datamodel/compbio/data/sequence/SequenceUtil.java

   1 /*\r
   2  * @(#)SequenceUtil.java 1.0 September 2009 Copyright (c) 2009 Peter Troshin\r
   3  * Jalview Web Services version: 2.0 This library is free software; you can\r
   4  * redistribute it and/or modify it under the terms of the Apache License\r
   5  * version 2 as published by the Apache Software Foundation This library is\r
   6  * distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY;\r
   7  * without even the implied warranty of MERCHANTABILITY or FITNESS FOR A\r
   8  * PARTICULAR PURPOSE. See the Apache License for more details. A copy of the\r
   9  * license is in apache_license.txt. It is also available here: see:\r
  10  * http://www.apache.org/licenses/LICENSE-2.0.txt Any republication or derived\r
  11  * work distributed in source code form must include this copyright and license\r
  12  * notice.\r
  13  */\r
  14 \r
  15 package compbio.data.sequence;\r
  16 \r
  17 import java.io.BufferedReader;\r
  18 import java.io.BufferedWriter;\r
  19 import java.io.Closeable;\r
  20 import java.io.File;\r
  21 import java.io.FileInputStream;\r
  22 import java.io.IOException;\r
  23 import java.io.InputStream;\r
  24 import java.io.InputStreamReader;\r
  25 import java.io.OutputStream;\r
  26 import java.io.OutputStreamWriter;\r
  27 import java.util.ArrayList;\r
  28 import java.util.HashMap;\r
  29 import java.util.HashSet;\r
  30 import java.util.List;\r
  31 import java.util.Map;\r
  32 import java.util.Scanner;\r
  33 import java.util.Set;\r
  34 import java.util.TreeSet;\r
  35 import java.util.logging.Level;\r
  36 import java.util.regex.Matcher;\r
  37 import java.util.regex.Pattern;\r
  38 \r
  39 import compbio.util.Util;\r
  40 \r
  41 /**\r
  42  * Utility class for operations on sequences\r
  43  * \r
  44  * @author Petr Troshin\r
  45  * @version 1.0\r
  46  */\r
  47 public final class SequenceUtil {\r
  48 \r
  49         /**\r
  50          * A whitespace character: [\t\n\x0B\f\r]\r
  51          */\r
  52         public static final Pattern WHITE_SPACE = Pattern.compile("\\s");\r
  53 \r
  54         /**\r
  55          * A digit\r
  56          */\r
  57         public static final Pattern DIGIT = Pattern.compile("\\d");\r
  58 \r
  59         /**\r
  60          * Non word\r
  61          */\r
  62         public static final Pattern NONWORD = Pattern.compile("\\W");\r
  63 \r
  64         /**\r
  65          * Valid Amino acids\r
  66          */\r
  67         public static final Pattern AA = Pattern.compile("[ARNDCQEGHILKMFPSTWYV]+",\r
  68                         Pattern.CASE_INSENSITIVE);\r
  69 \r
  70         /**\r
  71          * inversion of AA pattern\r
  72          */\r
  73         public static final Pattern NON_AA = Pattern.compile(\r
  74                         "[^ARNDCQEGHILKMFPSTWYV]+", Pattern.CASE_INSENSITIVE);\r
  75 \r
  76         /**\r
  77          * Same as AA pattern but with two additional letters - XU\r
  78          */\r
  79         public static final Pattern AMBIGUOUS_AA = Pattern.compile(\r
  80                         "[ARNDCQEGHILKMFPSTWYVXU]+", Pattern.CASE_INSENSITIVE);\r
  81 \r
  82         /**\r
  83          * Nucleotides a, t, g, c, u\r
  84          */\r
  85         public static final Pattern NUCLEOTIDE = Pattern.compile("[AGTCU]+",\r
  86                         Pattern.CASE_INSENSITIVE);\r
  87 \r
  88         /**\r
  89          * Ambiguous nucleotide\r
  90          */\r
  91         public static final Pattern AMBIGUOUS_NUCLEOTIDE = Pattern.compile(\r
  92                         "[AGTCRYMKSWHBVDNU]+", Pattern.CASE_INSENSITIVE); // see IUPAC\r
  93         /**\r
  94          * Non nucleotide\r
  95          */\r
  96         public static final Pattern NON_NUCLEOTIDE = Pattern.compile("[^AGTCU]+",\r
  97                         Pattern.CASE_INSENSITIVE);\r
  98 \r
  99         private SequenceUtil() {\r
 100         } // utility class, no instantiation\r
 101 \r
 102         /*\r
 103          * public static void write_PirSeq(OutputStream os, FastaSequence seq)\r
 104          * throws IOException { BufferedWriter pir_out = new BufferedWriter(new\r
 105          * OutputStreamWriter(os)); pir_out.write(">P1;" + seq.getId() +\r
 106          * SysPrefs.newlinechar); pir_out.write(seq.getSequence() +\r
