3 // FORESTER -- software libraries and applications
4 // for evolutionary biology research and applications.
6 // Copyright (C) 2008-2009 Christian M. Zmasek
7 // Copyright (C) 2008-2009 Burnham Institute for Medical Research
10 // This library is free software; you can redistribute it and/or
11 // modify it under the terms of the GNU Lesser General Public
12 // License as published by the Free Software Foundation; either
13 // version 2.1 of the License, or (at your option) any later version.
15 // This library is distributed in the hope that it will be useful,
16 // but WITHOUT ANY WARRANTY; without even the implied warranty of
17 // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
18 // Lesser General Public License for more details.
20 // You should have received a copy of the GNU Lesser General Public
21 // License along with this library; if not, write to the Free Software
22 // Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA
24 // Contact: phylosoft @ gmail . com
25 // WWW: www.phylosoft.org/forester
27 package org.forester.surfacing;
29 import java.io.BufferedWriter;
31 import java.io.FileWriter;
32 import java.io.IOException;
33 import java.io.Writer;
34 import java.text.DecimalFormat;
35 import java.text.NumberFormat;
36 import java.util.ArrayList;
37 import java.util.Arrays;
38 import java.util.Collections;
39 import java.util.Comparator;
40 import java.util.HashMap;
41 import java.util.HashSet;
42 import java.util.List;
44 import java.util.PriorityQueue;
46 import java.util.SortedMap;
47 import java.util.SortedSet;
48 import java.util.TreeMap;
49 import java.util.TreeSet;
50 import java.util.regex.Matcher;
51 import java.util.regex.Pattern;
53 import org.forester.application.surfacing;
54 import org.forester.evoinference.distance.NeighborJoining;
55 import org.forester.evoinference.matrix.character.BasicCharacterStateMatrix;
56 import org.forester.evoinference.matrix.character.CharacterStateMatrix;
57 import org.forester.evoinference.matrix.character.CharacterStateMatrix.BinaryStates;
58 import org.forester.evoinference.matrix.character.CharacterStateMatrix.Format;
59 import org.forester.evoinference.matrix.character.CharacterStateMatrix.GainLossStates;
60 import org.forester.evoinference.matrix.distance.DistanceMatrix;
61 import org.forester.go.GoId;
62 import org.forester.go.GoNameSpace;
63 import org.forester.go.GoTerm;
64 import org.forester.go.PfamToGoMapping;
65 import org.forester.io.parsers.nexus.NexusConstants;
66 import org.forester.io.writers.PhylogenyWriter;
67 import org.forester.phylogeny.Phylogeny;
68 import org.forester.phylogeny.PhylogenyMethods;
69 import org.forester.phylogeny.PhylogenyNode;
70 import org.forester.phylogeny.PhylogenyNodeI.NH_CONVERSION_SUPPORT_VALUE_STYLE;
71 import org.forester.phylogeny.data.BinaryCharacters;
72 import org.forester.phylogeny.data.Confidence;
73 import org.forester.phylogeny.iterators.PhylogenyNodeIterator;
74 import org.forester.surfacing.DomainSimilarityCalculator.Detailedness;
75 import org.forester.surfacing.DomainSimilarityCalculator.GoAnnotationOutput;
76 import org.forester.surfacing.GenomeWideCombinableDomains.GenomeWideCombinableDomainsSortOrder;
77 import org.forester.util.AsciiHistogram;
78 import org.forester.util.BasicDescriptiveStatistics;
79 import org.forester.util.BasicTable;
80 import org.forester.util.BasicTableParser;
81 import org.forester.util.DescriptiveStatistics;
82 import org.forester.util.ForesterUtil;
84 public final class SurfacingUtil {
86 private final static NumberFormat FORMATTER = new DecimalFormat( "0.0E0" );
87 private final static NumberFormat FORMATTER_3 = new DecimalFormat( "0.000" );
88 private static final Comparator<Domain> ASCENDING_CONFIDENCE_VALUE_ORDER = new Comparator<Domain>() {
91 public int compare( final Domain d1,
93 if ( d1.getPerSequenceEvalue() < d2
94 .getPerSequenceEvalue() ) {
98 .getPerSequenceEvalue() > d2
99 .getPerSequenceEvalue() ) {
103 return d1.compareTo( d2 );
107 public final static Pattern PATTERN_SP_STYLE_TAXONOMY = Pattern.compile( "^[A-Z0-9]{3,5}$" );
108 private static final boolean USE_LAST = true;
110 private SurfacingUtil() {
111 // Hidden constructor.
114 public static void addAllBinaryDomainCombinationToSet( final GenomeWideCombinableDomains genome,
115 final SortedSet<BinaryDomainCombination> binary_domain_combinations ) {
116 final SortedMap<DomainId, CombinableDomains> all_cd = genome.getAllCombinableDomainsIds();
117 for( final DomainId domain_id : all_cd.keySet() ) {
118 binary_domain_combinations.addAll( all_cd.get( domain_id ).toBinaryDomainCombinations() );
122 public static void addAllDomainIdsToSet( final GenomeWideCombinableDomains genome,
123 final SortedSet<DomainId> domain_ids ) {
124 final SortedSet<DomainId> domains = genome.getAllDomainIds();
125 for( final DomainId domain : domains ) {
126 domain_ids.add( domain );
130 public static void addHtmlHead( final Writer w, final String title ) throws IOException {
131 w.write( SurfacingConstants.NL );
133 w.write( "<title>" );
135 w.write( "</title>" );
136 w.write( SurfacingConstants.NL );
137 w.write( "<style>" );
138 w.write( SurfacingConstants.NL );
139 w.write( "a:visited { color : #6633FF; text-decoration : none; }" );
140 w.write( SurfacingConstants.NL );
141 w.write( "a:link { color : #6633FF; text-decoration : none; }" );
142 w.write( SurfacingConstants.NL );
143 w.write( "a:active { color : #99FF00; text-decoration : none; }" );
144 w.write( SurfacingConstants.NL );
145 w.write( "a:hover { color : #FFFFFF; background-color : #99FF00; text-decoration : none; }" );
146 w.write( SurfacingConstants.NL );
147 w.write( "td { text-align: left; vertical-align: top; font-family: Verdana, Arial, Helvetica; font-size: 8pt}" );
148 w.write( SurfacingConstants.NL );
149 w.write( "h1 { color : #0000FF; font-family: Verdana, Arial, Helvetica; font-size: 18pt; font-weight: bold }" );
150 w.write( SurfacingConstants.NL );
151 w.write( "h2 { color : #0000FF; font-family: Verdana, Arial, Helvetica; font-size: 16pt; font-weight: bold }" );
152 w.write( SurfacingConstants.NL );
153 w.write( "</style>" );
154 w.write( SurfacingConstants.NL );
155 w.write( "</head>" );
156 w.write( SurfacingConstants.NL );
159 public static DescriptiveStatistics calculateDescriptiveStatisticsForMeanValues( final Set<DomainSimilarity> similarities ) {
160 final DescriptiveStatistics stats = new BasicDescriptiveStatistics();
161 for( final DomainSimilarity similarity : similarities ) {
162 stats.addValue( similarity.getMeanSimilarityScore() );
167 private static void calculateIndependentDomainCombinationGains( final Phylogeny local_phylogeny_l,
168 final String outfilename_for_counts,
169 final String outfilename_for_dc,
170 final String outfilename_for_dc_for_go_mapping,
171 final String outfilename_for_dc_for_go_mapping_unique,
172 final String outfilename_for_rank_counts,
173 final String outfilename_for_ancestor_species_counts ) {
175 final BufferedWriter out_counts = new BufferedWriter( new FileWriter( outfilename_for_counts ) );
176 final BufferedWriter out_dc = new BufferedWriter( new FileWriter( outfilename_for_dc ) );
177 final BufferedWriter out_dc_for_go_mapping = new BufferedWriter( new FileWriter( outfilename_for_dc_for_go_mapping ) );
178 final BufferedWriter out_dc_for_go_mapping_unique = new BufferedWriter( new FileWriter( outfilename_for_dc_for_go_mapping_unique ) );
179 final SortedMap<String, Integer> dc_gain_counts = new TreeMap<String, Integer>();
180 for( final PhylogenyNodeIterator it = local_phylogeny_l.iteratorPostorder(); it.hasNext(); ) {
181 final PhylogenyNode n = it.next();
182 final Set<String> gained_dc = n.getNodeData().getBinaryCharacters().getGainedCharacters();
183 for( final String dc : gained_dc ) {
184 if ( dc_gain_counts.containsKey( dc ) ) {
185 dc_gain_counts.put( dc, dc_gain_counts.get( dc ) + 1 );
188 dc_gain_counts.put( dc, 1 );
192 final SortedMap<Integer, Integer> histogram = new TreeMap<Integer, Integer>();
193 final SortedMap<Integer, StringBuilder> domain_lists = new TreeMap<Integer, StringBuilder>();
194 final SortedMap<Integer, PriorityQueue<String>> domain_lists_go = new TreeMap<Integer, PriorityQueue<String>>();
195 final SortedMap<Integer, SortedSet<String>> domain_lists_go_unique = new TreeMap<Integer, SortedSet<String>>();
196 final Set<String> dcs = dc_gain_counts.keySet();
197 final SortedSet<String> more_than_once = new TreeSet<String>();
198 for( final String dc : dcs ) {
199 final int count = dc_gain_counts.get( dc );
200 if ( histogram.containsKey( count ) ) {
201 histogram.put( count, histogram.get( count ) + 1 );
202 domain_lists.get( count ).append( ", " + dc );
203 domain_lists_go.get( count ).addAll( splitDomainCombination( dc ) );
204 domain_lists_go_unique.get( count ).addAll( splitDomainCombination( dc ) );
207 histogram.put( count, 1 );
208 domain_lists.put( count, new StringBuilder( dc ) );
209 final PriorityQueue<String> q = new PriorityQueue<String>();
210 q.addAll( splitDomainCombination( dc ) );
211 domain_lists_go.put( count, q );
212 final SortedSet<String> set = new TreeSet<String>();
213 set.addAll( splitDomainCombination( dc ) );
214 domain_lists_go_unique.put( count, set );
217 more_than_once.add( dc );
220 final Set<Integer> histogram_keys = histogram.keySet();
221 for( final Integer histogram_key : histogram_keys ) {
222 final int count = histogram.get( histogram_key );
223 final StringBuilder dc = domain_lists.get( histogram_key );
224 out_counts.write( histogram_key + "\t" + count + ForesterUtil.LINE_SEPARATOR );
225 out_dc.write( histogram_key + "\t" + dc + ForesterUtil.LINE_SEPARATOR );
226 out_dc_for_go_mapping.write( "#" + histogram_key + ForesterUtil.LINE_SEPARATOR );
227 final Object[] sorted = domain_lists_go.get( histogram_key ).toArray();
228 Arrays.sort( sorted );
229 for( final Object domain : sorted ) {
230 out_dc_for_go_mapping.write( domain + ForesterUtil.LINE_SEPARATOR );
232 out_dc_for_go_mapping_unique.write( "#" + histogram_key + ForesterUtil.LINE_SEPARATOR );
233 for( final String domain : domain_lists_go_unique.get( histogram_key ) ) {
234 out_dc_for_go_mapping_unique.write( domain + ForesterUtil.LINE_SEPARATOR );
239 out_dc_for_go_mapping.close();
240 out_dc_for_go_mapping_unique.close();
242 final SortedMap<String, Integer> lca_rank_counts = new TreeMap<String, Integer>();
243 final SortedMap<String, Integer> lca_ancestor_species_counts = new TreeMap<String, Integer>();
244 for( final String dc : more_than_once ) {
245 final List<PhylogenyNode> nodes = new ArrayList<PhylogenyNode>();
246 for( final PhylogenyNodeIterator it = local_phylogeny_l.iteratorExternalForward(); it.hasNext(); ) {
247 final PhylogenyNode n = it.next();
248 if ( n.getNodeData().getBinaryCharacters().getGainedCharacters().contains( dc ) ) {
252 for( int i = 0; i < nodes.size() - 1; ++i ) {
253 for( int j = i + 1; j < nodes.size(); ++j ) {
254 final PhylogenyNode lca = PhylogenyMethods.getInstance().obtainLCA( nodes.get( i ),
256 String rank = "unknown";
257 if ( lca.getNodeData().isHasTaxonomy()
258 && !ForesterUtil.isEmpty( lca.getNodeData().getTaxonomy().getRank() ) ) {
259 rank = lca.getNodeData().getTaxonomy().getRank();
261 addToCountMap( lca_rank_counts, rank );
263 if ( lca.getNodeData().isHasTaxonomy()
264 && !ForesterUtil.isEmpty( lca.getNodeData().getTaxonomy().getScientificName() ) ) {
265 lca_species = lca.getNodeData().getTaxonomy().getScientificName();
267 else if ( lca.getNodeData().isHasTaxonomy()
268 && !ForesterUtil.isEmpty( lca.getNodeData().getTaxonomy().getCommonName() ) ) {
269 lca_species = lca.getNodeData().getTaxonomy().getCommonName();
272 lca_species = lca.getName();
274 addToCountMap( lca_ancestor_species_counts, lca_species );
278 final BufferedWriter out_for_rank_counts = new BufferedWriter( new FileWriter( outfilename_for_rank_counts ) );
279 final BufferedWriter out_for_ancestor_species_counts = new BufferedWriter( new FileWriter( outfilename_for_ancestor_species_counts ) );
280 ForesterUtil.map2writer( out_for_rank_counts, lca_rank_counts, "\t", ForesterUtil.LINE_SEPARATOR );
281 ForesterUtil.map2writer( out_for_ancestor_species_counts,
282 lca_ancestor_species_counts,
