X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;ds=sidebyside;f=forester%2Fjava%2Fsrc%2Forg%2Fforester%2Frio%2FRIO.java;h=0de8e9a04712e9ce9864ed9d0aa2b320fead6502;hb=f8ecfc1d254f7f413ee5c47fbb012b609f7f4fd1;hp=6e0706ef8294d216823aec93e1a4dfc322234360;hpb=3e6923f2f1a9802f78f7989b42590e5adfccac34;p=jalview.git diff --git a/forester/java/src/org/forester/rio/RIO.java b/forester/java/src/org/forester/rio/RIO.java index 6e0706e..0de8e9a 100644 --- a/forester/java/src/org/forester/rio/RIO.java +++ b/forester/java/src/org/forester/rio/RIO.java @@ -30,6 +30,7 @@ package org.forester.rio; import java.io.File; import java.io.FileNotFoundException; import java.io.IOException; +import java.text.DecimalFormat; import java.util.ArrayList; import java.util.Collections; import java.util.HashMap; @@ -65,6 +66,7 @@ import org.forester.util.ForesterUtil; public final class RIO { public static final int DEFAULT_RANGE = -1; + private static final int END_OF_GT = Integer.MAX_VALUE; private static IntMatrix _m; private Phylogeny[] _analyzed_gene_trees; private List _removed_gene_tree_nodes; @@ -76,8 +78,10 @@ public final class RIO { private final boolean _produce_log; private final boolean _verbose; private final REROOTING _rerooting; + private final Phylogeny _species_tree; + private Phylogeny _min_dub_gene_tree; - private RIO( final Phylogeny[] gene_trees, + private RIO( final IteratingPhylogenyParser p, final Phylogeny species_tree, final ALGORITHM algorithm, final REROOTING rerooting, @@ -85,15 +89,17 @@ public final class RIO { int first, int last, final boolean produce_log, - final boolean verbose ) throws IOException, SDIException, RIOException { + final boolean verbose, + final boolean transfer_taxonomy ) throws IOException, SDIException, RIOException { if ( ( last == DEFAULT_RANGE ) && ( first >= 0 ) ) { - last = gene_trees.length - 1; + last = END_OF_GT; } else if ( ( first == DEFAULT_RANGE ) && ( last >= 0 ) ) { first = 0; } removeSingleDescendentsNodes( species_tree, verbose ); - checkPreconditions( gene_trees, species_tree, rerooting, outgroup, first, last ); + p.reset(); + checkPreconditions( p, species_tree, rerooting, outgroup, first, last ); _produce_log = produce_log; _verbose = verbose; _rerooting = rerooting; @@ -104,24 +110,29 @@ public final class RIO { _analyzed_gene_trees = null; _removed_gene_tree_nodes = null; _duplications_stats = new BasicDescriptiveStatistics(); - inferOrthologs( gene_trees, species_tree, algorithm, outgroup, first, last ); + p.reset(); + inferOrthologs( p, species_tree, algorithm, outgroup, first, last, transfer_taxonomy ); + _species_tree = species_tree; } - private RIO( final IteratingPhylogenyParser p, + private RIO( final Phylogeny[] gene_trees, final Phylogeny species_tree, final ALGORITHM algorithm, final REROOTING rerooting, final String outgroup, int first, - final int last, + int last, final boolean produce_log, - final boolean verbose ) throws IOException, SDIException, RIOException { - if ( ( first == DEFAULT_RANGE ) && ( last >= 0 ) ) { + final boolean verbose, + final boolean transfer_taxonomy ) throws IOException, SDIException, RIOException { + if ( ( last == DEFAULT_RANGE ) && ( first >= 0 ) ) { + last = gene_trees.length - 1; + } + else if ( ( first == DEFAULT_RANGE ) && ( last >= 0 ) ) { first = 0; } removeSingleDescendentsNodes( species_tree, verbose ); - p.reset(); - checkPreconditions( p, species_tree, rerooting, outgroup, first, last ); + checkPreconditions( gene_trees, species_tree, rerooting, outgroup, first, last ); _produce_log = produce_log; _verbose = verbose; _rerooting = rerooting; @@ -132,8 +143,8 @@ public final class RIO { _analyzed_gene_trees = null; _removed_gene_tree_nodes = null; _duplications_stats = new BasicDescriptiveStatistics(); - p.reset(); - inferOrthologs( p, species_tree, algorithm, outgroup, first, last ); + inferOrthologs( gene_trees, species_tree, algorithm, outgroup, first, last, transfer_taxonomy ); + _species_tree = species_tree; } public final Phylogeny[] getAnalyzedGeneTrees() { @@ -154,6 +165,10 @@ public final class RIO { return _ext_nodes; } + public final TaxonomyComparisonBase getGSDIRtaxCompBase() { + return _gsdir_tax_comp_base; + } + /** * Returns the numbers of number of int nodes in gene trees analyzed (after * stripping). @@ -164,25 +179,116 @@ public final class RIO { return _int_nodes; } - public final TaxonomyComparisonBase getGSDIRtaxCompBase() { - return _gsdir_tax_comp_base; - } - public final