 107          * SysPrefs.newlinechar); pir_out.close(); } public static void\r
 108          * write_FastaSeq(OutputStream os, FastaSequence seq) throws IOException {\r
 109          * BufferedWriter fasta_out = new BufferedWriter( new\r
 110          * OutputStreamWriter(os)); fasta_out.write(">" + seq.getId() +\r
 111          * SysPrefs.newlinechar); fasta_out.write(seq.getSequence() +\r
 112          * SysPrefs.newlinechar); fasta_out.close(); }\r
 113          */\r
 114 \r
 115         /**\r
 116          * @return true is the sequence contains only letters a,c, t, g, u\r
 117          */\r
 118         public static boolean isNucleotideSequence(final FastaSequence s) {\r
 119                 return SequenceUtil.isNonAmbNucleotideSequence(s.getSequence());\r
 120         }\r
 121 \r
 122         /**\r
 123          * Ambiguous DNA chars : AGTCRYMKSWHBVDN // differs from protein in only one\r
 124          * (!) - B char\r
 125          */\r
 126         public static boolean isNonAmbNucleotideSequence(String sequence) {\r
 127                 sequence = SequenceUtil.cleanSequence(sequence);\r
 128                 if (SequenceUtil.DIGIT.matcher(sequence).find()) {\r
 129                         return false;\r
 130                 }\r
 131                 if (SequenceUtil.NON_NUCLEOTIDE.matcher(sequence).find()) {\r
 132                         return false;\r
 133                         /*\r
 134                          * System.out.format("I found the text starting at " +\r
 135                          * "index %d and ending at index %d.%n", nonDNAmatcher .start(),\r
 136                          * nonDNAmatcher.end());\r
 137                          */\r
 138                 }\r
 139                 final Matcher DNAmatcher = SequenceUtil.NUCLEOTIDE.matcher(sequence);\r
 140                 return DNAmatcher.find();\r
 141         }\r
 142 \r
 143         /**\r
 144          * Removes all whitespace chars in the sequence string\r
 145          * \r
 146          * @param sequence\r
 147          * @return cleaned up sequence\r
 148          */\r
 149         public static String cleanSequence(String sequence) {\r
 150                 assert sequence != null;\r
 151                 final Matcher m = SequenceUtil.WHITE_SPACE.matcher(sequence);\r
 152                 sequence = m.replaceAll("").toUpperCase();\r
 153                 return sequence;\r
 154         }\r
 155 \r
 156         /**\r
 157          * Removes all special characters and digits as well as whitespace chars\r
 158          * from the sequence\r
 159          * \r
 160          * @param sequence\r
 161          * @return cleaned up sequence\r
 162          */\r
 163         public static String deepCleanSequence(String sequence) {\r
 164                 sequence = SequenceUtil.cleanSequence(sequence);\r
 165                 sequence = SequenceUtil.DIGIT.matcher(sequence).replaceAll("");\r
 166                 sequence = SequenceUtil.NONWORD.matcher(sequence).replaceAll("");\r
 167                 final Pattern othernonSeqChars = Pattern.compile("[_-]+");\r
 168                 sequence = othernonSeqChars.matcher(sequence).replaceAll("");\r
 169                 return sequence;\r
 170         }\r
 171 \r
 172         /**\r
 173          * @param sequence\r
 174          * @return true is the sequence is a protein sequence, false overwise\r
 175          */\r
 176         public static boolean isProteinSequence(String sequence) {\r
 177                 sequence = SequenceUtil.cleanSequence(sequence);\r
 178                 if (SequenceUtil.isNonAmbNucleotideSequence(sequence)) {\r
 179                         return false;\r
 180                 }\r
 181                 if (SequenceUtil.DIGIT.matcher(sequence).find()) {\r
 182                         return false;\r
 183                 }\r
 184                 if (SequenceUtil.NON_AA.matcher(sequence).find()) {\r
 185                         return false;\r
 186                 }\r
 187                 final Matcher protmatcher = SequenceUtil.AA.matcher(sequence);\r
 188                 return protmatcher.find();\r
 189         }\r
 190 \r
 191         /**\r
 192          * Check whether the sequence confirms to amboguous protein sequence\r
 193          * \r
 194          * @param sequence\r
 195          * @return return true only if the sequence if ambiguous protein sequence\r
