1 // forester -- software libraries and applications
2 // for genomics and evolutionary biology research.
4 // Copyright (C) 2010 Christian M Zmasek
5 // Copyright (C) 2010 Sanford-Burnham Medical Research Institute
8 // This library is free software; you can redistribute it and/or
9 // modify it under the terms of the GNU Lesser General Public
10 // License as published by the Free Software Foundation; either
11 // version 2.1 of the License, or (at your option) any later version.
13 // This library is distributed in the hope that it will be useful,
14 // but WITHOUT ANY WARRANTY; without even the implied warranty of
15 // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
16 // Lesser General Public License for more details.
18 // You should have received a copy of the GNU Lesser General Public
19 // License along with this library; if not, write to the Free Software
20 // Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA
22 // Contact: phylosoft @ gmail . com
23 // WWW: www.phylosoft.org/forester
25 package org.forester.analysis;
27 import java.io.IOException;
28 import java.util.ArrayList;
29 import java.util.HashMap;
30 import java.util.List;
31 import java.util.SortedSet;
32 import java.util.TreeSet;
34 import org.forester.io.parsers.phyloxml.PhyloXmlDataFormatException;
35 import org.forester.phylogeny.Phylogeny;
36 import org.forester.phylogeny.PhylogenyNode;
37 import org.forester.phylogeny.data.Identifier;
38 import org.forester.phylogeny.data.Taxonomy;
39 import org.forester.phylogeny.iterators.PhylogenyNodeIterator;
40 import org.forester.util.ForesterUtil;
41 import org.forester.ws.uniprot.UniProtTaxonomy;
42 import org.forester.ws.uniprot.UniProtWsTools;
44 public final class AncestralTaxonomyInference {
46 private static final int MAX_CACHE_SIZE = 100000;
47 private static final int MAX_TAXONOMIES_TO_RETURN = 100;
48 private static final HashMap<String, UniProtTaxonomy> _sn_up_cache_map = new HashMap<String, UniProtTaxonomy>();
49 private static final HashMap<String, UniProtTaxonomy> _code_up_cache_map = new HashMap<String, UniProtTaxonomy>();
50 private static final HashMap<String, UniProtTaxonomy> _cn_up_cache_map = new HashMap<String, UniProtTaxonomy>();
51 private static final HashMap<String, UniProtTaxonomy> _id_up_cache_map = new HashMap<String, UniProtTaxonomy>();
53 synchronized private static void clearCachesIfTooLarge() {
54 if ( getSnTaxCacheMap().size() > MAX_CACHE_SIZE ) {
55 getSnTaxCacheMap().clear();
57 if ( getCnTaxCacheMap().size() > MAX_CACHE_SIZE ) {
58 getCnTaxCacheMap().clear();
60 if ( getCodeTaxCacheMap().size() > MAX_CACHE_SIZE ) {
61 getCodeTaxCacheMap().clear();
63 if ( getIdTaxCacheMap().size() > MAX_CACHE_SIZE ) {
64 getIdTaxCacheMap().clear();
68 synchronized private static HashMap<String, UniProtTaxonomy> getCnTaxCacheMap() {
69 return _cn_up_cache_map;
72 synchronized private static HashMap<String, UniProtTaxonomy> getCodeTaxCacheMap() {
73 return _code_up_cache_map;
76 synchronized private static HashMap<String, UniProtTaxonomy> getIdTaxCacheMap() {
77 return _id_up_cache_map;
80 synchronized private static HashMap<String, UniProtTaxonomy> getSnTaxCacheMap() {
81 return _sn_up_cache_map;
84 synchronized private static UniProtTaxonomy getTaxonomies( final HashMap<String, UniProtTaxonomy> cache,
86 final QUERY_TYPE qt ) throws IOException {
87 if ( cache.containsKey( query ) ) {
88 return cache.get( query ).copy();
91 List<UniProtTaxonomy> up_taxonomies = null;
94 up_taxonomies = getTaxonomiesFromId( query );
