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.Map.Entry;
45 import java.util.PriorityQueue;
47 import java.util.SortedMap;
48 import java.util.SortedSet;
49 import java.util.TreeMap;
50 import java.util.TreeSet;
51 import java.util.regex.Matcher;
52 import java.util.regex.Pattern;
54 import org.forester.application.surfacing;
55 import org.forester.evoinference.distance.NeighborJoining;
56 import org.forester.evoinference.matrix.character.BasicCharacterStateMatrix;
57 import org.forester.evoinference.matrix.character.CharacterStateMatrix;
58 import org.forester.evoinference.matrix.character.CharacterStateMatrix.BinaryStates;
59 import org.forester.evoinference.matrix.character.CharacterStateMatrix.Format;
60 import org.forester.evoinference.matrix.character.CharacterStateMatrix.GainLossStates;
61 import org.forester.evoinference.matrix.distance.DistanceMatrix;
62 import org.forester.go.GoId;
63 import org.forester.go.GoNameSpace;
64 import org.forester.go.GoTerm;
65 import org.forester.go.PfamToGoMapping;
66 import org.forester.io.parsers.nexus.NexusConstants;
67 import org.forester.io.writers.PhylogenyWriter;
68 import org.forester.phylogeny.Phylogeny;
69 import org.forester.phylogeny.PhylogenyMethods;
70 import org.forester.phylogeny.PhylogenyNode;
71 import org.forester.phylogeny.PhylogenyNodeI.NH_CONVERSION_SUPPORT_VALUE_STYLE;
72 import org.forester.phylogeny.data.BinaryCharacters;
73 import org.forester.phylogeny.data.Confidence;
74 import org.forester.phylogeny.iterators.PhylogenyNodeIterator;
75 import org.forester.protein.BasicProtein;
76 import org.forester.protein.BinaryDomainCombination;
77 import org.forester.protein.Domain;
78 import org.forester.protein.DomainId;
79 import org.forester.protein.Protein;
80 import org.forester.species.Species;
81 import org.forester.surfacing.DomainSimilarityCalculator.Detailedness;
82 import org.forester.surfacing.DomainSimilarityCalculator.GoAnnotationOutput;
83 import org.forester.surfacing.GenomeWideCombinableDomains.GenomeWideCombinableDomainsSortOrder;
84 import org.forester.util.AsciiHistogram;
85 import org.forester.util.BasicDescriptiveStatistics;
86 import org.forester.util.BasicTable;
87 import org.forester.util.BasicTableParser;
88 import org.forester.util.DescriptiveStatistics;
89 import org.forester.util.ForesterUtil;
91 public final class SurfacingUtil {
93 private final static NumberFormat FORMATTER = new DecimalFormat( "0.0E0" );
94 private final static NumberFormat FORMATTER_3 = new DecimalFormat( "0.000" );
95 private static final Comparator<Domain> ASCENDING_CONFIDENCE_VALUE_ORDER = new Comparator<Domain>() {
98 public int compare( final Domain d1,
100 if ( d1.getPerSequenceEvalue() < d2
101 .getPerSequenceEvalue() ) {
105 .getPerSequenceEvalue() > d2
106 .getPerSequenceEvalue() ) {
110 return d1.compareTo( d2 );
114 public final static Pattern PATTERN_SP_STYLE_TAXONOMY = Pattern.compile( "^[A-Z0-9]{3,5}$" );
115 private static final boolean USE_LAST = true;
117 private SurfacingUtil() {
118 // Hidden constructor.
121 public static void addAllBinaryDomainCombinationToSet( final GenomeWideCombinableDomains genome,
122 final SortedSet<BinaryDomainCombination> binary_domain_combinations ) {
123 final SortedMap<DomainId, CombinableDomains> all_cd = genome.getAllCombinableDomainsIds();
124 for( final DomainId domain_id : all_cd.keySet() ) {
125 binary_domain_combinations.addAll( all_cd.get( domain_id ).toBinaryDomainCombinations() );
129 public static void addAllDomainIdsToSet( final GenomeWideCombinableDomains genome,
130 final SortedSet<DomainId> domain_ids ) {
131 final SortedSet<DomainId> domains = genome.getAllDomainIds();
132 for( final DomainId domain : domains ) {
133 domain_ids.add( domain );
137 public static void addHtmlHead( final Writer w, final String title ) throws IOException {
138 w.write( SurfacingConstants.NL );
140 w.write( "<title>" );
142 w.write( "</title>" );
143 w.write( SurfacingConstants.NL );
144 w.write( "<style>" );
145 w.write( SurfacingConstants.NL );
146 w.write( "a:visited { color : #6633FF; text-decoration : none; }" );
147 w.write( SurfacingConstants.NL );
148 w.write( "a:link { color : #6633FF; text-decoration : none; }" );
149 w.write( SurfacingConstants.NL );
150 w.write( "a:active { color : #99FF00; text-decoration : none; }" );
151 w.write( SurfacingConstants.NL );
152 w.write( "a:hover { color : #FFFFFF; background-color : #99FF00; text-decoration : none; }" );
153 w.write( SurfacingConstants.NL );
154 w.write( "td { text-align: left; vertical-align: top; font-family: Verdana, Arial, Helvetica; font-size: 8pt}" );
155 w.write( SurfacingConstants.NL );
156 w.write( "h1 { color : #0000FF; font-family: Verdana, Arial, Helvetica; font-size: 18pt; font-weight: bold }" );
157 w.write( SurfacingConstants.NL );
158 w.write( "h2 { color : #0000FF; font-family: Verdana, Arial, Helvetica; font-size: 16pt; font-weight: bold }" );
159 w.write( SurfacingConstants.NL );
160 w.write( "</style>" );
161 w.write( SurfacingConstants.NL );
162 w.write( "</head>" );
163 w.write( SurfacingConstants.NL );
166 public static DescriptiveStatistics calculateDescriptiveStatisticsForMeanValues( final Set<DomainSimilarity> similarities ) {
167 final DescriptiveStatistics stats = new BasicDescriptiveStatistics();
168 for( final DomainSimilarity similarity : similarities ) {
169 stats.addValue( similarity.getMeanSimilarityScore() );
174 private static void calculateIndependentDomainCombinationGains( final Phylogeny local_phylogeny_l,
175 final String outfilename_for_counts,
176 final String outfilename_for_dc,
177 final String outfilename_for_dc_for_go_mapping,
178 final String outfilename_for_dc_for_go_mapping_unique,
179 final String outfilename_for_rank_counts,
180 final String outfilename_for_ancestor_species_counts,
181 final Map<String, DescriptiveStatistics> protein_length_stats_by_dc,
182 final Map<String, DescriptiveStatistics> domain_number_stats_by_dc ) {
184 final BufferedWriter out_counts = new BufferedWriter( new FileWriter( outfilename_for_counts ) );
185 final BufferedWriter out_dc = new BufferedWriter( new FileWriter( outfilename_for_dc ) );
186 final BufferedWriter out_dc_for_go_mapping = new BufferedWriter( new FileWriter( outfilename_for_dc_for_go_mapping ) );
187 final BufferedWriter out_dc_for_go_mapping_unique = new BufferedWriter( new FileWriter( outfilename_for_dc_for_go_mapping_unique ) );
188 final SortedMap<String, Integer> dc_gain_counts = new TreeMap<String, Integer>();
189 for( final PhylogenyNodeIterator it = local_phylogeny_l.iteratorPostorder(); it.hasNext(); ) {
190 final PhylogenyNode n = it.next();
191 final Set<String> gained_dc = n.getNodeData().getBinaryCharacters().getGainedCharacters();
192 for( final String dc : gained_dc ) {
193 if ( dc_gain_counts.containsKey( dc ) ) {
194 dc_gain_counts.put( dc, dc_gain_counts.get( dc ) + 1 );
197 dc_gain_counts.put( dc, 1 );
201 final SortedMap<Integer, Integer> histogram = new TreeMap<Integer, Integer>();
202 final SortedMap<Integer, StringBuilder> domain_lists = new TreeMap<Integer, StringBuilder>();
203 final SortedMap<Integer, DescriptiveStatistics> dc_reapp_counts_to_protein_length_stats = new TreeMap<Integer, DescriptiveStatistics>();
204 final SortedMap<Integer, DescriptiveStatistics> dc_reapp_counts_to_domain_number_stats = new TreeMap<Integer, DescriptiveStatistics>();
205 final SortedMap<Integer, PriorityQueue<String>> domain_lists_go = new TreeMap<Integer, PriorityQueue<String>>();
206 final SortedMap<Integer, SortedSet<String>> domain_lists_go_unique = new TreeMap<Integer, SortedSet<String>>();
207 final Set<String> dcs = dc_gain_counts.keySet();
208 final SortedSet<String> more_than_once = new TreeSet<String>();
209 for( final String dc : dcs ) {
210 final int count = dc_gain_counts.get( dc );
211 if ( histogram.containsKey( count ) ) {
212 histogram.put( count, histogram.get( count ) + 1 );
213 domain_lists.get( count ).append( ", " + dc );
214 domain_lists_go.get( count ).addAll( splitDomainCombination( dc ) );
215 domain_lists_go_unique.get( count ).addAll( splitDomainCombination( dc ) );
218 histogram.put( count, 1 );
219 domain_lists.put( count, new StringBuilder( dc ) );
220 final PriorityQueue<String> q = new PriorityQueue<String>();
221 q.addAll( splitDomainCombination( dc ) );
222 domain_lists_go.put( count, q );
223 final SortedSet<String> set = new TreeSet<String>();
224 set.addAll( splitDomainCombination( dc ) );
225 domain_lists_go_unique.put( count, set );
227 if ( protein_length_stats_by_dc != null ) {
228 if ( !dc_reapp_counts_to_protein_length_stats.containsKey( count ) ) {
229 dc_reapp_counts_to_protein_length_stats.put( count, new BasicDescriptiveStatistics() );
231 dc_reapp_counts_to_protein_length_stats.get( count ).addValue( protein_length_stats_by_dc.get( dc )
234 if ( domain_number_stats_by_dc != null ) {
235 if ( !dc_reapp_counts_to_domain_number_stats.containsKey( count ) ) {
236 dc_reapp_counts_to_domain_number_stats.put( count, new BasicDescriptiveStatistics() );
238 dc_reapp_counts_to_domain_number_stats.get( count ).addValue( domain_number_stats_by_dc.get( dc )
242 more_than_once.add( dc );
245 final Set<Integer> histogram_keys = histogram.keySet();
246 for( final Integer histogram_key : histogram_keys ) {
247 final int count = histogram.get( histogram_key );
248 final StringBuilder dc = domain_lists.get( histogram_key );
249 out_counts.write( histogram_key + "\t" + count + ForesterUtil.LINE_SEPARATOR );
250 out_dc.write( histogram_key + "\t" + dc + ForesterUtil.LINE_SEPARATOR );
251 out_dc_for_go_mapping.write( "#" + histogram_key + ForesterUtil.LINE_SEPARATOR );
252 final Object[] sorted = domain_lists_go.get( histogram_key ).toArray();
253 Arrays.sort( sorted );
254 for( final Object domain : sorted ) {
255 out_dc_for_go_mapping.write( domain + ForesterUtil.LINE_SEPARATOR );
257 out_dc_for_go_mapping_unique.write( "#" + histogram_key + ForesterUtil.LINE_SEPARATOR );
258 for( final String domain : domain_lists_go_unique.get( histogram_key ) ) {
259 out_dc_for_go_mapping_unique.write( domain + ForesterUtil.LINE_SEPARATOR );
264 out_dc_for_go_mapping.close();
265 out_dc_for_go_mapping_unique.close();
267 final SortedMap<String, Integer> lca_rank_counts = new TreeMap<String, Integer>();
268 final SortedMap<String, Integer> lca_ancestor_species_counts = new TreeMap<String, Integer>();
269 for( final String dc : more_than_once ) {
270 final List<PhylogenyNode> nodes = new ArrayList<PhylogenyNode>();
271 for( final PhylogenyNodeIterator it = local_phylogeny_l.iteratorExternalForward(); it.hasNext(); ) {
272 final PhylogenyNode n = it.next();
273 if ( n.getNodeData().getBinaryCharacters().getGainedCharacters().contains( dc ) ) {
277 for( int i = 0; i < nodes.size() - 1; ++i ) {
278 for( int j = i + 1; j < nodes.size(); ++j ) {
279 final PhylogenyNode lca = PhylogenyMethods.getInstance().obtainLCA( nodes.get( i ),
281 String rank = "unknown";
282 if ( lca.getNodeData().isHasTaxonomy()
283 && !ForesterUtil.isEmpty( lca.getNodeData().getTaxonomy().getRank() ) ) {
284 rank = lca.getNodeData().getTaxonomy().getRank();
286 addToCountMap( lca_rank_counts, rank );
288 if ( lca.getNodeData().isHasTaxonomy()
289 && !ForesterUtil.isEmpty( lca.getNodeData().getTaxonomy().getScientificName() ) ) {
290 lca_species = lca.getNodeData().getTaxonomy().getScientificName();
292 else if ( lca.getNodeData().isHasTaxonomy()
293 && !ForesterUtil.isEmpty( lca.getNodeData().getTaxonomy().getCommonName() ) ) {
294 lca_species = lca.getNodeData().getTaxonomy().getCommonName();
297 lca_species = lca.getName();
299 addToCountMap( lca_ancestor_species_counts, lca_species );
303 final BufferedWriter out_for_rank_counts = new BufferedWriter( new FileWriter( outfilename_for_rank_counts ) );
304 final BufferedWriter out_for_ancestor_species_counts = new BufferedWriter( new FileWriter( outfilename_for_ancestor_species_counts ) );
305 ForesterUtil.map2writer( out_for_rank_counts, lca_rank_counts, "\t", ForesterUtil.LINE_SEPARATOR );
306 ForesterUtil.map2writer( out_for_ancestor_species_counts,
307 lca_ancestor_species_counts,
309 ForesterUtil.LINE_SEPARATOR );
310 out_for_rank_counts.close();
311 out_for_ancestor_species_counts.close();
312 System.out.println( "Lengths: " );
313 if ( protein_length_stats_by_dc != null ) {
314 for( final Entry<?, ?> entry : dc_reapp_counts_to_protein_length_stats.entrySet() ) {
315 System.out.println( entry.getKey().toString() );