284 ForesterUtil.LINE_SEPARATOR );
285 out_for_rank_counts.close();
286 out_for_ancestor_species_counts.close();
288 catch ( final IOException e ) {
289 ForesterUtil.printWarningMessage( surfacing.PRG_NAME, "Failure to write: " + e );
291 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote independent domain combination gains fitch counts to ["
292 + outfilename_for_counts + "]" );
293 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote independent domain combination gains fitch lists to ["
294 + outfilename_for_dc + "]" );
295 ForesterUtil.programMessage( surfacing.PRG_NAME,
296 "Wrote independent domain combination gains fitch lists to (for GO mapping) ["
297 + outfilename_for_dc_for_go_mapping + "]" );
298 ForesterUtil.programMessage( surfacing.PRG_NAME,
299 "Wrote independent domain combination gains fitch lists to (for GO mapping, unique) ["
300 + outfilename_for_dc_for_go_mapping_unique + "]" );
303 private final static void addToCountMap( final Map<String, Integer> map, final String s ) {
304 if ( map.containsKey( s ) ) {
305 map.put( s, map.get( s ) + 1 );
312 public static int calculateOverlap( final Domain domain, final List<Boolean> covered_positions ) {
313 int overlap_count = 0;
314 for( int i = domain.getFrom(); i <= domain.getTo(); ++i ) {
315 if ( ( i < covered_positions.size() ) && ( covered_positions.get( i ) == true ) ) {
319 return overlap_count;
322 public static void checkForOutputFileWriteability( final File outfile ) {
323 final String error = ForesterUtil.isWritableFile( outfile );
324 if ( !ForesterUtil.isEmpty( error ) ) {
325 ForesterUtil.fatalError( surfacing.PRG_NAME, error );
329 private static SortedSet<String> collectAllDomainsChangedOnSubtree( final PhylogenyNode subtree_root,
330 final boolean get_gains ) {
331 final SortedSet<String> domains = new TreeSet<String>();
332 for( final PhylogenyNode descendant : PhylogenyMethods.getAllDescendants( subtree_root ) ) {
333 final BinaryCharacters chars = descendant.getNodeData().getBinaryCharacters();
335 domains.addAll( chars.getGainedCharacters() );
338 domains.addAll( chars.getLostCharacters() );
344 public static void collectChangedDomainCombinationsFromBinaryStatesMatrixAsListToFile( final CharacterStateMatrix<CharacterStateMatrix.GainLossStates> matrix,
345 final BinaryDomainCombination.DomainCombinationType dc_type,
346 final List<BinaryDomainCombination> all_binary_domains_combination_gained,
347 final boolean get_gains ) {
348 final SortedSet<String> sorted_ids = new TreeSet<String>();
349 for( int i = 0; i < matrix.getNumberOfIdentifiers(); ++i ) {
350 sorted_ids.add( matrix.getIdentifier( i ) );
352 for( final String id : sorted_ids ) {
353 for( int c = 0; c < matrix.getNumberOfCharacters(); ++c ) {
354 if ( ( get_gains && ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.GAIN ) )
355 || ( !get_gains && ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.LOSS ) ) ) {
356 if ( dc_type == BinaryDomainCombination.DomainCombinationType.DIRECTED_ADJACTANT ) {
357 all_binary_domains_combination_gained.add( AdjactantDirectedBinaryDomainCombination
358 .createInstance( matrix.getCharacter( c ) ) );
360 else if ( dc_type == BinaryDomainCombination.DomainCombinationType.DIRECTED ) {
361 all_binary_domains_combination_gained.add( DirectedBinaryDomainCombination
362 .createInstance( matrix.getCharacter( c ) ) );
365 all_binary_domains_combination_gained.add( BasicBinaryDomainCombination.createInstance( matrix
366 .getCharacter( c ) ) );
373 private static File createBaseDirForPerNodeDomainFiles( final String base_dir,
374 final boolean domain_combinations,
375 final CharacterStateMatrix.GainLossStates state,
376 final String outfile ) {
377 File per_node_go_mapped_domain_gain_loss_files_base_dir = new File( new File( outfile ).getParent()
378 + ForesterUtil.FILE_SEPARATOR + base_dir );
379 if ( !per_node_go_mapped_domain_gain_loss_files_base_dir.exists() ) {
380 per_node_go_mapped_domain_gain_loss_files_base_dir.mkdir();
382 if ( domain_combinations ) {
383 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
384 + ForesterUtil.FILE_SEPARATOR + "DC" );
387 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
388 + ForesterUtil.FILE_SEPARATOR + "DOMAINS" );
390 if ( !per_node_go_mapped_domain_gain_loss_files_base_dir.exists() ) {
391 per_node_go_mapped_domain_gain_loss_files_base_dir.mkdir();
393 if ( state == GainLossStates.GAIN ) {
394 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
395 + ForesterUtil.FILE_SEPARATOR + "GAINS" );
397 else if ( state == GainLossStates.LOSS ) {
398 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
399 + ForesterUtil.FILE_SEPARATOR + "LOSSES" );
402 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
403 + ForesterUtil.FILE_SEPARATOR + "PRESENT" );
405 if ( !per_node_go_mapped_domain_gain_loss_files_base_dir.exists() ) {
406 per_node_go_mapped_domain_gain_loss_files_base_dir.mkdir();
408 return per_node_go_mapped_domain_gain_loss_files_base_dir;
411 public static Map<DomainId, List<GoId>> createDomainIdToGoIdMap( final List<PfamToGoMapping> pfam_to_go_mappings ) {
412 final Map<DomainId, List<GoId>> domain_id_to_go_ids_map = new HashMap<DomainId, List<GoId>>( pfam_to_go_mappings
414 for( final PfamToGoMapping pfam_to_go : pfam_to_go_mappings ) {
415 if ( !domain_id_to_go_ids_map.containsKey( pfam_to_go.getKey() ) ) {
416 domain_id_to_go_ids_map.put( pfam_to_go.getKey(), new ArrayList<GoId>() );
418 domain_id_to_go_ids_map.get( pfam_to_go.getKey() ).add( pfam_to_go.getValue() );
420 return domain_id_to_go_ids_map;
423 public static Map<DomainId, Set<String>> createDomainIdToSecondaryFeaturesMap( final File secondary_features_map_file )
425 final BasicTable<String> primary_table = BasicTableParser.parse( secondary_features_map_file, "\t" );
426 final Map<DomainId, Set<String>> map = new TreeMap<DomainId, Set<String>>();
427 for( int r = 0; r < primary_table.getNumberOfRows(); ++r ) {
428 final DomainId domain_id = new DomainId( primary_table.getValue( 0, r ) );
429 if ( !map.containsKey( domain_id ) ) {
430 map.put( domain_id, new HashSet<String>() );
432 map.get( domain_id ).add( primary_table.getValue( 1, r ) );
437 public static Phylogeny createNjTreeBasedOnMatrixToFile( final File nj_tree_outfile, final DistanceMatrix distance ) {
438 checkForOutputFileWriteability( nj_tree_outfile );
439 final NeighborJoining nj = NeighborJoining.createInstance();
440 final Phylogeny phylogeny = nj.execute( distance );
441 phylogeny.setName( nj_tree_outfile.getName() );
442 writePhylogenyToFile( phylogeny, nj_tree_outfile.toString() );
446 private static SortedSet<BinaryDomainCombination> createSetOfAllBinaryDomainCombinationsPerGenome( final GenomeWideCombinableDomains gwcd ) {
447 final SortedMap<DomainId, CombinableDomains> cds = gwcd.getAllCombinableDomainsIds();
448 final SortedSet<BinaryDomainCombination> binary_combinations = new TreeSet<BinaryDomainCombination>();
449 for( final DomainId domain_id : cds.keySet() ) {
450 final CombinableDomains cd = cds.get( domain_id );
451 binary_combinations.addAll( cd.toBinaryDomainCombinations() );
453 return binary_combinations;
456 public static void decoratePrintableDomainSimilarities( final SortedSet<DomainSimilarity> domain_similarities,
457 final Detailedness detailedness,
458 final GoAnnotationOutput go_annotation_output,
459 final Map<GoId, GoTerm> go_id_to_term_map,
460 final GoNameSpace go_namespace_limit ) {
461 if ( ( go_namespace_limit != null ) && ( ( go_id_to_term_map == null ) || go_id_to_term_map.isEmpty() ) ) {
462 throw new IllegalArgumentException( "attempt to use a GO namespace limit without a GO id to term map" );
464 for( final DomainSimilarity domain_similarity : domain_similarities ) {
465 if ( domain_similarity instanceof PrintableDomainSimilarity ) {
466 final PrintableDomainSimilarity printable_domain_similarity = ( PrintableDomainSimilarity ) domain_similarity;
467 printable_domain_similarity.setDetailedness( detailedness );
468 printable_domain_similarity.setGoAnnotationOutput( go_annotation_output );
469 printable_domain_similarity.setGoIdToTermMap( go_id_to_term_map );
470 printable_domain_similarity.setGoNamespaceLimit( go_namespace_limit );
475 public static void executeDomainLengthAnalysis( final String[][] input_file_properties,
476 final int number_of_genomes,
477 final DomainLengthsTable domain_lengths_table,
478 final File outfile ) throws IOException {
479 final DecimalFormat df = new DecimalFormat( "#.00" );
480 checkForOutputFileWriteability( outfile );
481 final BufferedWriter out = new BufferedWriter( new FileWriter( outfile ) );
482 out.write( "MEAN BASED STATISTICS PER SPECIES" );
483 out.write( ForesterUtil.LINE_SEPARATOR );
484 out.write( domain_lengths_table.createMeanBasedStatisticsPerSpeciesTable().toString() );
485 out.write( ForesterUtil.LINE_SEPARATOR );
486 out.write( ForesterUtil.LINE_SEPARATOR );
487 final List<DomainLengths> domain_lengths_list = domain_lengths_table.getDomainLengthsList();
488 out.write( "OUTLIER SPECIES PER DOMAIN (Z>=1.5)" );
489 out.write( ForesterUtil.LINE_SEPARATOR );
490 for( final DomainLengths domain_lengths : domain_lengths_list ) {
491 final List<Species> species_list = domain_lengths.getMeanBasedOutlierSpecies( 1.5 );
492 if ( species_list.size() > 0 ) {
493 out.write( domain_lengths.getDomainId() + "\t" );
494 for( final Species species : species_list ) {
495 out.write( species + "\t" );
497 out.write( ForesterUtil.LINE_SEPARATOR );
498 // DescriptiveStatistics stats_for_domain = domain_lengths
499 // .calculateMeanBasedStatistics();
500 //AsciiHistogram histo = new AsciiHistogram( stats_for_domain );
501 //System.out.println( histo.toStringBuffer( 40, '=', 60, 4 ).toString() );
504 out.write( ForesterUtil.LINE_SEPARATOR );
505 out.write( ForesterUtil.LINE_SEPARATOR );
506 out.write( "OUTLIER SPECIES (Z 1.0)" );
507 out.write( ForesterUtil.LINE_SEPARATOR );
508 final DescriptiveStatistics stats_for_all_species = domain_lengths_table
509 .calculateMeanBasedStatisticsForAllSpecies();
510 out.write( stats_for_all_species.asSummary() );
511 out.write( ForesterUtil.LINE_SEPARATOR );
512 final AsciiHistogram histo = new AsciiHistogram( stats_for_all_species );
513 out.write( histo.toStringBuffer( 40, '=', 60, 4 ).toString() );
514 out.write( ForesterUtil.LINE_SEPARATOR );
515 final double population_sd = stats_for_all_species.sampleStandardDeviation();
516 final double population_mean = stats_for_all_species.arithmeticMean();
517 for( final Species species : domain_lengths_table.getSpecies() ) {
518 final double x = domain_lengths_table.calculateMeanBasedStatisticsForSpecies( species ).arithmeticMean();
519 final double z = ( x - population_mean ) / population_sd;
520 out.write( species + "\t" + z );
521 out.write( ForesterUtil.LINE_SEPARATOR );
523 out.write( ForesterUtil.LINE_SEPARATOR );
524 for( final Species species : domain_lengths_table.getSpecies() ) {
525 final DescriptiveStatistics stats_for_species = domain_lengths_table
526 .calculateMeanBasedStatisticsForSpecies( species );
527 final double x = stats_for_species.arithmeticMean();
528 final double z = ( x - population_mean ) / population_sd;
529 if ( ( z <= -1.0 ) || ( z >= 1.0 ) ) {
530 out.write( species + "\t" + df.format( z ) + "\t" + stats_for_species.asSummary() );
531 out.write( ForesterUtil.LINE_SEPARATOR );
535 // final List<HistogramData> histogram_datas = new ArrayList<HistogramData>();
536 // for( int i = 0; i < number_of_genomes; ++i ) {
537 // final Species species = new BasicSpecies( input_file_properties[ i ][ 0 ] );
539 // .add( new HistogramData( species.toString(), domain_lengths_table
540 // .calculateMeanBasedStatisticsForSpecies( species )
541 // .getDataAsDoubleArray(), 5, 600, null, 60 ) );
543 // final HistogramsFrame hf = new HistogramsFrame( histogram_datas );
544 // hf.setVisible( true );
550 * @param all_binary_domains_combination_lost_fitch
551 * @param consider_directedness_and_adjacency_for_bin_combinations
552 * @param all_binary_domains_combination_gained if null ignored, otherwise this is to list all binary domain combinations
553 * which were gained under unweighted (Fitch) parsimony.