97 up_taxonomies = getTaxonomiesFromTaxonomyCode( query );
100 up_taxonomies = getTaxonomiesFromScientificName( query );
103 up_taxonomies = getTaxonomiesFromCommonName( query );
106 throw new RuntimeException();
108 if ( ( up_taxonomies != null ) && ( up_taxonomies.size() == 1 ) ) {
109 final UniProtTaxonomy up_tax = up_taxonomies.get( 0 );
110 if ( !ForesterUtil.isEmpty( up_tax.getScientificName() ) ) {
111 getSnTaxCacheMap().put( up_tax.getScientificName(), up_tax );
113 if ( !ForesterUtil.isEmpty( up_tax.getCode() ) ) {
114 getCodeTaxCacheMap().put( up_tax.getCode(), up_tax );
116 if ( !ForesterUtil.isEmpty( up_tax.getCommonName() ) ) {
117 getCnTaxCacheMap().put( up_tax.getCommonName(), up_tax );
119 if ( !ForesterUtil.isEmpty( up_tax.getId() ) ) {
120 getIdTaxCacheMap().put( up_tax.getId(), up_tax );
130 synchronized private static List<UniProtTaxonomy> getTaxonomiesFromCommonName( final String query )
132 return UniProtWsTools.getTaxonomiesFromCommonNameStrict( query, MAX_TAXONOMIES_TO_RETURN );
135 synchronized private static List<UniProtTaxonomy> getTaxonomiesFromId( final String query ) throws IOException {
136 return UniProtWsTools.getTaxonomiesFromId( query, MAX_TAXONOMIES_TO_RETURN );
139 synchronized private static List<UniProtTaxonomy> getTaxonomiesFromScientificName( final String query )
141 return UniProtWsTools.getTaxonomiesFromScientificNameStrict( query, MAX_TAXONOMIES_TO_RETURN );
144 synchronized private static List<UniProtTaxonomy> getTaxonomiesFromTaxonomyCode( final String query )
146 return UniProtWsTools.getTaxonomiesFromTaxonomyCode( query, MAX_TAXONOMIES_TO_RETURN );
149 synchronized public static SortedSet<String> inferTaxonomyFromDescendents( final Phylogeny phy ) throws IOException {
150 clearCachesIfTooLarge();
151 final SortedSet<String> not_found = new TreeSet<String>();
152 for( final PhylogenyNodeIterator iter = phy.iteratorPostorder(); iter.hasNext(); ) {
153 final PhylogenyNode node = iter.next();
154 // final QUERY_TYPE qt = null;
155 // Taxonomy tax = null;
156 // if ( node.getNodeData().isHasTaxonomy() ) {
157 // tax = node.getNodeData().getTaxonomy();
159 // UniProtTaxonomy up_tax = null;
160 // if ( ( tax != null )
161 // && ( isHasAppropriateId( tax ) || !ForesterUtil.isEmpty(
162 // tax.getScientificName() )
163 // || !ForesterUtil.isEmpty( tax.getTaxonomyCode() ) ||
164 // !ForesterUtil.isEmpty( tax
165 // .getCommonName() ) ) ) {
166 // final String query = null;
167 // up_tax = obtainUniProtTaxonomy( tax, query, qt );
168 // if ( up_tax == null ) {
169 // not_found.add( query );
172 // updateTaxonomy( qt, node, tax, up_tax );
175 if ( !node.isExternal() ) {
176 inferTaxonomyFromDescendents( node, not_found );
182 synchronized private static void inferTaxonomyFromDescendents( final PhylogenyNode n,
183 final SortedSet<String> not_found )
185 if ( n.isExternal() ) {
186 throw new IllegalArgumentException( "attempt to infer taxonomy from descendants of external node" );
188 n.getNodeData().setTaxonomy( null );
189 final List<PhylogenyNode> descs = n.getDescendants();
190 final List<String[]> lineages = new ArrayList<String[]>();
191 int shortest_lin_length = Integer.MAX_VALUE;
192 for( final PhylogenyNode desc : descs ) {
193 if ( desc.getNodeData().isHasTaxonomy()
194 && ( isHasAppropriateId( desc.getNodeData().getTaxonomy() )
195 || !ForesterUtil.isEmpty( desc.getNodeData().getTaxonomy().getScientificName() )
196 || !ForesterUtil.isEmpty( desc.getNodeData().getTaxonomy().getTaxonomyCode() ) || !ForesterUtil