StringBuilder getLog() { return _log; } + final public Phylogeny getMinDuplicationsGeneTree() { + return _min_dub_gene_tree; + } + + public final IntMatrix getOrthologTable() { + return _m; + } + public final List getRemovedGeneTreeNodes() { return _removed_gene_tree_nodes; } + public final Phylogeny getSpeciesTree() { + return _species_tree; + } + + private final void inferOrthologs( final IteratingPhylogenyParser parser, + final Phylogeny species_tree, + final ALGORITHM algorithm, + final String outgroup, + int first, + final int last, + final boolean transfer_taxonomy ) throws SDIException, RIOException, + FileNotFoundException, IOException { + if ( !parser.hasNext() ) { + throw new RIOException( "no gene trees to analyze" ); + } + if ( log() ) { + preLog( -1, species_tree, algorithm, outgroup ); + } + if ( _verbose ) { + System.out.println(); + } + final DecimalFormat pf = new java.text.DecimalFormat( "000" ); + int gene_tree_ext_nodes = 0; + int i = 0; + int counter = 0; + final boolean no_range = ( first < 0 ) || ( last < first ); + while ( parser.hasNext() ) { + final Phylogeny gt = parser.next(); + if ( no_range || ( ( i >= first ) && ( i <= last ) ) ) { + if ( gt.isEmpty() ) { + throw new RIOException( "gene tree #" + i + " is empty" ); + } + if ( gt.getNumberOfExternalNodes() == 1 ) { + throw new RIOException( "gene tree #" + i + " has only one external node" ); + } + if ( _verbose ) { + ForesterUtil.updateProgress( i, pf ); + } + if ( counter == 0 ) { + if ( algorithm == ALGORITHM.SDIR ) { + // Removes from species_tree all species not found in gene_tree. + PhylogenyMethods.taxonomyBasedDeletionOfExternalNodes( gt, species_tree ); + if ( species_tree.isEmpty() ) { + throw new RIOException( "failed to establish species based mapping between gene and species trees" ); + } + } + gene_tree_ext_nodes = gt.getNumberOfExternalNodes(); + } + else if ( gene_tree_ext_nodes != gt.getNumberOfExternalNodes() ) { + throw new RIOException( "gene tree #" + i + " has a different number of external nodes (" + + gt.getNumberOfExternalNodes() + ") than the preceding gene tree(s) (" + + gene_tree_ext_nodes + ")" ); + } + if ( algorithm == ALGORITHM.SDIR ) { + // Removes from gene_tree all species not found in species_tree. + PhylogenyMethods.taxonomyBasedDeletionOfExternalNodes( species_tree, gt ); + if ( gt.isEmpty() ) { + throw new RIOException( "failed to establish species based mapping between gene and species trees" ); + } + } + final Phylogeny analyzed_gt = performOrthologInference( gt, + species_tree, + algorithm, + outgroup, + counter, + transfer_taxonomy ); + RIO.calculateOrthologTable( analyzed_gt, true, counter ); + ++counter; + } + ++i; + } + if ( ( first >= 0 ) && ( counter == 0 ) && ( i > 0 ) ) { + throw new RIOException( "attempt to analyze first gene tree #" + first + " in a set of " + i ); + } + if ( no_range ) { + first = 0; + } + if ( log() ) { + postLog( species_tree, first, first + counter - 1 ); + } + if ( _verbose ) { + System.out.println(); + System.out.println(); + } + } + private final void inferOrthologs( final Phylogeny[] gene_trees, final Phylogeny species_tree, final ALGORITHM algorithm, final String outgroup, final int first, - final int last ) throws SDIException, RIOException, FileNotFoundException, - IOException { + final int last, + final boolean transfer_taxonomy ) throws SDIException, RIOException, + FileNotFoundException, IOException { if ( algorithm == ALGORITHM.SDIR ) { // Removes from species_tree all species not found in gene_tree. PhylogenyMethods.taxonomyBasedDeletionOfExternalNodes( gene_trees[ 0 ], species_tree ); @@ -202,7 +308,7 @@ public final class RIO { my_gene_trees = gene_trees; } if ( log() ) { - preLog( gene_trees.length, species_tree, algorithm, outgroup, first, last ); + preLog( gene_trees.length, species_tree, algorithm, outgroup ); } if ( _verbose && ( my_gene_trees.length >= 4 ) ) { System.out.println(); @@ -211,6 +317,12 @@ public final class RIO { int gene_tree_ext_nodes = 0; for( int i = 0; i < my_gene_trees.length; ++i ) { final Phylogeny gt = my_gene_trees[ i ]; + if ( gt.isEmpty() ) { + throw new RIOException( "gene tree #" + i + " is empty" ); + } + if ( gt.getNumberOfExternalNodes() == 1 ) { + throw new RIOException( "gene tree #" + i + " has only one external node" ); + } if ( _verbose && ( my_gene_trees.length > 4 ) ) { ForesterUtil.updateProgress( ( ( double ) i ) / my_gene_trees.length ); } @@ -218,8 +330,8 @@ public final class