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.BasicSymmetricalDistanceMatrix;
62 import org.forester.evoinference.matrix.distance.DistanceMatrix;
63 import org.forester.go.GoId;
64 import org.forester.go.GoNameSpace;
65 import org.forester.go.GoTerm;
66 import org.forester.go.PfamToGoMapping;
67 import org.forester.io.parsers.nexus.NexusConstants;
68 import org.forester.io.writers.PhylogenyWriter;
69 import org.forester.msa.MsaCompactor.SORT_BY;
70 import org.forester.phylogeny.Phylogeny;
71 import org.forester.phylogeny.PhylogenyMethods;
72 import org.forester.phylogeny.PhylogenyNode;
73 import org.forester.phylogeny.PhylogenyNodeI.NH_CONVERSION_SUPPORT_VALUE_STYLE;
74 import org.forester.phylogeny.data.BinaryCharacters;
75 import org.forester.phylogeny.data.Confidence;
76 import org.forester.phylogeny.iterators.PhylogenyNodeIterator;
77 import org.forester.protein.BasicDomain;
78 import org.forester.protein.BasicProtein;
79 import org.forester.protein.BinaryDomainCombination;
80 import org.forester.protein.Domain;
81 import org.forester.protein.DomainId;
82 import org.forester.protein.Protein;
83 import org.forester.species.Species;
84 import org.forester.surfacing.DomainSimilarityCalculator.Detailedness;
85 import org.forester.surfacing.DomainSimilarityCalculator.GoAnnotationOutput;
86 import org.forester.surfacing.GenomeWideCombinableDomains.GenomeWideCombinableDomainsSortOrder;
87 import org.forester.util.AsciiHistogram;
88 import org.forester.util.BasicDescriptiveStatistics;
89 import org.forester.util.BasicTable;
90 import org.forester.util.BasicTableParser;
91 import org.forester.util.DescriptiveStatistics;
92 import org.forester.util.ForesterUtil;
94 public final class SurfacingUtil {
96 private final static NumberFormat FORMATTER = new DecimalFormat( "0.0E0" );
97 private final static NumberFormat FORMATTER_3 = new DecimalFormat( "0.000" );
98 private static final Comparator<Domain> ASCENDING_CONFIDENCE_VALUE_ORDER = new Comparator<Domain>() {
101 public int compare( final Domain d1,
103 if ( d1.getPerSequenceEvalue() < d2
104 .getPerSequenceEvalue() ) {
108 .getPerSequenceEvalue() > d2
109 .getPerSequenceEvalue() ) {
113 return d1.compareTo( d2 );
117 public final static Pattern PATTERN_SP_STYLE_TAXONOMY = Pattern.compile( "^[A-Z0-9]{3,5}$" );
118 private static final boolean USE_LAST = true;
120 private SurfacingUtil() {
121 // Hidden constructor.
124 public static void addAllBinaryDomainCombinationToSet( final GenomeWideCombinableDomains genome,
125 final SortedSet<BinaryDomainCombination> binary_domain_combinations ) {
126 final SortedMap<DomainId, CombinableDomains> all_cd = genome.getAllCombinableDomainsIds();
127 for( final DomainId domain_id : all_cd.keySet() ) {
128 binary_domain_combinations.addAll( all_cd.get( domain_id ).toBinaryDomainCombinations() );
132 public static void addAllDomainIdsToSet( final GenomeWideCombinableDomains genome,
133 final SortedSet<DomainId> domain_ids ) {
134 final SortedSet<DomainId> domains = genome.getAllDomainIds();
135 for( final DomainId domain : domains ) {
136 domain_ids.add( domain );
140 public static void addHtmlHead( final Writer w, final String title ) throws IOException {
141 w.write( SurfacingConstants.NL );
143 w.write( "<title>" );
145 w.write( "</title>" );
146 w.write( SurfacingConstants.NL );
147 w.write( "<style>" );
148 w.write( SurfacingConstants.NL );
149 w.write( "a:visited { color : #6633FF; text-decoration : none; }" );
150 w.write( SurfacingConstants.NL );
151 w.write( "a:link { color : #6633FF; text-decoration : none; }" );
152 w.write( SurfacingConstants.NL );
153 w.write( "a:active { color : #99FF00; text-decoration : none; }" );
154 w.write( SurfacingConstants.NL );
155 w.write( "a:hover { color : #FFFFFF; background-color : #99FF00; text-decoration : none; }" );
156 w.write( SurfacingConstants.NL );
157 w.write( "td { text-align: left; vertical-align: top; font-family: Verdana, Arial, Helvetica; font-size: 8pt}" );
158 w.write( SurfacingConstants.NL );
159 w.write( "h1 { color : #0000FF; font-family: Verdana, Arial, Helvetica; font-size: 18pt; font-weight: bold }" );
160 w.write( SurfacingConstants.NL );
161 w.write( "h2 { color : #0000FF; font-family: Verdana, Arial, Helvetica; font-size: 16pt; font-weight: bold }" );
162 w.write( SurfacingConstants.NL );
163 w.write( "</style>" );
164 w.write( SurfacingConstants.NL );
165 w.write( "</head>" );
166 w.write( SurfacingConstants.NL );
169 public static DescriptiveStatistics calculateDescriptiveStatisticsForMeanValues( final Set<DomainSimilarity> similarities ) {
170 final DescriptiveStatistics stats = new BasicDescriptiveStatistics();
171 for( final DomainSimilarity similarity : similarities ) {
172 stats.addValue( similarity.getMeanSimilarityScore() );
177 private static void calculateIndependentDomainCombinationGains( final Phylogeny local_phylogeny_l,
178 final String outfilename_for_counts,
179 final String outfilename_for_dc,
180 final String outfilename_for_dc_for_go_mapping,
181 final String outfilename_for_dc_for_go_mapping_unique,
182 final String outfilename_for_rank_counts,
183 final String outfilename_for_ancestor_species_counts,
184 final String outfilename_for_protein_stats,
185 final Map<String, DescriptiveStatistics> protein_length_stats_by_dc,
186 final Map<String, DescriptiveStatistics> domain_number_stats_by_dc,
187 final Map<String, DescriptiveStatistics> domain_length_stats_by_domain ) {
190 // if ( protein_length_stats_by_dc != null ) {
191 // for( final Entry<?, DescriptiveStatistics> entry : protein_length_stats_by_dc.entrySet() ) {
192 // System.out.print( entry.getKey().toString() );
193 // System.out.print( ": " );
194 // double[] a = entry.getValue().getDataAsDoubleArray();
195 // for( int i = 0; i < a.length; i++ ) {
196 // System.out.print( a[ i ] + " " );
198 // System.out.println();
201 // if ( domain_number_stats_by_dc != null ) {
202 // for( final Entry<?, DescriptiveStatistics> entry : domain_number_stats_by_dc.entrySet() ) {
203 // System.out.print( entry.getKey().toString() );
204 // System.out.print( ": " );
205 // double[] a = entry.getValue().getDataAsDoubleArray();
206 // for( int i = 0; i < a.length; i++ ) {
207 // System.out.print( a[ i ] + " " );
209 // System.out.println();
213 final BufferedWriter out_counts = new BufferedWriter( new FileWriter( outfilename_for_counts ) );
214 final BufferedWriter out_dc = new BufferedWriter( new FileWriter( outfilename_for_dc ) );
215 final BufferedWriter out_dc_for_go_mapping = new BufferedWriter( new FileWriter( outfilename_for_dc_for_go_mapping ) );
216 final BufferedWriter out_dc_for_go_mapping_unique = new BufferedWriter( new FileWriter( outfilename_for_dc_for_go_mapping_unique ) );
217 final SortedMap<String, Integer> dc_gain_counts = new TreeMap<String, Integer>();
218 for( final PhylogenyNodeIterator it = local_phylogeny_l.iteratorPostorder(); it.hasNext(); ) {
219 final PhylogenyNode n = it.next();
220 final Set<String> gained_dc = n.getNodeData().getBinaryCharacters().getGainedCharacters();
221 for( final String dc : gained_dc ) {
222 if ( dc_gain_counts.containsKey( dc ) ) {
223 dc_gain_counts.put( dc, dc_gain_counts.get( dc ) + 1 );
226 dc_gain_counts.put( dc, 1 );
230 final SortedMap<Integer, Integer> histogram = new TreeMap<Integer, Integer>();
231 final SortedMap<Integer, StringBuilder> domain_lists = new TreeMap<Integer, StringBuilder>();
232 final SortedMap<Integer, DescriptiveStatistics> dc_reapp_counts_to_protein_length_stats = new TreeMap<Integer, DescriptiveStatistics>();
233 final SortedMap<Integer, DescriptiveStatistics> dc_reapp_counts_to_domain_number_stats = new TreeMap<Integer, DescriptiveStatistics>();
234 final SortedMap<Integer, DescriptiveStatistics> dc_reapp_counts_to_domain_lengths_stats = new TreeMap<Integer, DescriptiveStatistics>();
235 final SortedMap<Integer, PriorityQueue<String>> domain_lists_go = new TreeMap<Integer, PriorityQueue<String>>();
236 final SortedMap<Integer, SortedSet<String>> domain_lists_go_unique = new TreeMap<Integer, SortedSet<String>>();
237 final Set<String> dcs = dc_gain_counts.keySet();
238 final SortedSet<String> more_than_once = new TreeSet<String>();
239 final DescriptiveStatistics gained_once_lengths_stats = new BasicDescriptiveStatistics();
240 final DescriptiveStatistics gained_once_domain_count_stats = new BasicDescriptiveStatistics();
241 final DescriptiveStatistics gained_multiple_times_lengths_stats = new BasicDescriptiveStatistics();
242 final DescriptiveStatistics gained_multiple_times_domain_count_stats = new BasicDescriptiveStatistics();
243 long gained_multiple_times_domain_length_sum = 0;
244 long gained_once_domain_length_sum = 0;
245 long gained_multiple_times_domain_length_count = 0;
246 long gained_once_domain_length_count = 0;
247 for( final String dc : dcs ) {
248 final int count = dc_gain_counts.get( dc );
249 if ( histogram.containsKey( count ) ) {
250 histogram.put( count, histogram.get( count ) + 1 );
251 domain_lists.get( count ).append( ", " + dc );
252 domain_lists_go.get( count ).addAll( splitDomainCombination( dc ) );
253 domain_lists_go_unique.get( count ).addAll( splitDomainCombination( dc ) );
256 histogram.put( count, 1 );
257 domain_lists.put( count, new StringBuilder( dc ) );
258 final PriorityQueue<String> q = new PriorityQueue<String>();
259 q.addAll( splitDomainCombination( dc ) );
260 domain_lists_go.put( count, q );
261 final SortedSet<String> set = new TreeSet<String>();
262 set.addAll( splitDomainCombination( dc ) );
263 domain_lists_go_unique.put( count, set );
265 if ( protein_length_stats_by_dc != null ) {
266 if ( !dc_reapp_counts_to_protein_length_stats.containsKey( count ) ) {
267 dc_reapp_counts_to_protein_length_stats.put( count, new BasicDescriptiveStatistics() );
269 dc_reapp_counts_to_protein_length_stats.get( count ).addValue( protein_length_stats_by_dc.get( dc )
272 if ( domain_number_stats_by_dc != null ) {
273 if ( !dc_reapp_counts_to_domain_number_stats.containsKey( count ) ) {
274 dc_reapp_counts_to_domain_number_stats.put( count, new BasicDescriptiveStatistics() );
276 dc_reapp_counts_to_domain_number_stats.get( count ).addValue( domain_number_stats_by_dc.get( dc )
279 if ( domain_length_stats_by_domain != null ) {
280 if ( !dc_reapp_counts_to_domain_lengths_stats.containsKey( count ) ) {
281 dc_reapp_counts_to_domain_lengths_stats.put( count, new BasicDescriptiveStatistics() );
283 final String[] ds = dc.split( "=" );
284 dc_reapp_counts_to_domain_lengths_stats.get( count ).addValue( domain_length_stats_by_domain
285 .get( ds[ 0 ] ).arithmeticMean() );
286 dc_reapp_counts_to_domain_lengths_stats.get( count ).addValue( domain_length_stats_by_domain
287 .get( ds[ 1 ] ).arithmeticMean() );
290 more_than_once.add( dc );
291 if ( protein_length_stats_by_dc != null ) {
292 final DescriptiveStatistics s = protein_length_stats_by_dc.get( dc );
293 for( final double element : s.getData() ) {
294 gained_multiple_times_lengths_stats.addValue( element );
297 if ( domain_number_stats_by_dc != null ) {
298 final DescriptiveStatistics s = domain_number_stats_by_dc.get( dc );
299 for( final double element : s.getData() ) {
300 gained_multiple_times_domain_count_stats.addValue( element );
303 if ( domain_length_stats_by_domain != null ) {
304 final String[] ds = dc.split( "=" );
305 final DescriptiveStatistics s0 = domain_length_stats_by_domain.get( ds[ 0 ] );
306 final DescriptiveStatistics s1 = domain_length_stats_by_domain.get( ds[ 1 ] );
307 for( final double element : s0.getData() ) {
308 gained_multiple_times_domain_length_sum += element;
309 ++gained_multiple_times_domain_length_count;
311 for( final double element : s1.getData() ) {
312 gained_multiple_times_domain_length_sum += element;
313 ++gained_multiple_times_domain_length_count;
318 if ( protein_length_stats_by_dc != null ) {
319 final DescriptiveStatistics s = protein_length_stats_by_dc.get( dc );
320 for( final double element : s.getData() ) {
321 gained_once_lengths_stats.addValue( element );
324 if ( domain_number_stats_by_dc != null ) {
325 final DescriptiveStatistics s = domain_number_stats_by_dc.get( dc );
326 for( final double element : s.getData() ) {
327 gained_once_domain_count_stats.addValue( element );
330 if ( domain_length_stats_by_domain != null ) {
331 final String[] ds = dc.split( "=" );
332 final DescriptiveStatistics s0 = domain_length_stats_by_domain.get( ds[ 0 ] );
333 final DescriptiveStatistics s1 = domain_length_stats_by_domain.get( ds[ 1 ] );
334 for( final double element : s0.getData() ) {
335 gained_once_domain_length_sum += element;
