forester/java/src/org/forester/io/parsers/nexus/NexusPhylogeniesParser.java

   1 // $Id:
   2 // FORESTER -- software libraries and applications
   3 // for evolutionary biology research and applications.
   4 //
   5 // Copyright (C) 2008-2009 Christian M. Zmasek
   6 // Copyright (C) 2008-2009 Burnham Institute for Medical Research
   7 // All rights reserved
   8 //
   9 // This library is free software; you can redistribute it and/or
  10 // modify it under the terms of the GNU Lesser General Public
  11 // License as published by the Free Software Foundation; either
  12 // version 2.1 of the License, or (at your option) any later version.
  13 //
  14 // This library is distributed in the hope that it will be useful,
  15 // but WITHOUT ANY WARRANTY; without even the implied warranty of
  16 // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
  17 // Lesser General Public License for more details.
  18 //
  19 // You should have received a copy of the GNU Lesser General Public
  20 // License along with this library; if not, write to the Free Software
  21 // Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA
  22 //
  23 // Contact: phylosoft @ gmail . com
  24 // WWW: https://sites.google.com/site/cmzmasek/home/software/forester
  25
  26 package org.forester.io.parsers.nexus;
  27
  28 import java.io.BufferedReader;
  29 import java.io.FileNotFoundException;
  30 import java.io.IOException;
  31 import java.util.ArrayList;
  32 import java.util.HashMap;
  33 import java.util.List;
  34 import java.util.Map;
  35 import java.util.regex.Matcher;
  36 import java.util.regex.Pattern;
  37
  38 import org.forester.archaeopteryx.AptxConstants;
  39 import org.forester.io.parsers.IteratingPhylogenyParser;
  40 import org.forester.io.parsers.PhylogenyParser;
  41 import org.forester.io.parsers.nhx.NHXFormatException;
  42 import org.forester.io.parsers.nhx.NHXParser;
  43 import org.forester.io.parsers.nhx.NHXParser.TAXONOMY_EXTRACTION;
  44 import org.forester.io.parsers.util.ParserUtils;
  45 import org.forester.io.parsers.util.PhylogenyParserException;
  46 import org.forester.phylogeny.Phylogeny;
  47 import org.forester.phylogeny.PhylogenyNode;
  48 import org.forester.phylogeny.data.Sequence;
  49 import org.forester.phylogeny.iterators.PhylogenyNodeIterator;
  50 import org.forester.sequence.BasicSequence;
  51 import org.forester.sequence.MolecularSequence;
  52 import org.forester.util.ForesterUtil;
  53
  54 public final class NexusPhylogeniesParser implements IteratingPhylogenyParser, PhylogenyParser {
  55
  56     final private static boolean DEBUG                               = false;
  57
  58     final private static String            begin_trees               = NexusConstants.BEGIN_TREES.toLowerCase();
  59     final private static String            end                       = NexusConstants.END.toLowerCase();
  60     final private static String            endblock                  = "endblock";
  61     final private static Pattern           ROOTEDNESS_PATTERN        = Pattern.compile( ".+=\\s*\\[&([R|U])\\].*" );
  62     final private static String            taxlabels                 = NexusConstants.TAXLABELS.toLowerCase();
  63     final private static Pattern           TITLE_PATTERN             = Pattern.compile( "TITLE.?\\s+([^;]+)",
  64                                                                                         Pattern.CASE_INSENSITIVE );
  65     final private static String            translate                 = NexusConstants.TRANSLATE.toLowerCase();
  66     final private static String            data                      = NexusConstants.BEGIN_CHARACTERS.toLowerCase();
  67     final private static String            characters                = NexusConstants.BEGIN_DATA.toLowerCase();
  68     final private static String            tree                      = NexusConstants.TREE.toLowerCase();
  69     final private static Pattern           TREE_NAME_PATTERN         = Pattern.compile( "\\s*.?Tree\\s+(.+?)\\s*=.+",
  70                                                                                         Pattern.CASE_INSENSITIVE );
  71     final private static Pattern           TRANSLATE_PATTERN         = Pattern.compile( "([0-9A-Za-z]+)\\s+(.+)" );
  72     final private static Pattern           ALN_PATTERN               = Pattern.compile( "(.+)\\s+([A-Za-z-_\\*\\?]+)" );
  73     final private static Pattern           DATATYPE_PATTERN          = Pattern.compile( "datatype\\s?.\\s?([a-z]+)" );
  74     //final private static Pattern           LINK_TAXA_PATTERN         = Pattern.compile( "link\\s+taxa\\s?.\\s?([^;]+)",
  75     //                                                                                    Pattern.CASE_INSENSITIVE );
  76     final private static String            utree                     = NexusConstants.UTREE.toLowerCase();
  77     private BufferedReader                 _br;
  78     private boolean                        _ignore_quotes_in_nh_data = AptxConstants.NH_PARSING_IGNORE_QUOTES_DEFAULT;
  79     private boolean                        _in_taxalabels;
  80     private boolean                        _in_translate;