RIO { gene_tree_ext_nodes = gt.getNumberOfExternalNodes(); } else if ( gene_tree_ext_nodes != gt.getNumberOfExternalNodes() ) { - throw new RIOException( "gene tree #" + ( i + 1 ) + " has a different number of external nodes (" - + gt.getNumberOfExternalNodes() + ") than the preceding gene trees (" + gene_tree_ext_nodes + throw new RIOException( "gene tree #" + i + " has a different number of external nodes (" + + gt.getNumberOfExternalNodes() + ") than the preceding gene tree(s) (" + gene_tree_ext_nodes + ")" ); } if ( algorithm == ALGORITHM.SDIR ) { @@ -229,10 +341,15 @@ public final class RIO { throw new RIOException( "failed to establish species based mapping between gene and species trees" ); } } - _analyzed_gene_trees[ i ] = performOrthologInference( gt, species_tree, algorithm, outgroup, i ); + _analyzed_gene_trees[ i ] = performOrthologInference( gt, + species_tree, + algorithm, + outgroup, + i, + transfer_taxonomy ); } if ( log() ) { - postLog( species_tree ); + postLog( species_tree, first, last ); } if ( _verbose && ( my_gene_trees.length > 4 ) ) { System.out.println(); @@ -240,87 +357,6 @@ public final class RIO { } } - private final void inferOrthologs( final IteratingPhylogenyParser parser, - final Phylogeny species_tree, - final ALGORITHM algorithm, - final String outgroup, - final int first, - final int last ) throws SDIException, RIOException, FileNotFoundException, - IOException { - if ( !parser.hasNext() ) { - throw new RIOException( "no gene trees to analyze" ); - } - // final Phylogeny g0 = parser.next(); - // if ( algorithm == ALGORITHM.SDIR ) { - // // Removes from species_tree all species not found in gene_tree. - // PhylogenyMethods.taxonomyBasedDeletionOfExternalNodes( g0, species_tree ); - // if ( species_tree.isEmpty() ) { - // throw new RIOException( "failed to establish species based mapping between gene and species trees" ); - // } - // } - // final Phylogeny[] my_gene_trees; - // if ( ( first >= 0 ) && ( last >= first ) && ( last < gene_trees.length ) ) { - // my_gene_trees = new Phylogeny[ ( 1 + last ) - first ]; - // int c = 0; - // for( int i = first; i <= last; ++i ) { - // my_gene_trees[ c++ ] = gene_trees[ i ]; - // } - // } - // else { - // my_gene_trees = gene_trees; - // } - if ( log() ) { - preLog( -1, species_tree, algorithm, outgroup, first, last ); - } - // if ( _verbose && ( my_gene_trees.length >= 4 ) ) { - // System.out.println(); - // } - // _analyzed_gene_trees = new Phylogeny[ my_gene_trees.length ]; - int gene_tree_ext_nodes = 0; - int i = 0; - while ( parser.hasNext() ) { - //for( int i = 0; i < my_gene_trees.length; ++i ) { - final Phylogeny gt = parser.next(); - // final Phylogeny gt = my_gene_trees[ i ]; - //if ( _verbose && ( my_gene_trees.length > 4 ) ) { - // ForesterUtil.updateProgress( ( ( double ) i ) / my_gene_trees.length ); - //} - if ( i == 0 ) { - if ( algorithm == ALGORITHM.SDIR ) { - // Removes from species_tree all species not found in gene_tree. - PhylogenyMethods.taxonomyBasedDeletionOfExternalNodes( gt, species_tree ); - if ( species_tree.isEmpty() ) { - throw new RIOException( "failed to establish species based mapping between gene and species trees" ); - } - } - gene_tree_ext_nodes = gt.getNumberOfExternalNodes(); - } - else if ( gene_tree_ext_nodes != gt.getNumberOfExternalNodes() ) { - throw new RIOException( "gene tree #" + ( i + 1 ) + " has a different number of external nodes (" - + gt.getNumberOfExternalNodes() + ") than the preceding gene trees (" + gene_tree_ext_nodes - + ")" ); - } - if ( algorithm == ALGORITHM.SDIR ) { - // Removes from gene_tree all species not found in species_tree. - PhylogenyMethods.taxonomyBasedDeletionOfExternalNodes( species_tree, gt ); - if ( gt.isEmpty() ) { - throw new RIOException( "failed to establish species based mapping between gene and species trees" ); - } - } - // _analyzed_gene_trees[ i ] = performOrthologInference( gt, species_tree, algorithm, outgroup, i ); - final Phylogeny analyzed_gt = performOrthologInference( gt, species_tree, algorithm, outgroup, i ); - RIO.calculateOrthologTable( analyzed_gt, true, i ); - ++i; - } - if ( log() ) { - postLog( species_tree ); - } - // if ( _verbose && ( my_gene_trees.length > 4 ) ) { - // System.out.println(); - // System.out.println(); - // } - } - private final boolean log() { return _produce_log; } @@ -360,7 +396,9 @@ public final class RIO { final Phylogeny species_tree, final ALGORITHM algorithm, final String outgroup, - final