336 ++gained_once_domain_length_count;
338 for( final double element : s1.getData() ) {
339 gained_once_domain_length_sum += element;
340 ++gained_once_domain_length_count;
345 final Set<Integer> histogram_keys = histogram.keySet();
346 for( final Integer histogram_key : histogram_keys ) {
347 final int count = histogram.get( histogram_key );
348 final StringBuilder dc = domain_lists.get( histogram_key );
349 out_counts.write( histogram_key + "\t" + count + ForesterUtil.LINE_SEPARATOR );
350 out_dc.write( histogram_key + "\t" + dc + ForesterUtil.LINE_SEPARATOR );
351 out_dc_for_go_mapping.write( "#" + histogram_key + ForesterUtil.LINE_SEPARATOR );
352 final Object[] sorted = domain_lists_go.get( histogram_key ).toArray();
353 Arrays.sort( sorted );
354 for( final Object domain : sorted ) {
355 out_dc_for_go_mapping.write( domain + ForesterUtil.LINE_SEPARATOR );
357 out_dc_for_go_mapping_unique.write( "#" + histogram_key + ForesterUtil.LINE_SEPARATOR );
358 for( final String domain : domain_lists_go_unique.get( histogram_key ) ) {
359 out_dc_for_go_mapping_unique.write( domain + ForesterUtil.LINE_SEPARATOR );
364 out_dc_for_go_mapping.close();
365 out_dc_for_go_mapping_unique.close();
366 final SortedMap<String, Integer> lca_rank_counts = new TreeMap<String, Integer>();
367 final SortedMap<String, Integer> lca_ancestor_species_counts = new TreeMap<String, Integer>();
368 for( final String dc : more_than_once ) {
369 final List<PhylogenyNode> nodes = new ArrayList<PhylogenyNode>();
370 for( final PhylogenyNodeIterator it = local_phylogeny_l.iteratorExternalForward(); it.hasNext(); ) {
371 final PhylogenyNode n = it.next();
372 if ( n.getNodeData().getBinaryCharacters().getGainedCharacters().contains( dc ) ) {
376 for( int i = 0; i < nodes.size() - 1; ++i ) {
377 for( int j = i + 1; j < nodes.size(); ++j ) {
378 final PhylogenyNode lca = PhylogenyMethods.obtainLCA( nodes.get( i ),
380 String rank = "unknown";
381 if ( lca.getNodeData().isHasTaxonomy()
382 && !ForesterUtil.isEmpty( lca.getNodeData().getTaxonomy().getRank() ) ) {
383 rank = lca.getNodeData().getTaxonomy().getRank();
385 addToCountMap( lca_rank_counts, rank );
387 if ( lca.getNodeData().isHasTaxonomy()
388 && !ForesterUtil.isEmpty( lca.getNodeData().getTaxonomy().getScientificName() ) ) {
389 lca_species = lca.getNodeData().getTaxonomy().getScientificName();
391 else if ( lca.getNodeData().isHasTaxonomy()
392 && !ForesterUtil.isEmpty( lca.getNodeData().getTaxonomy().getCommonName() ) ) {
393 lca_species = lca.getNodeData().getTaxonomy().getCommonName();
396 lca_species = lca.getName();
398 addToCountMap( lca_ancestor_species_counts, lca_species );
402 final BufferedWriter out_for_rank_counts = new BufferedWriter( new FileWriter( outfilename_for_rank_counts ) );
403 final BufferedWriter out_for_ancestor_species_counts = new BufferedWriter( new FileWriter( outfilename_for_ancestor_species_counts ) );
404 ForesterUtil.map2writer( out_for_rank_counts, lca_rank_counts, "\t", ForesterUtil.LINE_SEPARATOR );
405 ForesterUtil.map2writer( out_for_ancestor_species_counts,
406 lca_ancestor_species_counts,
408 ForesterUtil.LINE_SEPARATOR );
409 out_for_rank_counts.close();
410 out_for_ancestor_species_counts.close();
411 if ( !ForesterUtil.isEmpty( outfilename_for_protein_stats )
412 && ( ( domain_length_stats_by_domain != null ) || ( protein_length_stats_by_dc != null ) || ( domain_number_stats_by_dc != null ) ) ) {
413 final BufferedWriter w = new BufferedWriter( new FileWriter( outfilename_for_protein_stats ) );
414 w.write( "Domain Lengths: " );
416 if ( domain_length_stats_by_domain != null ) {
417 for( final Entry<Integer, DescriptiveStatistics> entry : dc_reapp_counts_to_domain_lengths_stats
419 w.write( entry.getKey().toString() );
420 w.write( "\t" + entry.getValue().arithmeticMean() );
421 w.write( "\t" + entry.getValue().median() );
428 w.write( "Protein Lengths: " );
430 if ( protein_length_stats_by_dc != null ) {
431 for( final Entry<Integer, DescriptiveStatistics> entry : dc_reapp_counts_to_protein_length_stats
433 w.write( entry.getKey().toString() );
434 w.write( "\t" + entry.getValue().arithmeticMean() );
435 w.write( "\t" + entry.getValue().median() );
442 w.write( "Number of domains: " );
444 if ( domain_number_stats_by_dc != null ) {
445 for( final Entry<Integer, DescriptiveStatistics> entry : dc_reapp_counts_to_domain_number_stats
447 w.write( entry.getKey().toString() );
448 w.write( "\t" + entry.getValue().arithmeticMean() );
449 w.write( "\t" + entry.getValue().median() );
456 w.write( "Gained once, domain lengths:" );
458 w.write( "N: " + gained_once_domain_length_count );
460 w.write( "Avg: " + ( ( double ) gained_once_domain_length_sum / gained_once_domain_length_count ) );
463 w.write( "Gained multiple times, domain lengths:" );
465 w.write( "N: " + gained_multiple_times_domain_length_count );
468 + ( ( double ) gained_multiple_times_domain_length_sum / gained_multiple_times_domain_length_count ) );
473 w.write( "Gained once, protein lengths:" );
475 w.write( gained_once_lengths_stats.toString() );
478 w.write( "Gained once, domain counts:" );
480 w.write( gained_once_domain_count_stats.toString() );
483 w.write( "Gained multiple times, protein lengths:" );
485 w.write( gained_multiple_times_lengths_stats.toString() );
488 w.write( "Gained multiple times, domain counts:" );
490 w.write( gained_multiple_times_domain_count_stats.toString() );
495 catch ( final IOException e ) {
496 ForesterUtil.printWarningMessage( surfacing.PRG_NAME, "Failure to write: " + e );
498 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote independent domain combination gains fitch counts to ["
499 + outfilename_for_counts + "]" );
500 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote independent domain combination gains fitch lists to ["
501 + outfilename_for_dc + "]" );
502 ForesterUtil.programMessage( surfacing.PRG_NAME,
503 "Wrote independent domain combination gains fitch lists to (for GO mapping) ["
504 + outfilename_for_dc_for_go_mapping + "]" );
505 ForesterUtil.programMessage( surfacing.PRG_NAME,
506 "Wrote independent domain combination gains fitch lists to (for GO mapping, unique) ["
507 + outfilename_for_dc_for_go_mapping_unique + "]" );
510 private final static void addToCountMap( final Map<String, Integer> map, final String s ) {
511 if ( map.containsKey( s ) ) {
512 map.put( s, map.get( s ) + 1 );
519 public static int calculateOverlap( final Domain domain, final List<Boolean> covered_positions ) {
520 int overlap_count = 0;
521 for( int i = domain.getFrom(); i <= domain.getTo(); ++i ) {
522 if ( ( i < covered_positions.size() ) && ( covered_positions.get( i ) == true ) ) {
526 return overlap_count;
529 public static void checkForOutputFileWriteability( final File outfile ) {
530 final String error = ForesterUtil.isWritableFile( outfile );
531 if ( !ForesterUtil.isEmpty( error ) ) {
532 ForesterUtil.fatalError( surfacing.PRG_NAME, error );
536 private static SortedSet<String> collectAllDomainsChangedOnSubtree( final PhylogenyNode subtree_root,
537 final boolean get_gains ) {
538 final SortedSet<String> domains = new TreeSet<String>();
539 for( final PhylogenyNode descendant : PhylogenyMethods.getAllDescendants( subtree_root ) ) {
540 final BinaryCharacters chars = descendant.getNodeData().getBinaryCharacters();
542 domains.addAll( chars.getGainedCharacters() );
545 domains.addAll( chars.getLostCharacters() );
551 public static void collectChangedDomainCombinationsFromBinaryStatesMatrixAsListToFile( final CharacterStateMatrix<CharacterStateMatrix.GainLossStates> matrix,
552 final BinaryDomainCombination.DomainCombinationType dc_type,
553 final List<BinaryDomainCombination> all_binary_domains_combination_gained,
554 final boolean get_gains ) {
555 final SortedSet<String> sorted_ids = new TreeSet<String>();
556 for( int i = 0; i < matrix.getNumberOfIdentifiers(); ++i ) {
557 sorted_ids.add( matrix.getIdentifier( i ) );
559 for( final String id : sorted_ids ) {
560 for( int c = 0; c < matrix.getNumberOfCharacters(); ++c ) {
561 if ( ( get_gains && ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.GAIN ) )
562 || ( !get_gains && ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.LOSS ) ) ) {
563 if ( dc_type == BinaryDomainCombination.DomainCombinationType.DIRECTED_ADJACTANT ) {
564 all_binary_domains_combination_gained.add( AdjactantDirectedBinaryDomainCombination
565 .createInstance( matrix.getCharacter( c ) ) );
567 else if ( dc_type == BinaryDomainCombination.DomainCombinationType.DIRECTED ) {
568 all_binary_domains_combination_gained.add( DirectedBinaryDomainCombination
569 .createInstance( matrix.getCharacter( c ) ) );
572 all_binary_domains_combination_gained.add( BasicBinaryDomainCombination.createInstance( matrix
573 .getCharacter( c ) ) );
580 private static File createBaseDirForPerNodeDomainFiles( final String base_dir,
581 final boolean domain_combinations,
582 final CharacterStateMatrix.GainLossStates state,
583 final String outfile ) {
584 File per_node_go_mapped_domain_gain_loss_files_base_dir = new File( new File( outfile ).getParent()
585 + ForesterUtil.FILE_SEPARATOR + base_dir );
586 if ( !per_node_go_mapped_domain_gain_loss_files_base_dir.exists() ) {
587 per_node_go_mapped_domain_gain_loss_files_base_dir.mkdir();
589 if ( domain_combinations ) {
590 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
591 + ForesterUtil.FILE_SEPARATOR + "DC" );
594 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
595 + ForesterUtil.FILE_SEPARATOR + "DOMAINS" );
597 if ( !per_node_go_mapped_domain_gain_loss_files_base_dir.exists() ) {
598 per_node_go_mapped_domain_gain_loss_files_base_dir.mkdir();
600 if ( state == GainLossStates.GAIN ) {
601 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
602 + ForesterUtil.FILE_SEPARATOR + "GAINS" );
604 else if ( state == GainLossStates.LOSS ) {
605 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
606 + ForesterUtil.FILE_SEPARATOR + "LOSSES" );
609 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
610 + ForesterUtil.FILE_SEPARATOR + "PRESENT" );
612 if ( !per_node_go_mapped_domain_gain_loss_files_base_dir.exists() ) {
613 per_node_go_mapped_domain_gain_loss_files_base_dir.mkdir();
615 return per_node_go_mapped_domain_gain_loss_files_base_dir;
618 public static Map<DomainId, List<GoId>> createDomainIdToGoIdMap( final List<PfamToGoMapping> pfam_to_go_mappings ) {
619 final Map<DomainId, List<GoId>> domain_id_to_go_ids_map = new HashMap<DomainId, List<GoId>>( pfam_to_go_mappings
621 for( final PfamToGoMapping pfam_to_go : pfam_to_go_mappings ) {
622 if ( !domain_id_to_go_ids_map.containsKey( pfam_to_go.getKey() ) ) {
623 domain_id_to_go_ids_map.put( pfam_to_go.getKey(), new ArrayList<GoId>() );
625 domain_id_to_go_ids_map.get( pfam_to_go.getKey() ).add( pfam_to_go.getValue() );
627 return domain_id_to_go_ids_map;
630 public static Map<DomainId, Set<String>> createDomainIdToSecondaryFeaturesMap( final File secondary_features_map_file )
632 final BasicTable<String> primary_table = BasicTableParser.parse( secondary_features_map_file, "\t" );
633 final Map<DomainId, Set<String>> map = new TreeMap<DomainId, Set<String>>();
634 for( int r = 0; r < primary_table.getNumberOfRows(); ++r ) {
635 final DomainId domain_id = new DomainId( primary_table.getValue( 0, r ) );
636 if ( !map.containsKey( domain_id ) ) {
637 map.put( domain_id, new HashSet<String>() );
639 map.get( domain_id ).add( primary_table.getValue( 1, r ) );
644 public static Phylogeny createNjTreeBasedOnMatrixToFile( final File nj_tree_outfile, final DistanceMatrix distance ) {
645 checkForOutputFileWriteability( nj_tree_outfile );
646 final NeighborJoining nj = NeighborJoining.createInstance();
647 final Phylogeny phylogeny = nj.execute( ( BasicSymmetricalDistanceMatrix ) distance );
648 phylogeny.setName( nj_tree_outfile.getName() );
649 writePhylogenyToFile( phylogeny, nj_tree_outfile.toString() );
653 private static SortedSet<BinaryDomainCombination> createSetOfAllBinaryDomainCombinationsPerGenome( final GenomeWideCombinableDomains gwcd ) {
654 final SortedMap<DomainId, CombinableDomains> cds = gwcd.getAllCombinableDomainsIds();
655 final SortedSet<BinaryDomainCombination> binary_combinations = new TreeSet<BinaryDomainCombination>();
656 for( final DomainId domain_id : cds.keySet() ) {
657 final CombinableDomains cd = cds.get( domain_id );
658 binary_combinations.addAll( cd.toBinaryDomainCombinations() );
660 return binary_combinations;
663 public static void decoratePrintableDomainSimilarities( final SortedSet<DomainSimilarity> domain_similarities,
