2 // FORESTER -- software libraries and applications
3 // for evolutionary biology research and applications.
5 // Copyright (C) 2008-2009 Christian M. Zmasek
6 // Copyright (C) 2008-2009 Burnham Institute for Medical Research
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.
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.
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
23 // Contact: phylosoft @ gmail . com
24 // WWW: www.phylosoft.org/forester
26 package org.forester.application;
29 import java.io.IOException;
30 import java.text.SimpleDateFormat;
31 import java.util.ArrayList;
32 import java.util.Date;
33 import java.util.List;
34 import java.util.SortedMap;
35 import java.util.SortedSet;
36 import java.util.TreeMap;
37 import java.util.TreeSet;
39 import org.forester.io.parsers.PhylogenyParser;
40 import org.forester.io.parsers.nhx.NHXParser;
41 import org.forester.io.parsers.phyloxml.PhyloXmlDataFormatException;
42 import org.forester.io.parsers.phyloxml.PhyloXmlParser;
43 import org.forester.io.parsers.util.ParserUtils;
44 import org.forester.io.writers.PhylogenyWriter;
45 import org.forester.phylogeny.Phylogeny;
46 import org.forester.phylogeny.PhylogenyMethods;
47 import org.forester.phylogeny.PhylogenyNode;
48 import org.forester.phylogeny.factories.ParserBasedPhylogenyFactory;
49 import org.forester.phylogeny.factories.PhylogenyFactory;
50 import org.forester.sdi.GSDI;
51 import org.forester.sdi.GSDIR;
52 import org.forester.sdi.SDI;
53 import org.forester.sdi.SDI.ALGORITHM;
54 import org.forester.sdi.SDI.TaxonomyComparisonBase;
55 import org.forester.sdi.SDIException;
56 import org.forester.sdi.SDIse;
57 import org.forester.util.CommandLineArguments;
58 import org.forester.util.EasyWriter;
59 import org.forester.util.ForesterConstants;
60 import org.forester.util.ForesterUtil;
62 public final class gsdi {
64 final static public boolean REPLACE_UNDERSCORES_IN_NH_SPECIES_TREE = true;
65 final static private String ALLOW_STRIPPING_OF_GENE_TREE_OPTION = "g";
66 final static private String SDISE_OPTION = "b";
67 final static private String GSDIR_OPTION = "r";
68 final static private String MOST_PARSIMONIOUS_OPTION = "m";
69 final static private String GUESS_FORMAT_OF_SPECIES_TREE = "q";
70 final static private String HELP_OPTION_1 = "help";
71 final static private String HELP_OPTION_2 = "h";
72 final static private String SUFFIX_FOR_SPECIES_TREE_USED = "_species_tree_used.xml";
73 final static private String LOGFILE_SUFFIX = "_gsdi_log.txt";
74 final static private String REMAPPED_SUFFIX = "_gsdi_remapped.txt";
75 final static private String PRG_NAME = "gsdi";
76 final static private String PRG_VERSION = "1.000";
77 final static private String PRG_DATE = "120629";
78 final static private String PRG_DESC = "general speciation duplication inference";
79 final static private String E_MAIL = "phylosoft@gmail.com";
80 final static private String WWW = "www.phylosoft.org/forester";
82 public static void main( final String args[] ) {
84 ForesterUtil.printProgramInformation( PRG_NAME,
90 ForesterUtil.getForesterLibraryInformation() );
91 CommandLineArguments cla = null;
93 cla = new CommandLineArguments( args );
95 catch ( final Exception e ) {
96 ForesterUtil.fatalError( PRG_NAME, e.getMessage() );
98 if ( cla.isOptionSet( gsdi.HELP_OPTION_1 ) || cla.isOptionSet( gsdi.HELP_OPTION_2 ) ) {
103 else if ( ( args.length < 2 ) || ( cla.getNumberOfNames() != 3 ) ) {
104 System.out.println();
105 System.out.println( "Wrong number of arguments." );
