2 * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
3 * Copyright (C) $$Year-Rel$$ The Jalview Authors
5 * This file is part of Jalview.
7 * Jalview is free software: you can redistribute it and/or
8 * modify it under the terms of the GNU General Public License
9 * as published by the Free Software Foundation, either version 3
10 * of the License, or (at your option) any later version.
12 * Jalview is distributed in the hope that it will be useful, but
13 * WITHOUT ANY WARRANTY; without even the implied warranty
14 * of MERCHANTABILITY or FITNESS FOR A PARTICULAR
15 * PURPOSE. See the GNU General Public License for more details.
17 * You should have received a copy of the GNU General Public License
18 * along with Jalview. If not, see <http://www.gnu.org/licenses/>.
19 * The Jalview Authors are detailed in the 'AUTHORS' file.
21 package jalview.analysis;
24 import static org.testng.AssertJUnit.assertEquals;
25 import static org.testng.AssertJUnit.assertFalse;
26 import static org.testng.AssertJUnit.assertNull;
27 import static org.testng.AssertJUnit.assertSame;
28 import static org.testng.AssertJUnit.assertTrue;
30 import jalview.datamodel.AlignedCodonFrame;
31 import jalview.datamodel.Alignment;
32 import jalview.datamodel.AlignmentAnnotation;
33 import jalview.datamodel.AlignmentI;
34 import jalview.datamodel.Annotation;
35 import jalview.datamodel.DBRefEntry;
36 import jalview.datamodel.Mapping;
37 import jalview.datamodel.SearchResults;
38 import jalview.datamodel.SearchResults.Match;
39 import jalview.datamodel.Sequence;
40 import jalview.datamodel.SequenceI;
41 import jalview.io.AppletFormatAdapter;
42 import jalview.io.FormatAdapter;
43 import jalview.util.MapList;
44 import jalview.util.MappingUtils;
46 import java.io.IOException;
47 import java.util.ArrayList;
48 import java.util.Arrays;
49 import java.util.Collections;
50 import java.util.HashSet;
51 import java.util.Iterator;
52 import java.util.LinkedHashSet;
53 import java.util.List;
57 import org.testng.annotations.Test;
59 public class AlignmentUtilsTests
62 private static final String TEST_DATA =
64 "#=GS D.melanogaster.1 AC AY119185.1/838-902\n" +
65 "#=GS D.melanogaster.2 AC AC092237.1/57223-57161\n" +
66 "#=GS D.melanogaster.3 AC AY060611.1/560-627\n" +
67 "D.melanogaster.1 G.AGCC.CU...AUGAUCGA\n" +
68 "#=GR D.melanogaster.1 SS ................((((\n" +
69 "D.melanogaster.2 C.AUUCAACU.UAUGAGGAU\n" +
70 "#=GR D.melanogaster.2 SS ................((((\n" +
71 "D.melanogaster.3 G.UGGCGCU..UAUGACGCA\n" +
72 "#=GR D.melanogaster.3 SS (.(((...(....(((((((\n" +
75 private static final String AA_SEQS_1 =
81 private static final String CDNA_SEQS_1 =
83 "AC-GG--CUC-CAA-CT\n" +
85 "-CG-TTA--ACG---AAGT\n";
87 private static final String CDNA_SEQS_2 =
94 // public static Sequence ts=new
95 // Sequence("short","ASDASDASDASDASDASDASDASDASDASDASDASDASD");
96 public static Sequence ts = new Sequence("short",
97 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklm");
99 @Test(groups ={ "Functional" })
100 public void testExpandContext()
102 AlignmentI al = new Alignment(new Sequence[] {});
103 for (int i = 4; i < 14; i += 2)
105 SequenceI s1=ts.deriveSequence().getSubSequence(i, i+7);
108 System.out.println(new AppletFormatAdapter().formatSequences("Clustal", al, true));
109 for (int flnk=-1;flnk<25; flnk++)
111 AlignmentI exp = AlignmentUtils.expandContext(al, flnk);
112 System.out.println("\nFlank size: " + flnk);
113 System.out.println(new AppletFormatAdapter().formatSequences(
114 "Clustal", exp, true));
118 * Full expansion to complete sequences
120 for (SequenceI sq : exp.getSequences())
122 String ung = sq.getSequenceAsString().replaceAll("-+", "");
123 final String errorMsg = "Flanking sequence not the same as original dataset sequence.\n"
126 + sq.getDatasetSequence().getSequenceAsString();
127 assertTrue(errorMsg, ung.equalsIgnoreCase(sq.getDatasetSequence()
128 .getSequenceAsString()));
134 * Last sequence is fully expanded, others have leading gaps to match
136 assertTrue(exp.getSequenceAt(4).getSequenceAsString()
138 assertTrue(exp.getSequenceAt(3).getSequenceAsString()
139 .startsWith("--abc"));
140 assertTrue(exp.getSequenceAt(2).getSequenceAsString()
141 .startsWith("----abc"));
142 assertTrue(exp.getSequenceAt(1).getSequenceAsString()
143 .startsWith("------abc"));
144 assertTrue(exp.getSequenceAt(0).getSequenceAsString()
145 .startsWith("--------abc"));
151 * Test that annotations are correctly adjusted by expandContext
153 @Test(groups ={ "Functional" })
154 public void testExpandContext_annotation()
156 AlignmentI al = new Alignment(new Sequence[]
158 SequenceI ds = new Sequence("Seq1", "ABCDEFGHI");
160 SequenceI seq1 = ds.deriveSequence().getSubSequence(3, 6);
161 al.addSequence(seq1);
164 * Annotate DEF with 4/5/6 respectively
166 Annotation[] anns = new Annotation[]
167 { new Annotation(4), new Annotation(5), new Annotation(6) };
168 AlignmentAnnotation ann = new AlignmentAnnotation("SS",
169 "secondary structure", anns);
170 seq1.addAlignmentAnnotation(ann);
173 * The annotations array should match aligned positions
175 assertEquals(3, ann.annotations.length);
176 assertEquals(4, ann.annotations[0].value, 0.001);
177 assertEquals(5, ann.annotations[1].value, 0.001);
178 assertEquals(6, ann.annotations[2].value, 0.001);
181 * Check annotation to sequence position mappings before expanding the
182 * sequence; these are set up in Sequence.addAlignmentAnnotation ->
183 * Annotation.setSequenceRef -> createSequenceMappings
185 assertNull(ann.getAnnotationForPosition(1));
186 assertNull(ann.getAnnotationForPosition(2));
187 assertNull(ann.getAnnotationForPosition(3));