664 final Detailedness detailedness,
665 final GoAnnotationOutput go_annotation_output,
666 final Map<GoId, GoTerm> go_id_to_term_map,
667 final GoNameSpace go_namespace_limit ) {
668 if ( ( go_namespace_limit != null ) && ( ( go_id_to_term_map == null ) || go_id_to_term_map.isEmpty() ) ) {
669 throw new IllegalArgumentException( "attempt to use a GO namespace limit without a GO id to term map" );
671 for( final DomainSimilarity domain_similarity : domain_similarities ) {
672 if ( domain_similarity instanceof PrintableDomainSimilarity ) {
673 final PrintableDomainSimilarity printable_domain_similarity = ( PrintableDomainSimilarity ) domain_similarity;
674 printable_domain_similarity.setDetailedness( detailedness );
675 printable_domain_similarity.setGoAnnotationOutput( go_annotation_output );
676 printable_domain_similarity.setGoIdToTermMap( go_id_to_term_map );
677 printable_domain_similarity.setGoNamespaceLimit( go_namespace_limit );
682 public static void executeDomainLengthAnalysis( final String[][] input_file_properties,
683 final int number_of_genomes,
684 final DomainLengthsTable domain_lengths_table,
685 final File outfile ) throws IOException {
686 final DecimalFormat df = new DecimalFormat( "#.00" );
687 checkForOutputFileWriteability( outfile );
688 final BufferedWriter out = new BufferedWriter( new FileWriter( outfile ) );
689 out.write( "MEAN BASED STATISTICS PER SPECIES" );
690 out.write( ForesterUtil.LINE_SEPARATOR );
691 out.write( domain_lengths_table.createMeanBasedStatisticsPerSpeciesTable().toString() );
692 out.write( ForesterUtil.LINE_SEPARATOR );
693 out.write( ForesterUtil.LINE_SEPARATOR );
694 final List<DomainLengths> domain_lengths_list = domain_lengths_table.getDomainLengthsList();
695 out.write( "OUTLIER SPECIES PER DOMAIN (Z>=1.5)" );
696 out.write( ForesterUtil.LINE_SEPARATOR );
697 for( final DomainLengths domain_lengths : domain_lengths_list ) {
698 final List<Species> species_list = domain_lengths.getMeanBasedOutlierSpecies( 1.5 );
699 if ( species_list.size() > 0 ) {
700 out.write( domain_lengths.getDomainId() + "\t" );
701 for( final Species species : species_list ) {
702 out.write( species + "\t" );
704 out.write( ForesterUtil.LINE_SEPARATOR );
705 // DescriptiveStatistics stats_for_domain = domain_lengths
706 // .calculateMeanBasedStatistics();
707 //AsciiHistogram histo = new AsciiHistogram( stats_for_domain );
708 //System.out.println( histo.toStringBuffer( 40, '=', 60, 4 ).toString() );
711 out.write( ForesterUtil.LINE_SEPARATOR );
712 out.write( ForesterUtil.LINE_SEPARATOR );
713 out.write( "OUTLIER SPECIES (Z 1.0)" );
714 out.write( ForesterUtil.LINE_SEPARATOR );
715 final DescriptiveStatistics stats_for_all_species = domain_lengths_table
716 .calculateMeanBasedStatisticsForAllSpecies();
717 out.write( stats_for_all_species.asSummary() );
718 out.write( ForesterUtil.LINE_SEPARATOR );
719 final AsciiHistogram histo = new AsciiHistogram( stats_for_all_species );
720 out.write( histo.toStringBuffer( 40, '=', 60, 4 ).toString() );
721 out.write( ForesterUtil.LINE_SEPARATOR );
722 final double population_sd = stats_for_all_species.sampleStandardDeviation();
723 final double population_mean = stats_for_all_species.arithmeticMean();
724 for( final Species species : domain_lengths_table.getSpecies() ) {
725 final double x = domain_lengths_table.calculateMeanBasedStatisticsForSpecies( species ).arithmeticMean();
726 final double z = ( x - population_mean ) / population_sd;
727 out.write( species + "\t" + z );
728 out.write( ForesterUtil.LINE_SEPARATOR );
730 out.write( ForesterUtil.LINE_SEPARATOR );
731 for( final Species species : domain_lengths_table.getSpecies() ) {
732 final DescriptiveStatistics stats_for_species = domain_lengths_table
733 .calculateMeanBasedStatisticsForSpecies( species );
734 final double x = stats_for_species.arithmeticMean();
735 final double z = ( x - population_mean ) / population_sd;
736 if ( ( z <= -1.0 ) || ( z >= 1.0 ) ) {
737 out.write( species + "\t" + df.format( z ) + "\t" + stats_for_species.asSummary() );
738 out.write( ForesterUtil.LINE_SEPARATOR );
742 // final List<HistogramData> histogram_datas = new ArrayList<HistogramData>();
743 // for( int i = 0; i < number_of_genomes; ++i ) {
744 // final Species species = new BasicSpecies( input_file_properties[ i ][ 0 ] );
746 // .add( new HistogramData( species.toString(), domain_lengths_table
747 // .calculateMeanBasedStatisticsForSpecies( species )
748 // .getDataAsDoubleArray(), 5, 600, null, 60 ) );
750 // final HistogramsFrame hf = new HistogramsFrame( histogram_datas );
751 // hf.setVisible( true );
757 * @param all_binary_domains_combination_lost_fitch
758 * @param consider_directedness_and_adjacency_for_bin_combinations
759 * @param all_binary_domains_combination_gained if null ignored, otherwise this is to list all binary domain combinations
760 * which were gained under unweighted (Fitch) parsimony.
762 public static void executeParsimonyAnalysis( final long random_number_seed_for_fitch_parsimony,
763 final boolean radomize_fitch_parsimony,
764 final String outfile_name,
765 final DomainParsimonyCalculator domain_parsimony,
766 final Phylogeny phylogeny,
767 final Map<DomainId, List<GoId>> domain_id_to_go_ids_map,
768 final Map<GoId, GoTerm> go_id_to_term_map,
769 final GoNameSpace go_namespace_limit,
770 final String parameters_str,
771 final Map<DomainId, Set<String>>[] domain_id_to_secondary_features_maps,
772 final SortedSet<DomainId> positive_filter,
773 final boolean output_binary_domain_combinations_for_graphs,
774 final List<BinaryDomainCombination> all_binary_domains_combination_gained_fitch,
775 final List<BinaryDomainCombination> all_binary_domains_combination_lost_fitch,
776 final BinaryDomainCombination.DomainCombinationType dc_type,
777 final Map<String, DescriptiveStatistics> protein_length_stats_by_dc,
778 final Map<String, DescriptiveStatistics> domain_number_stats_by_dc,
779 final Map<String, DescriptiveStatistics> domain_length_stats_by_domain ) {
780 final String sep = ForesterUtil.LINE_SEPARATOR + "###################" + ForesterUtil.LINE_SEPARATOR;
781 final String date_time = ForesterUtil.getCurrentDateTime();
782 final SortedSet<String> all_pfams_encountered = new TreeSet<String>();
783 final SortedSet<String> all_pfams_gained_as_domains = new TreeSet<String>();
784 final SortedSet<String> all_pfams_lost_as_domains = new TreeSet<String>();
785 final SortedSet<String> all_pfams_gained_as_dom_combinations = new TreeSet<String>();
786 final SortedSet<String> all_pfams_lost_as_dom_combinations = new TreeSet<String>();
787 writeToNexus( outfile_name, domain_parsimony, phylogeny );
790 Phylogeny local_phylogeny_l = phylogeny.copy();
791 if ( ( positive_filter != null ) && ( positive_filter.size() > 0 ) ) {
792 domain_parsimony.executeDolloParsimonyOnDomainPresence( positive_filter );
795 domain_parsimony.executeDolloParsimonyOnDomainPresence();
797 SurfacingUtil.writeMatrixToFile( domain_parsimony.getGainLossMatrix(), outfile_name
798 + surfacing.PARSIMONY_OUTPUT_GL_SUFFIX_DOLLO_DOMAINS, Format.FORESTER );
799 SurfacingUtil.writeMatrixToFile( domain_parsimony.getGainLossCountsMatrix(), outfile_name
800 + surfacing.PARSIMONY_OUTPUT_GL_COUNTS_SUFFIX_DOLLO_DOMAINS, Format.FORESTER );
801 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
802 CharacterStateMatrix.GainLossStates.GAIN,
803 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_GAINS_D,
805 ForesterUtil.LINE_SEPARATOR,
807 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
808 CharacterStateMatrix.GainLossStates.LOSS,
809 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_LOSSES_D,
811 ForesterUtil.LINE_SEPARATOR,
813 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(), null, outfile_name
814 + surfacing.PARSIMONY_OUTPUT_DOLLO_PRESENT_D, sep, ForesterUtil.LINE_SEPARATOR, null );
816 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
820 domain_parsimony.getGainLossMatrix(),
821 CharacterStateMatrix.GainLossStates.GAIN,
822 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_GAINS_HTML_D,
824 ForesterUtil.LINE_SEPARATOR,
825 "Dollo Parsimony | Gains | Domains",
827 domain_id_to_secondary_features_maps,
828 all_pfams_encountered,
829 all_pfams_gained_as_domains,
831 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
835 domain_parsimony.getGainLossMatrix(),
836 CharacterStateMatrix.GainLossStates.LOSS,
837 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_LOSSES_HTML_D,
839 ForesterUtil.LINE_SEPARATOR,
840 "Dollo Parsimony | Losses | Domains",
842 domain_id_to_secondary_features_maps,
843 all_pfams_encountered,
844 all_pfams_lost_as_domains,
846 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
850 domain_parsimony.getGainLossMatrix(),
852 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_PRESENT_HTML_D,
854 ForesterUtil.LINE_SEPARATOR,
855 "Dollo Parsimony | Present | Domains",
857 domain_id_to_secondary_features_maps,
858 all_pfams_encountered,
860 "_dollo_present_d" );
861 preparePhylogeny( local_phylogeny_l,
864 "Dollo parsimony on domain presence/absence",
865 "dollo_on_domains_" + outfile_name,
867 SurfacingUtil.writePhylogenyToFile( local_phylogeny_l, outfile_name
868 + surfacing.DOMAINS_PARSIMONY_TREE_OUTPUT_SUFFIX_DOLLO );
870 writeAllDomainsChangedOnAllSubtrees( local_phylogeny_l, true, outfile_name, "_dollo_all_gains_d" );
871 writeAllDomainsChangedOnAllSubtrees( local_phylogeny_l, false, outfile_name, "_dollo_all_losses_d" );
873 catch ( final IOException e ) {
875 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getLocalizedMessage() );
877 if ( domain_parsimony.calculateNumberOfBinaryDomainCombination() > 0 ) {
878 // FITCH DOMAIN COMBINATIONS
879 // -------------------------
880 local_phylogeny_l = phylogeny.copy();
881 String randomization = "no";
882 if ( radomize_fitch_parsimony ) {
883 domain_parsimony.executeFitchParsimonyOnBinaryDomainCombintion( random_number_seed_for_fitch_parsimony );
884 randomization = "yes, seed = " + random_number_seed_for_fitch_parsimony;
887 domain_parsimony.executeFitchParsimonyOnBinaryDomainCombintion( USE_LAST );
889 SurfacingUtil.writeMatrixToFile( domain_parsimony.getGainLossMatrix(), outfile_name
890 + surfacing.PARSIMONY_OUTPUT_GL_SUFFIX_FITCH_BINARY_COMBINATIONS, Format.FORESTER );
891 SurfacingUtil.writeMatrixToFile( domain_parsimony.getGainLossCountsMatrix(), outfile_name
892 + surfacing.PARSIMONY_OUTPUT_GL_COUNTS_SUFFIX_FITCH_BINARY_COMBINATIONS, Format.FORESTER );
894 .writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
895 CharacterStateMatrix.GainLossStates.GAIN,
896 outfile_name + surfacing.PARSIMONY_OUTPUT_FITCH_GAINS_BC,
898 ForesterUtil.LINE_SEPARATOR,
900 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
901 CharacterStateMatrix.GainLossStates.LOSS,
903 + surfacing.PARSIMONY_OUTPUT_FITCH_LOSSES_BC,
905 ForesterUtil.LINE_SEPARATOR,
907 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(), null, outfile_name
908 + surfacing.PARSIMONY_OUTPUT_FITCH_PRESENT_BC, sep, ForesterUtil.LINE_SEPARATOR, null );
909 if ( all_binary_domains_combination_gained_fitch != null ) {
910 collectChangedDomainCombinationsFromBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
912 all_binary_domains_combination_gained_fitch,
915 if ( all_binary_domains_combination_lost_fitch != null ) {
916 collectChangedDomainCombinationsFromBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
918 all_binary_domains_combination_lost_fitch,
921 if ( output_binary_domain_combinations_for_graphs ) {
923 .writeBinaryStatesMatrixAsListToFileForBinaryCombinationsForGraphAnalysis( domain_parsimony
924 .getGainLossMatrix(),
927 + surfacing.PARSIMONY_OUTPUT_FITCH_PRESENT_BC_OUTPUTFILE_SUFFIX_FOR_GRAPH_ANALYSIS,
929 ForesterUtil.LINE_SEPARATOR,
930 BinaryDomainCombination.OutputFormat.DOT );
933 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
937 domain_parsimony.getGainLossMatrix(),
938 CharacterStateMatrix.GainLossStates.GAIN,
939 outfile_name + surfacing.PARSIMONY_OUTPUT_FITCH_GAINS_HTML_BC,
941 ForesterUtil.LINE_SEPARATOR,
942 "Fitch Parsimony | Gains | Domain Combinations",
945 all_pfams_encountered,
946 all_pfams_gained_as_dom_combinations,
948 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
952 domain_parsimony.getGainLossMatrix(),
953 CharacterStateMatrix.GainLossStates.LOSS,
954 outfile_name + surfacing.PARSIMONY_OUTPUT_FITCH_LOSSES_HTML_BC,
956 ForesterUtil.LINE_SEPARATOR,
957 "Fitch Parsimony | Losses | Domain Combinations",
960 all_pfams_encountered,
961 all_pfams_lost_as_dom_combinations,
962 "_fitch_losses_dc" );