int i ) throws SDIException, RIOException { + final int i, + final boolean transfer_taxonomy ) throws SDIException, + RIOException { final Phylogeny assigned_tree; switch ( algorithm ) { case SDIR: { @@ -368,7 +406,7 @@ public final class RIO { break; } case GSDIR: { - assigned_tree = performOrthologInferenceByGSDI( gene_tree, species_tree, outgroup, i ); + assigned_tree = performOrthologInferenceByGSDI( gene_tree, species_tree, outgroup, i, transfer_taxonomy ); break; } default: { @@ -380,9 +418,9 @@ public final class RIO { _int_nodes = assigned_tree.getNumberOfInternalNodes(); } else if ( _ext_nodes != assigned_tree.getNumberOfExternalNodes() ) { - throw new RIOException( "after stripping gene tree #" + ( i + 1 ) - + " has a different number of external nodes (" + assigned_tree.getNumberOfExternalNodes() - + ") than the preceding gene trees (" + _ext_nodes + ")" ); + throw new RIOException( "after stripping gene tree #" + i + " has a different number of external nodes (" + + assigned_tree.getNumberOfExternalNodes() + ") than the preceding gene tree(s) (" + _ext_nodes + + ")" ); } return assigned_tree; } @@ -390,29 +428,27 @@ public final class RIO { private final Phylogeny performOrthologInferenceByGSDI( final Phylogeny gene_tree, final Phylogeny species_tree, final String outgroup, - final int i ) throws SDIException, RIOException { + final int i, + final boolean transfer_taxonomy ) throws SDIException, + RIOException { final Phylogeny assigned_tree; + final int dups; if ( _rerooting == REROOTING.BY_ALGORITHM ) { - final GSDIR gsdir = new GSDIR( gene_tree, species_tree, true, i == 0 ); - final List assigned_trees = gsdir.getMinDuplicationsSumGeneTrees(); + final GSDIR gsdir = new GSDIR( gene_tree, species_tree, true, i == 0, transfer_taxonomy ); + assigned_tree = gsdir.getMinDuplicationsSumGeneTree(); if ( i == 0 ) { _removed_gene_tree_nodes = gsdir.getStrippedExternalGeneTreeNodes(); for( final PhylogenyNode r : _removed_gene_tree_nodes ) { if ( !r.getNodeData().isHasTaxonomy() ) { - throw new RIOException( "node with no (appropriate) taxonomic information found in gene tree #1: " - + r.toString() ); + throw new RIOException( "node with no (appropriate) taxonomic information found in gene tree #" + + i + ": " + r.toString() ); } } } - final List shortests = GSDIR.getIndexesOfShortestTree( assigned_trees ); - assigned_tree = assigned_trees.get( shortests.get( 0 ) ); - if ( log() ) { - writeStatsToLog( i, gsdir, shortests ); - } if ( i == 0 ) { _gsdir_tax_comp_base = gsdir.getTaxCompBase(); } - _duplications_stats.addValue( gsdir.getMinDuplicationsSum() ); + dups = gsdir.getMinDuplicationsSum(); } else { if ( _rerooting == REROOTING.MIDPOINT ) { @@ -422,20 +458,24 @@ public final class RIO { final PhylogenyNode n = gene_tree.getNode( outgroup ); gene_tree.reRoot( n ); } - final GSDI gsdi = new GSDI( gene_tree, species_tree, true, true, true ); + final GSDI gsdi = new GSDI( gene_tree, species_tree, true, true, true, transfer_taxonomy ); _removed_gene_tree_nodes = gsdi.getStrippedExternalGeneTreeNodes(); for( final PhylogenyNode r : _removed_gene_tree_nodes ) { if ( !r.getNodeData().isHasTaxonomy() ) { - throw new RIOException( "node with no (appropriate) taxonomic information found in gene tree #1: " - + r.toString() ); + throw new RIOException( "node with no (appropriate) taxonomic information found in gene tree #" + i + + ": " + r.toString() ); } } assigned_tree = gene_tree; if ( i == 0 ) { _gsdir_tax_comp_base = gsdi.getTaxCompBase(); } - _duplications_stats.addValue( gsdi.getDuplicationsSum() ); + dups = gsdi.getDuplicationsSum(); + } + if ( ( i == 0 ) || ( dups < _duplications_stats.getMin() ) ) { + _min_dub_gene_tree = assigned_tree; } + _duplications_stats.addValue( dups ); return assigned_tree; } @@ -445,9 +485,9 @@ public final class RIO { return sdir.infer( gene_tree, species_tree, false, true, true, true, 1 )[ 0 ]; } - private final void postLog( final Phylogeny species_tree ) { + private final void postLog( final Phylogeny species_tree, final int first, final int last ) { log( "" ); - if ( getRemovedGeneTreeNodes().size() > 0 ) { + if ( ( getRemovedGeneTreeNodes() != null ) && ( getRemovedGeneTreeNodes().size() > 0 ) ) { logRemovedGeneTreeNodes(); } log( "Species tree external nodes (after stripping) : " + species_tree.getNumberOfExternalNodes() ); @@ -455,6 +495,9 @@ public final class RIO { + PhylogenyMethods.countNumberOfPolytomies( species_tree ) ); log( "Taxonomy linking based on : " + getGSDIRtaxCompBase() ); final java.text.DecimalFormat df = new java.text.DecimalFormat( "0.