188 assertEquals(4, ann.getAnnotationForPosition(4).value, 0.001);
189 assertEquals(5, ann.getAnnotationForPosition(5).value, 0.001);
190 assertEquals(6, ann.getAnnotationForPosition(6).value, 0.001);
191 assertNull(ann.getAnnotationForPosition(7));
192 assertNull(ann.getAnnotationForPosition(8));
193 assertNull(ann.getAnnotationForPosition(9));
196 * Expand the subsequence to the full sequence abcDEFghi
198 AlignmentI expanded = AlignmentUtils.expandContext(al, -1);
199 assertEquals("abcDEFghi", expanded.getSequenceAt(0)
200 .getSequenceAsString());
203 * Confirm the alignment and sequence have the same SS annotation,
204 * referencing the expanded sequence
206 ann = expanded.getSequenceAt(0).getAnnotation()[0];
207 assertSame(ann, expanded.getAlignmentAnnotation()[0]);
208 assertSame(expanded.getSequenceAt(0), ann.sequenceRef);
211 * The annotations array should have null values except for annotated
214 assertNull(ann.annotations[0]);
215 assertNull(ann.annotations[1]);
216 assertNull(ann.annotations[2]);
217 assertEquals(4, ann.annotations[3].value, 0.001);
218 assertEquals(5, ann.annotations[4].value, 0.001);
219 assertEquals(6, ann.annotations[5].value, 0.001);
220 assertNull(ann.annotations[6]);
221 assertNull(ann.annotations[7]);
222 assertNull(ann.annotations[8]);
225 * sequence position mappings should be unchanged
227 assertNull(ann.getAnnotationForPosition(1));
228 assertNull(ann.getAnnotationForPosition(2));
229 assertNull(ann.getAnnotationForPosition(3));
230 assertEquals(4, ann.getAnnotationForPosition(4).value, 0.001);
231 assertEquals(5, ann.getAnnotationForPosition(5).value, 0.001);
232 assertEquals(6, ann.getAnnotationForPosition(6).value, 0.001);
233 assertNull(ann.getAnnotationForPosition(7));
234 assertNull(ann.getAnnotationForPosition(8));
235 assertNull(ann.getAnnotationForPosition(9));
239 * Test method that returns a map of lists of sequences by sequence name.
241 * @throws IOException
243 @Test(groups ={ "Functional" })
244 public void testGetSequencesByName() throws IOException
246 final String data = ">Seq1Name\nKQYL\n" + ">Seq2Name\nRFPW\n"
247 + ">Seq1Name\nABCD\n";
248 AlignmentI al = loadAlignment(data, "FASTA");
249 Map<String, List<SequenceI>> map = AlignmentUtils
250 .getSequencesByName(al);
251 assertEquals(2, map.keySet().size());
252 assertEquals(2, map.get("Seq1Name").size());
253 assertEquals("KQYL", map.get("Seq1Name").get(0).getSequenceAsString());
254 assertEquals("ABCD", map.get("Seq1Name").get(1).getSequenceAsString());
255 assertEquals(1, map.get("Seq2Name").size());
256 assertEquals("RFPW", map.get("Seq2Name").get(0).getSequenceAsString());
259 * Helper method to load an alignment and ensure dataset sequences are set up.
264 * @throws IOException
266 protected AlignmentI loadAlignment(final String data, String format) throws IOException
268 AlignmentI a = new FormatAdapter().readFile(data,
269 AppletFormatAdapter.PASTE, format);
275 * Test mapping of protein to cDNA, for the case where we have no sequence
276 * cross-references, so mappings are made first-served 1-1 where sequences
279 * @throws IOException
281 @Test(groups ={ "Functional" })
282 public void testMapProteinToCdna_noXrefs() throws IOException
284 List<SequenceI> protseqs = new ArrayList<SequenceI>();
285 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
286 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
287 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
288 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
289 protein.setDataset(null);
291 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
292 dnaseqs.add(new Sequence("EMBL|A11111", "TCAGCACGC")); // = SAR
293 dnaseqs.add(new Sequence("EMBL|A22222", "GAGATACAA")); // = EIQ
294 dnaseqs.add(new Sequence("EMBL|A33333", "GAAATCCAG")); // = EIQ
295 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG")); // = EIQ
296 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[4]));
297 cdna.setDataset(null);
299 assertTrue(AlignmentUtils.mapProteinToCdna(protein, cdna));
301 // 3 mappings made, each from 1 to 1 sequence
302 assertEquals(3, protein.getCodonFrames().size());
303 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
304 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
305 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
307 // V12345 mapped to A22222
308 AlignedCodonFrame acf = protein.getCodonFrame(
309 protein.getSequenceAt(0)).get(0);
310 assertEquals(1, acf.getdnaSeqs().length);
311 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
312 acf.getdnaSeqs()[0]);
313 Mapping[] protMappings = acf.getProtMappings();
314 assertEquals(1, protMappings.length);
315 MapList mapList = protMappings[0].getMap();
316 assertEquals(3, mapList.getFromRatio());
317 assertEquals(1, mapList.getToRatio());
318 assertTrue(Arrays.equals(new int[]
319 { 1, 9 }, mapList.getFromRanges().get(0)));
320 assertEquals(1, mapList.getFromRanges().size());
321 assertTrue(Arrays.equals(new int[]
322 { 1, 3 }, mapList.getToRanges().get(0)));
323 assertEquals(1, mapList.getToRanges().size());
325 // V12346 mapped to A33333
326 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
327 assertEquals(1, acf.getdnaSeqs().length);
328 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
329 acf.getdnaSeqs()[0]);
331 // V12347 mapped to A11111
332 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
333 assertEquals(1, acf.getdnaSeqs().length);
334 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
335 acf.getdnaSeqs()[0]);
337 // no mapping involving the 'extra' A44444
338 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(3)).isEmpty());
342 * Test for the alignSequenceAs method that takes two sequences and a mapping.
