ranges = parseIUPredDomains(entry);\r
 344                                 score = new Score(type, ranges);\r
 345                         } else {\r
 346                                 // parse short | long\r
 347                                 float[] scores = parseIUPredScores(entry);\r
 348                                 score = new Score(type, scores);\r
 349                         }\r
 350                         entry.close();\r
 351                         seqs.put(name, score);\r
 352                 }\r
 353 \r
 354                 scan.close();\r
 355                 return seqs;\r
 356         }\r
 357 \r
 358         /**\r
 359          * # P53_HUMA\r
 360          * \r
 361          * Number of globular domains: 2\r
 362          * \r
 363          * globular domain 1. 98 - 269\r
 364          * \r
 365          * globular domain 2. 431 - 482\r
 366          * \r
 367          * >P53_HUMA\r
 368          * \r
 369          * meepqsdpsv epplsqetfs dlwkllpenn vlsplpsqam ddlmlspddi eqwftedpgp\r
 370          * \r
 371          * @param scan\r
 372          */\r
 373         private static TreeSet<Range> parseIUPredDomains(Scanner scan) {\r
 374                 String header = "Number of globular domains:";\r
 375                 String domainPref = "globular domain";\r
 376                 TreeSet<Range> ranges = new TreeSet<Range>();\r
 377                 String line = scan.nextLine().trim();\r
 378                 assert line.startsWith(header);\r
 379                 line = line.substring(header.length()).trim();\r
 380                 int domainNum = Integer.parseInt(line);\r
 381                 if (domainNum == 0) {\r
 382                         return ranges;\r
 383                 }\r
 384 \r
 385                 for (int i = 0; i < domainNum; i++) {\r
 386                         assert scan.hasNextLine();\r
 387                         line = scan.nextLine();\r
 388                         assert line.trim().startsWith(domainPref);\r
 389                         line = line.substring(line.indexOf(".") + 1).trim();\r
 390                         Range r = new Range(line.split("-"));\r
 391                         ranges.add(r);\r
 392                 }\r
 393 \r
 394                 return ranges;\r
 395         }\r
 396         /*\r
 397          * 1 M 0.9943\r
 398          * \r
 399          * 2 E 0.9917\r
 400          */\r
 401         private static float[] parseIUPredScores(Scanner scan)\r
 402                         throws UnknownFileFormatException {\r
 403                 List<String> annotation = new ArrayList<String>();\r
 404                 while (scan.hasNextLine()) {\r
 405                         String line = scan.nextLine().trim();\r
 406                         String[] val = line.split("\\s+");\r
 407                         annotation.add(val[2]);\r
 408                 }\r
 409                 return convertToNumber(annotation\r
 410                                 .toArray(new String[annotation.size()]));\r
 411         }\r
 412 \r
 413         public static Map<String, Score> readJRonn(final File result)\r
 414                         throws IOException, UnknownFileFormatException {\r
 415                 InputStream input = new FileInputStream(result);\r
 416                 Map<String, Score> sequences = readJRonn(input);\r
 417                 input.close();\r
 418                 return sequences;\r
 419         }\r
 420 \r
 421         /**\r
 422          * Reader for JRonn horizontal file format\r
 423          * \r
 424          * <pre>\r
 425          * &gtFoobar M G D T T A G 0.48 0.42\r
 426          * 0.42 0.48 0.52 0.53 0.54\r
 427          * \r
 428          * <pre>\r
 429          * Where all values are tab delimited\r
 430          * \r
 431          * @param inStream\r
 432          *            the InputStream connected to the JRonn output file\r
 433          * @return Map key=sequence name value=Score\r
 434          * @throws IOException\r
 435          *             is thrown if the inStream has problems accessing the data\r
 436          * @throws UnknownFileFormatException\r
 437          *             is thrown if the inStream represents an unknown source of\r
 438          * data, i.e. not a JRonn output\r
 439          */\r
 440         public static Map<String, Score> readJRonn(final InputStream inStream)\r
 441                         throws IOException, UnknownFileFormatException {\r