#" ); + if ( ( first >= 0 ) && ( last >= 0 ) ) { + log( "Gene trees analyzed range : " + first + "-" + last ); + } log( "Gene trees analyzed : " + _duplications_stats.getN() ); log( "Mean number of duplications : " + df.format( _duplications_stats.arithmeticMean() ) + " (sd: " + df.format( _duplications_stats.sampleStandardDeviation() ) + ")" + " (" @@ -476,9 +519,7 @@ public final class RIO { private final void preLog( final int gene_trees, final Phylogeny species_tree, final ALGORITHM algorithm, - final String outgroup, - final int first, - final int last ) { + final String outgroup ) { if ( gene_trees > 0 ) { log( "Number of gene trees (total) : " + gene_trees ); } @@ -506,111 +547,6 @@ public final class RIO { } } log( "Re-rooting : " + rs ); - if ( ( first >= 0 ) || ( last >= 0 ) ) { - log( "Gene trees analyzed range : " + first + "-" + last ); - } - if ( _rerooting == REROOTING.BY_ALGORITHM ) { - writeLogSubHeader(); - } - } - - private final void writeLogSubHeader() { - _log.append( ForesterUtil.LINE_SEPARATOR ); - _log.append( "Some information about duplication numbers in gene trees:" ); - _log.append( ForesterUtil.LINE_SEPARATOR ); - _log.append( "#" ); - _log.append( "\t" ); - _log.append( "re-rootings with minimal number of duplications" ); - _log.append( "/" ); - _log.append( "total root placements" ); - _log.append( "\t" ); - _log.append( "duplications range" ); - _log.append( "\t" ); - _log.append( "mininal duplication re-rootings with shortest tree heigth" ); - _log.append( ForesterUtil.LINE_SEPARATOR ); - } - - private final void writeStatsToLog( final int i, final GSDIR gsdir, final List shortests ) { - final BasicDescriptiveStatistics stats = gsdir.getDuplicationsSumStats(); - _log.append( i ); - _log.append( "\t" ); - _log.append( gsdir.getMinDuplicationsSumGeneTrees().size() ); - _log.append( "/" ); - _log.append( stats.getN() ); - _log.append( "\t" ); - _log.append( ( int ) stats.getMin() ); - _log.append( "-" ); - _log.append( ( int ) stats.getMax() ); - _log.append( "\t" ); - _log.append( shortests.size() ); - _log.append( ForesterUtil.LINE_SEPARATOR ); - } - - public IntMatrix getOrthologTable() { - return _m; - } - - private final static void calculateOrthologTable( final Phylogeny g, final boolean sort, final int counter ) - throws RIOException { - final List labels = new ArrayList(); - final Set labels_set = new HashSet(); - if ( counter == 0 ) { - for( final PhylogenyNode n : g.getExternalNodes() ) { - final String label = obtainLabel( labels_set, n ); - labels_set.add( label ); - labels.add( label ); - } - if ( sort ) { - Collections.sort( labels ); - } - _m = new IntMatrix( labels ); - } - updateCounts( _m, counter, g ); - } - - private final static String obtainLabel( final Set labels_set, final PhylogenyNode n ) throws RIOException { - String label; - if ( n.getNodeData().isHasSequence() && !ForesterUtil.isEmpty( n.getNodeData().getSequence().getName() ) ) { - label = n.getNodeData().getSequence().getName(); - } - else if ( n.getNodeData().isHasSequence() && !ForesterUtil.isEmpty( n.getNodeData().getSequence().getSymbol() ) ) { - label = n.getNodeData().getSequence().getSymbol(); - } - else if ( !ForesterUtil.isEmpty( n.getName() ) ) { - label = n.getName(); - } - else { - throw new RIOException( "node " + n + " has no appropriate label" ); - } - if ( labels_set.contains( label ) ) { - throw new RIOException( "label " + label + " is not unique" ); - } - return label; - } - - private final static void updateCounts( final IntMatrix m, final int counter, final Phylogeny g ) - throws RIOException { - PhylogenyMethods.preOrderReId( g ); - final HashMap map = PhylogenyMethods.createNameToExtNodeMap( g ); - for( int x = 0; x < m.size(); ++x ) { - final String mx = m.getLabel( x ); - final PhylogenyNode nx = map.get( mx ); - if ( nx == null ) { - throw new RIOException( "node \"" + mx + "\" not present in gene tree #" + counter ); - } - String my; - PhylogenyNode ny; - for( int y = 0; y < m.size(); ++y ) { - my = m.getLabel( y ); - ny = map.get( my ); - if ( ny == null ) { - throw new RIOException( "node \"" + my + "\" not present in gene tree #" + counter ); - } - if ( !PhylogenyMethods.calculateLCAonTreeWithIdsInPreOrder( nx, ny ).isDuplication() ) { - m.inreaseByOne( x, y ); - } - } - } } public final static IntMatrix calculateOrthologTable( final