197 .isEmpty( desc.getNodeData().getTaxonomy().getCommonName() ) ) ) {
198 final QUERY_TYPE qt = null;
199 final String query = null;
200 final UniProtTaxonomy up_tax = obtainUniProtTaxonomy( desc.getNodeData().getTaxonomy(), query, qt );
201 String[] lineage = null;
202 if ( up_tax != null ) {
203 lineage = obtainLineagePlusOwnScientificName( up_tax );
205 if ( ( lineage == null ) || ( lineage.length < 1 ) ) {
206 not_found.add( desc.getNodeData().getTaxonomy().asText().toString() );
209 if ( lineage.length < shortest_lin_length ) {
210 shortest_lin_length = lineage.length;
212 lineages.add( lineage );
215 String msg = "Node(s) with no or inappropriate taxonomic information found";
216 if ( !ForesterUtil.isEmpty( desc.getName() ) ) {
217 msg = "Node " + desc.getName() + " has no or inappropriate taxonomic information";
219 throw new IllegalArgumentException( msg );
222 String last_common_lineage = null;
223 if ( shortest_lin_length > 0 ) {
224 I: for( int i = 0; i < shortest_lin_length; ++i ) {
225 final String lineage_0 = lineages.get( 0 )[ i ];
226 for( int j = 1; j < lineages.size(); ++j ) {
227 if ( !lineage_0.equals( lineages.get( j )[ i ] ) ) {
231 last_common_lineage = lineage_0;
234 if ( last_common_lineage == null ) {
237 // if ( !n.getNodeData().isHasTaxonomy() ) {
238 // n.getNodeData().setTaxonomy( new Taxonomy() );
240 final Taxonomy tax = new Taxonomy();
241 n.getNodeData().setTaxonomy( tax );
242 tax.setScientificName( last_common_lineage );
243 final UniProtTaxonomy up_tax = obtainUniProtTaxonomyFromSn( last_common_lineage );
244 if ( up_tax != null ) {
245 if ( !ForesterUtil.isEmpty( up_tax.getRank() ) ) {
247 tax.setRank( up_tax.getRank().toLowerCase() );
249 catch ( final PhyloXmlDataFormatException ex ) {
253 if ( !ForesterUtil.isEmpty( up_tax.getId() ) ) {
254 tax.setIdentifier( new Identifier( up_tax.getId(), "uniprot" ) );
256 if ( !ForesterUtil.isEmpty( up_tax.getCommonName() ) ) {
257 tax.setCommonName( up_tax.getCommonName() );
259 if ( !ForesterUtil.isEmpty( up_tax.getSynonym() ) && !tax.getSynonyms().contains( up_tax.getSynonym() ) ) {
260 tax.getSynonyms().add( up_tax.getSynonym() );
263 for( final PhylogenyNode desc : descs ) {
264 if ( !desc.isExternal() && desc.getNodeData().isHasTaxonomy()
265 && desc.getNodeData().getTaxonomy().isEqual( tax ) ) {
266 desc.getNodeData().setTaxonomy( null );
271 synchronized private static boolean isHasAppropriateId( final Taxonomy tax ) {
272 return ( ( tax.getIdentifier() != null ) && ( !ForesterUtil.isEmpty( tax.getIdentifier().getValue() ) && ( tax
273 .getIdentifier().getProvider().equalsIgnoreCase( "ncbi" )
274 || tax.getIdentifier().getProvider().equalsIgnoreCase( "uniprot" ) || tax.getIdentifier().getProvider()
275 .equalsIgnoreCase( "uniprotkb" ) ) ) );
278 synchronized public static SortedSet<String> obtainDetailedTaxonomicInformation( final Phylogeny phy )
280 clearCachesIfTooLarge();
281 final SortedSet<String> not_found = new TreeSet<String>();
282 for( final PhylogenyNodeIterator iter = phy.iteratorPostorder(); iter.hasNext(); ) {
283 final PhylogenyNode node = iter.next();
284 final QUERY_TYPE qt = null;
286 if ( node.getNodeData().isHasTaxonomy() ) {
287 tax = node.getNodeData().getTaxonomy();
289 else if ( node.isExternal() ) {
290 if ( !ForesterUtil.isEmpty( node.getName() ) ) {
291 not_found.add( node.getName() );
294 not_found.add( node.toString() );
297 UniProtTaxonomy up_tax = null;
299 && ( isHasAppropriateId( tax ) || !ForesterUtil.isEmpty( tax.getScientificName() )
