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;
23 import static org.testng.AssertJUnit.assertEquals;
24 import static org.testng.AssertJUnit.assertFalse;
25 import static org.testng.AssertJUnit.assertNotNull;
26 import static org.testng.AssertJUnit.assertNull;
27 import static org.testng.AssertJUnit.assertSame;
28 import static org.testng.AssertJUnit.assertTrue;
30 import jalview.analysis.AlignmentUtils.DnaVariant;
31 import jalview.datamodel.AlignedCodonFrame;
32 import jalview.datamodel.Alignment;
33 import jalview.datamodel.AlignmentAnnotation;
34 import jalview.datamodel.AlignmentI;
35 import jalview.datamodel.Annotation;
36 import jalview.datamodel.DBRefEntry;
37 import jalview.datamodel.Mapping;
38 import jalview.datamodel.SearchResults;
39 import jalview.datamodel.SearchResults.Match;
40 import jalview.datamodel.Sequence;
41 import jalview.datamodel.SequenceFeature;
42 import jalview.datamodel.SequenceI;
43 import jalview.io.AppletFormatAdapter;
44 import jalview.io.FormatAdapter;
45 import jalview.util.MapList;
46 import jalview.util.MappingUtils;
48 import java.io.IOException;
49 import java.util.ArrayList;
50 import java.util.Arrays;
51 import java.util.LinkedHashMap;
52 import java.util.List;
54 import java.util.TreeMap;
56 import org.testng.annotations.Test;
58 public class AlignmentUtilsTests
60 public static Sequence ts = new Sequence("short",
61 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklm");
63 @Test(groups = { "Functional" })
64 public void testExpandContext()
66 AlignmentI al = new Alignment(new Sequence[] {});
67 for (int i = 4; i < 14; i += 2)
69 SequenceI s1 = ts.deriveSequence().getSubSequence(i, i + 7);
72 System.out.println(new AppletFormatAdapter().formatSequences("Clustal",
74 for (int flnk = -1; flnk < 25; flnk++)
76 AlignmentI exp = AlignmentUtils.expandContext(al, flnk);
77 System.out.println("\nFlank size: " + flnk);
78 System.out.println(new AppletFormatAdapter().formatSequences(
79 "Clustal", exp, true));
83 * Full expansion to complete sequences
85 for (SequenceI sq : exp.getSequences())
87 String ung = sq.getSequenceAsString().replaceAll("-+", "");
88 final String errorMsg = "Flanking sequence not the same as original dataset sequence.\n"
91 + sq.getDatasetSequence().getSequenceAsString();
92 assertTrue(errorMsg, ung.equalsIgnoreCase(sq.getDatasetSequence()
93 .getSequenceAsString()));
99 * Last sequence is fully expanded, others have leading gaps to match
101 assertTrue(exp.getSequenceAt(4).getSequenceAsString()
103 assertTrue(exp.getSequenceAt(3).getSequenceAsString()
104 .startsWith("--abc"));
105 assertTrue(exp.getSequenceAt(2).getSequenceAsString()
106 .startsWith("----abc"));
107 assertTrue(exp.getSequenceAt(1).getSequenceAsString()
108 .startsWith("------abc"));
109 assertTrue(exp.getSequenceAt(0).getSequenceAsString()
110 .startsWith("--------abc"));
116 * Test that annotations are correctly adjusted by expandContext
118 @Test(groups = { "Functional" })
119 public void testExpandContext_annotation()
121 AlignmentI al = new Alignment(new Sequence[] {});
122 SequenceI ds = new Sequence("Seq1", "ABCDEFGHI");
124 SequenceI seq1 = ds.deriveSequence().getSubSequence(3, 6);
125 al.addSequence(seq1);
128 * Annotate DEF with 4/5/6 respectively
130 Annotation[] anns = new Annotation[] { new Annotation(4),
131 new Annotation(5), new Annotation(6) };
132 AlignmentAnnotation ann = new AlignmentAnnotation("SS",
133 "secondary structure", anns);
134 seq1.addAlignmentAnnotation(ann);
137 * The annotations array should match aligned positions
139 assertEquals(3, ann.annotations.length);
140 assertEquals(4, ann.annotations[0].value, 0.001);
141 assertEquals(5, ann.annotations[1].value, 0.001);
142 assertEquals(6, ann.annotations[2].value, 0.001);
145 * Check annotation to sequence position mappings before expanding the
146 * sequence; these are set up in Sequence.addAlignmentAnnotation ->
147 * Annotation.setSequenceRef -> createSequenceMappings
149 assertNull(ann.getAnnotationForPosition(1));
150 assertNull(ann.getAnnotationForPosition(2));
151 assertNull(ann.getAnnotationForPosition(3));
152 assertEquals(4, ann.getAnnotationForPosition(4).value, 0.001);
153 assertEquals(5, ann.getAnnotationForPosition(5).value, 0.001);
154 assertEquals(6, ann.getAnnotationForPosition(6).value, 0.001);
155 assertNull(ann.getAnnotationForPosition(7));
156 assertNull(ann.getAnnotationForPosition(8));
157 assertNull(ann.getAnnotationForPosition(9));
160 * Expand the subsequence to the full sequence abcDEFghi
162 AlignmentI expanded = AlignmentUtils.expandContext(al, -1);
163 assertEquals("abcDEFghi", expanded.getSequenceAt(0)
164 .getSequenceAsString());
167 * Confirm the alignment and sequence have the same SS annotation,
168 * referencing the expanded sequence
170 ann = expanded.getSequenceAt(0).getAnnotation()[0];
171 assertSame(ann, expanded.getAlignmentAnnotation()[0]);
172 assertSame(expanded.getSequenceAt(0), ann.sequenceRef);
175 * The annotations array should have null values except for annotated
178 assertNull(ann.annotations[0]);
179 assertNull(ann.annotations[1]);
180 assertNull(ann.annotations[2]);
181 assertEquals(4, ann.annotations[3].value, 0.001);
182 assertEquals(5, ann.annotations[4].value, 0.001);
183 assertEquals(6, ann.annotations[5].value, 0.001);
184 assertNull(ann.annotations[6]);
185 assertNull(ann.annotations[7]);
186 assertNull(ann.annotations[8]);
189 * sequence position mappings should be unchanged
191 assertNull(ann.getAnnotationForPosition(1));
192 assertNull(ann.getAnnotationForPosition(2));
193 assertNull(ann.getAnnotationForPosition(3));
194 assertEquals(4, ann.getAnnotationForPosition(4).value, 0.001);
195 assertEquals(5, ann.getAnnotationForPosition(5).value, 0.001);
196 assertEquals(6, ann.getAnnotationForPosition(6).value, 0.001);
197 assertNull(ann.getAnnotationForPosition(7));
198 assertNull(ann.getAnnotationForPosition(8));
199 assertNull(ann.getAnnotationForPosition(9));
203 * Test method that returns a map of lists of sequences by sequence name.
205 * @throws IOException
207 @Test(groups = { "Functional" })
208 public void testGetSequencesByName() throws IOException
210 final String data = ">Seq1Name\nKQYL\n" + ">Seq2Name\nRFPW\n"
211 + ">Seq1Name\nABCD\n";
212 AlignmentI al = loadAlignment(data, "FASTA");
213 Map<String, List<SequenceI>> map = AlignmentUtils
214 .getSequencesByName(al);
215 assertEquals(2, map.keySet().size());
216 assertEquals(2, map.get("Seq1Name").size());
217 assertEquals("KQYL", map.get("Seq1Name").get(0).getSequenceAsString());
218 assertEquals("ABCD", map.get("Seq1Name").get(1).getSequenceAsString());
219 assertEquals(1, map.get("Seq2Name").size());
220 assertEquals("RFPW", map.get("Seq2Name").get(0).getSequenceAsString());
224 * Helper method to load an alignment and ensure dataset sequences are set up.
230 * @throws IOException
232 protected AlignmentI loadAlignment(final String data, String format)
235 AlignmentI a = new FormatAdapter().readFile(data,
236 AppletFormatAdapter.PASTE, format);
242 * Test mapping of protein to cDNA, for the case where we have no sequence
243 * cross-references, so mappings are made first-served 1-1 where sequences
246 * @throws IOException
248 @Test(groups = { "Functional" })
249 public void testMapProteinAlignmentToCdna_noXrefs() throws IOException
251 List<SequenceI> protseqs = new ArrayList<SequenceI>();
252 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
253 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
254 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
255 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
256 protein.setDataset(null);
258 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
259 dnaseqs.add(new Sequence("EMBL|A11111", "TCAGCACGC")); // = SAR
260 dnaseqs.add(new Sequence("EMBL|A22222", "GAGATACAA")); // = EIQ
261 dnaseqs.add(new Sequence("EMBL|A33333", "GAAATCCAG")); // = EIQ
262 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG")); // = EIQ
263 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[4]));
264 cdna.setDataset(null);
266 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
268 // 3 mappings made, each from 1 to 1 sequence
269 assertEquals(3, protein.getCodonFrames().size());
270 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
271 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
272 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
274 // V12345 mapped to A22222
275 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
277 assertEquals(1, acf.getdnaSeqs().length);
278 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
279 acf.getdnaSeqs()[0]);
280 Mapping[] protMappings = acf.getProtMappings();
281 assertEquals(1, protMappings.length);
282 MapList mapList = protMappings[0].getMap();
283 assertEquals(3, mapList.getFromRatio());
284 assertEquals(1, mapList.getToRatio());
285 assertTrue(Arrays.equals(new int[] { 1, 9 }, mapList.getFromRanges()
287 assertEquals(1, mapList.getFromRanges().size());
288 assertTrue(Arrays.equals(new int[] { 1, 3 },
289 mapList.getToRanges().get(0)));
290 assertEquals(1, mapList.getToRanges().size());
292 // V12346 mapped to A33333
293 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
294 assertEquals(1, acf.getdnaSeqs().length);
295 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
296 acf.getdnaSeqs()[0]);
298 // V12347 mapped to A11111
299 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
300 assertEquals(1, acf.getdnaSeqs().length);
301 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
302 acf.getdnaSeqs()[0]);
304 // no mapping involving the 'extra' A44444
305 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(3)).isEmpty());
309 * Test for the alignSequenceAs method that takes two sequences and a mapping.
311 @Test(groups = { "Functional" })
312 public void testAlignSequenceAs_withMapping_noIntrons()
314 MapList map = new MapList(new int[] { 1, 6 }, new int[] { 1, 2 }, 3, 1);
317 * No existing gaps in dna:
319 checkAlignSequenceAs("GGGAAA", "-A-L-", false, false, map,
323 * Now introduce gaps in dna but ignore them when realigning.
325 checkAlignSequenceAs("-G-G-G-A-A-A-", "-A-L-", false, false, map,
329 * Now include gaps in dna when realigning. First retaining 'mapped' gaps
330 * only, i.e. those within the exon region.
