2 // FORESTER -- software libraries and applications
3 // for evolutionary biology research and applications.
5 // Copyright (C) 2008-2009 Christian M. Zmasek
6 // Copyright (C) 2008-2009 Burnham Institute for Medical Research
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.
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.
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
23 // Contact: phylosoft @ gmail . com
24 // WWW: www.phylosoft.org/forester
26 package org.forester.sdi;
28 import java.util.ArrayList;
29 import java.util.HashMap;
30 import java.util.HashSet;
31 import java.util.List;
34 import java.util.SortedSet;
35 import java.util.TreeSet;
37 import org.forester.phylogeny.Phylogeny;
38 import org.forester.phylogeny.PhylogenyMethods;
39 import org.forester.phylogeny.PhylogenyNode;
40 import org.forester.phylogeny.data.Event;
41 import org.forester.phylogeny.data.Taxonomy;
42 import org.forester.phylogeny.iterators.PhylogenyNodeIterator;
43 import org.forester.util.ForesterUtil;
46 * Implements our algorithm for speciation - duplication inference (SDI). <p>
47 * The initialization is accomplished by: </p> <ul> <li>method
48 * "linkExtNodesOfG()" of class SDI: setting the links for the external nodes of
49 * the gene tree <li>"preorderReID(int)" from class Phylogeny: numbering of
50 * nodes of the species tree in preorder <li>the optional stripping of the
51 * species tree is accomplished by method "stripTree(Phylogeny,Phylogeny)" of
52 * class Phylogeny </ul> <p> The recursion part is accomplished by this class'
53 * method "geneTreePostOrderTraversal(PhylogenyNode)". <p> Requires JDK 1.5 or
56 * @see SDI#linkNodesOfG()
58 * @see Phylogeny#preorderReID(int)
61 * PhylogenyMethods#taxonomyBasedDeletionOfExternalNodes(Phylogeny,Phylogeny)
63 * @see #geneTreePostOrderTraversal(PhylogenyNode)
65 * @author Christian M. Zmasek
67 public final class GSDI extends SDI {
69 private final boolean _most_parsimonious_duplication_model;
70 private final boolean _strip_gene_tree;
71 private final boolean _strip_species_tree;
72 private int _speciation_or_duplication_events_sum;
73 private int _speciations_sum;
74 private final List<PhylogenyNode> _stripped_gene_tree_nodes;
75 private final List<PhylogenyNode> _stripped_species_tree_nodes;
76 private final Set<PhylogenyNode> _mapped_species_tree_nodes;
77 private TaxonomyComparisonBase _tax_comp_base;
78 private final SortedSet<String> _scientific_names_mapped_to_reduced_specificity;
80 public GSDI( final Phylogeny gene_tree,
81 final Phylogeny species_tree,
82 final boolean most_parsimonious_duplication_model,
83 final boolean strip_gene_tree,
84 final boolean strip_species_tree ) throws SDIException {
85 super( gene_tree, species_tree );
86 _speciation_or_duplication_events_sum = 0;
88 _most_parsimonious_duplication_model = most_parsimonious_duplication_model;
89 _duplications_sum = 0;
90 _strip_gene_tree = strip_gene_tree;
91 _strip_species_tree = strip_species_tree;
92 _stripped_gene_tree_nodes = new ArrayList<PhylogenyNode>();
93 _stripped_species_tree_nodes = new ArrayList<PhylogenyNode>();
94 _mapped_species_tree_nodes = new HashSet<PhylogenyNode>();
95 _scientific_names_mapped_to_reduced_specificity = new TreeSet<String>();
97 PhylogenyMethods.preOrderReId( getSpeciesTree() );
98 geneTreePostOrderTraversal();
101 GSDI( final Phylogeny gene_tree, final Phylogeny species_tree, final boolean most_parsimonious_duplication_model )
102 throws SDIException {
103 this( gene_tree, species_tree, most_parsimonious_duplication_model, false, false );
106 // s is the node on the species tree g maps to.
107 private final void determineEvent( final PhylogenyNode s, final PhylogenyNode g ) {
108 boolean oyako = false;
109 if ( ( g.getChildNode1().getLink() == s ) || ( g.getChildNode2().getLink() == s ) ) {
112 if ( g.getLink().getNumberOfDescendants() == 2 ) {
114 g.getNodeData().setEvent( createDuplicationEvent() );
117 g.getNodeData().setEvent( createSpeciationEvent() );
122 final Set<PhylogenyNode> set = new HashSet<PhylogenyNode>();
123 for( PhylogenyNode n : g.getChildNode1().getAllExternalDescendants() ) {
125 while ( n.getParent() != s ) {
133 boolean multiple = false;
134 for( PhylogenyNode n : g.getChildNode2().getAllExternalDescendants() ) {
136 while ( n.getParent() != s ) {
142 if ( set.contains( n ) ) {
148 g.getNodeData().setEvent( createDuplicationEvent() );
151 if ( _most_parsimonious_duplication_model ) {
152 g.getNodeData().setEvent( createSpeciationEvent() );
155 g.getNodeData().setEvent( createSingleSpeciationOrDuplicationEvent() );
160 g.getNodeData().setEvent( createSpeciationEvent() );
166 * Traverses the subtree of PhylogenyNode g in postorder, calculating the
167 * mapping function M, and determines which nodes represent speciation
168 * events and which ones duplication events.