  81     private boolean                        _in_tree;
  82     private boolean                        _in_trees_block;
  83     private boolean                        _in_data_block;
  84     private boolean                        _is_rooted;
  85     private String                         _datatype;
  86     private String                         _name;
  87     private Phylogeny                      _next;
  88     private Object                         _nexus_source;
  89     private StringBuilder                  _nh;
  90     private boolean                        _replace_underscores      = NHXParser.REPLACE_UNDERSCORES_DEFAULT;
  91     private boolean                        _rooted_info_present;
  92     private List<String>                   _taxlabels;
  93     private TAXONOMY_EXTRACTION            _taxonomy_extraction      = TAXONOMY_EXTRACTION.NO;
  94     private String                         _title;
  95     private Map<String, String>            _translate_map;
  96     private StringBuilder                  _translate_sb;
  97     private Map<String, MolecularSequence> _seqs;
  98     private final boolean                  _add_sequences            = true;
  99
 100     @Override
 101     public String getName() {
 102         return "Nexus Phylogenies Parser";
 103     }
 104
 105     @Override
 106     public final boolean hasNext() {
 107         return _next != null;
 108     }
 109
 110     @Override
 111     public final Phylogeny next() throws NHXFormatException, IOException {
 112         final Phylogeny phy = _next;
 113         getNext();
 114         return phy;
 115     }
 116
 117     @Override
 118     public final Phylogeny[] parse() throws IOException {
 119         final List<Phylogeny> l = new ArrayList<Phylogeny>();
 120         while ( hasNext() ) {
 121             l.add( next() );
 122         }
 123         final Phylogeny[] p = new Phylogeny[ l.size() ];
 124         for( int i = 0; i < l.size(); ++i ) {
 125             p[ i ] = l.get( i );
 126         }
 127         reset();
 128         return p;
 129     }
 130
 131     @Override
 132     public final void reset() throws FileNotFoundException, IOException {
 133         _taxlabels = new ArrayList<String>();
 134         _translate_map = new HashMap<String, String>();
 135         _nh = new StringBuilder();
 136         _name = "";
 137         _title = "";
 138         _translate_sb = null;
 139         _next = null;
 140         _in_trees_block = false;
 141         _in_taxalabels = false;
 142         _in_translate = false;
 143         _in_tree = false;
 144         _rooted_info_present = false;
 145         _is_rooted = false;
 146         _seqs = new HashMap<String, MolecularSequence>();
 147         _br = ParserUtils.createReader( _nexus_source );
 148         getNext();
 149     }
 150
 151     public final void setIgnoreQuotes( final boolean ignore_quotes_in_nh_data ) {
 152         _ignore_quotes_in_nh_data = ignore_quotes_in_nh_data;
 153     }
 154
 155     public final void setReplaceUnderscores( final boolean replace_underscores ) {
 156         _replace_underscores = replace_underscores;
 157     }
 158
 159     @Override
 160     public final void setSource( final Object nexus_source ) throws PhylogenyParserException, IOException {
 161         if ( nexus_source == null ) {
 162             throw new PhylogenyParserException( "attempt to parse null object" );
 163         }
 164         _nexus_source = nexus_source;
 165         reset();
 166     }
 167
 168     public final void setTaxonomyExtraction( final TAXONOMY_EXTRACTION taxonomy_extraction ) {
 169         _taxonomy_extraction = taxonomy_extraction;
 170     }
 171
 172     private final void createPhylogeny( final String title,
 173                                         final String name,
 174                                         final StringBuilder nhx,
 175                                         final boolean rooted_info_present,
 176                                         final boolean is_rooted ) throws IOException {
 177         _next = null;
 178         final NHXParser pars = new NHXParser();
 179         pars.setTaxonomyExtraction( _taxonomy_extraction );
 180         pars.setReplaceUnderscores( _replace_underscores );
 181         pars.setIgnoreQuotes( _ignore_quotes_in_nh_data );
 182         if ( rooted_info_present ) {
 183             pars.setGuessRootedness( false );
 184         }
 185         pars.setSource( nhx );
 186         final Phylogeny p = pars.next();
 187         if ( p == null ) {
 188             throw new PhylogenyParserException( "failed to create phylogeny" );
 189         }
 190         String myname = null;
 191         if ( !ForesterUtil.isEmpty( title ) && !ForesterUtil.isEmpty( name ) ) {
 192             myname = title.replace( '_', ' ' ).trim() + " (" + name.trim() + ")";
 193         }
 194         else if ( !ForesterUtil.isEmpty( title ) ) {
 195             myname = title.replace( '_', ' ' ).trim();
 196         }
 197         else if ( !ForesterUtil.isEmpty( name ) ) {
 198             myname = name.trim();
 199         }