332 checkAlignSequenceAs("-G-G--G-A--A-A-", "-A-L-", true, false, map,
333 "---G-G--G---A--A-A");
336 * Include all gaps in dna when realigning (within and without the exon
337 * region). The leading gap, and the gaps between codons, are subsumed by
338 * the protein alignment gap.
340 checkAlignSequenceAs("-G-GG--AA-A---", "-A-L-", true, true, map,
341 "---G-GG---AA-A---");
344 * Include only unmapped gaps in dna when realigning (outside the exon
345 * region). The leading gap, and the gaps between codons, are subsumed by
346 * the protein alignment gap.
348 checkAlignSequenceAs("-G-GG--AA-A-", "-A-L-", false, true, map,
353 * Test for the alignSequenceAs method that takes two sequences and a mapping.
355 @Test(groups = { "Functional" })
356 public void testAlignSequenceAs_withMapping_withIntrons()
359 * Exons at codon 2 (AAA) and 4 (TTT)
361 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
362 new int[] { 1, 2 }, 3, 1);
365 * Simple case: no gaps in dna
367 checkAlignSequenceAs("GGGAAACCCTTTGGG", "--A-L-", false, false, map,
368 "GGG---AAACCCTTTGGG");
371 * Add gaps to dna - but ignore when realigning.
373 checkAlignSequenceAs("-G-G-G--A--A---AC-CC-T-TT-GG-G-", "--A-L-",
374 false, false, map, "GGG---AAACCCTTTGGG");
377 * Add gaps to dna - include within exons only when realigning.
379 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
380 true, false, map, "GGG---A--A---ACCCT-TTGGG");
383 * Include gaps outside exons only when realigning.
385 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
386 false, true, map, "-G-G-GAAAC-CCTTT-GG-G-");
389 * Include gaps following first intron if we are 'preserving mapped gaps'
391 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
392 true, true, map, "-G-G-G--A--A---A-C-CC-T-TT-GG-G-");
395 * Include all gaps in dna when realigning.
397 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
398 true, true, map, "-G-G-G--A--A---A-C-CC-T-TT-GG-G-");
402 * Test for the case where not all of the protein sequence is mapped to cDNA.
404 @Test(groups = { "Functional" })
405 public void testAlignSequenceAs_withMapping_withUnmappedProtein()
408 * Exons at codon 2 (AAA) and 4 (TTT) mapped to A and P
410 final MapList map = new MapList(new int[] { 4, 6, 10, 12 }, new int[] {
414 * -L- 'aligns' ccc------
416 checkAlignSequenceAs("gggAAAcccTTTggg", "-A-L-P-", false, false, map,
417 "gggAAAccc------TTTggg");
421 * Helper method that performs and verifies the method under test.
424 * the sequence to be realigned
426 * the sequence whose alignment is to be copied
427 * @param preserveMappedGaps
428 * @param preserveUnmappedGaps
432 protected void checkAlignSequenceAs(final String alignee,
433 final String alignModel, final boolean preserveMappedGaps,
434 final boolean preserveUnmappedGaps, MapList map,
435 final String expected)
437 SequenceI alignMe = new Sequence("Seq1", alignee);
438 alignMe.createDatasetSequence();
439 SequenceI alignFrom = new Sequence("Seq2", alignModel);
440 alignFrom.createDatasetSequence();
441 AlignedCodonFrame acf = new AlignedCodonFrame();
442 acf.addMap(alignMe.getDatasetSequence(),
443 alignFrom.getDatasetSequence(), map);
445 AlignmentUtils.alignSequenceAs(alignMe, alignFrom, acf, "---", '-',
446 preserveMappedGaps, preserveUnmappedGaps);
447 assertEquals(expected, alignMe.getSequenceAsString());
451 * Test for the alignSequenceAs method where we preserve gaps in introns only.
453 @Test(groups = { "Functional" })
454 public void testAlignSequenceAs_keepIntronGapsOnly()
458 * Intron GGGAAA followed by exon CCCTTT
460 MapList map = new MapList(new int[] { 7, 12 }, new int[] { 1, 2 }, 3, 1);
462 checkAlignSequenceAs("GG-G-AA-A-C-CC-T-TT", "AL", false, true, map,
467 * Test the method that realigns protein to match mapped codon alignment.
469 @Test(groups = { "Functional" })
470 public void testAlignProteinAsDna()
472 // seq1 codons are [1,2,3] [4,5,6] [7,8,9] [10,11,12]
473 SequenceI dna1 = new Sequence("Seq1", "TGCCATTACCAG-");
474 // seq2 codons are [1,3,4] [5,6,7] [8,9,10] [11,12,13]
475 SequenceI dna2 = new Sequence("Seq2", "T-GCCATTACCAG");
476 // seq3 codons are [1,2,3] [4,5,7] [8,9,10] [11,12,13]
477 SequenceI dna3 = new Sequence("Seq3", "TGCCA-TTACCAG");
478 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
479 dna.setDataset(null);
481 // protein alignment will be realigned like dna
482 SequenceI prot1 = new Sequence("Seq1", "CHYQ");
483 SequenceI prot2 = new Sequence("Seq2", "CHYQ");
484 SequenceI prot3 = new Sequence("Seq3", "CHYQ");
485 SequenceI prot4 = new Sequence("Seq4", "R-QSV"); // unmapped, unchanged
486 AlignmentI protein = new Alignment(new SequenceI[] { prot1, prot2,
488 protein.setDataset(null);
490 MapList map = new MapList(new int[] { 1, 12 }, new int[] { 1, 4 }, 3, 1);
491 AlignedCodonFrame acf = new AlignedCodonFrame();
492 acf.addMap(dna1.getDatasetSequence(), prot1.getDatasetSequence(), map);
493 acf.addMap(dna2.getDatasetSequence(), prot2.getDatasetSequence(), map);
494 acf.addMap(dna3.getDatasetSequence(), prot3.getDatasetSequence(), map);
495 ArrayList<AlignedCodonFrame> acfs = new ArrayList<AlignedCodonFrame>();
497 protein.setCodonFrames(acfs);
500 * Translated codon order is [1,2,3] [1,3,4] [4,5,6] [4,5,7] [5,6,7] [7,8,9]
501 * [8,9,10] [10,11,12] [11,12,13]
503 AlignmentUtils.alignProteinAsDna(protein, dna);
504 assertEquals("C-H--Y-Q-", prot1.getSequenceAsString());
505 assertEquals("-C--H-Y-Q", prot2.getSequenceAsString());
506 assertEquals("C--H--Y-Q", prot3.getSequenceAsString());
507 assertEquals("R-QSV", prot4.getSequenceAsString());
511 * Test the method that tests whether a CDNA sequence translates to a protein
514 @Test(groups = { "Functional" })
515 public void testTranslatesAs()
517 // null arguments check
518 assertFalse(AlignmentUtils.translatesAs(null, 0, null));
519 assertFalse(AlignmentUtils.translatesAs(new char[] { 't' }, 0, null));
520 assertFalse(AlignmentUtils.translatesAs(null, 0, new char[] { 'a' }));
522 // straight translation
523 assertTrue(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(), 0,
524 "FPKG".toCharArray()));
525 // with extra start codon (not in protein)
526 assertTrue(AlignmentUtils.translatesAs("atgtttcccaaaggg".toCharArray(),
527 3, "FPKG".toCharArray()));
528 // with stop codon1 (not in protein)
529 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtaa".toCharArray(),
530 0, "FPKG".toCharArray()));
531 // with stop codon1 (in protein as *)
532 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtaa".toCharArray(),
533 0, "FPKG*".toCharArray()));
534 // with stop codon2 (not in protein)
535 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtag".toCharArray(),
536 0, "FPKG".toCharArray()));
537 // with stop codon3 (not in protein)
538 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtga".toCharArray(),
539 0, "FPKG".toCharArray()));
540 // with start and stop codon1
541 assertTrue(AlignmentUtils.translatesAs(
542 "atgtttcccaaagggtaa".toCharArray(), 3, "FPKG".toCharArray()));
543 // with start and stop codon1 (in protein as *)
544 assertTrue(AlignmentUtils.translatesAs(
545 "atgtttcccaaagggtaa".toCharArray(), 3, "FPKG*".toCharArray()));
546 // with start and stop codon2
547 assertTrue(AlignmentUtils.translatesAs(
548 "atgtttcccaaagggtag".toCharArray(), 3, "FPKG".toCharArray()));
549 // with start and stop codon3
550 assertTrue(AlignmentUtils.translatesAs(
551 "atgtttcccaaagggtga".toCharArray(), 3, "FPKG".toCharArray()));
553 // with embedded stop codons
554 assertTrue(AlignmentUtils.translatesAs(
555 "atgtttTAGcccaaaTAAgggtga".toCharArray(), 3,
556 "F*PK*G".toCharArray()));
559 assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(),
560 0, "FPMG".toCharArray()));
563 assertFalse(AlignmentUtils.translatesAs("tttcccaaagg".toCharArray(), 0,
564 "FPKG".toCharArray()));
567 assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(),
568 0, "FPK".toCharArray()));
570 // overlong dna (doesn't end in stop codon)
571 assertFalse(AlignmentUtils.translatesAs(
572 "tttcccaaagggttt".toCharArray(), 0, "FPKG".toCharArray()));
574 // dna + stop codon + more
575 assertFalse(AlignmentUtils.translatesAs(
576 "tttcccaaagggttaga".toCharArray(), 0, "FPKG".toCharArray()));
579 assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(),
580 0, "FPKGQ".toCharArray()));
584 * Test mapping of protein to cDNA, for cases where the cDNA has start and/or
585 * stop codons in addition to the protein coding sequence.