316 System.out.println( ": " );
317 System.out.println( entry.getValue().toString() );
320 System.out.println( "Number of domains: " );
321 if ( domain_number_stats_by_dc != null ) {
322 for( final Entry<?, ?> entry : dc_reapp_counts_to_domain_number_stats.entrySet() ) {
323 System.out.println( entry.getKey().toString() );
324 System.out.println( ": " );
325 System.out.println( entry.getValue().toString() );
329 catch ( final IOException e ) {
330 ForesterUtil.printWarningMessage( surfacing.PRG_NAME, "Failure to write: " + e );
332 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote independent domain combination gains fitch counts to ["
333 + outfilename_for_counts + "]" );
334 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote independent domain combination gains fitch lists to ["
335 + outfilename_for_dc + "]" );
336 ForesterUtil.programMessage( surfacing.PRG_NAME,
337 "Wrote independent domain combination gains fitch lists to (for GO mapping) ["
338 + outfilename_for_dc_for_go_mapping + "]" );
339 ForesterUtil.programMessage( surfacing.PRG_NAME,
340 "Wrote independent domain combination gains fitch lists to (for GO mapping, unique) ["
341 + outfilename_for_dc_for_go_mapping_unique + "]" );
344 private final static void addToCountMap( final Map<String, Integer> map, final String s ) {
345 if ( map.containsKey( s ) ) {
346 map.put( s, map.get( s ) + 1 );
353 public static int calculateOverlap( final Domain domain, final List<Boolean> covered_positions ) {
354 int overlap_count = 0;
355 for( int i = domain.getFrom(); i <= domain.getTo(); ++i ) {
356 if ( ( i < covered_positions.size() ) && ( covered_positions.get( i ) == true ) ) {
360 return overlap_count;
363 public static void checkForOutputFileWriteability( final File outfile ) {
364 final String error = ForesterUtil.isWritableFile( outfile );
365 if ( !ForesterUtil.isEmpty( error ) ) {
366 ForesterUtil.fatalError( surfacing.PRG_NAME, error );
370 private static SortedSet<String> collectAllDomainsChangedOnSubtree( final PhylogenyNode subtree_root,
371 final boolean get_gains ) {
372 final SortedSet<String> domains = new TreeSet<String>();
373 for( final PhylogenyNode descendant : PhylogenyMethods.getAllDescendants( subtree_root ) ) {
374 final BinaryCharacters chars = descendant.getNodeData().getBinaryCharacters();
376 domains.addAll( chars.getGainedCharacters() );
379 domains.addAll( chars.getLostCharacters() );
385 public static void collectChangedDomainCombinationsFromBinaryStatesMatrixAsListToFile( final CharacterStateMatrix<CharacterStateMatrix.GainLossStates> matrix,
386 final BinaryDomainCombination.DomainCombinationType dc_type,
387 final List<BinaryDomainCombination> all_binary_domains_combination_gained,
388 final boolean get_gains ) {
389 final SortedSet<String> sorted_ids = new TreeSet<String>();
390 for( int i = 0; i < matrix.getNumberOfIdentifiers(); ++i ) {
391 sorted_ids.add( matrix.getIdentifier( i ) );
393 for( final String id : sorted_ids ) {
394 for( int c = 0; c < matrix.getNumberOfCharacters(); ++c ) {
395 if ( ( get_gains && ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.GAIN ) )
396 || ( !get_gains && ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.LOSS ) ) ) {
397 if ( dc_type == BinaryDomainCombination.DomainCombinationType.DIRECTED_ADJACTANT ) {
398 all_binary_domains_combination_gained.add( AdjactantDirectedBinaryDomainCombination
399 .createInstance( matrix.getCharacter( c ) ) );
401 else if ( dc_type == BinaryDomainCombination.DomainCombinationType.DIRECTED ) {
402 all_binary_domains_combination_gained.add( DirectedBinaryDomainCombination
403 .createInstance( matrix.getCharacter( c ) ) );
406 all_binary_domains_combination_gained.add( BasicBinaryDomainCombination.createInstance( matrix
407 .getCharacter( c ) ) );
414 private static File createBaseDirForPerNodeDomainFiles( final String base_dir,
415 final boolean domain_combinations,
416 final CharacterStateMatrix.GainLossStates state,
417 final String outfile ) {
418 File per_node_go_mapped_domain_gain_loss_files_base_dir = new File( new File( outfile ).getParent()
419 + ForesterUtil.FILE_SEPARATOR + base_dir );
420 if ( !per_node_go_mapped_domain_gain_loss_files_base_dir.exists() ) {
421 per_node_go_mapped_domain_gain_loss_files_base_dir.mkdir();
423 if ( domain_combinations ) {
424 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
425 + ForesterUtil.FILE_SEPARATOR + "DC" );
428 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
429 + ForesterUtil.FILE_SEPARATOR + "DOMAINS" );
431 if ( !per_node_go_mapped_domain_gain_loss_files_base_dir.exists() ) {
432 per_node_go_mapped_domain_gain_loss_files_base_dir.mkdir();
434 if ( state == GainLossStates.GAIN ) {
435 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
436 + ForesterUtil.FILE_SEPARATOR + "GAINS" );
438 else if ( state == GainLossStates.LOSS ) {
439 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
440 + ForesterUtil.FILE_SEPARATOR + "LOSSES" );
443 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
444 + ForesterUtil.FILE_SEPARATOR + "PRESENT" );
446 if ( !per_node_go_mapped_domain_gain_loss_files_base_dir.exists() ) {
447 per_node_go_mapped_domain_gain_loss_files_base_dir.mkdir();
449 return per_node_go_mapped_domain_gain_loss_files_base_dir;
452 public static Map<DomainId, List<GoId>> createDomainIdToGoIdMap( final List<PfamToGoMapping> pfam_to_go_mappings ) {
453 final Map<DomainId, List<GoId>> domain_id_to_go_ids_map = new HashMap<DomainId, List<GoId>>( pfam_to_go_mappings
455 for( final PfamToGoMapping pfam_to_go : pfam_to_go_mappings ) {
456 if ( !domain_id_to_go_ids_map.containsKey( pfam_to_go.getKey() ) ) {
457 domain_id_to_go_ids_map.put( pfam_to_go.getKey(), new ArrayList<GoId>() );
459 domain_id_to_go_ids_map.get( pfam_to_go.getKey() ).add( pfam_to_go.getValue() );
461 return domain_id_to_go_ids_map;
464 public static Map<DomainId, Set<String>> createDomainIdToSecondaryFeaturesMap( final File secondary_features_map_file )
466 final BasicTable<String> primary_table = BasicTableParser.parse( secondary_features_map_file, "\t" );
467 final Map<DomainId, Set<String>> map = new TreeMap<DomainId, Set<String>>();
468 for( int r = 0; r < primary_table.getNumberOfRows(); ++r ) {
469 final DomainId domain_id = new DomainId( primary_table.getValue( 0, r ) );
470 if ( !map.containsKey( domain_id ) ) {
471 map.put( domain_id, new HashSet<String>() );
473 map.get( domain_id ).add( primary_table.getValue( 1, r ) );
478 public static Phylogeny createNjTreeBasedOnMatrixToFile( final File nj_tree_outfile, final DistanceMatrix distance ) {
479 checkForOutputFileWriteability( nj_tree_outfile );
480 final NeighborJoining nj = NeighborJoining.createInstance();
481 final Phylogeny phylogeny = nj.execute( distance );
482 phylogeny.setName( nj_tree_outfile.getName() );
483 writePhylogenyToFile( phylogeny, nj_tree_outfile.toString() );
487 private static SortedSet<BinaryDomainCombination> createSetOfAllBinaryDomainCombinationsPerGenome( final GenomeWideCombinableDomains gwcd ) {
488 final SortedMap<DomainId, CombinableDomains> cds = gwcd.getAllCombinableDomainsIds();
489 final SortedSet<BinaryDomainCombination> binary_combinations = new TreeSet<BinaryDomainCombination>();
490 for( final DomainId domain_id : cds.keySet() ) {
491 final CombinableDomains cd = cds.get( domain_id );
492 binary_combinations.addAll( cd.toBinaryDomainCombinations() );
494 return binary_combinations;
497 public static void decoratePrintableDomainSimilarities( final SortedSet<DomainSimilarity> domain_similarities,
498 final Detailedness detailedness,
499 final GoAnnotationOutput go_annotation_output,
500 final Map<GoId, GoTerm> go_id_to_term_map,
501 final GoNameSpace go_namespace_limit ) {
502 if ( ( go_namespace_limit != null ) && ( ( go_id_to_term_map == null ) || go_id_to_term_map.isEmpty() ) ) {
503 throw new IllegalArgumentException( "attempt to use a GO namespace limit without a GO id to term map" );
505 for( final DomainSimilarity domain_similarity : domain_similarities ) {
506 if ( domain_similarity instanceof PrintableDomainSimilarity ) {
507 final PrintableDomainSimilarity printable_domain_similarity = ( PrintableDomainSimilarity ) domain_similarity;
508 printable_domain_similarity.setDetailedness( detailedness );
509 printable_domain_similarity.setGoAnnotationOutput( go_annotation_output );
510 printable_domain_similarity.setGoIdToTermMap( go_id_to_term_map );
511 printable_domain_similarity.setGoNamespaceLimit( go_namespace_limit );
516 public static void executeDomainLengthAnalysis( final String[][] input_file_properties,
517 final int number_of_genomes,
518 final DomainLengthsTable domain_lengths_table,
519 final File outfile ) throws IOException {
520 final DecimalFormat df = new DecimalFormat( "#.00" );
521 checkForOutputFileWriteability( outfile );
522 final BufferedWriter out = new BufferedWriter( new FileWriter( outfile ) );
523 out.write( "MEAN BASED STATISTICS PER SPECIES" );
524 out.write( ForesterUtil.LINE_SEPARATOR );
525 out.write( domain_lengths_table.createMeanBasedStatisticsPerSpeciesTable().toString() );
526 out.write( ForesterUtil.LINE_SEPARATOR );
527 out.write( ForesterUtil.LINE_SEPARATOR );
528 final List<DomainLengths> domain_lengths_list = domain_lengths_table.getDomainLengthsList();
529 out.write( "OUTLIER SPECIES PER DOMAIN (Z>=1.5)" );
530 out.write( ForesterUtil.LINE_SEPARATOR );
531 for( final DomainLengths domain_lengths : domain_lengths_list ) {
532 final List<Species> species_list = domain_lengths.getMeanBasedOutlierSpecies( 1.5 );
533 if ( species_list.size() > 0 ) {
534 out.write( domain_lengths.getDomainId() + "\t" );
535 for( final Species species : species_list ) {
536 out.write( species + "\t" );
538 out.write( ForesterUtil.LINE_SEPARATOR );
539 // DescriptiveStatistics stats_for_domain = domain_lengths
540 // .calculateMeanBasedStatistics();
541 //AsciiHistogram histo = new AsciiHistogram( stats_for_domain );
542 //System.out.println( histo.toStringBuffer( 40, '=', 60, 4 ).toString() );
545 out.write( ForesterUtil.LINE_SEPARATOR );
546 out.write( ForesterUtil.LINE_SEPARATOR );
547 out.write( "OUTLIER SPECIES (Z 1.0)" );
548 out.write( ForesterUtil.LINE_SEPARATOR );
549 final DescriptiveStatistics stats_for_all_species = domain_lengths_table
550 .calculateMeanBasedStatisticsForAllSpecies();
551 out.write( stats_for_all_species.asSummary() );
552 out.write( ForesterUtil.LINE_SEPARATOR );
553 final AsciiHistogram histo = new AsciiHistogram( stats_for_all_species );
554 out.write( histo.toStringBuffer( 40, '=', 60, 4 ).toString() );
555 out.write( ForesterUtil.LINE_SEPARATOR );
556 final double population_sd = stats_for_all_species.sampleStandardDeviation();
557 final double population_mean = stats_for_all_species.arithmeticMean();
558 for( final Species species : domain_lengths_table.getSpecies() ) {
559 final double x = domain_lengths_table.calculateMeanBasedStatisticsForSpecies( species ).arithmeticMean();
560 final double z = ( x - population_mean ) / population_sd;
561 out.write( species + "\t" + z );
562 out.write( ForesterUtil.LINE_SEPARATOR );
564 out.write( ForesterUtil.LINE_SEPARATOR );
565 for( final Species species : domain_lengths_table.getSpecies() ) {
566 final DescriptiveStatistics stats_for_species = domain_lengths_table
567 .calculateMeanBasedStatisticsForSpecies( species );
568 final double x = stats_for_species.arithmeticMean();
569 final double z = ( x - population_mean ) / population_sd;
570 if ( ( z <= -1.0 ) || ( z >= 1.0 ) ) {
571 out.write( species + "\t" + df.format( z ) + "\t" + stats_for_species.asSummary() );
572 out.write( ForesterUtil.LINE_SEPARATOR );
576 // final List<HistogramData> histogram_datas = new ArrayList<HistogramData>();
577 // for( int i = 0; i < number_of_genomes; ++i ) {
578 // final Species species = new BasicSpecies( input_file_properties[ i ][ 0 ] );
580 // .add( new HistogramData( species.toString(), domain_lengths_table
581 // .calculateMeanBasedStatisticsForSpecies( species )
582 // .getDataAsDoubleArray(), 5, 600, null, 60 ) );
584 // final HistogramsFrame hf = new HistogramsFrame( histogram_datas );
585 // hf.setVisible( true );
591 * @param all_binary_domains_combination_lost_fitch
592 * @param consider_directedness_and_adjacency_for_bin_combinations
593 * @param all_binary_domains_combination_gained if null ignored, otherwise this is to list all binary domain combinations
594 * which were gained under unweighted (Fitch) parsimony.