 200         if ( !ForesterUtil.isEmpty( myname ) ) {
 201             p.setName( myname );
 202         }
 203         if ( rooted_info_present ) {
 204             p.setRooted( is_rooted );
 205         }
 206         if ( ( _taxlabels.size() > 0 ) || ( _translate_map.size() > 0 ) ) {
 207             final PhylogenyNodeIterator it = p.iteratorExternalForward();
 208             while ( it.hasNext() ) {
 209                 final PhylogenyNode node = it.next();
 210                 if ( ( _translate_map.size() > 0 ) && _translate_map.containsKey( node.getName() ) ) {
 211                     node.setName( _translate_map.get( node.getName() ).replaceAll( "['\"]+", "" ) );
 212                 }
 213                 else if ( _taxlabels.size() > 0 ) {
 214                     int i = -1;
 215                     try {
 216                         i = Integer.parseInt( node.getName() );
 217                     }
 218                     catch ( final NumberFormatException e ) {
 219                         // Ignore.
 220                     }
 221                     if ( i > 0 ) {
 222                         node.setName( _taxlabels.get( i - 1 ).replaceAll( "['\"]+", "" ) );
 223                     }
 224                 }
 225                 if ( !_replace_underscores && ( ( _taxonomy_extraction != TAXONOMY_EXTRACTION.NO ) ) ) {
 226                     ParserUtils.extractTaxonomyDataFromNodeName( node, _taxonomy_extraction );
 227                 }
 228                 else if ( _replace_underscores ) {
 229                     if ( !ForesterUtil.isEmpty( node.getName() ) ) {
 230                         node.setName( node.getName().replace( '_', ' ' ).trim() );
 231                     }
 232                 }
 233                 if ( _add_sequences ) {
 234                     if ( _seqs.containsKey( node.getName() ) ) {
 235                         final MolecularSequence s = _seqs.get( node.getName() );
 236                         //TODO need to check for uniqueness when adding seqs....
 237                         final Sequence ns = new Sequence( s );
 238                         ns.setMolecularSequenceAligned( true ); //TODO need to check if all same length
 239                         node.getNodeData().addSequence( ns );
 240                     }
 241                 }
 242             }
 243         }
 244         _next = p;
 245     }
 246
 247     private final void getNext() throws IOException, NHXFormatException {
 248         _next = null;
 249         String line;
 250         while ( ( line = _br.readLine() ) != null ) {
 251             if ( DEBUG ) {
 252                 System.out.println( line );
 253             }
 254             line = line.trim();
 255             if ( ( line.length() > 0 ) && !line.startsWith( "#" ) && !line.startsWith( ">" ) ) {
 256                 line = ForesterUtil.collapseWhiteSpace( line );
 257                 line = removeWhiteSpaceBeforeSemicolon( line );
 258                 final String line_lc = line.toLowerCase();
 259                 if ( line_lc.startsWith( begin_trees ) ) {
 260                     _in_trees_block = true;
 261                     _in_taxalabels = false;
 262                     _in_translate = false;
 263                     _in_data_block = false;
 264                     _datatype = null;
 265                     _title = "";
 266                 }
 267                 else if ( line_lc.startsWith( taxlabels ) ) {
 268                     //TODO need to be taxa block instead
 269                     _in_trees_block = false;
 270                     _in_taxalabels = true;
 271                     _in_translate = false;
 272                     _in_data_block = false;
 273                     _datatype = null;
 274                 }
 275                 else if ( line_lc.startsWith( translate ) ) {
 276                     _translate_sb = new StringBuilder();
 277                     _in_taxalabels = false;
 278                     _in_translate = true;
 279                     _in_data_block = false;
 280                     _datatype = null;
 281                 }
 282                 else if ( line_lc.startsWith( characters ) || line_lc.startsWith( data ) ) {
 283                     _in_taxalabels = false;
 284                     _in_trees_block = false;
 285                     _in_translate = false;
 286                     _in_data_block = true;
 287                     _datatype = null;
 288                 }
 289                 else if ( _in_trees_block ) {
 290                     if ( line_lc.startsWith( "title" ) ) {
 291                         final Matcher title_m = TITLE_PATTERN.matcher( line );
 292                         if ( title_m.lookingAt() ) {
 293                             _title = title_m.group( 1 );
 294                         }
 295                     }
 296                     else if ( line_lc.startsWith( "link" ) ) {
 297                         //final Matcher link_m = LINK_TAXA_PATTERN.matcher( line );
 298                         //if ( link_m.lookingAt() ) {
 299                             //final String link = link_m.group( 1 );  //TODO why?