 196          *         Return false otherwise. e.g. if the sequence is non-ambiguous\r
 197          *         protein or DNA\r
 198          */\r
 199         public static boolean isAmbiguosProtein(String sequence) {\r
 200                 sequence = SequenceUtil.cleanSequence(sequence);\r
 201                 if (SequenceUtil.isNonAmbNucleotideSequence(sequence)) {\r
 202                         return false;\r
 203                 }\r
 204                 if (SequenceUtil.DIGIT.matcher(sequence).find()) {\r
 205                         return false;\r
 206                 }\r
 207                 if (SequenceUtil.NON_AA.matcher(sequence).find()) {\r
 208                         return false;\r
 209                 }\r
 210                 if (SequenceUtil.AA.matcher(sequence).find()) {\r
 211                         return false;\r
 212                 }\r
 213                 final Matcher amb_prot = SequenceUtil.AMBIGUOUS_AA.matcher(sequence);\r
 214                 return amb_prot.find();\r
 215         }\r
 216 \r
 217         /**\r
 218          * Writes list of FastaSequeces into the outstream formatting the sequence\r
 219          * so that it contains width chars on each line\r
 220          * \r
 221          * @param outstream\r
 222          * @param sequences\r
 223          * @param width\r
 224          *            - the maximum number of characters to write in one line\r
 225          * @throws IOException\r
 226          */\r
 227         public static void writeFasta(final OutputStream outstream,\r
 228                         final List<FastaSequence> sequences, final int width)\r
 229                         throws IOException {\r
 230                 writeFastaKeepTheStream(outstream, sequences, width);\r
 231                 outstream.close();\r
 232         }\r
 233 \r
 234         public static void writeFastaKeepTheStream(final OutputStream outstream,\r
 235                         final List<FastaSequence> sequences, final int width)\r
 236                         throws IOException {\r
 237                 final OutputStreamWriter writer = new OutputStreamWriter(outstream);\r
 238                 final BufferedWriter fastawriter = new BufferedWriter(writer);\r
 239                 for (final FastaSequence fs : sequences) {\r
 240                         fastawriter.write(">" + fs.getId() + "\n");\r
 241                         fastawriter.write(fs.getFormatedSequence(width));\r
 242                         fastawriter.write("\n");\r
 243                 }\r
 244                 fastawriter.flush();\r
 245                 writer.flush();\r
 246         }\r
 247 \r
 248         /**\r
 249          * Reads fasta sequences from inStream into the list of FastaSequence\r
 250          * objects\r
 251          * \r
 252          * @param inStream\r
 253          *            from\r
 254          * @return list of FastaSequence objects\r
 255          * @throws IOException\r
 256          */\r
 257         public static List<FastaSequence> readFasta(final InputStream inStream)\r
 258                         throws IOException {\r
 259                 final List<FastaSequence> seqs = new ArrayList<FastaSequence>();\r
 260 \r
 261                 final BufferedReader infasta = new BufferedReader(\r
 262                                 new InputStreamReader(inStream, "UTF8"), 16000);\r
 263                 final Pattern pattern = Pattern.compile("//s+");\r
 264 \r
 265                 String line;\r
 266                 String sname = "", seqstr = null;\r
 267                 do {\r
 268                         line = infasta.readLine();\r
 269                         if ((line == null) || line.startsWith(">")) {\r
 270                                 if (seqstr != null) {\r
 271                                         seqs.add(new FastaSequence(sname.substring(1), seqstr));\r
 272                                 }\r
 273                                 sname = line; // remove >\r
 274                                 seqstr = "";\r
 275                         } else {\r
 276                                 final String subseq = pattern.matcher(line).replaceAll("");\r
 277                                 seqstr += subseq;\r
 278                         }\r
 279                 } while (line != null);\r
 280 \r
 281                 infasta.close();\r
 282                 return seqs;\r
