3 // FORESTER -- software libraries and applications
4 // for evolutionary biology research and applications.
6 // Copyright (C) 2008-2009 Christian M. Zmasek
7 // Copyright (C) 2008-2009 Burnham Institute for Medical Research
10 // This library is free software; you can redistribute it and/or
11 // modify it under the terms of the GNU Lesser General Public
12 // License as published by the Free Software Foundation; either
13 // version 2.1 of the License, or (at your option) any later version.
15 // This library is distributed in the hope that it will be useful,
16 // but WITHOUT ANY WARRANTY; without even the implied warranty of
17 // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
18 // Lesser General Public License for more details.
20 // You should have received a copy of the GNU Lesser General Public
21 // License along with this library; if not, write to the Free Software
22 // Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA
24 // Contact: phylosoft @ gmail . com
25 // WWW: www.phylosoft.org/forester
27 package org.forester.surfacing;
29 import java.io.BufferedWriter;
31 import java.io.FileWriter;
32 import java.io.IOException;
33 import java.io.Writer;
34 import java.text.DecimalFormat;
35 import java.text.NumberFormat;
36 import java.util.ArrayList;
37 import java.util.Arrays;
38 import java.util.Collections;
39 import java.util.Comparator;
40 import java.util.HashMap;
41 import java.util.HashSet;
42 import java.util.List;
44 import java.util.Map.Entry;
45 import java.util.PriorityQueue;
47 import java.util.SortedMap;
48 import java.util.SortedSet;
49 import java.util.TreeMap;
50 import java.util.TreeSet;
51 import java.util.regex.Matcher;
52 import java.util.regex.Pattern;
54 import org.forester.application.surfacing;
55 import org.forester.evoinference.distance.NeighborJoining;
56 import org.forester.evoinference.matrix.character.BasicCharacterStateMatrix;
57 import org.forester.evoinference.matrix.character.CharacterStateMatrix;
58 import org.forester.evoinference.matrix.character.CharacterStateMatrix.BinaryStates;
59 import org.forester.evoinference.matrix.character.CharacterStateMatrix.Format;
60 import org.forester.evoinference.matrix.character.CharacterStateMatrix.GainLossStates;
61 import org.forester.evoinference.matrix.distance.DistanceMatrix;
62 import org.forester.go.GoId;
63 import org.forester.go.GoNameSpace;
64 import org.forester.go.GoTerm;
65 import org.forester.go.PfamToGoMapping;
66 import org.forester.io.parsers.nexus.NexusConstants;
67 import org.forester.io.writers.PhylogenyWriter;
68 import org.forester.phylogeny.Phylogeny;
69 import org.forester.phylogeny.PhylogenyMethods;
70 import org.forester.phylogeny.PhylogenyNode;
71 import org.forester.phylogeny.PhylogenyNodeI.NH_CONVERSION_SUPPORT_VALUE_STYLE;
72 import org.forester.phylogeny.data.BinaryCharacters;
73 import org.forester.phylogeny.data.Confidence;
74 import org.forester.phylogeny.iterators.PhylogenyNodeIterator;
75 import org.forester.protein.BasicDomain;
76 import org.forester.protein.BasicProtein;
77 import org.forester.protein.BinaryDomainCombination;
78 import org.forester.protein.Domain;
79 import org.forester.protein.DomainId;
80 import org.forester.protein.Protein;
81 import org.forester.species.Species;
82 import org.forester.surfacing.DomainSimilarityCalculator.Detailedness;
83 import org.forester.surfacing.DomainSimilarityCalculator.GoAnnotationOutput;
84 import org.forester.surfacing.GenomeWideCombinableDomains.GenomeWideCombinableDomainsSortOrder;
85 import org.forester.util.AsciiHistogram;
86 import org.forester.util.BasicDescriptiveStatistics;
87 import org.forester.util.BasicTable;
88 import org.forester.util.BasicTableParser;
89 import org.forester.util.DescriptiveStatistics;
90 import org.forester.util.ForesterUtil;
92 public final class SurfacingUtil {
94 private final static NumberFormat FORMATTER = new DecimalFormat( "0.0E0" );
95 private final static NumberFormat FORMATTER_3 = new DecimalFormat( "0.000" );
96 private static final Comparator<Domain> ASCENDING_CONFIDENCE_VALUE_ORDER = new Comparator<Domain>() {
99 public int compare( final Domain d1,
101 if ( d1.getPerSequenceEvalue() < d2
102 .getPerSequenceEvalue() ) {
106 .getPerSequenceEvalue() > d2
107 .getPerSequenceEvalue() ) {
111 return d1.compareTo( d2 );
115 public final static Pattern PATTERN_SP_STYLE_TAXONOMY = Pattern.compile( "^[A-Z0-9]{3,5}$" );
116 private static final boolean USE_LAST = true;
118 private SurfacingUtil() {
119 // Hidden constructor.
122 public static void addAllBinaryDomainCombinationToSet( final GenomeWideCombinableDomains genome,
123 final SortedSet<BinaryDomainCombination> binary_domain_combinations ) {
124 final SortedMap<DomainId, CombinableDomains> all_cd = genome.getAllCombinableDomainsIds();
125 for( final DomainId domain_id : all_cd.keySet() ) {
126 binary_domain_combinations.addAll( all_cd.get( domain_id ).toBinaryDomainCombinations() );
130 public static void addAllDomainIdsToSet( final GenomeWideCombinableDomains genome,
131 final SortedSet<DomainId> domain_ids ) {
132 final SortedSet<DomainId> domains = genome.getAllDomainIds();
133 for( final DomainId domain : domains ) {
134 domain_ids.add( domain );
138 public static void addHtmlHead( final Writer w, final String title ) throws IOException {
139 w.write( SurfacingConstants.NL );
141 w.write( "<title>" );
143 w.write( "</title>" );
144 w.write( SurfacingConstants.NL );
145 w.write( "<style>" );
146 w.write( SurfacingConstants.NL );
147 w.write( "a:visited { color : #6633FF; text-decoration : none; }" );
148 w.write( SurfacingConstants.NL );
149 w.write( "a:link { color : #6633FF; text-decoration : none; }" );
150 w.write( SurfacingConstants.NL );
151 w.write( "a:active { color : #99FF00; text-decoration : none; }" );
152 w.write( SurfacingConstants.NL );
153 w.write( "a:hover { color : #FFFFFF; background-color : #99FF00; text-decoration : none; }" );
154 w.write( SurfacingConstants.NL );
155 w.write( "td { text-align: left; vertical-align: top; font-family: Verdana, Arial, Helvetica; font-size: 8pt}" );
156 w.write( SurfacingConstants.NL );
157 w.write( "h1 { color : #0000FF; font-family: Verdana, Arial, Helvetica; font-size: 18pt; font-weight: bold }" );
158 w.write( SurfacingConstants.NL );
159 w.write( "h2 { color : #0000FF; font-family: Verdana, Arial, Helvetica; font-size: 16pt; font-weight: bold }" );
160 w.write( SurfacingConstants.NL );
161 w.write( "</style>" );
162 w.write( SurfacingConstants.NL );
163 w.write( "</head>" );
164 w.write( SurfacingConstants.NL );
167 public static DescriptiveStatistics calculateDescriptiveStatisticsForMeanValues( final Set<DomainSimilarity> similarities ) {
168 final DescriptiveStatistics stats = new BasicDescriptiveStatistics();
169 for( final DomainSimilarity similarity : similarities ) {
170 stats.addValue( similarity.getMeanSimilarityScore() );
175 private static void calculateIndependentDomainCombinationGains( final Phylogeny local_phylogeny_l,
176 final String outfilename_for_counts,
177 final String outfilename_for_dc,
178 final String outfilename_for_dc_for_go_mapping,
179 final String outfilename_for_dc_for_go_mapping_unique,
180 final String outfilename_for_rank_counts,
181 final String outfilename_for_ancestor_species_counts,
182 final String outfilename_for_protein_stats,
183 final Map<String, DescriptiveStatistics> protein_length_stats_by_dc,
184 final Map<String, DescriptiveStatistics> domain_number_stats_by_dc,
185 final Map<String, DescriptiveStatistics> domain_length_stats_by_domain ) {
188 // if ( protein_length_stats_by_dc != null ) {
189 // for( final Entry<?, DescriptiveStatistics> entry : protein_length_stats_by_dc.entrySet() ) {
190 // System.out.print( entry.getKey().toString() );
191 // System.out.print( ": " );
192 // double[] a = entry.getValue().getDataAsDoubleArray();
193 // for( int i = 0; i < a.length; i++ ) {
194 // System.out.print( a[ i ] + " " );
196 // System.out.println();
199 // if ( domain_number_stats_by_dc != null ) {
200 // for( final Entry<?, DescriptiveStatistics> entry : domain_number_stats_by_dc.entrySet() ) {
201 // System.out.print( entry.getKey().toString() );
202 // System.out.print( ": " );
203 // double[] a = entry.getValue().getDataAsDoubleArray();
204 // for( int i = 0; i < a.length; i++ ) {
205 // System.out.print( a[ i ] + " " );
207 // System.out.println();
211 final BufferedWriter out_counts = new BufferedWriter( new FileWriter( outfilename_for_counts ) );
212 final BufferedWriter out_dc = new BufferedWriter( new FileWriter( outfilename_for_dc ) );
213 final BufferedWriter out_dc_for_go_mapping = new BufferedWriter( new FileWriter( outfilename_for_dc_for_go_mapping ) );
214 final BufferedWriter out_dc_for_go_mapping_unique = new BufferedWriter( new FileWriter( outfilename_for_dc_for_go_mapping_unique ) );
215 final SortedMap<String, Integer> dc_gain_counts = new TreeMap<String, Integer>();
216 for( final PhylogenyNodeIterator it = local_phylogeny_l.iteratorPostorder(); it.hasNext(); ) {
217 final PhylogenyNode n = it.next();
218 final Set<String> gained_dc = n.getNodeData().getBinaryCharacters().getGainedCharacters();
219 for( final String dc : gained_dc ) {
220 if ( dc_gain_counts.containsKey( dc ) ) {
221 dc_gain_counts.put( dc, dc_gain_counts.get( dc ) + 1 );
224 dc_gain_counts.put( dc, 1 );
228 final SortedMap<Integer, Integer> histogram = new TreeMap<Integer, Integer>();
229 final SortedMap<Integer, StringBuilder> domain_lists = new TreeMap<Integer, StringBuilder>();
230 final SortedMap<Integer, DescriptiveStatistics> dc_reapp_counts_to_protein_length_stats = new TreeMap<Integer, DescriptiveStatistics>();
231 final SortedMap<Integer, DescriptiveStatistics> dc_reapp_counts_to_domain_number_stats = new TreeMap<Integer, DescriptiveStatistics>();
232 final SortedMap<Integer, DescriptiveStatistics> dc_reapp_counts_to_domain_lengths_stats = new TreeMap<Integer, DescriptiveStatistics>();
233 final SortedMap<Integer, PriorityQueue<String>> domain_lists_go = new TreeMap<Integer, PriorityQueue<String>>();
234 final SortedMap<Integer, SortedSet<String>> domain_lists_go_unique = new TreeMap<Integer, SortedSet<String>>();
235 final Set<String> dcs = dc_gain_counts.keySet();
236 final SortedSet<String> more_than_once = new TreeSet<String>();
237 final DescriptiveStatistics gained_once_lengths_stats = new BasicDescriptiveStatistics();
238 final DescriptiveStatistics gained_once_domain_count_stats = new BasicDescriptiveStatistics();
239 final DescriptiveStatistics gained_multiple_times_lengths_stats = new BasicDescriptiveStatistics();
240 final DescriptiveStatistics gained_multiple_times_domain_count_stats = new BasicDescriptiveStatistics();
241 long gained_multiple_times_domain_length_sum = 0;
242 long gained_once_domain_length_sum = 0;
243 long gained_multiple_times_domain_length_count = 0;
244 long gained_once_domain_length_count = 0;
245 for( final String dc : dcs ) {
246 final int count = dc_gain_counts.get( dc );
247 if ( histogram.containsKey( count ) ) {
248 histogram.put( count, histogram.get( count ) + 1 );
249 domain_lists.get( count ).append( ", " + dc );
250 domain_lists_go.get( count ).addAll( splitDomainCombination( dc ) );
251 domain_lists_go_unique.get( count ).addAll( splitDomainCombination( dc ) );
254 histogram.put( count, 1 );
255 domain_lists.put( count, new StringBuilder( dc ) );
256 final PriorityQueue<String> q = new PriorityQueue<String>();
257 q.addAll( splitDomainCombination( dc ) );
258 domain_lists_go.put( count, q );
259 final SortedSet<String> set = new TreeSet<String>();
260 set.addAll( splitDomainCombination( dc ) );
261 domain_lists_go_unique.put( count, set );
263 if ( protein_length_stats_by_dc != null ) {
264 if ( !dc_reapp_counts_to_protein_length_stats.containsKey( count ) ) {
265 dc_reapp_counts_to_protein_length_stats.put( count, new BasicDescriptiveStatistics() );
267 dc_reapp_counts_to_protein_length_stats.get( count ).addValue( protein_length_stats_by_dc.get( dc )
270 if ( domain_number_stats_by_dc != null ) {
271 if ( !dc_reapp_counts_to_domain_number_stats.containsKey( count ) ) {
272 dc_reapp_counts_to_domain_number_stats.put( count, new BasicDescriptiveStatistics() );
274 dc_reapp_counts_to_domain_number_stats.get( count ).addValue( domain_number_stats_by_dc.get( dc )
277 if ( domain_length_stats_by_domain != null ) {
278 if ( !dc_reapp_counts_to_domain_lengths_stats.containsKey( count ) ) {
279 dc_reapp_counts_to_domain_lengths_stats.put( count, new BasicDescriptiveStatistics() );
281 final String[] ds = dc.split( "=" );
282 dc_reapp_counts_to_domain_lengths_stats.get( count ).addValue( domain_length_stats_by_domain
283 .get( ds[ 0 ] ).arithmeticMean() );
284 dc_reapp_counts_to_domain_lengths_stats.get( count ).addValue( domain_length_stats_by_domain
285 .get( ds[ 1 ] ).arithmeticMean() );
288 more_than_once.add( dc );
289 if ( protein_length_stats_by_dc != null ) {