344 @Test(groups ={ "Functional" })
345 public void testAlignSequenceAs_withMapping_noIntrons()
347 MapList map = new MapList(new int[]
352 * No existing gaps in dna:
354 checkAlignSequenceAs("GGGAAA", "-A-L-", false, false, map,
358 * Now introduce gaps in dna but ignore them when realigning.
360 checkAlignSequenceAs("-G-G-G-A-A-A-", "-A-L-", false, false, map,
364 * Now include gaps in dna when realigning. First retaining 'mapped' gaps
365 * only, i.e. those within the exon region.
367 checkAlignSequenceAs("-G-G--G-A--A-A-", "-A-L-", true, false, map,
368 "---G-G--G---A--A-A");
371 * Include all gaps in dna when realigning (within and without the exon
372 * region). The leading gap, and the gaps between codons, are subsumed by
373 * the protein alignment gap.
375 checkAlignSequenceAs("-G-GG--AA-A-", "-A-L-", true, true, map,
379 * Include only unmapped gaps in dna when realigning (outside the exon
380 * region). The leading gap, and the gaps between codons, are subsumed by
381 * the protein alignment gap.
383 checkAlignSequenceAs("-G-GG--AA-A-", "-A-L-", false, true, map,
388 * Test for the alignSequenceAs method that takes two sequences and a mapping.
390 @Test(groups ={ "Functional" })
391 public void testAlignSequenceAs_withMapping_withIntrons()
394 * Exons at codon 2 (AAA) and 4 (TTT)
396 MapList map = new MapList(new int[]
397 { 4, 6, 10, 12 }, new int[]
401 * Simple case: no gaps in dna
403 checkAlignSequenceAs("GGGAAACCCTTTGGG", "--A-L-", false, false, map,
404 "GGG---AAACCCTTTGGG");
407 * Add gaps to dna - but ignore when realigning.
409 checkAlignSequenceAs("-G-G-G--A--A---AC-CC-T-TT-GG-G-", "--A-L-",
410 false, false, map, "GGG---AAACCCTTTGGG");
413 * Add gaps to dna - include within exons only when realigning.
415 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
416 true, false, map, "GGG---A--A---ACCCT-TTGGG");
419 * Include gaps outside exons only when realigning.
421 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
422 false, true, map, "-G-G-GAAAC-CCTTT-GG-G-");
425 * Include gaps following first intron if we are 'preserving mapped gaps'
427 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
428 true, true, map, "-G-G-G--A--A---A-C-CC-T-TT-GG-G-");
431 * Include all gaps in dna when realigning.
433 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
434 true, true, map, "-G-G-G--A--A---A-C-CC-T-TT-GG-G-");
438 * Test for the case where not all of the protein sequence is mapped to cDNA.
440 @Test(groups ={ "Functional" })
441 public void testAlignSequenceAs_withMapping_withUnmappedProtein()
445 * Exons at codon 2 (AAA) and 4 (TTT) mapped to A and P
447 final MapList map = new MapList(new int[]
448 { 4, 6, 10, 12 }, new int[]
449 { 1, 1, 3, 3 }, 3, 1);
453 * Expect alignment does nothing (aborts realignment). Change this test
454 * first if different behaviour wanted.
456 checkAlignSequenceAs("GGGAAACCCTTTGGG", "-A-L-P-", false,
457 false, map, "GGGAAACCCTTTGGG");
461 * Helper method that performs and verifies the method under test.
465 * @param preserveMappedGaps
466 * @param preserveUnmappedGaps
470 protected void checkAlignSequenceAs(final String dnaSeq,
471 final String proteinSeq, final boolean preserveMappedGaps,
472 final boolean preserveUnmappedGaps, MapList map,
473 final String expected)
475 SequenceI dna = new Sequence("Seq1", dnaSeq);
476 dna.createDatasetSequence();
477 SequenceI protein = new Sequence("Seq1", proteinSeq);
478 protein.createDatasetSequence();
479 AlignedCodonFrame acf = new AlignedCodonFrame();
480 acf.addMap(dna.getDatasetSequence(), protein.getDatasetSequence(), map);
482 AlignmentUtils.alignSequenceAs(dna, protein, acf, "---", '-',
483 preserveMappedGaps, preserveUnmappedGaps);
484 assertEquals(expected, dna.getSequenceAsString());
488 * Test for the alignSequenceAs method where we preserve gaps in introns only.
490 @Test(groups ={ "Functional" })
491 public void testAlignSequenceAs_keepIntronGapsOnly()
495 * Intron GGGAAA followed by exon CCCTTT
497 MapList map = new MapList(new int[]
501 checkAlignSequenceAs("GG-G-AA-A-C-CC-T-TT", "AL",
502 false, true, map, "GG-G-AA-ACCCTTT");
506 * Test for the method that generates an aligned translated sequence from one
509 @Test(groups ={ "Functional" })
510 public void testGetAlignedTranslation_dnaLikeProtein()
512 // dna alignment will be replaced
513 SequenceI dna = new Sequence("Seq1", "T-G-CC-A--T-TAC-CAG-");
514 dna.createDatasetSequence();
515 // protein alignment will be 'applied' to dna
516 SequenceI protein = new Sequence("Seq1", "-CH-Y--Q-");
517 protein.createDatasetSequence();
518 MapList map = new MapList(new int[]
521 AlignedCodonFrame acf = new AlignedCodonFrame();
522 acf.addMap(dna.getDatasetSequence(), protein.getDatasetSequence(), map);
524 final SequenceI aligned = AlignmentUtils
525 .getAlignedTranslation(protein, '-', acf);
526 assertEquals("---TGCCAT---TAC------CAG---", aligned.getSequenceAsString());
527 assertSame(aligned.getDatasetSequence(), dna.getDatasetSequence());
531 * Test the method that realigns protein to match mapped codon alignment.