 442                 final Map<String, Score> seqs = new HashMap<String, Score>();\r
 443 \r
 444                 final BufferedReader infasta = new BufferedReader(\r
 445                                 new InputStreamReader(inStream, "UTF8"), 16000);\r
 446 \r
 447                 String line;\r
 448                 String sname = "";\r
 449                 do {\r
 450                         line = infasta.readLine();\r
 451                         if (line == null || line.isEmpty()) {\r
 452                                 // skip empty lines\r
 453                                 continue;\r
 454                         }\r
 455                         if (line.startsWith(">")) {\r
 456                                 // read name\r
 457                                 sname = line.trim().substring(1);\r
 458                                 // read sequence line\r
 459                                 line = infasta.readLine();\r
 460                                 final String sequence = line.replace("\t", "");\r
 461                                 // read annotation line\r
 462                                 line = infasta.readLine();\r
 463                                 String[] annotValues = line.split("\t");\r
 464                                 float[] annotation = convertToNumber(annotValues);\r
 465                                 if (annotation.length != sequence.length()) {\r
 466                                         throw new UnknownFileFormatException(\r
 467                                                         "File does not look like Jronn horizontally formatted output file!\n"\r
 468                                                                         + JRONN_WRONG_FORMAT_MESSAGE);\r
 469                                 }\r
 470                                 seqs.put(sname, new Score(DisorderMethod.JRonn, annotation));\r
 471                         }\r
 472                 } while (line != null);\r
 473 \r
 474                 infasta.close();\r
 475                 return seqs;\r
 476         }\r
 477 \r
 478         private static float[] convertToNumber(String[] annotValues)\r
 479                         throws UnknownFileFormatException {\r
 480                 float[] annotation = new float[annotValues.length];\r
 481                 try {\r
 482                         for (int i = 0; i < annotation.length; i++) {\r
 483                                 annotation[i] = Float.parseFloat(annotValues[i]);\r
 484                         }\r
 485                 } catch (NumberFormatException e) {\r
 486                         throw new UnknownFileFormatException(JRONN_WRONG_FORMAT_MESSAGE,\r
 487                                         e.getCause());\r
 488                 }\r
 489                 return annotation;\r
 490         }\r
 491 \r
 492         private static final String JRONN_WRONG_FORMAT_MESSAGE = "Jronn file must be in the following format:\n"\r
 493                         + ">sequence_name\n "\r
 494                         + "M    V       S\n"\r
 495                         + "0.43 0.22    0.65\n"\r
 496                         + "Where first line is the sequence name,\n"\r
 497                         + "second line is the tab delimited sequence,\n"\r
 498                         + "third line contains tab delimited disorder prediction values.\n"\r
 499                         + "No lines are allowed between these three. Additionally, the number of  "\r
 500                         + "sequence residues must be equal to the number of the disorder values.";\r
 501 \r
 502         /**\r
 503          * Closes the Closable and logs the exception if any\r
 504          * \r
 505          * @param log\r
 506          * @param stream\r
 507          */\r
 508         public final static void closeSilently(java.util.logging.Logger log,\r
 509                         Closeable stream) {\r
 510                 if (stream != null) {\r
 511                         try {\r
 512                                 stream.close();\r
 513                         } catch (IOException e) {\r
 514                                 log.log(Level.WARNING, e.getLocalizedMessage(), e.getCause());\r
 515                         }\r
 516                 }\r
 517         }\r
 518 \r
 519         /**\r
 520          * \r
 521          > Foobar_dundeefriends\r