596 public static void executeParsimonyAnalysis( final long random_number_seed_for_fitch_parsimony,
597 final boolean radomize_fitch_parsimony,
598 final String outfile_name,
599 final DomainParsimonyCalculator domain_parsimony,
600 final Phylogeny phylogeny,
601 final Map<DomainId, List<GoId>> domain_id_to_go_ids_map,
602 final Map<GoId, GoTerm> go_id_to_term_map,
603 final GoNameSpace go_namespace_limit,
604 final String parameters_str,
605 final Map<DomainId, Set<String>>[] domain_id_to_secondary_features_maps,
606 final SortedSet<DomainId> positive_filter,
607 final boolean output_binary_domain_combinations_for_graphs,
608 final List<BinaryDomainCombination> all_binary_domains_combination_gained_fitch,
609 final List<BinaryDomainCombination> all_binary_domains_combination_lost_fitch,
610 final BinaryDomainCombination.DomainCombinationType dc_type,
611 final Map<String, DescriptiveStatistics> protein_length_stats_by_dc,
612 final Map<String, DescriptiveStatistics> domain_number_stats_by_dc ) {
613 final String sep = ForesterUtil.LINE_SEPARATOR + "###################" + ForesterUtil.LINE_SEPARATOR;
614 final String date_time = ForesterUtil.getCurrentDateTime();
615 final SortedSet<String> all_pfams_encountered = new TreeSet<String>();
616 final SortedSet<String> all_pfams_gained_as_domains = new TreeSet<String>();
617 final SortedSet<String> all_pfams_lost_as_domains = new TreeSet<String>();
618 final SortedSet<String> all_pfams_gained_as_dom_combinations = new TreeSet<String>();
619 final SortedSet<String> all_pfams_lost_as_dom_combinations = new TreeSet<String>();
620 writeToNexus( outfile_name, domain_parsimony, phylogeny );
623 Phylogeny local_phylogeny_l = phylogeny.copy();
624 if ( ( positive_filter != null ) && ( positive_filter.size() > 0 ) ) {
625 domain_parsimony.executeDolloParsimonyOnDomainPresence( positive_filter );
628 domain_parsimony.executeDolloParsimonyOnDomainPresence();
630 SurfacingUtil.writeMatrixToFile( domain_parsimony.getGainLossMatrix(), outfile_name
631 + surfacing.PARSIMONY_OUTPUT_GL_SUFFIX_DOLLO_DOMAINS, Format.FORESTER );
632 SurfacingUtil.writeMatrixToFile( domain_parsimony.getGainLossCountsMatrix(), outfile_name
633 + surfacing.PARSIMONY_OUTPUT_GL_COUNTS_SUFFIX_DOLLO_DOMAINS, Format.FORESTER );
634 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
635 CharacterStateMatrix.GainLossStates.GAIN,
636 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_GAINS_D,
638 ForesterUtil.LINE_SEPARATOR,
640 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
641 CharacterStateMatrix.GainLossStates.LOSS,
642 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_LOSSES_D,
644 ForesterUtil.LINE_SEPARATOR,
646 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(), null, outfile_name
647 + surfacing.PARSIMONY_OUTPUT_DOLLO_PRESENT_D, sep, ForesterUtil.LINE_SEPARATOR, null );
649 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
653 domain_parsimony.getGainLossMatrix(),
654 CharacterStateMatrix.GainLossStates.GAIN,
655 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_GAINS_HTML_D,
657 ForesterUtil.LINE_SEPARATOR,
658 "Dollo Parsimony | Gains | Domains",
660 domain_id_to_secondary_features_maps,
661 all_pfams_encountered,
662 all_pfams_gained_as_domains,
664 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
668 domain_parsimony.getGainLossMatrix(),
669 CharacterStateMatrix.GainLossStates.LOSS,
670 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_LOSSES_HTML_D,
672 ForesterUtil.LINE_SEPARATOR,
673 "Dollo Parsimony | Losses | Domains",
675 domain_id_to_secondary_features_maps,
676 all_pfams_encountered,
677 all_pfams_lost_as_domains,
679 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
683 domain_parsimony.getGainLossMatrix(),
685 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_PRESENT_HTML_D,
687 ForesterUtil.LINE_SEPARATOR,
688 "Dollo Parsimony | Present | Domains",
690 domain_id_to_secondary_features_maps,
691 all_pfams_encountered,
693 "_dollo_present_d" );
694 preparePhylogeny( local_phylogeny_l,
697 "Dollo parsimony on domain presence/absence",
698 "dollo_on_domains_" + outfile_name,
700 SurfacingUtil.writePhylogenyToFile( local_phylogeny_l, outfile_name
701 + surfacing.DOMAINS_PARSIMONY_TREE_OUTPUT_SUFFIX_DOLLO );
703 writeAllDomainsChangedOnAllSubtrees( local_phylogeny_l, true, outfile_name, "_dollo_all_gains_d" );
704 writeAllDomainsChangedOnAllSubtrees( local_phylogeny_l, false, outfile_name, "_dollo_all_losses_d" );
706 catch ( final IOException e ) {
708 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getLocalizedMessage() );
710 if ( domain_parsimony.calculateNumberOfBinaryDomainCombination() > 0 ) {
711 // FITCH DOMAIN COMBINATIONS
712 // -------------------------
713 local_phylogeny_l = phylogeny.copy();
714 String randomization = "no";
715 if ( radomize_fitch_parsimony ) {
716 domain_parsimony.executeFitchParsimonyOnBinaryDomainCombintion( random_number_seed_for_fitch_parsimony );
717 randomization = "yes, seed = " + random_number_seed_for_fitch_parsimony;
720 domain_parsimony.executeFitchParsimonyOnBinaryDomainCombintion( USE_LAST );
722 SurfacingUtil.writeMatrixToFile( domain_parsimony.getGainLossMatrix(), outfile_name
723 + surfacing.PARSIMONY_OUTPUT_GL_SUFFIX_FITCH_BINARY_COMBINATIONS, Format.FORESTER );
724 SurfacingUtil.writeMatrixToFile( domain_parsimony.getGainLossCountsMatrix(), outfile_name
725 + surfacing.PARSIMONY_OUTPUT_GL_COUNTS_SUFFIX_FITCH_BINARY_COMBINATIONS, Format.FORESTER );
727 .writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
728 CharacterStateMatrix.GainLossStates.GAIN,
729 outfile_name + surfacing.PARSIMONY_OUTPUT_FITCH_GAINS_BC,
731 ForesterUtil.LINE_SEPARATOR,
733 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
734 CharacterStateMatrix.GainLossStates.LOSS,
736 + surfacing.PARSIMONY_OUTPUT_FITCH_LOSSES_BC,
738 ForesterUtil.LINE_SEPARATOR,
740 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(), null, outfile_name
741 + surfacing.PARSIMONY_OUTPUT_FITCH_PRESENT_BC, sep, ForesterUtil.LINE_SEPARATOR, null );
742 if ( all_binary_domains_combination_gained_fitch != null ) {
743 collectChangedDomainCombinationsFromBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
745 all_binary_domains_combination_gained_fitch,
748 if ( all_binary_domains_combination_lost_fitch != null ) {
749 collectChangedDomainCombinationsFromBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
751 all_binary_domains_combination_lost_fitch,
754 if ( output_binary_domain_combinations_for_graphs ) {
756 .writeBinaryStatesMatrixAsListToFileForBinaryCombinationsForGraphAnalysis( domain_parsimony
757 .getGainLossMatrix(),
760 + surfacing.PARSIMONY_OUTPUT_FITCH_PRESENT_BC_OUTPUTFILE_SUFFIX_FOR_GRAPH_ANALYSIS,
762 ForesterUtil.LINE_SEPARATOR,
763 BinaryDomainCombination.OutputFormat.DOT );
766 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
770 domain_parsimony.getGainLossMatrix(),
771 CharacterStateMatrix.GainLossStates.GAIN,
772 outfile_name + surfacing.PARSIMONY_OUTPUT_FITCH_GAINS_HTML_BC,
774 ForesterUtil.LINE_SEPARATOR,
775 "Fitch Parsimony | Gains | Domain Combinations",
778 all_pfams_encountered,
779 all_pfams_gained_as_dom_combinations,
781 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
785 domain_parsimony.getGainLossMatrix(),
786 CharacterStateMatrix.GainLossStates.LOSS,
787 outfile_name + surfacing.PARSIMONY_OUTPUT_FITCH_LOSSES_HTML_BC,
789 ForesterUtil.LINE_SEPARATOR,
790 "Fitch Parsimony | Losses | Domain Combinations",
793 all_pfams_encountered,
794 all_pfams_lost_as_dom_combinations,
795 "_fitch_losses_dc" );
796 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
800 domain_parsimony.getGainLossMatrix(),
802 outfile_name + surfacing.PARSIMONY_OUTPUT_FITCH_PRESENT_HTML_BC,
804 ForesterUtil.LINE_SEPARATOR,
805 "Fitch Parsimony | Present | Domain Combinations",
808 all_pfams_encountered,
810 "_fitch_present_dc" );
811 writeAllEncounteredPfamsToFile( domain_id_to_go_ids_map,
814 all_pfams_encountered );
815 writePfamsToFile( outfile_name + surfacing.ALL_PFAMS_GAINED_AS_DOMAINS_SUFFIX, all_pfams_gained_as_domains );
816 writePfamsToFile( outfile_name + surfacing.ALL_PFAMS_LOST_AS_DOMAINS_SUFFIX, all_pfams_lost_as_domains );
817 writePfamsToFile( outfile_name + surfacing.ALL_PFAMS_GAINED_AS_DC_SUFFIX,
818 all_pfams_gained_as_dom_combinations );
819 writePfamsToFile( outfile_name + surfacing.ALL_PFAMS_LOST_AS_DC_SUFFIX, all_pfams_lost_as_dom_combinations );
820 preparePhylogeny( local_phylogeny_l,
823 "Fitch parsimony on binary domain combination presence/absence randomization: "
825 "fitch_on_binary_domain_combinations_" + outfile_name,
827 SurfacingUtil.writePhylogenyToFile( local_phylogeny_l, outfile_name
828 + surfacing.BINARY_DOMAIN_COMBINATIONS_PARSIMONY_TREE_OUTPUT_SUFFIX_FITCH );
829 calculateIndependentDomainCombinationGains( local_phylogeny_l,
831 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_COUNTS_OUTPUT_SUFFIX,
833 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_OUTPUT_SUFFIX,
835 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_FOR_GO_MAPPING_OUTPUT_SUFFIX,
837 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_FOR_GO_MAPPING_OUTPUT_UNIQUE_SUFFIX,
838 outfile_name + "_indep_dc_gains_fitch_lca_ranks.txt",
839 outfile_name + "_indep_dc_gains_fitch_lca_taxonomies.txt",
840 protein_length_stats_by_dc,
841 domain_number_stats_by_dc );
845 public static void executeParsimonyAnalysisForSecondaryFeatures( final String outfile_name,
846 final DomainParsimonyCalculator secondary_features_parsimony,
847 final Phylogeny phylogeny,
848 final String parameters_str,
849 final Map<Species, MappingResults> mapping_results_map ) {
850 final String sep = ForesterUtil.LINE_SEPARATOR + "###################" + ForesterUtil.LINE_SEPARATOR;
851 final String date_time = ForesterUtil.getCurrentDateTime();
852 System.out.println();
853 writeToNexus( outfile_name + surfacing.NEXUS_SECONDARY_FEATURES,
854 secondary_features_parsimony.createMatrixOfSecondaryFeaturePresenceOrAbsence( null ),
856 Phylogeny local_phylogeny_copy = phylogeny.copy();
857 secondary_features_parsimony.executeDolloParsimonyOnSecondaryFeatures( mapping_results_map );
858 SurfacingUtil.writeMatrixToFile( secondary_features_parsimony.getGainLossMatrix(), outfile_name
859 + surfacing.PARSIMONY_OUTPUT_GL_SUFFIX_DOLLO_SECONDARY_FEATURES, Format.FORESTER );
860 SurfacingUtil.writeMatrixToFile( secondary_features_parsimony.getGainLossCountsMatrix(), outfile_name
861 + surfacing.PARSIMONY_OUTPUT_GL_COUNTS_SUFFIX_DOLLO_SECONDARY_FEATURES, Format.FORESTER );
863 .writeBinaryStatesMatrixAsListToFile( secondary_features_parsimony.getGainLossMatrix(),
864 CharacterStateMatrix.GainLossStates.GAIN,
866 + surfacing.PARSIMONY_OUTPUT_DOLLO_GAINS_SECONDARY_FEATURES,
868 ForesterUtil.LINE_SEPARATOR,
871 .writeBinaryStatesMatrixAsListToFile( secondary_features_parsimony.getGainLossMatrix(),
872 CharacterStateMatrix.GainLossStates.LOSS,
874 + surfacing.PARSIMONY_OUTPUT_DOLLO_LOSSES_SECONDARY_FEATURES,
876 ForesterUtil.LINE_SEPARATOR,
879 .writeBinaryStatesMatrixAsListToFile( secondary_features_parsimony.getGainLossMatrix(),
882 + surfacing.PARSIMONY_OUTPUT_DOLLO_PRESENT_SECONDARY_FEATURES,
884 ForesterUtil.LINE_SEPARATOR,
886 preparePhylogeny( local_phylogeny_copy,
887 secondary_features_parsimony,
889 "Dollo parsimony on secondary feature presence/absence",
890 "dollo_on_secondary_features_" + outfile_name,
892 SurfacingUtil.writePhylogenyToFile( local_phylogeny_copy, outfile_name
893 + surfacing.SECONDARY_FEATURES_PARSIMONY_TREE_OUTPUT_SUFFIX_DOLLO );
894 // FITCH DOMAIN COMBINATIONS
895 // -------------------------
896 local_phylogeny_copy = phylogeny.copy();
897 final String randomization = "no";
898 secondary_features_parsimony.executeFitchParsimonyOnBinaryDomainCombintionOnSecondaryFeatures( USE_LAST );
899 preparePhylogeny( local_phylogeny_copy,
900 secondary_features_parsimony,
902 "Fitch parsimony on secondary binary domain combination presence/absence randomization: "
904 "fitch_on_binary_domain_combinations_" + outfile_name,
906 SurfacingUtil.writePhylogenyToFile( local_phylogeny_copy, outfile_name
907 + surfacing.BINARY_DOMAIN_COMBINATIONS_PARSIMONY_TREE_OUTPUT_SUFFIX_FITCH_MAPPED );
908 calculateIndependentDomainCombinationGains( local_phylogeny_copy, outfile_name
909 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_COUNTS_MAPPED_OUTPUT_SUFFIX, outfile_name
910 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_MAPPED_OUTPUT_SUFFIX, outfile_name
911 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_FOR_GO_MAPPING_MAPPED_OUTPUT_SUFFIX, outfile_name
912 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_FOR_GO_MAPPING_MAPPED_OUTPUT_UNIQUE_SUFFIX, outfile_name
913 + "_MAPPED_indep_dc_gains_fitch_lca_ranks.txt", outfile_name
914 + "_MAPPED_indep_dc_gains_fitch_lca_taxonomies.txt", null, null );