533 @Test(groups ={ "Functional" })
534 public void testAlignProteinAsDna()
536 // seq1 codons are [1,2,3] [4,5,6] [7,8,9] [10,11,12]
537 SequenceI dna1 = new Sequence("Seq1", "TGCCATTACCAG-");
538 // seq2 codons are [1,3,4] [5,6,7] [8,9,10] [11,12,13]
539 SequenceI dna2 = new Sequence("Seq2", "T-GCCATTACCAG");
540 // seq3 codons are [1,2,3] [4,5,7] [8,9,10] [11,12,13]
541 SequenceI dna3 = new Sequence("Seq3", "TGCCA-TTACCAG");
542 AlignmentI dna = new Alignment(new SequenceI[]
543 { dna1, dna2, dna3 });
544 dna.setDataset(null);
546 // protein alignment will be realigned like dna
547 SequenceI prot1 = new Sequence("Seq1", "CHYQ");
548 SequenceI prot2 = new Sequence("Seq2", "CHYQ");
549 SequenceI prot3 = new Sequence("Seq3", "CHYQ");
550 SequenceI prot4 = new Sequence("Seq4", "R-QSV"); // unmapped, unchanged
551 AlignmentI protein = new Alignment(new SequenceI[]
554 protein.setDataset(null);
556 MapList map = new MapList(new int[]
559 AlignedCodonFrame acf = new AlignedCodonFrame();
560 acf.addMap(dna1.getDatasetSequence(), prot1.getDatasetSequence(), map);
561 acf.addMap(dna2.getDatasetSequence(), prot2.getDatasetSequence(), map);
562 acf.addMap(dna3.getDatasetSequence(), prot3.getDatasetSequence(), map);
563 protein.setCodonFrames(Collections.singleton(acf));
566 * Translated codon order is [1,2,3] [1,3,4] [4,5,6] [4,5,7] [5,6,7] [7,8,9]
567 * [8,9,10] [10,11,12] [11,12,13]
569 AlignmentUtils.alignProteinAsDna(protein, dna);
570 assertEquals("C-H--Y-Q-", prot1.getSequenceAsString());
571 assertEquals("-C--H-Y-Q", prot2.getSequenceAsString());
572 assertEquals("C--H--Y-Q", prot3.getSequenceAsString());
573 assertEquals("R-QSV", prot4.getSequenceAsString());
577 * Test the method that tests whether a CDNA sequence translates to a protein
580 @Test(groups ={ "Functional" })
581 public void testTranslatesAs()
583 assertTrue(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(), 0,
584 "FPKG".toCharArray()));
585 // with start codon (not in protein)
586 assertTrue(AlignmentUtils.translatesAs("atgtttcccaaaggg".toCharArray(),
587 3, "FPKG".toCharArray()));
588 // with stop codon1 (not in protein)
589 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtaa".toCharArray(),
590 0, "FPKG".toCharArray()));
591 // with stop codon1 (in protein as *)
592 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtaa".toCharArray(),
593 0, "FPKG*".toCharArray()));
594 // with stop codon2 (not in protein)
595 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtag".toCharArray(),
596 0, "FPKG".toCharArray()));
597 // with stop codon3 (not in protein)
598 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtga".toCharArray(),
599 0, "FPKG".toCharArray()));
600 // with start and stop codon1
601 assertTrue(AlignmentUtils.translatesAs(
602 "atgtttcccaaagggtaa".toCharArray(), 3, "FPKG".toCharArray()));
603 // with start and stop codon1 (in protein as *)
604 assertTrue(AlignmentUtils.translatesAs(
605 "atgtttcccaaagggtaa".toCharArray(), 3, "FPKG*".toCharArray()));
606 // with start and stop codon2
607 assertTrue(AlignmentUtils.translatesAs(
608 "atgtttcccaaagggtag".toCharArray(), 3, "FPKG".toCharArray()));
609 // with start and stop codon3
610 assertTrue(AlignmentUtils.translatesAs(
611 "atgtttcccaaagggtga".toCharArray(), 3, "FPKG".toCharArray()));
613 // with embedded stop codon
614 assertTrue(AlignmentUtils.translatesAs(
615 "atgtttTAGcccaaaTAAgggtga".toCharArray(), 3,
616 "F*PK*G".toCharArray()));
619 assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(),
621 "FPMG".toCharArray()));
625 * Test mapping of protein to cDNA, for cases where the cDNA has start and/or
626 * stop codons in addition to the protein coding sequence.
628 * @throws IOException
630 @Test(groups ={ "Functional" })
631 public void testMapProteinToCdna_withStartAndStopCodons()
634 List<SequenceI> protseqs = new ArrayList<SequenceI>();
635 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
636 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
637 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
638 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
639 protein.setDataset(null);
641 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
643 dnaseqs.add(new Sequence("EMBL|A11111", "ATGTCAGCACGC"));
645 dnaseqs.add(new Sequence("EMBL|A22222", "GAGATACAATAA"));
646 // = start +EIQ + stop
647 dnaseqs.add(new Sequence("EMBL|A33333", "ATGGAAATCCAGTAG"));
648 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG"));
649 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[4]));
650 cdna.setDataset(null);
652 assertTrue(AlignmentUtils.mapProteinToCdna(protein, cdna));
654 // 3 mappings made, each from 1 to 1 sequence
655 assertEquals(3, protein.getCodonFrames().size());
656 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
657 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
658 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
660 // V12345 mapped from A22222
661 AlignedCodonFrame acf = protein.getCodonFrame(
662 protein.getSequenceAt(0)).get(0);
663 assertEquals(1, acf.getdnaSeqs().length);
664 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
665 acf.getdnaSeqs()[0]);
666 Mapping[] protMappings = acf.getProtMappings();
667 assertEquals(1, protMappings.length);
668 MapList mapList = protMappings[0].getMap();
669 assertEquals(3, mapList.getFromRatio());
670 assertEquals(1, mapList.getToRatio());
671 assertTrue(Arrays.equals(new int[]
672 { 1, 9 }, mapList.getFromRanges().get(0)));
673 assertEquals(1, mapList.getFromRanges().size());
674 assertTrue(Arrays.equals(new int[]
675 { 1, 3 }, mapList.getToRanges().get(0)));
676 assertEquals(1, mapList.getToRanges().size());
678 // V12346 mapped from A33333 starting position 4
679 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
680 assertEquals(1, acf.getdnaSeqs().length);
681 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
682 acf.getdnaSeqs()[0]);
683 protMappings = acf.getProtMappings();
684 assertEquals(1, protMappings.length);
685 mapList = protMappings[0].getMap();
686 assertEquals(3, mapList.getFromRatio());
687 assertEquals(1, mapList.getToRatio());
688 assertTrue(Arrays.equals(new int[]
689 { 4, 12 }, mapList.getFromRanges().get(0)));
690 assertEquals(1, mapList.getFromRanges().size());
691 assertTrue(Arrays.equals(new int[]
692 { 1, 3 }, mapList.getToRanges().get(0)));
693 assertEquals(1, mapList.getToRanges().size());
695 // V12347 mapped to A11111 starting position 4
696 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
697 assertEquals(1, acf.getdnaSeqs().length);
698 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
699 acf.getdnaSeqs()[0]);
700 protMappings = acf.getProtMappings();
701 assertEquals(1, protMappings.length);
702 mapList = protMappings[0].getMap();
703 assertEquals(3, mapList.getFromRatio());
704 assertEquals(1, mapList.getToRatio());
705 assertTrue(Arrays.equals(new int[]
706 { 4, 12 }, mapList.getFromRanges().get(0)));
707 assertEquals(1, mapList.getFromRanges().size());
708 assertTrue(Arrays.equals(new int[]
709 { 1, 3 }, mapList.getToRanges().get(0)));
710 assertEquals(1, mapList.getToRanges().size());
712 // no mapping involving the 'extra' A44444
713 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(3)).isEmpty());
717 * Test mapping of protein to cDNA, for the case where we have some sequence
718 * cross-references. Verify that 1-to-many mappings are made where
719 * cross-references exist and sequences are mappable.