106 System.out.println();
110 final List<String> allowed_options = new ArrayList<String>();
111 allowed_options.add( gsdi.SDISE_OPTION );
112 allowed_options.add( gsdi.GSDIR_OPTION );
113 allowed_options.add( gsdi.GUESS_FORMAT_OF_SPECIES_TREE );
114 allowed_options.add( gsdi.MOST_PARSIMONIOUS_OPTION );
115 allowed_options.add( gsdi.ALLOW_STRIPPING_OF_GENE_TREE_OPTION );
116 final String dissallowed_options = cla.validateAllowedOptionsAsString( allowed_options );
117 if ( dissallowed_options.length() > 0 ) {
118 ForesterUtil.fatalError( gsdi.PRG_NAME, "unknown option(s): " + dissallowed_options );
122 catch ( final IOException e ) {
123 ForesterUtil.fatalError( gsdi.PRG_NAME, e.getMessage() );
127 private static void execute( final CommandLineArguments cla ) throws IOException {
128 ALGORITHM base_algorithm = ALGORITHM.GSDI;
129 boolean most_parsimonous_duplication_model = false;
130 boolean allow_stripping_of_gene_tree = false;
131 if ( cla.isOptionSet( gsdi.GSDIR_OPTION ) ) {
132 base_algorithm = ALGORITHM.GSDIR;
134 else if ( cla.isOptionSet( gsdi.SDISE_OPTION ) ) {
135 base_algorithm = ALGORITHM.SDI;
137 if ( cla.isOptionSet( gsdi.MOST_PARSIMONIOUS_OPTION ) ) {
138 if ( base_algorithm == ALGORITHM.SDI ) {
139 ForesterUtil.fatalError( gsdi.PRG_NAME, "Cannot use most parsimonious duplication mode with SDI" );
141 most_parsimonous_duplication_model = true;
143 if ( cla.isOptionSet( gsdi.ALLOW_STRIPPING_OF_GENE_TREE_OPTION ) ) {
144 if ( base_algorithm == ALGORITHM.SDI ) {
145 ForesterUtil.fatalError( gsdi.PRG_NAME, "Cannot allow stripping of gene tree with SDI" );
147 allow_stripping_of_gene_tree = true;
149 Phylogeny species_tree = null;
150 Phylogeny gene_tree = null;
151 File gene_tree_file = null;
152 File species_tree_file = null;
153 File out_file = null;
154 File log_file = null;
155 EasyWriter log_writer = null;
157 gene_tree_file = cla.getFile( 0 );
158 species_tree_file = cla.getFile( 1 );
159 out_file = cla.getFile( 2 );
160 log_file = new File( ForesterUtil.removeSuffix( out_file.toString() ) + LOGFILE_SUFFIX );
162 catch ( final IllegalArgumentException e ) {
163 ForesterUtil.fatalError( gsdi.PRG_NAME, "error in command line: " + e.getMessage() );
165 if ( ForesterUtil.isReadableFile( gene_tree_file ) != null ) {
166 ForesterUtil.fatalError( gsdi.PRG_NAME, ForesterUtil.isReadableFile( gene_tree_file ) );
168 if ( ForesterUtil.isReadableFile( species_tree_file ) != null ) {
169 ForesterUtil.fatalError( gsdi.PRG_NAME, ForesterUtil.isReadableFile( species_tree_file ) );
171 if ( ForesterUtil.isWritableFile( out_file ) != null ) {
172 ForesterUtil.fatalError( gsdi.PRG_NAME, ForesterUtil.isWritableFile( out_file ) );
174 if ( ForesterUtil.isWritableFile( log_file ) != null ) {
175 ForesterUtil.fatalError( gsdi.PRG_NAME, ForesterUtil.isWritableFile( log_file ) );
178 log_writer = ForesterUtil.createEasyWriter( log_file );
180 catch ( final IOException e ) {
181 ForesterUtil.fatalError( gsdi.PRG_NAME, "Failed to create [" + log_file + "]: " + e.getMessage() );
184 final PhylogenyFactory factory = ParserBasedPhylogenyFactory.getInstance();
185 gene_tree = factory.create( gene_tree_file, new PhyloXmlParser() )[ 0 ];
187 catch ( final IOException e ) {
189 "Failed to read gene tree from [" + gene_tree_file + "]: " + e.getMessage(),
193 final PhylogenyFactory factory = ParserBasedPhylogenyFactory.getInstance();
194 final PhylogenyParser p = ParserUtils.createParserDependingOnFileType( species_tree_file, true );