963 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
967 domain_parsimony.getGainLossMatrix(),
969 outfile_name + surfacing.PARSIMONY_OUTPUT_FITCH_PRESENT_HTML_BC,
971 ForesterUtil.LINE_SEPARATOR,
972 "Fitch Parsimony | Present | Domain Combinations",
975 all_pfams_encountered,
977 "_fitch_present_dc" );
978 writeAllEncounteredPfamsToFile( domain_id_to_go_ids_map,
981 all_pfams_encountered );
982 writePfamsToFile( outfile_name + surfacing.ALL_PFAMS_GAINED_AS_DOMAINS_SUFFIX, all_pfams_gained_as_domains );
983 writePfamsToFile( outfile_name + surfacing.ALL_PFAMS_LOST_AS_DOMAINS_SUFFIX, all_pfams_lost_as_domains );
984 writePfamsToFile( outfile_name + surfacing.ALL_PFAMS_GAINED_AS_DC_SUFFIX,
985 all_pfams_gained_as_dom_combinations );
986 writePfamsToFile( outfile_name + surfacing.ALL_PFAMS_LOST_AS_DC_SUFFIX, all_pfams_lost_as_dom_combinations );
987 preparePhylogeny( local_phylogeny_l,
990 "Fitch parsimony on binary domain combination presence/absence randomization: "
992 "fitch_on_binary_domain_combinations_" + outfile_name,
994 SurfacingUtil.writePhylogenyToFile( local_phylogeny_l, outfile_name
995 + surfacing.BINARY_DOMAIN_COMBINATIONS_PARSIMONY_TREE_OUTPUT_SUFFIX_FITCH );
996 calculateIndependentDomainCombinationGains( local_phylogeny_l,
998 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_COUNTS_OUTPUT_SUFFIX,
1000 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_OUTPUT_SUFFIX,
1002 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_FOR_GO_MAPPING_OUTPUT_SUFFIX,
1004 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_FOR_GO_MAPPING_OUTPUT_UNIQUE_SUFFIX,
1005 outfile_name + "_indep_dc_gains_fitch_lca_ranks.txt",
1006 outfile_name + "_indep_dc_gains_fitch_lca_taxonomies.txt",
1007 outfile_name + "_indep_dc_gains_fitch_protein_statistics.txt",
1008 protein_length_stats_by_dc,
1009 domain_number_stats_by_dc,
1010 domain_length_stats_by_domain );
1014 public static void executeParsimonyAnalysisForSecondaryFeatures( final String outfile_name,
1015 final DomainParsimonyCalculator secondary_features_parsimony,
1016 final Phylogeny phylogeny,
1017 final String parameters_str,
1018 final Map<Species, MappingResults> mapping_results_map ) {
1019 final String sep = ForesterUtil.LINE_SEPARATOR + "###################" + ForesterUtil.LINE_SEPARATOR;
1020 final String date_time = ForesterUtil.getCurrentDateTime();
1021 System.out.println();
1022 writeToNexus( outfile_name + surfacing.NEXUS_SECONDARY_FEATURES,
1023 secondary_features_parsimony.createMatrixOfSecondaryFeaturePresenceOrAbsence( null ),
1025 Phylogeny local_phylogeny_copy = phylogeny.copy();
1026 secondary_features_parsimony.executeDolloParsimonyOnSecondaryFeatures( mapping_results_map );
1027 SurfacingUtil.writeMatrixToFile( secondary_features_parsimony.getGainLossMatrix(), outfile_name
1028 + surfacing.PARSIMONY_OUTPUT_GL_SUFFIX_DOLLO_SECONDARY_FEATURES, Format.FORESTER );
1029 SurfacingUtil.writeMatrixToFile( secondary_features_parsimony.getGainLossCountsMatrix(), outfile_name
1030 + surfacing.PARSIMONY_OUTPUT_GL_COUNTS_SUFFIX_DOLLO_SECONDARY_FEATURES, Format.FORESTER );
1032 .writeBinaryStatesMatrixAsListToFile( secondary_features_parsimony.getGainLossMatrix(),
1033 CharacterStateMatrix.GainLossStates.GAIN,
1035 + surfacing.PARSIMONY_OUTPUT_DOLLO_GAINS_SECONDARY_FEATURES,
1037 ForesterUtil.LINE_SEPARATOR,
1040 .writeBinaryStatesMatrixAsListToFile( secondary_features_parsimony.getGainLossMatrix(),
1041 CharacterStateMatrix.GainLossStates.LOSS,
1043 + surfacing.PARSIMONY_OUTPUT_DOLLO_LOSSES_SECONDARY_FEATURES,
1045 ForesterUtil.LINE_SEPARATOR,
1048 .writeBinaryStatesMatrixAsListToFile( secondary_features_parsimony.getGainLossMatrix(),
1051 + surfacing.PARSIMONY_OUTPUT_DOLLO_PRESENT_SECONDARY_FEATURES,
1053 ForesterUtil.LINE_SEPARATOR,
1055 preparePhylogeny( local_phylogeny_copy,
1056 secondary_features_parsimony,
1058 "Dollo parsimony on secondary feature presence/absence",
1059 "dollo_on_secondary_features_" + outfile_name,
1061 SurfacingUtil.writePhylogenyToFile( local_phylogeny_copy, outfile_name
1062 + surfacing.SECONDARY_FEATURES_PARSIMONY_TREE_OUTPUT_SUFFIX_DOLLO );
1063 // FITCH DOMAIN COMBINATIONS
1064 // -------------------------
1065 local_phylogeny_copy = phylogeny.copy();
1066 final String randomization = "no";
1067 secondary_features_parsimony.executeFitchParsimonyOnBinaryDomainCombintionOnSecondaryFeatures( USE_LAST );
1068 preparePhylogeny( local_phylogeny_copy,
1069 secondary_features_parsimony,
1071 "Fitch parsimony on secondary binary domain combination presence/absence randomization: "
1073 "fitch_on_binary_domain_combinations_" + outfile_name,
1075 SurfacingUtil.writePhylogenyToFile( local_phylogeny_copy, outfile_name
1076 + surfacing.BINARY_DOMAIN_COMBINATIONS_PARSIMONY_TREE_OUTPUT_SUFFIX_FITCH_MAPPED );
1077 calculateIndependentDomainCombinationGains( local_phylogeny_copy, outfile_name
1078 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_COUNTS_MAPPED_OUTPUT_SUFFIX, outfile_name
1079 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_MAPPED_OUTPUT_SUFFIX, outfile_name
1080 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_FOR_GO_MAPPING_MAPPED_OUTPUT_SUFFIX, outfile_name
1081 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_FOR_GO_MAPPING_MAPPED_OUTPUT_UNIQUE_SUFFIX, outfile_name
1082 + "_MAPPED_indep_dc_gains_fitch_lca_ranks.txt", outfile_name
1083 + "_MAPPED_indep_dc_gains_fitch_lca_taxonomies.txt", null, null, null, null );
1086 public static void doit( final List<Protein> proteins,
1087 final List<DomainId> query_domain_ids_nc_order,
1089 final String separator,
1090 final String limit_to_species,
1091 final Map<String, List<Integer>> average_protein_lengths_by_dc ) throws IOException {
1092 for( final Protein protein : proteins ) {
1093 if ( ForesterUtil.isEmpty( limit_to_species )
1094 || protein.getSpecies().getSpeciesId().equalsIgnoreCase( limit_to_species ) ) {
1095 if ( protein.contains( query_domain_ids_nc_order, true ) ) {
1096 out.write( protein.getSpecies().getSpeciesId() );
1097 out.write( separator );
1098 out.write( protein.getProteinId().getId() );
1099 out.write( separator );
1101 final Set<DomainId> visited_domain_ids = new HashSet<DomainId>();
1102 boolean first = true;
1103 for( final Domain domain : protein.getProteinDomains() ) {
1104 if ( !visited_domain_ids.contains( domain.getDomainId() ) ) {
1105 visited_domain_ids.add( domain.getDomainId() );
1112 out.write( domain.getDomainId().getId() );
1114 out.write( "" + domain.getTotalCount() );
1119 out.write( separator );
1120 if ( !( ForesterUtil.isEmpty( protein.getDescription() ) || protein.getDescription()
1121 .equals( SurfacingConstants.NONE ) ) ) {
1122 out.write( protein.getDescription() );
1124 out.write( separator );
1125 if ( !( ForesterUtil.isEmpty( protein.getAccession() ) || protein.getAccession()
1126 .equals( SurfacingConstants.NONE ) ) ) {
1127 out.write( protein.getAccession() );
1129 out.write( SurfacingConstants.NL );
1136 public static void extractProteinNames( final List<Protein> proteins,
1137 final List<DomainId> query_domain_ids_nc_order,
1139 final String separator,
1140 final String limit_to_species ) throws IOException {
1141 for( final Protein protein : proteins ) {
1142 if ( ForesterUtil.isEmpty( limit_to_species )
1143 || protein.getSpecies().getSpeciesId().equalsIgnoreCase( limit_to_species ) ) {
1144 if ( protein.contains( query_domain_ids_nc_order, true ) ) {
1145 out.write( protein.getSpecies().getSpeciesId() );
1146 out.write( separator );
1147 out.write( protein.getProteinId().getId() );
1148 out.write( separator );
1150 final Set<DomainId> visited_domain_ids = new HashSet<DomainId>();
1151 boolean first = true;
1152 for( final Domain domain : protein.getProteinDomains() ) {
1153 if ( !visited_domain_ids.contains( domain.getDomainId() ) ) {
1154 visited_domain_ids.add( domain.getDomainId() );
1161 out.write( domain.getDomainId().getId() );
1163 out.write( "" + domain.getTotalCount() );
1168 out.write( separator );
1169 if ( !( ForesterUtil.isEmpty( protein.getDescription() ) || protein.getDescription()
1170 .equals( SurfacingConstants.NONE ) ) ) {
1171 out.write( protein.getDescription() );
1173 out.write( separator );
1174 if ( !( ForesterUtil.isEmpty( protein.getAccession() ) || protein.getAccession()
1175 .equals( SurfacingConstants.NONE ) ) ) {
1176 out.write( protein.getAccession() );
1178 out.write( SurfacingConstants.NL );
1185 public static void extractProteinNames( final SortedMap<Species, List<Protein>> protein_lists_per_species,
1186 final DomainId domain_id,
1188 final String separator,
1189 final String limit_to_species,
1190 final double domain_e_cutoff ) throws IOException {
1191 System.out.println( "Per domain E-value: " + domain_e_cutoff );
1192 for( final Species species : protein_lists_per_species.keySet() ) {
1193 for( final Protein protein : protein_lists_per_species.get( species ) ) {
1194 if ( ForesterUtil.isEmpty( limit_to_species )
1195 || protein.getSpecies().getSpeciesId().equalsIgnoreCase( limit_to_species ) ) {
1196 final List<Domain> domains = protein.getProteinDomains( domain_id );
1197 if ( domains.size() > 0 ) {
1198 out.write( protein.getSpecies().getSpeciesId() );
1199 out.write( separator );
1200 out.write( protein.getProteinId().getId() );
1201 out.write( separator );
1202 out.write( domain_id.toString() );
1203 out.write( separator );
1204 for( final Domain domain : domains ) {
1205 if ( domain_e_cutoff < 0 || domain.getPerDomainEvalue() <= domain_e_cutoff ) {
1207 out.write( domain.getFrom() + "-" + domain.getTo() );
1211 out.write( separator );
1213 final List<Domain> domain_list = new ArrayList<Domain>();
1215 for( final Domain domain : protein.getProteinDomains() ) {
1216 if ( domain_e_cutoff < 0 || domain.getPerDomainEvalue() <= domain_e_cutoff ) {
1217 domain_list.add( domain );
1221 Domain domain_ary[] = new Domain[ domain_list.size() ];
1223 for( int i = 0; i < domain_list.size(); ++i ) {
1224 domain_ary[ i ] = domain_list.get( i );
1227 Arrays.sort( domain_ary, new DomainComparator( true ) );
1230 boolean first = true;
1232 for( final Domain domain : domain_ary ) {
1239 out.write( domain.getDomainId().toString() );
1240 out.write( ":" + domain.getFrom() + "-" + domain.getTo() );
1241 out.write( ":" + domain.getPerDomainEvalue() );
1244 if ( !( ForesterUtil.isEmpty( protein.getDescription() ) || protein.getDescription()
1245 .equals( SurfacingConstants.NONE ) ) ) {
1246 out.write( protein.getDescription() );
1248 out.write( separator );
1249 if ( !( ForesterUtil.isEmpty( protein.getAccession() ) || protein.getAccession()
1250 .equals( SurfacingConstants.NONE ) ) ) {
1251 out.write( protein.getAccession() );
1253 out.write( SurfacingConstants.NL );
1261 public static SortedSet<DomainId> getAllDomainIds( final List<GenomeWideCombinableDomains> gwcd_list ) {
1262 final SortedSet<DomainId> all_domains_ids = new TreeSet<DomainId>();
1263 for( final GenomeWideCombinableDomains gwcd : gwcd_list ) {
1264 final Set<DomainId> all_domains = gwcd.getAllDomainIds();
1265 // for( final Domain domain : all_domains ) {
1266 all_domains_ids.addAll( all_domains );
1269 return all_domains_ids;
1272 public static SortedMap<String, Integer> getDomainCounts( final List<Protein> protein_domain_collections ) {
1273 final SortedMap<String, Integer> map = new TreeMap<String, Integer>();
1274 for( final Protein protein_domain_collection : protein_domain_collections ) {
1275 for( final Object name : protein_domain_collection.getProteinDomains() ) {
1276 final BasicDomain protein_domain = ( BasicDomain ) name;
1277 final String id = protein_domain.getDomainId().getId();
1278 if ( map.containsKey( id ) ) {
1279 map.put( id, map.get( id ) + 1 );
1289 public static int getNumberOfNodesLackingName( final Phylogeny p, final StringBuilder names ) {
1290 final PhylogenyNodeIterator it = p.iteratorPostorder();
1292 while ( it.hasNext() ) {
1293 final PhylogenyNode n = it.next();
1294 if ( ForesterUtil.isEmpty( n.getName() )
1295 && ( !n.getNodeData().isHasTaxonomy() || ForesterUtil.isEmpty( n.getNodeData().getTaxonomy()
1296 .getScientificName() ) )
1297 && ( !n.getNodeData().isHasTaxonomy() || ForesterUtil.isEmpty( n.getNodeData().getTaxonomy()
1298 .getCommonName() ) ) ) {
1299 if ( n.getParent() != null ) {
1300 names.append( " " );
1301 names.append( n.getParent().getName() );
1303 final List l = n.getAllExternalDescendants();
1304 for( final Object object : l ) {
1305 System.out.println( l.toString() );
1314 * Returns true is Domain domain falls in an uninterrupted stretch of
1315 * covered positions.