721 * @throws IOException
723 @Test(groups ={ "Functional" })
724 public void testMapProteinToCdna_withXrefs() throws IOException
726 List<SequenceI> protseqs = new ArrayList<SequenceI>();
727 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
728 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
729 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
730 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
731 protein.setDataset(null);
733 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
734 dnaseqs.add(new Sequence("EMBL|A11111", "TCAGCACGC")); // = SAR
735 dnaseqs.add(new Sequence("EMBL|A22222", "ATGGAGATACAA")); // = start + EIQ
736 dnaseqs.add(new Sequence("EMBL|A33333", "GAAATCCAG")); // = EIQ
737 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG")); // = EIQ
738 dnaseqs.add(new Sequence("EMBL|A55555", "GAGATTCAG")); // = EIQ
739 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[5]));
740 cdna.setDataset(null);
742 // Xref A22222 to V12345 (should get mapped)
743 dnaseqs.get(1).addDBRef(new DBRefEntry("UNIPROT", "1", "V12345"));
744 // Xref V12345 to A44444 (should get mapped)
745 protseqs.get(0).addDBRef(new DBRefEntry("EMBL", "1", "A44444"));
746 // Xref A33333 to V12347 (sequence mismatch - should not get mapped)
747 dnaseqs.get(2).addDBRef(new DBRefEntry("UNIPROT", "1", "V12347"));
748 // as V12345 is mapped to A22222 and A44444, this leaves V12346 unmapped.
749 // it should get paired up with the unmapped A33333
750 // A11111 should be mapped to V12347
751 // A55555 is spare and has no xref so is not mapped
753 assertTrue(AlignmentUtils.mapProteinToCdna(protein, cdna));
755 // 4 protein mappings made for 3 proteins, 2 to V12345, 1 each to V12346/7
756 assertEquals(3, protein.getCodonFrames().size());
757 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
758 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
759 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
761 // one mapping for each of the first 4 cDNA sequences
762 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(0)).size());
763 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(1)).size());
764 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(2)).size());
765 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(3)).size());
767 // V12345 mapped to A22222 and A44444
768 AlignedCodonFrame acf = protein.getCodonFrame(
769 protein.getSequenceAt(0)).get(0);
770 assertEquals(2, acf.getdnaSeqs().length);
771 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
772 acf.getdnaSeqs()[0]);
773 assertEquals(cdna.getSequenceAt(3).getDatasetSequence(),
774 acf.getdnaSeqs()[1]);
776 // V12346 mapped to A33333
777 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
778 assertEquals(1, acf.getdnaSeqs().length);
779 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
780 acf.getdnaSeqs()[0]);
782 // V12347 mapped to A11111
783 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
784 assertEquals(1, acf.getdnaSeqs().length);
785 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
786 acf.getdnaSeqs()[0]);
788 // no mapping involving the 'extra' A55555
789 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(4)).isEmpty());
793 * Test mapping of protein to cDNA, for the case where we have some sequence
794 * cross-references. Verify that once we have made an xref mapping we don't
795 * also map un-xrefd sequeces.
