X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;ds=inline;f=forester%2Fjava%2Fsrc%2Forg%2Fforester%2Fapplication%2Fgsdi.java;h=eab52df09d22bdf1d1ea05a4ec5df41ee039438a;hb=6005bacbd8aecc0d320b70568a24d321a30bc85c;hp=4c17f0e0c3ac33ee348108f10782eef9769fb60d;hpb=0ee466206ea1e3ac3025df1110538e3815c160b4;p=jalview.git diff --git a/forester/java/src/org/forester/application/gsdi.java b/forester/java/src/org/forester/application/gsdi.java index 4c17f0e..eab52df 100644 --- a/forester/java/src/org/forester/application/gsdi.java +++ b/forester/java/src/org/forester/application/gsdi.java @@ -48,6 +48,7 @@ import org.forester.phylogeny.PhylogenyNode; import org.forester.phylogeny.factories.ParserBasedPhylogenyFactory; import org.forester.phylogeny.factories.PhylogenyFactory; import org.forester.sdi.GSDI; +import org.forester.sdi.GSDIR; import org.forester.sdi.SDI; import org.forester.sdi.SDI.TaxonomyComparisonBase; import org.forester.sdi.SDIException; @@ -60,25 +61,25 @@ import org.forester.util.ForesterUtil; public final class gsdi { private enum BASE_ALGORITHM { - GSDI, SDI + GSDIR, GSDI, SDI } - final static public boolean REPLACE_UNDERSCORES_IN_NH_SPECIES_TREE = true; - final static private String ALLOW_STRIPPING_OF_GENE_TREE_OPTION = "g"; - final static private String SDI_OPTION = "b"; - final static private String MOST_PARSIMONIOUS_OPTION = "m"; - final static private String GUESS_FORMAT_OF_SPECIES_TREE = "q"; - final static private String HELP_OPTION_1 = "help"; - final static private String HELP_OPTION_2 = "h"; - final static private String DEFAULT_OUTFILE_SUFFIX = "_gsdi_out.xml"; - final static private String SUFFIX_FOR_LIST_OF_STIPPED_GENE_TREE_NODES = "_stripped_gene_tree_nodes.txt"; - final static private String SUFFIX_FOR_SPECIES_TREE_USED = "_species_tree_used.xml"; - final static private String LOGFILE_SUFFIX = "_gsdi_log.txt"; - final static private String PRG_NAME = "gsdi"; - final static private String PRG_VERSION = "0.901"; - final static private String PRG_DATE = "120608"; - final static private String PRG_DESC = "general speciation duplication inference"; - final static private String E_MAIL = "phylosoft@gmail.com"; - final static private String WWW = "www.phylosoft.org/forester"; + final static public boolean REPLACE_UNDERSCORES_IN_NH_SPECIES_TREE = true; + final static private String ALLOW_STRIPPING_OF_GENE_TREE_OPTION = "g"; + final static private String SDISE_OPTION = "b"; + final static private String GSDIR_OPTION = "r"; + final static private String MOST_PARSIMONIOUS_OPTION = "m"; + final static private String GUESS_FORMAT_OF_SPECIES_TREE = "q"; + final static private String HELP_OPTION_1 = "help"; + final static private String HELP_OPTION_2 = "h"; + final static private String SUFFIX_FOR_SPECIES_TREE_USED = "_species_tree_used.xml"; + final static private String LOGFILE_SUFFIX = "_gsdi_log.txt"; + final static private String REMAPPED_SUFFIX = "_gsdi_remapped.txt"; + final static private String PRG_NAME = "gsdi"; + final static private String PRG_VERSION = "1.000"; + final static private String PRG_DATE = "120629"; + final static private String PRG_DESC = "general speciation duplication inference"; + final static private String E_MAIL = "phylosoft@gmail.com"; + final static private String WWW = "www.phylosoft.org/forester"; public static void main( final String args[] ) { try { @@ -101,7 +102,7 @@ public final class gsdi { gsdi.print_help(); System.exit( 0 ); } - else if ( ( args.length < 2 ) || ( cla.getNumberOfNames() < 