290 final DescriptiveStatistics s = protein_length_stats_by_dc.get( dc );
291 for( final double element : s.getData() ) {
292 gained_multiple_times_lengths_stats.addValue( element );
295 if ( domain_number_stats_by_dc != null ) {
296 final DescriptiveStatistics s = domain_number_stats_by_dc.get( dc );
297 for( final double element : s.getData() ) {
298 gained_multiple_times_domain_count_stats.addValue( element );
301 if ( domain_length_stats_by_domain != null ) {
302 final String[] ds = dc.split( "=" );
303 final DescriptiveStatistics s0 = domain_length_stats_by_domain.get( ds[ 0 ] );
304 final DescriptiveStatistics s1 = domain_length_stats_by_domain.get( ds[ 1 ] );
305 for( final double element : s0.getData() ) {
306 gained_multiple_times_domain_length_sum += element;
307 ++gained_multiple_times_domain_length_count;
309 for( final double element : s1.getData() ) {
310 gained_multiple_times_domain_length_sum += element;
311 ++gained_multiple_times_domain_length_count;
316 if ( protein_length_stats_by_dc != null ) {
317 final DescriptiveStatistics s = protein_length_stats_by_dc.get( dc );
318 for( final double element : s.getData() ) {
319 gained_once_lengths_stats.addValue( element );
322 if ( domain_number_stats_by_dc != null ) {
323 final DescriptiveStatistics s = domain_number_stats_by_dc.get( dc );
324 for( final double element : s.getData() ) {
325 gained_once_domain_count_stats.addValue( element );
328 if ( domain_length_stats_by_domain != null ) {
329 final String[] ds = dc.split( "=" );
330 final DescriptiveStatistics s0 = domain_length_stats_by_domain.get( ds[ 0 ] );
331 final DescriptiveStatistics s1 = domain_length_stats_by_domain.get( ds[ 1 ] );
332 for( final double element : s0.getData() ) {
333 gained_once_domain_length_sum += element;
334 ++gained_once_domain_length_count;
336 for( final double element : s1.getData() ) {
337 gained_once_domain_length_sum += element;
338 ++gained_once_domain_length_count;
343 final Set<Integer> histogram_keys = histogram.keySet();
344 for( final Integer histogram_key : histogram_keys ) {
345 final int count = histogram.get( histogram_key );
346 final StringBuilder dc = domain_lists.get( histogram_key );
347 out_counts.write( histogram_key + "\t" + count + ForesterUtil.LINE_SEPARATOR );
348 out_dc.write( histogram_key + "\t" + dc + ForesterUtil.LINE_SEPARATOR );
349 out_dc_for_go_mapping.write( "#" + histogram_key + ForesterUtil.LINE_SEPARATOR );
350 final Object[] sorted = domain_lists_go.get( histogram_key ).toArray();
351 Arrays.sort( sorted );
352 for( final Object domain : sorted ) {
353 out_dc_for_go_mapping.write( domain + ForesterUtil.LINE_SEPARATOR );
355 out_dc_for_go_mapping_unique.write( "#" + histogram_key + ForesterUtil.LINE_SEPARATOR );
356 for( final String domain : domain_lists_go_unique.get( histogram_key ) ) {
357 out_dc_for_go_mapping_unique.write( domain + ForesterUtil.LINE_SEPARATOR );
362 out_dc_for_go_mapping.close();
363 out_dc_for_go_mapping_unique.close();
364 final SortedMap<String, Integer> lca_rank_counts = new TreeMap<String, Integer>();
365 final SortedMap<String, Integer> lca_ancestor_species_counts = new TreeMap<String, Integer>();
366 for( final String dc : more_than_once ) {
367 final List<PhylogenyNode> nodes = new ArrayList<PhylogenyNode>();
368 for( final PhylogenyNodeIterator it = local_phylogeny_l.iteratorExternalForward(); it.hasNext(); ) {
369 final PhylogenyNode n = it.next();
370 if ( n.getNodeData().getBinaryCharacters().getGainedCharacters().contains( dc ) ) {
374 for( int i = 0; i < nodes.size() - 1; ++i ) {
375 for( int j = i + 1; j < nodes.size(); ++j ) {
376 final PhylogenyNode lca = PhylogenyMethods.getInstance().obtainLCA( nodes.get( i ),
378 String rank = "unknown";
379 if ( lca.getNodeData().isHasTaxonomy()
380 && !ForesterUtil.isEmpty( lca.getNodeData().getTaxonomy().getRank() ) ) {
381 rank = lca.getNodeData().getTaxonomy().getRank();
383 addToCountMap( lca_rank_counts, rank );
385 if ( lca.getNodeData().isHasTaxonomy()
386 && !ForesterUtil.isEmpty( lca.getNodeData().getTaxonomy().getScientificName() ) ) {
387 lca_species = lca.getNodeData().getTaxonomy().getScientificName();
389 else if ( lca.getNodeData().isHasTaxonomy()
390 && !ForesterUtil.isEmpty( lca.getNodeData().getTaxonomy().getCommonName() ) ) {
391 lca_species = lca.getNodeData().getTaxonomy().getCommonName();
394 lca_species = lca.getName();
396 addToCountMap( lca_ancestor_species_counts, lca_species );
400 final BufferedWriter out_for_rank_counts = new BufferedWriter( new FileWriter( outfilename_for_rank_counts ) );
401 final BufferedWriter out_for_ancestor_species_counts = new BufferedWriter( new FileWriter( outfilename_for_ancestor_species_counts ) );
402 ForesterUtil.map2writer( out_for_rank_counts, lca_rank_counts, "\t", ForesterUtil.LINE_SEPARATOR );
403 ForesterUtil.map2writer( out_for_ancestor_species_counts,
404 lca_ancestor_species_counts,
406 ForesterUtil.LINE_SEPARATOR );
407 out_for_rank_counts.close();
408 out_for_ancestor_species_counts.close();
409 if ( !ForesterUtil.isEmpty( outfilename_for_protein_stats )
410 && ( ( domain_length_stats_by_domain != null ) || ( protein_length_stats_by_dc != null ) || ( domain_number_stats_by_dc != null ) ) ) {
411 final BufferedWriter w = new BufferedWriter( new FileWriter( outfilename_for_protein_stats ) );
412 w.write( "Domain Lengths: " );
414 if ( domain_length_stats_by_domain != null ) {
415 for( final Entry<Integer, DescriptiveStatistics> entry : dc_reapp_counts_to_domain_lengths_stats
417 w.write( entry.getKey().toString() );
418 w.write( "\t" + entry.getValue().arithmeticMean() );
419 w.write( "\t" + entry.getValue().median() );
426 w.write( "Protein Lengths: " );
428 if ( protein_length_stats_by_dc != null ) {
429 for( final Entry<Integer, DescriptiveStatistics> entry : dc_reapp_counts_to_protein_length_stats
431 w.write( entry.getKey().toString() );
432 w.write( "\t" + entry.getValue().arithmeticMean() );
433 w.write( "\t" + entry.getValue().median() );
440 w.write( "Number of domains: " );
442 if ( domain_number_stats_by_dc != null ) {
443 for( final Entry<Integer, DescriptiveStatistics> entry : dc_reapp_counts_to_domain_number_stats
445 w.write( entry.getKey().toString() );
446 w.write( "\t" + entry.getValue().arithmeticMean() );
447 w.write( "\t" + entry.getValue().median() );
454 w.write( "Gained once, domain lengths:" );
456 w.write( "N: " + gained_once_domain_length_count );
458 w.write( "Avg: " + ( ( double ) gained_once_domain_length_sum / gained_once_domain_length_count ) );
461 w.write( "Gained multiple times, domain lengths:" );
463 w.write( "N: " + gained_multiple_times_domain_length_count );
466 + ( ( double ) gained_multiple_times_domain_length_sum / gained_multiple_times_domain_length_count ) );
471 w.write( "Gained once, protein lengths:" );
473 w.write( gained_once_lengths_stats.toString() );
476 w.write( "Gained once, domain counts:" );
478 w.write( gained_once_domain_count_stats.toString() );
481 w.write( "Gained multiple times, protein lengths:" );
483 w.write( gained_multiple_times_lengths_stats.toString() );
486 w.write( "Gained multiple times, domain counts:" );
488 w.write( gained_multiple_times_domain_count_stats.toString() );
493 catch ( final IOException e ) {
494 ForesterUtil.printWarningMessage( surfacing.PRG_NAME, "Failure to write: " + e );
496 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote independent domain combination gains fitch counts to ["
497 + outfilename_for_counts + "]" );
498 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote independent domain combination gains fitch lists to ["
499 + outfilename_for_dc + "]" );
500 ForesterUtil.programMessage( surfacing.PRG_NAME,
501 "Wrote independent domain combination gains fitch lists to (for GO mapping) ["
502 + outfilename_for_dc_for_go_mapping + "]" );
503 ForesterUtil.programMessage( surfacing.PRG_NAME,
504 "Wrote independent domain combination gains fitch lists to (for GO mapping, unique) ["
505 + outfilename_for_dc_for_go_mapping_unique + "]" );
508 private final static void addToCountMap( final Map<String, Integer> map, final String s ) {
509 if ( map.containsKey( s ) ) {
510 map.put( s, map.get( s ) + 1 );
517 public static int calculateOverlap( final Domain domain, final List<Boolean> covered_positions ) {
518 int overlap_count = 0;
519 for( int i = domain.getFrom(); i <= domain.getTo(); ++i ) {
520 if ( ( i < covered_positions.size() ) && ( covered_positions.get( i ) == true ) ) {
524 return overlap_count;
527 public static void checkForOutputFileWriteability( final File outfile ) {
528 final String error = ForesterUtil.isWritableFile( outfile );
529 if ( !ForesterUtil.isEmpty( error ) ) {
530 ForesterUtil.fatalError( surfacing.PRG_NAME, error );
534 private static SortedSet<String> collectAllDomainsChangedOnSubtree( final PhylogenyNode subtree_root,
535 final boolean get_gains ) {
536 final SortedSet<String> domains = new TreeSet<String>();
537 for( final PhylogenyNode descendant : PhylogenyMethods.getAllDescendants( subtree_root ) ) {
538 final BinaryCharacters chars = descendant.getNodeData().getBinaryCharacters();
540 domains.addAll( chars.getGainedCharacters() );
543 domains.addAll( chars.getLostCharacters() );
549 public static void collectChangedDomainCombinationsFromBinaryStatesMatrixAsListToFile( final CharacterStateMatrix<CharacterStateMatrix.GainLossStates> matrix,
550 final BinaryDomainCombination.DomainCombinationType dc_type,
551 final List<BinaryDomainCombination> all_binary_domains_combination_gained,
552 final boolean get_gains ) {
553 final SortedSet<String> sorted_ids = new TreeSet<String>();
554 for( int i = 0; i < matrix.getNumberOfIdentifiers(); ++i ) {
555 sorted_ids.add( matrix.getIdentifier( i ) );
557 for( final String id : sorted_ids ) {
558 for( int c = 0; c < matrix.getNumberOfCharacters(); ++c ) {
559 if ( ( get_gains && ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.GAIN ) )
560 || ( !get_gains && ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.LOSS ) ) ) {
561 if ( dc_type == BinaryDomainCombination.DomainCombinationType.DIRECTED_ADJACTANT ) {
562 all_binary_domains_combination_gained.add( AdjactantDirectedBinaryDomainCombination
563 .createInstance( matrix.getCharacter( c ) ) );
565 else if ( dc_type == BinaryDomainCombination.DomainCombinationType.DIRECTED ) {
566 all_binary_domains_combination_gained.add( DirectedBinaryDomainCombination
567 .createInstance( matrix.getCharacter( c ) ) );
570 all_binary_domains_combination_gained.add( BasicBinaryDomainCombination.createInstance( matrix
571 .getCharacter( c ) ) );
578 private static File createBaseDirForPerNodeDomainFiles( final String base_dir,
579 final boolean domain_combinations,
580 final CharacterStateMatrix.GainLossStates state,
581 final String outfile ) {
582 File per_node_go_mapped_domain_gain_loss_files_base_dir = new File( new File( outfile ).getParent()
583 + ForesterUtil.FILE_SEPARATOR + base_dir );
584 if ( !per_node_go_mapped_domain_gain_loss_files_base_dir.exists() ) {
585 per_node_go_mapped_domain_gain_loss_files_base_dir.mkdir();
587 if ( domain_combinations ) {
588 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
589 + ForesterUtil.FILE_SEPARATOR + "DC" );
592 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
593 + ForesterUtil.FILE_SEPARATOR + "DOMAINS" );
595 if ( !per_node_go_mapped_domain_gain_loss_files_base_dir.exists() ) {
596 per_node_go_mapped_domain_gain_loss_files_base_dir.mkdir();
598 if ( state == GainLossStates.GAIN ) {
599 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
600 + ForesterUtil.FILE_SEPARATOR + "GAINS" );
602 else if ( state == GainLossStates.LOSS ) {
603 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
604 + ForesterUtil.FILE_SEPARATOR + "LOSSES" );
607 per_node_go_mapped_domain_gain_loss_files_base_dir = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
608 + ForesterUtil.FILE_SEPARATOR + "PRESENT" );
610 if ( !per_node_go_mapped_domain_gain_loss_files_base_dir.exists() ) {
611 per_node_go_mapped_domain_gain_loss_files_base_dir.mkdir();
613 return per_node_go_mapped_domain_gain_loss_files_base_dir;
616 public static Map<DomainId, List<GoId>> createDomainIdToGoIdMap( final List<PfamToGoMapping> pfam_to_go_mappings ) {
617 final Map<DomainId, List<GoId>> domain_id_to_go_ids_map = new HashMap<DomainId, List<GoId>>( pfam_to_go_mappings
619 for( final PfamToGoMapping pfam_to_go : pfam_to_go_mappings ) {
620 if ( !domain_id_to_go_ids_map.containsKey( pfam_to_go.getKey() ) ) {
621 domain_id_to_go_ids_map.put( pfam_to_go.getKey(), new ArrayList<GoId>() );
623 domain_id_to_go_ids_map.get( pfam_to_go.getKey() ).add( pfam_to_go.getValue() );
625 return domain_id_to_go_ids_map;
628 public static Map<DomainId, Set<String>> createDomainIdToSecondaryFeaturesMap( final File secondary_features_map_file )
630 final BasicTable<String> primary_table = BasicTableParser.parse( secondary_features_map_file, "\t" );