797 * @throws IOException
799 @Test(groups ={ "Functional" })
800 public void testMapProteinToCdna_prioritiseXrefs() throws IOException
802 List<SequenceI> protseqs = new ArrayList<SequenceI>();
803 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
804 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
805 AlignmentI protein = new Alignment(
806 protseqs.toArray(new SequenceI[protseqs.size()]));
807 protein.setDataset(null);
809 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
810 dnaseqs.add(new Sequence("EMBL|A11111", "GAAATCCAG")); // = EIQ
811 dnaseqs.add(new Sequence("EMBL|A22222", "GAAATTCAG")); // = EIQ
812 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[dnaseqs
814 cdna.setDataset(null);
816 // Xref A22222 to V12345 (should get mapped)
817 // A11111 should then be mapped to the unmapped V12346
818 dnaseqs.get(1).addDBRef(new DBRefEntry("UNIPROT", "1", "V12345"));
820 assertTrue(AlignmentUtils.mapProteinToCdna(protein, cdna));
822 // 2 protein mappings made
823 assertEquals(2, protein.getCodonFrames().size());
824 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
825 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
827 // one mapping for each of the cDNA sequences
828 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(0)).size());
829 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(1)).size());
831 // V12345 mapped to A22222
832 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
834 assertEquals(1, acf.getdnaSeqs().length);
835 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
836 acf.getdnaSeqs()[0]);
838 // V12346 mapped to A11111
839 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
840 assertEquals(1, acf.getdnaSeqs().length);
841 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
842 acf.getdnaSeqs()[0]);
846 * Test the method that shows or hides sequence annotations by type(s) and
849 @Test(groups ={ "Functional" })
850 public void testShowOrHideSequenceAnnotations()
852 SequenceI seq1 = new Sequence("Seq1", "AAA");
853 SequenceI seq2 = new Sequence("Seq2", "BBB");
854 SequenceI seq3 = new Sequence("Seq3", "CCC");
855 Annotation[] anns = new Annotation[]
856 { new Annotation(2f) };
857 AlignmentAnnotation ann1 = new AlignmentAnnotation("Structure", "ann1",
859 ann1.setSequenceRef(seq1);
860 AlignmentAnnotation ann2 = new AlignmentAnnotation("Structure", "ann2",
862 ann2.setSequenceRef(seq2);
863 AlignmentAnnotation ann3 = new AlignmentAnnotation("Structure", "ann3",
865 AlignmentAnnotation ann4 = new AlignmentAnnotation("Temp", "ann4", anns);
866 ann4.setSequenceRef(seq1);
867 AlignmentAnnotation ann5 = new AlignmentAnnotation("Temp", "ann5", anns);
868 ann5.setSequenceRef(seq2);
869 AlignmentAnnotation ann6 = new AlignmentAnnotation("Temp", "ann6", anns);
870 AlignmentI al = new Alignment(new SequenceI[] {seq1, seq2, seq3});
871 al.addAnnotation(ann1); // Structure for Seq1
872 al.addAnnotation(ann2); // Structure for Seq2
873 al.addAnnotation(ann3); // Structure for no sequence
874 al.addAnnotation(ann4); // Temp for seq1
875 al.addAnnotation(ann5); // Temp for seq2
876 al.addAnnotation(ann6); // Temp for no sequence
877 List<String> types = new ArrayList<String>();
878 List<SequenceI> scope = new ArrayList<SequenceI>();
881 * Set all sequence related Structure to hidden (ann1, ann2)
883 types.add("Structure");
884 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, false,
886 assertFalse(ann1.visible);
887 assertFalse(ann2.visible);
888 assertTrue(ann3.visible); // not sequence-related, not affected
889 assertTrue(ann4.visible); // not Structure, not affected
890 assertTrue(ann5.visible); // "
891 assertTrue(ann6.visible); // not sequence-related, not affected
894 * Set Temp in {seq1, seq3} to hidden
900 AlignmentUtils.showOrHideSequenceAnnotations(al, types, scope, false,
902 assertFalse(ann1.visible); // unchanged
903 assertFalse(ann2.visible); // unchanged
904 assertTrue(ann3.visible); // not sequence-related, not affected
905 assertFalse(ann4.visible); // Temp for seq1 hidden
906 assertTrue(ann5.visible); // not in scope, not affected
907 assertTrue(ann6.visible); // not sequence-related, not affected
910 * Set Temp in all sequences to hidden
916 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, false,
918 assertFalse(ann1.visible); // unchanged
919 assertFalse(ann2.visible); // unchanged
920 assertTrue(ann3.visible); // not sequence-related, not affected
921 assertFalse(ann4.visible); // Temp for seq1 hidden
922 assertFalse(ann5.visible); // Temp for seq2 hidden
923 assertTrue(ann6.visible); // not sequence-related, not affected
926 * Set all types in {seq1, seq3} to visible
932 AlignmentUtils.showOrHideSequenceAnnotations(al, types, scope, true,
934 assertTrue(ann1.visible); // Structure for seq1 set visible
935 assertFalse(ann2.visible); // not in scope, unchanged
936 assertTrue(ann3.visible); // not sequence-related, not affected
937 assertTrue(ann4.visible); // Temp for seq1 set visible
938 assertFalse(ann5.visible); // not in scope, unchanged
939 assertTrue(ann6.visible); // not sequence-related, not affected
942 * Set all types in all scope to hidden
944 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, true,
946 assertFalse(ann1.visible);
947 assertFalse(ann2.visible);
948 assertTrue(ann3.visible); // not sequence-related, not affected
949 assertFalse(ann4.visible);
950 assertFalse(ann5.visible);
951 assertTrue(ann6.visible); // not sequence-related, not affected
955 * Tests for the method that checks if one sequence cross-references another
957 @Test(groups ={ "Functional" })
958 public void testHasCrossRef()
960 assertFalse(AlignmentUtils.hasCrossRef(null, null));
961 SequenceI seq1 = new Sequence("EMBL|A12345", "ABCDEF");
962 assertFalse(AlignmentUtils.hasCrossRef(seq1, null));
963 assertFalse(AlignmentUtils.hasCrossRef(null, seq1));
964 SequenceI seq2 = new Sequence("UNIPROT|V20192", "ABCDEF");
965 assertFalse(AlignmentUtils.hasCrossRef(seq1, seq2));
968 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "v20193"));
969 assertFalse(AlignmentUtils.hasCrossRef(seq1, seq2));
971 // case-insensitive; version number is ignored
972 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "v20192"));
973 assertTrue(AlignmentUtils.hasCrossRef(seq1, seq2));
976 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
977 assertTrue(AlignmentUtils.hasCrossRef(seq1, seq2));
978 // test is one-way only
979 assertFalse(AlignmentUtils.hasCrossRef(seq2, seq1));
983 * Tests for the method that checks if either sequence cross-references the
986 @Test(groups ={ "Functional" })
987 public void testHaveCrossRef()
989 assertFalse(AlignmentUtils.hasCrossRef(null, null));
990 SequenceI seq1 = new Sequence("EMBL|A12345", "ABCDEF");
991 assertFalse(AlignmentUtils.haveCrossRef(seq1, null));
992 assertFalse(AlignmentUtils.haveCrossRef(null, seq1));
993 SequenceI seq2 = new Sequence("UNIPROT|V20192", "ABCDEF");
994 assertFalse(AlignmentUtils.haveCrossRef(seq1, seq2));
996 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
997 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
998 // next is true for haveCrossRef, false for hasCrossRef
999 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
1001 // now the other way round
1002 seq1.setDBRef(null);
1003 seq2.addDBRef(new DBRefEntry("EMBL", "1", "A12345"));
1004 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
1005 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
1008 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
1009 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
1010 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
1014 * Test the method that extracts the exon-only part of a dna alignment.
