X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=forester%2Fjava%2Fsrc%2Forg%2Fforester%2Fapplication%2Frio.java;h=13b2ade45566f2885ef6f756fb13c5a7a447632c;hb=89f36e42fb462f8d7b59def3ead995fb16a87b59;hp=ad0c33abc464ad6be03c0ceed00a63bd4a8ec303;hpb=e9ca0dc1764303d53fc6b9b087f33cdee53726ea;p=jalview.git diff --git a/forester/java/src/org/forester/application/rio.java b/forester/java/src/org/forester/application/rio.java index ad0c33a..13b2ade 100644 --- a/forester/java/src/org/forester/application/rio.java +++ b/forester/java/src/org/forester/application/rio.java @@ -28,39 +28,36 @@ package org.forester.application; import java.io.File; -import java.io.FileWriter; import java.io.IOException; -import java.io.PrintWriter; import java.util.ArrayList; import java.util.List; import org.forester.datastructures.IntMatrix; import org.forester.io.parsers.phyloxml.PhyloXmlParser; import org.forester.phylogeny.Phylogeny; +import org.forester.phylogeny.PhylogenyMethods; import org.forester.phylogeny.factories.ParserBasedPhylogenyFactory; import org.forester.phylogeny.factories.PhylogenyFactory; -import org.forester.sdi.RIO; -import org.forester.sdi.RIOException; +import org.forester.rio.RIO; +import org.forester.rio.RIO.REROOTING; +import org.forester.rio.RIOException; import org.forester.sdi.SDIException; +import org.forester.sdi.SDIutil.ALGORITHM; +import org.forester.util.BasicDescriptiveStatistics; import org.forester.util.CommandLineArguments; import org.forester.util.EasyWriter; import org.forester.util.ForesterUtil; public class rio { - final static private String PRG_NAME = "rio"; - final static private String PRG_VERSION = "3.00 beta 3"; - final static private String PRG_DATE = "2012.12.05"; - final static private String E_MAIL = "czmasek@burnham.org"; - final static private String WWW = "www.phylosoft.org/forester/"; - final static private String HELP_OPTION_1 = "help"; - final static private String HELP_OPTION_2 = "h"; - final static private String QUERY_OPTION = "q"; - final static private String SORT_OPTION = "s"; - final static private String OUTPUT_ULTRA_P_OPTION = "u"; - final static private String CUTOFF_ULTRA_P_OPTION = "cu"; - final static private String CUTOFF_ORTHO_OPTION = "co"; - final static private String TABLE_OUTPUT_OPTION = "t"; + final static private String PRG_NAME = "rio"; + final static private String PRG_VERSION = "4.000 beta 3"; + final static private String PRG_DATE = "2012.12.17"; + final static private String E_MAIL = "czmasek@burnham.org"; + final static private String WWW = "www.phylosoft.org/forester/"; + final static private String HELP_OPTION_1 = "help"; + final static private String HELP_OPTION_2 = "h"; + final static private String USE_SDIR = "b"; public static void main( final String[] args ) { ForesterUtil.printProgramInformation( PRG_NAME, @@ -80,98 +77,57 @@ public class rio { if ( cla.isOptionSet( HELP_OPTION_1 ) || cla.isOptionSet( HELP_OPTION_2 ) || ( args.length == 0 ) ) { printHelp(); } - if ( ( args.length < 2 ) || ( args.length > 10 ) ) { + if ( ( args.length < 3 ) || ( args.length > 8 ) ) { System.out.println(); System.out.println( "[" + PRG_NAME + "] incorrect number of arguments" ); System.out.println(); printHelp(); } final List allowed_options = new ArrayList(); - allowed_options.add( QUERY_OPTION ); - allowed_options.add( SORT_OPTION ); - allowed_options.add( CUTOFF_ULTRA_P_OPTION ); - allowed_options.add( CUTOFF_ORTHO_OPTION ); - allowed_options.add( TABLE_OUTPUT_OPTION ); - allowed_options.add( OUTPUT_ULTRA_P_OPTION ); + allowed_options.add( USE_SDIR ); final String dissallowed_options = cla.validateAllowedOptionsAsString( allowed_options ); if ( dissallowed_options.length() > 0 ) { ForesterUtil.fatalError( PRG_NAME, "unknown option(s): " + dissallowed_options ); } final File gene_trees_file = cla.getFile( 0 ); final File species_tree_file = cla.getFile( 1 ); - File outfile = null; - if ( cla.getNumberOfNames() > 2 ) { - outfile = cla.getFile( 2 ); + final File orthology_outtable = cla.getFile( 2 ); + final File logfile; + if ( cla.getNumberOfNames() > 3 ) { + logfile = cla.getFile( 3 ); + if ( logfile.exists() ) { + ForesterUtil.fatalError( PRG_NAME, "\"" + logfile + "\" already exists" ); + } + } + else { + logfile = null; } + final String outgroup = ""; ForesterUtil.fatalErrorIfFileNotReadable( PRG_NAME, gene_trees_file ); ForesterUtil.fatalErrorIfFileNotReadable( PRG_NAME, species_tree_file ); - if ( ( outfile != null ) && outfile.exists() ) { - ForesterUtil.fatalError( PRG_NAME, "[" + outfile + "] already exists" ); - } - String query = null; - if ( cla.isOptionSet( QUERY_OPTION ) ) { - query = cla.getOptionValue( QUERY_OPTION ); + if ( orthology_outtable.exists() ) { + ForesterUtil.fatalError( PRG_NAME, "\"" + orthology_outtable + "\" already exists" ); } - File table_outfile = null; - if ( cla.isOptionSet( TABLE_OUTPUT_OPTION ) ) { - table_outfile = new File( cla.getOptionValue( TABLE_OUTPUT_OPTION ) ); - if ( table_outfile.exists() ) { - ForesterUtil.fatalError( PRG_NAME, "[" + table_outfile + "] already exists" ); + boolean sdir = false; + if ( cla.isOptionSet( USE_SDIR ) ) { + sdir = true; + if ( logfile != null ) { + ForesterUtil.fatalError( PRG_NAME, "logfile output only for GSDIR algorithm" ); } } - boolean output_ultraparalogs = false; - if ( cla.isOptionSet( OUTPUT_ULTRA_P_OPTION ) ) { - output_ultraparalogs = true; - } - double cutoff_for_orthologs = 50; - double cutoff_for_ultra_paralogs = 50; - int sort = 1; - try { - if ( cla.isOptionSet( CUTOFF_ORTHO_OPTION ) ) { - cutoff_for_orthologs = cla.getOptionValueAsDouble( CUTOFF_ORTHO_OPTION ); - if ( query == null ) { - ForesterUtil.fatalError( PRG_NAME, "missing query name, type \"rio -h\" for help" ); - } - if ( outfile == null ) { - ForesterUtil.fatalError( PRG_NAME, "missing outfile, type \"rio -h\" for help" ); - } - } - if ( cla.isOptionSet( CUTOFF_ULTRA_P_OPTION ) ) { - cutoff_for_ultra_paralogs = cla.getOptionValueAsDouble( CUTOFF_ULTRA_P_OPTION ); - output_ultraparalogs = true; - } - if ( cla.isOptionSet( SORT_OPTION ) ) { - sort = cla.getOptionValueAsInt( SORT_OPTION ); - } - } - catch ( final Exception e ) { - ForesterUtil.fatalError( PRG_NAME, "error in command line: " + e.getLocalizedMessage() ); - } - if ( ( cutoff_for_orthologs < 0 ) || ( cutoff_for_ultra_paralogs < 0 ) || ( sort < 0 ) || ( sort > 2 ) ) { - ForesterUtil.fatalError( PRG_NAME, "numberical option out of range, type \"rio -h\" for help" ); - } - if ( ( ( query == null ) && ( ( outfile != null ) || output_ultraparalogs ) ) ) { - ForesterUtil.fatalError( PRG_NAME, "missing query name, type \"rio -h\" for help" ); - } - if ( ( output_ultraparalogs && ( outfile == null ) ) || ( ( query != null ) && ( outfile == null ) ) ) { - ForesterUtil.fatalError( PRG_NAME, "missing outfile, type \"rio -h\" for help" ); - } long time = 0; System.out.println( "Gene trees : " + gene_trees_file ); System.out.println( "Species tree : " + species_tree_file ); - if ( query != null ) { - System.out.println( "Query : " + query ); - System.out.println( "Outfile : " + outfile ); - System.out.println( "Sort : " + sort ); - System.out.println( "Cutoff for orthologs : " + cutoff_for_orthologs ); - if ( output_ultraparalogs ) { - System.out.println( "Cutoff for ultra paralogs : " + cutoff_for_ultra_paralogs ); + System.out.println( "All vs all orthology table: " + orthology_outtable ); + if ( !sdir ) { + if ( logfile != null ) { + System.out.println( "Logfile : " + logfile ); } + System.out.println( "Non binary species tree : allowed (GSDIR algorithm)" ); } - if ( table_outfile != null ) { - System.out.println( "Table output : " + table_outfile ); + else { + System.out.println( "Non binary species tree : disallowed (SDIR algorithm)" ); } - System.out.println(); time = System.currentTimeMillis(); Phylogeny species_tree = null; try { @@ -185,34 +141,55 @@ public class rio { if ( !species_tree.isRooted() ) { ForesterUtil.fatalError( PRG_NAME, "species tree is not rooted" ); } - if ( !species_tree.isCompletelyBinary() ) { - ForesterUtil.fatalError( PRG_NAME, "species tree is not completely binary" ); + final int o = PhylogenyMethods.countNumberOfOneDescendantNodes( species_tree ); + if ( o > 0 ) { + ForesterUtil.printWarningMessage( PRG_NAME, "species tree has " + o + + " internal nodes with only one descendent! Going