555 public static void executeParsimonyAnalysis( final long random_number_seed_for_fitch_parsimony,
556 final boolean radomize_fitch_parsimony,
557 final String outfile_name,
558 final DomainParsimonyCalculator domain_parsimony,
559 final Phylogeny phylogeny,
560 final Map<DomainId, List<GoId>> domain_id_to_go_ids_map,
561 final Map<GoId, GoTerm> go_id_to_term_map,
562 final GoNameSpace go_namespace_limit,
563 final String parameters_str,
564 final Map<DomainId, Set<String>>[] domain_id_to_secondary_features_maps,
565 final SortedSet<DomainId> positive_filter,
566 final boolean output_binary_domain_combinations_for_graphs,
567 final List<BinaryDomainCombination> all_binary_domains_combination_gained_fitch,
568 final List<BinaryDomainCombination> all_binary_domains_combination_lost_fitch,
569 final BinaryDomainCombination.DomainCombinationType dc_type ) {
570 final String sep = ForesterUtil.LINE_SEPARATOR + "###################" + ForesterUtil.LINE_SEPARATOR;
571 final String date_time = ForesterUtil.getCurrentDateTime();
572 final SortedSet<String> all_pfams_encountered = new TreeSet<String>();
573 final SortedSet<String> all_pfams_gained_as_domains = new TreeSet<String>();
574 final SortedSet<String> all_pfams_lost_as_domains = new TreeSet<String>();
575 final SortedSet<String> all_pfams_gained_as_dom_combinations = new TreeSet<String>();
576 final SortedSet<String> all_pfams_lost_as_dom_combinations = new TreeSet<String>();
577 writeToNexus( outfile_name, domain_parsimony, phylogeny );
580 Phylogeny local_phylogeny_l = phylogeny.copy();
581 if ( ( positive_filter != null ) && ( positive_filter.size() > 0 ) ) {
582 domain_parsimony.executeDolloParsimonyOnDomainPresence( positive_filter );
585 domain_parsimony.executeDolloParsimonyOnDomainPresence();
587 SurfacingUtil.writeMatrixToFile( domain_parsimony.getGainLossMatrix(), outfile_name
588 + surfacing.PARSIMONY_OUTPUT_GL_SUFFIX_DOLLO_DOMAINS, Format.FORESTER );
589 SurfacingUtil.writeMatrixToFile( domain_parsimony.getGainLossCountsMatrix(), outfile_name
590 + surfacing.PARSIMONY_OUTPUT_GL_COUNTS_SUFFIX_DOLLO_DOMAINS, Format.FORESTER );
591 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
592 CharacterStateMatrix.GainLossStates.GAIN,
593 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_GAINS_D,
595 ForesterUtil.LINE_SEPARATOR,
597 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
598 CharacterStateMatrix.GainLossStates.LOSS,
599 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_LOSSES_D,
601 ForesterUtil.LINE_SEPARATOR,
603 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(), null, outfile_name
604 + surfacing.PARSIMONY_OUTPUT_DOLLO_PRESENT_D, sep, ForesterUtil.LINE_SEPARATOR, null );
606 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
610 domain_parsimony.getGainLossMatrix(),
611 CharacterStateMatrix.GainLossStates.GAIN,
612 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_GAINS_HTML_D,
614 ForesterUtil.LINE_SEPARATOR,
615 "Dollo Parsimony | Gains | Domains",
617 domain_id_to_secondary_features_maps,
618 all_pfams_encountered,
619 all_pfams_gained_as_domains,
621 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
625 domain_parsimony.getGainLossMatrix(),
626 CharacterStateMatrix.GainLossStates.LOSS,
627 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_LOSSES_HTML_D,
629 ForesterUtil.LINE_SEPARATOR,
630 "Dollo Parsimony | Losses | Domains",
632 domain_id_to_secondary_features_maps,
633 all_pfams_encountered,
634 all_pfams_lost_as_domains,
636 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
640 domain_parsimony.getGainLossMatrix(),
642 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_PRESENT_HTML_D,
644 ForesterUtil.LINE_SEPARATOR,
645 "Dollo Parsimony | Present | Domains",
647 domain_id_to_secondary_features_maps,
648 all_pfams_encountered,
650 "_dollo_present_d" );
651 preparePhylogeny( local_phylogeny_l,
654 "Dollo parsimony on domain presence/absence",
655 "dollo_on_domains_" + outfile_name,
657 SurfacingUtil.writePhylogenyToFile( local_phylogeny_l, outfile_name
658 + surfacing.DOMAINS_PARSIMONY_TREE_OUTPUT_SUFFIX_DOLLO );
660 writeAllDomainsChangedOnAllSubtrees( local_phylogeny_l, true, outfile_name, "_dollo_all_gains_d" );
661 writeAllDomainsChangedOnAllSubtrees( local_phylogeny_l, false, outfile_name, "_dollo_all_losses_d" );
663 catch ( final IOException e ) {
665 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getLocalizedMessage() );
667 if ( domain_parsimony.calculateNumberOfBinaryDomainCombination() > 0 ) {
668 // FITCH DOMAIN COMBINATIONS
669 // -------------------------
670 local_phylogeny_l = phylogeny.copy();
671 String randomization = "no";
672 if ( radomize_fitch_parsimony ) {
673 domain_parsimony.executeFitchParsimonyOnBinaryDomainCombintion( random_number_seed_for_fitch_parsimony );
674 randomization = "yes, seed = " + random_number_seed_for_fitch_parsimony;
677 domain_parsimony.executeFitchParsimonyOnBinaryDomainCombintion( USE_LAST );
679 SurfacingUtil.writeMatrixToFile( domain_parsimony.getGainLossMatrix(), outfile_name
680 + surfacing.PARSIMONY_OUTPUT_GL_SUFFIX_FITCH_BINARY_COMBINATIONS, Format.FORESTER );
681 SurfacingUtil.writeMatrixToFile( domain_parsimony.getGainLossCountsMatrix(), outfile_name
682 + surfacing.PARSIMONY_OUTPUT_GL_COUNTS_SUFFIX_FITCH_BINARY_COMBINATIONS, Format.FORESTER );
684 .writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
685 CharacterStateMatrix.GainLossStates.GAIN,
686 outfile_name + surfacing.PARSIMONY_OUTPUT_FITCH_GAINS_BC,
688 ForesterUtil.LINE_SEPARATOR,
690 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
691 CharacterStateMatrix.GainLossStates.LOSS,
693 + surfacing.PARSIMONY_OUTPUT_FITCH_LOSSES_BC,
695 ForesterUtil.LINE_SEPARATOR,
697 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(), null, outfile_name
698 + surfacing.PARSIMONY_OUTPUT_FITCH_PRESENT_BC, sep, ForesterUtil.LINE_SEPARATOR, null );
699 if ( all_binary_domains_combination_gained_fitch != null ) {
700 collectChangedDomainCombinationsFromBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
702 all_binary_domains_combination_gained_fitch,
705 if ( all_binary_domains_combination_lost_fitch != null ) {
706 collectChangedDomainCombinationsFromBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
708 all_binary_domains_combination_lost_fitch,
711 if ( output_binary_domain_combinations_for_graphs ) {
713 .writeBinaryStatesMatrixAsListToFileForBinaryCombinationsForGraphAnalysis( domain_parsimony
714 .getGainLossMatrix(),
717 + surfacing.PARSIMONY_OUTPUT_FITCH_PRESENT_BC_OUTPUTFILE_SUFFIX_FOR_GRAPH_ANALYSIS,
719 ForesterUtil.LINE_SEPARATOR,
720 BinaryDomainCombination.OutputFormat.DOT );
723 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
727 domain_parsimony.getGainLossMatrix(),
728 CharacterStateMatrix.GainLossStates.GAIN,
729 outfile_name + surfacing.PARSIMONY_OUTPUT_FITCH_GAINS_HTML_BC,
731 ForesterUtil.LINE_SEPARATOR,
732 "Fitch Parsimony | Gains | Domain Combinations",
735 all_pfams_encountered,
736 all_pfams_gained_as_dom_combinations,
738 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
742 domain_parsimony.getGainLossMatrix(),
743 CharacterStateMatrix.GainLossStates.LOSS,
744 outfile_name + surfacing.PARSIMONY_OUTPUT_FITCH_LOSSES_HTML_BC,
746 ForesterUtil.LINE_SEPARATOR,
747 "Fitch Parsimony | Losses | Domain Combinations",
750 all_pfams_encountered,
751 all_pfams_lost_as_dom_combinations,
752 "_fitch_losses_dc" );
753 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
757 domain_parsimony.getGainLossMatrix(),
759 outfile_name + surfacing.PARSIMONY_OUTPUT_FITCH_PRESENT_HTML_BC,
761 ForesterUtil.LINE_SEPARATOR,
762 "Fitch Parsimony | Present | Domain Combinations",
765 all_pfams_encountered,
767 "_fitch_present_dc" );
768 writeAllEncounteredPfamsToFile( domain_id_to_go_ids_map,
771 all_pfams_encountered );
772 writePfamsToFile( outfile_name + surfacing.ALL_PFAMS_GAINED_AS_DOMAINS_SUFFIX, all_pfams_gained_as_domains );
773 writePfamsToFile( outfile_name + surfacing.ALL_PFAMS_LOST_AS_DOMAINS_SUFFIX, all_pfams_lost_as_domains );
774 writePfamsToFile( outfile_name + surfacing.ALL_PFAMS_GAINED_AS_DC_SUFFIX,
775 all_pfams_gained_as_dom_combinations );
776 writePfamsToFile( outfile_name + surfacing.ALL_PFAMS_LOST_AS_DC_SUFFIX, all_pfams_lost_as_dom_combinations );
777 preparePhylogeny( local_phylogeny_l,
780 "Fitch parsimony on binary domain combination presence/absence randomization: "
782 "fitch_on_binary_domain_combinations_" + outfile_name,
784 SurfacingUtil.writePhylogenyToFile( local_phylogeny_l, outfile_name
785 + surfacing.BINARY_DOMAIN_COMBINATIONS_PARSIMONY_TREE_OUTPUT_SUFFIX_FITCH );
786 calculateIndependentDomainCombinationGains( local_phylogeny_l, outfile_name
787 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_COUNTS_OUTPUT_SUFFIX, outfile_name
788 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_OUTPUT_SUFFIX, outfile_name
789 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_FOR_GO_MAPPING_OUTPUT_SUFFIX, outfile_name
790 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_FOR_GO_MAPPING_OUTPUT_UNIQUE_SUFFIX, outfile_name
791 + "_indep_dc_gains_fitch_lca_ranks.txt", outfile_name + "_indep_dc_gains_fitch_lca_taxonomies.txt" );
795 public static void executeParsimonyAnalysisForSecondaryFeatures( final String outfile_name,
796 final DomainParsimonyCalculator secondary_features_parsimony,
797 final Phylogeny phylogeny,
798 final String parameters_str,
799 final Map<Species, MappingResults> mapping_results_map ) {
800 final String sep = ForesterUtil.LINE_SEPARATOR + "###################" + ForesterUtil.LINE_SEPARATOR;
801 final String date_time = ForesterUtil.getCurrentDateTime();
802 System.out.println();
803 writeToNexus( outfile_name + surfacing.NEXUS_SECONDARY_FEATURES,
804 secondary_features_parsimony.createMatrixOfSecondaryFeaturePresenceOrAbsence( null ),
806 Phylogeny local_phylogeny_copy = phylogeny.copy();
807 secondary_features_parsimony.executeDolloParsimonyOnSecondaryFeatures( mapping_results_map );
808 SurfacingUtil.writeMatrixToFile( secondary_features_parsimony.getGainLossMatrix(), outfile_name
809 + surfacing.PARSIMONY_OUTPUT_GL_SUFFIX_DOLLO_SECONDARY_FEATURES, Format.FORESTER );
810 SurfacingUtil.writeMatrixToFile( secondary_features_parsimony.getGainLossCountsMatrix(), outfile_name
811 + surfacing.PARSIMONY_OUTPUT_GL_COUNTS_SUFFIX_DOLLO_SECONDARY_FEATURES, Format.FORESTER );
813 .writeBinaryStatesMatrixAsListToFile( secondary_features_parsimony.getGainLossMatrix(),
814 CharacterStateMatrix.GainLossStates.GAIN,
816 + surfacing.PARSIMONY_OUTPUT_DOLLO_GAINS_SECONDARY_FEATURES,
818 ForesterUtil.LINE_SEPARATOR,
821 .writeBinaryStatesMatrixAsListToFile( secondary_features_parsimony.getGainLossMatrix(),
822 CharacterStateMatrix.GainLossStates.LOSS,
824 + surfacing.PARSIMONY_OUTPUT_DOLLO_LOSSES_SECONDARY_FEATURES,
826 ForesterUtil.LINE_SEPARATOR,
829 .writeBinaryStatesMatrixAsListToFile( secondary_features_parsimony.getGainLossMatrix(),
832 + surfacing.PARSIMONY_OUTPUT_DOLLO_PRESENT_SECONDARY_FEATURES,
834 ForesterUtil.LINE_SEPARATOR,
836 preparePhylogeny( local_phylogeny_copy,
837 secondary_features_parsimony,
839 "Dollo parsimony on secondary feature presence/absence",
840 "dollo_on_secondary_features_" + outfile_name,
842 SurfacingUtil.writePhylogenyToFile( local_phylogeny_copy, outfile_name
843 + surfacing.SECONDARY_FEATURES_PARSIMONY_TREE_OUTPUT_SUFFIX_DOLLO );
844 // FITCH DOMAIN COMBINATIONS
845 // -------------------------
846 local_phylogeny_copy = phylogeny.copy();
847 final String randomization = "no";
848 secondary_features_parsimony.executeFitchParsimonyOnBinaryDomainCombintionOnSecondaryFeatures( USE_LAST );
849 preparePhylogeny( local_phylogeny_copy,
850 secondary_features_parsimony,
852 "Fitch parsimony on secondary binary domain combination presence/absence randomization: "
854 "fitch_on_binary_domain_combinations_" + outfile_name,
856 SurfacingUtil.writePhylogenyToFile( local_phylogeny_copy, outfile_name
857 + surfacing.BINARY_DOMAIN_COMBINATIONS_PARSIMONY_TREE_OUTPUT_SUFFIX_FITCH_MAPPED );
858 calculateIndependentDomainCombinationGains( local_phylogeny_copy, outfile_name
859 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_COUNTS_MAPPED_OUTPUT_SUFFIX, outfile_name