Phylogeny[] analyzed_gene_trees, final boolean sort ) @@ -642,7 +578,9 @@ public final class RIO { final int first, final int last, final boolean produce_log, - final boolean verbose ) throws IOException, SDIException, RIOException { + final boolean verbose, + final boolean transfer_taxonomy ) throws IOException, SDIException, + RIOException { final Phylogeny[] gene_trees = parseGeneTrees( gene_trees_file ); if ( gene_trees.length < 1 ) { throw new RIOException( "\"" + gene_trees_file + "\" is devoid of appropriate gene trees" ); @@ -652,7 +590,16 @@ public final class RIO { false, true, TAXONOMY_EXTRACTION.NO ); - return new RIO( gene_trees, species_tree, algorithm, rerooting, outgroup, first, last, produce_log, verbose ); + return new RIO( gene_trees, + species_tree, + algorithm, + rerooting, + outgroup, + first, + last, + produce_log, + verbose, + transfer_taxonomy ); } public final static RIO executeAnalysis( final File gene_trees_file, @@ -661,7 +608,9 @@ public final class RIO { final REROOTING rerooting, final String outgroup, final boolean produce_log, - final boolean verbose ) throws IOException, SDIException, RIOException { + final boolean verbose, + final boolean transfer_taxonomy ) throws IOException, SDIException, + RIOException { return new RIO( parseGeneTrees( gene_trees_file ), species_tree, algorithm, @@ -670,7 +619,64 @@ public final class RIO { DEFAULT_RANGE, DEFAULT_RANGE, produce_log, - verbose ); + verbose, + transfer_taxonomy ); + } + + public final static RIO executeAnalysis( final File gene_trees_file, + final Phylogeny species_tree, + final ALGORITHM algorithm, + final REROOTING rerooting, + final String outgroup, + final int first, + final int last, + final boolean produce_log, + final boolean verbose, + final boolean transfer_taxonomy ) throws IOException, SDIException, + RIOException { + return new RIO( parseGeneTrees( gene_trees_file ), + species_tree, + algorithm, + rerooting, + outgroup, + first, + last, + produce_log, + verbose, + transfer_taxonomy ); + } + + public final static RIO executeAnalysis( final IteratingPhylogenyParser p, + final File species_tree_file, + final ALGORITHM algorithm, + final REROOTING rerooting, + final String outgroup, + final int first, + final int last, + final boolean produce_log, + final boolean verbose, + final boolean transfer_taxonomy ) throws IOException, SDIException, + RIOException { + final Phylogeny g0 = p.next(); + if ( ( g0 == null ) || g0.isEmpty() || ( g0.getNumberOfExternalNodes() < 2 ) ) { + throw new RIOException( "input file does not seem to contain any gene trees" ); + } + final Phylogeny species_tree = SDIutil.parseSpeciesTree( g0, + species_tree_file, + false, + true, + TAXONOMY_EXTRACTION.NO ); + p.reset(); + return new RIO( p, + species_tree, + algorithm, + rerooting, + outgroup, + first, + last, + produce_log, + verbose, + transfer_taxonomy ); } public final static RIO executeAnalysis( final IteratingPhylogenyParser p, @@ -679,7 +685,9 @@ public final class RIO { final REROOTING rerooting, final String outgroup, final boolean produce_log, - final boolean verbose ) throws IOException, SDIException, RIOException { + final boolean verbose, + final boolean transfer_taxonomy ) throws IOException, SDIException, + RIOException { return new RIO( p, species_tree, algorithm, @@ -688,10 +696,11 @@ public final class RIO { DEFAULT_RANGE, DEFAULT_RANGE, produce_log, - verbose ); + verbose, + transfer_taxonomy ); } - public final static RIO executeAnalysis( final File gene_trees_file, + public final static RIO executeAnalysis( final IteratingPhylogenyParser p, final Phylogeny species_tree, final ALGORITHM algorithm, final REROOTING rerooting, @@ -699,8 +708,10 @@ public final class RIO { final int first, final int last, final boolean produce_log, - final boolean verbose ) throws IOException, SDIException, RIOException { - return new RIO( parseGeneTrees( gene_trees_file ), + final boolean verbose, + final boolean transfer_taxonomy ) throws IOException, SDIException, + RIOException { + return new RIO( p, species_tree, algorithm, rerooting, @@ -708,7 +719,8 @@ public final class RIO { first, last, produce_log, - verbose ); + verbose, + transfer_taxonomy ); } public final static RIO executeAnalysis( final Phylogeny[] gene_trees, final Phylogeny species_tree ) @@ -721,6 +733,7 @@ public final class RIO { DEFAULT_RANGE, DEFAULT_RANGE, false, + false, false ); } @@ -730,7 +743,9 @@ public final class RIO { final REROOTING rerooting, final String