917 public static void doit( final List<Protein> proteins,
918 final List<DomainId> query_domain_ids_nc_order,
920 final String separator,
921 final String limit_to_species,
922 final Map<String, List<Integer>> average_protein_lengths_by_dc ) throws IOException {
923 for( final Protein protein : proteins ) {
924 if ( ForesterUtil.isEmpty( limit_to_species )
925 || protein.getSpecies().getSpeciesId().equalsIgnoreCase( limit_to_species ) ) {
926 if ( protein.contains( query_domain_ids_nc_order, true ) ) {
927 out.write( protein.getSpecies().getSpeciesId() );
928 out.write( separator );
929 out.write( protein.getProteinId().getId() );
930 out.write( separator );
932 final Set<DomainId> visited_domain_ids = new HashSet<DomainId>();
933 boolean first = true;
934 for( final Domain domain : protein.getProteinDomains() ) {
935 if ( !visited_domain_ids.contains( domain.getDomainId() ) ) {
936 visited_domain_ids.add( domain.getDomainId() );
943 out.write( domain.getDomainId().getId() );
945 out.write( "" + domain.getTotalCount() );
950 out.write( separator );
951 if ( !( ForesterUtil.isEmpty( protein.getDescription() ) || protein.getDescription()
952 .equals( SurfacingConstants.NONE ) ) ) {
953 out.write( protein.getDescription() );
955 out.write( separator );
956 if ( !( ForesterUtil.isEmpty( protein.getAccession() ) || protein.getAccession()
957 .equals( SurfacingConstants.NONE ) ) ) {
958 out.write( protein.getAccession() );
960 out.write( SurfacingConstants.NL );
967 public static void extractProteinNames( final List<Protein> proteins,
968 final List<DomainId> query_domain_ids_nc_order,
970 final String separator,
971 final String limit_to_species ) throws IOException {
972 for( final Protein protein : proteins ) {
973 if ( ForesterUtil.isEmpty( limit_to_species )
974 || protein.getSpecies().getSpeciesId().equalsIgnoreCase( limit_to_species ) ) {
975 if ( protein.contains( query_domain_ids_nc_order, true ) ) {
976 out.write( protein.getSpecies().getSpeciesId() );
977 out.write( separator );
978 out.write( protein.getProteinId().getId() );
979 out.write( separator );
981 final Set<DomainId> visited_domain_ids = new HashSet<DomainId>();
982 boolean first = true;
983 for( final Domain domain : protein.getProteinDomains() ) {
984 if ( !visited_domain_ids.contains( domain.getDomainId() ) ) {
985 visited_domain_ids.add( domain.getDomainId() );
992 out.write( domain.getDomainId().getId() );
994 out.write( "" + domain.getTotalCount() );
999 out.write( separator );
1000 if ( !( ForesterUtil.isEmpty( protein.getDescription() ) || protein.getDescription()
1001 .equals( SurfacingConstants.NONE ) ) ) {
1002 out.write( protein.getDescription() );
1004 out.write( separator );
1005 if ( !( ForesterUtil.isEmpty( protein.getAccession() ) || protein.getAccession()
1006 .equals( SurfacingConstants.NONE ) ) ) {
1007 out.write( protein.getAccession() );
1009 out.write( SurfacingConstants.NL );
1016 public static void extractProteinNames( final SortedMap<Species, List<Protein>> protein_lists_per_species,
1017 final DomainId domain_id,
1019 final String separator,
1020 final String limit_to_species ) throws IOException {
1021 for( final Species species : protein_lists_per_species.keySet() ) {
1022 for( final Protein protein : protein_lists_per_species.get( species ) ) {
1023 if ( ForesterUtil.isEmpty( limit_to_species )
1024 || protein.getSpecies().getSpeciesId().equalsIgnoreCase( limit_to_species ) ) {
1025 final List<Domain> domains = protein.getProteinDomains( domain_id );
1026 if ( domains.size() > 0 ) {
1027 final DescriptiveStatistics stats = new BasicDescriptiveStatistics();
1028 for( final Domain domain : domains ) {
1029 stats.addValue( domain.getPerSequenceEvalue() );
1031 out.write( protein.getSpecies().getSpeciesId() );
1032 out.write( separator );
1033 out.write( protein.getProteinId().getId() );
1034 out.write( separator );
1035 out.write( "[" + FORMATTER.format( stats.median() ) + "]" );
1036 out.write( separator );
1037 if ( !( ForesterUtil.isEmpty( protein.getDescription() ) || protein.getDescription()
1038 .equals( SurfacingConstants.NONE ) ) ) {
1039 out.write( protein.getDescription() );
1041 out.write( separator );
1042 if ( !( ForesterUtil.isEmpty( protein.getAccession() ) || protein.getAccession()
1043 .equals( SurfacingConstants.NONE ) ) ) {
1044 out.write( protein.getAccession() );
1046 out.write( SurfacingConstants.NL );
1054 public static SortedSet<DomainId> getAllDomainIds( final List<GenomeWideCombinableDomains> gwcd_list ) {
1055 final SortedSet<DomainId> all_domains_ids = new TreeSet<DomainId>();
1056 for( final GenomeWideCombinableDomains gwcd : gwcd_list ) {
1057 final Set<DomainId> all_domains = gwcd.getAllDomainIds();
1058 // for( final Domain domain : all_domains ) {
1059 all_domains_ids.addAll( all_domains );
1062 return all_domains_ids;
1065 public static SortedMap<String, Integer> getDomainCounts( final List<Protein> protein_domain_collections ) {
1066 final SortedMap<String, Integer> map = new TreeMap<String, Integer>();
1067 for( final Protein protein_domain_collection : protein_domain_collections ) {
1068 for( final Object name : protein_domain_collection.getProteinDomains() ) {
1069 final BasicDomain protein_domain = ( BasicDomain ) name;
1070 final String id = protein_domain.getDomainId().getId();
1071 if ( map.containsKey( id ) ) {
1072 map.put( id, map.get( id ) + 1 );
1082 public static int getNumberOfNodesLackingName( final Phylogeny p, final StringBuilder names ) {
1083 final PhylogenyNodeIterator it = p.iteratorPostorder();
1085 while ( it.hasNext() ) {
1086 final PhylogenyNode n = it.next();
1087 if ( ForesterUtil.isEmpty( n.getName() )
1088 && ( !n.getNodeData().isHasTaxonomy() || ForesterUtil.isEmpty( n.getNodeData().getTaxonomy()
1089 .getScientificName() ) )
1090 && ( !n.getNodeData().isHasTaxonomy() || ForesterUtil.isEmpty( n.getNodeData().getTaxonomy()
1091 .getCommonName() ) ) ) {
1092 if ( n.getParent() != null ) {
1093 names.append( " " );
1094 names.append( n.getParent().getName() );
1096 final List l = n.getAllExternalDescendants();
1097 for( final Object object : l ) {
1098 System.out.println( l.toString() );
1107 * Returns true is Domain domain falls in an uninterrupted stretch of
1108 * covered positions.
1111 * @param covered_positions
1114 public static boolean isEngulfed( final Domain domain, final List<Boolean> covered_positions ) {
1115 for( int i = domain.getFrom(); i <= domain.getTo(); ++i ) {
1116 if ( ( i >= covered_positions.size() ) || ( covered_positions.get( i ) != true ) ) {
1123 public static void preparePhylogeny( final Phylogeny p,
1124 final DomainParsimonyCalculator domain_parsimony,
1125 final String date_time,
1126 final String method,
1128 final String parameters_str ) {
1129 domain_parsimony.decoratePhylogenyWithDomains( p );
1130 final StringBuilder desc = new StringBuilder();
1131 desc.append( "[Method: " + method + "] [Date: " + date_time + "] " );
1132 desc.append( "[Cost: " + domain_parsimony.getCost() + "] " );
1133 desc.append( "[Gains: " + domain_parsimony.getTotalGains() + "] " );
1134 desc.append( "[Losses: " + domain_parsimony.getTotalLosses() + "] " );
1135 desc.append( "[Unchanged: " + domain_parsimony.getTotalUnchanged() + "] " );
1136 desc.append( "[Parameters: " + parameters_str + "]" );
1138 p.setDescription( desc.toString() );
1139 p.setConfidence( new Confidence( domain_parsimony.getCost(), "parsimony" ) );
1140 p.setRerootable( false );
1141 p.setRooted( true );
1145 * species | protein id | n-terminal domain | c-terminal domain | n-terminal domain per domain E-value | c-terminal domain per domain E-value
1149 static public StringBuffer proteinToDomainCombinations( final Protein protein,
1150 final String protein_id,
1151 final String separator ) {
1152 final StringBuffer sb = new StringBuffer();
1153 if ( protein.getSpecies() == null ) {
1154 throw new IllegalArgumentException( "species must not be null" );
1156 if ( ForesterUtil.isEmpty( protein.getSpecies().getSpeciesId() ) ) {
1157 throw new IllegalArgumentException( "species id must not be empty" );
1159 final List<Domain> domains = protein.getProteinDomains();
1160 if ( domains.size() > 1 ) {
1161 final Map<String, Integer> counts = new HashMap<String, Integer>();
1162 for( final Domain domain : domains ) {
1163 final String id = domain.getDomainId().getId();
1164 if ( counts.containsKey( id ) ) {
1165 counts.put( id, counts.get( id ) + 1 );
1168 counts.put( id, 1 );
1171 final Set<String> dcs = new HashSet<String>();
1172 for( int i = 1; i < domains.size(); ++i ) {
1173 for( int j = 0; j < i; ++j ) {
1174 Domain domain_n = domains.get( i );
1175 Domain domain_c = domains.get( j );
1176 if ( domain_n.getFrom() > domain_c.getFrom() ) {
1177 domain_n = domains.get( j );
1178 domain_c = domains.get( i );
1180 final String dc = domain_n.getDomainId().getId() + domain_c.getDomainId().getId();
1181 if ( !dcs.contains( dc ) ) {
1183 sb.append( protein.getSpecies() );
1184 sb.append( separator );
1185 sb.append( protein_id );
1186 sb.append( separator );
1187 sb.append( domain_n.getDomainId().getId() );
1188 sb.append( separator );
1189 sb.append( domain_c.getDomainId().getId() );
1190 sb.append( separator );
1191 sb.append( domain_n.getPerDomainEvalue() );
1192 sb.append( separator );
1193 sb.append( domain_c.getPerDomainEvalue() );
1194 sb.append( separator );
1195 sb.append( counts.get( domain_n.getDomainId().getId() ) );
1196 sb.append( separator );
1197 sb.append( counts.get( domain_c.getDomainId().getId() ) );
1198 sb.append( ForesterUtil.LINE_SEPARATOR );
1203 else if ( domains.size() == 1 ) {
1204 sb.append( protein.getSpecies() );
1205 sb.append( separator );
1206 sb.append( protein_id );
1207 sb.append( separator );
1208 sb.append( domains.get( 0 ).getDomainId().getId() );
1209 sb.append( separator );
1210 sb.append( separator );
1211 sb.append( domains.get( 0 ).getPerDomainEvalue() );
1212 sb.append( separator );
1213 sb.append( separator );
1215 sb.append( separator );
1216 sb.append( ForesterUtil.LINE_SEPARATOR );
1219 sb.append( protein.getSpecies() );
1220 sb.append( separator );
1221 sb.append( protein_id );
1222 sb.append( separator );
1223 sb.append( separator );
1224 sb.append( separator );
1225 sb.append( separator );
1226 sb.append( separator );
1227 sb.append( separator );
1228 sb.append( ForesterUtil.LINE_SEPARATOR );
1235 * Example regarding engulfment: ------------0.1 ----------0.2 --0.3 =>
1236 * domain with 0.3 is ignored
1238 * -----------0.1 ----------0.2 --0.3 => domain with 0.3 is ignored
1241 * ------------0.1 ----------0.3 --0.2 => domains with 0.3 and 0.2 are _not_
1244 * @param max_allowed_overlap
1245 * maximal allowed overlap (inclusive) to be still considered not
1246 * overlapping (zero or negative value to allow any overlap)
1247 * @param remove_engulfed_domains
1248 * to remove domains which are completely engulfed by coverage of
1249 * domains with better support
1253 public static Protein removeOverlappingDomains( final int max_allowed_overlap,
1254 final boolean remove_engulfed_domains,
1255 final Protein protein ) {
1256 final Protein pruned_protein = new BasicProtein( protein.getProteinId().getId(), protein.getSpecies()
1257 .getSpeciesId(), protein.getLength() );
1258 final List<Domain> sorted = SurfacingUtil.sortDomainsWithAscendingConfidenceValues( protein );
1259 final List<Boolean> covered_positions = new ArrayList<Boolean>();
1260 for( final Domain domain : sorted ) {
1261 if ( ( ( max_allowed_overlap < 0 ) || ( SurfacingUtil.calculateOverlap( domain, covered_positions ) <= max_allowed_overlap ) )
1262 && ( !remove_engulfed_domains || !isEngulfed( domain, covered_positions ) ) ) {
1263 final int covered_positions_size = covered_positions.size();
1264 for( int i = covered_positions_size; i < domain.getFrom(); ++i ) {
1265 covered_positions.add( false );
1267 final int new_covered_positions_size = covered_positions.size();
1268 for( int i = domain.getFrom(); i <= domain.getTo(); ++i ) {
1269 if ( i < new_covered_positions_size ) {
1270 covered_positions.set( i, true );
1273 covered_positions.add( true );
1276 pruned_protein.addProteinDomain( domain );
1279 return pruned_protein;
1282 public static List<Domain> sortDomainsWithAscendingConfidenceValues( final Protein protein ) {
1283 final List<Domain> domains = new ArrayList<Domain>();
1284 for( final Domain d : protein.getProteinDomains() ) {
1287 Collections.sort( domains, SurfacingUtil.ASCENDING_CONFIDENCE_VALUE_ORDER );
1291 private static List<String> splitDomainCombination( final String dc ) {
1292 final String[] s = dc.split( "=" );
1293 if ( s.length != 2 ) {
1294 ForesterUtil.printErrorMessage( surfacing.PRG_NAME, "Stringyfied domain combination has illegal format: "
1298 final List<String> l = new ArrayList<String>( 2 );
