X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;ds=sidebyside;f=forester%2Fjava%2Fsrc%2Forg%2Fforester%2Fapplication%2Frio.java;h=13b2ade45566f2885ef6f756fb13c5a7a447632c;hb=89f36e42fb462f8d7b59def3ead995fb16a87b59;hp=397329cc60ffdbab7cffee4b332bb71acd6e77b8;hpb=caebe5e946fc15f9042dc70cac985910c6205e3c;p=jalview.git diff --git a/forester/java/src/org/forester/application/rio.java b/forester/java/src/org/forester/application/rio.java index 397329c..13b2ade 100644 --- a/forester/java/src/org/forester/application/rio.java +++ b/forester/java/src/org/forester/application/rio.java @@ -35,12 +35,15 @@ 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.rio.RIO; +import org.forester.rio.RIO.REROOTING; import org.forester.rio.RIOException; -import org.forester.sdi.SDI.ALGORITHM; 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; @@ -48,8 +51,8 @@ import org.forester.util.ForesterUtil; public class rio { final static private String PRG_NAME = "rio"; - final static private String PRG_VERSION = "3.00 beta 4"; - final static private String PRG_DATE = "2012.12.10"; + 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"; @@ -74,7 +77,7 @@ 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(); @@ -88,34 +91,43 @@ public class rio { } final File gene_trees_file = cla.getFile( 0 ); final File species_tree_file = cla.getFile( 1 ); - final File othology_outtable = 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 ( othology_outtable.exists() ) { - ForesterUtil.fatalError( PRG_NAME, "\"" + othology_outtable + "\" already exists" ); + if ( orthology_outtable.exists() ) { + ForesterUtil.fatalError( PRG_NAME, "\"" + orthology_outtable + "\" 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" ); + } } long time = 0; System.out.println( "Gene trees : " + gene_trees_file ); System.out.println( "Species tree : " + species_tree_file ); - System.out.println( "All vs all orthology table: " + othology_outtable ); + 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)" ); } else { System.out.println( "Non binary species tree : disallowed (SDIR algorithm)" ); } - System.out.println(); time = System.currentTimeMillis(); Phylogeny species_tree = null; try { @@ -129,6 +141,15 @@ public class rio { if ( !species_tree.isRooted() ) { ForesterUtil.fatalError( PRG_NAME, "species tree is not rooted" ); } + 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; @@ -137,8 +158,38 @@ public class rio { algorithm = ALGORITHM.GSDIR; } try { - final RIO rio = new RIO( gene_trees_file, species_tree, algorithm ); - tableOutput( othology_outtable, rio ); + 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() ); + } + 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() ); + } + 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() ); @@ -152,20 +203,67 @@ public class rio { catch ( final Exception e ) { ForesterUtil.unexpectedFatalError( PRG_NAME, e ); } - if ( othology_outtable != null ) { - ForesterUtil.programMessage( PRG_NAME, "wrote results to \"" + othology_outtable + "\"" ); - } 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(), 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.###" ); @@ -180,13 +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(); @@ -194,30 +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] [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 ); - } }