outgroup, final boolean produce_log, - final boolean verbose ) throws IOException, SDIException, RIOException { + final boolean verbose, + final boolean transfer_taxonomy ) throws IOException, SDIException, + RIOException { return new RIO( gene_trees, species_tree, algorithm, @@ -739,7 +754,8 @@ public final class RIO { DEFAULT_RANGE, DEFAULT_RANGE, produce_log, - verbose ); + verbose, + transfer_taxonomy ); } public final static RIO executeAnalysis( final Phylogeny[] gene_trees, @@ -750,35 +766,59 @@ public final class RIO { final int first, final int last, final boolean produce_log, - final boolean verbose ) throws IOException, SDIException, RIOException { - return new RIO( gene_trees, species_tree, algorithm, rerooting, outgroup, first, last, produce_log, verbose ); + final boolean verbose, + final boolean transfer_taxonomy ) throws IOException, SDIException, + RIOException { + return new RIO( gene_trees, + species_tree, + algorithm, + rerooting, + outgroup, + first, + last, + produce_log, + verbose, + transfer_taxonomy ); } - public final static RIO executeAnalysis( final IteratingPhylogenyParser p, - final Phylogeny species_tree, - final ALGORITHM algorithm, - final REROOTING rerooting, - final String outgroup, - final int first, - final int last, - final boolean produce_log, - final boolean verbose ) throws IOException, SDIException, RIOException { - return new RIO( p, species_tree, algorithm, rerooting, outgroup, first, last, produce_log, verbose ); + private final static void calculateOrthologTable( final Phylogeny g, final boolean sort, final int counter ) + throws RIOException { + if ( counter == 0 ) { + final List labels = new ArrayList(); + final Set labels_set = new HashSet(); + for( final PhylogenyNode n : g.getExternalNodes() ) { + final String label = obtainLabel( labels_set, n ); + labels_set.add( label ); + labels.add( label ); + } + if ( sort ) { + Collections.sort( labels ); + } + _m = new IntMatrix( labels ); + } + updateCounts( _m, counter, g ); } - private final static void checkPreconditions( final Phylogeny[] gene_trees, + private final static void checkPreconditions( final IteratingPhylogenyParser p, final Phylogeny species_tree, final REROOTING rerooting, final String outgroup, final int first, - final int last ) throws RIOException { + final int last ) throws RIOException, IOException { + final Phylogeny g0 = p.next(); + if ( ( g0 == null ) || g0.isEmpty() ) { + throw new RIOException( "input file does not seem to contain any gene trees" ); + } + if ( g0.getNumberOfExternalNodes() < 2 ) { + throw new RIOException( "input file does not seem to contain any useable gene trees" ); + } if ( !species_tree.isRooted() ) { throw new RIOException( "species tree is not rooted" ); } if ( !( ( last == DEFAULT_RANGE ) && ( first == DEFAULT_RANGE ) ) - && ( ( last < first ) || ( last >= gene_trees.length ) || ( last < 0 ) || ( first < 0 ) ) ) { + && ( ( last < first ) || ( last < 0 ) || ( first < 0 ) ) ) { throw new RIOException( "attempt to set range (0-based) of gene to analyze to: from " + first + " to " - + last + " (out of " + gene_trees.length + ")" ); + + last ); } if ( ( rerooting == REROOTING.OUTGROUP ) && ForesterUtil.isEmpty( outgroup ) ) { throw new RIOException( "outgroup not set for midpoint rooting" ); @@ -786,13 +826,12 @@ public final class RIO { if ( ( rerooting != REROOTING.OUTGROUP ) && !ForesterUtil.isEmpty( outgroup ) ) { throw new RIOException( "outgroup only used for midpoint rooting" ); } - if ( ( rerooting == REROOTING.MIDPOINT ) - && ( PhylogenyMethods.calculateMaxDistanceToRoot( gene_trees[ 0 ] ) <= 0 ) ) { + if ( ( rerooting == REROOTING.MIDPOINT ) && ( PhylogenyMethods.calculateMaxDistanceToRoot( g0 ) <= 0 ) ) { throw new RIOException( "attempt to use midpoint rooting on gene trees which seem to have no (positive) branch lengths (cladograms)" ); } if ( rerooting == REROOTING.OUTGROUP ) { try { - gene_trees[ 0 ].getNode( outgroup ); + g0.getNode( outgroup ); } catch ( final IllegalArgumentException e ) { throw new RIOException( "cannot perform re-rooting by outgroup: " + e.getLocalizedMessage() ); @@ -800,23 +839,19 @@ public final class RIO { } } - private final static void checkPreconditions( final IteratingPhylogenyParser p, + private final static void checkPreconditions( final Phylogeny[] gene_trees, final Phylogeny species_tree, final REROOTING rerooting, final String outgroup, final int first, - final int last ) throws RIOException, IOException { - final Phylogeny g0 = p.next(); - if ( ( g0 == null ) || g0.isEmpty() || ( g0.getNumberOfExternalNodes() < 2 ) ) { - throw new RIOException( "input file does not seem to contain any gene trees" ); - } + final int last ) throws RIOException { if ( !species_tree.isRooted() ) { throw new RIOException( "species tree is not rooted" ); } if ( !