860 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_MAPPED_OUTPUT_SUFFIX, outfile_name
861 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_FOR_GO_MAPPING_MAPPED_OUTPUT_SUFFIX, outfile_name
862 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_FOR_GO_MAPPING_MAPPED_OUTPUT_UNIQUE_SUFFIX, outfile_name
863 + "_MAPPED_indep_dc_gains_fitch_lca_ranks.txt", outfile_name
864 + "_MAPPED_indep_dc_gains_fitch_lca_taxonomies.txt" );
867 public static void extractProteinNames( final List<Protein> proteins,
868 final List<DomainId> query_domain_ids_nc_order,
870 final String separator,
871 final String limit_to_species ) throws IOException {
872 for( final Protein protein : proteins ) {
873 if ( ForesterUtil.isEmpty( limit_to_species )
874 || protein.getSpecies().getSpeciesId().equalsIgnoreCase( limit_to_species ) ) {
875 if ( protein.contains( query_domain_ids_nc_order, true ) ) {
876 out.write( protein.getSpecies().getSpeciesId() );
877 out.write( separator );
878 out.write( protein.getProteinId().getId() );
879 out.write( separator );
881 final Set<DomainId> visited_domain_ids = new HashSet<DomainId>();
882 boolean first = true;
883 for( final Domain domain : protein.getProteinDomains() ) {
884 if ( !visited_domain_ids.contains( domain.getDomainId() ) ) {
885 visited_domain_ids.add( domain.getDomainId() );
892 out.write( domain.getDomainId().getId() );
894 out.write( "" + domain.getTotalCount() );
899 out.write( separator );
900 if ( !( ForesterUtil.isEmpty( protein.getDescription() ) || protein.getDescription()
901 .equals( SurfacingConstants.NONE ) ) ) {
902 out.write( protein.getDescription() );
904 out.write( separator );
905 if ( !( ForesterUtil.isEmpty( protein.getAccession() ) || protein.getAccession()
906 .equals( SurfacingConstants.NONE ) ) ) {
907 out.write( protein.getAccession() );
909 out.write( SurfacingConstants.NL );
916 public static void extractProteinNames( final SortedMap<Species, List<Protein>> protein_lists_per_species,
917 final DomainId domain_id,
919 final String separator,
920 final String limit_to_species ) throws IOException {
921 for( final Species species : protein_lists_per_species.keySet() ) {
922 for( final Protein protein : protein_lists_per_species.get( species ) ) {
923 if ( ForesterUtil.isEmpty( limit_to_species )
924 || protein.getSpecies().getSpeciesId().equalsIgnoreCase( limit_to_species ) ) {
925 final List<Domain> domains = protein.getProteinDomains( domain_id );
926 if ( domains.size() > 0 ) {
927 final DescriptiveStatistics stats = new BasicDescriptiveStatistics();
928 for( final Domain domain : domains ) {
929 stats.addValue( domain.getPerSequenceEvalue() );
931 out.write( protein.getSpecies().getSpeciesId() );
932 out.write( separator );
933 out.write( protein.getProteinId().getId() );
934 out.write( separator );
935 out.write( "[" + FORMATTER.format( stats.median() ) + "]" );
936 out.write( separator );
937 if ( !( ForesterUtil.isEmpty( protein.getDescription() ) || protein.getDescription()
938 .equals( SurfacingConstants.NONE ) ) ) {
939 out.write( protein.getDescription() );
941 out.write( separator );
942 if ( !( ForesterUtil.isEmpty( protein.getAccession() ) || protein.getAccession()
943 .equals( SurfacingConstants.NONE ) ) ) {
944 out.write( protein.getAccession() );
946 out.write( SurfacingConstants.NL );
954 public static SortedSet<DomainId> getAllDomainIds( final List<GenomeWideCombinableDomains> gwcd_list ) {
955 final SortedSet<DomainId> all_domains_ids = new TreeSet<DomainId>();
956 for( final GenomeWideCombinableDomains gwcd : gwcd_list ) {
957 final Set<DomainId> all_domains = gwcd.getAllDomainIds();
958 // for( final Domain domain : all_domains ) {
959 all_domains_ids.addAll( all_domains );
962 return all_domains_ids;
965 public static SortedMap<String, Integer> getDomainCounts( final List<Protein> protein_domain_collections ) {
966 final SortedMap<String, Integer> map = new TreeMap<String, Integer>();
967 for( final Protein protein_domain_collection : protein_domain_collections ) {
968 for( final Object name : protein_domain_collection.getProteinDomains() ) {
969 final BasicDomain protein_domain = ( BasicDomain ) name;
970 final String id = protein_domain.getDomainId().getId();
971 if ( map.containsKey( id ) ) {
972 map.put( id, map.get( id ) + 1 );
982 public static int getNumberOfNodesLackingName( final Phylogeny p, final StringBuilder names ) {
983 final PhylogenyNodeIterator it = p.iteratorPostorder();
985 while ( it.hasNext() ) {
986 final PhylogenyNode n = it.next();
987 if ( ForesterUtil.isEmpty( n.getName() )
988 && ( !n.getNodeData().isHasTaxonomy() || ForesterUtil.isEmpty( n.getNodeData().getTaxonomy()
989 .getScientificName() ) )
990 && ( !n.getNodeData().isHasTaxonomy() || ForesterUtil.isEmpty( n.getNodeData().getTaxonomy()
991 .getCommonName() ) ) ) {
992 if ( n.getParent() != null ) {
994 names.append( n.getParent().getName() );
996 final List l = n.getAllExternalDescendants();
997 for( final Object object : l ) {
998 System.out.println( l.toString() );
1007 * Returns true is Domain domain falls in an uninterrupted stretch of
1008 * covered positions.
1011 * @param covered_positions
1014 public static boolean isEngulfed( final Domain domain, final List<Boolean> covered_positions ) {
1015 for( int i = domain.getFrom(); i <= domain.getTo(); ++i ) {
1016 if ( ( i >= covered_positions.size() ) || ( covered_positions.get( i ) != true ) ) {
1023 public static void preparePhylogeny( final Phylogeny p,
1024 final DomainParsimonyCalculator domain_parsimony,
1025 final String date_time,
1026 final String method,
1028 final String parameters_str ) {
1029 domain_parsimony.decoratePhylogenyWithDomains( p );
1030 final StringBuilder desc = new StringBuilder();
1031 desc.append( "[Method: " + method + "] [Date: " + date_time + "] " );
1032 desc.append( "[Cost: " + domain_parsimony.getCost() + "] " );
1033 desc.append( "[Gains: " + domain_parsimony.getTotalGains() + "] " );
1034 desc.append( "[Losses: " + domain_parsimony.getTotalLosses() + "] " );
1035 desc.append( "[Unchanged: " + domain_parsimony.getTotalUnchanged() + "] " );
1036 desc.append( "[Parameters: " + parameters_str + "]" );
1038 p.setDescription( desc.toString() );
1039 p.setConfidence( new Confidence( domain_parsimony.getCost(), "parsimony" ) );
1040 p.setRerootable( false );
1041 p.setRooted( true );
1045 * species | protein id | n-terminal domain | c-terminal domain | n-terminal domain per domain E-value | c-terminal domain per domain E-value
1049 static public StringBuffer proteinToDomainCombinations( final Protein protein,
1050 final String protein_id,
1051 final String separator ) {
1052 final StringBuffer sb = new StringBuffer();
1053 if ( protein.getSpecies() == null ) {
1054 throw new IllegalArgumentException( "species must not be null" );
1056 if ( ForesterUtil.isEmpty( protein.getSpecies().getSpeciesId() ) ) {
1057 throw new IllegalArgumentException( "species id must not be empty" );
1059 final List<Domain> domains = protein.getProteinDomains();
1060 if ( domains.size() > 1 ) {
1061 final Map<String, Integer> counts = new HashMap<String, Integer>();
1062 for( final Domain domain : domains ) {
1063 final String id = domain.getDomainId().getId();
1064 if ( counts.containsKey( id ) ) {
1065 counts.put( id, counts.get( id ) + 1 );
1068 counts.put( id, 1 );
1071 final Set<String> dcs = new HashSet<String>();
1072 for( int i = 1; i < domains.size(); ++i ) {
1073 for( int j = 0; j < i; ++j ) {
1074 Domain domain_n = domains.get( i );
1075 Domain domain_c = domains.get( j );
1076 if ( domain_n.getFrom() > domain_c.getFrom() ) {
1077 domain_n = domains.get( j );
1078 domain_c = domains.get( i );
1080 final String dc = domain_n.getDomainId().getId() + domain_c.getDomainId().getId();
1081 if ( !dcs.contains( dc ) ) {
1083 sb.append( protein.getSpecies() );
1084 sb.append( separator );
1085 sb.append( protein_id );
1086 sb.append( separator );
1087 sb.append( domain_n.getDomainId().getId() );
1088 sb.append( separator );
1089 sb.append( domain_c.getDomainId().getId() );
1090 sb.append( separator );
1091 sb.append( domain_n.getPerDomainEvalue() );
1092 sb.append( separator );
1093 sb.append( domain_c.getPerDomainEvalue() );
1094 sb.append( separator );
1095 sb.append( counts.get( domain_n.getDomainId().getId() ) );
1096 sb.append( separator );
1097 sb.append( counts.get( domain_c.getDomainId().getId() ) );
1098 sb.append( ForesterUtil.LINE_SEPARATOR );
1103 else if ( domains.size() == 1 ) {
1104 sb.append( protein.getSpecies() );
1105 sb.append( separator );
1106 sb.append( protein_id );
1107 sb.append( separator );
1108 sb.append( domains.get( 0 ).getDomainId().getId() );
1109 sb.append( separator );
1110 sb.append( separator );
1111 sb.append( domains.get( 0 ).getPerDomainEvalue() );
1112 sb.append( separator );
1113 sb.append( separator );
1115 sb.append( separator );
1116 sb.append( ForesterUtil.LINE_SEPARATOR );
1119 sb.append( protein.getSpecies() );
1120 sb.append( separator );
1121 sb.append( protein_id );
1122 sb.append( separator );
1123 sb.append( separator );
1124 sb.append( separator );
1125 sb.append( separator );
1126 sb.append( separator );
1127 sb.append( separator );
1128 sb.append( ForesterUtil.LINE_SEPARATOR );
1135 * Example regarding engulfment: ------------0.1 ----------0.2 --0.3 =>
1136 * domain with 0.3 is ignored
1138 * -----------0.1 ----------0.2 --0.3 => domain with 0.3 is ignored
1141 * ------------0.1 ----------0.3 --0.2 => domains with 0.3 and 0.2 are _not_
1144 * @param max_allowed_overlap
1145 * maximal allowed overlap (inclusive) to be still considered not
1146 * overlapping (zero or negative value to allow any overlap)
1147 * @param remove_engulfed_domains
1148 * to remove domains which are completely engulfed by coverage of
1149 * domains with better support
1153 public static Protein removeOverlappingDomains( final int max_allowed_overlap,
1154 final boolean remove_engulfed_domains,
1155 final Protein protein ) {
1156 final Protein pruned_protein = new BasicProtein( protein.getProteinId().getId(), protein.getSpecies()
1158 final List<Domain> sorted = SurfacingUtil.sortDomainsWithAscendingConfidenceValues( protein );
1159 final List<Boolean> covered_positions = new ArrayList<Boolean>();
1160 for( final Domain domain : sorted ) {
1161 if ( ( ( max_allowed_overlap < 0 ) || ( SurfacingUtil.calculateOverlap( domain, covered_positions ) <= max_allowed_overlap ) )
1162 && ( !remove_engulfed_domains || !isEngulfed( domain, covered_positions ) ) ) {
1163 final int covered_positions_size = covered_positions.size();
1164 for( int i = covered_positions_size; i < domain.getFrom(); ++i ) {
1165 covered_positions.add( false );
1167 final int new_covered_positions_size = covered_positions.size();
1168 for( int i = domain.getFrom(); i <= domain.getTo(); ++i ) {
1169 if ( i < new_covered_positions_size ) {
1170 covered_positions.set( i, true );
1173 covered_positions.add( true );
1176 pruned_protein.addProteinDomain( domain );
1179 return pruned_protein;
1182 public static List<Domain> sortDomainsWithAscendingConfidenceValues( final Protein protein ) {
1183 final List<Domain> domains = new ArrayList<Domain>();
1184 for( final Domain d : protein.getProteinDomains() ) {
1187 Collections.sort( domains, SurfacingUtil.ASCENDING_CONFIDENCE_VALUE_ORDER );
1191 private static List<String> splitDomainCombination( final String dc ) {
1192 final String[] s = dc.split( "=" );
1193 if ( s.length != 2 ) {
1194 ForesterUtil.printErrorMessage( surfacing.PRG_NAME, "Stringyfied domain combination has illegal format: "
1198 final List<String> l = new ArrayList<String>( 2 );
1204 public static void writeAllDomainsChangedOnAllSubtrees( final Phylogeny p,
1205 final boolean get_gains,
1206 final String outdir,
1207 final String suffix_for_filename ) throws IOException {
1208 CharacterStateMatrix.GainLossStates state = CharacterStateMatrix.GainLossStates.GAIN;
1210 state = CharacterStateMatrix.GainLossStates.LOSS;
1212 final File base_dir = createBaseDirForPerNodeDomainFiles( surfacing.BASE_DIRECTORY_PER_SUBTREE_DOMAIN_GAIN_LOSS_FILES,
1216 for( final PhylogenyNodeIterator it = p.iteratorPostorder(); it.hasNext(); ) {
1217 final PhylogenyNode node = it.next();
1218 if ( !node.isExternal() ) {