631 final Map<DomainId, Set<String>> map = new TreeMap<DomainId, Set<String>>();
632 for( int r = 0; r < primary_table.getNumberOfRows(); ++r ) {
633 final DomainId domain_id = new DomainId( primary_table.getValue( 0, r ) );
634 if ( !map.containsKey( domain_id ) ) {
635 map.put( domain_id, new HashSet<String>() );
637 map.get( domain_id ).add( primary_table.getValue( 1, r ) );
642 public static Phylogeny createNjTreeBasedOnMatrixToFile( final File nj_tree_outfile, final DistanceMatrix distance ) {
643 checkForOutputFileWriteability( nj_tree_outfile );
644 final NeighborJoining nj = NeighborJoining.createInstance();
645 final Phylogeny phylogeny = nj.execute( distance );
646 phylogeny.setName( nj_tree_outfile.getName() );
647 writePhylogenyToFile( phylogeny, nj_tree_outfile.toString() );
651 private static SortedSet<BinaryDomainCombination> createSetOfAllBinaryDomainCombinationsPerGenome( final GenomeWideCombinableDomains gwcd ) {
652 final SortedMap<DomainId, CombinableDomains> cds = gwcd.getAllCombinableDomainsIds();
653 final SortedSet<BinaryDomainCombination> binary_combinations = new TreeSet<BinaryDomainCombination>();
654 for( final DomainId domain_id : cds.keySet() ) {
655 final CombinableDomains cd = cds.get( domain_id );
656 binary_combinations.addAll( cd.toBinaryDomainCombinations() );
658 return binary_combinations;
661 public static void decoratePrintableDomainSimilarities( final SortedSet<DomainSimilarity> domain_similarities,
662 final Detailedness detailedness,
663 final GoAnnotationOutput go_annotation_output,
664 final Map<GoId, GoTerm> go_id_to_term_map,
665 final GoNameSpace go_namespace_limit ) {
666 if ( ( go_namespace_limit != null ) && ( ( go_id_to_term_map == null ) || go_id_to_term_map.isEmpty() ) ) {
667 throw new IllegalArgumentException( "attempt to use a GO namespace limit without a GO id to term map" );
669 for( final DomainSimilarity domain_similarity : domain_similarities ) {
670 if ( domain_similarity instanceof PrintableDomainSimilarity ) {
671 final PrintableDomainSimilarity printable_domain_similarity = ( PrintableDomainSimilarity ) domain_similarity;
672 printable_domain_similarity.setDetailedness( detailedness );
673 printable_domain_similarity.setGoAnnotationOutput( go_annotation_output );
674 printable_domain_similarity.setGoIdToTermMap( go_id_to_term_map );
675 printable_domain_similarity.setGoNamespaceLimit( go_namespace_limit );
680 public static void executeDomainLengthAnalysis( final String[][] input_file_properties,
681 final int number_of_genomes,
682 final DomainLengthsTable domain_lengths_table,
683 final File outfile ) throws IOException {
684 final DecimalFormat df = new DecimalFormat( "#.00" );
685 checkForOutputFileWriteability( outfile );
686 final BufferedWriter out = new BufferedWriter( new FileWriter( outfile ) );
687 out.write( "MEAN BASED STATISTICS PER SPECIES" );
688 out.write( ForesterUtil.LINE_SEPARATOR );
689 out.write( domain_lengths_table.createMeanBasedStatisticsPerSpeciesTable().toString() );
690 out.write( ForesterUtil.LINE_SEPARATOR );
691 out.write( ForesterUtil.LINE_SEPARATOR );
692 final List<DomainLengths> domain_lengths_list = domain_lengths_table.getDomainLengthsList();
693 out.write( "OUTLIER SPECIES PER DOMAIN (Z>=1.5)" );
694 out.write( ForesterUtil.LINE_SEPARATOR );
695 for( final DomainLengths domain_lengths : domain_lengths_list ) {
696 final List<Species> species_list = domain_lengths.getMeanBasedOutlierSpecies( 1.5 );
697 if ( species_list.size() > 0 ) {
698 out.write( domain_lengths.getDomainId() + "\t" );
699 for( final Species species : species_list ) {
700 out.write( species + "\t" );
702 out.write( ForesterUtil.LINE_SEPARATOR );
703 // DescriptiveStatistics stats_for_domain = domain_lengths
704 // .calculateMeanBasedStatistics();
705 //AsciiHistogram histo = new AsciiHistogram( stats_for_domain );
706 //System.out.println( histo.toStringBuffer( 40, '=', 60, 4 ).toString() );
709 out.write( ForesterUtil.LINE_SEPARATOR );
710 out.write( ForesterUtil.LINE_SEPARATOR );
711 out.write( "OUTLIER SPECIES (Z 1.0)" );
712 out.write( ForesterUtil.LINE_SEPARATOR );
713 final DescriptiveStatistics stats_for_all_species = domain_lengths_table
714 .calculateMeanBasedStatisticsForAllSpecies();
715 out.write( stats_for_all_species.asSummary() );
716 out.write( ForesterUtil.LINE_SEPARATOR );
717 final AsciiHistogram histo = new AsciiHistogram( stats_for_all_species );
718 out.write( histo.toStringBuffer( 40, '=', 60, 4 ).toString() );
719 out.write( ForesterUtil.LINE_SEPARATOR );
720 final double population_sd = stats_for_all_species.sampleStandardDeviation();
721 final double population_mean = stats_for_all_species.arithmeticMean();
722 for( final Species species : domain_lengths_table.getSpecies() ) {
723 final double x = domain_lengths_table.calculateMeanBasedStatisticsForSpecies( species ).arithmeticMean();
724 final double z = ( x - population_mean ) / population_sd;
725 out.write( species + "\t" + z );
726 out.write( ForesterUtil.LINE_SEPARATOR );
728 out.write( ForesterUtil.LINE_SEPARATOR );
729 for( final Species species : domain_lengths_table.getSpecies() ) {
730 final DescriptiveStatistics stats_for_species = domain_lengths_table
731 .calculateMeanBasedStatisticsForSpecies( species );
732 final double x = stats_for_species.arithmeticMean();
733 final double z = ( x - population_mean ) / population_sd;
734 if ( ( z <= -1.0 ) || ( z >= 1.0 ) ) {
735 out.write( species + "\t" + df.format( z ) + "\t" + stats_for_species.asSummary() );
736 out.write( ForesterUtil.LINE_SEPARATOR );
740 // final List<HistogramData> histogram_datas = new ArrayList<HistogramData>();
741 // for( int i = 0; i < number_of_genomes; ++i ) {
742 // final Species species = new BasicSpecies( input_file_properties[ i ][ 0 ] );
744 // .add( new HistogramData( species.toString(), domain_lengths_table
745 // .calculateMeanBasedStatisticsForSpecies( species )
746 // .getDataAsDoubleArray(), 5, 600, null, 60 ) );
748 // final HistogramsFrame hf = new HistogramsFrame( histogram_datas );
749 // hf.setVisible( true );
755 * @param all_binary_domains_combination_lost_fitch
756 * @param consider_directedness_and_adjacency_for_bin_combinations
757 * @param all_binary_domains_combination_gained if null ignored, otherwise this is to list all binary domain combinations
758 * which were gained under unweighted (Fitch) parsimony.
760 public static void executeParsimonyAnalysis( final long random_number_seed_for_fitch_parsimony,
761 final boolean radomize_fitch_parsimony,
762 final String outfile_name,
763 final DomainParsimonyCalculator domain_parsimony,
764 final Phylogeny phylogeny,
765 final Map<DomainId, List<GoId>> domain_id_to_go_ids_map,
766 final Map<GoId, GoTerm> go_id_to_term_map,
767 final GoNameSpace go_namespace_limit,
768 final String parameters_str,
769 final Map<DomainId, Set<String>>[] domain_id_to_secondary_features_maps,
770 final SortedSet<DomainId> positive_filter,
771 final boolean output_binary_domain_combinations_for_graphs,
772 final List<BinaryDomainCombination> all_binary_domains_combination_gained_fitch,
773 final List<BinaryDomainCombination> all_binary_domains_combination_lost_fitch,
774 final BinaryDomainCombination.DomainCombinationType dc_type,
775 final Map<String, DescriptiveStatistics> protein_length_stats_by_dc,
776 final Map<String, DescriptiveStatistics> domain_number_stats_by_dc,
777 final Map<String, DescriptiveStatistics> domain_length_stats_by_domain ) {
778 final String sep = ForesterUtil.LINE_SEPARATOR + "###################" + ForesterUtil.LINE_SEPARATOR;
779 final String date_time = ForesterUtil.getCurrentDateTime();
780 final SortedSet<String> all_pfams_encountered = new TreeSet<String>();
781 final SortedSet<String> all_pfams_gained_as_domains = new TreeSet<String>();
782 final SortedSet<String> all_pfams_lost_as_domains = new TreeSet<String>();
783 final SortedSet<String> all_pfams_gained_as_dom_combinations = new TreeSet<String>();
784 final SortedSet<String> all_pfams_lost_as_dom_combinations = new TreeSet<String>();
785 writeToNexus( outfile_name, domain_parsimony, phylogeny );
788 Phylogeny local_phylogeny_l = phylogeny.copy();
789 if ( ( positive_filter != null ) && ( positive_filter.size() > 0 ) ) {
790 domain_parsimony.executeDolloParsimonyOnDomainPresence( positive_filter );
793 domain_parsimony.executeDolloParsimonyOnDomainPresence();
795 SurfacingUtil.writeMatrixToFile( domain_parsimony.getGainLossMatrix(), outfile_name
796 + surfacing.PARSIMONY_OUTPUT_GL_SUFFIX_DOLLO_DOMAINS, Format.FORESTER );
797 SurfacingUtil.writeMatrixToFile( domain_parsimony.getGainLossCountsMatrix(), outfile_name
798 + surfacing.PARSIMONY_OUTPUT_GL_COUNTS_SUFFIX_DOLLO_DOMAINS, Format.FORESTER );
799 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
800 CharacterStateMatrix.GainLossStates.GAIN,
801 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_GAINS_D,
803 ForesterUtil.LINE_SEPARATOR,
805 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
806 CharacterStateMatrix.GainLossStates.LOSS,
807 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_LOSSES_D,
809 ForesterUtil.LINE_SEPARATOR,
811 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(), null, outfile_name
812 + surfacing.PARSIMONY_OUTPUT_DOLLO_PRESENT_D, sep, ForesterUtil.LINE_SEPARATOR, null );
814 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
818 domain_parsimony.getGainLossMatrix(),
819 CharacterStateMatrix.GainLossStates.GAIN,
820 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_GAINS_HTML_D,
822 ForesterUtil.LINE_SEPARATOR,
823 "Dollo Parsimony | Gains | Domains",
825 domain_id_to_secondary_features_maps,
826 all_pfams_encountered,
827 all_pfams_gained_as_domains,
829 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
833 domain_parsimony.getGainLossMatrix(),
834 CharacterStateMatrix.GainLossStates.LOSS,
835 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_LOSSES_HTML_D,
837 ForesterUtil.LINE_SEPARATOR,
838 "Dollo Parsimony | Losses | Domains",
840 domain_id_to_secondary_features_maps,
841 all_pfams_encountered,
842 all_pfams_lost_as_domains,
844 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
848 domain_parsimony.getGainLossMatrix(),
850 outfile_name + surfacing.PARSIMONY_OUTPUT_DOLLO_PRESENT_HTML_D,
852 ForesterUtil.LINE_SEPARATOR,
853 "Dollo Parsimony | Present | Domains",
855 domain_id_to_secondary_features_maps,
856 all_pfams_encountered,
858 "_dollo_present_d" );
859 preparePhylogeny( local_phylogeny_l,
862 "Dollo parsimony on domain presence/absence",
863 "dollo_on_domains_" + outfile_name,
865 SurfacingUtil.writePhylogenyToFile( local_phylogeny_l, outfile_name
866 + surfacing.DOMAINS_PARSIMONY_TREE_OUTPUT_SUFFIX_DOLLO );
868 writeAllDomainsChangedOnAllSubtrees( local_phylogeny_l, true, outfile_name, "_dollo_all_gains_d" );
869 writeAllDomainsChangedOnAllSubtrees( local_phylogeny_l, false, outfile_name, "_dollo_all_losses_d" );
871 catch ( final IOException e ) {
873 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getLocalizedMessage() );
875 if ( domain_parsimony.calculateNumberOfBinaryDomainCombination() > 0 ) {
876 // FITCH DOMAIN COMBINATIONS
877 // -------------------------
878 local_phylogeny_l = phylogeny.copy();
879 String randomization = "no";
880 if ( radomize_fitch_parsimony ) {
881 domain_parsimony.executeFitchParsimonyOnBinaryDomainCombintion( random_number_seed_for_fitch_parsimony );
882 randomization = "yes, seed = " + random_number_seed_for_fitch_parsimony;
885 domain_parsimony.executeFitchParsimonyOnBinaryDomainCombintion( USE_LAST );
887 SurfacingUtil.writeMatrixToFile( domain_parsimony.getGainLossMatrix(), outfile_name
888 + surfacing.PARSIMONY_OUTPUT_GL_SUFFIX_FITCH_BINARY_COMBINATIONS, Format.FORESTER );
889 SurfacingUtil.writeMatrixToFile( domain_parsimony.getGainLossCountsMatrix(), outfile_name
890 + surfacing.PARSIMONY_OUTPUT_GL_COUNTS_SUFFIX_FITCH_BINARY_COMBINATIONS, Format.FORESTER );
892 .writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
893 CharacterStateMatrix.GainLossStates.GAIN,
894 outfile_name + surfacing.PARSIMONY_OUTPUT_FITCH_GAINS_BC,
896 ForesterUtil.LINE_SEPARATOR,
898 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
899 CharacterStateMatrix.GainLossStates.LOSS,
901 + surfacing.PARSIMONY_OUTPUT_FITCH_LOSSES_BC,
903 ForesterUtil.LINE_SEPARATOR,
905 SurfacingUtil.writeBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(), null, outfile_name
906 + surfacing.PARSIMONY_OUTPUT_FITCH_PRESENT_BC, sep, ForesterUtil.LINE_SEPARATOR, null );
907 if ( all_binary_domains_combination_gained_fitch != null ) {
908 collectChangedDomainCombinationsFromBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
910 all_binary_domains_combination_gained_fitch,
913 if ( all_binary_domains_combination_lost_fitch != null ) {
914 collectChangedDomainCombinationsFromBinaryStatesMatrixAsListToFile( domain_parsimony.getGainLossMatrix(),