1318 * @param covered_positions
1321 public static boolean isEngulfed( final Domain domain, final List<Boolean> covered_positions ) {
1322 for( int i = domain.getFrom(); i <= domain.getTo(); ++i ) {
1323 if ( ( i >= covered_positions.size() ) || ( covered_positions.get( i ) != true ) ) {
1330 public static void preparePhylogeny( final Phylogeny p,
1331 final DomainParsimonyCalculator domain_parsimony,
1332 final String date_time,
1333 final String method,
1335 final String parameters_str ) {
1336 domain_parsimony.decoratePhylogenyWithDomains( p );
1337 final StringBuilder desc = new StringBuilder();
1338 desc.append( "[Method: " + method + "] [Date: " + date_time + "] " );
1339 desc.append( "[Cost: " + domain_parsimony.getCost() + "] " );
1340 desc.append( "[Gains: " + domain_parsimony.getTotalGains() + "] " );
1341 desc.append( "[Losses: " + domain_parsimony.getTotalLosses() + "] " );
1342 desc.append( "[Unchanged: " + domain_parsimony.getTotalUnchanged() + "] " );
1343 desc.append( "[Parameters: " + parameters_str + "]" );
1345 p.setDescription( desc.toString() );
1346 p.setConfidence( new Confidence( domain_parsimony.getCost(), "parsimony" ) );
1347 p.setRerootable( false );
1348 p.setRooted( true );
1352 * species | protein id | n-terminal domain | c-terminal domain | n-terminal domain per domain E-value | c-terminal domain per domain E-value
1356 static public StringBuffer proteinToDomainCombinations( final Protein protein,
1357 final String protein_id,
1358 final String separator ) {
1359 final StringBuffer sb = new StringBuffer();
1360 if ( protein.getSpecies() == null ) {
1361 throw new IllegalArgumentException( "species must not be null" );
1363 if ( ForesterUtil.isEmpty( protein.getSpecies().getSpeciesId() ) ) {
1364 throw new IllegalArgumentException( "species id must not be empty" );
1366 final List<Domain> domains = protein.getProteinDomains();
1367 if ( domains.size() > 1 ) {
1368 final Map<String, Integer> counts = new HashMap<String, Integer>();
1369 for( final Domain domain : domains ) {
1370 final String id = domain.getDomainId().getId();
1371 if ( counts.containsKey( id ) ) {
1372 counts.put( id, counts.get( id ) + 1 );
1375 counts.put( id, 1 );
1378 final Set<String> dcs = new HashSet<String>();
1379 for( int i = 1; i < domains.size(); ++i ) {
1380 for( int j = 0; j < i; ++j ) {
1381 Domain domain_n = domains.get( i );
1382 Domain domain_c = domains.get( j );
1383 if ( domain_n.getFrom() > domain_c.getFrom() ) {
1384 domain_n = domains.get( j );
1385 domain_c = domains.get( i );
1387 final String dc = domain_n.getDomainId().getId() + domain_c.getDomainId().getId();
1388 if ( !dcs.contains( dc ) ) {
1390 sb.append( protein.getSpecies() );
1391 sb.append( separator );
1392 sb.append( protein_id );
1393 sb.append( separator );
1394 sb.append( domain_n.getDomainId().getId() );
1395 sb.append( separator );
1396 sb.append( domain_c.getDomainId().getId() );
1397 sb.append( separator );
1398 sb.append( domain_n.getPerDomainEvalue() );
1399 sb.append( separator );
1400 sb.append( domain_c.getPerDomainEvalue() );
1401 sb.append( separator );
1402 sb.append( counts.get( domain_n.getDomainId().getId() ) );
1403 sb.append( separator );
1404 sb.append( counts.get( domain_c.getDomainId().getId() ) );
1405 sb.append( ForesterUtil.LINE_SEPARATOR );
1410 else if ( domains.size() == 1 ) {
1411 sb.append( protein.getSpecies() );
1412 sb.append( separator );
1413 sb.append( protein_id );
1414 sb.append( separator );
1415 sb.append( domains.get( 0 ).getDomainId().getId() );
1416 sb.append( separator );
1417 sb.append( separator );
1418 sb.append( domains.get( 0 ).getPerDomainEvalue() );
1419 sb.append( separator );
1420 sb.append( separator );
1422 sb.append( separator );
1423 sb.append( ForesterUtil.LINE_SEPARATOR );
1426 sb.append( protein.getSpecies() );
1427 sb.append( separator );
1428 sb.append( protein_id );
1429 sb.append( separator );
1430 sb.append( separator );
1431 sb.append( separator );
1432 sb.append( separator );
1433 sb.append( separator );
1434 sb.append( separator );
1435 sb.append( ForesterUtil.LINE_SEPARATOR );
1442 * Example regarding engulfment: ------------0.1 ----------0.2 --0.3 =>
1443 * domain with 0.3 is ignored
1445 * -----------0.1 ----------0.2 --0.3 => domain with 0.3 is ignored
1448 * ------------0.1 ----------0.3 --0.2 => domains with 0.3 and 0.2 are _not_
1451 * @param max_allowed_overlap
1452 * maximal allowed overlap (inclusive) to be still considered not
1453 * overlapping (zero or negative value to allow any overlap)
1454 * @param remove_engulfed_domains
1455 * to remove domains which are completely engulfed by coverage of
1456 * domains with better support
1460 public static Protein removeOverlappingDomains( final int max_allowed_overlap,
1461 final boolean remove_engulfed_domains,
1462 final Protein protein ) {
1463 final Protein pruned_protein = new BasicProtein( protein.getProteinId().getId(), protein.getSpecies()
1464 .getSpeciesId(), protein.getLength() );
1465 final List<Domain> sorted = SurfacingUtil.sortDomainsWithAscendingConfidenceValues( protein );
1466 final List<Boolean> covered_positions = new ArrayList<Boolean>();
1467 for( final Domain domain : sorted ) {
1468 if ( ( ( max_allowed_overlap < 0 ) || ( SurfacingUtil.calculateOverlap( domain, covered_positions ) <= max_allowed_overlap ) )
1469 && ( !remove_engulfed_domains || !isEngulfed( domain, covered_positions ) ) ) {
1470 final int covered_positions_size = covered_positions.size();
1471 for( int i = covered_positions_size; i < domain.getFrom(); ++i ) {
1472 covered_positions.add( false );
1474 final int new_covered_positions_size = covered_positions.size();
1475 for( int i = domain.getFrom(); i <= domain.getTo(); ++i ) {
1476 if ( i < new_covered_positions_size ) {
1477 covered_positions.set( i, true );
1480 covered_positions.add( true );
1483 pruned_protein.addProteinDomain( domain );
1486 return pruned_protein;
1489 public static List<Domain> sortDomainsWithAscendingConfidenceValues( final Protein protein ) {
1490 final List<Domain> domains = new ArrayList<Domain>();
1491 for( final Domain d : protein.getProteinDomains() ) {
1494 Collections.sort( domains, SurfacingUtil.ASCENDING_CONFIDENCE_VALUE_ORDER );
1498 private static List<String> splitDomainCombination( final String dc ) {
1499 final String[] s = dc.split( "=" );
1500 if ( s.length != 2 ) {
1501 ForesterUtil.printErrorMessage( surfacing.PRG_NAME, "Stringyfied domain combination has illegal format: "
1505 final List<String> l = new ArrayList<String>( 2 );
1511 public static void writeAllDomainsChangedOnAllSubtrees( final Phylogeny p,
1512 final boolean get_gains,
1513 final String outdir,
1514 final String suffix_for_filename ) throws IOException {
1515 CharacterStateMatrix.GainLossStates state = CharacterStateMatrix.GainLossStates.GAIN;
1517 state = CharacterStateMatrix.GainLossStates.LOSS;
1519 final File base_dir = createBaseDirForPerNodeDomainFiles( surfacing.BASE_DIRECTORY_PER_SUBTREE_DOMAIN_GAIN_LOSS_FILES,
1523 for( final PhylogenyNodeIterator it = p.iteratorPostorder(); it.hasNext(); ) {
1524 final PhylogenyNode node = it.next();
1525 if ( !node.isExternal() ) {
1526 final SortedSet<String> domains = collectAllDomainsChangedOnSubtree( node, get_gains );
1527 if ( domains.size() > 0 ) {
1528 final Writer writer = ForesterUtil.createBufferedWriter( base_dir + ForesterUtil.FILE_SEPARATOR
1529 + node.getName() + suffix_for_filename );
1530 for( final String domain : domains ) {
1531 writer.write( domain );
1532 writer.write( ForesterUtil.LINE_SEPARATOR );
1540 private static void writeAllEncounteredPfamsToFile( final Map<DomainId, List<GoId>> domain_id_to_go_ids_map,
1541 final Map<GoId, GoTerm> go_id_to_term_map,
1542 final String outfile_name,
1543 final SortedSet<String> all_pfams_encountered ) {
1544 final File all_pfams_encountered_file = new File( outfile_name + surfacing.ALL_PFAMS_ENCOUNTERED_SUFFIX );
1545 final File all_pfams_encountered_with_go_annotation_file = new File( outfile_name
1546 + surfacing.ALL_PFAMS_ENCOUNTERED_WITH_GO_ANNOTATION_SUFFIX );
1547 final File encountered_pfams_summary_file = new File( outfile_name + surfacing.ENCOUNTERED_PFAMS_SUMMARY_SUFFIX );
1548 int biological_process_counter = 0;
1549 int cellular_component_counter = 0;
1550 int molecular_function_counter = 0;
1551 int pfams_with_mappings_counter = 0;
1552 int pfams_without_mappings_counter = 0;
1553 int pfams_without_mappings_to_bp_or_mf_counter = 0;
1554 int pfams_with_mappings_to_bp_or_mf_counter = 0;
1556 final Writer all_pfams_encountered_writer = new BufferedWriter( new FileWriter( all_pfams_encountered_file ) );
1557 final Writer all_pfams_encountered_with_go_annotation_writer = new BufferedWriter( new FileWriter( all_pfams_encountered_with_go_annotation_file ) );
1558 final Writer summary_writer = new BufferedWriter( new FileWriter( encountered_pfams_summary_file ) );
1559 summary_writer.write( "# Pfam to GO mapping summary" );
1560 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1561 summary_writer.write( "# Actual summary is at the end of this file." );
1562 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1563 summary_writer.write( "# Encountered Pfams without a GO mapping:" );
1564 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1565 for( final String pfam : all_pfams_encountered ) {
1566 all_pfams_encountered_writer.write( pfam );
1567 all_pfams_encountered_writer.write( ForesterUtil.LINE_SEPARATOR );
1568 final DomainId domain_id = new DomainId( pfam );
1569 if ( domain_id_to_go_ids_map.containsKey( domain_id ) ) {
1570 ++pfams_with_mappings_counter;
1571 all_pfams_encountered_with_go_annotation_writer.write( pfam );
1572 all_pfams_encountered_with_go_annotation_writer.write( ForesterUtil.LINE_SEPARATOR );
1573 final List<GoId> go_ids = domain_id_to_go_ids_map.get( domain_id );
1574 boolean maps_to_bp = false;
1575 boolean maps_to_cc = false;
1576 boolean maps_to_mf = false;
1577 for( final GoId go_id : go_ids ) {
1578 final GoTerm go_term = go_id_to_term_map.get( go_id );
1579 if ( go_term.getGoNameSpace().isBiologicalProcess() ) {
1582 else if ( go_term.getGoNameSpace().isCellularComponent() ) {
1585 else if ( go_term.getGoNameSpace().isMolecularFunction() ) {
1590 ++biological_process_counter;
1593 ++cellular_component_counter;
1596 ++molecular_function_counter;
1598 if ( maps_to_bp || maps_to_mf ) {
1599 ++pfams_with_mappings_to_bp_or_mf_counter;
1602 ++pfams_without_mappings_to_bp_or_mf_counter;
1606 ++pfams_without_mappings_to_bp_or_mf_counter;
1607 ++pfams_without_mappings_counter;
1608 summary_writer.write( pfam );
1609 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1612 all_pfams_encountered_writer.close();
1613 all_pfams_encountered_with_go_annotation_writer.close();
1614 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote all [" + all_pfams_encountered.size()
1615 + "] encountered Pfams to: \"" + all_pfams_encountered_file + "\"" );
1616 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote all [" + pfams_with_mappings_counter
1617 + "] encountered Pfams with GO mappings to: \"" + all_pfams_encountered_with_go_annotation_file
1619 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote summary (including all ["
1620 + pfams_without_mappings_counter + "] encountered Pfams without GO mappings) to: \""
1621 + encountered_pfams_summary_file + "\"" );
1622 ForesterUtil.programMessage( surfacing.PRG_NAME, "Sum of Pfams encountered : "
1623 + all_pfams_encountered.size() );
1624 ForesterUtil.programMessage( surfacing.PRG_NAME, "Pfams without a mapping : "
1625 + pfams_without_mappings_counter + " ["
1626 + ( 100 * pfams_without_mappings_counter / all_pfams_encountered.size() ) + "%]" );
1627 ForesterUtil.programMessage( surfacing.PRG_NAME, "Pfams without mapping to proc. or func. : "
1628 + pfams_without_mappings_to_bp_or_mf_counter + " ["
1629 + ( 100 * pfams_without_mappings_to_bp_or_mf_counter / all_pfams_encountered.size() ) + "%]" );
1630 ForesterUtil.programMessage( surfacing.PRG_NAME,
1631 "Pfams with a mapping : " + pfams_with_mappings_counter
1633 + ( 100 * pfams_with_mappings_counter / all_pfams_encountered.size() )
1635 ForesterUtil.programMessage( surfacing.PRG_NAME, "Pfams with a mapping to proc. or func. : "
1636 + pfams_with_mappings_to_bp_or_mf_counter + " ["
1637 + ( 100 * pfams_with_mappings_to_bp_or_mf_counter / all_pfams_encountered.size() ) + "%]" );
1638 ForesterUtil.programMessage( surfacing.PRG_NAME,