( ( last == DEFAULT_RANGE ) && ( first == DEFAULT_RANGE ) ) - && ( ( last < first ) || ( last < 0 ) || ( first < 0 ) ) ) { + && ( ( last < first ) || ( last >= gene_trees.length ) || ( last < 0 ) || ( first < 0 ) ) ) { throw new RIOException( "attempt to set range (0-based) of gene to analyze to: from " + first + " to " - + last ); + + last + " (out of " + gene_trees.length + ")" ); } if ( ( rerooting == REROOTING.OUTGROUP ) && ForesterUtil.isEmpty( outgroup ) ) { throw new RIOException( "outgroup not set for midpoint rooting" ); @@ -824,12 +859,13 @@ public final class RIO { if ( ( rerooting != REROOTING.OUTGROUP ) && !ForesterUtil.isEmpty( outgroup ) ) { throw new RIOException( "outgroup only used for midpoint rooting" ); } - if ( ( rerooting == REROOTING.MIDPOINT ) && ( PhylogenyMethods.calculateMaxDistanceToRoot( g0 ) <= 0 ) ) { + if ( ( rerooting == REROOTING.MIDPOINT ) + && ( PhylogenyMethods.calculateMaxDistanceToRoot( gene_trees[ 0 ] ) <= 0 ) ) { throw new RIOException( "attempt to use midpoint rooting on gene trees which seem to have no (positive) branch lengths (cladograms)" ); } if ( rerooting == REROOTING.OUTGROUP ) { try { - g0.getNode( outgroup ); + gene_trees[ 0 ].getNode( outgroup ); } catch ( final IllegalArgumentException e ) { throw new RIOException( "cannot perform re-rooting by outgroup: " + e.getLocalizedMessage() ); @@ -837,6 +873,30 @@ public final class RIO { } } + private final static String obtainLabel( final Set labels_set, final PhylogenyNode n ) throws RIOException { + String label; + if ( n.getNodeData().isHasSequence() && !ForesterUtil.isEmpty( n.getNodeData().getSequence().getName() ) ) { + label = n.getNodeData().getSequence().getName(); + } + else if ( n.getNodeData().isHasSequence() && !ForesterUtil.isEmpty( n.getNodeData().getSequence().getSymbol() ) ) { + label = n.getNodeData().getSequence().getSymbol(); + } + else if ( n.getNodeData().isHasSequence() + && !ForesterUtil.isEmpty( n.getNodeData().getSequence().getGeneName() ) ) { + label = n.getNodeData().getSequence().getGeneName(); + } + else if ( !ForesterUtil.isEmpty( n.getName() ) ) { + label = n.getName(); + } + else { + throw new RIOException( "node " + n + " has no appropriate label" ); + } + if ( labels_set.contains( label ) ) { + throw new RIOException( "label " + label + " is not unique" ); + } + return label; + } + private final static Phylogeny[] parseGeneTrees( final File gene_trees_file ) throws FileNotFoundException, IOException { final PhylogenyFactory factory = ParserBasedPhylogenyFactory.getInstance(); @@ -845,13 +905,13 @@ public final class RIO { final NHXParser nhx = ( NHXParser ) p; nhx.setReplaceUnderscores( false ); nhx.setIgnoreQuotes( true ); - nhx.setTaxonomyExtraction( NHXParser.TAXONOMY_EXTRACTION.YES ); + nhx.setTaxonomyExtraction( TAXONOMY_EXTRACTION.AGGRESSIVE ); } else if ( p instanceof NexusPhylogeniesParser ) { final NexusPhylogeniesParser nex = ( NexusPhylogeniesParser ) p; nex.setReplaceUnderscores( false ); nex.setIgnoreQuotes( true ); - nex.setTaxonomyExtraction( TAXONOMY_EXTRACTION.YES ); + nex.setTaxonomyExtraction( TAXONOMY_EXTRACTION.AGGRESSIVE ); } return factory.create( gene_trees_file, p ); } @@ -867,6 +927,31 @@ public final class RIO { } } + private final static void updateCounts( final IntMatrix m, final int counter, final Phylogeny g ) + throws RIOException { + PhylogenyMethods.preOrderReId( g ); + final HashMap map = PhylogenyMethods.createNameToExtNodeMap( g ); + for( int x = 0; x < m.size(); ++x ) { + final String mx = m.getLabel( x ); + final PhylogenyNode nx = map.get( mx ); + if ( nx == null ) { + throw new RIOException( "node \"" + mx + "\" not present in gene tree #" + counter ); + } + String my; + PhylogenyNode ny; + for( int y = 0; y < m.size(); ++y ) { + my = m.getLabel( y ); + ny = map.get( my ); + if ( ny == null ) { + throw new RIOException( "node \"" + my + "\" not present in gene tree #" + counter ); + } + if ( !PhylogenyMethods.calculateLCAonTreeWithIdsInPreOrder( nx, ny ).isDuplication() ) { + m.inreaseByOne( x, y ); + } + } + } + } + public enum REROOTING { NONE, BY_ALGORITHM, MIDPOINT, OUTGROUP; }