 522          * \r
 523          * # COILS 34-41, 50-58, 83-91, 118-127, 160-169, 191-220, 243-252, 287-343\r
 524          * \r
 525          * # REM465 355-368\r
 526          * \r
 527          * # HOTLOOPS 190-204\r
 528          * \r
 529          * # RESIDUE COILS REM465 HOTLOOPS\r
 530          * \r
 531          * M 0.86010 0.88512 0.37094\r
 532          * \r
 533          * T 0.79983 0.85864 0.44331\r
 534          * \r
 535          * >Next Sequence name\r
 536          * \r
 537          * \r
 538          * @param input\r
 539          *            the InputStream\r
 540          * @return Map key=sequence name, value=set of score\r
 541          * @throws IOException\r
 542          * @throws UnknownFileFormatException\r
 543          */\r
 544         public static HashMap<String, Set<Score>> readDisembl(\r
 545                         final InputStream input) throws IOException,\r
 546                         UnknownFileFormatException {\r
 547                 Scanner scan = new Scanner(input);\r
 548                 scan.useDelimiter(">");\r
 549                 if (!scan.hasNext()) {\r
 550                         throw new UnknownFileFormatException(\r
 551                                         "In Disembl score format each sequence score is expected "\r
 552                                                         + "to start from the line: >Sequence name "\r
 553                                                         + " No such line was found!");\r
 554                 }\r
 555 \r
 556                 HashMap<String, Set<Score>> results = new HashMap<String, Set<Score>>();\r
 557                 int seqCounter = 0;\r
 558                 while (scan.hasNext()) {\r
 559                         seqCounter++;\r
 560                         String singleSeq = scan.next();\r
 561                         Scanner scansingle = new Scanner(singleSeq);\r
 562                         if (!scansingle.hasNextLine()) {\r
 563                                 throw new RuntimeException(\r
 564                                                 "The input looks like an incomplete disembl file - cannot parse!");\r
 565                         }\r
 566 \r
 567                         StringBuffer seqbuffer = new StringBuffer();\r
 568                         ArrayList<Float> coils = new ArrayList<Float>();\r
 569                         ArrayList<Float> rem = new ArrayList<Float>();\r
 570                         ArrayList<Float> hotloops = new ArrayList<Float>();\r
 571 \r
 572                         String sequenceName = scansingle.nextLine().trim();\r
 573                         TreeSet<Range> coilsR = parseRanges(DisemblResult.COILS,\r
 574                                         scansingle.nextLine());\r
 575                         TreeSet<Range> rem465R = parseRanges(DisemblResult.REM465,\r
 576                                         scansingle.nextLine());\r
 577                         TreeSet<Range> loopsR = parseRanges(DisemblResult.HOTLOOPS,\r
 578                                         scansingle.nextLine());\r
 579 \r
 580                         String title = scansingle.nextLine();\r
 581                         assert title.startsWith("# RESIDUE COILS REM465 HOTLOOPS") : ">Sequence_name must follow column title: # RESIDUE COILS REM465 HOTLOOPS!";\r
 582 \r
 583                         while (scansingle.hasNext()) {\r
 584                                 seqbuffer.append(scansingle.next());\r
 585                                 coils.add(scansingle.nextFloat());\r
 586                                 rem.add(scansingle.nextFloat());\r
 587                                 hotloops.add(scansingle.nextFloat());\r
 588                         }\r
 589                         /*\r
 590                          * Also possible FastaSequence fs = new FastaSequence(sequenceName,\r
 591                          * seqbuffer.toString());\r
 592                          */\r
 593                         HashSet<Score> scores = new HashSet<Score>();\r
 594                         scores.add(new Score(DisemblResult.COILS, coils, coilsR));\r
 595                         scores.add(new Score(DisemblResult.REM465, rem, rem465R));\r
 596                         scores.add(new Score(DisemblResult.HOTLOOPS, hotloops, loopsR));\r
 597                         results.put(sequenceName, scores);\r
 598 \r
 599                         scansingle.close();\r