1639 "Pfams with mapping to biological process: " + biological_process_counter
1641 + ( 100 * biological_process_counter / all_pfams_encountered.size() )
1643 ForesterUtil.programMessage( surfacing.PRG_NAME,
1644 "Pfams with mapping to molecular function: " + molecular_function_counter
1646 + ( 100 * molecular_function_counter / all_pfams_encountered.size() )
1648 ForesterUtil.programMessage( surfacing.PRG_NAME,
1649 "Pfams with mapping to cellular component: " + cellular_component_counter
1651 + ( 100 * cellular_component_counter / all_pfams_encountered.size() )
1653 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1654 summary_writer.write( "# Sum of Pfams encountered : " + all_pfams_encountered.size() );
1655 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1656 summary_writer.write( "# Pfams without a mapping : " + pfams_without_mappings_counter
1657 + " [" + ( 100 * pfams_without_mappings_counter / all_pfams_encountered.size() ) + "%]" );
1658 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1659 summary_writer.write( "# Pfams without mapping to proc. or func. : "
1660 + pfams_without_mappings_to_bp_or_mf_counter + " ["
1661 + ( 100 * pfams_without_mappings_to_bp_or_mf_counter / all_pfams_encountered.size() ) + "%]" );
1662 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1663 summary_writer.write( "# Pfams with a mapping : " + pfams_with_mappings_counter + " ["
1664 + ( 100 * pfams_with_mappings_counter / all_pfams_encountered.size() ) + "%]" );
1665 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1666 summary_writer.write( "# Pfams with a mapping to proc. or func. : "
1667 + pfams_with_mappings_to_bp_or_mf_counter + " ["
1668 + ( 100 * pfams_with_mappings_to_bp_or_mf_counter / all_pfams_encountered.size() ) + "%]" );
1669 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1670 summary_writer.write( "# Pfams with mapping to biological process: " + biological_process_counter + " ["
1671 + ( 100 * biological_process_counter / all_pfams_encountered.size() ) + "%]" );
1672 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1673 summary_writer.write( "# Pfams with mapping to molecular function: " + molecular_function_counter + " ["
1674 + ( 100 * molecular_function_counter / all_pfams_encountered.size() ) + "%]" );
1675 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1676 summary_writer.write( "# Pfams with mapping to cellular component: " + cellular_component_counter + " ["
1677 + ( 100 * cellular_component_counter / all_pfams_encountered.size() ) + "%]" );
1678 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1679 summary_writer.close();
1681 catch ( final IOException e ) {
1682 ForesterUtil.printWarningMessage( surfacing.PRG_NAME, "Failure to write: " + e );
1686 public static void writeBinaryDomainCombinationsFileForGraphAnalysis( final String[][] input_file_properties,
1687 final File output_dir,
1688 final GenomeWideCombinableDomains gwcd,
1690 final GenomeWideCombinableDomainsSortOrder dc_sort_order ) {
1691 File dc_outfile_dot = new File( input_file_properties[ i ][ 0 ]
1692 + surfacing.DOMAIN_COMBINITONS_OUTPUTFILE_SUFFIX_FOR_GRAPH_ANALYSIS );
1693 if ( output_dir != null ) {
1694 dc_outfile_dot = new File( output_dir + ForesterUtil.FILE_SEPARATOR + dc_outfile_dot );
1696 checkForOutputFileWriteability( dc_outfile_dot );
1697 final SortedSet<BinaryDomainCombination> binary_combinations = createSetOfAllBinaryDomainCombinationsPerGenome( gwcd );
1699 final BufferedWriter out_dot = new BufferedWriter( new FileWriter( dc_outfile_dot ) );
1700 for( final BinaryDomainCombination bdc : binary_combinations ) {
1701 out_dot.write( bdc.toGraphDescribingLanguage( BinaryDomainCombination.OutputFormat.DOT, null, null )
1703 out_dot.write( SurfacingConstants.NL );
1707 catch ( final IOException e ) {
1708 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1710 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote binary domain combination for \""
1711 + input_file_properties[ i ][ 0 ] + "\" (" + input_file_properties[ i ][ 1 ] + ", "
1712 + input_file_properties[ i ][ 2 ] + ") to: \"" + dc_outfile_dot + "\"" );
1715 public static void writeBinaryStatesMatrixAsListToFile( final CharacterStateMatrix<CharacterStateMatrix.GainLossStates> matrix,
1716 final CharacterStateMatrix.GainLossStates state,
1717 final String filename,
1718 final String indentifier_characters_separator,
1719 final String character_separator,
1720 final Map<String, String> descriptions ) {
1721 final File outfile = new File( filename );
1722 checkForOutputFileWriteability( outfile );
1723 final SortedSet<String> sorted_ids = new TreeSet<String>();
1724 for( int i = 0; i < matrix.getNumberOfIdentifiers(); ++i ) {
1725 sorted_ids.add( matrix.getIdentifier( i ) );
1728 final BufferedWriter out = new BufferedWriter( new FileWriter( outfile ) );
1729 for( final String id : sorted_ids ) {
1730 out.write( indentifier_characters_separator );
1731 out.write( "#" + id );
1732 out.write( indentifier_characters_separator );
1733 for( int c = 0; c < matrix.getNumberOfCharacters(); ++c ) {
1735 // using null to indicate either UNCHANGED_PRESENT or GAIN.
1736 if ( ( matrix.getState( id, c ) == state )
1737 || ( ( state == null ) && ( ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.GAIN ) || ( matrix
1738 .getState( id, c ) == CharacterStateMatrix.GainLossStates.UNCHANGED_PRESENT ) ) ) ) {
1739 out.write( matrix.getCharacter( c ) );
1740 if ( ( descriptions != null ) && !descriptions.isEmpty()
1741 && descriptions.containsKey( matrix.getCharacter( c ) ) ) {
1743 out.write( descriptions.get( matrix.getCharacter( c ) ) );
1745 out.write( character_separator );
1752 catch ( final IOException e ) {
1753 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1755 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote characters list: \"" + filename + "\"" );
1758 public static void writeBinaryStatesMatrixAsListToFileForBinaryCombinationsForGraphAnalysis( final CharacterStateMatrix<CharacterStateMatrix.GainLossStates> matrix,
1759 final CharacterStateMatrix.GainLossStates state,
1760 final String filename,
1761 final String indentifier_characters_separator,
1762 final String character_separator,
1763 final BinaryDomainCombination.OutputFormat bc_output_format ) {
1764 final File outfile = new File( filename );
1765 checkForOutputFileWriteability( outfile );
1766 final SortedSet<String> sorted_ids = new TreeSet<String>();
1767 for( int i = 0; i < matrix.getNumberOfIdentifiers(); ++i ) {
1768 sorted_ids.add( matrix.getIdentifier( i ) );
1771 final BufferedWriter out = new BufferedWriter( new FileWriter( outfile ) );
1772 for( final String id : sorted_ids ) {
1773 out.write( indentifier_characters_separator );
1774 out.write( "#" + id );
1775 out.write( indentifier_characters_separator );
1776 for( int c = 0; c < matrix.getNumberOfCharacters(); ++c ) {
1778 // using null to indicate either UNCHANGED_PRESENT or GAIN.
1779 if ( ( matrix.getState( id, c ) == state )
1780 || ( ( state == null ) && ( ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.GAIN ) || ( matrix
1781 .getState( id, c ) == CharacterStateMatrix.GainLossStates.UNCHANGED_PRESENT ) ) ) ) {
1782 BinaryDomainCombination bdc = null;
1784 bdc = BasicBinaryDomainCombination.createInstance( matrix.getCharacter( c ) );
1786 catch ( final Exception e ) {
1787 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getLocalizedMessage() );
1789 out.write( bdc.toGraphDescribingLanguage( bc_output_format, null, null ).toString() );
1790 out.write( character_separator );
1797 catch ( final IOException e ) {
1798 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1800 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote characters list: \"" + filename + "\"" );
1803 public static void writeBinaryStatesMatrixToList( final Map<DomainId, List<GoId>> domain_id_to_go_ids_map,
1804 final Map<GoId, GoTerm> go_id_to_term_map,
1805 final GoNameSpace go_namespace_limit,
1806 final boolean domain_combinations,
1807 final CharacterStateMatrix<CharacterStateMatrix.GainLossStates> matrix,
1808 final CharacterStateMatrix.GainLossStates state,
1809 final String filename,
1810 final String indentifier_characters_separator,
1811 final String character_separator,
1812 final String title_for_html,
1813 final String prefix_for_html,
1814 final Map<DomainId, Set<String>>[] domain_id_to_secondary_features_maps,
1815 final SortedSet<String> all_pfams_encountered,
1816 final SortedSet<String> pfams_gained_or_lost,
1817 final String suffix_for_per_node_events_file ) {
1818 if ( ( go_namespace_limit != null ) && ( ( go_id_to_term_map == null ) || ( go_id_to_term_map.size() < 1 ) ) ) {
1819 throw new IllegalArgumentException( "attempt to use GO namespace limit without a GO-id to term map" );
1821 else if ( ( ( domain_id_to_go_ids_map == null ) || ( domain_id_to_go_ids_map.size() < 1 ) ) ) {
1822 throw new IllegalArgumentException( "attempt to output detailed HTML without a Pfam to GO map" );
1824 else if ( ( ( go_id_to_term_map == null ) || ( go_id_to_term_map.size() < 1 ) ) ) {
1825 throw new IllegalArgumentException( "attempt to output detailed HTML without a GO-id to term map" );
1827 final File outfile = new File( filename );
1828 checkForOutputFileWriteability( outfile );
1829 final SortedSet<String> sorted_ids = new TreeSet<String>();
1830 for( int i = 0; i < matrix.getNumberOfIdentifiers(); ++i ) {
1831 sorted_ids.add( matrix.getIdentifier( i ) );
1834 final Writer out = new BufferedWriter( new FileWriter( outfile ) );
1835 final File per_node_go_mapped_domain_gain_loss_files_base_dir = createBaseDirForPerNodeDomainFiles( surfacing.BASE_DIRECTORY_PER_NODE_DOMAIN_GAIN_LOSS_FILES,
1836 domain_combinations,
1839 Writer per_node_go_mapped_domain_gain_loss_outfile_writer = null;
1840 File per_node_go_mapped_domain_gain_loss_outfile = null;
1841 int per_node_counter = 0;
1842 out.write( "<html>" );
1843 out.write( SurfacingConstants.NL );
1844 addHtmlHead( out, title_for_html );
1845 out.write( SurfacingConstants.NL );
1846 out.write( "<body>" );
1847 out.write( SurfacingConstants.NL );
1848 out.write( "<h1>" );
1849 out.write( SurfacingConstants.NL );
1850 out.write( title_for_html );
1851 out.write( SurfacingConstants.NL );
1852 out.write( "</h1>" );
1853 out.write( SurfacingConstants.NL );
1854 out.write( "<table>" );
1855 out.write( SurfacingConstants.NL );
1856 for( final String id : sorted_ids ) {
1857 final Matcher matcher = PATTERN_SP_STYLE_TAXONOMY.matcher( id );
1858 if ( matcher.matches() ) {
1861 out.write( "<tr>" );
1862 out.write( "<td>" );
1863 out.write( "<a href=\"#" + id + "\">" + id + "</a>" );
1864 out.write( "</td>" );
1865 out.write( "</tr>" );
1866 out.write( SurfacingConstants.NL );
1868 out.write( "</table>" );
1869 out.write( SurfacingConstants.NL );
1870 for( final String id : sorted_ids ) {
1871 final Matcher matcher = PATTERN_SP_STYLE_TAXONOMY.matcher( id );
1872 if ( matcher.matches() ) {
1875 out.write( SurfacingConstants.NL );
1876 out.write( "<h2>" );
1877 out.write( "<a name=\"" + id + "\">" + id + "</a>" );
1878 writeTaxonomyLinks( out, id );
1879 out.write( "</h2>" );
1880 out.write( SurfacingConstants.NL );
1881 out.write( "<table>" );
1882 out.write( SurfacingConstants.NL );
1883 out.write( "<tr>" );
1884 out.write( "<td><b>" );
1885 out.write( "Pfam domain(s)" );
1886 out.write( "</b></td><td><b>" );
1887 out.write( "GO term acc" );
1888 out.write( "</b></td><td><b>" );
1889 out.write( "GO term" );
1890 out.write( "</b></td><td><b>" );
1891 out.write( "GO namespace" );
1892 out.write( "</b></td>" );
1893 out.write( "</tr>" );
1894 out.write( SurfacingConstants.NL );
1895 out.write( "</tr>" );
1896 out.write( SurfacingConstants.NL );
1897 per_node_counter = 0;
1898 if ( matrix.getNumberOfCharacters() > 0 ) {
1899 per_node_go_mapped_domain_gain_loss_outfile = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
1900 + ForesterUtil.FILE_SEPARATOR + id + suffix_for_per_node_events_file );
1901 SurfacingUtil.checkForOutputFileWriteability( per_node_go_mapped_domain_gain_loss_outfile );
1902 per_node_go_mapped_domain_gain_loss_outfile_writer = ForesterUtil
1903 .createBufferedWriter( per_node_go_mapped_domain_gain_loss_outfile );
1906 per_node_go_mapped_domain_gain_loss_outfile = null;
1907 per_node_go_mapped_domain_gain_loss_outfile_writer = null;
1909 for( int c = 0; c < matrix.getNumberOfCharacters(); ++c ) {
1911 // using null to indicate either UNCHANGED_PRESENT or GAIN.