1304 public static void writeAllDomainsChangedOnAllSubtrees( final Phylogeny p,
1305 final boolean get_gains,
1306 final String outdir,
1307 final String suffix_for_filename ) throws IOException {
1308 CharacterStateMatrix.GainLossStates state = CharacterStateMatrix.GainLossStates.GAIN;
1310 state = CharacterStateMatrix.GainLossStates.LOSS;
1312 final File base_dir = createBaseDirForPerNodeDomainFiles( surfacing.BASE_DIRECTORY_PER_SUBTREE_DOMAIN_GAIN_LOSS_FILES,
1316 for( final PhylogenyNodeIterator it = p.iteratorPostorder(); it.hasNext(); ) {
1317 final PhylogenyNode node = it.next();
1318 if ( !node.isExternal() ) {
1319 final SortedSet<String> domains = collectAllDomainsChangedOnSubtree( node, get_gains );
1320 if ( domains.size() > 0 ) {
1321 final Writer writer = ForesterUtil.createBufferedWriter( base_dir + ForesterUtil.FILE_SEPARATOR
1322 + node.getName() + suffix_for_filename );
1323 for( final String domain : domains ) {
1324 writer.write( domain );
1325 writer.write( ForesterUtil.LINE_SEPARATOR );
1333 private static void writeAllEncounteredPfamsToFile( final Map<DomainId, List<GoId>> domain_id_to_go_ids_map,
1334 final Map<GoId, GoTerm> go_id_to_term_map,
1335 final String outfile_name,
1336 final SortedSet<String> all_pfams_encountered ) {
1337 final File all_pfams_encountered_file = new File( outfile_name + surfacing.ALL_PFAMS_ENCOUNTERED_SUFFIX );
1338 final File all_pfams_encountered_with_go_annotation_file = new File( outfile_name
1339 + surfacing.ALL_PFAMS_ENCOUNTERED_WITH_GO_ANNOTATION_SUFFIX );
1340 final File encountered_pfams_summary_file = new File( outfile_name + surfacing.ENCOUNTERED_PFAMS_SUMMARY_SUFFIX );
1341 int biological_process_counter = 0;
1342 int cellular_component_counter = 0;
1343 int molecular_function_counter = 0;
1344 int pfams_with_mappings_counter = 0;
1345 int pfams_without_mappings_counter = 0;
1346 int pfams_without_mappings_to_bp_or_mf_counter = 0;
1347 int pfams_with_mappings_to_bp_or_mf_counter = 0;
1349 final Writer all_pfams_encountered_writer = new BufferedWriter( new FileWriter( all_pfams_encountered_file ) );
1350 final Writer all_pfams_encountered_with_go_annotation_writer = new BufferedWriter( new FileWriter( all_pfams_encountered_with_go_annotation_file ) );
1351 final Writer summary_writer = new BufferedWriter( new FileWriter( encountered_pfams_summary_file ) );
1352 summary_writer.write( "# Pfam to GO mapping summary" );
1353 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1354 summary_writer.write( "# Actual summary is at the end of this file." );
1355 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1356 summary_writer.write( "# Encountered Pfams without a GO mapping:" );
1357 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1358 for( final String pfam : all_pfams_encountered ) {
1359 all_pfams_encountered_writer.write( pfam );
1360 all_pfams_encountered_writer.write( ForesterUtil.LINE_SEPARATOR );
1361 final DomainId domain_id = new DomainId( pfam );
1362 if ( domain_id_to_go_ids_map.containsKey( domain_id ) ) {
1363 ++pfams_with_mappings_counter;
1364 all_pfams_encountered_with_go_annotation_writer.write( pfam );
1365 all_pfams_encountered_with_go_annotation_writer.write( ForesterUtil.LINE_SEPARATOR );
1366 final List<GoId> go_ids = domain_id_to_go_ids_map.get( domain_id );
1367 boolean maps_to_bp = false;
1368 boolean maps_to_cc = false;
1369 boolean maps_to_mf = false;
1370 for( final GoId go_id : go_ids ) {
1371 final GoTerm go_term = go_id_to_term_map.get( go_id );
1372 if ( go_term.getGoNameSpace().isBiologicalProcess() ) {
1375 else if ( go_term.getGoNameSpace().isCellularComponent() ) {
1378 else if ( go_term.getGoNameSpace().isMolecularFunction() ) {
1383 ++biological_process_counter;
1386 ++cellular_component_counter;
1389 ++molecular_function_counter;
1391 if ( maps_to_bp || maps_to_mf ) {
1392 ++pfams_with_mappings_to_bp_or_mf_counter;
1395 ++pfams_without_mappings_to_bp_or_mf_counter;
1399 ++pfams_without_mappings_to_bp_or_mf_counter;
1400 ++pfams_without_mappings_counter;
1401 summary_writer.write( pfam );
1402 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1405 all_pfams_encountered_writer.close();
1406 all_pfams_encountered_with_go_annotation_writer.close();
1407 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote all [" + all_pfams_encountered.size()
1408 + "] encountered Pfams to: \"" + all_pfams_encountered_file + "\"" );
1409 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote all [" + pfams_with_mappings_counter
1410 + "] encountered Pfams with GO mappings to: \"" + all_pfams_encountered_with_go_annotation_file
1412 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote summary (including all ["
1413 + pfams_without_mappings_counter + "] encountered Pfams without GO mappings) to: \""
1414 + encountered_pfams_summary_file + "\"" );
1415 ForesterUtil.programMessage( surfacing.PRG_NAME, "Sum of Pfams encountered : "
1416 + all_pfams_encountered.size() );
1417 ForesterUtil.programMessage( surfacing.PRG_NAME, "Pfams without a mapping : "
1418 + pfams_without_mappings_counter + " ["
1419 + ( 100 * pfams_without_mappings_counter / all_pfams_encountered.size() ) + "%]" );
1420 ForesterUtil.programMessage( surfacing.PRG_NAME, "Pfams without mapping to proc. or func. : "
1421 + pfams_without_mappings_to_bp_or_mf_counter + " ["
1422 + ( 100 * pfams_without_mappings_to_bp_or_mf_counter / all_pfams_encountered.size() ) + "%]" );
1423 ForesterUtil.programMessage( surfacing.PRG_NAME,
1424 "Pfams with a mapping : " + pfams_with_mappings_counter
1426 + ( 100 * pfams_with_mappings_counter / all_pfams_encountered.size() )
1428 ForesterUtil.programMessage( surfacing.PRG_NAME, "Pfams with a mapping to proc. or func. : "
1429 + pfams_with_mappings_to_bp_or_mf_counter + " ["
1430 + ( 100 * pfams_with_mappings_to_bp_or_mf_counter / all_pfams_encountered.size() ) + "%]" );
1431 ForesterUtil.programMessage( surfacing.PRG_NAME,
1432 "Pfams with mapping to biological process: " + biological_process_counter
1434 + ( 100 * biological_process_counter / all_pfams_encountered.size() )
1436 ForesterUtil.programMessage( surfacing.PRG_NAME,
1437 "Pfams with mapping to molecular function: " + molecular_function_counter
1439 + ( 100 * molecular_function_counter / all_pfams_encountered.size() )
1441 ForesterUtil.programMessage( surfacing.PRG_NAME,
1442 "Pfams with mapping to cellular component: " + cellular_component_counter
1444 + ( 100 * cellular_component_counter / all_pfams_encountered.size() )
1446 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1447 summary_writer.write( "# Sum of Pfams encountered : " + all_pfams_encountered.size() );
1448 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1449 summary_writer.write( "# Pfams without a mapping : " + pfams_without_mappings_counter
1450 + " [" + ( 100 * pfams_without_mappings_counter / all_pfams_encountered.size() ) + "%]" );
1451 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1452 summary_writer.write( "# Pfams without mapping to proc. or func. : "
1453 + pfams_without_mappings_to_bp_or_mf_counter + " ["
1454 + ( 100 * pfams_without_mappings_to_bp_or_mf_counter / all_pfams_encountered.size() ) + "%]" );
1455 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1456 summary_writer.write( "# Pfams with a mapping : " + pfams_with_mappings_counter + " ["
1457 + ( 100 * pfams_with_mappings_counter / all_pfams_encountered.size() ) + "%]" );
1458 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1459 summary_writer.write( "# Pfams with a mapping to proc. or func. : "
1460 + pfams_with_mappings_to_bp_or_mf_counter + " ["
1461 + ( 100 * pfams_with_mappings_to_bp_or_mf_counter / all_pfams_encountered.size() ) + "%]" );
1462 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1463 summary_writer.write( "# Pfams with mapping to biological process: " + biological_process_counter + " ["
1464 + ( 100 * biological_process_counter / all_pfams_encountered.size() ) + "%]" );
1465 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1466 summary_writer.write( "# Pfams with mapping to molecular function: " + molecular_function_counter + " ["
1467 + ( 100 * molecular_function_counter / all_pfams_encountered.size() ) + "%]" );
1468 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1469 summary_writer.write( "# Pfams with mapping to cellular component: " + cellular_component_counter + " ["
1470 + ( 100 * cellular_component_counter / all_pfams_encountered.size() ) + "%]" );
1471 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1472 summary_writer.close();
1474 catch ( final IOException e ) {
1475 ForesterUtil.printWarningMessage( surfacing.PRG_NAME, "Failure to write: " + e );
1479 public static void writeBinaryDomainCombinationsFileForGraphAnalysis( final String[][] input_file_properties,
1480 final File output_dir,
1481 final GenomeWideCombinableDomains gwcd,
1483 final GenomeWideCombinableDomainsSortOrder dc_sort_order ) {
1484 File dc_outfile_dot = new File( input_file_properties[ i ][ 0 ]
1485 + surfacing.DOMAIN_COMBINITONS_OUTPUTFILE_SUFFIX_FOR_GRAPH_ANALYSIS );
1486 if ( output_dir != null ) {
1487 dc_outfile_dot = new File( output_dir + ForesterUtil.FILE_SEPARATOR + dc_outfile_dot );
1489 checkForOutputFileWriteability( dc_outfile_dot );
1490 final SortedSet<BinaryDomainCombination> binary_combinations = createSetOfAllBinaryDomainCombinationsPerGenome( gwcd );
1492 final BufferedWriter out_dot = new BufferedWriter( new FileWriter( dc_outfile_dot ) );
1493 for( final BinaryDomainCombination bdc : binary_combinations ) {
1494 out_dot.write( bdc.toGraphDescribingLanguage( BinaryDomainCombination.OutputFormat.DOT, null, null )
1496 out_dot.write( SurfacingConstants.NL );
1500 catch ( final IOException e ) {
1501 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1503 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote binary domain combination for \""
1504 + input_file_properties[ i ][ 0 ] + "\" (" + input_file_properties[ i ][ 1 ] + ", "
1505 + input_file_properties[ i ][ 2 ] + ") to: \"" + dc_outfile_dot + "\"" );
1508 public static void writeBinaryStatesMatrixAsListToFile( final CharacterStateMatrix<CharacterStateMatrix.GainLossStates> matrix,
1509 final CharacterStateMatrix.GainLossStates state,
1510 final String filename,
1511 final String indentifier_characters_separator,
1512 final String character_separator,
1513 final Map<String, String> descriptions ) {
1514 final File outfile = new File( filename );
1515 checkForOutputFileWriteability( outfile );
1516 final SortedSet<String> sorted_ids = new TreeSet<String>();
1517 for( int i = 0; i < matrix.getNumberOfIdentifiers(); ++i ) {
1518 sorted_ids.add( matrix.getIdentifier( i ) );
1521 final BufferedWriter out = new BufferedWriter( new FileWriter( outfile ) );
1522 for( final String id : sorted_ids ) {
1523 out.write( indentifier_characters_separator );
1524 out.write( "#" + id );
1525 out.write( indentifier_characters_separator );
1526 for( int c = 0; c < matrix.getNumberOfCharacters(); ++c ) {
1528 // using null to indicate either UNCHANGED_PRESENT or GAIN.
1529 if ( ( matrix.getState( id, c ) == state )
1530 || ( ( state == null ) && ( ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.GAIN ) || ( matrix
1531 .getState( id, c ) == CharacterStateMatrix.GainLossStates.UNCHANGED_PRESENT ) ) ) ) {
1532 out.write( matrix.getCharacter( c ) );
1533 if ( ( descriptions != null ) && !descriptions.isEmpty()
1534 && descriptions.containsKey( matrix.getCharacter( c ) ) ) {
1536 out.write( descriptions.get( matrix.getCharacter( c ) ) );
1538 out.write( character_separator );
1545 catch ( final IOException e ) {
1546 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1548 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote characters list: \"" + filename + "\"" );
1551 public static void writeBinaryStatesMatrixAsListToFileForBinaryCombinationsForGraphAnalysis( final CharacterStateMatrix<CharacterStateMatrix.GainLossStates> matrix,
1552 final CharacterStateMatrix.GainLossStates state,
1553 final String filename,
1554 final String indentifier_characters_separator,
1555 final String character_separator,
1556 final BinaryDomainCombination.OutputFormat bc_output_format ) {
1557 final File outfile = new File( filename );
1558 checkForOutputFileWriteability( outfile );
1559 final SortedSet<String> sorted_ids = new TreeSet<String>();
1560 for( int i = 0; i < matrix.getNumberOfIdentifiers(); ++i ) {
1561 sorted_ids.add( matrix.getIdentifier( i ) );
1564 final BufferedWriter out = new BufferedWriter( new FileWriter( outfile ) );
1565 for( final String id : sorted_ids ) {
1566 out.write( indentifier_characters_separator );
1567 out.write( "#" + id );
1568 out.write( indentifier_characters_separator );
1569 for( int c = 0; c < matrix.getNumberOfCharacters(); ++c ) {
1571 // using null to indicate either UNCHANGED_PRESENT or GAIN.