2 ) || ( cla.getNumberOfNames() > 3 ) ) { + else if ( ( args.length < 2 ) || ( cla.getNumberOfNames() != 3 ) ) { System.out.println(); System.out.println( "Wrong number of arguments." ); System.out.println(); @@ -109,7 +110,8 @@ public final class gsdi { System.exit( -1 ); } final List allowed_options = new ArrayList(); - allowed_options.add( gsdi.SDI_OPTION ); + allowed_options.add( gsdi.SDISE_OPTION ); + allowed_options.add( gsdi.GSDIR_OPTION ); allowed_options.add( gsdi.GUESS_FORMAT_OF_SPECIES_TREE ); allowed_options.add( gsdi.MOST_PARSIMONIOUS_OPTION ); allowed_options.add( gsdi.ALLOW_STRIPPING_OF_GENE_TREE_OPTION ); @@ -127,23 +129,22 @@ public final class gsdi { private static void execute( final CommandLineArguments cla ) throws IOException { BASE_ALGORITHM base_algorithm = BASE_ALGORITHM.GSDI; boolean most_parsimonous_duplication_model = false; - boolean species_tree_in_phyloxml = true; boolean allow_stripping_of_gene_tree = false; - if ( cla.isOptionSet( gsdi.SDI_OPTION ) ) { + if ( cla.isOptionSet( gsdi.GSDIR_OPTION ) ) { + base_algorithm = BASE_ALGORITHM.GSDIR; + } + else if ( cla.isOptionSet( gsdi.SDISE_OPTION ) ) { base_algorithm = BASE_ALGORITHM.SDI; } if ( cla.isOptionSet( gsdi.MOST_PARSIMONIOUS_OPTION ) ) { - if ( base_algorithm != BASE_ALGORITHM.GSDI ) { - ForesterUtil.fatalError( gsdi.PRG_NAME, "Can only use most parsimonious duplication mode with GSDI" ); + if ( base_algorithm == BASE_ALGORITHM.SDI ) { + ForesterUtil.fatalError( gsdi.PRG_NAME, "Cannot use most parsimonious duplication mode with SDI" ); } most_parsimonous_duplication_model = true; } - if ( cla.isOptionSet( gsdi.GUESS_FORMAT_OF_SPECIES_TREE ) ) { - species_tree_in_phyloxml = false; - } if ( cla.isOptionSet( gsdi.ALLOW_STRIPPING_OF_GENE_TREE_OPTION ) ) { - if ( base_algorithm != BASE_ALGORITHM.GSDI ) { - ForesterUtil.fatalError( gsdi.PRG_NAME, "Can only allow stripping of gene tree with GSDI" ); + if ( base_algorithm == BASE_ALGORITHM.SDI ) { + ForesterUtil.fatalError( gsdi.PRG_NAME, "Cannot allow stripping of gene tree with SDI" ); } allow_stripping_of_gene_tree = true; } @@ -157,12 +158,7 @@ public final class gsdi { try { gene_tree_file = cla.getFile( 0 ); species_tree_file = cla.getFile( 1 ); - if ( cla.getNumberOfNames() == 3 ) { - out_file = cla.getFile( 2 ); - } - else { - out_file = new File( ForesterUtil.removeSuffix( gene_tree_file.toString() ) + DEFAULT_OUTFILE_SUFFIX ); - } + out_file = cla.getFile( 2 ); log_file = new File( ForesterUtil.removeSuffix( out_file.toString() ) + LOGFILE_SUFFIX ); } catch ( final IllegalArgumentException e ) { @@ -191,16 +187,17 @@ public final class gsdi { gene_tree = factory.create( gene_tree_file, new PhyloXmlParser() )[ 0 ]; } catch ( final IOException e ) { - ForesterUtil.fatalError( gsdi.PRG_NAME, - "Failed to read gene tree from [" + gene_tree_file + "]: " + e.getMessage() ); + fatalError( "ERROR", + "Failed to read gene tree from [" + gene_tree_file + "]: " + e.getMessage(), + log_writer ); } try { final PhylogenyFactory factory = ParserBasedPhylogenyFactory.getInstance(); - if ( species_tree_in_phyloxml ) { - species_tree = factory.create( species_tree_file, new PhyloXmlParser() )[ 0 ]; + final PhylogenyParser p = ParserUtils.createParserDependingOnFileType( species_tree_file, true ); + if ( p instanceof PhyloXmlParser ) { + species_tree = factory.create( species_tree_file, p )[ 0 ]; } else { - final PhylogenyParser p = ParserUtils.createParserDependingOnFileType( species_tree_file, true ); if ( REPLACE_UNDERSCORES_IN_NH_SPECIES_TREE && ( p instanceof NHXParser ) ) { ( ( NHXParser ) p ).setReplaceUnderscores( true ); } @@ -215,9 +212,8 @@ public final class gsdi { true ); } catch ( final PhyloXmlDataFormatException e ) { - ForesterUtil.fatalError( gsdi.PRG_NAME, - "Failed to transfer general node name to scientific name, in [" - + species_tree_file + "]: " + e.getMessage() ); + fatalError( "USER ERROR", "Failed to transfer general node name to scientific name, in [" + + species_tree_file + "]: " + e.getMessage(), log_writer ); } break; case CODE: @@ -228,9 +224,8 @@ public final class gsdi { true ); } catch ( final PhyloXmlDataFormatException e ) { - ForesterUtil.fatalError( gsdi.PRG_NAME, - "Failed to transfer general node name to taxonomy code, in [" - + species_tree_file + "]: " + e.getMessage() ); + fatalError( "USER ERROR", "Failed to transfer general node name to taxonomy code, in [" + + species_tree_file + "]: " + e.getMessage(), log_writer ); } break; case ID: @@ -240,32 +235,30 @@ public final class gsdi { true ); } catch ( final PhyloXmlDataFormatException e ) { - ForesterUtil.fatalError( gsdi.PRG_NAME, - "Failed to transfer general node name to taxonomy id, in [" - + species_tree_file + "]: " + e.getMessage() ); + fatalError( "USER ERROR", "Failed to transfer general node name to taxonomy id, in [" + + species_tree_file + "]: " + e.getMessage(), log_writer ); } break; default: - ForesterUtil.fatalError( gsdi.PRG_NAME, "unable to determine comparison base" ); + fatalError( "UNEXPECTED ERROR", "unable to determine comparison base", log_writer ); } } } catch ( final IOException e ) { - ForesterUtil.fatalError( gsdi.PRG_NAME, "Failed to read species tree from [" + species_tree_file + "]: " - + e.getMessage() ); + fatalError( "ERROR", + "Failed to read species tree from [" + species_tree_file + "]: " + e.getMessage(), + log_writer ); } gene_tree.setRooted( true ); species_tree.setRooted( true ); if ( !gene_tree.isCompletelyBinary() ) { - log_writer.println( "User Error: gene tree is not completely binary" ); - log_writer.close(); - ForesterUtil.fatalError( gsdi.PRG_NAME, "gene tree is not completely binary" ); + fatalError( "user error", "gene tree is not completely binary", log_writer ); } - if ( base_algorithm != BASE_ALGORITHM.GSDI ) { + if ( base_algorithm == BASE_ALGORITHM.SDI ) { if ( !species_tree.isCompletelyBinary() ) { - log_writer.println( "User Error: species tree is not completely binary, use GSDI instead" ); - log_writer.close(); - ForesterUtil.fatalError( gsdi.PRG_NAME, "species tree is not completely binary, use GSDI instead" ); + fatalError( "user error", + "species tree is not completely binary, use GSDI or GSDIR instead", + log_writer ); } } log_writer.println( PRG_NAME + " - " + PRG_DESC ); @@ -292,41 +285,48 @@ public final class gsdi { SDI sdi = null; final long start_time = new Date().getTime(); try { - if ( base_algorithm == BASE_ALGORITHM.GSDI ) { + if ( ( base_algorithm == BASE_ALGORITHM.GSDI ) || ( base_algorithm == BASE_ALGORITHM.GSDIR ) ) { + if ( base_algorithm == BASE_ALGORITHM.GSDI ) { + System.out.println( "Algorithm : GSDI" ); + log_writer.println( "Algorithm : GSDI" ); + } + else if ( base_algorithm == BASE_ALGORITHM.GSDIR ) { + System.out.println( "Algorithm : GSDIR" ); + log_writer.println( "Algorithm : GSDIR" ); + } System.out.println( "Use most parsimonous duplication model : " + most_parsimonous_duplication_model ); System.out.println( "Allow stripping of gene tree nodes : " + allow_stripping_of_gene_tree ); log_writer.println( "Use most parsimonous duplication model : " + most_parsimonous_duplication_model ); log_writer.println( "Allow stripping of gene tree nodes : " + allow_stripping_of_gene_tree ); log_writer.flush(); - sdi = new GSDI( gene_tree, - species_tree, - most_parsimonous_duplication_model, - allow_stripping_of_gene_tree, - true ); + if ( base_algorithm == BASE_ALGORITHM.GSDI ) { + sdi = new GSDI( gene_tree, + species_tree, + most_parsimonous_duplication_model, + allow_stripping_of_gene_tree, + true ); + } + else if ( base_algorithm == BASE_ALGORITHM.GSDIR ) { + sdi = new GSDIR( gene_tree, species_tree, allow_stripping_of_gene_tree, 1 ); + } } else { System.out.println(); - System.out.println( "Using SDIse algorithm" ); - log_writer.println( "Using SDIse algorithm" ); + System.out.println( "Algorithm : SDI" ); + log_writer.println( "Algorithm : SDI" ); log_writer.flush(); sdi = new SDIse( gene_tree, species_tree ); } } catch ( final SDIException e ) { - log_writer.println( "User Error: " + e.getLocalizedMessage() ); - log_writer.close(); - ForesterUtil.fatalError( PRG_NAME, e.getLocalizedMessage() ); + fatalError( "user error", e.getLocalizedMessage(), log_writer ); } catch ( final IOException e ) { - log_writer.println( "Error: " + e ); - log_writer.close(); - ForesterUtil.fatalError( PRG_NAME, e.toString() ); + fatalError( "error", e.toString(), log_writer ); } catch ( final Exception e ) { - log_writer.println( "Error: " + e ); - log_writer.close(); e.printStackTrace(); - System.exit( -1 ); + fatalError( "unexpected error", e.toString(), log_writer ); } System.out.println( "Running time (excluding I/O) : " + ( new Date().getTime() - start_time ) + "ms" ); @@ -337,33 +337,62 @@ public final class gsdi { System.out.println( "Mapping based on : " + gsdi.getTaxCompBase() ); log_writer.println( "Mapping based on : " + gsdi.getTaxCompBase() ); } + if ( ( base_algorithm == BASE_ALGORITHM.GSDIR ) ) { + final GSDIR gsdir = ( GSDIR ) sdi; + System.out.println( "Mapping based on : " + gsdir.getTaxCompBase() ); + log_writer.println( "Mapping based on : " + gsdir.getTaxCompBase() ); + System.out.println( "Minimal duplications sum : " + gsdir.getMinDuplicationsSum() ); + log_writer.println( "Minimal duplications sum : " + gsdir.getMinDuplicationsSum() ); + System.out.println( "Duplications sum statistics : " + gsdir.getMinDuplicationsSum() ); + log_writer.println( "Duplications sum statistics : " + gsdir.getMinDuplicationsSum() ); + } try { final PhylogenyWriter writer = new PhylogenyWriter(); - writer.toPhyloXML( out_file, gene_tree, 0 ); + if ( base_algorithm == BASE_ALGORITHM.GSDIR ) { + writer.toPhyloXML( out_file, + ( ( GSDIR ) sdi ).getMinDuplicationsSumGeneTrees(), + 0, + ForesterUtil.LINE_SEPARATOR ); + } + else { + writer.toPhyloXML( out_file, gene_tree, 0 ); + } } catch ( final IOException e ) { - ForesterUtil.fatalError( PRG_NAME, "Failed to write to [" + out_file + "]: " + e.getMessage() ); + ForesterUtil.fatalError( PRG_NAME, + "Failed to write to [" + out_file.getCanonicalPath() + "]: " + e.getMessage() ); } - System.out.println( "Wrote resulting gene tree to : " + out_file ); - log_writer.println( "Wrote resulting gene tree to : " + out_file ); + System.out.println( "Wrote resulting gene tree to : " + out_file.getCanonicalPath() ); + log_writer.println( "Wrote resulting gene tree to : " + out_file.getCanonicalPath() ); if ( base_algorithm == BASE_ALGORITHM.SDI ) { sdi.computeMappingCostL(); System.out.println( "Mapping cost : " + sdi.computeMappingCostL() ); log_writer.println( "Mapping cost : " + sdi.computeMappingCostL() ); } - else if ( ( base_algorithm == BASE_ALGORITHM.GSDI ) ) { + else if ( ( base_algorithm == BASE_ALGORITHM.GSDI ) || ( base_algorithm == BASE_ALGORITHM.GSDIR ) ) { final GSDI gsdi = ( GSDI ) sdi; - final File species_tree_used_file = new File( out_file + SUFFIX_FOR_SPECIES_TREE_USED ); + final File species_tree_used_file = new File( ForesterUtil.removeSuffix( out_file.toString() ) + + SUFFIX_FOR_SPECIES_TREE_USED ); try { final PhylogenyWriter writer = new PhylogenyWriter(); writer.toPhyloXML( species_tree_used_file, gsdi.getSpeciesTree(), 0 ); } catch ( final IOException e ) { - ForesterUtil.fatalError( PRG_NAME, - "Failed to write to [" + species_tree_used_file + "]: " + e.getMessage() ); + ForesterUtil.fatalError( PRG_NAME, "Failed to write to [" + species_tree_used_file.getCanonicalPath() + + "]: " + e.getMessage() ); + } + System.out.println( "Wrote (stripped) species tree to : " + + species_tree_used_file.getCanonicalPath() ); + log_writer.println( "Wrote (stripped) species tree to : " + + species_tree_used_file.getCanonicalPath() ); + if ( ( gsdi.getReMappedScientificNamesFromGeneTree() != null ) + && !gsdi.getReMappedScientificNamesFromGeneTree().isEmpty() ) { + System.out.println( "Number of gene tree species remapped : " + + gsdi.getReMappedScientificNamesFromGeneTree().size() ); + log_writer.println( "Number of gene tree species remapped : " + + gsdi.getReMappedScientificNamesFromGeneTree().size() ); + writeToRemappedFile( out_file, gsdi.getReMappedScientificNamesFromGeneTree(), log_writer ); } - System.out.println( "Wrote (stripped) species tree to : " + species_tree_used_file ); - log_writer.println( "Wrote (stripped) species tree to : " + species_tree_used_file ); } System.out.println( "Number of external nodes in gene tree : " + gene_tree.getNumberOfExternalNodes() ); log_writer.println( "Number of external nodes in gene tree : " + gene_tree.getNumberOfExternalNodes() ); @@ -401,15 +430,26 @@ public final class gsdi { printMappedNodesToLog( log_writer, gsdi ); log_writer.println(); printStrippedGeneTreeNodesToLog( log_writer, gsdi ); - log_writer.println(); - printStrippedSpeciesTreeNodesToLog( log_writer, gsdi ); } System.out.println(); - System.out.println( "Wrote log to : " + log_file ); + System.out.println( "Wrote log to : " + log_file.getCanonicalPath() ); System.out.println(); log_writer.close(); } + private static void writeToRemappedFile( final File out_file, + final SortedSet remapped, + final EasyWriter log_writer ) throws IOException { + final File file = new File( ForesterUtil.removeSuffix( out_file.toString() ) + REMAPPED_SUFFIX ); + final EasyWriter remapped_writer = ForesterUtil.createEasyWriter( file ); + for( final String s : remapped ) { + remapped_writer.println( s ); + } + remapped_writer.close(); + System.out.println( "Wrote remapped gene tree species to : " + file.getCanonicalPath() ); + log_writer.println( "Wrote remapped gene tree species to : " + file.getCanonicalPath() ); + } + private static void printMappedNodesToLog( final EasyWriter log_writer, final GSDI gsdi ) throws IOException { final SortedSet ss = new TreeSet(); for( final PhylogenyNode n : gsdi.getMappedExternalSpeciesTreeNodes() ) { @@ -421,6 +461,20 @@ public final class gsdi { } } + private static void fatalError( final String type, final String msg, final EasyWriter log_writer ) { + try { + log_writer.flush(); + log_writer.println(); + log_writer.print( type.toUpperCase() + ": " ); + log_writer.println( msg ); + log_writer.close(); + } + catch ( final IOException e ) { + e.printStackTrace(); + } + ForesterUtil.fatalError( gsdi.PRG_NAME, msg ); + } + private static void printStrippedGeneTreeNodesToLog( final EasyWriter log_writer, final GSDI gsdi ) throws IOException { final SortedMap sm = new TreeMap(); @@ -445,45 +499,32 @@ public final class gsdi { } } - private static void printStrippedSpeciesTreeNodesToLog( final EasyWriter log_writer, final GSDI gsdi ) - throws IOException { - final SortedSet ss = new TreeSet(); - for( final PhylogenyNode n : gsdi.getStrippedSpeciesTreeNodes() ) { - ss.add( n.toString() ); - } - log_writer.println( "The following " + ss.size() + " nodes were stripped from the species tree: " ); - for( final String n : ss ) { - log_writer.println( " " + n ); - } - } - private static void print_help() { System.out.println( "Usage: " + gsdi.PRG_NAME - + " [-options] [outfile]" ); + + " [-options] " ); System.out.println(); System.out.println( "Options:" ); - // System.out.println( " -" + gsdi.ALLOW_STRIPPING_OF_GENE_TREE_OPTION + ": to allow stripping of gene tree nodes without a matching species in the species tree (writes list of stripped nodes to " + ); - System.out.println( " -" + gsdi.SDI_OPTION + ": to use SDI algorithm instead of GSDI algorithm" );//TODO gsdi.ALLOW_STRIPPING_OF_GENE_TREE_OPTION not allowed + System.out.println( " -" + gsdi.ALLOW_STRIPPING_OF_GENE_TREE_OPTION + + ": to allow stripping of gene tree nodes without a matching species" ); System.out.println( " -" + gsdi.MOST_PARSIMONIOUS_OPTION + ": use most parimonious duplication model for GSDI: " ); System.out.println( " assign nodes as speciations which would otherwise be assiged" ); - System.out.println( " as unknown because of polytomies in the species tree" ); + System.out.println( " as potential duplications due to polytomies in the species tree" ); System.out.println( " -" + gsdi.GUESS_FORMAT_OF_SPECIES_TREE + ": to allow species tree in other formats than phyloXML (i.e. Newick, NHX, Nexus)" ); - System.out.println(); - System.out.println( "Species tree:" ); - System.out - .println( " In phyloXML format (unless option -" - + gsdi.GUESS_FORMAT_OF_SPECIES_TREE - + " is used, in which case the species matching between gene tree and species tree must be via scientific name), with taxonomy data in appropriate fields" ); + System.out.println( " -" + gsdi.SDISE_OPTION + + ": to use SDIse algorithm instead of GSDI algorithm (for binary species trees)" ); + System.out.println( " -" + gsdi.GSDIR_OPTION + + ": to use GSDIR algorithm instead of GSDI algorithm (re-rooting)" ); System.out.println(); System.out.println( "Gene tree:" ); - System.out.println( " In phyloXM format, with taxonomy and sequence data in appropriate fields" ); + System.out.println( " in phyloXM format, with taxonomy and sequence data in appropriate fields" ); + System.out.println(); + System.out.println( "Species tree:" ); + System.out.println( " in phyloXML format (unless option -" + gsdi.GUESS_FORMAT_OF_SPECIES_TREE + " is used)" ); System.out.println(); - System.out.println( "Example:" ); - // System.out.println( "gsdi - // System.out.println(); - System.out.println( "Note -- GSDI algorithm is under development" ); + System.out.println( "Example: gsdi -" + ALLOW_STRIPPING_OF_GENE_TREE_OPTION + + " gene_tree.xml tree_of_life.xml out.xml" ); System.out.println(); } }