to strip them." ); + PhylogenyMethods.deleteInternalNodesWithOnlyOneDescendent( species_tree ); + if ( PhylogenyMethods.countNumberOfOneDescendantNodes( species_tree ) > 0 ) { + ForesterUtil.unexpectedFatalError( PRG_NAME, "stripping of one-desc nodes failed" ); + } + } + final ALGORITHM algorithm; + if ( sdir ) { + algorithm = ALGORITHM.SDIR; + } + else { + algorithm = ALGORITHM.GSDIR; } try { - final RIO rio; - if ( ForesterUtil.isEmpty( query ) ) { - rio = new RIO( gene_trees_file, species_tree ); + final RIO rio = RIO.executeAnalysis( gene_trees_file, + species_tree, + algorithm, + REROOTING.BY_ALGORITHM, + outgroup, + logfile != null, + true ); + if ( algorithm == ALGORITHM.GSDIR ) { + ForesterUtil.programMessage( PRG_NAME, "taxonomy linking based on: " + rio.getGSDIRtaxCompBase() ); } - else { - rio = new RIO( gene_trees_file, species_tree, query ); - } - if ( outfile != null ) { - final StringBuilder output = new StringBuilder(); - output.append( rio.inferredOrthologsToString( query, sort, cutoff_for_orthologs ) ); - if ( output_ultraparalogs ) { - output.append( "\n\nUltra paralogs:\n" ); - output.append( rio.inferredUltraParalogsToString( query, cutoff_for_ultra_paralogs ) ); - } - output.append( "\n\nSort priority: " + RIO.getOrder( sort ) ); - output.append( "\nExt nodes : " + rio.getExtNodesOfAnalyzedGeneTrees() ); - output.append( "\nSamples : " + rio.getNumberOfSamples() + "\n" ); - final PrintWriter out = new PrintWriter( new FileWriter( outfile ), true ); - out.println( output ); - out.close(); + tableOutput( orthology_outtable, rio ); + if ( ( algorithm != ALGORITHM.SDIR ) && ( logfile != null ) ) { + writeLogFile( logfile, + rio, + species_tree_file, + gene_trees_file, + orthology_outtable, + PRG_NAME, + PRG_VERSION, + PRG_DATE, + ForesterUtil.getForesterLibraryInformation() ); } - if ( table_outfile != null ) { - tableOutput( table_outfile, rio ); + final BasicDescriptiveStatistics stats = rio.getDuplicationsStatistics(); + final java.text.DecimalFormat df = new java.text.DecimalFormat( "0.#" ); + ForesterUtil.programMessage( PRG_NAME, + "Mean number of duplications : " + df.format( stats.arithmeticMean() ) + + " (sd: " + df.format( stats.sampleStandardDeviation() ) + ")" ); + if ( stats.getN() > 3 ) { + ForesterUtil.programMessage( PRG_NAME, "Median number of duplications: " + df.format( stats.median() ) ); } + ForesterUtil.programMessage( PRG_NAME, "Minimum duplications : " + ( int ) stats.getMin() ); + ForesterUtil.programMessage( PRG_NAME, "Maximum duplications : " + ( int ) stats.getMax() ); } catch ( final RIOException e ) { ForesterUtil.fatalError( PRG_NAME, e.getLocalizedMessage() ); @@ -226,20 +203,67 @@ public class rio { catch ( final Exception e ) { ForesterUtil.unexpectedFatalError( PRG_NAME, e ); } - if ( outfile != null ) { - ForesterUtil.programMessage( PRG_NAME, "wrote results to \"" + outfile + "\"" ); - } time = System.currentTimeMillis() - time; ForesterUtil.programMessage( PRG_NAME, "time: " + time + "ms" ); ForesterUtil.programMessage( PRG_NAME, "OK" ); System.exit( 0 ); } + private final static void printHelp() { + System.out.println( "Usage" ); + System.out.println(); + System.out + .println( PRG_NAME + + " [options] [logfile]" ); + System.out.println(); + System.out.println( " Options" ); + System.out.println( " -" + USE_SDIR + + " : to use SDIR instead of GSDIR (faster, but non-binary species trees are disallowed)" ); + System.out.println(); + System.out.println( " Formats" ); + System.out.println( " The species tree is expected to be in phyloXML format." ); + System.out + .println( " The gene trees ideally are in phyloXML as well, but can also be in New Hamphshire (Newick)" ); + System.out.println( " or Nexus format as long as species information can be extracted from the gene names" ); + System.out.println( " (e.g. \"HUMAN\" from \"BCL2_HUMAN\")." ); + System.out.println(); + System.out.println( " Examples" ); + System.out.println( " \"rio gene_trees.nh species.xml outtable.tsv log.txt\"" ); + System.out.println(); + System.out.println( " More information: http://code.google.com/p/forester/wiki/RIO" ); + System.out.println(); + System.exit( -1 ); + } + private static void tableOutput( final File table_outfile, final RIO rio ) throws IOException, RIOException { - final IntMatrix m = RIO.calculateOrthologTable( rio.getAnalyzedGeneTrees() ); + final IntMatrix m = RIO.calculateOrthologTable( rio.getAnalyzedGeneTrees(), true ); writeTable( table_outfile, rio, m ); } + private static void writeLogFile( final File logfile, + final RIO rio, + final File species_tree_file, + final File gene_trees_file, + final File outtable, + final String prg_name, + final String prg_v, + final String prg_date, + final String f ) throws IOException { + final EasyWriter out = ForesterUtil.createEasyWriter( logfile ); + out.println( prg_name ); + out.println( "version : " + prg_v ); + out.println( "date : " + prg_date ); + out.println( "based on: " + f ); + out.println( "----------------------------------" ); + out.println( "Gene trees : " + gene_trees_file ); + out.println( "Species tree : " + species_tree_file ); + out.println( "All vs all orthology table : " + outtable ); + out.flush(); + out.println( rio.getLog().toString() ); + out.close(); + ForesterUtil.programMessage( PRG_NAME, "wrote log to \"" + logfile + "\"" ); + } + private static void writeTable( final File table_outfile, final RIO rio, final IntMatrix m ) throws IOException { final EasyWriter w = ForesterUtil.createEasyWriter( table_outfile ); final java.text.DecimalFormat df = new java.text.DecimalFormat( "0.###" ); @@ -254,13 +278,13 @@ public class rio { for( int y = 0; y < m.size(); ++y ) { w.print( "\t" ); if ( x == y ) { - if ( m.get( x, y ) != rio.getNumberOfSamples() ) { + if ( m.get( x, y ) != rio.getAnalyzedGeneTrees().length ) { ForesterUtil.unexpectedFatalError( PRG_NAME, "diagonal value is off" ); } w.print( "-" ); } else { - w.print( df.format( ( ( double ) m.get( x, y ) ) / rio.getNumberOfSamples() ) ); + w.print( df.format( ( ( double ) m.get( x, y ) ) / rio.getAnalyzedGeneTrees().length ) ); } } w.println(); @@ -268,42 +292,4 @@ public class rio { w.close(); ForesterUtil.programMessage( PRG_NAME, "wrote table to \"" + table_outfile + "\"" ); } - - private final static void printHelp() { - System.out.println( "Usage" ); - System.out.println(); - System.out.println( PRG_NAME + " [options] [outfile]" ); - System.out.println(); - System.out.println( " Options" ); - System.out.println( " -" + CUTOFF_ORTHO_OPTION + " : cutoff for ortholog output (default: 50)" ); - System.out.println( " -" + TABLE_OUTPUT_OPTION - + " : file-name for output table of all vs. all ortholgy support" ); - System.out.println( " -" + QUERY_OPTION - + " : name for query (sequence/node), if this is used, [outfile] is required as well" ); - System.out.println( " -" + SORT_OPTION + " : sort (default: 1)" ); - System.out.println( " -" + OUTPUT_ULTRA_P_OPTION - + " : to output ultra-paralogs (species specific expansions/paralogs)" ); - System.out.println( " -" + CUTOFF_ULTRA_P_OPTION + " : cutoff for ultra-paralog output (default: 50)" ); - System.out.println(); - System.out.println( " Note" ); - System.out.println( " Either output of all vs. all ortholgy support with -t= and/or output for" ); - System.out.println( " one query sequence with -q= and a [outfile] are required." ); - System.out.println(); - System.out.println( " Sort" ); - System.out.println( RIO.getOrderHelp().toString() ); - System.out.println( " Formats" ); - System.out.println( " The species tree is expected to be in phyloXML format." ); - System.out - .println( " The gene trees ideally are in phyloXML as well, but can also be in New Hamphshire (Newick)" ); - System.out.println( " or Nexus format as long as species information can be extracted from the gene names" ); - System.out.println( " (e.g. \"HUMAN\" from \"BCL2_HUMAN\")." ); - System.out.println(); - System.out.println( " Examples" ); - System.out.println( " \"rio gene_trees.nh species.xml outfile -q=BCL2_HUMAN -t=outtable -u -cu=60 -co=60\"" ); - System.out.println( " \"rio gene_trees.nh species.xml -t=outtable\"" ); - System.out.println(); - System.out.println( " More information: http://code.google.com/p/forester/wiki/RIO" ); - System.out.println(); - System.exit( -1 ); - } }