1572 if ( ( matrix.getState( id, c ) == state )
1573 || ( ( state == null ) && ( ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.GAIN ) || ( matrix
1574 .getState( id, c ) == CharacterStateMatrix.GainLossStates.UNCHANGED_PRESENT ) ) ) ) {
1575 BinaryDomainCombination bdc = null;
1577 bdc = BasicBinaryDomainCombination.createInstance( matrix.getCharacter( c ) );
1579 catch ( final Exception e ) {
1580 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getLocalizedMessage() );
1582 out.write( bdc.toGraphDescribingLanguage( bc_output_format, null, null ).toString() );
1583 out.write( character_separator );
1590 catch ( final IOException e ) {
1591 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1593 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote characters list: \"" + filename + "\"" );
1596 public static void writeBinaryStatesMatrixToList( final Map<DomainId, List<GoId>> domain_id_to_go_ids_map,
1597 final Map<GoId, GoTerm> go_id_to_term_map,
1598 final GoNameSpace go_namespace_limit,
1599 final boolean domain_combinations,
1600 final CharacterStateMatrix<CharacterStateMatrix.GainLossStates> matrix,
1601 final CharacterStateMatrix.GainLossStates state,
1602 final String filename,
1603 final String indentifier_characters_separator,
1604 final String character_separator,
1605 final String title_for_html,
1606 final String prefix_for_html,
1607 final Map<DomainId, Set<String>>[] domain_id_to_secondary_features_maps,
1608 final SortedSet<String> all_pfams_encountered,
1609 final SortedSet<String> pfams_gained_or_lost,
1610 final String suffix_for_per_node_events_file ) {
1611 if ( ( go_namespace_limit != null ) && ( ( go_id_to_term_map == null ) || ( go_id_to_term_map.size() < 1 ) ) ) {
1612 throw new IllegalArgumentException( "attempt to use GO namespace limit without a GO-id to term map" );
1614 else if ( ( ( domain_id_to_go_ids_map == null ) || ( domain_id_to_go_ids_map.size() < 1 ) ) ) {
1615 throw new IllegalArgumentException( "attempt to output detailed HTML without a Pfam to GO map" );
1617 else if ( ( ( go_id_to_term_map == null ) || ( go_id_to_term_map.size() < 1 ) ) ) {
1618 throw new IllegalArgumentException( "attempt to output detailed HTML without a GO-id to term map" );
1620 final File outfile = new File( filename );
1621 checkForOutputFileWriteability( outfile );
1622 final SortedSet<String> sorted_ids = new TreeSet<String>();
1623 for( int i = 0; i < matrix.getNumberOfIdentifiers(); ++i ) {
1624 sorted_ids.add( matrix.getIdentifier( i ) );
1627 final Writer out = new BufferedWriter( new FileWriter( outfile ) );
1628 final File per_node_go_mapped_domain_gain_loss_files_base_dir = createBaseDirForPerNodeDomainFiles( surfacing.BASE_DIRECTORY_PER_NODE_DOMAIN_GAIN_LOSS_FILES,
1629 domain_combinations,
1632 Writer per_node_go_mapped_domain_gain_loss_outfile_writer = null;
1633 File per_node_go_mapped_domain_gain_loss_outfile = null;
1634 int per_node_counter = 0;
1635 out.write( "<html>" );
1636 out.write( SurfacingConstants.NL );
1637 addHtmlHead( out, title_for_html );
1638 out.write( SurfacingConstants.NL );
1639 out.write( "<body>" );
1640 out.write( SurfacingConstants.NL );
1641 out.write( "<h1>" );
1642 out.write( SurfacingConstants.NL );
1643 out.write( title_for_html );
1644 out.write( SurfacingConstants.NL );
1645 out.write( "</h1>" );
1646 out.write( SurfacingConstants.NL );
1647 out.write( "<table>" );
1648 out.write( SurfacingConstants.NL );
1649 for( final String id : sorted_ids ) {
1650 final Matcher matcher = PATTERN_SP_STYLE_TAXONOMY.matcher( id );
1651 if ( matcher.matches() ) {
1654 out.write( "<tr>" );
1655 out.write( "<td>" );
1656 out.write( "<a href=\"#" + id + "\">" + id + "</a>" );
1657 out.write( "</td>" );
1658 out.write( "</tr>" );
1659 out.write( SurfacingConstants.NL );
1661 out.write( "</table>" );
1662 out.write( SurfacingConstants.NL );
1663 for( final String id : sorted_ids ) {
1664 final Matcher matcher = PATTERN_SP_STYLE_TAXONOMY.matcher( id );
1665 if ( matcher.matches() ) {
1668 out.write( SurfacingConstants.NL );
1669 out.write( "<h2>" );
1670 out.write( "<a name=\"" + id + "\">" + id + "</a>" );
1671 writeTaxonomyLinks( out, id );
1672 out.write( "</h2>" );
1673 out.write( SurfacingConstants.NL );
1674 out.write( "<table>" );
1675 out.write( SurfacingConstants.NL );
1676 out.write( "<tr>" );
1677 out.write( "<td><b>" );
1678 out.write( "Pfam domain(s)" );
1679 out.write( "</b></td><td><b>" );
1680 out.write( "GO term acc" );
1681 out.write( "</b></td><td><b>" );
1682 out.write( "GO term" );
1683 out.write( "</b></td><td><b>" );
1684 out.write( "GO namespace" );
1685 out.write( "</b></td>" );
1686 out.write( "</tr>" );
1687 out.write( SurfacingConstants.NL );
1688 out.write( "</tr>" );
1689 out.write( SurfacingConstants.NL );
1690 per_node_counter = 0;
1691 if ( matrix.getNumberOfCharacters() > 0 ) {
1692 per_node_go_mapped_domain_gain_loss_outfile = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
1693 + ForesterUtil.FILE_SEPARATOR + id + suffix_for_per_node_events_file );
1694 SurfacingUtil.checkForOutputFileWriteability( per_node_go_mapped_domain_gain_loss_outfile );
1695 per_node_go_mapped_domain_gain_loss_outfile_writer = ForesterUtil
1696 .createBufferedWriter( per_node_go_mapped_domain_gain_loss_outfile );
1699 per_node_go_mapped_domain_gain_loss_outfile = null;
1700 per_node_go_mapped_domain_gain_loss_outfile_writer = null;
1702 for( int c = 0; c < matrix.getNumberOfCharacters(); ++c ) {
1704 // using null to indicate either UNCHANGED_PRESENT or GAIN.
1705 if ( ( matrix.getState( id, c ) == state )
1706 || ( ( state == null ) && ( ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.UNCHANGED_PRESENT ) || ( matrix
1707 .getState( id, c ) == CharacterStateMatrix.GainLossStates.GAIN ) ) ) ) {
1708 final String character = matrix.getCharacter( c );
1709 String domain_0 = "";
1710 String domain_1 = "";
1711 if ( character.indexOf( BinaryDomainCombination.SEPARATOR ) > 0 ) {
1712 final String[] s = character.split( BinaryDomainCombination.SEPARATOR );
1713 if ( s.length != 2 ) {
1714 throw new AssertionError( "this should not have happened: unexpected format for domain combination: ["
1715 + character + "]" );
1721 domain_0 = character;
1723 writeDomainData( domain_id_to_go_ids_map,
1730 character_separator,
1731 domain_id_to_secondary_features_maps,
1733 all_pfams_encountered.add( domain_0 );
1734 if ( pfams_gained_or_lost != null ) {
1735 pfams_gained_or_lost.add( domain_0 );
1737 if ( !ForesterUtil.isEmpty( domain_1 ) ) {
1738 all_pfams_encountered.add( domain_1 );
1739 if ( pfams_gained_or_lost != null ) {
1740 pfams_gained_or_lost.add( domain_1 );
1743 if ( per_node_go_mapped_domain_gain_loss_outfile_writer != null ) {
1744 writeDomainsToIndividualFilePerTreeNode( per_node_go_mapped_domain_gain_loss_outfile_writer,
1751 if ( per_node_go_mapped_domain_gain_loss_outfile_writer != null ) {
1752 per_node_go_mapped_domain_gain_loss_outfile_writer.close();
1753 if ( per_node_counter < 1 ) {
1754 per_node_go_mapped_domain_gain_loss_outfile.delete();
1756 per_node_counter = 0;
1758 out.write( "</table>" );
1759 out.write( SurfacingConstants.NL );
1760 out.write( "<hr>" );
1761 out.write( SurfacingConstants.NL );
1762 } // for( final String id : sorted_ids ) {
1763 out.write( "</body>" );
1764 out.write( SurfacingConstants.NL );
1765 out.write( "</html>" );
1766 out.write( SurfacingConstants.NL );
1770 catch ( final IOException e ) {
1771 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1773 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote characters detailed HTML list: \"" + filename + "\"" );
1776 public static void writeBinaryStatesMatrixToListORIGIG( final Map<DomainId, List<GoId>> domain_id_to_go_ids_map,
1777 final Map<GoId, GoTerm> go_id_to_term_map,
1778 final GoNameSpace go_namespace_limit,
1779 final boolean domain_combinations,
1780 final CharacterStateMatrix<CharacterStateMatrix.GainLossStates> matrix,
1781 final CharacterStateMatrix.GainLossStates state,
1782 final String filename,
1783 final String indentifier_characters_separator,
1784 final String character_separator,
1785 final String title_for_html,
1786 final String prefix_for_html,
1787 final Map<DomainId, Set<String>>[] domain_id_to_secondary_features_maps,
1788 final SortedSet<String> all_pfams_encountered,
1789 final SortedSet<String> pfams_gained_or_lost,
1790 final String suffix_for_per_node_events_file ) {
1791 if ( ( go_namespace_limit != null ) && ( ( go_id_to_term_map == null ) || ( go_id_to_term_map.size() < 1 ) ) ) {
1792 throw new IllegalArgumentException( "attempt to use GO namespace limit without a GO-id to term map" );
1794 else if ( ( ( domain_id_to_go_ids_map == null ) || ( domain_id_to_go_ids_map.size() < 1 ) ) ) {
1795 throw new IllegalArgumentException( "attempt to output detailed HTML without a Pfam to GO map" );
1797 else if ( ( ( go_id_to_term_map == null ) || ( go_id_to_term_map.size() < 1 ) ) ) {
1798 throw new IllegalArgumentException( "attempt to output detailed HTML without a GO-id to term map" );
1800 final File outfile = new File( filename );
1801 checkForOutputFileWriteability( outfile );
1802 final SortedSet<String> sorted_ids = new TreeSet<String>();
1803 for( int i = 0; i < matrix.getNumberOfIdentifiers(); ++i ) {
1804 sorted_ids.add( matrix.getIdentifier( i ) );
1807 final Writer out = new BufferedWriter( new FileWriter( outfile ) );
1808 final File per_node_go_mapped_domain_gain_loss_files_base_dir = createBaseDirForPerNodeDomainFiles( surfacing.BASE_DIRECTORY_PER_NODE_DOMAIN_GAIN_LOSS_FILES,
1809 domain_combinations,
1812 Writer per_node_go_mapped_domain_gain_loss_outfile_writer = null;
1813 File per_node_go_mapped_domain_gain_loss_outfile = null;
1814 int per_node_counter = 0;
1815 out.write( "<html>" );
1816 out.write( SurfacingConstants.NL );
1817 addHtmlHead( out, title_for_html );
1818 out.write( SurfacingConstants.NL );
1819 out.write( "<body>" );
1820 out.write( SurfacingConstants.NL );
1821 out.write( "<h1>" );
1822 out.write( SurfacingConstants.NL );
1823 out.write( title_for_html );
1824 out.write( SurfacingConstants.NL );
1825 out.write( "</h1>" );
1826 out.write( SurfacingConstants.NL );
1827 out.write( "<table>" );
1828 out.write( SurfacingConstants.NL );
1829 for( final String id : sorted_ids ) {
1830 out.write( "<tr>" );
1831 out.write( "<td>" );
1832 out.write( "<a href=\"#" + id + "\">" + id + "</a>" );
1833 writeTaxonomyLinks( out, id );
1834 out.write( "</td>" );
1835 out.write( "</tr>" );
1836 out.write( SurfacingConstants.NL );
1838 out.write( "</table>" );
1839 out.write( SurfacingConstants.NL );
1840 for( final String id : sorted_ids ) {
1841 out.write( SurfacingConstants.NL );
1842 out.write( "<h2>" );
1843 out.write( "<a name=\"" + id + "\">" + id + "</a>" );
1844 writeTaxonomyLinks( out, id );
1845 out.write( "</h2>" );
1846 out.write( SurfacingConstants.NL );
1847 out.write( "<table>" );
1848 out.write( SurfacingConstants.NL );
1849 out.write( "<tr>" );
1850 out.write( "<td><b>" );
1851 out.write( "Pfam domain(s)" );
1852 out.write( "</b></td><td><b>" );
1853 out.write( "GO term acc" );
1854 out.write( "</b></td><td><b>" );
1855 out.write( "GO term" );
1856 out.write( "</b></td><td><b>" );
1857 out.write( "Penultimate GO term" );
1858 out.write( "</b></td><td><b>" );
1859 out.write( "GO namespace" );
1860 out.write( "</b></td>" );
1861 out.write( "</tr>" );
1862 out.write( SurfacingConstants.NL );
1863 out.write( "</tr>" );
1864 out.write( SurfacingConstants.NL );
1865 per_node_counter = 0;
1866 if ( matrix.getNumberOfCharacters() > 0 ) {
1867 per_node_go_mapped_domain_gain_loss_outfile = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
1868 + ForesterUtil.FILE_SEPARATOR + id + suffix_for_per_node_events_file );
1869 SurfacingUtil.checkForOutputFileWriteability( per_node_go_mapped_domain_gain_loss_outfile );
1870 per_node_go_mapped_domain_gain_loss_outfile_writer = ForesterUtil
1871 .createBufferedWriter( per_node_go_mapped_domain_gain_loss_outfile );
1874 per_node_go_mapped_domain_gain_loss_outfile = null;
1875 per_node_go_mapped_domain_gain_loss_outfile_writer = null;
1877 for( int c = 0; c < matrix.getNumberOfCharacters(); ++c ) {
1879 // using null to indicate either UNCHANGED_PRESENT or GAIN.