 300                        // }
 301                     }
 302                     else if ( line_lc.startsWith( end ) || line_lc.startsWith( endblock ) ) {
 303                         _in_trees_block = false;
 304                         _in_tree = false;
 305                         _in_translate = false;
 306                         if ( _nh.length() > 0 ) {
 307                             createPhylogeny( _title, _name, _nh, _rooted_info_present, _is_rooted );
 308                             _nh = new StringBuilder();
 309                             _name = "";
 310                             _rooted_info_present = false;
 311                             _is_rooted = false;
 312                             if ( _next != null ) {
 313                                 return;
 314                             }
 315                         }
 316                     }
 317                     else if ( line_lc.startsWith( tree ) || ( line_lc.startsWith( utree ) ) ) {
 318                         boolean might = false;
 319                         if ( _nh.length() > 0 ) {
 320                             might = true;
 321                             createPhylogeny( _title, _name, _nh, _rooted_info_present, _is_rooted );
 322                             _nh = new StringBuilder();
 323                             _name = "";
 324                             _rooted_info_present = false;
 325                             _is_rooted = false;
 326                         }
 327                         _in_tree = true;
 328                         _nh.append( line.substring( line.indexOf( '=' ) ) );
 329                         final Matcher name_matcher = TREE_NAME_PATTERN.matcher( line );
 330                         if ( name_matcher.matches() ) {
 331                             _name = name_matcher.group( 1 );
 332                             _name = _name.replaceAll( "['\"]+", "" );
 333                         }
 334                         final Matcher rootedness_matcher = ROOTEDNESS_PATTERN.matcher( line );
 335                         if ( rootedness_matcher.matches() ) {
 336                             final String s = rootedness_matcher.group( 1 );
 337                             line = line.replaceAll( "\\[\\&.\\]", "" );
 338                             _rooted_info_present = true;
 339                             if ( s.toUpperCase().equals( "R" ) ) {
 340                                 _is_rooted = true;
 341                             }
 342                         }
 343                         if ( might && ( _next != null ) ) {
 344                             return;
 345                         }
 346                     }
 347                     else if ( _in_tree && !_in_translate ) {
 348                         _nh.append( line );
 349                     }
 350                     if ( !line_lc.startsWith( "title" ) && !line_lc.startsWith( "link" ) && !_in_translate
 351                             && !line_lc.startsWith( end ) && !line_lc.startsWith( endblock ) && line_lc.endsWith( ";" ) ) {
 352                         _in_tree = false;
 353                         _in_translate = false;
 354                         createPhylogeny( _title, _name, _nh, _rooted_info_present, _is_rooted );
 355                         _nh = new StringBuilder();
 356                         _name = "";
 357                         _rooted_info_present = false;
 358                         _is_rooted = false;
 359                         if ( _next != null ) {
 360                             return;
 361                         }
 362                     }
 363                 }
 364                 if ( _in_taxalabels ) {
 365                     if ( line_lc.startsWith( end ) || line_lc.startsWith( endblock ) ) {
 366                         _in_taxalabels = false;
 367                     }
 368                     else {
 369                         final String[] labels = line.split( "\\s+" );
 370                         for( String label : labels ) {
 371                             if ( !label.toLowerCase().equals( taxlabels ) ) {
 372                                 if ( label.endsWith( ";" ) ) {
 373                                     _in_taxalabels = false;
 374                                     label = label.substring( 0, label.length() - 1 );
 375                                 }
 376                                 if ( label.length() > 0 ) {
 377                                     _taxlabels.add( label );
 378                                 }
 379                             }
 380                         }
 381                     }
 382                 }
 383                 if ( _in_translate ) {
 384                     if ( line_lc.startsWith( end ) || line_lc.startsWith( endblock ) ) {