1219 final SortedSet<String> domains = collectAllDomainsChangedOnSubtree( node, get_gains );
1220 if ( domains.size() > 0 ) {
1221 final Writer writer = ForesterUtil.createBufferedWriter( base_dir + ForesterUtil.FILE_SEPARATOR
1222 + node.getName() + suffix_for_filename );
1223 for( final String domain : domains ) {
1224 writer.write( domain );
1225 writer.write( ForesterUtil.LINE_SEPARATOR );
1233 private static void writeAllEncounteredPfamsToFile( final Map<DomainId, List<GoId>> domain_id_to_go_ids_map,
1234 final Map<GoId, GoTerm> go_id_to_term_map,
1235 final String outfile_name,
1236 final SortedSet<String> all_pfams_encountered ) {
1237 final File all_pfams_encountered_file = new File( outfile_name + surfacing.ALL_PFAMS_ENCOUNTERED_SUFFIX );
1238 final File all_pfams_encountered_with_go_annotation_file = new File( outfile_name
1239 + surfacing.ALL_PFAMS_ENCOUNTERED_WITH_GO_ANNOTATION_SUFFIX );
1240 final File encountered_pfams_summary_file = new File( outfile_name + surfacing.ENCOUNTERED_PFAMS_SUMMARY_SUFFIX );
1241 int biological_process_counter = 0;
1242 int cellular_component_counter = 0;
1243 int molecular_function_counter = 0;
1244 int pfams_with_mappings_counter = 0;
1245 int pfams_without_mappings_counter = 0;
1246 int pfams_without_mappings_to_bp_or_mf_counter = 0;
1247 int pfams_with_mappings_to_bp_or_mf_counter = 0;
1249 final Writer all_pfams_encountered_writer = new BufferedWriter( new FileWriter( all_pfams_encountered_file ) );
1250 final Writer all_pfams_encountered_with_go_annotation_writer = new BufferedWriter( new FileWriter( all_pfams_encountered_with_go_annotation_file ) );
1251 final Writer summary_writer = new BufferedWriter( new FileWriter( encountered_pfams_summary_file ) );
1252 summary_writer.write( "# Pfam to GO mapping summary" );
1253 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1254 summary_writer.write( "# Actual summary is at the end of this file." );
1255 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1256 summary_writer.write( "# Encountered Pfams without a GO mapping:" );
1257 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1258 for( final String pfam : all_pfams_encountered ) {
1259 all_pfams_encountered_writer.write( pfam );
1260 all_pfams_encountered_writer.write( ForesterUtil.LINE_SEPARATOR );
1261 final DomainId domain_id = new DomainId( pfam );
1262 if ( domain_id_to_go_ids_map.containsKey( domain_id ) ) {
1263 ++pfams_with_mappings_counter;
1264 all_pfams_encountered_with_go_annotation_writer.write( pfam );
1265 all_pfams_encountered_with_go_annotation_writer.write( ForesterUtil.LINE_SEPARATOR );
1266 final List<GoId> go_ids = domain_id_to_go_ids_map.get( domain_id );
1267 boolean maps_to_bp = false;
1268 boolean maps_to_cc = false;
1269 boolean maps_to_mf = false;
1270 for( final GoId go_id : go_ids ) {
1271 final GoTerm go_term = go_id_to_term_map.get( go_id );
1272 if ( go_term.getGoNameSpace().isBiologicalProcess() ) {
1275 else if ( go_term.getGoNameSpace().isCellularComponent() ) {
1278 else if ( go_term.getGoNameSpace().isMolecularFunction() ) {
1283 ++biological_process_counter;
1286 ++cellular_component_counter;
1289 ++molecular_function_counter;
1291 if ( maps_to_bp || maps_to_mf ) {
1292 ++pfams_with_mappings_to_bp_or_mf_counter;
1295 ++pfams_without_mappings_to_bp_or_mf_counter;
1299 ++pfams_without_mappings_to_bp_or_mf_counter;
1300 ++pfams_without_mappings_counter;
1301 summary_writer.write( pfam );
1302 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1305 all_pfams_encountered_writer.close();
1306 all_pfams_encountered_with_go_annotation_writer.close();
1307 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote all [" + all_pfams_encountered.size()
1308 + "] encountered Pfams to: \"" + all_pfams_encountered_file + "\"" );
1309 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote all [" + pfams_with_mappings_counter
1310 + "] encountered Pfams with GO mappings to: \"" + all_pfams_encountered_with_go_annotation_file
1312 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote summary (including all ["
1313 + pfams_without_mappings_counter + "] encountered Pfams without GO mappings) to: \""
1314 + encountered_pfams_summary_file + "\"" );
1315 ForesterUtil.programMessage( surfacing.PRG_NAME, "Sum of Pfams encountered : "
1316 + all_pfams_encountered.size() );
1317 ForesterUtil.programMessage( surfacing.PRG_NAME, "Pfams without a mapping : "
1318 + pfams_without_mappings_counter + " ["
1319 + ( 100 * pfams_without_mappings_counter / all_pfams_encountered.size() ) + "%]" );
1320 ForesterUtil.programMessage( surfacing.PRG_NAME, "Pfams without mapping to proc. or func. : "
1321 + pfams_without_mappings_to_bp_or_mf_counter + " ["
1322 + ( 100 * pfams_without_mappings_to_bp_or_mf_counter / all_pfams_encountered.size() ) + "%]" );
1323 ForesterUtil.programMessage( surfacing.PRG_NAME,
1324 "Pfams with a mapping : " + pfams_with_mappings_counter
1326 + ( 100 * pfams_with_mappings_counter / all_pfams_encountered.size() )
1328 ForesterUtil.programMessage( surfacing.PRG_NAME, "Pfams with a mapping to proc. or func. : "
1329 + pfams_with_mappings_to_bp_or_mf_counter + " ["
1330 + ( 100 * pfams_with_mappings_to_bp_or_mf_counter / all_pfams_encountered.size() ) + "%]" );
1331 ForesterUtil.programMessage( surfacing.PRG_NAME,
1332 "Pfams with mapping to biological process: " + biological_process_counter
1334 + ( 100 * biological_process_counter / all_pfams_encountered.size() )
1336 ForesterUtil.programMessage( surfacing.PRG_NAME,
1337 "Pfams with mapping to molecular function: " + molecular_function_counter
1339 + ( 100 * molecular_function_counter / all_pfams_encountered.size() )
1341 ForesterUtil.programMessage( surfacing.PRG_NAME,
1342 "Pfams with mapping to cellular component: " + cellular_component_counter
1344 + ( 100 * cellular_component_counter / all_pfams_encountered.size() )
1346 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1347 summary_writer.write( "# Sum of Pfams encountered : " + all_pfams_encountered.size() );
1348 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1349 summary_writer.write( "# Pfams without a mapping : " + pfams_without_mappings_counter
1350 + " [" + ( 100 * pfams_without_mappings_counter / all_pfams_encountered.size() ) + "%]" );
1351 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1352 summary_writer.write( "# Pfams without mapping to proc. or func. : "
1353 + pfams_without_mappings_to_bp_or_mf_counter + " ["
1354 + ( 100 * pfams_without_mappings_to_bp_or_mf_counter / all_pfams_encountered.size() ) + "%]" );
1355 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1356 summary_writer.write( "# Pfams with a mapping : " + pfams_with_mappings_counter + " ["
1357 + ( 100 * pfams_with_mappings_counter / all_pfams_encountered.size() ) + "%]" );
1358 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1359 summary_writer.write( "# Pfams with a mapping to proc. or func. : "
1360 + pfams_with_mappings_to_bp_or_mf_counter + " ["
1361 + ( 100 * pfams_with_mappings_to_bp_or_mf_counter / all_pfams_encountered.size() ) + "%]" );
1362 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1363 summary_writer.write( "# Pfams with mapping to biological process: " + biological_process_counter + " ["
1364 + ( 100 * biological_process_counter / all_pfams_encountered.size() ) + "%]" );
1365 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1366 summary_writer.write( "# Pfams with mapping to molecular function: " + molecular_function_counter + " ["
1367 + ( 100 * molecular_function_counter / all_pfams_encountered.size() ) + "%]" );
1368 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1369 summary_writer.write( "# Pfams with mapping to cellular component: " + cellular_component_counter + " ["
1370 + ( 100 * cellular_component_counter / all_pfams_encountered.size() ) + "%]" );
1371 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1372 summary_writer.close();
1374 catch ( final IOException e ) {
1375 ForesterUtil.printWarningMessage( surfacing.PRG_NAME, "Failure to write: " + e );
1379 public static void writeBinaryDomainCombinationsFileForGraphAnalysis( final String[][] input_file_properties,
1380 final File output_dir,
1381 final GenomeWideCombinableDomains gwcd,
1383 final GenomeWideCombinableDomainsSortOrder dc_sort_order ) {
1384 File dc_outfile_dot = new File( input_file_properties[ i ][ 0 ]
1385 + surfacing.DOMAIN_COMBINITONS_OUTPUTFILE_SUFFIX_FOR_GRAPH_ANALYSIS );
1386 if ( output_dir != null ) {
1387 dc_outfile_dot = new File( output_dir + ForesterUtil.FILE_SEPARATOR + dc_outfile_dot );
1389 checkForOutputFileWriteability( dc_outfile_dot );
1390 final SortedSet<BinaryDomainCombination> binary_combinations = createSetOfAllBinaryDomainCombinationsPerGenome( gwcd );
1392 final BufferedWriter out_dot = new BufferedWriter( new FileWriter( dc_outfile_dot ) );
1393 for( final BinaryDomainCombination bdc : binary_combinations ) {
1394 out_dot.write( bdc.toGraphDescribingLanguage( BinaryDomainCombination.OutputFormat.DOT, null, null )
1396 out_dot.write( SurfacingConstants.NL );
1400 catch ( final IOException e ) {
1401 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1403 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote binary domain combination for \""
1404 + input_file_properties[ i ][ 0 ] + "\" (" + input_file_properties[ i ][ 1 ] + ", "
1405 + input_file_properties[ i ][ 2 ] + ") to: \"" + dc_outfile_dot + "\"" );
1408 public static void writeBinaryStatesMatrixAsListToFile( final CharacterStateMatrix<CharacterStateMatrix.GainLossStates> matrix,
1409 final CharacterStateMatrix.GainLossStates state,
1410 final String filename,
1411 final String indentifier_characters_separator,
1412 final String character_separator,
1413 final Map<String, String> descriptions ) {
1414 final File outfile = new File( filename );
1415 checkForOutputFileWriteability( outfile );
1416 final SortedSet<String> sorted_ids = new TreeSet<String>();
1417 for( int i = 0; i < matrix.getNumberOfIdentifiers(); ++i ) {
1418 sorted_ids.add( matrix.getIdentifier( i ) );
1421 final BufferedWriter out = new BufferedWriter( new FileWriter( outfile ) );
1422 for( final String id : sorted_ids ) {
1423 out.write( indentifier_characters_separator );
1424 out.write( "#" + id );
1425 out.write( indentifier_characters_separator );
1426 for( int c = 0; c < matrix.getNumberOfCharacters(); ++c ) {
1428 // using null to indicate either UNCHANGED_PRESENT or GAIN.
1429 if ( ( matrix.getState( id, c ) == state )
1430 || ( ( state == null ) && ( ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.GAIN ) || ( matrix
1431 .getState( id, c ) == CharacterStateMatrix.GainLossStates.UNCHANGED_PRESENT ) ) ) ) {
1432 out.write( matrix.getCharacter( c ) );
1433 if ( ( descriptions != null ) && !descriptions.isEmpty()
1434 && descriptions.containsKey( matrix.getCharacter( c ) ) ) {
1436 out.write( descriptions.get( matrix.getCharacter( c ) ) );
1438 out.write( character_separator );
1445 catch ( final IOException e ) {
1446 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1448 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote characters list: \"" + filename + "\"" );
1451 public static void writeBinaryStatesMatrixAsListToFileForBinaryCombinationsForGraphAnalysis( final CharacterStateMatrix<CharacterStateMatrix.GainLossStates> matrix,
1452 final CharacterStateMatrix.GainLossStates state,
1453 final String filename,
1454 final String indentifier_characters_separator,
1455 final String character_separator,
1456 final BinaryDomainCombination.OutputFormat bc_output_format ) {
1457 final File outfile = new File( filename );
1458 checkForOutputFileWriteability( outfile );
1459 final SortedSet<String> sorted_ids = new TreeSet<String>();
1460 for( int i = 0; i < matrix.getNumberOfIdentifiers(); ++i ) {
1461 sorted_ids.add( matrix.getIdentifier( i ) );
1464 final BufferedWriter out = new BufferedWriter( new FileWriter( outfile ) );
1465 for( final String id : sorted_ids ) {
1466 out.write( indentifier_characters_separator );
1467 out.write( "#" + id );
1468 out.write( indentifier_characters_separator );
1469 for( int c = 0; c < matrix.getNumberOfCharacters(); ++c ) {
1471 // using null to indicate either UNCHANGED_PRESENT or GAIN.