1880 if ( ( matrix.getState( id, c ) == state )
1881 || ( ( state == null ) && ( ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.UNCHANGED_PRESENT ) || ( matrix
1882 .getState( id, c ) == CharacterStateMatrix.GainLossStates.GAIN ) ) ) ) {
1883 final String character = matrix.getCharacter( c );
1884 String domain_0 = "";
1885 String domain_1 = "";
1886 if ( character.indexOf( BinaryDomainCombination.SEPARATOR ) > 0 ) {
1887 final String[] s = character.split( BinaryDomainCombination.SEPARATOR );
1888 if ( s.length != 2 ) {
1889 throw new AssertionError( "this should not have happened: unexpected format for domain combination: ["
1890 + character + "]" );
1896 domain_0 = character;
1898 writeDomainData( domain_id_to_go_ids_map,
1905 character_separator,
1906 domain_id_to_secondary_features_maps,
1908 all_pfams_encountered.add( domain_0 );
1909 if ( pfams_gained_or_lost != null ) {
1910 pfams_gained_or_lost.add( domain_0 );
1912 if ( !ForesterUtil.isEmpty( domain_1 ) ) {
1913 all_pfams_encountered.add( domain_1 );
1914 if ( pfams_gained_or_lost != null ) {
1915 pfams_gained_or_lost.add( domain_1 );
1918 if ( per_node_go_mapped_domain_gain_loss_outfile_writer != null ) {
1919 writeDomainsToIndividualFilePerTreeNode( per_node_go_mapped_domain_gain_loss_outfile_writer,
1926 if ( per_node_go_mapped_domain_gain_loss_outfile_writer != null ) {
1927 per_node_go_mapped_domain_gain_loss_outfile_writer.close();
1928 if ( per_node_counter < 1 ) {
1929 per_node_go_mapped_domain_gain_loss_outfile.delete();
1931 per_node_counter = 0;
1933 out.write( "</table>" );
1934 out.write( SurfacingConstants.NL );
1935 out.write( "<hr>" );
1936 out.write( SurfacingConstants.NL );
1937 } // for( final String id : sorted_ids ) {
1938 out.write( "</body>" );
1939 out.write( SurfacingConstants.NL );
1940 out.write( "</html>" );
1941 out.write( SurfacingConstants.NL );
1945 catch ( final IOException e ) {
1946 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1948 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote characters detailed HTML list: \"" + filename + "\"" );
1951 public static void writeDomainCombinationsCountsFile( final String[][] input_file_properties,
1952 final File output_dir,
1953 final Writer per_genome_domain_promiscuity_statistics_writer,
1954 final GenomeWideCombinableDomains gwcd,
1956 final GenomeWideCombinableDomains.GenomeWideCombinableDomainsSortOrder dc_sort_order ) {
1957 File dc_outfile = new File( input_file_properties[ i ][ 0 ]
1958 + surfacing.DOMAIN_COMBINITON_COUNTS_OUTPUTFILE_SUFFIX );
1959 if ( output_dir != null ) {
1960 dc_outfile = new File( output_dir + ForesterUtil.FILE_SEPARATOR + dc_outfile );
1962 checkForOutputFileWriteability( dc_outfile );
1964 final BufferedWriter out = new BufferedWriter( new FileWriter( dc_outfile ) );
1965 out.write( gwcd.toStringBuilder( dc_sort_order ).toString() );
1968 catch ( final IOException e ) {
1969 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1971 final DescriptiveStatistics stats = gwcd.getPerGenomeDomainPromiscuityStatistics();
1973 per_genome_domain_promiscuity_statistics_writer.write( input_file_properties[ i ][ 0 ] + "\t" );
1974 per_genome_domain_promiscuity_statistics_writer.write( FORMATTER_3.format( stats.arithmeticMean() ) + "\t" );
1975 if ( stats.getN() < 2 ) {
1976 per_genome_domain_promiscuity_statistics_writer.write( "n/a" + "\t" );
1979 per_genome_domain_promiscuity_statistics_writer.write( FORMATTER_3.format( stats
1980 .sampleStandardDeviation() ) + "\t" );
1982 per_genome_domain_promiscuity_statistics_writer.write( FORMATTER_3.format( stats.median() ) + "\t" );
1983 per_genome_domain_promiscuity_statistics_writer.write( ( int ) stats.getMin() + "\t" );
1984 per_genome_domain_promiscuity_statistics_writer.write( ( int ) stats.getMax() + "\t" );
1985 per_genome_domain_promiscuity_statistics_writer.write( stats.getN() + "\t" );
1986 final SortedSet<DomainId> mpds = gwcd.getMostPromiscuosDomain();
1987 for( final DomainId mpd : mpds ) {
1988 per_genome_domain_promiscuity_statistics_writer.write( mpd.getId() + " " );
1990 per_genome_domain_promiscuity_statistics_writer.write( ForesterUtil.LINE_SEPARATOR );
1992 catch ( final IOException e ) {
1993 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1995 if ( input_file_properties[ i ].length == 3 ) {
1996 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote domain combination counts for \""
1997 + input_file_properties[ i ][ 0 ] + "\" (" + input_file_properties[ i ][ 1 ] + ", "
1998 + input_file_properties[ i ][ 2 ] + ") to: \"" + dc_outfile + "\"" );
2001 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote domain combination counts for \""
2002 + input_file_properties[ i ][ 0 ] + "\" (" + input_file_properties[ i ][ 1 ] + ") to: \""
2003 + dc_outfile + "\"" );
2007 private static void writeDomainData( final Map<DomainId, List<GoId>> domain_id_to_go_ids_map,
2008 final Map<GoId, GoTerm> go_id_to_term_map,
2009 final GoNameSpace go_namespace_limit,
2011 final String domain_0,
2012 final String domain_1,
2013 final String prefix_for_html,
2014 final String character_separator_for_non_html_output,
2015 final Map<DomainId, Set<String>>[] domain_id_to_secondary_features_maps,
2016 final Set<GoId> all_go_ids ) throws IOException {
2017 boolean any_go_annotation_present = false;
2018 boolean first_has_no_go = false;
2019 int domain_count = 2; // To distinguish between domains and binary domain combinations.
2020 if ( ForesterUtil.isEmpty( domain_1 ) ) {
2023 // The following has a difficult to understand logic.
2024 for( int d = 0; d < domain_count; ++d ) {
2025 List<GoId> go_ids = null;
2026 boolean go_annotation_present = false;
2028 final DomainId domain_id = new DomainId( domain_0 );
2029 if ( domain_id_to_go_ids_map.containsKey( domain_id ) ) {
2030 go_annotation_present = true;
2031 any_go_annotation_present = true;
2032 go_ids = domain_id_to_go_ids_map.get( domain_id );
2035 first_has_no_go = true;
2039 final DomainId domain_id = new DomainId( domain_1 );
2040 if ( domain_id_to_go_ids_map.containsKey( domain_id ) ) {
2041 go_annotation_present = true;
2042 any_go_annotation_present = true;
2043 go_ids = domain_id_to_go_ids_map.get( domain_id );
2046 if ( go_annotation_present ) {
2047 boolean first = ( ( d == 0 ) || ( ( d == 1 ) && first_has_no_go ) );
2048 for( final GoId go_id : go_ids ) {
2049 out.write( "<tr>" );
2052 writeDomainIdsToHtml( out,
2056 domain_id_to_secondary_features_maps );
2059 out.write( "<td></td>" );
2061 if ( !go_id_to_term_map.containsKey( go_id ) ) {
2062 throw new IllegalArgumentException( "GO-id [" + go_id + "] not found in GO-id to GO-term map" );
2064 final GoTerm go_term = go_id_to_term_map.get( go_id );
2065 if ( ( go_namespace_limit == null ) || go_namespace_limit.equals( go_term.getGoNameSpace() ) ) {
2066 // final String top = GoUtils.getPenultimateGoTerm( go_term, go_id_to_term_map ).getName();
2067 final String go_id_str = go_id.getId();
2068 out.write( "<td>" );
2069 out.write( "<a href=\"" + SurfacingConstants.AMIGO_LINK + go_id_str
2070 + "\" target=\"amigo_window\">" + go_id_str + "</a>" );
2071 out.write( "</td><td>" );
2072 out.write( go_term.getName() );
2073 if ( domain_count == 2 ) {
2074 out.write( " (" + d + ")" );
2076 out.write( "</td><td>" );
2077 // out.write( top );
2078 // out.write( "</td><td>" );
2080 out.write( go_term.getGoNameSpace().toShortString() );
2082 out.write( "</td>" );
2083 if ( all_go_ids != null ) {
2084 all_go_ids.add( go_id );
2088 out.write( "<td>" );
2089 out.write( "</td><td>" );
2090 out.write( "</td><td>" );
2091 out.write( "</td><td>" );
2092 out.write( "</td>" );
2094 out.write( "</tr>" );
2095 out.write( SurfacingConstants.NL );
2098 } // for( int d = 0; d < domain_count; ++d )
2099 if ( !any_go_annotation_present ) {
2100 out.write( "<tr>" );
2101 writeDomainIdsToHtml( out, domain_0, domain_1, prefix_for_html, domain_id_to_secondary_features_maps );
2102 out.write( "<td>" );
2103 out.write( "</td><td>" );
2104 out.write( "</td><td>" );
2105 out.write( "</td><td>" );
2106 out.write( "</td>" );
2107 out.write( "</tr>" );
2108 out.write( SurfacingConstants.NL );
2112 private static void writeDomainIdsToHtml( final Writer out,
2113 final String domain_0,
2114 final String domain_1,
2115 final String prefix_for_detailed_html,
2116 final Map<DomainId, Set<String>>[] domain_id_to_secondary_features_maps )
2117 throws IOException {
2118 out.write( "<td>" );
2119 if ( !ForesterUtil.isEmpty( prefix_for_detailed_html ) ) {
2120 out.write( prefix_for_detailed_html );
2123 out.write( "<a href=\"" + SurfacingConstants.PFAM_FAMILY_ID_LINK + domain_0 + "\">" + domain_0 + "</a>" );
2124 out.write( "</td>" );
2127 public static DescriptiveStatistics writeDomainSimilaritiesToFile( final StringBuilder html_desc,
2128 final StringBuilder html_title,
2129 final Writer single_writer,
2130 Map<Character, Writer> split_writers,
2131 final SortedSet<DomainSimilarity> similarities,
2132 final boolean treat_as_binary,
2133 final List<Species> species_order,
2134 final PrintableDomainSimilarity.PRINT_OPTION print_option,
2135 final DomainSimilarity.DomainSimilaritySortField sort_field,
2136 final DomainSimilarity.DomainSimilarityScoring scoring,
2137 final boolean verbose ) throws IOException {
2138 final DescriptiveStatistics stats = new BasicDescriptiveStatistics();
2139 String histogram_title = null;
2140 switch ( sort_field ) {
2141 case ABS_MAX_COUNTS_DIFFERENCE:
2142 if ( treat_as_binary ) {
2143 histogram_title = "absolute counts difference:";
2146 histogram_title = "absolute (maximal) counts difference:";
2149 case MAX_COUNTS_DIFFERENCE:
2150 if ( treat_as_binary ) {
2151 histogram_title = "counts difference:";
2154 histogram_title = "(maximal) counts difference:";
2158 histogram_title = "score mean:";
2161 histogram_title = "score minimum:";
2164 histogram_title = "score maximum:";
2166 case MAX_DIFFERENCE:
2167 if ( treat_as_binary ) {
2168 histogram_title = "difference:";
2171 histogram_title = "(maximal) difference:";
2175 histogram_title = "score mean:";
2178 histogram_title = "score standard deviation:";
2181 histogram_title = "species number:";
2184 throw new AssertionError( "Unknown sort field: " + sort_field );
2186 for( final DomainSimilarity similarity : similarities ) {
2187 switch ( sort_field ) {
2188 case ABS_MAX_COUNTS_DIFFERENCE:
2189 stats.addValue( Math.abs( similarity.getMaximalDifferenceInCounts() ) );
2191 case MAX_COUNTS_DIFFERENCE:
2192 stats.addValue( similarity.getMaximalDifferenceInCounts() );
2195 stats.addValue( similarity.getMeanSimilarityScore() );
2198 stats.addValue( similarity.getMinimalSimilarityScore() );
2201 stats.addValue( similarity.getMaximalSimilarityScore() );
2203 case MAX_DIFFERENCE:
2204 stats.addValue( similarity.getMaximalDifference() );
2207 stats.addValue( similarity.getMeanSimilarityScore() );
2210 stats.addValue( similarity.getStandardDeviationOfSimilarityScore() );
2213 stats.addValue( similarity.getSpecies().size() );
2216 throw new AssertionError( "Unknown sort field: " + sort_field );
2220 // final HistogramData[] hists = new HistogramData[ 1 ];
2223 // List<HistogramDataItem> data_items = new
2224 // ArrayList<HistogramDataItem>();
2225 // double[] values = stats.getDataAsDoubleArray();
2226 // for( int i = 0; i < values.length; i++ ) {
2227 // HistogramDataItem data_item = new BasicHistogramDataItem( "", values[
2229 // data_items.add( data_item );
2233 // HistogramData hd0 = new HistogramData( "name",
2241 // hists[ 0 ] = hd0;
2243 // final HistogramsFrame hf = new HistogramsFrame( hists );
2244 // hf.setVisible( true );
2246 AsciiHistogram histo = null;
2247 if ( stats.getMin() < stats.getMin() ) {
2248 histo = new AsciiHistogram( stats, histogram_title );
2251 if ( histo != null ) {
2252 System.out.println( histo.toStringBuffer( 20, '|', 40, 5 ) );
2254 System.out.println();
2255 System.out.println( "N : " + stats.getN() );
2256 System.out.println( "Min : " + stats.getMin() );