587 * @throws IOException
589 @Test(groups = { "Functional" })
590 public void testMapProteinAlignmentToCdna_withStartAndStopCodons()
593 List<SequenceI> protseqs = new ArrayList<SequenceI>();
594 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
595 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
596 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
597 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
598 protein.setDataset(null);
600 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
602 dnaseqs.add(new Sequence("EMBL|A11111", "ATGTCAGCACGC"));
604 dnaseqs.add(new Sequence("EMBL|A22222", "GAGATACAATAA"));
605 // = start +EIQ + stop
606 dnaseqs.add(new Sequence("EMBL|A33333", "ATGGAAATCCAGTAG"));
607 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG"));
608 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[4]));
609 cdna.setDataset(null);
611 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
613 // 3 mappings made, each from 1 to 1 sequence
614 assertEquals(3, protein.getCodonFrames().size());
615 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
616 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
617 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
619 // V12345 mapped from A22222
620 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
622 assertEquals(1, acf.getdnaSeqs().length);
623 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
624 acf.getdnaSeqs()[0]);
625 Mapping[] protMappings = acf.getProtMappings();
626 assertEquals(1, protMappings.length);
627 MapList mapList = protMappings[0].getMap();
628 assertEquals(3, mapList.getFromRatio());
629 assertEquals(1, mapList.getToRatio());
630 assertTrue(Arrays.equals(new int[] { 1, 9 }, mapList.getFromRanges()
632 assertEquals(1, mapList.getFromRanges().size());
633 assertTrue(Arrays.equals(new int[] { 1, 3 },
634 mapList.getToRanges().get(0)));
635 assertEquals(1, mapList.getToRanges().size());
637 // V12346 mapped from A33333 starting position 4
638 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
639 assertEquals(1, acf.getdnaSeqs().length);
640 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
641 acf.getdnaSeqs()[0]);
642 protMappings = acf.getProtMappings();
643 assertEquals(1, protMappings.length);
644 mapList = protMappings[0].getMap();
645 assertEquals(3, mapList.getFromRatio());
646 assertEquals(1, mapList.getToRatio());
647 assertTrue(Arrays.equals(new int[] { 4, 12 }, mapList.getFromRanges()
649 assertEquals(1, mapList.getFromRanges().size());
650 assertTrue(Arrays.equals(new int[] { 1, 3 },
651 mapList.getToRanges().get(0)));
652 assertEquals(1, mapList.getToRanges().size());
654 // V12347 mapped to A11111 starting position 4
655 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
656 assertEquals(1, acf.getdnaSeqs().length);
657 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
658 acf.getdnaSeqs()[0]);
659 protMappings = acf.getProtMappings();
660 assertEquals(1, protMappings.length);
661 mapList = protMappings[0].getMap();
662 assertEquals(3, mapList.getFromRatio());
663 assertEquals(1, mapList.getToRatio());
664 assertTrue(Arrays.equals(new int[] { 4, 12 }, mapList.getFromRanges()
666 assertEquals(1, mapList.getFromRanges().size());
667 assertTrue(Arrays.equals(new int[] { 1, 3 },
668 mapList.getToRanges().get(0)));
669 assertEquals(1, mapList.getToRanges().size());
671 // no mapping involving the 'extra' A44444
672 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(3)).isEmpty());
676 * Test mapping of protein to cDNA, for the case where we have some sequence
677 * cross-references. Verify that 1-to-many mappings are made where
678 * cross-references exist and sequences are mappable.
680 * @throws IOException
682 @Test(groups = { "Functional" })
683 public void testMapProteinAlignmentToCdna_withXrefs() throws IOException
685 List<SequenceI> protseqs = new ArrayList<SequenceI>();
686 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
687 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
688 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
689 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
690 protein.setDataset(null);
692 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
693 dnaseqs.add(new Sequence("EMBL|A11111", "TCAGCACGC")); // = SAR
694 dnaseqs.add(new Sequence("EMBL|A22222", "ATGGAGATACAA")); // = start + EIQ
695 dnaseqs.add(new Sequence("EMBL|A33333", "GAAATCCAG")); // = EIQ
696 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG")); // = EIQ
697 dnaseqs.add(new Sequence("EMBL|A55555", "GAGATTCAG")); // = EIQ
698 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[5]));
699 cdna.setDataset(null);
701 // Xref A22222 to V12345 (should get mapped)
702 dnaseqs.get(1).addDBRef(new DBRefEntry("UNIPROT", "1", "V12345"));
703 // Xref V12345 to A44444 (should get mapped)
704 protseqs.get(0).addDBRef(new DBRefEntry("EMBL", "1", "A44444"));
705 // Xref A33333 to V12347 (sequence mismatch - should not get mapped)
706 dnaseqs.get(2).addDBRef(new DBRefEntry("UNIPROT", "1", "V12347"));
707 // as V12345 is mapped to A22222 and A44444, this leaves V12346 unmapped.
708 // it should get paired up with the unmapped A33333
709 // A11111 should be mapped to V12347
710 // A55555 is spare and has no xref so is not mapped
712 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
714 // 4 protein mappings made for 3 proteins, 2 to V12345, 1 each to V12346/7
715 assertEquals(3, protein.getCodonFrames().size());
716 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
717 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
718 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
720 // one mapping for each of the first 4 cDNA sequences
721 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(0)).size());
722 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(1)).size());
723 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(2)).size());
724 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(3)).size());
726 // V12345 mapped to A22222 and A44444
727 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
729 assertEquals(2, acf.getdnaSeqs().length);
730 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
731 acf.getdnaSeqs()[0]);
732 assertEquals(cdna.getSequenceAt(3).getDatasetSequence(),
733 acf.getdnaSeqs()[1]);
735 // V12346 mapped to A33333
736 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
737 assertEquals(1, acf.getdnaSeqs().length);
738 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
739 acf.getdnaSeqs()[0]);
741 // V12347 mapped to A11111
742 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
743 assertEquals(1, acf.getdnaSeqs().length);
744 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
745 acf.getdnaSeqs()[0]);
747 // no mapping involving the 'extra' A55555
748 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(4)).isEmpty());
752 * Test mapping of protein to cDNA, for the case where we have some sequence
753 * cross-references. Verify that once we have made an xref mapping we don't
754 * also map un-xrefd sequeces.
756 * @throws IOException
758 @Test(groups = { "Functional" })
759 public void testMapProteinAlignmentToCdna_prioritiseXrefs()
762 List<SequenceI> protseqs = new ArrayList<SequenceI>();
763 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
764 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
765 AlignmentI protein = new Alignment(
766 protseqs.toArray(new SequenceI[protseqs.size()]));
767 protein.setDataset(null);
769 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
770 dnaseqs.add(new Sequence("EMBL|A11111", "GAAATCCAG")); // = EIQ
771 dnaseqs.add(new Sequence("EMBL|A22222", "GAAATTCAG")); // = EIQ
772 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[dnaseqs
774 cdna.setDataset(null);
776 // Xref A22222 to V12345 (should get mapped)
777 // A11111 should then be mapped to the unmapped V12346
778 dnaseqs.get(1).addDBRef(new DBRefEntry("UNIPROT", "1", "V12345"));
780 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
782 // 2 protein mappings made
783 assertEquals(2, protein.getCodonFrames().size());
784 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
785 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
787 // one mapping for each of the cDNA sequences
788 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(0)).size());
789 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(1)).size());
791 // V12345 mapped to A22222
792 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
794 assertEquals(1, acf.getdnaSeqs().length);
795 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
796 acf.getdnaSeqs()[0]);
798 // V12346 mapped to A11111
799 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
800 assertEquals(1, acf.getdnaSeqs().length);
801 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
802 acf.getdnaSeqs()[0]);
806 * Test the method that shows or hides sequence annotations by type(s) and
809 @Test(groups = { "Functional" })
810 public void testShowOrHideSequenceAnnotations()
812 SequenceI seq1 = new Sequence("Seq1", "AAA");
813 SequenceI seq2 = new Sequence("Seq2", "BBB");
814 SequenceI seq3 = new Sequence("Seq3", "CCC");
815 Annotation[] anns = new Annotation[] { new Annotation(2f) };
816 AlignmentAnnotation ann1 = new AlignmentAnnotation("Structure", "ann1",
818 ann1.setSequenceRef(seq1);
819 AlignmentAnnotation ann2 = new AlignmentAnnotation("Structure", "ann2",
821 ann2.setSequenceRef(seq2);
822 AlignmentAnnotation ann3 = new AlignmentAnnotation("Structure", "ann3",
824 AlignmentAnnotation ann4 = new AlignmentAnnotation("Temp", "ann4", anns);
825 ann4.setSequenceRef(seq1);
826 AlignmentAnnotation ann5 = new AlignmentAnnotation("Temp", "ann5", anns);
827 ann5.setSequenceRef(seq2);
828 AlignmentAnnotation ann6 = new AlignmentAnnotation("Temp", "ann6", anns);
829 AlignmentI al = new Alignment(new SequenceI[] { seq1, seq2, seq3 });
830 al.addAnnotation(ann1); // Structure for Seq1
831 al.addAnnotation(ann2); // Structure for Seq2
832 al.addAnnotation(ann3); // Structure for no sequence
833 al.addAnnotation(ann4); // Temp for seq1
834 al.addAnnotation(ann5); // Temp for seq2
835 al.addAnnotation(ann6); // Temp for no sequence
836 List<String> types = new ArrayList<String>();
837 List<SequenceI> scope = new ArrayList<SequenceI>();
840 * Set all sequence related Structure to hidden (ann1, ann2)
842 types.add("Structure");
843 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, false,
845 assertFalse(ann1.visible);
846 assertFalse(ann2.visible);
847 assertTrue(ann3.visible); // not sequence-related, not affected
848 assertTrue(ann4.visible); // not Structure, not affected
849 assertTrue(ann5.visible); // "
850 assertTrue(ann6.visible); // not sequence-related, not affected
853 * Set Temp in {seq1, seq3} to hidden
859 AlignmentUtils.showOrHideSequenceAnnotations(al, types, scope, false,
861 assertFalse(ann1.visible); // unchanged
862 assertFalse(ann2.visible); // unchanged
863 assertTrue(ann3.visible); // not sequence-related, not affected
864 assertFalse(ann4.visible); // Temp for seq1 hidden
865 assertTrue(ann5.visible); // not in scope, not affected
866 assertTrue(ann6.visible); // not sequence-related, not affected
869 * Set Temp in all sequences to hidden
875 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, false,
877 assertFalse(ann1.visible); // unchanged
878 assertFalse(ann2.visible); // unchanged
879 assertTrue(ann3.visible); // not sequence-related, not affected
880 assertFalse(ann4.visible); // Temp for seq1 hidden
881 assertFalse(ann5.visible); // Temp for seq2 hidden
882 assertTrue(ann6.visible); // not sequence-related, not affected
885 * Set all types in {seq1, seq3} to visible
891 AlignmentUtils.showOrHideSequenceAnnotations(al, types, scope, true,
893 assertTrue(ann1.visible); // Structure for seq1 set visible
894 assertFalse(ann2.visible); // not in scope, unchanged
895 assertTrue(ann3.visible); // not sequence-related, not affected
896 assertTrue(ann4.visible); // Temp for seq1 set visible
897 assertFalse(ann5.visible); // not in scope, unchanged
898 assertTrue(ann6.visible); // not sequence-related, not affected
901 * Set all types in all scope to hidden
903 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, true,
905 assertFalse(ann1.visible);
906 assertFalse(ann2.visible);
907 assertTrue(ann3.visible); // not sequence-related, not affected
908 assertFalse(ann4.visible);
909 assertFalse(ann5.visible);
910 assertTrue(ann6.visible); // not sequence-related, not affected
914 * Tests for the method that checks if one sequence cross-references another
916 @Test(groups = { "Functional" })
917 public void testHasCrossRef()
919 assertFalse(AlignmentUtils.hasCrossRef(null, null));
920 SequenceI seq1 = new Sequence("EMBL|A12345", "ABCDEF");
921 assertFalse(AlignmentUtils.hasCrossRef(seq1, null));
922 assertFalse(AlignmentUtils.hasCrossRef(null, seq1));
923 SequenceI seq2 = new Sequence("UNIPROT|V20192", "ABCDEF");
924 assertFalse(AlignmentUtils.hasCrossRef(seq1, seq2));
927 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "v20193"));
928 assertFalse(AlignmentUtils.hasCrossRef(seq1, seq2));
930 // case-insensitive; version number is ignored
931 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "v20192"));
932 assertTrue(AlignmentUtils.hasCrossRef(seq1, seq2));
935 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
936 assertTrue(AlignmentUtils.hasCrossRef(seq1, seq2));
937 // test is one-way only
938 assertFalse(AlignmentUtils.hasCrossRef(seq2, seq1));
942 * Tests for the method that checks if either sequence cross-references the
945 @Test(groups = { "Functional" })
946 public void testHaveCrossRef()
948 assertFalse(AlignmentUtils.hasCrossRef(null, null));
949 SequenceI seq1 = new Sequence("EMBL|A12345", "ABCDEF");
950 assertFalse(AlignmentUtils.haveCrossRef(seq1, null));
951 assertFalse(AlignmentUtils.haveCrossRef(null, seq1));
952 SequenceI seq2 = new Sequence("UNIPROT|V20192", "ABCDEF");
953 assertFalse(AlignmentUtils.haveCrossRef(seq1, seq2));
955 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
956 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
957 // next is true for haveCrossRef, false for hasCrossRef
958 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
960 // now the other way round
961 seq1.setDBRefs(null);
962 seq2.addDBRef(new DBRefEntry("EMBL", "1", "A12345"));
963 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
964 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
967 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
968 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
969 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
973 * Test the method that extracts the cds-only part of a dna alignment.