1016 @Test(groups ={ "Functional" })
1017 public void testMakeExonAlignment()
1019 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
1020 SequenceI dna2 = new Sequence("dna2", "GGGcccTTTaaaCCC");
1021 SequenceI pep1 = new Sequence("pep1", "GF");
1022 SequenceI pep2 = new Sequence("pep2", "GFP");
1023 dna1.createDatasetSequence();
1024 dna2.createDatasetSequence();
1025 pep1.createDatasetSequence();
1026 pep2.createDatasetSequence();
1028 Set<AlignedCodonFrame> mappings = new HashSet<AlignedCodonFrame>();
1029 MapList map = new MapList(new int[]
1030 { 4, 6, 10, 12 }, new int[]
1032 AlignedCodonFrame acf = new AlignedCodonFrame();
1033 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
1035 map = new MapList(new int[]
1036 { 1, 3, 7, 9, 13, 15 }, new int[]
1038 acf = new AlignedCodonFrame();
1039 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map);
1042 AlignmentI exons = AlignmentUtils.makeExonAlignment(new SequenceI[]
1043 { dna1, dna2 }, mappings);
1044 assertEquals(2, exons.getSequences().size());
1045 assertEquals("GGGTTT", exons.getSequenceAt(0).getSequenceAsString());
1046 assertEquals("GGGTTTCCC", exons.getSequenceAt(1).getSequenceAsString());
1049 * Verify updated mappings
1051 assertEquals(2, mappings.size());
1054 * Mapping from pep1 to GGGTTT in first new exon sequence
1056 List<AlignedCodonFrame> pep1Mapping = MappingUtils
1057 .findMappingsForSequence(pep1, mappings);
1058 assertEquals(1, pep1Mapping.size());
1060 SearchResults sr = MappingUtils.buildSearchResults(pep1, 1, mappings);
1061 assertEquals(1, sr.getResults().size());
1062 Match m = sr.getResults().get(0);
1063 assertEquals(exons.getSequenceAt(0).getDatasetSequence(),
1065 assertEquals(1, m.getStart());
1066 assertEquals(3, m.getEnd());
1068 sr = MappingUtils.buildSearchResults(pep1, 2, mappings);
1069 m = sr.getResults().get(0);
1070 assertEquals(exons.getSequenceAt(0).getDatasetSequence(),
1072 assertEquals(4, m.getStart());
1073 assertEquals(6, m.getEnd());
1076 * Mapping from pep2 to GGGTTTCCC in second new exon sequence
1078 List<AlignedCodonFrame> pep2Mapping = MappingUtils
1079 .findMappingsForSequence(pep2, mappings);
1080 assertEquals(1, pep2Mapping.size());
1082 sr = MappingUtils.buildSearchResults(pep2, 1, mappings);
1083 assertEquals(1, sr.getResults().size());
1084 m = sr.getResults().get(0);
1085 assertEquals(exons.getSequenceAt(1).getDatasetSequence(),
1087 assertEquals(1, m.getStart());
1088 assertEquals(3, m.getEnd());
1090 sr = MappingUtils.buildSearchResults(pep2, 2, mappings);
1091 m = sr.getResults().get(0);
1092 assertEquals(exons.getSequenceAt(1).getDatasetSequence(),
1094 assertEquals(4, m.getStart());
1095 assertEquals(6, m.getEnd());
1097 sr = MappingUtils.buildSearchResults(pep2, 3, mappings);
1098 m = sr.getResults().get(0);
1099 assertEquals(exons.getSequenceAt(1).getDatasetSequence(),
1101 assertEquals(7, m.getStart());
1102 assertEquals(9, m.getEnd());
1106 * Test the method that makes an exon-only sequence from a DNA sequence and
1107 * its product mapping. Test includes the expected case that the DNA sequence
1108 * already has a protein product (Uniprot translation) which in turn has an
1109 * x-ref to the EMBLCDS record.
1111 @Test(groups ={ "Functional" })
1112 public void testMakeExonSequences()
1114 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
1115 SequenceI pep1 = new Sequence("pep1", "GF");
1116 dna1.createDatasetSequence();
1117 pep1.createDatasetSequence();
1118 pep1.getDatasetSequence().addDBRef(
1119 new DBRefEntry("EMBLCDS", "2", "A12345"));
1122 * Make the mapping from dna to protein. The protein sequence has a DBRef to
1125 Set<AlignedCodonFrame> mappings = new HashSet<AlignedCodonFrame>();
1126 MapList map = new MapList(new int[]
1127 { 4, 6, 10, 12 }, new int[]
1129 AlignedCodonFrame acf = new AlignedCodonFrame();
1130 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
1133 AlignedCodonFrame newMapping = new AlignedCodonFrame();
1134 List<SequenceI> exons = AlignmentUtils.makeExonSequences(dna1, acf,
1136 assertEquals(1, exons.size());
1137 SequenceI exon = exons.get(0);
1139 assertEquals("GGGTTT", exon.getSequenceAsString());
1140 assertEquals("dna1|A12345", exon.getName());
1141 assertEquals(1, exon.getDBRef().length);
1142 DBRefEntry cdsRef = exon.getDBRef()[0];
1143 assertEquals("EMBLCDS", cdsRef.getSource());
1144 assertEquals("2", cdsRef.getVersion());
1145 assertEquals("A12345", cdsRef.getAccessionId());
1149 * Test the method that makes an exon-only alignment from a DNA sequence and
1150 * its product mappings, for the case where there are multiple exon mappings
1151 * to different protein products.
1153 @Test(groups ={ "Functional" })
1154 public void testMakeExonAlignment_multipleProteins()
1156 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
1157 SequenceI pep1 = new Sequence("pep1", "GF"); // GGGTTT
1158 SequenceI pep2 = new Sequence("pep2", "KP"); // aaaccc
1159 SequenceI pep3 = new Sequence("pep3", "KF"); // aaaTTT
1160 dna1.createDatasetSequence();
1161 pep1.createDatasetSequence();
1162 pep2.createDatasetSequence();
1163 pep3.createDatasetSequence();
1164 pep1.getDatasetSequence().addDBRef(
1165 new DBRefEntry("EMBLCDS", "2", "A12345"));
1166 pep2.getDatasetSequence().addDBRef(
1167 new DBRefEntry("EMBLCDS", "3", "A12346"));
1168 pep3.getDatasetSequence().addDBRef(
1169 new DBRefEntry("EMBLCDS", "4", "A12347"));
1172 * Make the mappings from dna to protein. Using LinkedHashset is a
1173 * convenience so results are in the input order. There is no assertion that
1174 * the generated exon sequences are in any particular order.