195 if ( p instanceof PhyloXmlParser ) {
196 species_tree = factory.create( species_tree_file, p )[ 0 ];
199 if ( REPLACE_UNDERSCORES_IN_NH_SPECIES_TREE && ( p instanceof NHXParser ) ) {
200 ( ( NHXParser ) p ).setReplaceUnderscores( true );
202 species_tree = factory.create( species_tree_file, p )[ 0 ];
203 final TaxonomyComparisonBase comp_base = GSDI.determineTaxonomyComparisonBase( gene_tree );
204 switch ( comp_base ) {
205 case SCIENTIFIC_NAME:
208 .transferNodeNameToField( species_tree,
209 PhylogenyMethods.PhylogenyNodeField.TAXONOMY_SCIENTIFIC_NAME,
212 catch ( final PhyloXmlDataFormatException e ) {
213 fatalError( "USER ERROR", "Failed to transfer general node name to scientific name, in ["
214 + species_tree_file + "]: " + e.getMessage(), log_writer );
220 .transferNodeNameToField( species_tree,
221 PhylogenyMethods.PhylogenyNodeField.TAXONOMY_CODE,
224 catch ( final PhyloXmlDataFormatException e ) {
225 fatalError( "USER ERROR", "Failed to transfer general node name to taxonomy code, in ["
226 + species_tree_file + "]: " + e.getMessage(), log_writer );
231 PhylogenyMethods.transferNodeNameToField( species_tree,
232 PhylogenyMethods.PhylogenyNodeField.TAXONOMY_ID,
235 catch ( final PhyloXmlDataFormatException e ) {
236 fatalError( "USER ERROR", "Failed to transfer general node name to taxonomy id, in ["
237 + species_tree_file + "]: " + e.getMessage(), log_writer );
241 fatalError( "UNEXPECTED ERROR", "unable to determine comparison base", log_writer );
245 catch ( final IOException e ) {
247 "Failed to read species tree from [" + species_tree_file + "]: " + e.getMessage(),
250 gene_tree.setRooted( true );
251 species_tree.setRooted( true );
252 if ( !gene_tree.isCompletelyBinary() ) {
253 fatalError( "user error", "gene tree is not completely binary", log_writer );
255 if ( base_algorithm == ALGORITHM.SDI ) {
256 if ( !species_tree.isCompletelyBinary() ) {
257 fatalError( "user error",
258 "species tree is not completely binary, use GSDI or GSDIR instead",
262 log_writer.println( PRG_NAME + " - " + PRG_DESC );
263 log_writer.println( " version : " + PRG_VERSION );
264 log_writer.println( " date : " + PRG_DATE );
265 log_writer.println( " forester version: " + ForesterConstants.FORESTER_VERSION );
266 log_writer.println();
267 log_writer.println( "Start time : "
268 + new SimpleDateFormat( "yyyyMMdd HH:mm:ss" ).format( new Date() ) );
269 System.out.println( "Start time : "
270 + new SimpleDateFormat( "yyyyMMdd HH:mm:ss" ).format( new Date() ) );
271 log_writer.println( "Gene tree file : " + gene_tree_file.getCanonicalPath() );
272 System.out.println( "Gene tree file : " + gene_tree_file.getCanonicalPath() );
273 log_writer.println( "Gene tree name : "
274 + ( ForesterUtil.isEmpty( gene_tree.getName() ) ? "" : gene_tree.getName() ) );
275 System.out.println( "Gene tree name : "
276 + ( ForesterUtil.isEmpty( gene_tree.getName() ) ? "" : gene_tree.getName() ) );
277 log_writer.println( "Species tree file : " + species_tree_file.getCanonicalPath() );
278 System.out.println( "Species tree file : " + species_tree_file.getCanonicalPath() );
279 log_writer.println( "Species tree name : "
280 + ( ForesterUtil.isEmpty( species_tree.getName() ) ? "" : gene_tree.getName() ) );
281 System.out.println( "Species tree name : "
282 + ( ForesterUtil.isEmpty( species_tree.getName() ) ? "" : gene_tree.getName() ) );
284 final long start_time = new Date().getTime();
286 if ( ( base_algorithm == ALGORITHM.GSDI ) || ( base_algorithm == ALGORITHM.GSDIR ) ) {