 600                 }\r
 601                 scan.close();\r
 602                 input.close();\r
 603                 return results;\r
 604         }\r
 605 \r
 606         /**\r
 607          * Parsing:\r
 608          * \r
 609          * # COILS 34-41, 50-58, 83-91, 118-127, 160-169, 191-220, 243-252, 287-343,\r
 610          * 350-391, 429-485, 497-506, 539-547\r
 611          * \r
 612          * # REM465 355-368\r
 613          * \r
 614          * # HOTLOOPS 190-204\r
 615          * \r
 616          * @param lines\r
 617          * @return\r
 618          */\r
 619         private static TreeSet<Range> parseRanges(Enum resultType, String lines) {\r
 620                 TreeSet<Range> ranges = new TreeSet<Range>();\r
 621 \r
 622                 Scanner scan = new Scanner(lines);\r
 623 \r
 624                 assert scan.hasNext();\r
 625                 String del = scan.next();\r
 626                 assert "#".equals(del); // pass delimiter #\r
 627                 String type = scan.next(); // pass enum name e.g. COILS\r
 628                 assert resultType.toString().equalsIgnoreCase(type) : "Unknown result type: "\r
 629                                 + resultType.toString();\r
 630 \r
 631                 // beginning of the ranges\r
 632                 scan.useDelimiter(",");\r
 633                 while (scan.hasNext()) {\r
 634                         String range = scan.next();\r
 635                         if (!Util.isEmpty(range)) {\r
 636                                 ranges.add(new Range(range.split("-")));\r
 637                         }\r
 638                 }\r
 639                 return ranges;\r
 640         }\r
 641 \r
 642         /**\r
 643          * \r
 644          > Foobar_dundeefriends\r
 645          * \r
 646          * # COILS 34-41, 50-58, 83-91, 118-127, 160-169, 191-220, 243-252, 287-343\r
 647          * \r
 648          * # REM465 355-368\r
 649          * \r
 650          * # HOTLOOPS 190-204\r
 651          * \r
 652          * # RESIDUE COILS REM465 HOTLOOPS\r
 653          * \r
 654          * M 0.86010 0.88512 0.37094\r
 655          * \r
 656          * T 0.79983 0.85864 0.44331\r
 657          * \r
 658          * >Next Sequence name\r
 659          * \r
 660          * \r
 661          * @param input\r
 662          * @return Map key=sequence name, value=set of score\r
 663          * @throws IOException\r
 664          * @throws UnknownFileFormatException\r
 665          */\r
 666         public static HashMap<String, Set<Score>> readGlobPlot(\r
 667                         final InputStream input) throws IOException,\r
 668                         UnknownFileFormatException {\r
 669                 Scanner scan = new Scanner(input);\r
 670                 scan.useDelimiter(">");\r
 671                 if (!scan.hasNext()) {\r
 672                         throw new UnknownFileFormatException(\r
 673                                         "In GlobPlot score format each sequence score is expected "\r
 674                                                         + "to start from the line: >Sequence name "\r
 675                                                         + " No such line was found!");\r
 676                 }\r
 677 \r
 678                 HashMap<String, Set<Score>> results = new HashMap<String, Set<Score>>();\r
 679                 int seqCounter = 0;\r
 680                 while (scan.hasNext()) {\r
 681                         seqCounter++;\r
 682                         String singleSeq = scan.next();\r
 683                         Scanner scansingle = new Scanner(singleSeq);\r
 684                         if (!scansingle.hasNextLine()) {\r
 685                                 throw new RuntimeException(\r
 686                                                 "The input looks like an incomplete GlobPlot file - cannot parse!");\r
 687                         }\r
 688 \r
 689                         StringBuffer seqbuffer = new StringBuffer();\r
 690                         ArrayList<Float> dydxScore = new ArrayList<Float>();\r
 691                         ArrayList<Float> rawScore = new ArrayList<Float>();\r
 692                         ArrayList<Float> smoothedScore = new ArrayList<Float>();\r
 693 \r
 694                         String sequenceName = scansingle.nextLine().trim();\r
 695                         TreeSet<Range> domsR = parseRanges(GlobProtResult.GlobDoms,\r