916 all_binary_domains_combination_lost_fitch,
919 if ( output_binary_domain_combinations_for_graphs ) {
921 .writeBinaryStatesMatrixAsListToFileForBinaryCombinationsForGraphAnalysis( domain_parsimony
922 .getGainLossMatrix(),
925 + surfacing.PARSIMONY_OUTPUT_FITCH_PRESENT_BC_OUTPUTFILE_SUFFIX_FOR_GRAPH_ANALYSIS,
927 ForesterUtil.LINE_SEPARATOR,
928 BinaryDomainCombination.OutputFormat.DOT );
931 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
935 domain_parsimony.getGainLossMatrix(),
936 CharacterStateMatrix.GainLossStates.GAIN,
937 outfile_name + surfacing.PARSIMONY_OUTPUT_FITCH_GAINS_HTML_BC,
939 ForesterUtil.LINE_SEPARATOR,
940 "Fitch Parsimony | Gains | Domain Combinations",
943 all_pfams_encountered,
944 all_pfams_gained_as_dom_combinations,
946 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
950 domain_parsimony.getGainLossMatrix(),
951 CharacterStateMatrix.GainLossStates.LOSS,
952 outfile_name + surfacing.PARSIMONY_OUTPUT_FITCH_LOSSES_HTML_BC,
954 ForesterUtil.LINE_SEPARATOR,
955 "Fitch Parsimony | Losses | Domain Combinations",
958 all_pfams_encountered,
959 all_pfams_lost_as_dom_combinations,
960 "_fitch_losses_dc" );
961 writeBinaryStatesMatrixToList( domain_id_to_go_ids_map,
965 domain_parsimony.getGainLossMatrix(),
967 outfile_name + surfacing.PARSIMONY_OUTPUT_FITCH_PRESENT_HTML_BC,
969 ForesterUtil.LINE_SEPARATOR,
970 "Fitch Parsimony | Present | Domain Combinations",
973 all_pfams_encountered,
975 "_fitch_present_dc" );
976 writeAllEncounteredPfamsToFile( domain_id_to_go_ids_map,
979 all_pfams_encountered );
980 writePfamsToFile( outfile_name + surfacing.ALL_PFAMS_GAINED_AS_DOMAINS_SUFFIX, all_pfams_gained_as_domains );
981 writePfamsToFile( outfile_name + surfacing.ALL_PFAMS_LOST_AS_DOMAINS_SUFFIX, all_pfams_lost_as_domains );
982 writePfamsToFile( outfile_name + surfacing.ALL_PFAMS_GAINED_AS_DC_SUFFIX,
983 all_pfams_gained_as_dom_combinations );
984 writePfamsToFile( outfile_name + surfacing.ALL_PFAMS_LOST_AS_DC_SUFFIX, all_pfams_lost_as_dom_combinations );
985 preparePhylogeny( local_phylogeny_l,
988 "Fitch parsimony on binary domain combination presence/absence randomization: "
990 "fitch_on_binary_domain_combinations_" + outfile_name,
992 SurfacingUtil.writePhylogenyToFile( local_phylogeny_l, outfile_name
993 + surfacing.BINARY_DOMAIN_COMBINATIONS_PARSIMONY_TREE_OUTPUT_SUFFIX_FITCH );
994 calculateIndependentDomainCombinationGains( local_phylogeny_l,
996 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_COUNTS_OUTPUT_SUFFIX,
998 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_OUTPUT_SUFFIX,
1000 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_FOR_GO_MAPPING_OUTPUT_SUFFIX,
1002 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_FOR_GO_MAPPING_OUTPUT_UNIQUE_SUFFIX,
1003 outfile_name + "_indep_dc_gains_fitch_lca_ranks.txt",
1004 outfile_name + "_indep_dc_gains_fitch_lca_taxonomies.txt",
1005 outfile_name + "_indep_dc_gains_fitch_protein_statistics.txt",
1006 protein_length_stats_by_dc,
1007 domain_number_stats_by_dc,
1008 domain_length_stats_by_domain );
1012 public static void executeParsimonyAnalysisForSecondaryFeatures( final String outfile_name,
1013 final DomainParsimonyCalculator secondary_features_parsimony,
1014 final Phylogeny phylogeny,
1015 final String parameters_str,
1016 final Map<Species, MappingResults> mapping_results_map ) {
1017 final String sep = ForesterUtil.LINE_SEPARATOR + "###################" + ForesterUtil.LINE_SEPARATOR;
1018 final String date_time = ForesterUtil.getCurrentDateTime();
1019 System.out.println();
1020 writeToNexus( outfile_name + surfacing.NEXUS_SECONDARY_FEATURES,
1021 secondary_features_parsimony.createMatrixOfSecondaryFeaturePresenceOrAbsence( null ),
1023 Phylogeny local_phylogeny_copy = phylogeny.copy();
1024 secondary_features_parsimony.executeDolloParsimonyOnSecondaryFeatures( mapping_results_map );
1025 SurfacingUtil.writeMatrixToFile( secondary_features_parsimony.getGainLossMatrix(), outfile_name
1026 + surfacing.PARSIMONY_OUTPUT_GL_SUFFIX_DOLLO_SECONDARY_FEATURES, Format.FORESTER );
1027 SurfacingUtil.writeMatrixToFile( secondary_features_parsimony.getGainLossCountsMatrix(), outfile_name
1028 + surfacing.PARSIMONY_OUTPUT_GL_COUNTS_SUFFIX_DOLLO_SECONDARY_FEATURES, Format.FORESTER );
1030 .writeBinaryStatesMatrixAsListToFile( secondary_features_parsimony.getGainLossMatrix(),
1031 CharacterStateMatrix.GainLossStates.GAIN,
1033 + surfacing.PARSIMONY_OUTPUT_DOLLO_GAINS_SECONDARY_FEATURES,
1035 ForesterUtil.LINE_SEPARATOR,
1038 .writeBinaryStatesMatrixAsListToFile( secondary_features_parsimony.getGainLossMatrix(),
1039 CharacterStateMatrix.GainLossStates.LOSS,
1041 + surfacing.PARSIMONY_OUTPUT_DOLLO_LOSSES_SECONDARY_FEATURES,
1043 ForesterUtil.LINE_SEPARATOR,
1046 .writeBinaryStatesMatrixAsListToFile( secondary_features_parsimony.getGainLossMatrix(),
1049 + surfacing.PARSIMONY_OUTPUT_DOLLO_PRESENT_SECONDARY_FEATURES,
1051 ForesterUtil.LINE_SEPARATOR,
1053 preparePhylogeny( local_phylogeny_copy,
1054 secondary_features_parsimony,
1056 "Dollo parsimony on secondary feature presence/absence",
1057 "dollo_on_secondary_features_" + outfile_name,
1059 SurfacingUtil.writePhylogenyToFile( local_phylogeny_copy, outfile_name
1060 + surfacing.SECONDARY_FEATURES_PARSIMONY_TREE_OUTPUT_SUFFIX_DOLLO );
1061 // FITCH DOMAIN COMBINATIONS
1062 // -------------------------
1063 local_phylogeny_copy = phylogeny.copy();
1064 final String randomization = "no";
1065 secondary_features_parsimony.executeFitchParsimonyOnBinaryDomainCombintionOnSecondaryFeatures( USE_LAST );
1066 preparePhylogeny( local_phylogeny_copy,
1067 secondary_features_parsimony,
1069 "Fitch parsimony on secondary binary domain combination presence/absence randomization: "
1071 "fitch_on_binary_domain_combinations_" + outfile_name,
1073 SurfacingUtil.writePhylogenyToFile( local_phylogeny_copy, outfile_name
1074 + surfacing.BINARY_DOMAIN_COMBINATIONS_PARSIMONY_TREE_OUTPUT_SUFFIX_FITCH_MAPPED );
1075 calculateIndependentDomainCombinationGains( local_phylogeny_copy, outfile_name
1076 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_COUNTS_MAPPED_OUTPUT_SUFFIX, outfile_name
1077 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_MAPPED_OUTPUT_SUFFIX, outfile_name
1078 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_FOR_GO_MAPPING_MAPPED_OUTPUT_SUFFIX, outfile_name
1079 + surfacing.INDEPENDENT_DC_GAINS_FITCH_PARS_DC_FOR_GO_MAPPING_MAPPED_OUTPUT_UNIQUE_SUFFIX, outfile_name
1080 + "_MAPPED_indep_dc_gains_fitch_lca_ranks.txt", outfile_name
1081 + "_MAPPED_indep_dc_gains_fitch_lca_taxonomies.txt", null, null, null, null );
1084 public static void doit( final List<Protein> proteins,
1085 final List<DomainId> query_domain_ids_nc_order,
1087 final String separator,
1088 final String limit_to_species,
1089 final Map<String, List<Integer>> average_protein_lengths_by_dc ) throws IOException {
1090 for( final Protein protein : proteins ) {
1091 if ( ForesterUtil.isEmpty( limit_to_species )
1092 || protein.getSpecies().getSpeciesId().equalsIgnoreCase( limit_to_species ) ) {
1093 if ( protein.contains( query_domain_ids_nc_order, true ) ) {
1094 out.write( protein.getSpecies().getSpeciesId() );
1095 out.write( separator );
1096 out.write( protein.getProteinId().getId() );
1097 out.write( separator );
1099 final Set<DomainId> visited_domain_ids = new HashSet<DomainId>();
1100 boolean first = true;
1101 for( final Domain domain : protein.getProteinDomains() ) {
1102 if ( !visited_domain_ids.contains( domain.getDomainId() ) ) {
1103 visited_domain_ids.add( domain.getDomainId() );
1110 out.write( domain.getDomainId().getId() );
1112 out.write( "" + domain.getTotalCount() );
1117 out.write( separator );
1118 if ( !( ForesterUtil.isEmpty( protein.getDescription() ) || protein.getDescription()
1119 .equals( SurfacingConstants.NONE ) ) ) {
1120 out.write( protein.getDescription() );
1122 out.write( separator );
1123 if ( !( ForesterUtil.isEmpty( protein.getAccession() ) || protein.getAccession()
1124 .equals( SurfacingConstants.NONE ) ) ) {
1125 out.write( protein.getAccession() );
1127 out.write( SurfacingConstants.NL );
1134 public static void extractProteinNames( final List<Protein> proteins,
1135 final List<DomainId> query_domain_ids_nc_order,
1137 final String separator,
1138 final String limit_to_species ) throws IOException {
1139 for( final Protein protein : proteins ) {
1140 if ( ForesterUtil.isEmpty( limit_to_species )
1141 || protein.getSpecies().getSpeciesId().equalsIgnoreCase( limit_to_species ) ) {
1142 if ( protein.contains( query_domain_ids_nc_order, true ) ) {
1143 out.write( protein.getSpecies().getSpeciesId() );
1144 out.write( separator );
1145 out.write( protein.getProteinId().getId() );
1146 out.write( separator );
1148 final Set<DomainId> visited_domain_ids = new HashSet<DomainId>();
1149 boolean first = true;
1150 for( final Domain domain : protein.getProteinDomains() ) {
1151 if ( !visited_domain_ids.contains( domain.getDomainId() ) ) {
1152 visited_domain_ids.add( domain.getDomainId() );
1159 out.write( domain.getDomainId().getId() );
1161 out.write( "" + domain.getTotalCount() );
1166 out.write( separator );
1167 if ( !( ForesterUtil.isEmpty( protein.getDescription() ) || protein.getDescription()
1168 .equals( SurfacingConstants.NONE ) ) ) {
1169 out.write( protein.getDescription() );
1171 out.write( separator );
1172 if ( !( ForesterUtil.isEmpty( protein.getAccession() ) || protein.getAccession()
1173 .equals( SurfacingConstants.NONE ) ) ) {
1174 out.write( protein.getAccession() );
1176 out.write( SurfacingConstants.NL );
1183 public static void extractProteinNames( final SortedMap<Species, List<Protein>> protein_lists_per_species,
1184 final DomainId domain_id,
1186 final String separator,
1187 final String limit_to_species ) throws IOException {
1188 for( final Species species : protein_lists_per_species.keySet() ) {
1189 for( final Protein protein : protein_lists_per_species.get( species ) ) {
1190 if ( ForesterUtil.isEmpty( limit_to_species )
1191 || protein.getSpecies().getSpeciesId().equalsIgnoreCase( limit_to_species ) ) {
1192 final List<Domain> domains = protein.getProteinDomains( domain_id );
1193 if ( domains.size() > 0 ) {
1194 final DescriptiveStatistics stats = new BasicDescriptiveStatistics();
1195 for( final Domain domain : domains ) {
1196 stats.addValue( domain.getPerSequenceEvalue() );
1198 out.write( protein.getSpecies().getSpeciesId() );
1199 out.write( separator );
1200 out.write( protein.getProteinId().getId() );
1201 out.write( separator );
1202 out.write( "[" + FORMATTER.format( stats.median() ) + "]" );
1203 out.write( separator );
1204 if ( !( ForesterUtil.isEmpty( protein.getDescription() ) || protein.getDescription()
1205 .equals( SurfacingConstants.NONE ) ) ) {
1206 out.write( protein.getDescription() );
1208 out.write( separator );
1209 if ( !( ForesterUtil.isEmpty( protein.getAccession() ) || protein.getAccession()
1210 .equals( SurfacingConstants.NONE ) ) ) {
1211 out.write( protein.getAccession() );
1213 out.write( SurfacingConstants.NL );
1221 public static SortedSet<DomainId> getAllDomainIds( final List<GenomeWideCombinableDomains> gwcd_list ) {
1222 final SortedSet<DomainId> all_domains_ids = new TreeSet<DomainId>();
1223 for( final GenomeWideCombinableDomains gwcd : gwcd_list ) {
1224 final Set<DomainId> all_domains = gwcd.getAllDomainIds();
1225 // for( final Domain domain : all_domains ) {
1226 all_domains_ids.addAll( all_domains );
1229 return all_domains_ids;
1232 public static SortedMap<String, Integer> getDomainCounts( final List<Protein> protein_domain_collections ) {
1233 final SortedMap<String, Integer> map = new TreeMap<String, Integer>();
1234 for( final Protein protein_domain_collection : protein_domain_collections ) {
1235 for( final Object name : protein_domain_collection.getProteinDomains() ) {
1236 final BasicDomain protein_domain = ( BasicDomain ) name;
1237 final String id = protein_domain.getDomainId().getId();
1238 if ( map.containsKey( id ) ) {
1239 map.put( id, map.get( id ) + 1 );
1249 public static int getNumberOfNodesLackingName( final Phylogeny p, final StringBuilder names ) {
1250 final PhylogenyNodeIterator it = p.iteratorPostorder();
1252 while ( it.hasNext() ) {
1253 final PhylogenyNode n = it.next();
1254 if ( ForesterUtil.isEmpty( n.getName() )
1255 && ( !n.getNodeData().isHasTaxonomy() || ForesterUtil.isEmpty( n.getNodeData().getTaxonomy()
1256 .getScientificName() ) )
1257 && ( !n.getNodeData().isHasTaxonomy() || ForesterUtil.isEmpty( n.getNodeData().getTaxonomy()
1258 .getCommonName() ) ) ) {
1259 if ( n.getParent() != null ) {
1260 names.append( " " );
1261 names.append( n.getParent().getName() );
1263 final List l = n.getAllExternalDescendants();
1264 for( final Object object : l ) {
1265 System.out.println( l.toString() );
1274 * Returns true is Domain domain falls in an uninterrupted stretch of
1275 * covered positions.