287 if ( base_algorithm == ALGORITHM.GSDI ) {
288 System.out.println( "Algorithm : GSDI" );
289 log_writer.println( "Algorithm : GSDI" );
291 else if ( base_algorithm == ALGORITHM.GSDIR ) {
292 System.out.println( "Algorithm : GSDIR" );
293 log_writer.println( "Algorithm : GSDIR" );
295 System.out.println( "Use most parsimonous duplication model : " + most_parsimonous_duplication_model );
296 System.out.println( "Allow stripping of gene tree nodes : " + allow_stripping_of_gene_tree );
297 log_writer.println( "Use most parsimonous duplication model : " + most_parsimonous_duplication_model );
298 log_writer.println( "Allow stripping of gene tree nodes : " + allow_stripping_of_gene_tree );
300 if ( base_algorithm == ALGORITHM.GSDI ) {
301 sdi = new GSDI( gene_tree,
303 most_parsimonous_duplication_model,
304 allow_stripping_of_gene_tree,
307 else if ( base_algorithm == ALGORITHM.GSDIR ) {
308 sdi = new GSDIR( gene_tree, species_tree, allow_stripping_of_gene_tree, 1 );
312 System.out.println();
313 System.out.println( "Algorithm : SDI" );
314 log_writer.println( "Algorithm : SDI" );
316 sdi = new SDIse( gene_tree, species_tree );
319 catch ( final SDIException e ) {
320 fatalError( "user error", e.getLocalizedMessage(), log_writer );
322 catch ( final IOException e ) {
323 fatalError( "error", e.toString(), log_writer );
325 catch ( final Exception e ) {
327 fatalError( "unexpected error", e.toString(), log_writer );
329 System.out.println( "Running time (excluding I/O) : " + ( new Date().getTime() - start_time )
331 log_writer.println( "Running time (excluding I/O) : " + ( new Date().getTime() - start_time )
333 if ( ( base_algorithm == ALGORITHM.GSDI ) ) {
334 final GSDI gsdi = ( GSDI ) sdi;
335 System.out.println( "Mapping based on : " + gsdi.getTaxCompBase() );
336 log_writer.println( "Mapping based on : " + gsdi.getTaxCompBase() );
338 if ( ( base_algorithm == ALGORITHM.GSDIR ) ) {
339 final GSDIR gsdir = ( GSDIR ) sdi;
340 System.out.println( "Mapping based on : " + gsdir.getTaxCompBase() );
341 log_writer.println( "Mapping based on : " + gsdir.getTaxCompBase() );
342 System.out.println( "Minimal duplications sum : " + gsdir.getMinDuplicationsSum() );
343 log_writer.println( "Minimal duplications sum : " + gsdir.getMinDuplicationsSum() );
344 System.out.println( "Duplications sum statistics : " + gsdir.getMinDuplicationsSum() );
345 log_writer.println( "Duplications sum statistics : " + gsdir.getMinDuplicationsSum() );
348 final PhylogenyWriter writer = new PhylogenyWriter();
349 if ( base_algorithm == ALGORITHM.GSDIR ) {
350 writer.toPhyloXML( out_file,
351 ( ( GSDIR ) sdi ).getMinDuplicationsSumGeneTrees(),
353 ForesterUtil.LINE_SEPARATOR );
356 writer.toPhyloXML( out_file, gene_tree, 0 );
359 catch ( final IOException e ) {
360 ForesterUtil.fatalError( PRG_NAME,
361 "Failed to write to [" + out_file.getCanonicalPath() + "]: " + e.getMessage() );
363 System.out.println( "Wrote resulting gene tree to : " + out_file.getCanonicalPath() );
364 log_writer.println( "Wrote resulting gene tree to : " + out_file.getCanonicalPath() );
365 if ( base_algorithm == ALGORITHM.SDI ) {
366 sdi.computeMappingCostL();
367 System.out.println( "Mapping cost : " + sdi.computeMappingCostL() );
368 log_writer.println( "Mapping cost : " + sdi.computeMappingCostL() );
370 else if ( ( base_algorithm == ALGORITHM.GSDI ) || ( base_algorithm == ALGORITHM.GSDIR ) ) {
371 final GSDI gsdi = ( GSDI ) sdi;