 385                         _in_translate = false;
 386                     }
 387                     else {
 388                         _translate_sb.append( " " );
 389                         _translate_sb.append( line.trim() );
 390                         if ( line.endsWith( ";" ) ) {
 391                             _in_translate = false;
 392                             setTranslateKeyValuePairs( _translate_sb );
 393                         }
 394                     }
 395                 }
 396                 if ( _in_data_block ) {
 397                     if ( line_lc.startsWith( end ) || line_lc.startsWith( endblock ) ) {
 398                         _in_data_block = false;
 399                         _datatype = null;
 400                     }
 401                     else if ( line_lc.startsWith( "link" ) ) {
 402                      //   final Matcher link_m = LINK_TAXA_PATTERN.matcher( line );
 403                      //   if ( link_m.lookingAt() ) {
 404                      //       final String link = link_m.group( 1 );
 405                      //   }
 406                     }
 407                     else {
 408                         final Matcher datatype_matcher = DATATYPE_PATTERN.matcher( line_lc );
 409                         if ( datatype_matcher.find() ) {
 410                             _datatype = datatype_matcher.group( 1 );
 411                         }
 412                         else {
 413                             if ( ( _datatype != null )
 414                                     && ( _datatype.equals( "protein" ) || _datatype.equals( "dna" ) || _datatype
 415                                             .equals( "rna" ) ) ) {
 416                                 if ( line.endsWith( ";" ) ) {
 417                                     _in_data_block = false;
 418                                     line = line.substring( 0, line.length() - 1 );
 419                                 }
 420                                 final Matcher aln_matcher = ALN_PATTERN.matcher( line );
 421                                 if ( aln_matcher.matches() ) {
 422                                     final String id = aln_matcher.group( 1 );
 423                                     final String seq = aln_matcher.group( 2 );
 424                                     MolecularSequence s = null;
 425                                     if ( _datatype.equals( "protein" ) ) {
 426                                         s = BasicSequence.createAaSequence( id, seq );
 427                                     }
 428                                     else if ( _datatype.equals( "dna" ) ) {
 429                                         s = BasicSequence.createDnaSequence( id, seq );
 430                                     }
 431                                     else {
 432                                         s = BasicSequence.createRnaSequence( id, seq );
 433                                     }
 434                                     _seqs.put( id, s );
 435                                 }
 436                             }
 437                         }
 438                     }
 439                 }
 440             }
 441         }
 442         if ( _nh.length() > 0 ) {
 443             createPhylogeny( _title, _name, _nh, _rooted_info_present, _is_rooted );
 444             if ( _next != null ) {
 445                 return;
 446             }
 447         }
 448     }
 449
 450     private final void setTranslateKeyValuePairs( final StringBuilder translate_sb ) throws IOException {
 451         String s = translate_sb.toString().trim();
 452         if ( s.endsWith( ";" ) ) {
 453             s = s.substring( 0, s.length() - 1 ).trim();
 454         }
 455         for( String pair : s.split( "," ) ) {
 456             String key = "";
 457             String value = "";
 458             final int ti = pair.toLowerCase().indexOf( "translate" );
 459             if ( ti > -1 ) {
 460                 pair = pair.substring( ti + 9 );
 461             }
 462             final Matcher m = TRANSLATE_PATTERN.matcher( pair );
 463             if ( m.find() ) {
 464                 key = m.group( 1 );
 465                 value = m.group( 2 ).replaceAll( "\'", "" ).replaceAll( "\"", "" ).trim();
 466             }
 467             else {
 468                 throw new IOException( "ill-formatted translate values: " + pair );
 469             }
 470             if ( value.endsWith( ";" ) ) {
 471                 value = value.substring( 0, value.length() - 1 );
 472             }
 473             _translate_map.put( key, value );
 474         }
 475     }
 476
 477     private final static String removeWhiteSpaceBeforeSemicolon( final String s ) {
 478         return s.replaceAll( "\\s+;", ";" );
 479     }
 480 }