170 * Preconditions: Mapping M for external nodes must have been calculated and
171 * the species tree must be labeled in preorder.
175 final void geneTreePostOrderTraversal() {
176 for( final PhylogenyNodeIterator it = getGeneTree().iteratorPostorder(); it.hasNext(); ) {
177 final PhylogenyNode g = it.next();
178 if ( g.isInternal() ) {
179 PhylogenyNode s1 = g.getChildNode1().getLink();
180 PhylogenyNode s2 = g.getChildNode2().getLink();
182 if ( s1.getId() > s2.getId() ) {
190 determineEvent( s1, g );
195 private final Event createDuplicationEvent() {
196 final Event event = Event.createSingleDuplicationEvent();
201 private final Event createSingleSpeciationOrDuplicationEvent() {
202 final Event event = Event.createSingleSpeciationOrDuplicationEvent();
203 ++_speciation_or_duplication_events_sum;
207 private final Event createSpeciationEvent() {
208 final Event event = Event.createSingleSpeciationEvent();
213 public final int getSpeciationOrDuplicationEventsSum() {
214 return _speciation_or_duplication_events_sum;
217 public final int getSpeciationsSum() {
218 return _speciations_sum;
222 * This allows for linking of internal nodes of the species tree (as opposed
223 * to just external nodes, as in the method it overrides.
224 * @throws SDIException
228 final void linkNodesOfG() throws SDIException {
229 final Map<String, PhylogenyNode> species_to_node_map = new HashMap<String, PhylogenyNode>();
230 final List<PhylogenyNode> species_tree_ext_nodes = new ArrayList<PhylogenyNode>();
231 _tax_comp_base = determineTaxonomyComparisonBase( _gene_tree );
232 // Stringyfied taxonomy is the key, node is the value.
233 for( final PhylogenyNodeIterator iter = _species_tree.iteratorExternalForward(); iter.hasNext(); ) {
234 final PhylogenyNode s = iter.next();
235 species_tree_ext_nodes.add( s );
236 if ( s.getNodeData().isHasTaxonomy() ) {
237 final String tax_str = taxonomyToString( s, _tax_comp_base );
238 if ( !ForesterUtil.isEmpty( tax_str ) ) {
239 if ( species_to_node_map.containsKey( tax_str ) ) {
240 throw new SDIException( "taxonomy \"" + s + "\" is not unique in species tree" );
242 species_to_node_map.put( tax_str, s );
246 // Retrieve the reference to the node with a matching stringyfied taxonomy.
247 for( final PhylogenyNodeIterator iter = _gene_tree.iteratorExternalForward(); iter.hasNext(); ) {
248 final PhylogenyNode g = iter.next();
249 if ( !g.getNodeData().isHasTaxonomy() ) {
250 if ( _strip_gene_tree ) {
251 _stripped_gene_tree_nodes.add( g );
254 throw new SDIException( "gene tree node \"" + g + "\" has no taxonomic data" );
258 final String tax_str = taxonomyToString( g, _tax_comp_base );
259 if ( ForesterUtil.isEmpty( tax_str ) ) {
260 if ( _strip_gene_tree ) {
261 _stripped_gene_tree_nodes.add( g );
264 throw new SDIException( "gene tree node \"" + g + "\" has no appropriate taxonomic data" );
268 PhylogenyNode s = species_to_node_map.get( tax_str );
269 if ( ( _tax_comp_base == TaxonomyComparisonBase.SCIENTIFIC_NAME ) && ( s == null )
270 && ( ForesterUtil.countChars( tax_str, ' ' ) > 1 ) ) {
271 s = tryMapByRemovingOverlySpecificData( species_to_node_map, tax_str );
274 if ( _strip_gene_tree ) {
275 _stripped_gene_tree_nodes.add( g );
278 throw new SDIException( "taxonomy \"" + g.getNodeData().getTaxonomy()
279 + "\" not present in species tree" );
284 _mapped_species_tree_nodes.add( s );
289 if ( _strip_gene_tree ) {
291 if ( getGeneTree().isEmpty() || ( getGeneTree().getNumberOfExternalNodes() < 2 ) ) {
292 throw new SDIException( "species could not be mapped between gene tree and species tree" );
295 if ( _strip_species_tree ) {
296 stripSpeciesTree( species_tree_ext_nodes );
300 private final PhylogenyNode tryMapByRemovingOverlySpecificData( final Map<String, PhylogenyNode> species_to_node_map,
301 final String tax_str ) {
302 PhylogenyNode s = tryMapByRemovingOverlySpecificData( species_to_node_map, tax_str, " (" );
304 if ( ForesterUtil.countChars( tax_str, ' ' ) == 2 ) {
305 final String new_tax_str = tax_str.substring( 0, tax_str.lastIndexOf( ' ' ) ).trim();