1912 if ( ( matrix.getState( id, c ) == state )
1913 || ( ( state == null ) && ( ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.UNCHANGED_PRESENT ) || ( matrix
1914 .getState( id, c ) == CharacterStateMatrix.GainLossStates.GAIN ) ) ) ) {
1915 final String character = matrix.getCharacter( c );
1916 String domain_0 = "";
1917 String domain_1 = "";
1918 if ( character.indexOf( BinaryDomainCombination.SEPARATOR ) > 0 ) {
1919 final String[] s = character.split( BinaryDomainCombination.SEPARATOR );
1920 if ( s.length != 2 ) {
1921 throw new AssertionError( "this should not have happened: unexpected format for domain combination: ["
1922 + character + "]" );
1928 domain_0 = character;
1930 writeDomainData( domain_id_to_go_ids_map,
1937 character_separator,
1938 domain_id_to_secondary_features_maps,
1940 all_pfams_encountered.add( domain_0 );
1941 if ( pfams_gained_or_lost != null ) {
1942 pfams_gained_or_lost.add( domain_0 );
1944 if ( !ForesterUtil.isEmpty( domain_1 ) ) {
1945 all_pfams_encountered.add( domain_1 );
1946 if ( pfams_gained_or_lost != null ) {
1947 pfams_gained_or_lost.add( domain_1 );
1950 if ( per_node_go_mapped_domain_gain_loss_outfile_writer != null ) {
1951 writeDomainsToIndividualFilePerTreeNode( per_node_go_mapped_domain_gain_loss_outfile_writer,
1958 if ( per_node_go_mapped_domain_gain_loss_outfile_writer != null ) {
1959 per_node_go_mapped_domain_gain_loss_outfile_writer.close();
1960 if ( per_node_counter < 1 ) {
1961 per_node_go_mapped_domain_gain_loss_outfile.delete();
1963 per_node_counter = 0;
1965 out.write( "</table>" );
1966 out.write( SurfacingConstants.NL );
1967 out.write( "<hr>" );
1968 out.write( SurfacingConstants.NL );
1969 } // for( final String id : sorted_ids ) {
1970 out.write( "</body>" );
1971 out.write( SurfacingConstants.NL );
1972 out.write( "</html>" );
1973 out.write( SurfacingConstants.NL );
1977 catch ( final IOException e ) {
1978 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1980 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote characters detailed HTML list: \"" + filename + "\"" );
1983 public static void writeDomainCombinationsCountsFile( final String[][] input_file_properties,
1984 final File output_dir,
1985 final Writer per_genome_domain_promiscuity_statistics_writer,
1986 final GenomeWideCombinableDomains gwcd,
1988 final GenomeWideCombinableDomains.GenomeWideCombinableDomainsSortOrder dc_sort_order ) {
1989 File dc_outfile = new File( input_file_properties[ i ][ 0 ]
1990 + surfacing.DOMAIN_COMBINITON_COUNTS_OUTPUTFILE_SUFFIX );
1991 if ( output_dir != null ) {
1992 dc_outfile = new File( output_dir + ForesterUtil.FILE_SEPARATOR + dc_outfile );
1994 checkForOutputFileWriteability( dc_outfile );
1996 final BufferedWriter out = new BufferedWriter( new FileWriter( dc_outfile ) );
1997 out.write( gwcd.toStringBuilder( dc_sort_order ).toString() );
2000 catch ( final IOException e ) {
2001 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
2003 final DescriptiveStatistics stats = gwcd.getPerGenomeDomainPromiscuityStatistics();
2005 per_genome_domain_promiscuity_statistics_writer.write( input_file_properties[ i ][ 0 ] + "\t" );
2006 per_genome_domain_promiscuity_statistics_writer.write( FORMATTER_3.format( stats.arithmeticMean() ) + "\t" );
2007 if ( stats.getN() < 2 ) {
2008 per_genome_domain_promiscuity_statistics_writer.write( "n/a" + "\t" );
2011 per_genome_domain_promiscuity_statistics_writer.write( FORMATTER_3.format( stats
2012 .sampleStandardDeviation() ) + "\t" );
2014 per_genome_domain_promiscuity_statistics_writer.write( FORMATTER_3.format( stats.median() ) + "\t" );
2015 per_genome_domain_promiscuity_statistics_writer.write( ( int ) stats.getMin() + "\t" );
2016 per_genome_domain_promiscuity_statistics_writer.write( ( int ) stats.getMax() + "\t" );
2017 per_genome_domain_promiscuity_statistics_writer.write( stats.getN() + "\t" );
2018 final SortedSet<DomainId> mpds = gwcd.getMostPromiscuosDomain();
2019 for( final DomainId mpd : mpds ) {
2020 per_genome_domain_promiscuity_statistics_writer.write( mpd.getId() + " " );
2022 per_genome_domain_promiscuity_statistics_writer.write( ForesterUtil.LINE_SEPARATOR );
2024 catch ( final IOException e ) {
2025 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
2027 if ( input_file_properties[ i ].length == 3 ) {
2028 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote domain combination counts for \""
2029 + input_file_properties[ i ][ 0 ] + "\" (" + input_file_properties[ i ][ 1 ] + ", "
2030 + input_file_properties[ i ][ 2 ] + ") to: \"" + dc_outfile + "\"" );
2033 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote domain combination counts for \""
2034 + input_file_properties[ i ][ 0 ] + "\" (" + input_file_properties[ i ][ 1 ] + ") to: \""
2035 + dc_outfile + "\"" );
2039 private static void writeDomainData( final Map<DomainId, List<GoId>> domain_id_to_go_ids_map,
2040 final Map<GoId, GoTerm> go_id_to_term_map,
2041 final GoNameSpace go_namespace_limit,
2043 final String domain_0,
2044 final String domain_1,
2045 final String prefix_for_html,
2046 final String character_separator_for_non_html_output,
2047 final Map<DomainId, Set<String>>[] domain_id_to_secondary_features_maps,
2048 final Set<GoId> all_go_ids ) throws IOException {
2049 boolean any_go_annotation_present = false;
2050 boolean first_has_no_go = false;
2051 int domain_count = 2; // To distinguish between domains and binary domain combinations.
2052 if ( ForesterUtil.isEmpty( domain_1 ) ) {
2055 // The following has a difficult to understand logic.
2056 for( int d = 0; d < domain_count; ++d ) {
2057 List<GoId> go_ids = null;
2058 boolean go_annotation_present = false;
2060 final DomainId domain_id = new DomainId( domain_0 );
2061 if ( domain_id_to_go_ids_map.containsKey( domain_id ) ) {
2062 go_annotation_present = true;
2063 any_go_annotation_present = true;
2064 go_ids = domain_id_to_go_ids_map.get( domain_id );
2067 first_has_no_go = true;
2071 final DomainId domain_id = new DomainId( domain_1 );
2072 if ( domain_id_to_go_ids_map.containsKey( domain_id ) ) {
2073 go_annotation_present = true;
2074 any_go_annotation_present = true;
2075 go_ids = domain_id_to_go_ids_map.get( domain_id );
2078 if ( go_annotation_present ) {
2079 boolean first = ( ( d == 0 ) || ( ( d == 1 ) && first_has_no_go ) );
2080 for( final GoId go_id : go_ids ) {
2081 out.write( "<tr>" );
2084 writeDomainIdsToHtml( out,
2088 domain_id_to_secondary_features_maps );
2091 out.write( "<td></td>" );
2093 if ( !go_id_to_term_map.containsKey( go_id ) ) {
2094 throw new IllegalArgumentException( "GO-id [" + go_id + "] not found in GO-id to GO-term map" );
2096 final GoTerm go_term = go_id_to_term_map.get( go_id );
2097 if ( ( go_namespace_limit == null ) || go_namespace_limit.equals( go_term.getGoNameSpace() ) ) {
2098 // final String top = GoUtils.getPenultimateGoTerm( go_term, go_id_to_term_map ).getName();
2099 final String go_id_str = go_id.getId();
2100 out.write( "<td>" );
2101 out.write( "<a href=\"" + SurfacingConstants.AMIGO_LINK + go_id_str
2102 + "\" target=\"amigo_window\">" + go_id_str + "</a>" );
2103 out.write( "</td><td>" );
2104 out.write( go_term.getName() );
2105 if ( domain_count == 2 ) {
2106 out.write( " (" + d + ")" );
2108 out.write( "</td><td>" );
2109 // out.write( top );
2110 // out.write( "</td><td>" );
2112 out.write( go_term.getGoNameSpace().toShortString() );
2114 out.write( "</td>" );
2115 if ( all_go_ids != null ) {
2116 all_go_ids.add( go_id );
2120 out.write( "<td>" );
2121 out.write( "</td><td>" );
2122 out.write( "</td><td>" );
2123 out.write( "</td><td>" );
2124 out.write( "</td>" );
2126 out.write( "</tr>" );
2127 out.write( SurfacingConstants.NL );
2130 } // for( int d = 0; d < domain_count; ++d )
2131 if ( !any_go_annotation_present ) {
2132 out.write( "<tr>" );
2133 writeDomainIdsToHtml( out, domain_0, domain_1, prefix_for_html, domain_id_to_secondary_features_maps );
2134 out.write( "<td>" );
2135 out.write( "</td><td>" );
2136 out.write( "</td><td>" );
2137 out.write( "</td><td>" );
2138 out.write( "</td>" );
2139 out.write( "</tr>" );
2140 out.write( SurfacingConstants.NL );
2144 private static void writeDomainIdsToHtml( final Writer out,
2145 final String domain_0,
2146 final String domain_1,
2147 final String prefix_for_detailed_html,
2148 final Map<DomainId, Set<String>>[] domain_id_to_secondary_features_maps )
2149 throws IOException {
2150 out.write( "<td>" );
2151 if ( !ForesterUtil.isEmpty( prefix_for_detailed_html ) ) {
2152 out.write( prefix_for_detailed_html );
2155 out.write( "<a href=\"" + SurfacingConstants.PFAM_FAMILY_ID_LINK + domain_0 + "\">" + domain_0 + "</a>" );
2156 out.write( "</td>" );
2159 public static DescriptiveStatistics writeDomainSimilaritiesToFile( final StringBuilder html_desc,
2160 final StringBuilder html_title,
2161 final Writer single_writer,
2162 Map<Character, Writer> split_writers,
2163 final SortedSet<DomainSimilarity> similarities,
2164 final boolean treat_as_binary,
2165 final List<Species> species_order,
2166 final PrintableDomainSimilarity.PRINT_OPTION print_option,
2167 final DomainSimilarity.DomainSimilaritySortField sort_field,
2168 final DomainSimilarity.DomainSimilarityScoring scoring,
2169 final boolean verbose ) throws IOException {
2170 final DescriptiveStatistics stats = new BasicDescriptiveStatistics();
2171 String histogram_title = null;
2172 switch ( sort_field ) {
2173 case ABS_MAX_COUNTS_DIFFERENCE:
2174 if ( treat_as_binary ) {
2175 histogram_title = "absolute counts difference:";
2178 histogram_title = "absolute (maximal) counts difference:";
2181 case MAX_COUNTS_DIFFERENCE:
2182 if ( treat_as_binary ) {
2183 histogram_title = "counts difference:";
2186 histogram_title = "(maximal) counts difference:";
2190 histogram_title = "score mean:";
2193 histogram_title = "score minimum:";
2196 histogram_title = "score maximum:";
2198 case MAX_DIFFERENCE:
2199 if ( treat_as_binary ) {
2200 histogram_title = "difference:";
2203 histogram_title = "(maximal) difference:";
2207 histogram_title = "score mean:";
2210 histogram_title = "score standard deviation:";
2213 histogram_title = "species number:";
2216 throw new AssertionError( "Unknown sort field: " + sort_field );
2218 for( final DomainSimilarity similarity : similarities ) {
2219 switch ( sort_field ) {
2220 case ABS_MAX_COUNTS_DIFFERENCE:
2221 stats.addValue( Math.abs( similarity.getMaximalDifferenceInCounts() ) );
2223 case MAX_COUNTS_DIFFERENCE:
2224 stats.addValue( similarity.getMaximalDifferenceInCounts() );
2227 stats.addValue( similarity.getMeanSimilarityScore() );
2230 stats.addValue( similarity.getMinimalSimilarityScore() );
2233 stats.addValue( similarity.getMaximalSimilarityScore() );
2235 case MAX_DIFFERENCE:
2236 stats.addValue( similarity.getMaximalDifference() );
2239 stats.addValue( similarity.getMeanSimilarityScore() );
2242 stats.addValue( similarity.getStandardDeviationOfSimilarityScore() );
2245 stats.addValue( similarity.getSpecies().size() );
2248 throw new AssertionError( "Unknown sort field: " + sort_field );
2252 // final HistogramData[] hists = new HistogramData[ 1 ];
2255 // List<HistogramDataItem> data_items = new
2256 // ArrayList<HistogramDataItem>();
2257 // double[] values = stats.getDataAsDoubleArray();
2258 // for( int i = 0; i < values.length; i++ ) {
2259 // HistogramDataItem data_item = new BasicHistogramDataItem( "", values[
2261 // data_items.add( data_item );
2265 // HistogramData hd0 = new HistogramData( "name",
2273 // hists[ 0 ] = hd0;
2275 // final HistogramsFrame hf = new HistogramsFrame( hists );
2276 // hf.setVisible( true );
2278 AsciiHistogram histo = null;
2279 if ( stats.getMin() < stats.getMin() ) {
2280 histo = new AsciiHistogram( stats, histogram_title );
2283 if ( histo != null ) {
2284 System.out.println( histo.toStringBuffer( 20, '|', 40, 5 ) );
2286 System.out.println();
2287 System.out.println( "N : " + stats.getN() );
2288 System.out.println( "Min : " + stats.getMin() );
2289 System.out.println( "Max : " + stats.getMax() );
2290 System.out.println( "Mean : " + stats.arithmeticMean() );
2291 if ( stats.getN() > 1 ) {
2292 System.out.println( "SD : " + stats.sampleStandardDeviation() );