975 @Test(groups = { "Functional" })
976 public void testMakeCdsAlignment()
980 * dna1 --> [4, 6] [10,12] --> pep1
981 * dna2 --> [1, 3] [7, 9] [13,15] --> pep2
983 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
984 SequenceI dna2 = new Sequence("dna2", "GGGcccTTTaaaCCC");
985 SequenceI pep1 = new Sequence("pep1", "GF");
986 SequenceI pep2 = new Sequence("pep2", "GFP");
987 pep1.addDBRef(new DBRefEntry("UNIPROT", "0", "pep1"));
988 pep2.addDBRef(new DBRefEntry("UNIPROT", "0", "pep2"));
989 dna1.createDatasetSequence();
990 dna2.createDatasetSequence();
991 pep1.createDatasetSequence();
992 pep2.createDatasetSequence();
993 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
994 dna.setDataset(null);
997 * put a variant feature on dna2 base 8
998 * - should transfer to cds2 base 5
1000 dna2.addSequenceFeature(new SequenceFeature("variant", "hgmd", 8, 8,
1004 * need a sourceDbRef if we are to construct dbrefs to the CDS
1005 * sequence from the dna contig sequences
1007 DBRefEntry dbref = new DBRefEntry("ENSEMBL", "0", "dna1");
1008 dna1.getDatasetSequence().addDBRef(dbref);
1009 org.testng.Assert.assertEquals(dbref, dna1.getPrimaryDBRefs().get(0));
1010 dbref = new DBRefEntry("ENSEMBL", "0", "dna2");
1011 dna2.getDatasetSequence().addDBRef(dbref);
1012 org.testng.Assert.assertEquals(dbref, dna2.getPrimaryDBRefs().get(0));
1015 * CDS sequences are 'discovered' from dna-to-protein mappings on the alignment
1016 * dataset (e.g. added from dbrefs by CrossRef.findXrefSequences)
1018 MapList mapfordna1 = new MapList(new int[] { 4, 6, 10, 12 }, new int[] {
1020 AlignedCodonFrame acf = new AlignedCodonFrame();
1021 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(),
1023 dna.addCodonFrame(acf);
1024 MapList mapfordna2 = new MapList(new int[] { 1, 3, 7, 9, 13, 15 },
1025 new int[] { 1, 3 }, 3, 1);
1026 acf = new AlignedCodonFrame();
1027 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(),
1029 dna.addCodonFrame(acf);
1032 * In this case, mappings originally came from matching Uniprot accessions - so need an xref on dna involving those regions. These are normally constructed from CDS annotation
1034 DBRefEntry dna1xref = new DBRefEntry("UNIPROT", "ENSEMBL", "pep1",
1035 new Mapping(mapfordna1));
1036 dna1.getDatasetSequence().addDBRef(dna1xref);
1037 DBRefEntry dna2xref = new DBRefEntry("UNIPROT", "ENSEMBL", "pep2",
1038 new Mapping(mapfordna2));
1039 dna2.getDatasetSequence().addDBRef(dna2xref);
1042 * execute method under test:
1044 AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
1045 dna1, dna2 }, dna.getDataset(), null);
1048 * verify cds sequences
1050 assertEquals(2, cds.getSequences().size());
1051 assertEquals("GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
1052 assertEquals("GGGTTTCCC", cds.getSequenceAt(1).getSequenceAsString());
1055 * verify shared, extended alignment dataset
1057 assertSame(dna.getDataset(), cds.getDataset());
1058 SequenceI cds1Dss = cds.getSequenceAt(0).getDatasetSequence();
1059 SequenceI cds2Dss = cds.getSequenceAt(1).getDatasetSequence();
1060 assertTrue(dna.getDataset().getSequences().contains(cds1Dss));
1061 assertTrue(dna.getDataset().getSequences().contains(cds2Dss));
1064 * verify CDS has a dbref with mapping to peptide
1066 assertNotNull(cds1Dss.getDBRefs());
1067 assertEquals(2, cds1Dss.getDBRefs().length);
1068 dbref = cds1Dss.getDBRefs()[0];
1069 assertEquals(dna1xref.getSource(), dbref.getSource());
1070 // version is via ensembl's primary ref
1071 assertEquals(dna1xref.getVersion(), dbref.getVersion());
1072 assertEquals(dna1xref.getAccessionId(), dbref.getAccessionId());
1073 assertNotNull(dbref.getMap());
1074 assertSame(pep1.getDatasetSequence(), dbref.getMap().getTo());
1075 MapList cdsMapping = new MapList(new int[] { 1, 6 },
1076 new int[] { 1, 2 }, 3, 1);
1077 assertEquals(cdsMapping, dbref.getMap().getMap());
1080 * verify peptide has added a dbref with reverse mapping to CDS
1082 assertNotNull(pep1.getDBRefs());
1083 // FIXME pep1.getDBRefs() is 1 - is that the correct behaviour ?
1084 assertEquals(2, pep1.getDBRefs().length);
1085 dbref = pep1.getDBRefs()[1];
1086 assertEquals("ENSEMBL", dbref.getSource());
1087 assertEquals("0", dbref.getVersion());
1088 assertEquals("CDS|dna1", dbref.getAccessionId());
1089 assertNotNull(dbref.getMap());
1090 assertSame(cds1Dss, dbref.getMap().getTo());
1091 assertEquals(cdsMapping.getInverse(), dbref.getMap().getMap());
1094 * Verify mappings from CDS to peptide, cDNA to CDS, and cDNA to peptide
1095 * the mappings are on the shared alignment dataset
1096 * 6 mappings, 2*(DNA->CDS), 2*(DNA->Pep), 2*(CDS->Pep)
1098 List<AlignedCodonFrame> cdsMappings = cds.getDataset().getCodonFrames();
1099 assertEquals(6, cdsMappings.size());
1102 * verify that mapping sets for dna and cds alignments are different
1103 * [not current behaviour - all mappings are on the alignment dataset]
1105 // select -> subselect type to test.
1106 // Assert.assertNotSame(dna.getCodonFrames(), cds.getCodonFrames());
1107 // assertEquals(4, dna.getCodonFrames().size());
1108 // assertEquals(4, cds.getCodonFrames().size());
1111 * Two mappings involve pep1 (dna to pep1, cds to pep1)
1112 * Mapping from pep1 to GGGTTT in first new exon sequence
1114 List<AlignedCodonFrame> pep1Mappings = MappingUtils
1115 .findMappingsForSequence(pep1, cdsMappings);
1116 assertEquals(2, pep1Mappings.size());
1117 List<AlignedCodonFrame> mappings = MappingUtils
1118 .findMappingsForSequence(cds.getSequenceAt(0), pep1Mappings);
1119 assertEquals(1, mappings.size());
1122 SearchResults sr = MappingUtils.buildSearchResults(pep1, 1, mappings);
1123 assertEquals(1, sr.getResults().size());
1124 Match m = sr.getResults().get(0);
1125 assertSame(cds1Dss, m.getSequence());
1126 assertEquals(1, m.getStart());
1127 assertEquals(3, m.getEnd());
1129 sr = MappingUtils.buildSearchResults(pep1, 2, mappings);
1130 m = sr.getResults().get(0);
1131 assertSame(cds1Dss, m.getSequence());
1132 assertEquals(4, m.getStart());
1133 assertEquals(6, m.getEnd());
1136 * Two mappings involve pep2 (dna to pep2, cds to pep2)
1137 * Verify mapping from pep2 to GGGTTTCCC in second new exon sequence
1139 List<AlignedCodonFrame> pep2Mappings = MappingUtils
1140 .findMappingsForSequence(pep2, cdsMappings);
1141 assertEquals(2, pep2Mappings.size());
1142 mappings = MappingUtils.findMappingsForSequence(cds.getSequenceAt(1),
1144 assertEquals(1, mappings.size());
1146 sr = MappingUtils.buildSearchResults(pep2, 1, mappings);
1147 assertEquals(1, sr.getResults().size());
1148 m = sr.getResults().get(0);
1149 assertSame(cds2Dss, m.getSequence());
1150 assertEquals(1, m.getStart());
1151 assertEquals(3, m.getEnd());
1153 sr = MappingUtils.buildSearchResults(pep2, 2, mappings);
1154 m = sr.getResults().get(0);
1155 assertSame(cds2Dss, m.getSequence());
1156 assertEquals(4, m.getStart());
1157 assertEquals(6, m.getEnd());
1159 sr = MappingUtils.buildSearchResults(pep2, 3, mappings);
1160 m = sr.getResults().get(0);
1161 assertSame(cds2Dss, m.getSequence());
1162 assertEquals(7, m.getStart());
1163 assertEquals(9, m.getEnd());
1166 * check cds2 acquired a variant feature in position 5
1168 SequenceFeature[] sfs = cds2Dss.getSequenceFeatures();
1170 assertEquals(1, sfs.length);
1171 assertEquals("variant", sfs[0].type);
1172 assertEquals(5, sfs[0].begin);
1173 assertEquals(5, sfs[0].end);
1177 * Test the method that makes a cds-only alignment from a DNA sequence and its
1178 * product mappings, for the case where there are multiple exon mappings to
1179 * different protein products.