 696                                         scansingle.nextLine());\r
 697                         TreeSet<Range> disorderR = parseRanges(GlobProtResult.Disorder,\r
 698                                         scansingle.nextLine());\r
 699 \r
 700                         String title = scansingle.nextLine();\r
 701                         assert title.startsWith("# RESIDUE      DYDX") : ">Sequence_name must follow column title: # RESIDUE DYDX RAW SMOOTHED!";\r
 702 \r
 703                         while (scansingle.hasNext()) {\r
 704                                 seqbuffer.append(scansingle.next());\r
 705                                 dydxScore.add(scansingle.nextFloat());\r
 706                                 rawScore.add(scansingle.nextFloat());\r
 707                                 smoothedScore.add(scansingle.nextFloat());\r
 708                         }\r
 709                         /*\r
 710                          * Also possible FastaSequence fs = new FastaSequence(sequenceName,\r
 711                          * seqbuffer.toString());\r
 712                          */\r
 713                         Set<Score> scores = new TreeSet<Score>();\r
 714                         scores.add(new Score(GlobProtResult.Disorder, disorderR));\r
 715                         scores.add(new Score(GlobProtResult.GlobDoms, domsR));\r
 716                         scores.add(new Score(GlobProtResult.Dydx, dydxScore));\r
 717                         scores.add(new Score(GlobProtResult.RawScore, rawScore));\r
 718                         scores.add(new Score(GlobProtResult.SmoothedScore, smoothedScore));\r
 719                         results.put(sequenceName, scores);\r
 720 \r
 721                         scansingle.close();\r
 722                 }\r
 723                 scan.close();\r
 724                 input.close();\r
 725                 return results;\r
 726         }\r
 727         /**\r
 728          * Read AACon result with no alignment files. This method leaves incoming\r
 729          * InputStream open!\r
 730          * \r
 731          * @param results\r
 732          *            output file of AAConservation\r
 733          * @return Map with keys {@link ConservationMethod} -> float[]\r
 734          */\r
 735         public static HashSet<Score> readAAConResults(InputStream results) {\r
 736                 if (results == null) {\r
 737                         throw new NullPointerException(\r
 738                                         "InputStream with results must be provided");\r
 739                 }\r
 740                 HashSet<Score> annotations = new HashSet<Score>();\r
 741                 Scanner sc = new Scanner(results);\r
 742                 sc.useDelimiter("#");\r
 743                 while (sc.hasNext()) {\r
 744                         String line = sc.next();\r
 745                         int spacePos = line.indexOf(" ");\r
 746                         assert spacePos > 0 : "Space is expected as delimited between method "\r
 747                                         + "name and values!";\r
 748                         String methodLine = line.substring(0, spacePos);\r
 749                         ConservationMethod method = ConservationMethod\r
 750                                         .getMethod(methodLine);\r
 751                         assert method != null : "Method " + methodLine\r
 752                                         + " is not recognized! ";\r
 753                         Scanner valuesScanner = new Scanner(line.substring(spacePos));\r
 754                         ArrayList<Float> values = new ArrayList<Float>();\r
 755                         while (valuesScanner.hasNextDouble()) {\r
 756                                 Double value = valuesScanner.nextDouble();\r
 757                                 values.add(value.floatValue());\r
 758                         }\r
 759                         annotations.add(new Score(method, values));\r
 760                 }\r
 761                 return annotations;\r
 762         }\r
 763 \r
 764         \r
 765         \r
 766 \r
 767         /**\r
 768          * Reads and parses Fasta or Clustal formatted file into a list of\r
 769          * FastaSequence objects\r
 770          * \r
 771          * @param inFilePath\r
 772          *            the path to the input file\r