1278 * @param covered_positions
1281 public static boolean isEngulfed( final Domain domain, final List<Boolean> covered_positions ) {
1282 for( int i = domain.getFrom(); i <= domain.getTo(); ++i ) {
1283 if ( ( i >= covered_positions.size() ) || ( covered_positions.get( i ) != true ) ) {
1290 public static void preparePhylogeny( final Phylogeny p,
1291 final DomainParsimonyCalculator domain_parsimony,
1292 final String date_time,
1293 final String method,
1295 final String parameters_str ) {
1296 domain_parsimony.decoratePhylogenyWithDomains( p );
1297 final StringBuilder desc = new StringBuilder();
1298 desc.append( "[Method: " + method + "] [Date: " + date_time + "] " );
1299 desc.append( "[Cost: " + domain_parsimony.getCost() + "] " );
1300 desc.append( "[Gains: " + domain_parsimony.getTotalGains() + "] " );
1301 desc.append( "[Losses: " + domain_parsimony.getTotalLosses() + "] " );
1302 desc.append( "[Unchanged: " + domain_parsimony.getTotalUnchanged() + "] " );
1303 desc.append( "[Parameters: " + parameters_str + "]" );
1305 p.setDescription( desc.toString() );
1306 p.setConfidence( new Confidence( domain_parsimony.getCost(), "parsimony" ) );
1307 p.setRerootable( false );
1308 p.setRooted( true );
1312 * species | protein id | n-terminal domain | c-terminal domain | n-terminal domain per domain E-value | c-terminal domain per domain E-value
1316 static public StringBuffer proteinToDomainCombinations( final Protein protein,
1317 final String protein_id,
1318 final String separator ) {
1319 final StringBuffer sb = new StringBuffer();
1320 if ( protein.getSpecies() == null ) {
1321 throw new IllegalArgumentException( "species must not be null" );
1323 if ( ForesterUtil.isEmpty( protein.getSpecies().getSpeciesId() ) ) {
1324 throw new IllegalArgumentException( "species id must not be empty" );
1326 final List<Domain> domains = protein.getProteinDomains();
1327 if ( domains.size() > 1 ) {
1328 final Map<String, Integer> counts = new HashMap<String, Integer>();
1329 for( final Domain domain : domains ) {
1330 final String id = domain.getDomainId().getId();
1331 if ( counts.containsKey( id ) ) {
1332 counts.put( id, counts.get( id ) + 1 );
1335 counts.put( id, 1 );
1338 final Set<String> dcs = new HashSet<String>();
1339 for( int i = 1; i < domains.size(); ++i ) {
1340 for( int j = 0; j < i; ++j ) {
1341 Domain domain_n = domains.get( i );
1342 Domain domain_c = domains.get( j );
1343 if ( domain_n.getFrom() > domain_c.getFrom() ) {
1344 domain_n = domains.get( j );
1345 domain_c = domains.get( i );
1347 final String dc = domain_n.getDomainId().getId() + domain_c.getDomainId().getId();
1348 if ( !dcs.contains( dc ) ) {
1350 sb.append( protein.getSpecies() );
1351 sb.append( separator );
1352 sb.append( protein_id );
1353 sb.append( separator );
1354 sb.append( domain_n.getDomainId().getId() );
1355 sb.append( separator );
1356 sb.append( domain_c.getDomainId().getId() );
1357 sb.append( separator );
1358 sb.append( domain_n.getPerDomainEvalue() );
1359 sb.append( separator );
1360 sb.append( domain_c.getPerDomainEvalue() );
1361 sb.append( separator );
1362 sb.append( counts.get( domain_n.getDomainId().getId() ) );
1363 sb.append( separator );
1364 sb.append( counts.get( domain_c.getDomainId().getId() ) );
1365 sb.append( ForesterUtil.LINE_SEPARATOR );
1370 else if ( domains.size() == 1 ) {
1371 sb.append( protein.getSpecies() );
1372 sb.append( separator );
1373 sb.append( protein_id );
1374 sb.append( separator );
1375 sb.append( domains.get( 0 ).getDomainId().getId() );
1376 sb.append( separator );
1377 sb.append( separator );
1378 sb.append( domains.get( 0 ).getPerDomainEvalue() );
1379 sb.append( separator );
1380 sb.append( separator );
1382 sb.append( separator );
1383 sb.append( ForesterUtil.LINE_SEPARATOR );
1386 sb.append( protein.getSpecies() );
1387 sb.append( separator );
1388 sb.append( protein_id );
1389 sb.append( separator );
1390 sb.append( separator );
1391 sb.append( separator );
1392 sb.append( separator );
1393 sb.append( separator );
1394 sb.append( separator );
1395 sb.append( ForesterUtil.LINE_SEPARATOR );
1402 * Example regarding engulfment: ------------0.1 ----------0.2 --0.3 =>
1403 * domain with 0.3 is ignored
1405 * -----------0.1 ----------0.2 --0.3 => domain with 0.3 is ignored
1408 * ------------0.1 ----------0.3 --0.2 => domains with 0.3 and 0.2 are _not_
1411 * @param max_allowed_overlap
1412 * maximal allowed overlap (inclusive) to be still considered not
1413 * overlapping (zero or negative value to allow any overlap)
1414 * @param remove_engulfed_domains
1415 * to remove domains which are completely engulfed by coverage of
1416 * domains with better support
1420 public static Protein removeOverlappingDomains( final int max_allowed_overlap,
1421 final boolean remove_engulfed_domains,
1422 final Protein protein ) {
1423 final Protein pruned_protein = new BasicProtein( protein.getProteinId().getId(), protein.getSpecies()
1424 .getSpeciesId(), protein.getLength() );
1425 final List<Domain> sorted = SurfacingUtil.sortDomainsWithAscendingConfidenceValues( protein );
1426 final List<Boolean> covered_positions = new ArrayList<Boolean>();
1427 for( final Domain domain : sorted ) {
1428 if ( ( ( max_allowed_overlap < 0 ) || ( SurfacingUtil.calculateOverlap( domain, covered_positions ) <= max_allowed_overlap ) )
1429 && ( !remove_engulfed_domains || !isEngulfed( domain, covered_positions ) ) ) {
1430 final int covered_positions_size = covered_positions.size();
1431 for( int i = covered_positions_size; i < domain.getFrom(); ++i ) {
1432 covered_positions.add( false );
1434 final int new_covered_positions_size = covered_positions.size();
1435 for( int i = domain.getFrom(); i <= domain.getTo(); ++i ) {
1436 if ( i < new_covered_positions_size ) {
1437 covered_positions.set( i, true );
1440 covered_positions.add( true );
1443 pruned_protein.addProteinDomain( domain );
1446 return pruned_protein;
1449 public static List<Domain> sortDomainsWithAscendingConfidenceValues( final Protein protein ) {
1450 final List<Domain> domains = new ArrayList<Domain>();
1451 for( final Domain d : protein.getProteinDomains() ) {
1454 Collections.sort( domains, SurfacingUtil.ASCENDING_CONFIDENCE_VALUE_ORDER );
1458 private static List<String> splitDomainCombination( final String dc ) {
1459 final String[] s = dc.split( "=" );
1460 if ( s.length != 2 ) {
1461 ForesterUtil.printErrorMessage( surfacing.PRG_NAME, "Stringyfied domain combination has illegal format: "
1465 final List<String> l = new ArrayList<String>( 2 );
1471 public static void writeAllDomainsChangedOnAllSubtrees( final Phylogeny p,
1472 final boolean get_gains,
1473 final String outdir,
1474 final String suffix_for_filename ) throws IOException {
1475 CharacterStateMatrix.GainLossStates state = CharacterStateMatrix.GainLossStates.GAIN;
1477 state = CharacterStateMatrix.GainLossStates.LOSS;
1479 final File base_dir = createBaseDirForPerNodeDomainFiles( surfacing.BASE_DIRECTORY_PER_SUBTREE_DOMAIN_GAIN_LOSS_FILES,
1483 for( final PhylogenyNodeIterator it = p.iteratorPostorder(); it.hasNext(); ) {
1484 final PhylogenyNode node = it.next();
1485 if ( !node.isExternal() ) {
1486 final SortedSet<String> domains = collectAllDomainsChangedOnSubtree( node, get_gains );
1487 if ( domains.size() > 0 ) {
1488 final Writer writer = ForesterUtil.createBufferedWriter( base_dir + ForesterUtil.FILE_SEPARATOR
1489 + node.getName() + suffix_for_filename );
1490 for( final String domain : domains ) {
1491 writer.write( domain );
1492 writer.write( ForesterUtil.LINE_SEPARATOR );
1500 private static void writeAllEncounteredPfamsToFile( final Map<DomainId, List<GoId>> domain_id_to_go_ids_map,
1501 final Map<GoId, GoTerm> go_id_to_term_map,
1502 final String outfile_name,
1503 final SortedSet<String> all_pfams_encountered ) {
1504 final File all_pfams_encountered_file = new File( outfile_name + surfacing.ALL_PFAMS_ENCOUNTERED_SUFFIX );
1505 final File all_pfams_encountered_with_go_annotation_file = new File( outfile_name
1506 + surfacing.ALL_PFAMS_ENCOUNTERED_WITH_GO_ANNOTATION_SUFFIX );
1507 final File encountered_pfams_summary_file = new File( outfile_name + surfacing.ENCOUNTERED_PFAMS_SUMMARY_SUFFIX );
1508 int biological_process_counter = 0;
1509 int cellular_component_counter = 0;
1510 int molecular_function_counter = 0;
1511 int pfams_with_mappings_counter = 0;
1512 int pfams_without_mappings_counter = 0;
1513 int pfams_without_mappings_to_bp_or_mf_counter = 0;
1514 int pfams_with_mappings_to_bp_or_mf_counter = 0;
1516 final Writer all_pfams_encountered_writer = new BufferedWriter( new FileWriter( all_pfams_encountered_file ) );
1517 final Writer all_pfams_encountered_with_go_annotation_writer = new BufferedWriter( new FileWriter( all_pfams_encountered_with_go_annotation_file ) );
1518 final Writer summary_writer = new BufferedWriter( new FileWriter( encountered_pfams_summary_file ) );
1519 summary_writer.write( "# Pfam to GO mapping summary" );
1520 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1521 summary_writer.write( "# Actual summary is at the end of this file." );
1522 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1523 summary_writer.write( "# Encountered Pfams without a GO mapping:" );
1524 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1525 for( final String pfam : all_pfams_encountered ) {
1526 all_pfams_encountered_writer.write( pfam );
1527 all_pfams_encountered_writer.write( ForesterUtil.LINE_SEPARATOR );
1528 final DomainId domain_id = new DomainId( pfam );
1529 if ( domain_id_to_go_ids_map.containsKey( domain_id ) ) {
1530 ++pfams_with_mappings_counter;
1531 all_pfams_encountered_with_go_annotation_writer.write( pfam );
1532 all_pfams_encountered_with_go_annotation_writer.write( ForesterUtil.LINE_SEPARATOR );
1533 final List<GoId> go_ids = domain_id_to_go_ids_map.get( domain_id );
1534 boolean maps_to_bp = false;
1535 boolean maps_to_cc = false;
1536 boolean maps_to_mf = false;
1537 for( final GoId go_id : go_ids ) {
1538 final GoTerm go_term = go_id_to_term_map.get( go_id );
1539 if ( go_term.getGoNameSpace().isBiologicalProcess() ) {
1542 else if ( go_term.getGoNameSpace().isCellularComponent() ) {
1545 else if ( go_term.getGoNameSpace().isMolecularFunction() ) {
1550 ++biological_process_counter;
1553 ++cellular_component_counter;
1556 ++molecular_function_counter;
1558 if ( maps_to_bp || maps_to_mf ) {
1559 ++pfams_with_mappings_to_bp_or_mf_counter;
1562 ++pfams_without_mappings_to_bp_or_mf_counter;
1566 ++pfams_without_mappings_to_bp_or_mf_counter;
1567 ++pfams_without_mappings_counter;
1568 summary_writer.write( pfam );
1569 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1572 all_pfams_encountered_writer.close();
1573 all_pfams_encountered_with_go_annotation_writer.close();
1574 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote all [" + all_pfams_encountered.size()
1575 + "] encountered Pfams to: \"" + all_pfams_encountered_file + "\"" );
1576 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote all [" + pfams_with_mappings_counter
1577 + "] encountered Pfams with GO mappings to: \"" + all_pfams_encountered_with_go_annotation_file
1579 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote summary (including all ["
1580 + pfams_without_mappings_counter + "] encountered Pfams without GO mappings) to: \""
1581 + encountered_pfams_summary_file + "\"" );
1582 ForesterUtil.programMessage( surfacing.PRG_NAME, "Sum of Pfams encountered : "
1583 + all_pfams_encountered.size() );
1584 ForesterUtil.programMessage( surfacing.PRG_NAME, "Pfams without a mapping : "
1585 + pfams_without_mappings_counter + " ["
1586 + ( 100 * pfams_without_mappings_counter / all_pfams_encountered.size() ) + "%]" );
1587 ForesterUtil.programMessage( surfacing.PRG_NAME, "Pfams without mapping to proc. or func. : "
1588 + pfams_without_mappings_to_bp_or_mf_counter + " ["
1589 + ( 100 * pfams_without_mappings_to_bp_or_mf_counter / all_pfams_encountered.size() ) + "%]" );
1590 ForesterUtil.programMessage( surfacing.PRG_NAME,
1591 "Pfams with a mapping : " + pfams_with_mappings_counter
1593 + ( 100 * pfams_with_mappings_counter / all_pfams_encountered.size() )
1595 ForesterUtil.programMessage( surfacing.PRG_NAME, "Pfams with a mapping to proc. or func. : "
1596 + pfams_with_mappings_to_bp_or_mf_counter + " ["
1597 + ( 100 * pfams_with_mappings_to_bp_or_mf_counter / all_pfams_encountered.size() ) + "%]" );
1598 ForesterUtil.programMessage( surfacing.PRG_NAME,