1181 @Test(groups = { "Functional" })
1182 public void testMakeCdsAlignment_multipleProteins()
1184 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
1185 SequenceI pep1 = new Sequence("pep1", "GF"); // GGGTTT
1186 SequenceI pep2 = new Sequence("pep2", "KP"); // aaaccc
1187 SequenceI pep3 = new Sequence("pep3", "KF"); // aaaTTT
1188 dna1.createDatasetSequence();
1189 pep1.createDatasetSequence();
1190 pep2.createDatasetSequence();
1191 pep3.createDatasetSequence();
1192 pep1.getDatasetSequence().addDBRef(
1193 new DBRefEntry("EMBLCDS", "2", "A12345"));
1194 pep2.getDatasetSequence().addDBRef(
1195 new DBRefEntry("EMBLCDS", "3", "A12346"));
1196 pep3.getDatasetSequence().addDBRef(
1197 new DBRefEntry("EMBLCDS", "4", "A12347"));
1200 * Create the CDS alignment
1202 AlignmentI dna = new Alignment(new SequenceI[] { dna1 });
1203 dna.setDataset(null);
1206 * Make the mappings from dna to protein
1208 // map ...GGG...TTT to GF
1209 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1210 new int[] { 1, 2 }, 3, 1);
1211 AlignedCodonFrame acf = new AlignedCodonFrame();
1212 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
1213 dna.addCodonFrame(acf);
1215 // map aaa...ccc to KP
1216 map = new MapList(new int[] { 1, 3, 7, 9 }, new int[] { 1, 2 }, 3, 1);
1217 acf = new AlignedCodonFrame();
1218 acf.addMap(dna1.getDatasetSequence(), pep2.getDatasetSequence(), map);
1219 dna.addCodonFrame(acf);
1221 // map aaa......TTT to KF
1222 map = new MapList(new int[] { 1, 3, 10, 12 }, new int[] { 1, 2 }, 3, 1);
1223 acf = new AlignedCodonFrame();
1224 acf.addMap(dna1.getDatasetSequence(), pep3.getDatasetSequence(), map);
1225 dna.addCodonFrame(acf);
1228 * execute method under test
1230 AlignmentI cdsal = AlignmentUtils.makeCdsAlignment(
1231 new SequenceI[] { dna1 }, dna.getDataset(), null);
1234 * Verify we have 3 cds sequences, mapped to pep1/2/3 respectively
1236 List<SequenceI> cds = cdsal.getSequences();
1237 assertEquals(3, cds.size());
1240 * verify shared, extended alignment dataset
1242 assertSame(cdsal.getDataset(), dna.getDataset());
1243 assertTrue(dna.getDataset().getSequences()
1244 .contains(cds.get(0).getDatasetSequence()));
1245 assertTrue(dna.getDataset().getSequences()
1246 .contains(cds.get(1).getDatasetSequence()));
1247 assertTrue(dna.getDataset().getSequences()
1248 .contains(cds.get(2).getDatasetSequence()));
1251 * verify aligned cds sequences and their xrefs
1253 SequenceI cdsSeq = cds.get(0);
1254 assertEquals("GGGTTT", cdsSeq.getSequenceAsString());
1255 // assertEquals("dna1|A12345", cdsSeq.getName());
1256 assertEquals("CDS|dna1", cdsSeq.getName());
1257 // assertEquals(1, cdsSeq.getDBRefs().length);
1258 // DBRefEntry cdsRef = cdsSeq.getDBRefs()[0];
1259 // assertEquals("EMBLCDS", cdsRef.getSource());
1260 // assertEquals("2", cdsRef.getVersion());
1261 // assertEquals("A12345", cdsRef.getAccessionId());
1263 cdsSeq = cds.get(1);
1264 assertEquals("aaaccc", cdsSeq.getSequenceAsString());
1265 // assertEquals("dna1|A12346", cdsSeq.getName());
1266 assertEquals("CDS|dna1", cdsSeq.getName());
1267 // assertEquals(1, cdsSeq.getDBRefs().length);
1268 // cdsRef = cdsSeq.getDBRefs()[0];
1269 // assertEquals("EMBLCDS", cdsRef.getSource());
1270 // assertEquals("3", cdsRef.getVersion());
1271 // assertEquals("A12346", cdsRef.getAccessionId());
1273 cdsSeq = cds.get(2);
1274 assertEquals("aaaTTT", cdsSeq.getSequenceAsString());
1275 // assertEquals("dna1|A12347", cdsSeq.getName());
1276 assertEquals("CDS|dna1", cdsSeq.getName());
1277 // assertEquals(1, cdsSeq.getDBRefs().length);
1278 // cdsRef = cdsSeq.getDBRefs()[0];
1279 // assertEquals("EMBLCDS", cdsRef.getSource());
1280 // assertEquals("4", cdsRef.getVersion());
1281 // assertEquals("A12347", cdsRef.getAccessionId());
1284 * Verify there are mappings from each cds sequence to its protein product
1285 * and also to its dna source
1287 List<AlignedCodonFrame> newMappings = cdsal.getCodonFrames();
1290 * 6 mappings involve dna1 (to pep1/2/3, cds1/2/3)
1292 List<AlignedCodonFrame> dnaMappings = MappingUtils
1293 .findMappingsForSequence(dna1, newMappings);
1294 assertEquals(6, dnaMappings.size());
1299 List<AlignedCodonFrame> mappings = MappingUtils
1300 .findMappingsForSequence(pep1, dnaMappings);
1301 assertEquals(1, mappings.size());
1302 assertEquals(1, mappings.get(0).getMappings().size());
1303 assertSame(pep1.getDatasetSequence(), mappings.get(0).getMappings()
1304 .get(0).getMapping().getTo());
1309 List<AlignedCodonFrame> dnaToCds1Mappings = MappingUtils
1310 .findMappingsForSequence(cds.get(0), dnaMappings);
1311 Mapping mapping = dnaToCds1Mappings.get(0).getMappings().get(0)
1313 assertSame(cds.get(0).getDatasetSequence(), mapping.getTo());
1314 assertEquals("G(1) in CDS should map to G(4) in DNA", 4, mapping
1315 .getMap().getToPosition(1));
1320 mappings = MappingUtils.findMappingsForSequence(pep2, dnaMappings);
1321 assertEquals(1, mappings.size());
1322 assertEquals(1, mappings.get(0).getMappings().size());
1323 assertSame(pep2.getDatasetSequence(), mappings.get(0).getMappings()
1324 .get(0).getMapping().getTo());
1329 List<AlignedCodonFrame> dnaToCds2Mappings = MappingUtils
1330 .findMappingsForSequence(cds.get(1), dnaMappings);
1331 mapping = dnaToCds2Mappings.get(0).getMappings().get(0).getMapping();
1332 assertSame(cds.get(1).getDatasetSequence(), mapping.getTo());
1333 assertEquals("c(4) in CDS should map to c(7) in DNA", 7, mapping
1334 .getMap().getToPosition(4));
1339 mappings = MappingUtils.findMappingsForSequence(pep3, dnaMappings);
1340 assertEquals(1, mappings.size());
1341 assertEquals(1, mappings.get(0).getMappings().size());
1342 assertSame(pep3.getDatasetSequence(), mappings.get(0).getMappings()
1343 .get(0).getMapping().getTo());
1348 List<AlignedCodonFrame> dnaToCds3Mappings = MappingUtils
1349 .findMappingsForSequence(cds.get(2), dnaMappings);
1350 mapping = dnaToCds3Mappings.get(0).getMappings().get(0).getMapping();
1351 assertSame(cds.get(2).getDatasetSequence(), mapping.getTo());
1352 assertEquals("T(4) in CDS should map to T(10) in DNA", 10, mapping
1353 .getMap().getToPosition(4));
1356 @Test(groups = { "Functional" })
1357 public void testIsMappable()
1359 SequenceI dna1 = new Sequence("dna1", "cgCAGtgGT");
1360 SequenceI aa1 = new Sequence("aa1", "RSG");
1361 AlignmentI al1 = new Alignment(new SequenceI[] { dna1 });
1362 AlignmentI al2 = new Alignment(new SequenceI[] { aa1 });
1364 assertFalse(AlignmentUtils.isMappable(null, null));
1365 assertFalse(AlignmentUtils.isMappable(al1, null));
1366 assertFalse(AlignmentUtils.isMappable(null, al1));
1367 assertFalse(AlignmentUtils.isMappable(al1, al1));
1368 assertFalse(AlignmentUtils.isMappable(al2, al2));
1370 assertTrue(AlignmentUtils.isMappable(al1, al2));
1371 assertTrue(AlignmentUtils.isMappable(al2, al1));
1375 * Test creating a mapping when the sequences involved do not start at residue
1378 * @throws IOException
1380 @Test(groups = { "Functional" })
1381 public void testMapCdnaToProtein_forSubsequence() throws IOException
1383 SequenceI prot = new Sequence("UNIPROT|V12345", "E-I--Q", 10, 12);
1384 prot.createDatasetSequence();
1386 SequenceI dna = new Sequence("EMBL|A33333", "GAA--AT-C-CAG", 40, 48);
1387 dna.createDatasetSequence();
1389 MapList map = AlignmentUtils.mapCdnaToProtein(prot, dna);
1390 assertEquals(10, map.getToLowest());
1391 assertEquals(12, map.getToHighest());
1392 assertEquals(40, map.getFromLowest());
1393 assertEquals(48, map.getFromHighest());
1397 * Test for the alignSequenceAs method where we have protein mapped to protein
1399 @Test(groups = { "Functional" })
1400 public void testAlignSequenceAs_mappedProteinProtein()
1403 SequenceI alignMe = new Sequence("Match", "MGAASEV");
1404 alignMe.createDatasetSequence();
1405 SequenceI alignFrom = new Sequence("Query", "LQTGYMGAASEVMFSPTRR");
1406 alignFrom.createDatasetSequence();
1408 AlignedCodonFrame acf = new AlignedCodonFrame();
1409 // this is like a domain or motif match of part of a peptide sequence
1410 MapList map = new MapList(new int[] { 6, 12 }, new int[] { 1, 7 }, 1, 1);
1411 acf.addMap(alignFrom.getDatasetSequence(),
1412 alignMe.getDatasetSequence(), map);
1414 AlignmentUtils.alignSequenceAs(alignMe, alignFrom, acf, "-", '-', true,
1416 assertEquals("-----MGAASEV-------", alignMe.getSequenceAsString());
1420 * Test for the alignSequenceAs method where there are trailing unmapped
1421 * residues in the model sequence
1423 @Test(groups = { "Functional" })
1424 public void testAlignSequenceAs_withTrailingPeptide()
1426 // map first 3 codons to KPF; G is a trailing unmapped residue
1427 MapList map = new MapList(new int[] { 1, 9 }, new int[] { 1, 3 }, 3, 1);
1429 checkAlignSequenceAs("AAACCCTTT", "K-PFG", true, true, map,
1434 * Tests for transferring features between mapped sequences
1436 @Test(groups = { "Functional" })
1437 public void testTransferFeatures()
1439 SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