300 || !ForesterUtil.isEmpty( tax.getTaxonomyCode() ) || !ForesterUtil.isEmpty( tax
301 .getCommonName() ) ) ) {
302 up_tax = obtainUniProtTaxonomy( tax, null, qt );
303 if ( up_tax != null ) {
304 updateTaxonomy( qt, node, tax, up_tax );
307 not_found.add( tax.toString() );
314 synchronized private static String[] obtainLineagePlusOwnScientificName( final UniProtTaxonomy up_tax ) {
315 final String[] lineage = up_tax.getLineage();
316 final String[] lin_plus_self = new String[ lineage.length + 1 ];
317 for( int i = 0; i < lineage.length; ++i ) {
318 lin_plus_self[ i ] = lineage[ i ];
320 lin_plus_self[ lineage.length ] = up_tax.getScientificName();
321 return lin_plus_self;
324 synchronized private static UniProtTaxonomy obtainUniProtTaxonomy( final Taxonomy tax, String query, QUERY_TYPE qt )
326 if ( isHasAppropriateId( tax ) ) {
327 query = tax.getIdentifier().getValue();
329 return getTaxonomies( getIdTaxCacheMap(), query, qt );
331 else if ( !ForesterUtil.isEmpty( tax.getScientificName() ) ) {
332 query = tax.getScientificName();
334 return getTaxonomies( getSnTaxCacheMap(), query, qt );
336 else if ( !ForesterUtil.isEmpty( tax.getTaxonomyCode() ) ) {
337 query = tax.getTaxonomyCode();
338 qt = QUERY_TYPE.CODE;
339 return getTaxonomies( getCodeTaxCacheMap(), query, qt );
342 query = tax.getCommonName();
344 return getTaxonomies( getCnTaxCacheMap(), query, qt );
348 synchronized private static UniProtTaxonomy obtainUniProtTaxonomyFromSn( final String sn ) throws IOException {
349 UniProtTaxonomy up_tax = null;
350 if ( getSnTaxCacheMap().containsKey( sn ) ) {
351 up_tax = getSnTaxCacheMap().get( sn ).copy();
354 final List<UniProtTaxonomy> up_taxonomies = getTaxonomiesFromScientificName( sn );
355 if ( ( up_taxonomies != null ) && ( up_taxonomies.size() == 1 ) ) {
356 up_tax = up_taxonomies.get( 0 );
357 getSnTaxCacheMap().put( sn, up_tax );
358 if ( !ForesterUtil.isEmpty( up_tax.getCode() ) ) {
359 getCodeTaxCacheMap().put( up_tax.getCode(), up_tax );
361 if ( !ForesterUtil.isEmpty( up_tax.getCommonName() ) ) {
362 getCnTaxCacheMap().put( up_tax.getCommonName(), up_tax );
364 if ( !ForesterUtil.isEmpty( up_tax.getId() ) ) {
365 getIdTaxCacheMap().put( up_tax.getId(), up_tax );
372 synchronized private static void updateTaxonomy( final QUERY_TYPE qt,
373 final PhylogenyNode node,
375 final UniProtTaxonomy up_tax ) {
376 if ( ( qt != QUERY_TYPE.SN ) && !ForesterUtil.isEmpty( up_tax.getScientificName() )
377 && ForesterUtil.isEmpty( tax.getScientificName() ) ) {
378 tax.setScientificName( up_tax.getScientificName() );
380 if ( node.isExternal()
381 && ( ( qt != QUERY_TYPE.CODE ) && !ForesterUtil.isEmpty( up_tax.getCode() ) && ForesterUtil
382 .isEmpty( tax.getTaxonomyCode() ) ) ) {
383 tax.setTaxonomyCode( up_tax.getCode() );
385 if ( ( qt != QUERY_TYPE.CN ) && !ForesterUtil.isEmpty( up_tax.getCommonName() )
386 && ForesterUtil.isEmpty( tax.getCommonName() ) ) {
387 tax.setCommonName( up_tax.getCommonName() );
389 if ( !ForesterUtil.isEmpty( up_tax.getSynonym() ) && !tax.getSynonyms().contains( up_tax.getSynonym() ) ) {
390 tax.getSynonyms().add( up_tax.getSynonym() );
392 if ( !ForesterUtil.isEmpty( up_tax.getRank() ) && ForesterUtil.isEmpty( tax.getRank() ) ) {
394 tax.setRank( up_tax.getRank().toLowerCase() );
396 catch ( final PhyloXmlDataFormatException ex ) {
400 if ( ( qt != QUERY_TYPE.ID ) && !ForesterUtil.isEmpty( up_tax.getId() ) && ( tax.getIdentifier() == null ) ) {
401 tax.setIdentifier( new Identifier( up_tax.getId(), "uniprot" ) );
405 private enum QUERY_TYPE {