1472 if ( ( matrix.getState( id, c ) == state )
1473 || ( ( state == null ) && ( ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.GAIN ) || ( matrix
1474 .getState( id, c ) == CharacterStateMatrix.GainLossStates.UNCHANGED_PRESENT ) ) ) ) {
1475 BinaryDomainCombination bdc = null;
1477 bdc = BasicBinaryDomainCombination.createInstance( matrix.getCharacter( c ) );
1479 catch ( final Exception e ) {
1480 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getLocalizedMessage() );
1482 out.write( bdc.toGraphDescribingLanguage( bc_output_format, null, null ).toString() );
1483 out.write( character_separator );
1490 catch ( final IOException e ) {
1491 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1493 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote characters list: \"" + filename + "\"" );
1496 public static void writeBinaryStatesMatrixToList( final Map<DomainId, List<GoId>> domain_id_to_go_ids_map,
1497 final Map<GoId, GoTerm> go_id_to_term_map,
1498 final GoNameSpace go_namespace_limit,
1499 final boolean domain_combinations,
1500 final CharacterStateMatrix<CharacterStateMatrix.GainLossStates> matrix,
1501 final CharacterStateMatrix.GainLossStates state,
1502 final String filename,
1503 final String indentifier_characters_separator,
1504 final String character_separator,
1505 final String title_for_html,
1506 final String prefix_for_html,
1507 final Map<DomainId, Set<String>>[] domain_id_to_secondary_features_maps,
1508 final SortedSet<String> all_pfams_encountered,
1509 final SortedSet<String> pfams_gained_or_lost,
1510 final String suffix_for_per_node_events_file ) {
1511 if ( ( go_namespace_limit != null ) && ( ( go_id_to_term_map == null ) || ( go_id_to_term_map.size() < 1 ) ) ) {
1512 throw new IllegalArgumentException( "attempt to use GO namespace limit without a GO-id to term map" );
1514 else if ( ( ( domain_id_to_go_ids_map == null ) || ( domain_id_to_go_ids_map.size() < 1 ) ) ) {
1515 throw new IllegalArgumentException( "attempt to output detailed HTML without a Pfam to GO map" );
1517 else if ( ( ( go_id_to_term_map == null ) || ( go_id_to_term_map.size() < 1 ) ) ) {
1518 throw new IllegalArgumentException( "attempt to output detailed HTML without a GO-id to term map" );
1520 final File outfile = new File( filename );
1521 checkForOutputFileWriteability( outfile );
1522 final SortedSet<String> sorted_ids = new TreeSet<String>();
1523 for( int i = 0; i < matrix.getNumberOfIdentifiers(); ++i ) {
1524 sorted_ids.add( matrix.getIdentifier( i ) );
1527 final Writer out = new BufferedWriter( new FileWriter( outfile ) );
1528 final File per_node_go_mapped_domain_gain_loss_files_base_dir = createBaseDirForPerNodeDomainFiles( surfacing.BASE_DIRECTORY_PER_NODE_DOMAIN_GAIN_LOSS_FILES,
1529 domain_combinations,
1532 Writer per_node_go_mapped_domain_gain_loss_outfile_writer = null;
1533 File per_node_go_mapped_domain_gain_loss_outfile = null;
1534 int per_node_counter = 0;
1535 out.write( "<html>" );
1536 out.write( SurfacingConstants.NL );
1537 addHtmlHead( out, title_for_html );
1538 out.write( SurfacingConstants.NL );
1539 out.write( "<body>" );
1540 out.write( SurfacingConstants.NL );
1541 out.write( "<h1>" );
1542 out.write( SurfacingConstants.NL );
1543 out.write( title_for_html );
1544 out.write( SurfacingConstants.NL );
1545 out.write( "</h1>" );
1546 out.write( SurfacingConstants.NL );
1547 out.write( "<table>" );
1548 out.write( SurfacingConstants.NL );
1549 for( final String id : sorted_ids ) {
1550 final Matcher matcher = PATTERN_SP_STYLE_TAXONOMY.matcher( id );
1551 if ( matcher.matches() ) {
1554 out.write( "<tr>" );
1555 out.write( "<td>" );
1556 out.write( "<a href=\"#" + id + "\">" + id + "</a>" );
1557 out.write( "</td>" );
1558 out.write( "</tr>" );
1559 out.write( SurfacingConstants.NL );
1561 out.write( "</table>" );
1562 out.write( SurfacingConstants.NL );
1563 for( final String id : sorted_ids ) {
1564 final Matcher matcher = PATTERN_SP_STYLE_TAXONOMY.matcher( id );
1565 if ( matcher.matches() ) {
1568 out.write( SurfacingConstants.NL );
1569 out.write( "<h2>" );
1570 out.write( "<a name=\"" + id + "\">" + id + "</a>" );
1571 writeTaxonomyLinks( out, id );
1572 out.write( "</h2>" );
1573 out.write( SurfacingConstants.NL );
1574 out.write( "<table>" );
1575 out.write( SurfacingConstants.NL );
1576 out.write( "<tr>" );
1577 out.write( "<td><b>" );
1578 out.write( "Pfam domain(s)" );
1579 out.write( "</b></td><td><b>" );
1580 out.write( "GO term acc" );
1581 out.write( "</b></td><td><b>" );
1582 out.write( "GO term" );
1583 out.write( "</b></td><td><b>" );
1584 out.write( "GO namespace" );
1585 out.write( "</b></td>" );
1586 out.write( "</tr>" );
1587 out.write( SurfacingConstants.NL );
1588 out.write( "</tr>" );
1589 out.write( SurfacingConstants.NL );
1590 per_node_counter = 0;
1591 if ( matrix.getNumberOfCharacters() > 0 ) {
1592 per_node_go_mapped_domain_gain_loss_outfile = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
1593 + ForesterUtil.FILE_SEPARATOR + id + suffix_for_per_node_events_file );
1594 SurfacingUtil.checkForOutputFileWriteability( per_node_go_mapped_domain_gain_loss_outfile );
1595 per_node_go_mapped_domain_gain_loss_outfile_writer = ForesterUtil
1596 .createBufferedWriter( per_node_go_mapped_domain_gain_loss_outfile );
1599 per_node_go_mapped_domain_gain_loss_outfile = null;
1600 per_node_go_mapped_domain_gain_loss_outfile_writer = null;
1602 for( int c = 0; c < matrix.getNumberOfCharacters(); ++c ) {
1604 // using null to indicate either UNCHANGED_PRESENT or GAIN.
1605 if ( ( matrix.getState( id, c ) == state )
1606 || ( ( state == null ) && ( ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.UNCHANGED_PRESENT ) || ( matrix
1607 .getState( id, c ) == CharacterStateMatrix.GainLossStates.GAIN ) ) ) ) {
1608 final String character = matrix.getCharacter( c );
1609 String domain_0 = "";
1610 String domain_1 = "";
1611 if ( character.indexOf( BinaryDomainCombination.SEPARATOR ) > 0 ) {
1612 final String[] s = character.split( BinaryDomainCombination.SEPARATOR );
1613 if ( s.length != 2 ) {
1614 throw new AssertionError( "this should not have happened: unexpected format for domain combination: ["
1615 + character + "]" );
1621 domain_0 = character;
1623 writeDomainData( domain_id_to_go_ids_map,
1630 character_separator,
1631 domain_id_to_secondary_features_maps,
1633 all_pfams_encountered.add( domain_0 );
1634 if ( pfams_gained_or_lost != null ) {
1635 pfams_gained_or_lost.add( domain_0 );
1637 if ( !ForesterUtil.isEmpty( domain_1 ) ) {
1638 all_pfams_encountered.add( domain_1 );
1639 if ( pfams_gained_or_lost != null ) {
1640 pfams_gained_or_lost.add( domain_1 );
1643 if ( per_node_go_mapped_domain_gain_loss_outfile_writer != null ) {
1644 writeDomainsToIndividualFilePerTreeNode( per_node_go_mapped_domain_gain_loss_outfile_writer,
1651 if ( per_node_go_mapped_domain_gain_loss_outfile_writer != null ) {
1652 per_node_go_mapped_domain_gain_loss_outfile_writer.close();
1653 if ( per_node_counter < 1 ) {
1654 per_node_go_mapped_domain_gain_loss_outfile.delete();
1656 per_node_counter = 0;
1658 out.write( "</table>" );
1659 out.write( SurfacingConstants.NL );
1660 out.write( "<hr>" );
1661 out.write( SurfacingConstants.NL );
1662 } // for( final String id : sorted_ids ) {
1663 out.write( "</body>" );
1664 out.write( SurfacingConstants.NL );
1665 out.write( "</html>" );
1666 out.write( SurfacingConstants.NL );
1670 catch ( final IOException e ) {
1671 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1673 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote characters detailed HTML list: \"" + filename + "\"" );
1676 public static void writeBinaryStatesMatrixToListORIGIG( final Map<DomainId, List<GoId>> domain_id_to_go_ids_map,
1677 final Map<GoId, GoTerm> go_id_to_term_map,
1678 final GoNameSpace go_namespace_limit,
1679 final boolean domain_combinations,
1680 final CharacterStateMatrix<CharacterStateMatrix.GainLossStates> matrix,
1681 final CharacterStateMatrix.GainLossStates state,
1682 final String filename,
1683 final String indentifier_characters_separator,
1684 final String character_separator,
1685 final String title_for_html,
1686 final String prefix_for_html,
1687 final Map<DomainId, Set<String>>[] domain_id_to_secondary_features_maps,
1688 final SortedSet<String> all_pfams_encountered,
1689 final SortedSet<String> pfams_gained_or_lost,
1690 final String suffix_for_per_node_events_file ) {
1691 if ( ( go_namespace_limit != null ) && ( ( go_id_to_term_map == null ) || ( go_id_to_term_map.size() < 1 ) ) ) {
1692 throw new IllegalArgumentException( "attempt to use GO namespace limit without a GO-id to term map" );
1694 else if ( ( ( domain_id_to_go_ids_map == null ) || ( domain_id_to_go_ids_map.size() < 1 ) ) ) {
1695 throw new IllegalArgumentException( "attempt to output detailed HTML without a Pfam to GO map" );
1697 else if ( ( ( go_id_to_term_map == null ) || ( go_id_to_term_map.size() < 1 ) ) ) {
1698 throw new IllegalArgumentException( "attempt to output detailed HTML without a GO-id to term map" );
1700 final File outfile = new File( filename );
1701 checkForOutputFileWriteability( outfile );
1702 final SortedSet<String> sorted_ids = new TreeSet<String>();
1703 for( int i = 0; i < matrix.getNumberOfIdentifiers(); ++i ) {
1704 sorted_ids.add( matrix.getIdentifier( i ) );
1707 final Writer out = new BufferedWriter( new FileWriter( outfile ) );
1708 final File per_node_go_mapped_domain_gain_loss_files_base_dir = createBaseDirForPerNodeDomainFiles( surfacing.BASE_DIRECTORY_PER_NODE_DOMAIN_GAIN_LOSS_FILES,
1709 domain_combinations,
1712 Writer per_node_go_mapped_domain_gain_loss_outfile_writer = null;
1713 File per_node_go_mapped_domain_gain_loss_outfile = null;
1714 int per_node_counter = 0;
1715 out.write( "<html>" );
1716 out.write( SurfacingConstants.NL );
1717 addHtmlHead( out, title_for_html );
1718 out.write( SurfacingConstants.NL );
1719 out.write( "<body>" );
1720 out.write( SurfacingConstants.NL );
1721 out.write( "<h1>" );
1722 out.write( SurfacingConstants.NL );
1723 out.write( title_for_html );
1724 out.write( SurfacingConstants.NL );
1725 out.write( "</h1>" );
1726 out.write( SurfacingConstants.NL );
1727 out.write( "<table>" );
1728 out.write( SurfacingConstants.NL );
1729 for( final String id : sorted_ids ) {
1730 out.write( "<tr>" );
1731 out.write( "<td>" );
1732 out.write( "<a href=\"#" + id + "\">" + id + "</a>" );
1733 writeTaxonomyLinks( out, id );
1734 out.write( "</td>" );
1735 out.write( "</tr>" );
1736 out.write( SurfacingConstants.NL );
1738 out.write( "</table>" );
1739 out.write( SurfacingConstants.NL );
1740 for( final String id : sorted_ids ) {
1741 out.write( SurfacingConstants.NL );
1742 out.write( "<h2>" );
1743 out.write( "<a name=\"" + id + "\">" + id + "</a>" );
1744 writeTaxonomyLinks( out, id );
1745 out.write( "</h2>" );
1746 out.write( SurfacingConstants.NL );
1747 out.write( "<table>" );
1748 out.write( SurfacingConstants.NL );
1749 out.write( "<tr>" );
1750 out.write( "<td><b>" );
1751 out.write( "Pfam domain(s)" );
1752 out.write( "</b></td><td><b>" );
1753 out.write( "GO term acc" );
1754 out.write( "</b></td><td><b>" );
1755 out.write( "GO term" );
1756 out.write( "</b></td><td><b>" );
1757 out.write( "Penultimate GO term" );
1758 out.write( "</b></td><td><b>" );
1759 out.write( "GO namespace" );
1760 out.write( "</b></td>" );
1761 out.write( "</tr>" );
1762 out.write( SurfacingConstants.NL );
1763 out.write( "</tr>" );
1764 out.write( SurfacingConstants.NL );
1765 per_node_counter = 0;
1766 if ( matrix.getNumberOfCharacters() > 0 ) {
1767 per_node_go_mapped_domain_gain_loss_outfile = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
1768 + ForesterUtil.FILE_SEPARATOR + id + suffix_for_per_node_events_file );
1769 SurfacingUtil.checkForOutputFileWriteability( per_node_go_mapped_domain_gain_loss_outfile );
1770 per_node_go_mapped_domain_gain_loss_outfile_writer = ForesterUtil
1771 .createBufferedWriter( per_node_go_mapped_domain_gain_loss_outfile );
1774 per_node_go_mapped_domain_gain_loss_outfile = null;
1775 per_node_go_mapped_domain_gain_loss_outfile_writer = null;
1777 for( int c = 0; c < matrix.getNumberOfCharacters(); ++c ) {
1779 // using null to indicate either UNCHANGED_PRESENT or GAIN.