2295 System.out.println( "SD : n/a" );
2297 System.out.println( "Median : " + stats.median() );
2298 if ( stats.getN() > 1 ) {
2299 System.out.println( "Pearsonian skewness : " + stats.pearsonianSkewness() );
2302 System.out.println( "Pearsonian skewness : n/a" );
2305 if ( ( single_writer != null ) && ( ( split_writers == null ) || split_writers.isEmpty() ) ) {
2306 split_writers = new HashMap<Character, Writer>();
2307 split_writers.put( '_', single_writer );
2309 switch ( print_option ) {
2310 case SIMPLE_TAB_DELIMITED:
2313 for( final Character key : split_writers.keySet() ) {
2314 final Writer w = split_writers.get( key );
2315 w.write( "<html>" );
2316 w.write( SurfacingConstants.NL );
2318 addHtmlHead( w, "DCs (" + html_title + ") " + key.toString().toUpperCase() );
2321 addHtmlHead( w, "DCs (" + html_title + ")" );
2323 w.write( SurfacingConstants.NL );
2324 w.write( "<body>" );
2325 w.write( SurfacingConstants.NL );
2326 w.write( html_desc.toString() );
2327 w.write( SurfacingConstants.NL );
2330 w.write( SurfacingConstants.NL );
2331 w.write( "<tt><pre>" );
2332 w.write( SurfacingConstants.NL );
2333 if ( histo != null ) {
2334 w.write( histo.toStringBuffer( 20, '|', 40, 5 ).toString() );
2335 w.write( SurfacingConstants.NL );
2337 w.write( "</pre></tt>" );
2338 w.write( SurfacingConstants.NL );
2339 w.write( "<table>" );
2340 w.write( SurfacingConstants.NL );
2341 w.write( "<tr><td>N: </td><td>" + stats.getN() + "</td></tr>" );
2342 w.write( SurfacingConstants.NL );
2343 w.write( "<tr><td>Min: </td><td>" + stats.getMin() + "</td></tr>" );
2344 w.write( SurfacingConstants.NL );
2345 w.write( "<tr><td>Max: </td><td>" + stats.getMax() + "</td></tr>" );
2346 w.write( SurfacingConstants.NL );
2347 w.write( "<tr><td>Mean: </td><td>" + stats.arithmeticMean() + "</td></tr>" );
2348 w.write( SurfacingConstants.NL );
2349 if ( stats.getN() > 1 ) {
2350 w.write( "<tr><td>SD: </td><td>" + stats.sampleStandardDeviation() + "</td></tr>" );
2353 w.write( "<tr><td>SD: </td><td>n/a</td></tr>" );
2355 w.write( SurfacingConstants.NL );
2356 w.write( "<tr><td>Median: </td><td>" + stats.median() + "</td></tr>" );
2357 w.write( SurfacingConstants.NL );
2358 if ( stats.getN() > 1 ) {
2359 w.write( "<tr><td>Pearsonian skewness: </td><td>" + stats.pearsonianSkewness() + "</td></tr>" );
2362 w.write( "<tr><td>Pearsonian skewness: </td><td>n/a</td></tr>" );
2364 w.write( SurfacingConstants.NL );
2365 w.write( "</table>" );
2366 w.write( SurfacingConstants.NL );
2368 w.write( SurfacingConstants.NL );
2370 w.write( SurfacingConstants.NL );
2372 w.write( SurfacingConstants.NL );
2373 w.write( "<table>" );
2374 w.write( SurfacingConstants.NL );
2378 for( final Writer w : split_writers.values() ) {
2379 w.write( SurfacingConstants.NL );
2381 for( final DomainSimilarity similarity : similarities ) {
2382 if ( ( species_order != null ) && !species_order.isEmpty() ) {
2383 ( ( PrintableDomainSimilarity ) similarity ).setSpeciesOrder( species_order );
2385 if ( single_writer != null ) {
2386 single_writer.write( similarity.toStringBuffer( print_option ).toString() );
2389 Writer local_writer = split_writers.get( ( similarity.getDomainId().getId().charAt( 0 ) + "" )
2390 .toLowerCase().charAt( 0 ) );
2391 if ( local_writer == null ) {
2392 local_writer = split_writers.get( '0' );
2394 local_writer.write( similarity.toStringBuffer( print_option ).toString() );
2396 for( final Writer w : split_writers.values() ) {
2397 w.write( SurfacingConstants.NL );
2400 switch ( print_option ) {
2402 for( final Writer w : split_writers.values() ) {
2403 w.write( SurfacingConstants.NL );
2404 w.write( "</table>" );
2405 w.write( SurfacingConstants.NL );
2406 w.write( "</font>" );
2407 w.write( SurfacingConstants.NL );
2408 w.write( "</body>" );
2409 w.write( SurfacingConstants.NL );
2410 w.write( "</html>" );
2411 w.write( SurfacingConstants.NL );
2415 for( final Writer w : split_writers.values() ) {
2421 private static void writeDomainsToIndividualFilePerTreeNode( final Writer individual_files_writer,
2422 final String domain_0,
2423 final String domain_1 ) throws IOException {
2424 individual_files_writer.write( domain_0 );
2425 individual_files_writer.write( ForesterUtil.LINE_SEPARATOR );
2426 if ( !ForesterUtil.isEmpty( domain_1 ) ) {
2427 individual_files_writer.write( domain_1 );
2428 individual_files_writer.write( ForesterUtil.LINE_SEPARATOR );
2432 public static void writeMatrixToFile( final CharacterStateMatrix<?> matrix,
2433 final String filename,
2434 final Format format ) {
2435 final File outfile = new File( filename );
2436 checkForOutputFileWriteability( outfile );
2438 final BufferedWriter out = new BufferedWriter( new FileWriter( outfile ) );
2439 matrix.toWriter( out, format );
2443 catch ( final IOException e ) {
2444 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
2446 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote matrix: \"" + filename + "\"" );
2449 public static void writeMatrixToFile( final File matrix_outfile, final List<DistanceMatrix> matrices ) {
2450 checkForOutputFileWriteability( matrix_outfile );
2452 final BufferedWriter out = new BufferedWriter( new FileWriter( matrix_outfile ) );
2453 for( final DistanceMatrix distance_matrix : matrices ) {
2454 out.write( distance_matrix.toStringBuffer( DistanceMatrix.Format.PHYLIP ).toString() );
2455 out.write( ForesterUtil.LINE_SEPARATOR );
2460 catch ( final IOException e ) {
2461 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
2463 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote distance matrices to \"" + matrix_outfile + "\"" );
2466 private static void writePfamsToFile( final String outfile_name, final SortedSet<String> pfams ) {
2468 final Writer writer = new BufferedWriter( new FileWriter( new File( outfile_name ) ) );
2469 for( final String pfam : pfams ) {
2470 writer.write( pfam );
2471 writer.write( ForesterUtil.LINE_SEPARATOR );
2474 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote " + pfams.size() + " pfams to [" + outfile_name
2477 catch ( final IOException e ) {
2478 ForesterUtil.printWarningMessage( surfacing.PRG_NAME, "Failure to write: " + e );
2482 public static void writePhylogenyToFile( final Phylogeny phylogeny, final String filename ) {
2483 final PhylogenyWriter writer = new PhylogenyWriter();
2485 writer.toPhyloXML( new File( filename ), phylogeny, 1 );
2487 catch ( final IOException e ) {
2488 ForesterUtil.printWarningMessage( surfacing.PRG_NAME, "failed to write phylogeny to \"" + filename + "\": "
2491 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote phylogeny to \"" + filename + "\"" );
2494 public static void writeTaxonomyLinks( final Writer writer, final String species ) throws IOException {
2495 if ( ( species.length() > 1 ) && ( species.indexOf( '_' ) < 1 ) ) {
2496 final Matcher matcher = PATTERN_SP_STYLE_TAXONOMY.matcher( species );
2497 writer.write( " [" );
2498 if ( matcher.matches() ) {
2499 writer.write( "<a href=\"" + SurfacingConstants.UNIPROT_LINK + species
2500 + "\" target=\"taxonomy_window\">uniprot</a>" );
2503 writer.write( "<a href=\"" + SurfacingConstants.EOL_LINK + species
2504 + "\" target=\"taxonomy_window\">eol</a>" );
2505 writer.write( "|" );
2506 writer.write( "<a href=\"" + SurfacingConstants.TOL_LINK + species
2507 + "\" target=\"taxonomy_window\">tol</a>" );
2509 writer.write( "]" );
2513 private static void writeToNexus( final String outfile_name,
2514 final CharacterStateMatrix<BinaryStates> matrix,
2515 final Phylogeny phylogeny ) {
2516 if ( !( matrix instanceof BasicCharacterStateMatrix ) ) {
2517 throw new IllegalArgumentException( "can only write matrices of type [" + BasicCharacterStateMatrix.class
2520 final BasicCharacterStateMatrix<BinaryStates> my_matrix = ( org.forester.evoinference.matrix.character.BasicCharacterStateMatrix<BinaryStates> ) matrix;
2521 final List<Phylogeny> phylogenies = new ArrayList<Phylogeny>( 1 );
2522 phylogenies.add( phylogeny );
2524 final BufferedWriter w = new BufferedWriter( new FileWriter( outfile_name ) );
2525 w.write( NexusConstants.NEXUS );
2526 w.write( ForesterUtil.LINE_SEPARATOR );
2527 my_matrix.writeNexusTaxaBlock( w );
2528 my_matrix.writeNexusBinaryChractersBlock( w );
2529 PhylogenyWriter.writeNexusTreesBlock( w, phylogenies, NH_CONVERSION_SUPPORT_VALUE_STYLE.NONE );
2532 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote Nexus file: \"" + outfile_name + "\"" );
2534 catch ( final IOException e ) {
2535 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
2539 private static void writeToNexus( final String outfile_name,
2540 final DomainParsimonyCalculator domain_parsimony,
2541 final Phylogeny phylogeny ) {
2542 writeToNexus( outfile_name + surfacing.NEXUS_EXTERNAL_DOMAINS,
2543 domain_parsimony.createMatrixOfDomainPresenceOrAbsence(),
2545 writeToNexus( outfile_name + surfacing.NEXUS_EXTERNAL_DOMAIN_COMBINATIONS,
2546 domain_parsimony.createMatrixOfBinaryDomainCombinationPresenceOrAbsence(),
2550 public static void domainsPerProteinsStatistics( final String genome,
2551 final List<Protein> protein_list,
2552 final DescriptiveStatistics all_genomes_domains_per_potein_stats,
2553 final SortedMap<Integer, Integer> all_genomes_domains_per_potein_histo,
2554 final SortedSet<String> domains_which_are_always_single,
2555 final SortedSet<String> domains_which_are_sometimes_single_sometimes_not,
2556 final SortedSet<String> domains_which_never_single,
2557 final Writer writer ) {
2558 final DescriptiveStatistics stats = new BasicDescriptiveStatistics();
2559 for( final Protein protein : protein_list ) {
2560 final int domains = protein.getNumberOfProteinDomains();
2561 //System.out.println( domains );
2562 stats.addValue( domains );
2563 all_genomes_domains_per_potein_stats.addValue( domains );
2564 if ( !all_genomes_domains_per_potein_histo.containsKey( domains ) ) {
2565 all_genomes_domains_per_potein_histo.put( domains, 1 );
2568 all_genomes_domains_per_potein_histo.put( domains,
2569 1 + all_genomes_domains_per_potein_histo.get( domains ) );
2571 if ( domains == 1 ) {
2572 final String domain = protein.getProteinDomain( 0 ).getDomainId().getId();
2573 if ( !domains_which_are_sometimes_single_sometimes_not.contains( domain ) ) {
2574 if ( domains_which_never_single.contains( domain ) ) {
2575 domains_which_never_single.remove( domain );
2576 domains_which_are_sometimes_single_sometimes_not.add( domain );
2579 domains_which_are_always_single.add( domain );
2583 else if ( domains > 1 ) {
2584 for( final Domain d : protein.getProteinDomains() ) {
2585 final String domain = d.getDomainId().getId();
2586 // System.out.println( domain );
2587 if ( !domains_which_are_sometimes_single_sometimes_not.contains( domain ) ) {
2588 if ( domains_which_are_always_single.contains( domain ) ) {
2589 domains_which_are_always_single.remove( domain );
2590 domains_which_are_sometimes_single_sometimes_not.add( domain );
2593 domains_which_never_single.add( domain );
2600 writer.write( genome );
2601 writer.write( "\t" );
2602 if ( stats.getN() >= 1 ) {
2603 writer.write( stats.arithmeticMean() + "" );
2604 writer.write( "\t" );
2605 if ( stats.getN() >= 2 ) {
2606 writer.write( stats.sampleStandardDeviation() + "" );
2611 writer.write( "\t" );
2612 writer.write( stats.median() + "" );
2613 writer.write( "\t" );
2614 writer.write( stats.getN() + "" );
2615 writer.write( "\t" );
2616 writer.write( stats.getMin() + "" );
2617 writer.write( "\t" );
2618 writer.write( stats.getMax() + "" );
2621 writer.write( "\t" );
2622 writer.write( "\t" );
2623 writer.write( "\t" );
2624 writer.write( "0" );
2625 writer.write( "\t" );
2626 writer.write( "\t" );
2628 writer.write( "\n" );
2630 catch ( final IOException e ) {
2631 e.printStackTrace();
2635 final static class DomainComparator implements Comparator<Domain> {
2637 final private boolean _ascending;
2639 public DomainComparator( final boolean ascending ) {
2640 _ascending = ascending;
2644 public final int compare( final Domain d0, final Domain d1 ) {
2646 if ( d0.getFrom() < d1.getFrom() ) {
2647 return _ascending ? -1 : 1;
2649 else if ( d0.getFrom() > d1.getFrom() ) {
2650 return _ascending ? 1 : -1;