 283         }\r
 284 \r
 285         /**\r
 286          * Writes FastaSequence in the file, each sequence will take one line only\r
 287          * \r
 288          * @param os\r
 289          * @param sequences\r
 290          * @throws IOException\r
 291          */\r
 292         public static void writeFasta(final OutputStream os,\r
 293                         final List<FastaSequence> sequences) throws IOException {\r
 294                 final OutputStreamWriter outWriter = new OutputStreamWriter(os);\r
 295                 final BufferedWriter fasta_out = new BufferedWriter(outWriter);\r
 296                 for (final FastaSequence fs : sequences) {\r
 297                         fasta_out.write(fs.getOnelineFasta());\r
 298                 }\r
 299                 fasta_out.close();\r
 300                 outWriter.close();\r
 301         }\r
 302 \r
 303         public static Map<String, Score> readJRonn(final File result)\r
 304                         throws IOException, UnknownFileFormatException {\r
 305                 InputStream input = new FileInputStream(result);\r
 306                 Map<String, Score> sequences = readJRonn(input);\r
 307                 input.close();\r
 308                 return sequences;\r
 309         }\r
 310 \r
 311         /**\r
 312          * Reader for JRonn horizontal file format\r
 313          * \r
 314          * <pre>\r
 315          * &gtFoobar M G D T T A G 0.48 0.42\r
 316          * 0.42 0.48 0.52 0.53 0.54\r
 317          * \r
 318          * <pre>\r
 319          * Where all values are tab delimited\r
 320          * \r
 321          * @param inStream\r
 322          *            the InputStream connected to the JRonn output file\r
 323          * @return List of {@link AnnotatedSequence} objects\r
 324          * @throws IOException\r
 325          *             is thrown if the inStream has problems accessing the data\r
 326          * @throws UnknownFileFormatException\r
 327          *             is thrown if the inStream represents an unknown source of\r
 328          * data, i.e. not a JRonn output\r
 329          */\r
 330         public static Map<String, Score> readJRonn(final InputStream inStream)\r
 331                         throws IOException, UnknownFileFormatException {\r
 332                 final Map<String, Score> seqs = new HashMap<String, Score>();\r
 333 \r
 334                 final BufferedReader infasta = new BufferedReader(\r
 335                                 new InputStreamReader(inStream, "UTF8"), 16000);\r
 336 \r
 337                 String line;\r
 338                 String sname = "";\r
 339                 do {\r
 340                         line = infasta.readLine();\r
 341                         if (line == null || line.isEmpty()) {\r
 342                                 // skip empty lines\r
 343                                 continue;\r
 344                         }\r
 345                         if (line.startsWith(">")) {\r
 346                                 // read name\r
 347                                 sname = line.trim().substring(1);\r
 348                                 // read sequence line\r
 349                                 line = infasta.readLine();\r
 350                                 final String sequence = line.replace("\t", "");\r
 351                                 // read annotation line\r
 352                                 line = infasta.readLine();\r
 353                                 String[] annotValues = line.split("\t");\r
 354                                 float[] annotation = convertToNumber(annotValues);\r
 355                                 if (annotation.length != sequence.length()) {\r
 356                                         throw new UnknownFileFormatException(\r
 357                                                         "File does not look like Jronn horizontally formatted output file!\n"\r
 358                                                                         + JRONN_WRONG_FORMAT_MESSAGE);\r
 359                                 }\r
 360                                 seqs.put(sname, new Score(DisorderMethod.JRonn, annotation));\r
 361                         }\r
 362                 } while (line != null);\r
 363 \r
 364                 infasta.close();\r
 365                 return seqs;\r