1599 "Pfams with mapping to biological process: " + biological_process_counter
1601 + ( 100 * biological_process_counter / all_pfams_encountered.size() )
1603 ForesterUtil.programMessage( surfacing.PRG_NAME,
1604 "Pfams with mapping to molecular function: " + molecular_function_counter
1606 + ( 100 * molecular_function_counter / all_pfams_encountered.size() )
1608 ForesterUtil.programMessage( surfacing.PRG_NAME,
1609 "Pfams with mapping to cellular component: " + cellular_component_counter
1611 + ( 100 * cellular_component_counter / all_pfams_encountered.size() )
1613 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1614 summary_writer.write( "# Sum of Pfams encountered : " + all_pfams_encountered.size() );
1615 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1616 summary_writer.write( "# Pfams without a mapping : " + pfams_without_mappings_counter
1617 + " [" + ( 100 * pfams_without_mappings_counter / all_pfams_encountered.size() ) + "%]" );
1618 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1619 summary_writer.write( "# Pfams without mapping to proc. or func. : "
1620 + pfams_without_mappings_to_bp_or_mf_counter + " ["
1621 + ( 100 * pfams_without_mappings_to_bp_or_mf_counter / all_pfams_encountered.size() ) + "%]" );
1622 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1623 summary_writer.write( "# Pfams with a mapping : " + pfams_with_mappings_counter + " ["
1624 + ( 100 * pfams_with_mappings_counter / all_pfams_encountered.size() ) + "%]" );
1625 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1626 summary_writer.write( "# Pfams with a mapping to proc. or func. : "
1627 + pfams_with_mappings_to_bp_or_mf_counter + " ["
1628 + ( 100 * pfams_with_mappings_to_bp_or_mf_counter / all_pfams_encountered.size() ) + "%]" );
1629 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1630 summary_writer.write( "# Pfams with mapping to biological process: " + biological_process_counter + " ["
1631 + ( 100 * biological_process_counter / all_pfams_encountered.size() ) + "%]" );
1632 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1633 summary_writer.write( "# Pfams with mapping to molecular function: " + molecular_function_counter + " ["
1634 + ( 100 * molecular_function_counter / all_pfams_encountered.size() ) + "%]" );
1635 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1636 summary_writer.write( "# Pfams with mapping to cellular component: " + cellular_component_counter + " ["
1637 + ( 100 * cellular_component_counter / all_pfams_encountered.size() ) + "%]" );
1638 summary_writer.write( ForesterUtil.LINE_SEPARATOR );
1639 summary_writer.close();
1641 catch ( final IOException e ) {
1642 ForesterUtil.printWarningMessage( surfacing.PRG_NAME, "Failure to write: " + e );
1646 public static void writeBinaryDomainCombinationsFileForGraphAnalysis( final String[][] input_file_properties,
1647 final File output_dir,
1648 final GenomeWideCombinableDomains gwcd,
1650 final GenomeWideCombinableDomainsSortOrder dc_sort_order ) {
1651 File dc_outfile_dot = new File( input_file_properties[ i ][ 0 ]
1652 + surfacing.DOMAIN_COMBINITONS_OUTPUTFILE_SUFFIX_FOR_GRAPH_ANALYSIS );
1653 if ( output_dir != null ) {
1654 dc_outfile_dot = new File( output_dir + ForesterUtil.FILE_SEPARATOR + dc_outfile_dot );
1656 checkForOutputFileWriteability( dc_outfile_dot );
1657 final SortedSet<BinaryDomainCombination> binary_combinations = createSetOfAllBinaryDomainCombinationsPerGenome( gwcd );
1659 final BufferedWriter out_dot = new BufferedWriter( new FileWriter( dc_outfile_dot ) );
1660 for( final BinaryDomainCombination bdc : binary_combinations ) {
1661 out_dot.write( bdc.toGraphDescribingLanguage( BinaryDomainCombination.OutputFormat.DOT, null, null )
1663 out_dot.write( SurfacingConstants.NL );
1667 catch ( final IOException e ) {
1668 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1670 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote binary domain combination for \""
1671 + input_file_properties[ i ][ 0 ] + "\" (" + input_file_properties[ i ][ 1 ] + ", "
1672 + input_file_properties[ i ][ 2 ] + ") to: \"" + dc_outfile_dot + "\"" );
1675 public static void writeBinaryStatesMatrixAsListToFile( final CharacterStateMatrix<CharacterStateMatrix.GainLossStates> matrix,
1676 final CharacterStateMatrix.GainLossStates state,
1677 final String filename,
1678 final String indentifier_characters_separator,
1679 final String character_separator,
1680 final Map<String, String> descriptions ) {
1681 final File outfile = new File( filename );
1682 checkForOutputFileWriteability( outfile );
1683 final SortedSet<String> sorted_ids = new TreeSet<String>();
1684 for( int i = 0; i < matrix.getNumberOfIdentifiers(); ++i ) {
1685 sorted_ids.add( matrix.getIdentifier( i ) );
1688 final BufferedWriter out = new BufferedWriter( new FileWriter( outfile ) );
1689 for( final String id : sorted_ids ) {
1690 out.write( indentifier_characters_separator );
1691 out.write( "#" + id );
1692 out.write( indentifier_characters_separator );
1693 for( int c = 0; c < matrix.getNumberOfCharacters(); ++c ) {
1695 // using null to indicate either UNCHANGED_PRESENT or GAIN.
1696 if ( ( matrix.getState( id, c ) == state )
1697 || ( ( state == null ) && ( ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.GAIN ) || ( matrix
1698 .getState( id, c ) == CharacterStateMatrix.GainLossStates.UNCHANGED_PRESENT ) ) ) ) {
1699 out.write( matrix.getCharacter( c ) );
1700 if ( ( descriptions != null ) && !descriptions.isEmpty()
1701 && descriptions.containsKey( matrix.getCharacter( c ) ) ) {
1703 out.write( descriptions.get( matrix.getCharacter( c ) ) );
1705 out.write( character_separator );
1712 catch ( final IOException e ) {
1713 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1715 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote characters list: \"" + filename + "\"" );
1718 public static void writeBinaryStatesMatrixAsListToFileForBinaryCombinationsForGraphAnalysis( final CharacterStateMatrix<CharacterStateMatrix.GainLossStates> matrix,
1719 final CharacterStateMatrix.GainLossStates state,
1720 final String filename,
1721 final String indentifier_characters_separator,
1722 final String character_separator,
1723 final BinaryDomainCombination.OutputFormat bc_output_format ) {
1724 final File outfile = new File( filename );
1725 checkForOutputFileWriteability( outfile );
1726 final SortedSet<String> sorted_ids = new TreeSet<String>();
1727 for( int i = 0; i < matrix.getNumberOfIdentifiers(); ++i ) {
1728 sorted_ids.add( matrix.getIdentifier( i ) );
1731 final BufferedWriter out = new BufferedWriter( new FileWriter( outfile ) );
1732 for( final String id : sorted_ids ) {
1733 out.write( indentifier_characters_separator );
1734 out.write( "#" + id );
1735 out.write( indentifier_characters_separator );
1736 for( int c = 0; c < matrix.getNumberOfCharacters(); ++c ) {
1738 // using null to indicate either UNCHANGED_PRESENT or GAIN.
1739 if ( ( matrix.getState( id, c ) == state )
1740 || ( ( state == null ) && ( ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.GAIN ) || ( matrix
1741 .getState( id, c ) == CharacterStateMatrix.GainLossStates.UNCHANGED_PRESENT ) ) ) ) {
1742 BinaryDomainCombination bdc = null;
1744 bdc = BasicBinaryDomainCombination.createInstance( matrix.getCharacter( c ) );
1746 catch ( final Exception e ) {
1747 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getLocalizedMessage() );
1749 out.write( bdc.toGraphDescribingLanguage( bc_output_format, null, null ).toString() );
1750 out.write( character_separator );
1757 catch ( final IOException e ) {
1758 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1760 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote characters list: \"" + filename + "\"" );
1763 public static void writeBinaryStatesMatrixToList( final Map<DomainId, List<GoId>> domain_id_to_go_ids_map,
1764 final Map<GoId, GoTerm> go_id_to_term_map,
1765 final GoNameSpace go_namespace_limit,
1766 final boolean domain_combinations,
1767 final CharacterStateMatrix<CharacterStateMatrix.GainLossStates> matrix,
1768 final CharacterStateMatrix.GainLossStates state,
1769 final String filename,
1770 final String indentifier_characters_separator,
1771 final String character_separator,
1772 final String title_for_html,
1773 final String prefix_for_html,
1774 final Map<DomainId, Set<String>>[] domain_id_to_secondary_features_maps,
1775 final SortedSet<String> all_pfams_encountered,
1776 final SortedSet<String> pfams_gained_or_lost,
1777 final String suffix_for_per_node_events_file ) {
1778 if ( ( go_namespace_limit != null ) && ( ( go_id_to_term_map == null ) || ( go_id_to_term_map.size() < 1 ) ) ) {
1779 throw new IllegalArgumentException( "attempt to use GO namespace limit without a GO-id to term map" );
1781 else if ( ( ( domain_id_to_go_ids_map == null ) || ( domain_id_to_go_ids_map.size() < 1 ) ) ) {
1782 throw new IllegalArgumentException( "attempt to output detailed HTML without a Pfam to GO map" );
1784 else if ( ( ( go_id_to_term_map == null ) || ( go_id_to_term_map.size() < 1 ) ) ) {
1785 throw new IllegalArgumentException( "attempt to output detailed HTML without a GO-id to term map" );
1787 final File outfile = new File( filename );
1788 checkForOutputFileWriteability( outfile );
1789 final SortedSet<String> sorted_ids = new TreeSet<String>();
1790 for( int i = 0; i < matrix.getNumberOfIdentifiers(); ++i ) {
1791 sorted_ids.add( matrix.getIdentifier( i ) );
1794 final Writer out = new BufferedWriter( new FileWriter( outfile ) );
1795 final File per_node_go_mapped_domain_gain_loss_files_base_dir = createBaseDirForPerNodeDomainFiles( surfacing.BASE_DIRECTORY_PER_NODE_DOMAIN_GAIN_LOSS_FILES,
1796 domain_combinations,
1799 Writer per_node_go_mapped_domain_gain_loss_outfile_writer = null;
1800 File per_node_go_mapped_domain_gain_loss_outfile = null;
1801 int per_node_counter = 0;
1802 out.write( "<html>" );
1803 out.write( SurfacingConstants.NL );
1804 addHtmlHead( out, title_for_html );
1805 out.write( SurfacingConstants.NL );
1806 out.write( "<body>" );
1807 out.write( SurfacingConstants.NL );
1808 out.write( "<h1>" );
1809 out.write( SurfacingConstants.NL );
1810 out.write( title_for_html );
1811 out.write( SurfacingConstants.NL );
1812 out.write( "</h1>" );
1813 out.write( SurfacingConstants.NL );
1814 out.write( "<table>" );
1815 out.write( SurfacingConstants.NL );
1816 for( final String id : sorted_ids ) {
1817 final Matcher matcher = PATTERN_SP_STYLE_TAXONOMY.matcher( id );
1818 if ( matcher.matches() ) {
1821 out.write( "<tr>" );
1822 out.write( "<td>" );
1823 out.write( "<a href=\"#" + id + "\">" + id + "</a>" );
1824 out.write( "</td>" );
1825 out.write( "</tr>" );
1826 out.write( SurfacingConstants.NL );
1828 out.write( "</table>" );
1829 out.write( SurfacingConstants.NL );
1830 for( final String id : sorted_ids ) {
1831 final Matcher matcher = PATTERN_SP_STYLE_TAXONOMY.matcher( id );
1832 if ( matcher.matches() ) {
1835 out.write( SurfacingConstants.NL );
1836 out.write( "<h2>" );
1837 out.write( "<a name=\"" + id + "\">" + id + "</a>" );
1838 writeTaxonomyLinks( out, id );
1839 out.write( "</h2>" );
1840 out.write( SurfacingConstants.NL );
1841 out.write( "<table>" );
1842 out.write( SurfacingConstants.NL );
1843 out.write( "<tr>" );
1844 out.write( "<td><b>" );
1845 out.write( "Pfam domain(s)" );
1846 out.write( "</b></td><td><b>" );
1847 out.write( "GO term acc" );
1848 out.write( "</b></td><td><b>" );
1849 out.write( "GO term" );
1850 out.write( "</b></td><td><b>" );
1851 out.write( "GO namespace" );
1852 out.write( "</b></td>" );
1853 out.write( "</tr>" );
1854 out.write( SurfacingConstants.NL );
1855 out.write( "</tr>" );
1856 out.write( SurfacingConstants.NL );
1857 per_node_counter = 0;
1858 if ( matrix.getNumberOfCharacters() > 0 ) {
1859 per_node_go_mapped_domain_gain_loss_outfile = new File( per_node_go_mapped_domain_gain_loss_files_base_dir
1860 + ForesterUtil.FILE_SEPARATOR + id + suffix_for_per_node_events_file );
1861 SurfacingUtil.checkForOutputFileWriteability( per_node_go_mapped_domain_gain_loss_outfile );
1862 per_node_go_mapped_domain_gain_loss_outfile_writer = ForesterUtil
1863 .createBufferedWriter( per_node_go_mapped_domain_gain_loss_outfile );
1866 per_node_go_mapped_domain_gain_loss_outfile = null;
1867 per_node_go_mapped_domain_gain_loss_outfile_writer = null;
1869 for( int c = 0; c < matrix.getNumberOfCharacters(); ++c ) {
1871 // using null to indicate either UNCHANGED_PRESENT or GAIN.