1440 SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
1443 dna.addSequenceFeature(new SequenceFeature("type1", "desc1", 1, 2, 1f,
1445 // partial overlap - to [1, 1]
1446 dna.addSequenceFeature(new SequenceFeature("type2", "desc2", 3, 4, 2f,
1448 // exact overlap - to [1, 3]
1449 dna.addSequenceFeature(new SequenceFeature("type3", "desc3", 4, 6, 3f,
1451 // spanning overlap - to [2, 5]
1452 dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
1454 // exactly overlaps whole mapped range [1, 6]
1455 dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
1457 // no overlap (internal)
1458 dna.addSequenceFeature(new SequenceFeature("type6", "desc6", 7, 9, 6f,
1460 // no overlap (3' end)
1461 dna.addSequenceFeature(new SequenceFeature("type7", "desc7", 13, 15,
1463 // overlap (3' end) - to [6, 6]
1464 dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
1466 // extended overlap - to [6, +]
1467 dna.addSequenceFeature(new SequenceFeature("type9", "desc9", 12, 13,
1470 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1471 new int[] { 1, 6 }, 1, 1);
1474 * transferFeatures() will build 'partial overlap' for regions
1475 * that partially overlap 5' or 3' (start or end) of target sequence
1477 AlignmentUtils.transferFeatures(dna, cds, map, null);
1478 SequenceFeature[] sfs = cds.getSequenceFeatures();
1479 assertEquals(6, sfs.length);
1481 SequenceFeature sf = sfs[0];
1482 assertEquals("type2", sf.getType());
1483 assertEquals("desc2", sf.getDescription());
1484 assertEquals(2f, sf.getScore());
1485 assertEquals(1, sf.getBegin());
1486 assertEquals(1, sf.getEnd());
1489 assertEquals("type3", sf.getType());
1490 assertEquals("desc3", sf.getDescription());
1491 assertEquals(3f, sf.getScore());
1492 assertEquals(1, sf.getBegin());
1493 assertEquals(3, sf.getEnd());
1496 assertEquals("type4", sf.getType());
1497 assertEquals(2, sf.getBegin());
1498 assertEquals(5, sf.getEnd());
1501 assertEquals("type5", sf.getType());
1502 assertEquals(1, sf.getBegin());
1503 assertEquals(6, sf.getEnd());
1506 assertEquals("type8", sf.getType());
1507 assertEquals(6, sf.getBegin());
1508 assertEquals(6, sf.getEnd());
1511 assertEquals("type9", sf.getType());
1512 assertEquals(6, sf.getBegin());
1513 assertEquals(6, sf.getEnd());
1517 * Tests for transferring features between mapped sequences
1519 @Test(groups = { "Functional" })
1520 public void testTransferFeatures_withOmit()
1522 SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
1523 SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
1525 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1526 new int[] { 1, 6 }, 1, 1);
1528 // [5, 11] maps to [2, 5]
1529 dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
1531 // [4, 12] maps to [1, 6]
1532 dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
1534 // [12, 12] maps to [6, 6]
1535 dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
1538 // desc4 and desc8 are the 'omit these' varargs
1539 AlignmentUtils.transferFeatures(dna, cds, map, null, "type4", "type8");
1540 SequenceFeature[] sfs = cds.getSequenceFeatures();
1541 assertEquals(1, sfs.length);
1543 SequenceFeature sf = sfs[0];
1544 assertEquals("type5", sf.getType());
1545 assertEquals(1, sf.getBegin());
1546 assertEquals(6, sf.getEnd());
1550 * Tests for transferring features between mapped sequences
1552 @Test(groups = { "Functional" })
1553 public void testTransferFeatures_withSelect()
1555 SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
1556 SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
1558 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1559 new int[] { 1, 6 }, 1, 1);
1561 // [5, 11] maps to [2, 5]
1562 dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
1564 // [4, 12] maps to [1, 6]
1565 dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
1567 // [12, 12] maps to [6, 6]
1568 dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
1571 // "type5" is the 'select this type' argument
1572 AlignmentUtils.transferFeatures(dna, cds, map, "type5");
1573 SequenceFeature[] sfs = cds.getSequenceFeatures();
1574 assertEquals(1, sfs.length);
1576 SequenceFeature sf = sfs[0];
1577 assertEquals("type5", sf.getType());
1578 assertEquals(1, sf.getBegin());
1579 assertEquals(6, sf.getEnd());
1583 * Test the method that extracts the cds-only part of a dna alignment, for the
1584 * case where the cds should be aligned to match its nucleotide sequence.
1586 @Test(groups = { "Functional" })
1587 public void testMakeCdsAlignment_alternativeTranscripts()
1589 SequenceI dna1 = new Sequence("dna1", "aaaGGGCC-----CTTTaaaGGG");
1590 // alternative transcript of same dna skips CCC codon
1591 SequenceI dna2 = new Sequence("dna2", "aaaGGGCC-----cttTaaaGGG");
1592 // dna3 has no mapping (protein product) so should be ignored here
1593 SequenceI dna3 = new Sequence("dna3", "aaaGGGCCCCCGGGcttTaaaGGG");
1594 SequenceI pep1 = new Sequence("pep1", "GPFG");
1595 SequenceI pep2 = new Sequence("pep2", "GPG");
1596 dna1.createDatasetSequence();
1597 dna2.createDatasetSequence();
1598 dna3.createDatasetSequence();
1599 pep1.createDatasetSequence();
1600 pep2.createDatasetSequence();
1602 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
1603 dna.setDataset(null);
1605 MapList map = new MapList(new int[] { 4, 12, 16, 18 },
1606 new int[] { 1, 4 }, 3, 1);
1607 AlignedCodonFrame acf = new AlignedCodonFrame();
1608 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
1609 dna.addCodonFrame(acf);
1610 map = new MapList(new int[] { 4, 8, 12, 12, 16, 18 },
1611 new int[] { 1, 3 }, 3, 1);
1612 acf = new AlignedCodonFrame();
1613 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map);
1614 dna.addCodonFrame(acf);
1616 AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
1617 dna1, dna2, dna3 }, dna.getDataset(), null);
1618 List<SequenceI> cdsSeqs = cds.getSequences();
1619 assertEquals(2, cdsSeqs.size());
1620 assertEquals("GGGCCCTTTGGG", cdsSeqs.get(0).getSequenceAsString());
1621 assertEquals("GGGCCTGGG", cdsSeqs.get(1).getSequenceAsString());
1624 * verify shared, extended alignment dataset
1626 assertSame(dna.getDataset(), cds.getDataset());
1627 assertTrue(dna.getDataset().getSequences()
1628 .contains(cdsSeqs.get(0).getDatasetSequence()));
1629 assertTrue(dna.getDataset().getSequences()
1630 .contains(cdsSeqs.get(1).getDatasetSequence()));
1633 * Verify 6 mappings: dna1 to cds1, cds1 to pep1, dna1 to pep1
1634 * and the same for dna2/cds2/pep2
1636 List<AlignedCodonFrame> mappings = cds.getCodonFrames();
1637 assertEquals(6, mappings.size());
1640 * 2 mappings involve pep1
1642 List<AlignedCodonFrame> pep1Mappings = MappingUtils
1643 .findMappingsForSequence(pep1, mappings);
1644 assertEquals(2, pep1Mappings.size());
1647 * Get mapping of pep1 to cds1 and verify it
1648 * maps GPFG to 1-3,4-6,7-9,10-12
1650 List<AlignedCodonFrame> pep1CdsMappings = MappingUtils
1651 .findMappingsForSequence(cds.getSequenceAt(0), pep1Mappings);
1652 assertEquals(1, pep1CdsMappings.size());
1653 SearchResults sr = MappingUtils.buildSearchResults(pep1, 1,
1655 assertEquals(1, sr.getResults().size());
1656 Match m = sr.getResults().get(0);
1657 assertEquals(cds.getSequenceAt(0).getDatasetSequence(), m.getSequence());
1658 assertEquals(1, m.getStart());
1659 assertEquals(3, m.getEnd());
1660 sr = MappingUtils.buildSearchResults(pep1, 2, pep1CdsMappings);
1661 m = sr.getResults().get(0);
1662 assertEquals(4, m.getStart());
1663 assertEquals(6, m.getEnd());
1664 sr = MappingUtils.buildSearchResults(pep1, 3, pep1CdsMappings);
1665 m = sr.getResults().get(0);
1666 assertEquals(7, m.getStart());
1667 assertEquals(9, m.getEnd());
1668 sr = MappingUtils.buildSearchResults(pep1, 4, pep1CdsMappings);
1669 m = sr.getResults().get(0);
1670 assertEquals(10, m.getStart());
1671 assertEquals(12, m.getEnd());
1674 * Get mapping of pep2 to cds2 and verify it
1675 * maps GPG in pep2 to 1-3,4-6,7-9 in second CDS sequence
1677 List<AlignedCodonFrame> pep2Mappings = MappingUtils
1678 .findMappingsForSequence(pep2, mappings);
1679 assertEquals(2, pep2Mappings.size());
1680 List<AlignedCodonFrame> pep2CdsMappings = MappingUtils
1681 .findMappingsForSequence(cds.getSequenceAt(1), pep2Mappings);
1682 assertEquals(1, pep2CdsMappings.size());
1683 sr = MappingUtils.buildSearchResults(pep2, 1, pep2CdsMappings);
1684 assertEquals(1, sr.getResults().size());
1685 m = sr.getResults().get(0);
1686 assertEquals(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
1687 assertEquals(1, m.getStart());
1688 assertEquals(3, m.getEnd());
1689 sr = MappingUtils.buildSearchResults(pep2, 2, pep2CdsMappings);
1690 m = sr.getResults().get(0);
1691 assertEquals(4, m.getStart());
1692 assertEquals(6, m.getEnd());
1693 sr = MappingUtils.buildSearchResults(pep2, 3, pep2CdsMappings);
1694 m = sr.getResults().get(0);
1695 assertEquals(7, m.getStart());
1696 assertEquals(9, m.getEnd());
1700 * Test the method that realigns protein to match mapped codon alignment.