 366         }\r
 367 \r
 368         private static float[] convertToNumber(String[] annotValues)\r
 369                         throws UnknownFileFormatException {\r
 370                 float[] annotation = new float[annotValues.length];\r
 371                 try {\r
 372                         for (int i = 0; i < annotation.length; i++) {\r
 373                                 annotation[i] = Float.parseFloat(annotValues[i]);\r
 374                         }\r
 375                 } catch (NumberFormatException e) {\r
 376                         throw new UnknownFileFormatException(JRONN_WRONG_FORMAT_MESSAGE,\r
 377                                         e.getCause());\r
 378                 }\r
 379                 return annotation;\r
 380         }\r
 381 \r
 382         private static final String JRONN_WRONG_FORMAT_MESSAGE = "Jronn file must be in the following format:\n"\r
 383                         + ">sequence_name\n "\r
 384                         + "M    V       S\n"\r
 385                         + "0.43 0.22    0.65\n"\r
 386                         + "Where first line is the sequence name,\n"\r
 387                         + "second line is the tab delimited sequence,\n"\r
 388                         + "third line contains tab delimited disorder prediction values.\n"\r
 389                         + "No lines are allowed between these three. Additionally, the number of  "\r
 390                         + "sequence residues must be equal to the number of the disorder values.";\r
 391 \r
 392         /**\r
 393          * Closes the Closable and logs the exception if any\r
 394          * \r
 395          * @param log\r
 396          * @param stream\r
 397          */\r
 398         public final static void closeSilently(java.util.logging.Logger log,\r
 399                         Closeable stream) {\r
 400                 if (stream != null) {\r
 401                         try {\r
 402                                 stream.close();\r
 403                         } catch (IOException e) {\r
 404                                 log.log(Level.WARNING, e.getLocalizedMessage(), e.getCause());\r
 405                         }\r
 406                 }\r
 407         }\r
 408 \r
 409         /**\r
 410          * \r
 411          > Foobar_dundeefriends\r
 412          * \r
 413          * # COILS 34-41, 50-58, 83-91, 118-127, 160-169, 191-220, 243-252, 287-343\r
 414          * \r
 415          * # REM465 355-368\r
 416          * \r
 417          * # HOTLOOPS 190-204\r
 418          * \r
 419          * # RESIDUE COILS REM465 HOTLOOPS\r
 420          * \r
 421          * M 0.86010 0.88512 0.37094\r
 422          * \r
 423          * T 0.79983 0.85864 0.44331\r
 424          * \r
 425          * >Next Sequence name\r
 426          * \r
 427          * \r
 428          * @param input\r
 429          * @return\r
 430          * @throws IOException\r
 431          * @throws UnknownFileFormatException\r
 432          */\r
 433         public static HashMap<String, Set<Score>> readDisembl(\r
 434                         final InputStream input) throws IOException,\r
 435                         UnknownFileFormatException {\r
 436                 Scanner scan = new Scanner(input);\r
 437                 scan.useDelimiter(">");\r
 438                 if (!scan.hasNext()) {\r
 439                         throw new UnknownFileFormatException(\r
 440                                         "In Disembl score format each sequence score is expected "\r
 441                                                         + "to start from the line: >Sequence name "\r
 442                                                         + " No such line was found!");\r
 443                 }\r
 444 \r
 445                 HashMap<String, Set<Score>> results = new HashMap<String, Set<Score>>();\r
 446                 int seqCounter = 0;\r
 447                 while (scan.hasNext()) {\r
 448                         seqCounter++;\r
 449                         String singleSeq = scan.next();\r
 450                         Scanner scansingle = new Scanner(singleSeq);\r
 451                         if (!scansingle.hasNextLine()) {\r
 452                                 throw new RuntimeException(\r
 453                                                 "The input looks like an incomplete disembl file - cannot parse!");\r