 773          * @throws IOException\r
 774          *             if the file denoted by inFilePath cannot be read\r
 775          * @throws UnknownFileFormatException\r
 776          *             if the inFilePath points to the file which format cannot be\r
 777          *             recognised\r
 778          * @return the List of FastaSequence objects\r
 779          * \r
 780          */\r
 781         public static List<FastaSequence> openInputStream(String inFilePath)\r
 782                         throws IOException, UnknownFileFormatException {\r
 783 \r
 784                 // This stream gets closed in isValidClustalFile method\r
 785                 InputStream inStrForValidation = new FileInputStream(inFilePath);\r
 786                 // This stream is closed in the calling methods\r
 787                 InputStream inStr = new FileInputStream(inFilePath);\r
 788                 List<FastaSequence> fastaSeqs = null;\r
 789                 if (ClustalAlignmentUtil.isValidClustalFile(inStrForValidation)) {\r
 790                         Alignment al = ClustalAlignmentUtil.readClustalFile(inStr);\r
 791                         // alignment cannot be null see\r
 792                         // ClustalAlignmentUtil.readClustalFile(inStr);\r
 793                         fastaSeqs = al.getSequences();\r
 794                 } else {\r
 795                         fastaSeqs = SequenceUtil.readFasta(inStr);\r
 796                 }\r
 797                 return fastaSeqs;\r
 798         }\r
 799 \r
 800         // This can't possibly be right for all cases!\r
 801         // but it will do for now\r
 802         \r
 803         // As for the metadata. This function doesnt know what program\r
 804         // generated it. How to handle the metadata!?\r
 805         \r
 806         public static void writeClustal(OutputStream outStream,\r
 807                         List<FastaSequence> sequences, char gapChar) \r
 808                         throws IOException {\r
 809                 \r
 810                 BufferedWriter writer = new BufferedWriter(\r
 811                                 new OutputStreamWriter(outStream));\r
 812                 // will give AlignmentMetadata default type of CLUSTAL for now\r
 813                 AlignmentMetadata al = new AlignmentMetadata(Program.CLUSTAL, gapChar);\r
 814                 \r
 815                 ClustalAlignmentUtil.writeClustalAlignment(writer, \r
 816                                 new Alignment(sequences, al));\r
 817                 \r
 818         }\r
 819 \r
 820 }\r
 821 \r
 822 enum DisemblResult {\r
 823         /** These contains ranges and scores */\r
 824         COILS, REM465, HOTLOOPS\r
 825 }\r
 826 enum GlobProtResult {\r
 827         /** This a range with no scores */\r
 828         GlobDoms,\r
 829         /** This a range with no scores */\r
 830         Disorder,\r
 831         /** This a score with no range */\r
 832         Dydx,\r
 833         /** This a score with no range */\r
 834         SmoothedScore,\r
 835         /** This a score with no range */\r
 836         RawScore\r
 837 }\r
 838 \r
 839 enum IUPredResult {\r
 840         /**\r
 841          * Short disorder\r
 842          */\r
 843         Short,\r
 844         /**\r
 845          * Long disorder\r
 846          */\r
 847         Long,\r
 848         /**\r
 849          * Globular domains\r
 850          */\r
 851         Glob;\r
 852 \r
 853         static IUPredResult getType(File file) {\r
 854                 assert file != null;\r
 855                 String name = file.getName();\r
 856                 if (name.endsWith(Long.toString().toLowerCase())) {\r
 857                         return Long;\r
 858                 }\r
 859                 if (name.endsWith(Short.toString().toLowerCase())) {\r
 860                         return Short;\r
 861                 }\r
 862                 if (name.endsWith(Glob.toString().toLowerCase())) {\r
 863                         return Glob;\r
 864                 }\r
 865                 throw new AssertionError(\r
 866                                 "IUPred result file type cannot be recognised! "\r
 867                                                 + "\nFile must ends with one of [glob, long or short]"\r
 868                                                 + "\n but given file name was: " + file.getName());\r
 869         }\r
 870 }\r