forester/java/src/org/forester/sdi/SDIx.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: www.phylosoft.org/forester
  25
  26 package org.forester.sdi;
  27
  28 import java.io.File;
  29 import java.io.IOException;
  30 import java.util.ArrayList;
  31 import java.util.HashSet;
  32 import java.util.List;
  33 import java.util.Set;
  34
  35 import org.forester.io.parsers.phyloxml.PhyloXmlParser;
  36 import org.forester.io.writers.PhylogenyWriter;
  37 import org.forester.phylogeny.Phylogeny;
  38 import org.forester.phylogeny.PhylogenyMethods;
  39 import org.forester.phylogeny.PhylogenyNode;
  40 import org.forester.phylogeny.data.Taxonomy;
  41 import org.forester.phylogeny.factories.ParserBasedPhylogenyFactory;
  42 import org.forester.phylogeny.factories.PhylogenyFactory;
  43 import org.forester.phylogeny.iterators.PhylogenyNodeIterator;
  44 import org.forester.util.ForesterUtil;
  45
  46 public class SDIx {
  47
  48     public SDIx() {
  49     }
  50
  51     private void analyze( final Phylogeny gene_tree,
  52                           final String gene_tree_file_name,
  53                           final Phylogeny[] species_trees,
  54                           final File out_dir ) throws IOException {
  55         final boolean minimize_cost = true;
  56         final boolean minimize_sum_of_dup = true;
  57         final boolean minimize_height = true;
  58         final int trees_to_return = 1;
  59         System.out.println( gene_tree_file_name + ": " + gene_tree.getName() );
  60         final Set<Taxonomy> species_tree_species = getAllExternalSpecies( species_trees[ 0 ] );
  61         final PhylogenyWriter w = new PhylogenyWriter();
  62         for( final Phylogeny species_tree : species_trees ) {
  63             PhylogenyMethods.deleteExternalNodesPositiveSelection( species_tree_species, gene_tree );
  64             if ( gene_tree.isEmpty() ) {
  65                 System.out.println( " >> empty: " + gene_tree_file_name + ": " + gene_tree.getName() );
  66                 continue;
  67             }
  68             final File outfile = new File( out_dir + ForesterUtil.FILE_SEPARATOR + gene_tree_file_name );
  69             if ( outfile.exists() ) {
  70                 System.out
  71                         .println( " >> already exists, skipping: " + gene_tree_file_name + ": " + gene_tree.getName() );
  72             }
  73             final SDIR sdir = new SDIR();
  74             final Phylogeny[] analyzed_gene_trees = sdir.infer( gene_tree,
  75                                                                 species_tree,
  76                                                                 minimize_cost,
  77                                                                 minimize_sum_of_dup,
  78                                                                 minimize_height,
  79                                                                 true,
  80                                                                 trees_to_return );
  81             final int duplications = sdir.getMinimalDuplications();
  82             final int mapping_cost = sdir.getMinimalMappingCost();
  83             final List<Phylogeny> phys = new ArrayList<Phylogeny>();
  84             for( final Phylogeny phy : analyzed_gene_trees ) {
  85                 phys.add( phy );
  86             }
  87             w.toPhyloXML( outfile, phys, 0, ForesterUtil.LINE_SEPARATOR );
  88         }
  89     }
  90
  91     private void checkSpeciesTreesForEqualNumberOfExtNodes( final Phylogeny[] species_trees ) {
  92         int ext_nodes = -1;
  93         for( final Phylogeny phylogeny : species_trees ) {
  94             if ( ext_nodes < 0 ) {
  95                 ext_nodes = phylogeny.getNumberOfExternalNodes();
  96             }
  97             else if ( ext_nodes != phylogeny.getNumberOfExternalNodes() ) {
  98                 throw new IllegalArgumentException( "species trees must have all the same number of external nodes" );
  99             }
 100         }
 101     }
 102
 103     public void method1( final List<File> gene_tree_files, final Phylogeny[] species_trees, final File out_dir )
 104             throws IOException {
 105         checkSpeciesTreesForEqualNumberOfExtNodes( species_trees );
 106         final PhylogenyFactory factory = ParserBasedPhylogenyFactory.getInstance();
 107         for( final File gene_tree_file : gene_tree_files ) {
 108             if ( ForesterUtil.isReadableFile( gene_tree_file ) != null ) {
 109                 throw new IOException( "[" + gene_tree_file + "] is not readable" );
 110             }
 111             Phylogeny[] gene_trees = null;
 112             gene_trees = factory.create( gene_tree_file, new PhyloXmlParser() );
 113             if ( gene_trees.length != 1 ) {
 114                 throw new IOException( "[" + gene_tree_file + "] contains " + gene_trees.length
 115                         + " gene trees, expecting precisely one" );
 116             }
 117             analyze( gene_trees[ 0 ], gene_tree_file.getName(), species_trees, out_dir );
 118         }
 119     }
 120
 121     private static Set<Taxonomy> getAllExternalSpecies( final Phylogeny phy ) {
 122         final Set<Taxonomy> specs = new HashSet<Taxonomy>();
 123         for( final PhylogenyNodeIterator it = phy.iteratorExternalForward(); it.hasNext(); ) {
 124             final PhylogenyNode n = it.next();
 125             if ( n.getNodeData().isHasTaxonomy() ) {
 126                 specs.add( n.getNodeData().getTaxonomy() );
 127             }
 128             else {
 129                 throw new IllegalArgumentException( "node " + n.getId() + " has no taxonomic data" );
 130             }
 131         }
 132         return specs;
 133     }
 134 }