 454                         }\r
 455 \r
 456                         StringBuffer seqbuffer = new StringBuffer();\r
 457                         ArrayList<Float> coils = new ArrayList<Float>();\r
 458                         ArrayList<Float> rem = new ArrayList<Float>();\r
 459                         ArrayList<Float> hotloops = new ArrayList<Float>();\r
 460 \r
 461                         String sequenceName = scansingle.nextLine().trim();\r
 462                         TreeSet<Range> coilsR = parseRanges(DisemblResult.COILS,\r
 463                                         scansingle.nextLine());\r
 464                         TreeSet<Range> rem465R = parseRanges(DisemblResult.REM465,\r
 465                                         scansingle.nextLine());\r
 466                         TreeSet<Range> loopsR = parseRanges(DisemblResult.HOTLOOPS,\r
 467                                         scansingle.nextLine());\r
 468 \r
 469                         String title = scansingle.nextLine();\r
 470                         assert title.startsWith("# RESIDUE COILS REM465 HOTLOOPS") : ">Sequence_name must follow column title: # RESIDUE COILS REM465 HOTLOOPS!";\r
 471 \r
 472                         while (scansingle.hasNext()) {\r
 473                                 seqbuffer.append(scansingle.next());\r
 474                                 coils.add(scansingle.nextFloat());\r
 475                                 rem.add(scansingle.nextFloat());\r
 476                                 hotloops.add(scansingle.nextFloat());\r
 477                         }\r
 478                         /*\r
 479                          * Also possible FastaSequence fs = new FastaSequence(sequenceName,\r
 480                          * seqbuffer.toString());\r
 481                          */\r
 482                         HashSet<Score> scores = new HashSet<Score>();\r
 483                         scores.add(new Score(DisemblResult.COILS, coils, coilsR));\r
 484                         scores.add(new Score(DisemblResult.HOTLOOPS, hotloops, rem465R));\r
 485                         scores.add(new Score(DisemblResult.REM465, rem, loopsR));\r
 486                         results.put(sequenceName, scores);\r
 487 \r
 488                         scansingle.close();\r
 489                 }\r
 490                 scan.close();\r
 491                 input.close();\r
 492                 return results;\r
 493         }\r
 494 \r
 495         /**\r
 496          * Parsing:\r
 497          * \r
 498          * # COILS 34-41, 50-58, 83-91, 118-127, 160-169, 191-220, 243-252, 287-343,\r
 499          * 350-391, 429-485, 497-506, 539-547\r
 500          * \r
 501          * # REM465 355-368\r
 502          * \r
 503          * # HOTLOOPS 190-204\r
 504          * \r
 505          * @param lines\r
 506          * @return\r
 507          */\r
 508         private static TreeSet<Range> parseRanges(Enum resultType, String lines) {\r
 509                 TreeSet<Range> ranges = new TreeSet<Range>();\r
 510 \r
 511                 Scanner scan = new Scanner(lines);\r
 512 \r
 513                 assert scan.hasNext();\r
 514                 String del = scan.next();\r
 515                 assert "#".equals(del); // pass delimiter #\r
 516                 String type = scan.next(); // pass enum name e.g. COILS\r
 517                 assert resultType.toString().equalsIgnoreCase(type) : "Unknown result type: "\r
 518                                 + resultType.toString();\r
 519 \r
 520                 // beginning of the ranges\r
 521                 scan.useDelimiter(",");\r
 522                 while (scan.hasNext()) {\r
 523                         String range = scan.next();\r
 524                         if (!Util.isEmpty(range)) {\r
 525                                 ranges.add(new Range(range.split("-")));\r
 526                         }\r
 527                 }\r
 528                 return ranges;\r
 529         }\r
 530 \r
 531         /**\r
 532          * \r
 533          > Foobar_dundeefriends\r
 534          * \r
 535          * # COILS 34-41, 50-58, 83-91, 118-127, 160-169, 191-220, 243-252, 287-343\r
 536          * \r
 537          * # REM465 355-368\r
 538          * \r
 539          * # HOTLOOPS 190-204\r
 540          * \r
 541          * # RESIDUE COILS REM465 HOTLOOPS\r
 542          * \r