1872 if ( ( matrix.getState( id, c ) == state )
1873 || ( ( state == null ) && ( ( matrix.getState( id, c ) == CharacterStateMatrix.GainLossStates.UNCHANGED_PRESENT ) || ( matrix
1874 .getState( id, c ) == CharacterStateMatrix.GainLossStates.GAIN ) ) ) ) {
1875 final String character = matrix.getCharacter( c );
1876 String domain_0 = "";
1877 String domain_1 = "";
1878 if ( character.indexOf( BinaryDomainCombination.SEPARATOR ) > 0 ) {
1879 final String[] s = character.split( BinaryDomainCombination.SEPARATOR );
1880 if ( s.length != 2 ) {
1881 throw new AssertionError( "this should not have happened: unexpected format for domain combination: ["
1882 + character + "]" );
1888 domain_0 = character;
1890 writeDomainData( domain_id_to_go_ids_map,
1897 character_separator,
1898 domain_id_to_secondary_features_maps,
1900 all_pfams_encountered.add( domain_0 );
1901 if ( pfams_gained_or_lost != null ) {
1902 pfams_gained_or_lost.add( domain_0 );
1904 if ( !ForesterUtil.isEmpty( domain_1 ) ) {
1905 all_pfams_encountered.add( domain_1 );
1906 if ( pfams_gained_or_lost != null ) {
1907 pfams_gained_or_lost.add( domain_1 );
1910 if ( per_node_go_mapped_domain_gain_loss_outfile_writer != null ) {
1911 writeDomainsToIndividualFilePerTreeNode( per_node_go_mapped_domain_gain_loss_outfile_writer,
1918 if ( per_node_go_mapped_domain_gain_loss_outfile_writer != null ) {
1919 per_node_go_mapped_domain_gain_loss_outfile_writer.close();
1920 if ( per_node_counter < 1 ) {
1921 per_node_go_mapped_domain_gain_loss_outfile.delete();
1923 per_node_counter = 0;
1925 out.write( "</table>" );
1926 out.write( SurfacingConstants.NL );
1927 out.write( "<hr>" );
1928 out.write( SurfacingConstants.NL );
1929 } // for( final String id : sorted_ids ) {
1930 out.write( "</body>" );
1931 out.write( SurfacingConstants.NL );
1932 out.write( "</html>" );
1933 out.write( SurfacingConstants.NL );
1937 catch ( final IOException e ) {
1938 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1940 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote characters detailed HTML list: \"" + filename + "\"" );
1943 public static void writeDomainCombinationsCountsFile( final String[][] input_file_properties,
1944 final File output_dir,
1945 final Writer per_genome_domain_promiscuity_statistics_writer,
1946 final GenomeWideCombinableDomains gwcd,
1948 final GenomeWideCombinableDomains.GenomeWideCombinableDomainsSortOrder dc_sort_order ) {
1949 File dc_outfile = new File( input_file_properties[ i ][ 0 ]
1950 + surfacing.DOMAIN_COMBINITON_COUNTS_OUTPUTFILE_SUFFIX );
1951 if ( output_dir != null ) {
1952 dc_outfile = new File( output_dir + ForesterUtil.FILE_SEPARATOR + dc_outfile );
1954 checkForOutputFileWriteability( dc_outfile );
1956 final BufferedWriter out = new BufferedWriter( new FileWriter( dc_outfile ) );
1957 out.write( gwcd.toStringBuilder( dc_sort_order ).toString() );
1960 catch ( final IOException e ) {
1961 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1963 final DescriptiveStatistics stats = gwcd.getPerGenomeDomainPromiscuityStatistics();
1965 per_genome_domain_promiscuity_statistics_writer.write( input_file_properties[ i ][ 0 ] + "\t" );
1966 per_genome_domain_promiscuity_statistics_writer.write( FORMATTER_3.format( stats.arithmeticMean() ) + "\t" );
1967 if ( stats.getN() < 2 ) {
1968 per_genome_domain_promiscuity_statistics_writer.write( "n/a" + "\t" );
1971 per_genome_domain_promiscuity_statistics_writer.write( FORMATTER_3.format( stats
1972 .sampleStandardDeviation() ) + "\t" );
1974 per_genome_domain_promiscuity_statistics_writer.write( FORMATTER_3.format( stats.median() ) + "\t" );
1975 per_genome_domain_promiscuity_statistics_writer.write( ( int ) stats.getMin() + "\t" );
1976 per_genome_domain_promiscuity_statistics_writer.write( ( int ) stats.getMax() + "\t" );
1977 per_genome_domain_promiscuity_statistics_writer.write( stats.getN() + "\t" );
1978 final SortedSet<DomainId> mpds = gwcd.getMostPromiscuosDomain();
1979 for( final DomainId mpd : mpds ) {
1980 per_genome_domain_promiscuity_statistics_writer.write( mpd.getId() + " " );
1982 per_genome_domain_promiscuity_statistics_writer.write( ForesterUtil.LINE_SEPARATOR );
1984 catch ( final IOException e ) {
1985 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
1987 if ( input_file_properties[ i ].length == 3 ) {
1988 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote domain combination counts for \""
1989 + input_file_properties[ i ][ 0 ] + "\" (" + input_file_properties[ i ][ 1 ] + ", "
1990 + input_file_properties[ i ][ 2 ] + ") to: \"" + dc_outfile + "\"" );
1993 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote domain combination counts for \""
1994 + input_file_properties[ i ][ 0 ] + "\" (" + input_file_properties[ i ][ 1 ] + ") to: \""
1995 + dc_outfile + "\"" );
1999 private static void writeDomainData( final Map<DomainId, List<GoId>> domain_id_to_go_ids_map,
2000 final Map<GoId, GoTerm> go_id_to_term_map,
2001 final GoNameSpace go_namespace_limit,
2003 final String domain_0,
2004 final String domain_1,
2005 final String prefix_for_html,
2006 final String character_separator_for_non_html_output,
2007 final Map<DomainId, Set<String>>[] domain_id_to_secondary_features_maps,
2008 final Set<GoId> all_go_ids ) throws IOException {
2009 boolean any_go_annotation_present = false;
2010 boolean first_has_no_go = false;
2011 int domain_count = 2; // To distinguish between domains and binary domain combinations.
2012 if ( ForesterUtil.isEmpty( domain_1 ) ) {
2015 // The following has a difficult to understand logic.
2016 for( int d = 0; d < domain_count; ++d ) {
2017 List<GoId> go_ids = null;
2018 boolean go_annotation_present = false;
2020 final DomainId domain_id = new DomainId( domain_0 );
2021 if ( domain_id_to_go_ids_map.containsKey( domain_id ) ) {
2022 go_annotation_present = true;
2023 any_go_annotation_present = true;
2024 go_ids = domain_id_to_go_ids_map.get( domain_id );
2027 first_has_no_go = true;
2031 final DomainId domain_id = new DomainId( domain_1 );
2032 if ( domain_id_to_go_ids_map.containsKey( domain_id ) ) {
2033 go_annotation_present = true;
2034 any_go_annotation_present = true;
2035 go_ids = domain_id_to_go_ids_map.get( domain_id );
2038 if ( go_annotation_present ) {
2039 boolean first = ( ( d == 0 ) || ( ( d == 1 ) && first_has_no_go ) );
2040 for( final GoId go_id : go_ids ) {
2041 out.write( "<tr>" );
2044 writeDomainIdsToHtml( out,
2048 domain_id_to_secondary_features_maps );
2051 out.write( "<td></td>" );
2053 if ( !go_id_to_term_map.containsKey( go_id ) ) {
2054 throw new IllegalArgumentException( "GO-id [" + go_id + "] not found in GO-id to GO-term map" );
2056 final GoTerm go_term = go_id_to_term_map.get( go_id );
2057 if ( ( go_namespace_limit == null ) || go_namespace_limit.equals( go_term.getGoNameSpace() ) ) {
2058 // final String top = GoUtils.getPenultimateGoTerm( go_term, go_id_to_term_map ).getName();
2059 final String go_id_str = go_id.getId();
2060 out.write( "<td>" );
2061 out.write( "<a href=\"" + SurfacingConstants.AMIGO_LINK + go_id_str
2062 + "\" target=\"amigo_window\">" + go_id_str + "</a>" );
2063 out.write( "</td><td>" );
2064 out.write( go_term.getName() );
2065 if ( domain_count == 2 ) {
2066 out.write( " (" + d + ")" );
2068 out.write( "</td><td>" );
2069 // out.write( top );
2070 // out.write( "</td><td>" );
2072 out.write( go_term.getGoNameSpace().toShortString() );
2074 out.write( "</td>" );
2075 if ( all_go_ids != null ) {
2076 all_go_ids.add( go_id );
2080 out.write( "<td>" );
2081 out.write( "</td><td>" );
2082 out.write( "</td><td>" );
2083 out.write( "</td><td>" );
2084 out.write( "</td>" );
2086 out.write( "</tr>" );
2087 out.write( SurfacingConstants.NL );
2090 } // for( int d = 0; d < domain_count; ++d )
2091 if ( !any_go_annotation_present ) {
2092 out.write( "<tr>" );
2093 writeDomainIdsToHtml( out, domain_0, domain_1, prefix_for_html, domain_id_to_secondary_features_maps );
2094 out.write( "<td>" );
2095 out.write( "</td><td>" );
2096 out.write( "</td><td>" );
2097 out.write( "</td><td>" );
2098 out.write( "</td>" );
2099 out.write( "</tr>" );
2100 out.write( SurfacingConstants.NL );
2104 private static void writeDomainIdsToHtml( final Writer out,
2105 final String domain_0,
2106 final String domain_1,
2107 final String prefix_for_detailed_html,
2108 final Map<DomainId, Set<String>>[] domain_id_to_secondary_features_maps )
2109 throws IOException {
2110 out.write( "<td>" );
2111 if ( !ForesterUtil.isEmpty( prefix_for_detailed_html ) ) {
2112 out.write( prefix_for_detailed_html );
2115 out.write( "<a href=\"" + SurfacingConstants.PFAM_FAMILY_ID_LINK + domain_0 + "\">" + domain_0 + "</a>" );
2116 out.write( "</td>" );
2119 public static DescriptiveStatistics writeDomainSimilaritiesToFile( final StringBuilder html_desc,
2120 final StringBuilder html_title,
2121 final Writer single_writer,
2122 Map<Character, Writer> split_writers,
2123 final SortedSet<DomainSimilarity> similarities,
2124 final boolean treat_as_binary,
2125 final List<Species> species_order,
2126 final PrintableDomainSimilarity.PRINT_OPTION print_option,
2127 final DomainSimilarity.DomainSimilaritySortField sort_field,
2128 final DomainSimilarity.DomainSimilarityScoring scoring,
2129 final boolean verbose ) throws IOException {
2130 final DescriptiveStatistics stats = new BasicDescriptiveStatistics();
2131 String histogram_title = null;
2132 switch ( sort_field ) {
2133 case ABS_MAX_COUNTS_DIFFERENCE:
2134 if ( treat_as_binary ) {
2135 histogram_title = "absolute counts difference:";
2138 histogram_title = "absolute (maximal) counts difference:";
2141 case MAX_COUNTS_DIFFERENCE:
2142 if ( treat_as_binary ) {
2143 histogram_title = "counts difference:";
2146 histogram_title = "(maximal) counts difference:";
2150 histogram_title = "score mean:";
2153 histogram_title = "score minimum:";
2156 histogram_title = "score maximum:";
2158 case MAX_DIFFERENCE:
2159 if ( treat_as_binary ) {
2160 histogram_title = "difference:";
2163 histogram_title = "(maximal) difference:";
2167 histogram_title = "score mean:";
2170 histogram_title = "score standard deviation:";
2173 histogram_title = "species number:";
2176 throw new AssertionError( "Unknown sort field: " + sort_field );
2178 for( final DomainSimilarity similarity : similarities ) {
2179 switch ( sort_field ) {
2180 case ABS_MAX_COUNTS_DIFFERENCE:
2181 stats.addValue( Math.abs( similarity.getMaximalDifferenceInCounts() ) );
2183 case MAX_COUNTS_DIFFERENCE:
2184 stats.addValue( similarity.getMaximalDifferenceInCounts() );
2187 stats.addValue( similarity.getMeanSimilarityScore() );
2190 stats.addValue( similarity.getMinimalSimilarityScore() );
2193 stats.addValue( similarity.getMaximalSimilarityScore() );
2195 case MAX_DIFFERENCE:
2196 stats.addValue( similarity.getMaximalDifference() );
2199 stats.addValue( similarity.getMeanSimilarityScore() );
2202 stats.addValue( similarity.getStandardDeviationOfSimilarityScore() );
2205 stats.addValue( similarity.getSpecies().size() );
2208 throw new AssertionError( "Unknown sort field: " + sort_field );
2212 // final HistogramData[] hists = new HistogramData[ 1 ];
2215 // List<HistogramDataItem> data_items = new
2216 // ArrayList<HistogramDataItem>();
2217 // double[] values = stats.getDataAsDoubleArray();
2218 // for( int i = 0; i < values.length; i++ ) {
2219 // HistogramDataItem data_item = new BasicHistogramDataItem( "", values[
2221 // data_items.add( data_item );
2225 // HistogramData hd0 = new HistogramData( "name",
2233 // hists[ 0 ] = hd0;
2235 // final HistogramsFrame hf = new HistogramsFrame( hists );
2236 // hf.setVisible( true );
2238 AsciiHistogram histo = null;
2239 if ( stats.getMin() < stats.getMin() ) {
2240 histo = new AsciiHistogram( stats, histogram_title );
2243 if ( histo != null ) {
2244 System.out.println( histo.toStringBuffer( 20, '|', 40, 5 ) );
2246 System.out.println();
2247 System.out.println( "N : " + stats.getN() );
2248 System.out.println( "Min : " + stats.getMin() );