1780 if ( ( matrix.getState( id, c ) == state )
1781 || ( ( state == null ) && ( ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.UNCHANGED_PRESENT ) || ( matrix
1782 .getState( id, c ) == CharacterStateMatrix.GainLossStates.GAIN ) ) ) ) {
1783 final String character = matrix.getCharacter( c );
1784 String domain_0 = "";
1785 String domain_1 = "";
1786 if ( character.indexOf( BinaryDomainCombination.SEPARATOR ) > 0 ) {
1787 final String[] s = character.split( BinaryDomainCombination.SEPARATOR );
1788 if ( s.length != 2 ) {
1789 throw new AssertionError( "this should not have happened: unexpected format for domain combination: ["
1790 + character + "]" );
1796 domain_0 = character;
1798 writeDomainData( domain_id_to_go_ids_map,
1805 character_separator,
1806 domain_id_to_secondary_features_maps,
1808 all_pfams_encountered.add( domain_0 );
1809 if ( pfams_gained_or_lost != null ) {
1810 pfams_gained_or_lost.add( domain_0 );
1812 if ( !ForesterUtil.isEmpty( domain_1 ) ) {
1813 all_pfams_encountered.add( domain_1 );
1814 if ( pfams_gained_or_lost != null ) {
1815 pfams_gained_or_lost.add( domain_1 );
1818 if ( per_node_go_mapped_domain_gain_loss_outfile_writer != null ) {
1819 writeDomainsToIndividualFilePerTreeNode( per_node_go_mapped_domain_gain_loss_outfile_writer,
1826 if ( per_node_go_mapped_domain_gain_loss_outfile_writer != null ) {
1827 per_node_go_mapped_domain_gain_loss_outfile_writer.close();
1828 if ( per_node_counter < 1 ) {
1829 per_node_go_mapped_domain_gain_loss_outfile.delete();
1831 per_node_counter = 0;
1833 out.write( "</table>" );
1834 out.write( SurfacingConstants.NL );
1835 out.write( "<hr>" );
1836 out.write( SurfacingConstants.NL );
1837 } // for( final String id : sorted_ids ) {
1838 out.write( "</body>" );
1839 out.write( SurfacingConstants.NL );
1840 out.write( "</html>" );
1841 out.write( SurfacingConstants.NL );
1845 catch ( final IOException e ) {
1846 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1848 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote characters detailed HTML list: \"" + filename + "\"" );
1851 public static void writeDomainCombinationsCountsFile( final String[][] input_file_properties,
1852 final File output_dir,
1853 final Writer per_genome_domain_promiscuity_statistics_writer,
1854 final GenomeWideCombinableDomains gwcd,
1856 final GenomeWideCombinableDomains.GenomeWideCombinableDomainsSortOrder dc_sort_order ) {
1857 File dc_outfile = new File( input_file_properties[ i ][ 0 ]
1858 + surfacing.DOMAIN_COMBINITON_COUNTS_OUTPUTFILE_SUFFIX );
1859 if ( output_dir != null ) {
1860 dc_outfile = new File( output_dir + ForesterUtil.FILE_SEPARATOR + dc_outfile );
1862 checkForOutputFileWriteability( dc_outfile );
1864 final BufferedWriter out = new BufferedWriter( new FileWriter( dc_outfile ) );
1865 out.write( gwcd.toStringBuilder( dc_sort_order ).toString() );
1868 catch ( final IOException e ) {
1869 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1871 final DescriptiveStatistics stats = gwcd.getPerGenomeDomainPromiscuityStatistics();
1873 per_genome_domain_promiscuity_statistics_writer.write( input_file_properties[ i ][ 0 ] + "\t" );
1874 per_genome_domain_promiscuity_statistics_writer.write( FORMATTER_3.format( stats.arithmeticMean() ) + "\t" );
1875 if ( stats.getN() < 2 ) {
1876 per_genome_domain_promiscuity_statistics_writer.write( "n/a" + "\t" );
1879 per_genome_domain_promiscuity_statistics_writer.write( FORMATTER_3.format( stats
1880 .sampleStandardDeviation() ) + "\t" );
1882 per_genome_domain_promiscuity_statistics_writer.write( FORMATTER_3.format( stats.median() ) + "\t" );
1883 per_genome_domain_promiscuity_statistics_writer.write( ( int ) stats.getMin() + "\t" );
1884 per_genome_domain_promiscuity_statistics_writer.write( ( int ) stats.getMax() + "\t" );
1885 per_genome_domain_promiscuity_statistics_writer.write( stats.getN() + "\t" );
1886 final SortedSet<DomainId> mpds = gwcd.getMostPromiscuosDomain();
1887 for( final DomainId mpd : mpds ) {
1888 per_genome_domain_promiscuity_statistics_writer.write( mpd.getId() + " " );
1890 per_genome_domain_promiscuity_statistics_writer.write( ForesterUtil.LINE_SEPARATOR );
1892 catch ( final IOException e ) {
1893 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1895 if ( input_file_properties[ i ].length == 3 ) {
1896 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote domain combination counts for \""
1897 + input_file_properties[ i ][ 0 ] + "\" (" + input_file_properties[ i ][ 1 ] + ", "
1898 + input_file_properties[ i ][ 2 ] + ") to: \"" + dc_outfile + "\"" );
1901 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote domain combination counts for \""
1902 + input_file_properties[ i ][ 0 ] + "\" (" + input_file_properties[ i ][ 1 ] + ") to: \""
1903 + dc_outfile + "\"" );
1907 private static void writeDomainData( final Map<DomainId, List<GoId>> domain_id_to_go_ids_map,
1908 final Map<GoId, GoTerm> go_id_to_term_map,
1909 final GoNameSpace go_namespace_limit,
1911 final String domain_0,
1912 final String domain_1,
1913 final String prefix_for_html,
1914 final String character_separator_for_non_html_output,
1915 final Map<DomainId, Set<String>>[] domain_id_to_secondary_features_maps,
1916 final Set<GoId> all_go_ids ) throws IOException {
1917 boolean any_go_annotation_present = false;
1918 boolean first_has_no_go = false;
1919 int domain_count = 2; // To distinguish between domains and binary domain combinations.
1920 if ( ForesterUtil.isEmpty( domain_1 ) ) {
1923 // The following has a difficult to understand logic.
1924 for( int d = 0; d < domain_count; ++d ) {
1925 List<GoId> go_ids = null;
1926 boolean go_annotation_present = false;
1928 final DomainId domain_id = new DomainId( domain_0 );
1929 if ( domain_id_to_go_ids_map.containsKey( domain_id ) ) {
1930 go_annotation_present = true;
1931 any_go_annotation_present = true;
1932 go_ids = domain_id_to_go_ids_map.get( domain_id );
1935 first_has_no_go = true;
1939 final DomainId domain_id = new DomainId( domain_1 );
1940 if ( domain_id_to_go_ids_map.containsKey( domain_id ) ) {
1941 go_annotation_present = true;
1942 any_go_annotation_present = true;
1943 go_ids = domain_id_to_go_ids_map.get( domain_id );
1946 if ( go_annotation_present ) {
1947 boolean first = ( ( d == 0 ) || ( ( d == 1 ) && first_has_no_go ) );
1948 for( final GoId go_id : go_ids ) {
1949 out.write( "<tr>" );
1952 writeDomainIdsToHtml( out,
1956 domain_id_to_secondary_features_maps );
1959 out.write( "<td></td>" );
1961 if ( !go_id_to_term_map.containsKey( go_id ) ) {
1962 throw new IllegalArgumentException( "GO-id [" + go_id + "] not found in GO-id to GO-term map" );
1964 final GoTerm go_term = go_id_to_term_map.get( go_id );
1965 if ( ( go_namespace_limit == null ) || go_namespace_limit.equals( go_term.getGoNameSpace() ) ) {
1966 // final String top = GoUtils.getPenultimateGoTerm( go_term, go_id_to_term_map ).getName();
1967 final String go_id_str = go_id.getId();
1968 out.write( "<td>" );
1969 out.write( "<a href=\"" + SurfacingConstants.AMIGO_LINK + go_id_str
1970 + "\" target=\"amigo_window\">" + go_id_str + "</a>" );
1971 out.write( "</td><td>" );
1972 out.write( go_term.getName() );
1973 if ( domain_count == 2 ) {
1974 out.write( " (" + d + ")" );
1976 out.write( "</td><td>" );
1977 // out.write( top );
1978 // out.write( "</td><td>" );
1980 out.write( go_term.getGoNameSpace().toShortString() );
1982 out.write( "</td>" );
1983 if ( all_go_ids != null ) {
1984 all_go_ids.add( go_id );
1988 out.write( "<td>" );
1989 out.write( "</td><td>" );
1990 out.write( "</td><td>" );
1991 out.write( "</td><td>" );
1992 out.write( "</td>" );
1994 out.write( "</tr>" );
1995 out.write( SurfacingConstants.NL );
1998 } // for( int d = 0; d < domain_count; ++d )
1999 if ( !any_go_annotation_present ) {
2000 out.write( "<tr>" );
2001 writeDomainIdsToHtml( out, domain_0, domain_1, prefix_for_html, domain_id_to_secondary_features_maps );
2002 out.write( "<td>" );
2003 out.write( "</td><td>" );
2004 out.write( "</td><td>" );
2005 out.write( "</td><td>" );
2006 out.write( "</td>" );
2007 out.write( "</tr>" );
2008 out.write( SurfacingConstants.NL );
2012 private static void writeDomainIdsToHtml( final Writer out,
2013 final String domain_0,
2014 final String domain_1,
2015 final String prefix_for_detailed_html,
2016 final Map<DomainId, Set<String>>[] domain_id_to_secondary_features_maps )
2017 throws IOException {
2018 out.write( "<td>" );
2019 if ( !ForesterUtil.isEmpty( prefix_for_detailed_html ) ) {
2020 out.write( prefix_for_detailed_html );
2023 out.write( "<a href=\"" + SurfacingConstants.PFAM_FAMILY_ID_LINK + domain_0 + "\">" + domain_0 + "</a>" );
2024 out.write( "</td>" );
2027 public static DescriptiveStatistics writeDomainSimilaritiesToFile( final StringBuilder html_desc,
2028 final StringBuilder html_title,
2029 final Writer single_writer,
2030 Map<Character, Writer> split_writers,
2031 final SortedSet<DomainSimilarity> similarities,
2032 final boolean treat_as_binary,
2033 final List<Species> species_order,
2034 final PrintableDomainSimilarity.PRINT_OPTION print_option,
2035 final DomainSimilarity.DomainSimilaritySortField sort_field,
2036 final DomainSimilarity.DomainSimilarityScoring scoring,
2037 final boolean verbose ) throws IOException {
2038 final DescriptiveStatistics stats = new BasicDescriptiveStatistics();
2039 String histogram_title = null;
2040 switch ( sort_field ) {
2041 case ABS_MAX_COUNTS_DIFFERENCE:
2042 if ( treat_as_binary ) {
2043 histogram_title = "absolute counts difference:";
2046 histogram_title = "absolute (maximal) counts difference:";
2049 case MAX_COUNTS_DIFFERENCE:
2050 if ( treat_as_binary ) {
2051 histogram_title = "counts difference:";
2054 histogram_title = "(maximal) counts difference:";
2058 histogram_title = "score mean:";
2061 histogram_title = "score minimum:";
2064 histogram_title = "score maximum:";
2066 case MAX_DIFFERENCE:
2067 if ( treat_as_binary ) {
2068 histogram_title = "difference:";
2071 histogram_title = "(maximal) difference:";
2075 histogram_title = "score mean:";
2078 histogram_title = "score standard deviation:";
2081 histogram_title = "species number:";
2084 throw new AssertionError( "Unknown sort field: " + sort_field );
2086 for( final DomainSimilarity similarity : similarities ) {
2087 switch ( sort_field ) {
2088 case ABS_MAX_COUNTS_DIFFERENCE:
2089 stats.addValue( Math.abs( similarity.getMaximalDifferenceInCounts() ) );
2091 case MAX_COUNTS_DIFFERENCE:
2092 stats.addValue( similarity.getMaximalDifferenceInCounts() );
2095 stats.addValue( similarity.getMeanSimilarityScore() );
2098 stats.addValue( similarity.getMinimalSimilarityScore() );
2101 stats.addValue( similarity.getMaximalSimilarityScore() );
2103 case MAX_DIFFERENCE:
2104 stats.addValue( similarity.getMaximalDifference() );
2107 stats.addValue( similarity.getMeanSimilarityScore() );
2110 stats.addValue( similarity.getStandardDeviationOfSimilarityScore() );
2113 stats.addValue( similarity.getSpecies().size() );
2116 throw new AssertionError( "Unknown sort field: " + sort_field );
2120 // final HistogramData[] hists = new HistogramData[ 1 ];
2123 // List<HistogramDataItem> data_items = new
2124 // ArrayList<HistogramDataItem>();
2125 // double[] values = stats.getDataAsDoubleArray();
2126 // for( int i = 0; i < values.length; i++ ) {
2127 // HistogramDataItem data_item = new BasicHistogramDataItem( "", values[
2129 // data_items.add( data_item );
2133 // HistogramData hd0 = new HistogramData( "name",
2141 // hists[ 0 ] = hd0;
2143 // final HistogramsFrame hf = new HistogramsFrame( hists );
2144 // hf.setVisible( true );
2146 AsciiHistogram histo = null;
2147 if ( stats.getMin() < stats.getMin() ) {
2148 histo = new AsciiHistogram( stats, histogram_title );
2151 if ( histo != null ) {
2152 System.out.println( histo.toStringBuffer( 20, '|', 40, 5 ) );
2154 System.out.println();
2155 System.out.println( "N : " + stats.getN() );
2156 System.out.println( "Min : " + stats.getMin() );