1702 @Test(groups = { "Functional" })
1703 public void testAlignProteinAsDna_incompleteStartCodon()
1705 // seq1: incomplete start codon (not mapped), then [3, 11]
1706 SequenceI dna1 = new Sequence("Seq1", "ccAAA-TTT-GGG-");
1707 // seq2 codons are [4, 5], [8, 11]
1708 SequenceI dna2 = new Sequence("Seq2", "ccaAA-ttT-GGG-");
1709 // seq3 incomplete start codon at 'tt'
1710 SequenceI dna3 = new Sequence("Seq3", "ccaaa-ttt-GGG-");
1711 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
1712 dna.setDataset(null);
1714 // prot1 has 'X' for incomplete start codon (not mapped)
1715 SequenceI prot1 = new Sequence("Seq1", "XKFG"); // X for incomplete start
1716 SequenceI prot2 = new Sequence("Seq2", "NG");
1717 SequenceI prot3 = new Sequence("Seq3", "XG"); // X for incomplete start
1718 AlignmentI protein = new Alignment(new SequenceI[] { prot1, prot2,
1720 protein.setDataset(null);
1722 // map dna1 [3, 11] to prot1 [2, 4] KFG
1723 MapList map = new MapList(new int[] { 3, 11 }, new int[] { 2, 4 }, 3, 1);
1724 AlignedCodonFrame acf = new AlignedCodonFrame();
1725 acf.addMap(dna1.getDatasetSequence(), prot1.getDatasetSequence(), map);
1727 // map dna2 [4, 5] [8, 11] to prot2 [1, 2] NG
1728 map = new MapList(new int[] { 4, 5, 8, 11 }, new int[] { 1, 2 }, 3, 1);
1729 acf.addMap(dna2.getDatasetSequence(), prot2.getDatasetSequence(), map);
1731 // map dna3 [9, 11] to prot3 [2, 2] G
1732 map = new MapList(new int[] { 9, 11 }, new int[] { 2, 2 }, 3, 1);
1733 acf.addMap(dna3.getDatasetSequence(), prot3.getDatasetSequence(), map);
1735 ArrayList<AlignedCodonFrame> acfs = new ArrayList<AlignedCodonFrame>();
1737 protein.setCodonFrames(acfs);
1740 * verify X is included in the aligned proteins, and placed just
1741 * before the first mapped residue
1742 * CCT is between CCC and TTT
1744 AlignmentUtils.alignProteinAsDna(protein, dna);
1745 assertEquals("XK-FG", prot1.getSequenceAsString());
1746 assertEquals("--N-G", prot2.getSequenceAsString());
1747 assertEquals("---XG", prot3.getSequenceAsString());
1751 * Tests for the method that maps the subset of a dna sequence that has CDS
1752 * (or subtype) feature - case where the start codon is incomplete.
1754 @Test(groups = "Functional")
1755 public void testFindCdsPositions_fivePrimeIncomplete()
1757 SequenceI dnaSeq = new Sequence("dna", "aaagGGCCCaaaTTTttt");
1758 dnaSeq.createDatasetSequence();
1759 SequenceI ds = dnaSeq.getDatasetSequence();
1761 // CDS for dna 5-6 (incomplete codon), 7-9
1762 SequenceFeature sf = new SequenceFeature("CDS", "", 5, 9, 0f, null);
1763 sf.setPhase("2"); // skip 2 bases to start of next codon
1764 ds.addSequenceFeature(sf);
1765 // CDS for dna 13-15
1766 sf = new SequenceFeature("CDS_predicted", "", 13, 15, 0f, null);
1767 ds.addSequenceFeature(sf);
1769 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
1772 * check the mapping starts with the first complete codon
1774 assertEquals(6, MappingUtils.getLength(ranges));
1775 assertEquals(2, ranges.size());
1776 assertEquals(7, ranges.get(0)[0]);
1777 assertEquals(9, ranges.get(0)[1]);
1778 assertEquals(13, ranges.get(1)[0]);
1779 assertEquals(15, ranges.get(1)[1]);
1783 * Tests for the method that maps the subset of a dna sequence that has CDS
1784 * (or subtype) feature.
1786 @Test(groups = "Functional")
1787 public void testFindCdsPositions()
1789 SequenceI dnaSeq = new Sequence("dna", "aaaGGGcccAAATTTttt");
1790 dnaSeq.createDatasetSequence();
1791 SequenceI ds = dnaSeq.getDatasetSequence();
1793 // CDS for dna 10-12
1794 SequenceFeature sf = new SequenceFeature("CDS_predicted", "", 10, 12,
1797 ds.addSequenceFeature(sf);
1799 sf = new SequenceFeature("CDS", "", 4, 6, 0f, null);
1801 ds.addSequenceFeature(sf);
1802 // exon feature should be ignored here
1803 sf = new SequenceFeature("exon", "", 7, 9, 0f, null);
1804 ds.addSequenceFeature(sf);
1806 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
1808 * verify ranges { [4-6], [12-10] }
1809 * note CDS ranges are ordered ascending even if the CDS
1812 assertEquals(6, MappingUtils.getLength(ranges));
1813 assertEquals(2, ranges.size());
1814 assertEquals(4, ranges.get(0)[0]);
1815 assertEquals(6, ranges.get(0)[1]);
1816 assertEquals(10, ranges.get(1)[0]);
1817 assertEquals(12, ranges.get(1)[1]);
1821 * Test the method that computes a map of codon variants for each protein
1822 * position from "sequence_variant" features on dna
1824 @Test(groups = "Functional")
1825 public void testBuildDnaVariantsMap()
1827 SequenceI dna = new Sequence("dna", "atgAAATTTGGGCCCtag");
1828 MapList map = new MapList(new int[] { 1, 18 }, new int[] { 1, 5 }, 3, 1);
1831 * first with no variants on dna
1833 LinkedHashMap<Integer, List<DnaVariant>[]> variantsMap = AlignmentUtils
1834 .buildDnaVariantsMap(dna, map);
1835 assertTrue(variantsMap.isEmpty());
1838 * single allele codon 1, on base 1
1840 SequenceFeature sf1 = new SequenceFeature("sequence_variant", "", 1, 1,
1842 sf1.setValue("alleles", "T");
1843 sf1.setValue("ID", "sequence_variant:rs758803211");
1844 dna.addSequenceFeature(sf1);
1847 * two alleles codon 2, on bases 2 and 3 (distinct variants)
1849 SequenceFeature sf2 = new SequenceFeature("sequence_variant", "", 5, 5,
1851 sf2.setValue("alleles", "T");
1852 sf2.setValue("ID", "sequence_variant:rs758803212");
1853 dna.addSequenceFeature(sf2);
1854 SequenceFeature sf3 = new SequenceFeature("sequence_variant", "", 6, 6,
1856 sf3.setValue("alleles", "G");
1857 sf3.setValue("ID", "sequence_variant:rs758803213");
1858 dna.addSequenceFeature(sf3);
1861 * two alleles codon 3, both on base 2 (one variant)
1863 SequenceFeature sf4 = new SequenceFeature("sequence_variant", "", 8, 8,
1865 sf4.setValue("alleles", "C, G");
1866 sf4.setValue("ID", "sequence_variant:rs758803214");
1867 dna.addSequenceFeature(sf4);
1869 // no alleles on codon 4
1872 * alleles on codon 5 on all 3 bases (distinct variants)
1874 SequenceFeature sf5 = new SequenceFeature("sequence_variant", "", 13,
1876 sf5.setValue("alleles", "C, G"); // (C duplicates given base value)
1877 sf5.setValue("ID", "sequence_variant:rs758803215");
1878 dna.addSequenceFeature(sf5);
1879 SequenceFeature sf6 = new SequenceFeature("sequence_variant", "", 14,
1881 sf6.setValue("alleles", "g, a"); // should force to upper-case
1882 sf6.setValue("ID", "sequence_variant:rs758803216");
1883 dna.addSequenceFeature(sf6);
1884 SequenceFeature sf7 = new SequenceFeature("sequence_variant", "", 15,
1886 sf7.setValue("alleles", "A, T");
1887 sf7.setValue("ID", "sequence_variant:rs758803217");
1888 dna.addSequenceFeature(sf7);
1891 * build map - expect variants on positions 1, 2, 3, 5
1893 variantsMap = AlignmentUtils.buildDnaVariantsMap(dna, map);
1894 assertEquals(4, variantsMap.size());
1897 * protein residue 1: variant on codon (ATG) base 1, not on 2 or 3
1899 List<DnaVariant>[] pep1Variants = variantsMap.get(1);
1900 assertEquals(3, pep1Variants.length);
1901 assertEquals(1, pep1Variants[0].size());
1902 assertEquals("A", pep1Variants[0].get(0).base); // codon[1] base
1903 assertSame(sf1, pep1Variants[0].get(0).variant); // codon[1] variant
1904 assertEquals(1, pep1Variants[1].size());
1905 assertEquals("T", pep1Variants[1].get(0).base); // codon[2] base
1906 assertNull(pep1Variants[1].get(0).variant); // no variant here
1907 assertEquals(1, pep1Variants[2].size());
1908 assertEquals("G", pep1Variants[2].get(0).base); // codon[3] base
1909 assertNull(pep1Variants[2].get(0).variant); // no variant here
1912 * protein residue 2: variants on codon (AAA) bases 2 and 3
1914 List<DnaVariant>[] pep2Variants = variantsMap.get(2);
1915 assertEquals(3, pep2Variants.length);
1916 assertEquals(1, pep2Variants[0].size());
1917 // codon[1] base recorded while processing variant on codon[2]
1918 assertEquals("A", pep2Variants[0].get(0).base);
1919 assertNull(pep2Variants[0].get(0).variant); // no variant here
1920 // codon[2] base and variant:
1921 assertEquals(1, pep2Variants[1].size());
1922 assertEquals("A", pep2Variants[1].get(0).base);
1923 assertSame(sf2, pep2Variants[1].get(0).variant);
1924 // codon[3] base was recorded when processing codon[2] variant
1925 // and then the variant for codon[3] added to it
1926 assertEquals(1, pep2Variants[2].size());
1927 assertEquals("A", pep2Variants[2].get(0).base);
1928 assertSame(sf3, pep2Variants[2].get(0).variant);
1931 * protein residue 3: variants on codon (TTT) base 2 only
1933 List<DnaVariant>[] pep3Variants = variantsMap.get(3);
1934 assertEquals(3, pep3Variants.length);
1935 assertEquals(1, pep3Variants[0].size());
1936 assertEquals("T", pep3Variants[0].get(0).base); // codon[1] base
1937 assertNull(pep3Variants[0].get(0).variant); // no variant here
1938 assertEquals(1, pep3Variants[1].size());
1939 assertEquals("T", pep3Variants[1].get(0).base); // codon[2] base
1940 assertSame(sf4, pep3Variants[1].get(0).variant); // codon[2] variant
1941 assertEquals(1, pep3Variants[2].size());
1942 assertEquals("T", pep3Variants[2].get(0).base); // codon[3] base
1943 assertNull(pep3Variants[2].get(0).variant); // no variant here
1946 * three variants on protein position 5
1948 List<DnaVariant>[] pep5Variants = variantsMap.get(5);
1949 assertEquals(3, pep5Variants.length);