 543          * M 0.86010 0.88512 0.37094\r
 544          * \r
 545          * T 0.79983 0.85864 0.44331\r
 546          * \r
 547          * >Next Sequence name\r
 548          * \r
 549          * \r
 550          * @param input\r
 551          * @return\r
 552          * @throws IOException\r
 553          * @throws UnknownFileFormatException\r
 554          */\r
 555         public static HashMap<String, Set<Score>> readGlobPlot(\r
 556                         final InputStream input) throws IOException,\r
 557                         UnknownFileFormatException {\r
 558                 Scanner scan = new Scanner(input);\r
 559                 scan.useDelimiter(">");\r
 560                 if (!scan.hasNext()) {\r
 561                         throw new UnknownFileFormatException(\r
 562                                         "In GlobPlot score format each sequence score is expected "\r
 563                                                         + "to start from the line: >Sequence name "\r
 564                                                         + " No such line was found!");\r
 565                 }\r
 566 \r
 567                 HashMap<String, Set<Score>> results = new HashMap<String, Set<Score>>();\r
 568                 int seqCounter = 0;\r
 569                 while (scan.hasNext()) {\r
 570                         seqCounter++;\r
 571                         String singleSeq = scan.next();\r
 572                         Scanner scansingle = new Scanner(singleSeq);\r
 573                         if (!scansingle.hasNextLine()) {\r
 574                                 throw new RuntimeException(\r
 575                                                 "The input looks like an incomplete GlobPlot file - cannot parse!");\r
 576                         }\r
 577 \r
 578                         StringBuffer seqbuffer = new StringBuffer();\r
 579                         ArrayList<Float> dydxScore = new ArrayList<Float>();\r
 580                         ArrayList<Float> rawScore = new ArrayList<Float>();\r
 581                         ArrayList<Float> smoothedScore = new ArrayList<Float>();\r
 582 \r
 583                         String sequenceName = scansingle.nextLine().trim();\r
 584                         TreeSet<Range> domsR = parseRanges(GlobProtResult.GlobDoms,\r
 585                                         scansingle.nextLine());\r
 586                         TreeSet<Range> disorderR = parseRanges(GlobProtResult.Disorder,\r
 587                                         scansingle.nextLine());\r
 588 \r
 589                         String title = scansingle.nextLine();\r
 590                         assert title.startsWith("# RESIDUE      DYDX") : ">Sequence_name must follow column title: # RESIDUE DYDX RAW SMOOTHED!";\r
 591 \r
 592                         while (scansingle.hasNext()) {\r
 593                                 seqbuffer.append(scansingle.next());\r
 594                                 dydxScore.add(scansingle.nextFloat());\r
 595                                 rawScore.add(scansingle.nextFloat());\r
 596                                 smoothedScore.add(scansingle.nextFloat());\r
 597                         }\r
 598                         /*\r
 599                          * Also possible FastaSequence fs = new FastaSequence(sequenceName,\r
 600                          * seqbuffer.toString());\r
 601                          */\r
 602                         HashSet<Score> scores = new HashSet<Score>();\r
 603                         scores.add(new Score(GlobProtResult.Disorder, disorderR));\r
 604                         scores.add(new Score(GlobProtResult.GlobDoms, domsR));\r
 605                         scores.add(new Score(GlobProtResult.Dydx, dydxScore));\r
 606                         scores.add(new Score(GlobProtResult.RawScore, rawScore));\r
 607                         scores.add(new Score(GlobProtResult.SmoothedScore, smoothedScore));\r
 608                         results.put(sequenceName, scores);\r
 609 \r
 610                         scansingle.close();\r
 611                 }\r
 612                 scan.close();\r
 613                 input.close();\r
 614                 return results;\r
 615         }\r
 616         /**\r
 617          * Read AACon result with no alignment files. This method leaves incoming\r
 618          * InputStream open!\r