372 final File species_tree_used_file = new File( ForesterUtil.removeSuffix( out_file.toString() )
373 + SUFFIX_FOR_SPECIES_TREE_USED );
375 final PhylogenyWriter writer = new PhylogenyWriter();
376 writer.toPhyloXML( species_tree_used_file, gsdi.getSpeciesTree(), 0 );
378 catch ( final IOException e ) {
379 ForesterUtil.fatalError( PRG_NAME, "Failed to write to [" + species_tree_used_file.getCanonicalPath()
380 + "]: " + e.getMessage() );
382 System.out.println( "Wrote (stripped) species tree to : "
383 + species_tree_used_file.getCanonicalPath() );
384 log_writer.println( "Wrote (stripped) species tree to : "
385 + species_tree_used_file.getCanonicalPath() );
386 if ( ( gsdi.getReMappedScientificNamesFromGeneTree() != null )
387 && !gsdi.getReMappedScientificNamesFromGeneTree().isEmpty() ) {
388 System.out.println( "Number of gene tree species remapped : "
389 + gsdi.getReMappedScientificNamesFromGeneTree().size() );
390 log_writer.println( "Number of gene tree species remapped : "
391 + gsdi.getReMappedScientificNamesFromGeneTree().size() );
392 writeToRemappedFile( out_file, gsdi.getReMappedScientificNamesFromGeneTree(), log_writer );
395 System.out.println( "Number of external nodes in gene tree : " + gene_tree.getNumberOfExternalNodes() );
396 log_writer.println( "Number of external nodes in gene tree : " + gene_tree.getNumberOfExternalNodes() );
397 System.out.println( "Number of external nodes in species tree : "
398 + sdi.getSpeciesTree().getNumberOfExternalNodes() );
399 log_writer.println( "Number of external nodes in species tree : "
400 + sdi.getSpeciesTree().getNumberOfExternalNodes() );
401 if ( ( base_algorithm == ALGORITHM.GSDI ) ) {
402 final GSDI gsdi = ( GSDI ) sdi;
403 final int poly = PhylogenyMethods.countNumberOfPolytomies( gsdi.getSpeciesTree() );
404 System.out.println( "Number of polytomies in species tree : " + poly );
405 log_writer.println( "Number of polytomies in species tree : " + poly );
406 System.out.println( "External nodes stripped from gene tree : "
407 + gsdi.getStrippedExternalGeneTreeNodes().size() );
408 log_writer.println( "External nodes stripped from gene tree : "
409 + gsdi.getStrippedExternalGeneTreeNodes().size() );
410 System.out.println( "External nodes stripped from species tree: "
411 + gsdi.getStrippedSpeciesTreeNodes().size() );
412 log_writer.println( "External nodes stripped from species tree: "
413 + gsdi.getStrippedSpeciesTreeNodes().size() );
415 System.out.println();
416 System.out.println( "Number of duplications : " + sdi.getDuplicationsSum() );
417 log_writer.println( "Number of duplications : " + sdi.getDuplicationsSum() );
418 if ( ( base_algorithm == ALGORITHM.GSDI ) ) {
419 final GSDI gsdi = ( GSDI ) sdi;
420 if ( !most_parsimonous_duplication_model ) {
421 final int u = gsdi.getSpeciationOrDuplicationEventsSum();
422 System.out.println( "Number of potential duplications : " + u );
423 log_writer.println( "Number of potential duplications : " + u );
425 System.out.println( "Number of speciations : " + gsdi.getSpeciationsSum() );
426 log_writer.println( "Number of speciations : " + gsdi.getSpeciationsSum() );
427 log_writer.println();
428 printMappedNodesToLog( log_writer, gsdi );
429 log_writer.println();
430 printStrippedGeneTreeNodesToLog( log_writer, gsdi );
432 System.out.println();
433 System.out.println( "Wrote log to : " + log_file.getCanonicalPath() );
434 System.out.println();
438 private static void writeToRemappedFile( final File out_file,
439 final SortedSet<String> remapped,