306 s = species_to_node_map.get( new_tax_str );
308 addScientificNamesMappedToReducedSpecificity( tax_str, new_tax_str );
313 for( final String t : new String[] { " subspecies ", " strain ", " variety ", " varietas ", " subvariety ",
314 " form ", " subform ", " cultivar ", " section ", " subsection " } ) {
315 s = tryMapByRemovingOverlySpecificData( species_to_node_map, tax_str, t );
324 private final PhylogenyNode tryMapByRemovingOverlySpecificData( final Map<String, PhylogenyNode> species_to_node_map,
325 final String tax_str,
326 final String term ) {
327 final int i = tax_str.indexOf( term );
329 final String new_tax_str = tax_str.substring( 0, i ).trim();
330 final PhylogenyNode s = species_to_node_map.get( new_tax_str );
332 addScientificNamesMappedToReducedSpecificity( tax_str, new_tax_str );
339 private final void addScientificNamesMappedToReducedSpecificity( final String s1, final String s2 ) {
340 _scientific_names_mapped_to_reduced_specificity.add( s1 + " -> " + s2 );
343 public final SortedSet<String> getReMappedScientificNamesFromGeneTree() {
344 return _scientific_names_mapped_to_reduced_specificity;
347 public TaxonomyComparisonBase getTaxCompBase() {
348 return _tax_comp_base;
351 private void stripSpeciesTree( final List<PhylogenyNode> species_tree_ext_nodes ) {
352 for( final PhylogenyNode s : species_tree_ext_nodes ) {
353 if ( !_mapped_species_tree_nodes.contains( s ) ) {
354 _species_tree.deleteSubtree( s, true );
355 _stripped_species_tree_nodes.add( s );
358 _species_tree.clearHashIdToNodeMap();
359 _species_tree.externalNodesHaveChanged();
362 public List<PhylogenyNode> getStrippedSpeciesTreeNodes() {
363 return _stripped_species_tree_nodes;
366 private void stripGeneTree() {
367 for( final PhylogenyNode g : _stripped_gene_tree_nodes ) {
368 _gene_tree.deleteSubtree( g, true );
370 _gene_tree.clearHashIdToNodeMap();
371 _gene_tree.externalNodesHaveChanged();
374 public Set<PhylogenyNode> getMappedExternalSpeciesTreeNodes() {
375 return _mapped_species_tree_nodes;
378 public static TaxonomyComparisonBase determineTaxonomyComparisonBase( final Phylogeny gene_tree ) {
379 int with_id_count = 0;
380 int with_code_count = 0;
381 int with_sn_count = 0;
383 for( final PhylogenyNodeIterator iter = gene_tree.iteratorExternalForward(); iter.hasNext(); ) {
384 final PhylogenyNode g = iter.next();
385 if ( g.getNodeData().isHasTaxonomy() ) {
386 final Taxonomy tax = g.getNodeData().getTaxonomy();
387 if ( ( tax.getIdentifier() != null ) && !ForesterUtil.isEmpty( tax.getIdentifier().getValue() ) ) {
388 if ( ++with_id_count > max ) {
392 if ( !ForesterUtil.isEmpty( tax.getTaxonomyCode() ) ) {
393 if ( ++with_code_count > max ) {
394 max = with_code_count;
397 if ( !ForesterUtil.isEmpty( tax.getScientificName() ) ) {
398 if ( ++with_sn_count > max ) {
405 throw new IllegalArgumentException( "gene tree has no taxonomic data" );
407 else if ( max == 1 ) {
408 throw new IllegalArgumentException( "gene tree has only one node with taxonomic data" );
410 else if ( max == with_id_count ) {
411 return SDI.TaxonomyComparisonBase.ID;
413 else if ( max == with_sn_count ) {
414 return SDI.TaxonomyComparisonBase.SCIENTIFIC_NAME;
417 return SDI.TaxonomyComparisonBase.CODE;
421 public List<PhylogenyNode> getStrippedExternalGeneTreeNodes() {
422 return _stripped_gene_tree_nodes;
426 public final String toString() {
427 final StringBuffer sb = new StringBuffer();
428 sb.append( "Most parsimonious duplication model: " + _most_parsimonious_duplication_model );
429 sb.append( ForesterUtil.getLineSeparator() );
430 sb.append( "Speciations sum : " + getSpeciationsSum() );
431 sb.append( ForesterUtil.getLineSeparator() );
432 sb.append( "Duplications sum : " + getDuplicationsSum() );
433 sb.append( ForesterUtil.getLineSeparator() );
434 if ( !_most_parsimonious_duplication_model ) {
435 sb.append( "Speciation or duplications sum : " + getSpeciationOrDuplicationEventsSum() );
436 sb.append( ForesterUtil.getLineSeparator() );
438 sb.append( "mapping cost L : " + computeMappingCostL() );
439 return sb.toString();