 619          * \r
 620          * @param results\r
 621          *            output file of AAConservation\r
 622          * @return Map with keys {@link ConservationMethod} -> float[]\r
 623          */\r
 624         public static HashSet<Score> readAAConResults(InputStream results) {\r
 625                 if (results == null) {\r
 626                         throw new NullPointerException(\r
 627                                         "InputStream with results must be provided");\r
 628                 }\r
 629                 HashSet<Score> annotations = new HashSet<Score>();\r
 630                 Scanner sc = new Scanner(results);\r
 631                 sc.useDelimiter("#");\r
 632                 while (sc.hasNext()) {\r
 633                         String line = sc.next();\r
 634                         int spacePos = line.indexOf(" ");\r
 635                         assert spacePos > 0 : "Space is expected as delimited between method "\r
 636                                         + "name and values!";\r
 637                         String methodLine = line.substring(0, spacePos);\r
 638                         ConservationMethod method = ConservationMethod\r
 639                                         .getMethod(methodLine);\r
 640                         assert method != null : "Method " + methodLine\r
 641                                         + " is not recognized! ";\r
 642                         Scanner valuesScanner = new Scanner(line.substring(spacePos));\r
 643                         ArrayList<Float> values = new ArrayList<Float>();\r
 644                         while (valuesScanner.hasNextDouble()) {\r
 645                                 Double value = valuesScanner.nextDouble();\r
 646                                 values.add(value.floatValue());\r
 647                         }\r
 648                         annotations.add(new Score(method, values));\r
 649                 }\r
 650                 return annotations;\r
 651         }\r
 652 \r
 653         /**\r
 654          * Reads and parses Fasta or Clustal formatted file into a list of\r
 655          * FastaSequence objects\r
 656          * \r
 657          * @param inFilePath\r
 658          *            the path to the input file\r
 659          * @throws IOException\r
 660          *             if the file denoted by inFilePath cannot be read\r
 661          * @throws UnknownFileFormatException\r
 662          *             if the inFilePath points to the file which format cannot be\r
 663          *             recognised\r
 664          * @return the List of FastaSequence objects\r
 665          * \r
 666          */\r
 667         public static List<FastaSequence> openInputStream(String inFilePath)\r
 668                         throws IOException, UnknownFileFormatException {\r
 669 \r
 670                 // This stream gets closed in isValidClustalFile method\r
 671                 InputStream inStrForValidation = new FileInputStream(inFilePath);\r
 672                 // This stream is closed in the calling methods\r
 673                 InputStream inStr = new FileInputStream(inFilePath);\r
 674                 List<FastaSequence> fastaSeqs = null;\r
 675                 if (ClustalAlignmentUtil.isValidClustalFile(inStrForValidation)) {\r
 676                         Alignment al = ClustalAlignmentUtil.readClustalFile(inStr);\r
 677                         // alignment cannot be null see\r
 678                         // ClustalAlignmentUtil.readClustalFile(inStr);\r
 679                         fastaSeqs = al.getSequences();\r
 680                 } else {\r
 681                         fastaSeqs = SequenceUtil.readFasta(inStr);\r
 682                 }\r
 683                 return fastaSeqs;\r
 684         }\r
 685 \r
 686 }\r
 687 \r
 688 enum DisemblResult {\r
 689         /** These contains ranges and scores */\r
 690         COILS, REM465, HOTLOOPS\r
 691 }\r
 692 enum GlobProtResult {\r
 693         /** This a range with no scores */\r
 694         GlobDoms,\r
 695         /** This a range with no scores */\r
 696         Disorder,\r
 697         /** This a score with no range */\r
 698         Dydx,\r
 699         /** This a score with no range */\r
 700         SmoothedScore,\r
 701         /** This a score with no range */\r
 702         RawScore\r
 703 }