1176 Set<AlignedCodonFrame> mappings = new LinkedHashSet<AlignedCodonFrame>();
1177 // map ...GGG...TTT to GF
1178 MapList map = new MapList(new int[]
1179 { 4, 6, 10, 12 }, new int[]
1181 AlignedCodonFrame acf = new AlignedCodonFrame();
1182 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
1185 // map aaa...ccc to KP
1186 map = new MapList(new int[]
1187 { 1, 3, 7, 9 }, new int[]
1189 acf = new AlignedCodonFrame();
1190 acf.addMap(dna1.getDatasetSequence(), pep2.getDatasetSequence(), map);
1193 // map aaa......TTT to KF
1194 map = new MapList(new int[]
1195 { 1, 3, 10, 12 }, new int[]
1197 acf = new AlignedCodonFrame();
1198 acf.addMap(dna1.getDatasetSequence(), pep3.getDatasetSequence(), map);
1202 * Create the Exon alignment; also replaces the dna-to-protein mappings with
1203 * exon-to-protein and exon-to-dna mappings
1205 AlignmentI exal = AlignmentUtils.makeExonAlignment(new SequenceI[]
1206 { dna1 }, mappings);
1209 * Verify we have 3 exon sequences, mapped to pep1/2/3 respectively
1211 List<SequenceI> exons = exal.getSequences();
1212 assertEquals(3, exons.size());
1214 SequenceI exon = exons.get(0);
1215 assertEquals("GGGTTT", exon.getSequenceAsString());
1216 assertEquals("dna1|A12345", exon.getName());
1217 assertEquals(1, exon.getDBRef().length);
1218 DBRefEntry cdsRef = exon.getDBRef()[0];
1219 assertEquals("EMBLCDS", cdsRef.getSource());
1220 assertEquals("2", cdsRef.getVersion());
1221 assertEquals("A12345", cdsRef.getAccessionId());
1223 exon = exons.get(1);
1224 assertEquals("aaaccc", exon.getSequenceAsString());
1225 assertEquals("dna1|A12346", exon.getName());
1226 assertEquals(1, exon.getDBRef().length);
1227 cdsRef = exon.getDBRef()[0];
1228 assertEquals("EMBLCDS", cdsRef.getSource());
1229 assertEquals("3", cdsRef.getVersion());
1230 assertEquals("A12346", cdsRef.getAccessionId());
1232 exon = exons.get(2);
1233 assertEquals("aaaTTT", exon.getSequenceAsString());
1234 assertEquals("dna1|A12347", exon.getName());
1235 assertEquals(1, exon.getDBRef().length);
1236 cdsRef = exon.getDBRef()[0];
1237 assertEquals("EMBLCDS", cdsRef.getSource());
1238 assertEquals("4", cdsRef.getVersion());
1239 assertEquals("A12347", cdsRef.getAccessionId());
1242 * Verify there are mappings from each exon sequence to its protein product
1243 * and also to its dna source
1245 Iterator<AlignedCodonFrame> newMappingsIterator = mappings.iterator();
1247 // mappings for dna1 - exon1 - pep1
1248 AlignedCodonFrame exonMapping = newMappingsIterator.next();
1249 List<Mapping> dnaMappings = exonMapping.getMappingsForSequence(dna1);
1250 assertEquals(1, dnaMappings.size());
1251 assertSame(exons.get(0).getDatasetSequence(), dnaMappings.get(0)
1253 assertEquals("G(1) in CDS should map to G(4) in DNA", 4, dnaMappings
1254 .get(0).getMap().getToPosition(1));
1255 List<Mapping> peptideMappings = exonMapping
1256 .getMappingsForSequence(pep1);
1257 assertEquals(1, peptideMappings.size());
1258 assertSame(pep1.getDatasetSequence(), peptideMappings.get(0).getTo());
1260 // mappings for dna1 - exon2 - pep2
1261 exonMapping = newMappingsIterator.next();
1262 dnaMappings = exonMapping.getMappingsForSequence(dna1);
1263 assertEquals(1, dnaMappings.size());
1264 assertSame(exons.get(1).getDatasetSequence(), dnaMappings.get(0)
1266 assertEquals("c(4) in CDS should map to c(7) in DNA", 7, dnaMappings
1267 .get(0).getMap().getToPosition(4));
1268 peptideMappings = exonMapping.getMappingsForSequence(pep2);
1269 assertEquals(1, peptideMappings.size());
1270 assertSame(pep2.getDatasetSequence(), peptideMappings.get(0).getTo());
1272 // mappings for dna1 - exon3 - pep3
1273 exonMapping = newMappingsIterator.next();
1274 dnaMappings = exonMapping.getMappingsForSequence(dna1);
1275 assertEquals(1, dnaMappings.size());
1276 assertSame(exons.get(2).getDatasetSequence(), dnaMappings.get(0)
1278 assertEquals("T(4) in CDS should map to T(10) in DNA", 10, dnaMappings
1279 .get(0).getMap().getToPosition(4));
1280 peptideMappings = exonMapping.getMappingsForSequence(pep3);
1281 assertEquals(1, peptideMappings.size());
1282 assertSame(pep3.getDatasetSequence(), peptideMappings.get(0).getTo());
1285 @Test(groups = { "Functional" })
1286 public void testIsMappable()
1288 SequenceI dna1 = new Sequence("dna1", "cgCAGtgGT");
1289 SequenceI aa1 = new Sequence("aa1", "RSG");
1290 AlignmentI al1 = new Alignment(new SequenceI[] { dna1 });
1291 AlignmentI al2 = new Alignment(new SequenceI[] { aa1 });
1293 assertFalse(AlignmentUtils.isMappable(null, null));
1294 assertFalse(AlignmentUtils.isMappable(al1, null));
1295 assertFalse(AlignmentUtils.isMappable(null, al1));
1296 assertFalse(AlignmentUtils.isMappable(al1, al1));
1297 assertFalse(AlignmentUtils.isMappable(al2, al2));
1299 assertTrue(AlignmentUtils.isMappable(al1, al2));
1300 assertTrue(AlignmentUtils.isMappable(al2, al1));