2257 System.out.println( "Max : " + stats.getMax() );
2258 System.out.println( "Mean : " + stats.arithmeticMean() );
2259 if ( stats.getN() > 1 ) {
2260 System.out.println( "SD : " + stats.sampleStandardDeviation() );
2263 System.out.println( "SD : n/a" );
2265 System.out.println( "Median : " + stats.median() );
2266 if ( stats.getN() > 1 ) {
2267 System.out.println( "Pearsonian skewness : " + stats.pearsonianSkewness() );
2270 System.out.println( "Pearsonian skewness : n/a" );
2273 if ( ( single_writer != null ) && ( ( split_writers == null ) || split_writers.isEmpty() ) ) {
2274 split_writers = new HashMap<Character, Writer>();
2275 split_writers.put( '_', single_writer );
2277 switch ( print_option ) {
2278 case SIMPLE_TAB_DELIMITED:
2281 for( final Character key : split_writers.keySet() ) {
2282 final Writer w = split_writers.get( key );
2283 w.write( "<html>" );
2284 w.write( SurfacingConstants.NL );
2286 addHtmlHead( w, "DCs (" + html_title + ") " + key.toString().toUpperCase() );
2289 addHtmlHead( w, "DCs (" + html_title + ")" );
2291 w.write( SurfacingConstants.NL );
2292 w.write( "<body>" );
2293 w.write( SurfacingConstants.NL );
2294 w.write( html_desc.toString() );
2295 w.write( SurfacingConstants.NL );
2298 w.write( SurfacingConstants.NL );
2299 w.write( "<tt><pre>" );
2300 w.write( SurfacingConstants.NL );
2301 if ( histo != null ) {
2302 w.write( histo.toStringBuffer( 20, '|', 40, 5 ).toString() );
2303 w.write( SurfacingConstants.NL );
2305 w.write( "</pre></tt>" );
2306 w.write( SurfacingConstants.NL );
2307 w.write( "<table>" );
2308 w.write( SurfacingConstants.NL );
2309 w.write( "<tr><td>N: </td><td>" + stats.getN() + "</td></tr>" );
2310 w.write( SurfacingConstants.NL );
2311 w.write( "<tr><td>Min: </td><td>" + stats.getMin() + "</td></tr>" );
2312 w.write( SurfacingConstants.NL );
2313 w.write( "<tr><td>Max: </td><td>" + stats.getMax() + "</td></tr>" );
2314 w.write( SurfacingConstants.NL );
2315 w.write( "<tr><td>Mean: </td><td>" + stats.arithmeticMean() + "</td></tr>" );
2316 w.write( SurfacingConstants.NL );
2317 if ( stats.getN() > 1 ) {
2318 w.write( "<tr><td>SD: </td><td>" + stats.sampleStandardDeviation() + "</td></tr>" );
2321 w.write( "<tr><td>SD: </td><td>n/a</td></tr>" );
2323 w.write( SurfacingConstants.NL );
2324 w.write( "<tr><td>Median: </td><td>" + stats.median() + "</td></tr>" );
2325 w.write( SurfacingConstants.NL );
2326 if ( stats.getN() > 1 ) {
2327 w.write( "<tr><td>Pearsonian skewness: </td><td>" + stats.pearsonianSkewness() + "</td></tr>" );
2330 w.write( "<tr><td>Pearsonian skewness: </td><td>n/a</td></tr>" );
2332 w.write( SurfacingConstants.NL );
2333 w.write( "</table>" );
2334 w.write( SurfacingConstants.NL );
2336 w.write( SurfacingConstants.NL );
2338 w.write( SurfacingConstants.NL );
2340 w.write( SurfacingConstants.NL );
2341 w.write( "<table>" );
2342 w.write( SurfacingConstants.NL );
2346 for( final Writer w : split_writers.values() ) {
2347 w.write( SurfacingConstants.NL );
2349 for( final DomainSimilarity similarity : similarities ) {
2350 if ( ( species_order != null ) && !species_order.isEmpty() ) {
2351 ( ( PrintableDomainSimilarity ) similarity ).setSpeciesOrder( species_order );
2353 if ( single_writer != null ) {
2354 single_writer.write( similarity.toStringBuffer( print_option ).toString() );
2357 Writer local_writer = split_writers.get( ( similarity.getDomainId().getId().charAt( 0 ) + "" )
2358 .toLowerCase().charAt( 0 ) );
2359 if ( local_writer == null ) {
2360 local_writer = split_writers.get( '0' );
2362 local_writer.write( similarity.toStringBuffer( print_option ).toString() );
2364 for( final Writer w : split_writers.values() ) {
2365 w.write( SurfacingConstants.NL );
2368 switch ( print_option ) {
2370 for( final Writer w : split_writers.values() ) {
2371 w.write( SurfacingConstants.NL );
2372 w.write( "</table>" );
2373 w.write( SurfacingConstants.NL );
2374 w.write( "</font>" );
2375 w.write( SurfacingConstants.NL );
2376 w.write( "</body>" );
2377 w.write( SurfacingConstants.NL );
2378 w.write( "</html>" );
2379 w.write( SurfacingConstants.NL );
2383 for( final Writer w : split_writers.values() ) {
2389 private static void writeDomainsToIndividualFilePerTreeNode( final Writer individual_files_writer,
2390 final String domain_0,
2391 final String domain_1 ) throws IOException {
2392 individual_files_writer.write( domain_0 );
2393 individual_files_writer.write( ForesterUtil.LINE_SEPARATOR );
2394 if ( !ForesterUtil.isEmpty( domain_1 ) ) {
2395 individual_files_writer.write( domain_1 );
2396 individual_files_writer.write( ForesterUtil.LINE_SEPARATOR );
2400 public static void writeMatrixToFile( final CharacterStateMatrix<?> matrix,
2401 final String filename,
2402 final Format format ) {
2403 final File outfile = new File( filename );
2404 checkForOutputFileWriteability( outfile );
2406 final BufferedWriter out = new BufferedWriter( new FileWriter( outfile ) );
2407 matrix.toWriter( out, format );
2411 catch ( final IOException e ) {
2412 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
2414 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote matrix: \"" + filename + "\"" );
2417 public static void writeMatrixToFile( final File matrix_outfile, final List<DistanceMatrix> matrices ) {
2418 checkForOutputFileWriteability( matrix_outfile );
2420 final BufferedWriter out = new BufferedWriter( new FileWriter( matrix_outfile ) );
2421 for( final DistanceMatrix distance_matrix : matrices ) {
2422 out.write( distance_matrix.toStringBuffer( DistanceMatrix.Format.PHYLIP ).toString() );
2423 out.write( ForesterUtil.LINE_SEPARATOR );
2428 catch ( final IOException e ) {
2429 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
2431 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote distance matrices to \"" + matrix_outfile + "\"" );
2434 private static void writePfamsToFile( final String outfile_name, final SortedSet<String> pfams ) {
2436 final Writer writer = new BufferedWriter( new FileWriter( new File( outfile_name ) ) );
2437 for( final String pfam : pfams ) {
2438 writer.write( pfam );
2439 writer.write( ForesterUtil.LINE_SEPARATOR );
2442 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote " + pfams.size() + " pfams to [" + outfile_name
2445 catch ( final IOException e ) {
2446 ForesterUtil.printWarningMessage( surfacing.PRG_NAME, "Failure to write: " + e );
2450 public static void writePhylogenyToFile( final Phylogeny phylogeny, final String filename ) {
2451 final PhylogenyWriter writer = new PhylogenyWriter();
2453 writer.toPhyloXML( new File( filename ), phylogeny, 1 );
2455 catch ( final IOException e ) {
2456 ForesterUtil.printWarningMessage( surfacing.PRG_NAME, "failed to write phylogeny to \"" + filename + "\": "
2459 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote phylogeny to \"" + filename + "\"" );
2462 public static void writeTaxonomyLinks( final Writer writer, final String species ) throws IOException {
2463 if ( ( species.length() > 1 ) && ( species.indexOf( '_' ) < 1 ) ) {
2464 final Matcher matcher = PATTERN_SP_STYLE_TAXONOMY.matcher( species );
2465 writer.write( " [" );
2466 if ( matcher.matches() ) {
2467 writer.write( "<a href=\"" + SurfacingConstants.UNIPROT_LINK + species
2468 + "\" target=\"taxonomy_window\">uniprot</a>" );
2471 writer.write( "<a href=\"" + SurfacingConstants.EOL_LINK + species
2472 + "\" target=\"taxonomy_window\">eol</a>" );
2473 writer.write( "|" );
2474 writer.write( "<a href=\"" + SurfacingConstants.TOL_LINK + species
2475 + "\" target=\"taxonomy_window\">tol</a>" );
2477 writer.write( "]" );
2481 private static void writeToNexus( final String outfile_name,
2482 final CharacterStateMatrix<BinaryStates> matrix,
2483 final Phylogeny phylogeny ) {
2484 if ( !( matrix instanceof BasicCharacterStateMatrix ) ) {
2485 throw new IllegalArgumentException( "can only write matrices of type [" + BasicCharacterStateMatrix.class
2488 final BasicCharacterStateMatrix<BinaryStates> my_matrix = ( org.forester.evoinference.matrix.character.BasicCharacterStateMatrix<BinaryStates> ) matrix;
2489 final List<Phylogeny> phylogenies = new ArrayList<Phylogeny>( 1 );
2490 phylogenies.add( phylogeny );
2492 final BufferedWriter w = new BufferedWriter( new FileWriter( outfile_name ) );
2493 w.write( NexusConstants.NEXUS );
2494 w.write( ForesterUtil.LINE_SEPARATOR );
2495 my_matrix.writeNexusTaxaBlock( w );
2496 my_matrix.writeNexusBinaryChractersBlock( w );
2497 PhylogenyWriter.writeNexusTreesBlock( w, phylogenies, NH_CONVERSION_SUPPORT_VALUE_STYLE.NONE );
2500 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote Nexus file: \"" + outfile_name + "\"" );
2502 catch ( final IOException e ) {
2503 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
2507 private static void writeToNexus( final String outfile_name,
2508 final DomainParsimonyCalculator domain_parsimony,
2509 final Phylogeny phylogeny ) {
2510 writeToNexus( outfile_name + surfacing.NEXUS_EXTERNAL_DOMAINS,
2511 domain_parsimony.createMatrixOfDomainPresenceOrAbsence(),
2513 writeToNexus( outfile_name + surfacing.NEXUS_EXTERNAL_DOMAIN_COMBINATIONS,
2514 domain_parsimony.createMatrixOfBinaryDomainCombinationPresenceOrAbsence(),
2518 public static void domainsPerProteinsStatistics( final String genome,
2519 final List<Protein> protein_list,
2520 final DescriptiveStatistics all_genomes_domains_per_potein_stats,
2521 final SortedMap<Integer, Integer> all_genomes_domains_per_potein_histo,
2522 final SortedSet<String> domains_which_are_always_single,
2523 final SortedSet<String> domains_which_are_sometimes_single_sometimes_not,
2524 final SortedSet<String> domains_which_never_single,
2525 final Writer writer ) {
2526 final DescriptiveStatistics stats = new BasicDescriptiveStatistics();
2527 for( final Protein protein : protein_list ) {
2528 final int domains = protein.getNumberOfProteinDomains();
2529 //System.out.println( domains );
2530 stats.addValue( domains );
2531 all_genomes_domains_per_potein_stats.addValue( domains );
2532 if ( !all_genomes_domains_per_potein_histo.containsKey( domains ) ) {
2533 all_genomes_domains_per_potein_histo.put( domains, 1 );
2536 all_genomes_domains_per_potein_histo.put( domains,
2537 1 + all_genomes_domains_per_potein_histo.get( domains ) );
2539 if ( domains == 1 ) {
2540 final String domain = protein.getProteinDomain( 0 ).getDomainId().getId();
2541 if ( !domains_which_are_sometimes_single_sometimes_not.contains( domain ) ) {
2542 if ( domains_which_never_single.contains( domain ) ) {
2543 domains_which_never_single.remove( domain );
2544 domains_which_are_sometimes_single_sometimes_not.add( domain );
2547 domains_which_are_always_single.add( domain );
2551 else if ( domains > 1 ) {
2552 for( final Domain d : protein.getProteinDomains() ) {
2553 final String domain = d.getDomainId().getId();
2554 // System.out.println( domain );
2555 if ( !domains_which_are_sometimes_single_sometimes_not.contains( domain ) ) {
2556 if ( domains_which_are_always_single.contains( domain ) ) {
2557 domains_which_are_always_single.remove( domain );
2558 domains_which_are_sometimes_single_sometimes_not.add( domain );
2561 domains_which_never_single.add( domain );
2568 writer.write( genome );
2569 writer.write( "\t" );
2570 if ( stats.getN() >= 1 ) {
2571 writer.write( stats.arithmeticMean() + "" );
2572 writer.write( "\t" );
2573 if ( stats.getN() >= 2 ) {
2574 writer.write( stats.sampleStandardDeviation() + "" );
2579 writer.write( "\t" );
2580 writer.write( stats.median() + "" );
2581 writer.write( "\t" );
2582 writer.write( stats.getN() + "" );
2583 writer.write( "\t" );
2584 writer.write( stats.getMin() + "" );
2585 writer.write( "\t" );
2586 writer.write( stats.getMax() + "" );
2589 writer.write( "\t" );
2590 writer.write( "\t" );
2591 writer.write( "\t" );
2592 writer.write( "0" );
2593 writer.write( "\t" );
2594 writer.write( "\t" );
2596 writer.write( "\n" );
2598 catch ( final IOException e ) {
2599 e.printStackTrace();