2157 System.out.println( "Max : " + stats.getMax() );
2158 System.out.println( "Mean : " + stats.arithmeticMean() );
2159 if ( stats.getN() > 1 ) {
2160 System.out.println( "SD : " + stats.sampleStandardDeviation() );
2163 System.out.println( "SD : n/a" );
2165 System.out.println( "Median : " + stats.median() );
2166 if ( stats.getN() > 1 ) {
2167 System.out.println( "Pearsonian skewness : " + stats.pearsonianSkewness() );
2170 System.out.println( "Pearsonian skewness : n/a" );
2173 if ( ( single_writer != null ) && ( ( split_writers == null ) || split_writers.isEmpty() ) ) {
2174 split_writers = new HashMap<Character, Writer>();
2175 split_writers.put( '_', single_writer );
2177 switch ( print_option ) {
2178 case SIMPLE_TAB_DELIMITED:
2181 for( final Character key : split_writers.keySet() ) {
2182 final Writer w = split_writers.get( key );
2183 w.write( "<html>" );
2184 w.write( SurfacingConstants.NL );
2186 addHtmlHead( w, "DCs (" + html_title + ") " + key.toString().toUpperCase() );
2189 addHtmlHead( w, "DCs (" + html_title + ")" );
2191 w.write( SurfacingConstants.NL );
2192 w.write( "<body>" );
2193 w.write( SurfacingConstants.NL );
2194 w.write( html_desc.toString() );
2195 w.write( SurfacingConstants.NL );
2198 w.write( SurfacingConstants.NL );
2199 w.write( "<tt><pre>" );
2200 w.write( SurfacingConstants.NL );
2201 if ( histo != null ) {
2202 w.write( histo.toStringBuffer( 20, '|', 40, 5 ).toString() );
2203 w.write( SurfacingConstants.NL );
2205 w.write( "</pre></tt>" );
2206 w.write( SurfacingConstants.NL );
2207 w.write( "<table>" );
2208 w.write( SurfacingConstants.NL );
2209 w.write( "<tr><td>N: </td><td>" + stats.getN() + "</td></tr>" );
2210 w.write( SurfacingConstants.NL );
2211 w.write( "<tr><td>Min: </td><td>" + stats.getMin() + "</td></tr>" );
2212 w.write( SurfacingConstants.NL );
2213 w.write( "<tr><td>Max: </td><td>" + stats.getMax() + "</td></tr>" );
2214 w.write( SurfacingConstants.NL );
2215 w.write( "<tr><td>Mean: </td><td>" + stats.arithmeticMean() + "</td></tr>" );
2216 w.write( SurfacingConstants.NL );
2217 if ( stats.getN() > 1 ) {
2218 w.write( "<tr><td>SD: </td><td>" + stats.sampleStandardDeviation() + "</td></tr>" );
2221 w.write( "<tr><td>SD: </td><td>n/a</td></tr>" );
2223 w.write( SurfacingConstants.NL );
2224 w.write( "<tr><td>Median: </td><td>" + stats.median() + "</td></tr>" );
2225 w.write( SurfacingConstants.NL );
2226 if ( stats.getN() > 1 ) {
2227 w.write( "<tr><td>Pearsonian skewness: </td><td>" + stats.pearsonianSkewness() + "</td></tr>" );
2230 w.write( "<tr><td>Pearsonian skewness: </td><td>n/a</td></tr>" );
2232 w.write( SurfacingConstants.NL );
2233 w.write( "</table>" );
2234 w.write( SurfacingConstants.NL );
2236 w.write( SurfacingConstants.NL );
2238 w.write( SurfacingConstants.NL );
2240 w.write( SurfacingConstants.NL );
2241 w.write( "<table>" );
2242 w.write( SurfacingConstants.NL );
2246 for( final Writer w : split_writers.values() ) {
2247 w.write( SurfacingConstants.NL );
2249 for( final DomainSimilarity similarity : similarities ) {
2250 if ( ( species_order != null ) && !species_order.isEmpty() ) {
2251 ( ( PrintableDomainSimilarity ) similarity ).setSpeciesOrder( species_order );
2253 if ( single_writer != null ) {
2254 single_writer.write( similarity.toStringBuffer( print_option ).toString() );
2257 Writer local_writer = split_writers.get( ( similarity.getDomainId().getId().charAt( 0 ) + "" )
2258 .toLowerCase().charAt( 0 ) );
2259 if ( local_writer == null ) {
2260 local_writer = split_writers.get( '0' );
2262 local_writer.write( similarity.toStringBuffer( print_option ).toString() );
2264 for( final Writer w : split_writers.values() ) {
2265 w.write( SurfacingConstants.NL );
2268 switch ( print_option ) {
2270 for( final Writer w : split_writers.values() ) {
2271 w.write( SurfacingConstants.NL );
2272 w.write( "</table>" );
2273 w.write( SurfacingConstants.NL );
2274 w.write( "</font>" );
2275 w.write( SurfacingConstants.NL );
2276 w.write( "</body>" );
2277 w.write( SurfacingConstants.NL );
2278 w.write( "</html>" );
2279 w.write( SurfacingConstants.NL );
2283 for( final Writer w : split_writers.values() ) {
2289 private static void writeDomainsToIndividualFilePerTreeNode( final Writer individual_files_writer,
2290 final String domain_0,
2291 final String domain_1 ) throws IOException {
2292 individual_files_writer.write( domain_0 );
2293 individual_files_writer.write( ForesterUtil.LINE_SEPARATOR );
2294 if ( !ForesterUtil.isEmpty( domain_1 ) ) {
2295 individual_files_writer.write( domain_1 );
2296 individual_files_writer.write( ForesterUtil.LINE_SEPARATOR );
2300 public static void writeMatrixToFile( final CharacterStateMatrix<?> matrix,
2301 final String filename,
2302 final Format format ) {
2303 final File outfile = new File( filename );
2304 checkForOutputFileWriteability( outfile );
2306 final BufferedWriter out = new BufferedWriter( new FileWriter( outfile ) );
2307 matrix.toWriter( out, format );
2311 catch ( final IOException e ) {
2312 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
2314 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote matrix: \"" + filename + "\"" );
2317 public static void writeMatrixToFile( final File matrix_outfile, final List<DistanceMatrix> matrices ) {
2318 checkForOutputFileWriteability( matrix_outfile );
2320 final BufferedWriter out = new BufferedWriter( new FileWriter( matrix_outfile ) );
2321 for( final DistanceMatrix distance_matrix : matrices ) {
2322 out.write( distance_matrix.toStringBuffer( DistanceMatrix.Format.PHYLIP ).toString() );
2323 out.write( ForesterUtil.LINE_SEPARATOR );
2328 catch ( final IOException e ) {
2329 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
2331 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote distance matrices to \"" + matrix_outfile + "\"" );
2334 private static void writePfamsToFile( final String outfile_name, final SortedSet<String> pfams ) {
2336 final Writer writer = new BufferedWriter( new FileWriter( new File( outfile_name ) ) );
2337 for( final String pfam : pfams ) {
2338 writer.write( pfam );
2339 writer.write( ForesterUtil.LINE_SEPARATOR );
2342 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote " + pfams.size() + " pfams to [" + outfile_name
2345 catch ( final IOException e ) {
2346 ForesterUtil.printWarningMessage( surfacing.PRG_NAME, "Failure to write: " + e );
2350 public static void writePhylogenyToFile( final Phylogeny phylogeny, final String filename ) {
2351 final PhylogenyWriter writer = new PhylogenyWriter();
2353 writer.toPhyloXML( new File( filename ), phylogeny, 1 );
2355 catch ( final IOException e ) {
2356 ForesterUtil.printWarningMessage( surfacing.PRG_NAME, "failed to write phylogeny to \"" + filename + "\": "
2359 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote phylogeny to \"" + filename + "\"" );
2362 public static void writeTaxonomyLinks( final Writer writer, final String species ) throws IOException {
2363 if ( ( species.length() > 1 ) && ( species.indexOf( '_' ) < 1 ) ) {
2364 final Matcher matcher = PATTERN_SP_STYLE_TAXONOMY.matcher( species );
2365 writer.write( " [" );
2366 if ( matcher.matches() ) {
2367 writer.write( "<a href=\"" + SurfacingConstants.UNIPROT_LINK + species
2368 + "\" target=\"taxonomy_window\">uniprot</a>" );
2371 writer.write( "<a href=\"" + SurfacingConstants.EOL_LINK + species
2372 + "\" target=\"taxonomy_window\">eol</a>" );
2373 writer.write( "|" );
2374 writer.write( "<a href=\"" + SurfacingConstants.TOL_LINK + species
2375 + "\" target=\"taxonomy_window\">tol</a>" );
2377 writer.write( "]" );
2381 private static void writeToNexus( final String outfile_name,
2382 final CharacterStateMatrix<BinaryStates> matrix,
2383 final Phylogeny phylogeny ) {
2384 if ( !( matrix instanceof BasicCharacterStateMatrix ) ) {
2385 throw new IllegalArgumentException( "can only write matrices of type [" + BasicCharacterStateMatrix.class
2388 final BasicCharacterStateMatrix<BinaryStates> my_matrix = ( org.forester.evoinference.matrix.character.BasicCharacterStateMatrix<BinaryStates> ) matrix;
2389 final List<Phylogeny> phylogenies = new ArrayList<Phylogeny>( 1 );
2390 phylogenies.add( phylogeny );
2392 final BufferedWriter w = new BufferedWriter( new FileWriter( outfile_name ) );
2393 w.write( NexusConstants.NEXUS );
2394 w.write( ForesterUtil.LINE_SEPARATOR );
2395 my_matrix.writeNexusTaxaBlock( w );
2396 my_matrix.writeNexusBinaryChractersBlock( w );
2397 PhylogenyWriter.writeNexusTreesBlock( w, phylogenies, NH_CONVERSION_SUPPORT_VALUE_STYLE.NONE );
2400 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote Nexus file: \"" + outfile_name + "\"" );
2402 catch ( final IOException e ) {
2403 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
2407 private static void writeToNexus( final String outfile_name,
2408 final DomainParsimonyCalculator domain_parsimony,
2409 final Phylogeny phylogeny ) {
2410 writeToNexus( outfile_name + surfacing.NEXUS_EXTERNAL_DOMAINS,
2411 domain_parsimony.createMatrixOfDomainPresenceOrAbsence(),
2413 writeToNexus( outfile_name + surfacing.NEXUS_EXTERNAL_DOMAIN_COMBINATIONS,
2414 domain_parsimony.createMatrixOfBinaryDomainCombinationPresenceOrAbsence(),
2418 public static void domainsPerProteinsStatistics( final String genome,
2419 final List<Protein> protein_list,
2420 final DescriptiveStatistics all_genomes_domains_per_potein_stats,
2421 final SortedMap<Integer, Integer> all_genomes_domains_per_potein_histo,
2422 final SortedSet<String> domains_which_are_always_single,
2423 final SortedSet<String> domains_which_are_sometimes_single_sometimes_not,
2424 final SortedSet<String> domains_which_never_single,
2425 final Writer writer ) {
2426 final DescriptiveStatistics stats = new BasicDescriptiveStatistics();
2427 for( final Protein protein : protein_list ) {
2428 final int domains = protein.getNumberOfProteinDomains();
2429 //System.out.println( domains );
2430 stats.addValue( domains );
2431 all_genomes_domains_per_potein_stats.addValue( domains );
2432 if ( !all_genomes_domains_per_potein_histo.containsKey( domains ) ) {
2433 all_genomes_domains_per_potein_histo.put( domains, 1 );
2436 all_genomes_domains_per_potein_histo.put( domains,
2437 1 + all_genomes_domains_per_potein_histo.get( domains ) );
2439 if ( domains == 1 ) {
2440 final String domain = protein.getProteinDomain( 0 ).getDomainId().getId();
2441 if ( !domains_which_are_sometimes_single_sometimes_not.contains( domain ) ) {
2442 if ( domains_which_never_single.contains( domain ) ) {
2443 domains_which_never_single.remove( domain );
2444 domains_which_are_sometimes_single_sometimes_not.add( domain );
2447 domains_which_are_always_single.add( domain );
2451 else if ( domains > 1 ) {
2452 for( final Domain d : protein.getProteinDomains() ) {
2453 final String domain = d.getDomainId().getId();
2454 // System.out.println( domain );
2455 if ( !domains_which_are_sometimes_single_sometimes_not.contains( domain ) ) {
2456 if ( domains_which_are_always_single.contains( domain ) ) {
2457 domains_which_are_always_single.remove( domain );
2458 domains_which_are_sometimes_single_sometimes_not.add( domain );
2461 domains_which_never_single.add( domain );
2468 writer.write( genome );
2469 writer.write( "\t" );
2470 if ( stats.getN() >= 1 ) {
2471 writer.write( stats.arithmeticMean() + "" );
2472 writer.write( "\t" );
2473 if ( stats.getN() >= 2 ) {
2474 writer.write( stats.sampleStandardDeviation() + "" );
2479 writer.write( "\t" );
2480 writer.write( stats.median() + "" );
2481 writer.write( "\t" );
2482 writer.write( stats.getN() + "" );
2483 writer.write( "\t" );
2484 writer.write( stats.getMin() + "" );
2485 writer.write( "\t" );
2486 writer.write( stats.getMax() + "" );
2489 writer.write( "\t" );
2490 writer.write( "\t" );
2491 writer.write( "\t" );
2492 writer.write( "0" );
2493 writer.write( "\t" );
2494 writer.write( "\t" );
2496 writer.write( "\n" );
2498 catch ( final IOException e ) {
2499 e.printStackTrace();