2249 System.out.println( "Max : " + stats.getMax() );
2250 System.out.println( "Mean : " + stats.arithmeticMean() );
2251 if ( stats.getN() > 1 ) {
2252 System.out.println( "SD : " + stats.sampleStandardDeviation() );
2255 System.out.println( "SD : n/a" );
2257 System.out.println( "Median : " + stats.median() );
2258 if ( stats.getN() > 1 ) {
2259 System.out.println( "Pearsonian skewness : " + stats.pearsonianSkewness() );
2262 System.out.println( "Pearsonian skewness : n/a" );
2265 if ( ( single_writer != null ) && ( ( split_writers == null ) || split_writers.isEmpty() ) ) {
2266 split_writers = new HashMap<Character, Writer>();
2267 split_writers.put( '_', single_writer );
2269 switch ( print_option ) {
2270 case SIMPLE_TAB_DELIMITED:
2273 for( final Character key : split_writers.keySet() ) {
2274 final Writer w = split_writers.get( key );
2275 w.write( "<html>" );
2276 w.write( SurfacingConstants.NL );
2278 addHtmlHead( w, "DCs (" + html_title + ") " + key.toString().toUpperCase() );
2281 addHtmlHead( w, "DCs (" + html_title + ")" );
2283 w.write( SurfacingConstants.NL );
2284 w.write( "<body>" );
2285 w.write( SurfacingConstants.NL );
2286 w.write( html_desc.toString() );
2287 w.write( SurfacingConstants.NL );
2290 w.write( SurfacingConstants.NL );
2291 w.write( "<tt><pre>" );
2292 w.write( SurfacingConstants.NL );
2293 if ( histo != null ) {
2294 w.write( histo.toStringBuffer( 20, '|', 40, 5 ).toString() );
2295 w.write( SurfacingConstants.NL );
2297 w.write( "</pre></tt>" );
2298 w.write( SurfacingConstants.NL );
2299 w.write( "<table>" );
2300 w.write( SurfacingConstants.NL );
2301 w.write( "<tr><td>N: </td><td>" + stats.getN() + "</td></tr>" );
2302 w.write( SurfacingConstants.NL );
2303 w.write( "<tr><td>Min: </td><td>" + stats.getMin() + "</td></tr>" );
2304 w.write( SurfacingConstants.NL );
2305 w.write( "<tr><td>Max: </td><td>" + stats.getMax() + "</td></tr>" );
2306 w.write( SurfacingConstants.NL );
2307 w.write( "<tr><td>Mean: </td><td>" + stats.arithmeticMean() + "</td></tr>" );
2308 w.write( SurfacingConstants.NL );
2309 if ( stats.getN() > 1 ) {
2310 w.write( "<tr><td>SD: </td><td>" + stats.sampleStandardDeviation() + "</td></tr>" );
2313 w.write( "<tr><td>SD: </td><td>n/a</td></tr>" );
2315 w.write( SurfacingConstants.NL );
2316 w.write( "<tr><td>Median: </td><td>" + stats.median() + "</td></tr>" );
2317 w.write( SurfacingConstants.NL );
2318 if ( stats.getN() > 1 ) {
2319 w.write( "<tr><td>Pearsonian skewness: </td><td>" + stats.pearsonianSkewness() + "</td></tr>" );
2322 w.write( "<tr><td>Pearsonian skewness: </td><td>n/a</td></tr>" );
2324 w.write( SurfacingConstants.NL );
2325 w.write( "</table>" );
2326 w.write( SurfacingConstants.NL );
2328 w.write( SurfacingConstants.NL );
2330 w.write( SurfacingConstants.NL );
2332 w.write( SurfacingConstants.NL );
2333 w.write( "<table>" );
2334 w.write( SurfacingConstants.NL );
2338 for( final Writer w : split_writers.values() ) {
2339 w.write( SurfacingConstants.NL );
2341 for( final DomainSimilarity similarity : similarities ) {
2342 if ( ( species_order != null ) && !species_order.isEmpty() ) {
2343 ( ( PrintableDomainSimilarity ) similarity ).setSpeciesOrder( species_order );
2345 if ( single_writer != null ) {
2346 single_writer.write( similarity.toStringBuffer( print_option ).toString() );
2349 Writer local_writer = split_writers.get( ( similarity.getDomainId().getId().charAt( 0 ) + "" )
2350 .toLowerCase().charAt( 0 ) );
2351 if ( local_writer == null ) {
2352 local_writer = split_writers.get( '0' );
2354 local_writer.write( similarity.toStringBuffer( print_option ).toString() );
2356 for( final Writer w : split_writers.values() ) {
2357 w.write( SurfacingConstants.NL );
2360 switch ( print_option ) {
2362 for( final Writer w : split_writers.values() ) {
2363 w.write( SurfacingConstants.NL );
2364 w.write( "</table>" );
2365 w.write( SurfacingConstants.NL );
2366 w.write( "</font>" );
2367 w.write( SurfacingConstants.NL );
2368 w.write( "</body>" );
2369 w.write( SurfacingConstants.NL );
2370 w.write( "</html>" );
2371 w.write( SurfacingConstants.NL );
2375 for( final Writer w : split_writers.values() ) {
2381 private static void writeDomainsToIndividualFilePerTreeNode( final Writer individual_files_writer,
2382 final String domain_0,
2383 final String domain_1 ) throws IOException {
2384 individual_files_writer.write( domain_0 );
2385 individual_files_writer.write( ForesterUtil.LINE_SEPARATOR );
2386 if ( !ForesterUtil.isEmpty( domain_1 ) ) {
2387 individual_files_writer.write( domain_1 );
2388 individual_files_writer.write( ForesterUtil.LINE_SEPARATOR );
2392 public static void writeMatrixToFile( final CharacterStateMatrix<?> matrix,
2393 final String filename,
2394 final Format format ) {
2395 final File outfile = new File( filename );
2396 checkForOutputFileWriteability( outfile );
2398 final BufferedWriter out = new BufferedWriter( new FileWriter( outfile ) );
2399 matrix.toWriter( out, format );
2403 catch ( final IOException e ) {
2404 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
2406 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote matrix: \"" + filename + "\"" );
2409 public static void writeMatrixToFile( final File matrix_outfile, final List<DistanceMatrix> matrices ) {
2410 checkForOutputFileWriteability( matrix_outfile );
2412 final BufferedWriter out = new BufferedWriter( new FileWriter( matrix_outfile ) );
2413 for( final DistanceMatrix distance_matrix : matrices ) {
2414 out.write( distance_matrix.toStringBuffer( DistanceMatrix.Format.PHYLIP ).toString() );
2415 out.write( ForesterUtil.LINE_SEPARATOR );
2420 catch ( final IOException e ) {
2421 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
2423 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote distance matrices to \"" + matrix_outfile + "\"" );
2426 private static void writePfamsToFile( final String outfile_name, final SortedSet<String> pfams ) {
2428 final Writer writer = new BufferedWriter( new FileWriter( new File( outfile_name ) ) );
2429 for( final String pfam : pfams ) {
2430 writer.write( pfam );
2431 writer.write( ForesterUtil.LINE_SEPARATOR );
2434 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote " + pfams.size() + " pfams to [" + outfile_name
2437 catch ( final IOException e ) {
2438 ForesterUtil.printWarningMessage( surfacing.PRG_NAME, "Failure to write: " + e );
2442 public static void writePhylogenyToFile( final Phylogeny phylogeny, final String filename ) {
2443 final PhylogenyWriter writer = new PhylogenyWriter();
2445 writer.toPhyloXML( new File( filename ), phylogeny, 1 );
2447 catch ( final IOException e ) {
2448 ForesterUtil.printWarningMessage( surfacing.PRG_NAME, "failed to write phylogeny to \"" + filename + "\": "
2451 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote phylogeny to \"" + filename + "\"" );
2454 public static void writeTaxonomyLinks( final Writer writer, final String species ) throws IOException {
2455 if ( ( species.length() > 1 ) && ( species.indexOf( '_' ) < 1 ) ) {
2456 final Matcher matcher = PATTERN_SP_STYLE_TAXONOMY.matcher( species );
2457 writer.write( " [" );
2458 if ( matcher.matches() ) {
2459 writer.write( "<a href=\"" + SurfacingConstants.UNIPROT_LINK + species
2460 + "\" target=\"taxonomy_window\">uniprot</a>" );
2463 writer.write( "<a href=\"" + SurfacingConstants.EOL_LINK + species
2464 + "\" target=\"taxonomy_window\">eol</a>" );
2465 writer.write( "|" );
2466 writer.write( "<a href=\"" + SurfacingConstants.TOL_LINK + species
2467 + "\" target=\"taxonomy_window\">tol</a>" );
2469 writer.write( "]" );
2473 private static void writeToNexus( final String outfile_name,
2474 final CharacterStateMatrix<BinaryStates> matrix,
2475 final Phylogeny phylogeny ) {
2476 if ( !( matrix instanceof BasicCharacterStateMatrix ) ) {
2477 throw new IllegalArgumentException( "can only write matrices of type [" + BasicCharacterStateMatrix.class
2480 final BasicCharacterStateMatrix<BinaryStates> my_matrix = ( org.forester.evoinference.matrix.character.BasicCharacterStateMatrix<BinaryStates> ) matrix;
2481 final List<Phylogeny> phylogenies = new ArrayList<Phylogeny>( 1 );
2482 phylogenies.add( phylogeny );
2484 final BufferedWriter w = new BufferedWriter( new FileWriter( outfile_name ) );
2485 w.write( NexusConstants.NEXUS );
2486 w.write( ForesterUtil.LINE_SEPARATOR );
2487 my_matrix.writeNexusTaxaBlock( w );
2488 my_matrix.writeNexusBinaryChractersBlock( w );
2489 PhylogenyWriter.writeNexusTreesBlock( w, phylogenies, NH_CONVERSION_SUPPORT_VALUE_STYLE.NONE );
2492 ForesterUtil.programMessage( surfacing.PRG_NAME, "Wrote Nexus file: \"" + outfile_name + "\"" );
2494 catch ( final IOException e ) {
2495 ForesterUtil.fatalError( surfacing.PRG_NAME, e.getMessage() );
2499 private static void writeToNexus( final String outfile_name,
2500 final DomainParsimonyCalculator domain_parsimony,
2501 final Phylogeny phylogeny ) {
2502 writeToNexus( outfile_name + surfacing.NEXUS_EXTERNAL_DOMAINS,
2503 domain_parsimony.createMatrixOfDomainPresenceOrAbsence(),
2505 writeToNexus( outfile_name + surfacing.NEXUS_EXTERNAL_DOMAIN_COMBINATIONS,
2506 domain_parsimony.createMatrixOfBinaryDomainCombinationPresenceOrAbsence(),
2510 public static void domainsPerProteinsStatistics( final String genome,
2511 final List<Protein> protein_list,
2512 final DescriptiveStatistics all_genomes_domains_per_potein_stats,
2513 final SortedMap<Integer, Integer> all_genomes_domains_per_potein_histo,
2514 final SortedSet<String> domains_which_are_always_single,
2515 final SortedSet<String> domains_which_are_sometimes_single_sometimes_not,
2516 final SortedSet<String> domains_which_never_single,
2517 final Writer writer ) {
2518 final DescriptiveStatistics stats = new BasicDescriptiveStatistics();
2519 for( final Protein protein : protein_list ) {
2520 final int domains = protein.getNumberOfProteinDomains();
2521 //System.out.println( domains );
2522 stats.addValue( domains );
2523 all_genomes_domains_per_potein_stats.addValue( domains );
2524 if ( !all_genomes_domains_per_potein_histo.containsKey( domains ) ) {
2525 all_genomes_domains_per_potein_histo.put( domains, 1 );
2528 all_genomes_domains_per_potein_histo.put( domains,
2529 1 + all_genomes_domains_per_potein_histo.get( domains ) );
2531 if ( domains == 1 ) {
2532 final String domain = protein.getProteinDomain( 0 ).getDomainId().getId();
2533 if ( !domains_which_are_sometimes_single_sometimes_not.contains( domain ) ) {
2534 if ( domains_which_never_single.contains( domain ) ) {
2535 domains_which_never_single.remove( domain );
2536 domains_which_are_sometimes_single_sometimes_not.add( domain );
2539 domains_which_are_always_single.add( domain );
2543 else if ( domains > 1 ) {
2544 for( final Domain d : protein.getProteinDomains() ) {
2545 final String domain = d.getDomainId().getId();
2546 // System.out.println( domain );
2547 if ( !domains_which_are_sometimes_single_sometimes_not.contains( domain ) ) {
2548 if ( domains_which_are_always_single.contains( domain ) ) {
2549 domains_which_are_always_single.remove( domain );
2550 domains_which_are_sometimes_single_sometimes_not.add( domain );
2553 domains_which_never_single.add( domain );
2560 writer.write( genome );
2561 writer.write( "\t" );
2562 if ( stats.getN() >= 1 ) {
2563 writer.write( stats.arithmeticMean() + "" );
2564 writer.write( "\t" );
2565 if ( stats.getN() >= 2 ) {
2566 writer.write( stats.sampleStandardDeviation() + "" );
2571 writer.write( "\t" );
2572 writer.write( stats.median() + "" );
2573 writer.write( "\t" );
2574 writer.write( stats.getN() + "" );
2575 writer.write( "\t" );
2576 writer.write( stats.getMin() + "" );
2577 writer.write( "\t" );
2578 writer.write( stats.getMax() + "" );
2581 writer.write( "\t" );
2582 writer.write( "\t" );
2583 writer.write( "\t" );
2584 writer.write( "0" );
2585 writer.write( "\t" );
2586 writer.write( "\t" );
2588 writer.write( "\n" );
2590 catch ( final IOException e ) {
2591 e.printStackTrace();