440 final EasyWriter log_writer ) throws IOException {
441 final File file = new File( ForesterUtil.removeSuffix( out_file.toString() ) + REMAPPED_SUFFIX );
442 final EasyWriter remapped_writer = ForesterUtil.createEasyWriter( file );
443 for( final String s : remapped ) {
444 remapped_writer.println( s );
446 remapped_writer.close();
447 System.out.println( "Wrote remapped gene tree species to : " + file.getCanonicalPath() );
448 log_writer.println( "Wrote remapped gene tree species to : " + file.getCanonicalPath() );
451 private static void printMappedNodesToLog( final EasyWriter log_writer, final GSDI gsdi ) throws IOException {
452 final SortedSet<String> ss = new TreeSet<String>();
453 for( final PhylogenyNode n : gsdi.getMappedExternalSpeciesTreeNodes() ) {
454 ss.add( n.toString() );
456 log_writer.println( "The following " + ss.size() + " species were used: " );
457 for( final String s : ss ) {
458 log_writer.println( " " + s );
462 private static void fatalError( final String type, final String msg, final EasyWriter log_writer ) {
465 log_writer.println();
466 log_writer.print( type.toUpperCase() + ": " );
467 log_writer.println( msg );
470 catch ( final IOException e ) {
473 ForesterUtil.fatalError( gsdi.PRG_NAME, msg );
476 private static void printStrippedGeneTreeNodesToLog( final EasyWriter log_writer, final GSDI gsdi )
478 final SortedMap<String, Integer> sm = new TreeMap<String, Integer>();
479 for( final PhylogenyNode n : gsdi.getStrippedExternalGeneTreeNodes() ) {
480 final String s = n.toString();
481 if ( sm.containsKey( s ) ) {
482 sm.put( s, sm.get( s ) + 1 );
488 log_writer.println( "The following " + sm.size() + " nodes were stripped from the gene tree: " );
489 for( final String s : sm.keySet() ) {
490 final int count = sm.get( s );
492 log_writer.println( " " + s );
495 log_writer.println( " " + s + " [" + count + "]" );
500 private static void print_help() {
501 System.out.println( "Usage: " + gsdi.PRG_NAME
502 + " [-options] <gene tree in phyloXML format> <species tree> <outfile>" );
503 System.out.println();
504 System.out.println( "Options:" );
505 System.out.println( " -" + gsdi.ALLOW_STRIPPING_OF_GENE_TREE_OPTION
506 + ": to allow stripping of gene tree nodes without a matching species" );
507 System.out.println( " -" + gsdi.MOST_PARSIMONIOUS_OPTION
508 + ": use most parimonious duplication model for GSDI: " );
509 System.out.println( " assign nodes as speciations which would otherwise be assiged" );
510 System.out.println( " as potential duplications due to polytomies in the species tree" );
511 System.out.println( " -" + gsdi.GUESS_FORMAT_OF_SPECIES_TREE
512 + ": to allow species tree in other formats than phyloXML (i.e. Newick, NHX, Nexus)" );
513 System.out.println( " -" + gsdi.SDISE_OPTION
514 + ": to use SDIse algorithm instead of GSDI algorithm (for binary species trees)" );
515 System.out.println( " -" + gsdi.GSDIR_OPTION
516 + ": to use GSDIR algorithm instead of GSDI algorithm (re-rooting)" );
517 System.out.println();
518 System.out.println( "Gene tree:" );
519 System.out.println( " in phyloXM format, with taxonomy and sequence data in appropriate fields" );
520 System.out.println();
521 System.out.println( "Species tree:" );
522 System.out.println( " in phyloXML format (unless option -" + gsdi.GUESS_FORMAT_OF_SPECIES_TREE + " is used)" );
523 System.out.println();
524 System.out.println( "Example: gsdi -" + ALLOW_STRIPPING_OF_GENE_TREE_OPTION
525 + " gene_tree.xml tree_of_life.xml out.xml" );
526 System.out.println();