1950 assertEquals(1, pep5Variants[0].size());
1951 assertEquals("C", pep5Variants[0].get(0).base); // codon[1] base
1952 assertSame(sf5, pep5Variants[0].get(0).variant); // codon[1] variant
1953 assertEquals(1, pep5Variants[1].size());
1954 assertEquals("C", pep5Variants[1].get(0).base); // codon[2] base
1955 assertSame(sf6, pep5Variants[1].get(0).variant); // codon[2] variant
1956 assertEquals(1, pep5Variants[2].size());
1957 assertEquals("C", pep5Variants[2].get(0).base); // codon[3] base
1958 assertSame(sf7, pep5Variants[2].get(0).variant); // codon[3] variant
1962 * Tests for the method that computes all peptide variants given codon
1965 @Test(groups = "Functional")
1966 public void testComputePeptideVariants()
1969 * scenario: AAATTTCCC codes for KFP
1971 * GAA -> E source: Ensembl
1972 * CAA -> Q source: dbSNP
1973 * AAG synonymous source: COSMIC
1974 * AAT -> N source: Ensembl
1975 * ...TTC synonymous source: dbSNP
1976 * ......CAC,CGC -> H,R source: COSMIC
1977 * (one variant with two alleles)
1979 SequenceI peptide = new Sequence("pep/10-12", "KFP");
1982 * two distinct variants for codon 1 position 1
1983 * second one has clinical significance
1985 String ensembl = "Ensembl";
1986 String dbSnp = "dbSNP";
1987 String cosmic = "COSMIC";
1988 SequenceFeature sf1 = new SequenceFeature("sequence_variant", "", 1, 1,
1990 sf1.setValue("alleles", "A,G"); // GAA -> E
1991 sf1.setValue("ID", "var1.125A>G");
1992 SequenceFeature sf2 = new SequenceFeature("sequence_variant", "", 1, 1,
1994 sf2.setValue("alleles", "A,C"); // CAA -> Q
1995 sf2.setValue("ID", "var2");
1996 sf2.setValue("clinical_significance", "Dodgy");
1997 SequenceFeature sf3 = new SequenceFeature("sequence_variant", "", 3, 3,
1999 sf3.setValue("alleles", "A,G"); // synonymous
2000 sf3.setValue("ID", "var3");
2001 sf3.setValue("clinical_significance", "None");
2002 SequenceFeature sf4 = new SequenceFeature("sequence_variant", "", 3, 3,
2004 sf4.setValue("alleles", "A,T"); // AAT -> N
2005 sf4.setValue("ID", "sequence_variant:var4"); // prefix gets stripped off
2006 sf4.setValue("clinical_significance", "Benign");
2007 SequenceFeature sf5 = new SequenceFeature("sequence_variant", "", 6, 6,
2009 sf5.setValue("alleles", "T,C"); // synonymous
2010 sf5.setValue("ID", "var5");
2011 sf5.setValue("clinical_significance", "Bad");
2012 SequenceFeature sf6 = new SequenceFeature("sequence_variant", "", 8, 8,
2014 sf6.setValue("alleles", "C,A,G"); // CAC,CGC -> H,R
2015 sf6.setValue("ID", "var6");
2016 sf6.setValue("clinical_significance", "Good");
2018 List<DnaVariant> codon1Variants = new ArrayList<DnaVariant>();
2019 List<DnaVariant> codon2Variants = new ArrayList<DnaVariant>();
2020 List<DnaVariant> codon3Variants = new ArrayList<DnaVariant>();
2021 List<DnaVariant> codonVariants[] = new ArrayList[3];
2022 codonVariants[0] = codon1Variants;
2023 codonVariants[1] = codon2Variants;
2024 codonVariants[2] = codon3Variants;
2027 * compute variants for protein position 1
2029 codon1Variants.add(new DnaVariant("A", sf1));
2030 codon1Variants.add(new DnaVariant("A", sf2));
2031 codon2Variants.add(new DnaVariant("A"));
2032 codon2Variants.add(new DnaVariant("A"));
2033 codon3Variants.add(new DnaVariant("A", sf3));
2034 codon3Variants.add(new DnaVariant("A", sf4));
2035 AlignmentUtils.computePeptideVariants(peptide, 1, codonVariants);
2038 * compute variants for protein position 2
2040 codon1Variants.clear();
2041 codon2Variants.clear();
2042 codon3Variants.clear();
2043 codon1Variants.add(new DnaVariant("T"));
2044 codon2Variants.add(new DnaVariant("T"));
2045 codon3Variants.add(new DnaVariant("T", sf5));
2046 AlignmentUtils.computePeptideVariants(peptide, 2, codonVariants);
2049 * compute variants for protein position 3
2051 codon1Variants.clear();
2052 codon2Variants.clear();
2053 codon3Variants.clear();
2054 codon1Variants.add(new DnaVariant("C"));
2055 codon2Variants.add(new DnaVariant("C", sf6));
2056 codon3Variants.add(new DnaVariant("C"));
2057 AlignmentUtils.computePeptideVariants(peptide, 3, codonVariants);
2060 * verify added sequence features for
2061 * var1 K -> E Ensembl
2063 * var4 K -> N Ensembl
2064 * var6 P -> H COSMIC
2065 * var6 P -> R COSMIC
2067 SequenceFeature[] sfs = peptide.getSequenceFeatures();
2068 assertEquals(5, sfs.length);
2070 SequenceFeature sf = sfs[0];
2071 assertEquals(1, sf.getBegin());
2072 assertEquals(1, sf.getEnd());
2073 assertEquals("p.Lys1Glu", sf.getDescription());
2074 assertEquals("var1.125A>G", sf.getValue("ID"));
2075 assertNull(sf.getValue("clinical_significance"));
2076 assertEquals("ID=var1.125A>G", sf.getAttributes());
2077 assertEquals(1, sf.links.size());
2078 // link to variation is urlencoded
2080 "p.Lys1Glu var1.125A>G|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var1.125A%3EG",
2082 assertEquals(ensembl, sf.getFeatureGroup());
2085 assertEquals(1, sf.getBegin());
2086 assertEquals(1, sf.getEnd());
2087 assertEquals("p.Lys1Gln", sf.getDescription());
2088 assertEquals("var2", sf.getValue("ID"));
2089 assertEquals("Dodgy", sf.getValue("clinical_significance"));
2090 assertEquals("ID=var2;clinical_significance=Dodgy", sf.getAttributes());
2091 assertEquals(1, sf.links.size());
2093 "p.Lys1Gln var2|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var2",
2095 assertEquals(dbSnp, sf.getFeatureGroup());
2098 assertEquals(1, sf.getBegin());
2099 assertEquals(1, sf.getEnd());
2100 assertEquals("p.Lys1Asn", sf.getDescription());
2101 assertEquals("var4", sf.getValue("ID"));
2102 assertEquals("Benign", sf.getValue("clinical_significance"));
2103 assertEquals("ID=var4;clinical_significance=Benign", sf.getAttributes());
2104 assertEquals(1, sf.links.size());
2106 "p.Lys1Asn var4|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var4",
2108 assertEquals(ensembl, sf.getFeatureGroup());
2110 // var5 generates two distinct protein variant features
2112 assertEquals(3, sf.getBegin());
2113 assertEquals(3, sf.getEnd());
2114 assertEquals("p.Pro3His", sf.getDescription());
2115 assertEquals("var6", sf.getValue("ID"));
2116 assertEquals("Good", sf.getValue("clinical_significance"));
2117 assertEquals("ID=var6;clinical_significance=Good", sf.getAttributes());
2118 assertEquals(1, sf.links.size());
2120 "p.Pro3His var6|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var6",
2122 assertEquals(cosmic, sf.getFeatureGroup());
2125 assertEquals(3, sf.getBegin());
2126 assertEquals(3, sf.getEnd());
2127 assertEquals("p.Pro3Arg", sf.getDescription());
2128 assertEquals("var6", sf.getValue("ID"));
2129 assertEquals("Good", sf.getValue("clinical_significance"));
2130 assertEquals("ID=var6;clinical_significance=Good", sf.getAttributes());
2131 assertEquals(1, sf.links.size());
2133 "p.Pro3Arg var6|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var6",
2135 assertEquals(cosmic, sf.getFeatureGroup());
2139 * Tests for the method that maps the subset of a dna sequence that has CDS
2140 * (or subtype) feature, with CDS strand = '-' (reverse)
2142 // test turned off as currently findCdsPositions is not strand-dependent
2143 // left in case it comes around again...
2144 @Test(groups = "Functional", enabled = false)
2145 public void testFindCdsPositions_reverseStrand()
2147 SequenceI dnaSeq = new Sequence("dna", "aaaGGGcccAAATTTttt");
2148 dnaSeq.createDatasetSequence();
2149 SequenceI ds = dnaSeq.getDatasetSequence();
2152 SequenceFeature sf = new SequenceFeature("CDS", "", 4, 6, 0f, null);
2154 ds.addSequenceFeature(sf);
2155 // exon feature should be ignored here
2156 sf = new SequenceFeature("exon", "", 7, 9, 0f, null);
2157 ds.addSequenceFeature(sf);
2158 // CDS for dna 10-12
2159 sf = new SequenceFeature("CDS_predicted", "", 10, 12, 0f, null);
2161 ds.addSequenceFeature(sf);
2163 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
2165 * verify ranges { [12-10], [6-4] }
2167 assertEquals(6, MappingUtils.getLength(ranges));
2168 assertEquals(2, ranges.size());
2169 assertEquals(12, ranges.get(0)[0]);
2170 assertEquals(10, ranges.get(0)[1]);
2171 assertEquals(6, ranges.get(1)[0]);
2172 assertEquals(4, ranges.get(1)[1]);
2176 * Tests for the method that maps the subset of a dna sequence that has CDS
2177 * (or subtype) feature - reverse strand case where the start codon is
2180 @Test(groups = "Functional", enabled = false)
2181 // test turned off as currently findCdsPositions is not strand-dependent
2182 // left in case it comes around again...
2183 public void testFindCdsPositions_reverseStrandThreePrimeIncomplete()
2185 SequenceI dnaSeq = new Sequence("dna", "aaagGGCCCaaaTTTttt");
2186 dnaSeq.createDatasetSequence();
2187 SequenceI ds = dnaSeq.getDatasetSequence();
2190 SequenceFeature sf = new SequenceFeature("CDS", "", 5, 9, 0f, null);
2192 ds.addSequenceFeature(sf);
2193 // CDS for dna 13-15
2194 sf = new SequenceFeature("CDS_predicted", "", 13, 15, 0f, null);
2196 sf.setPhase("2"); // skip 2 bases to start of next codon
2197 ds.addSequenceFeature(sf);
2199 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
2202 * check the mapping starts with the first complete codon
2203 * expect ranges [13, 13], [9, 5]
2205 assertEquals(6, MappingUtils.getLength(ranges));
2206 assertEquals(2, ranges.size());
2207 assertEquals(13, ranges.get(0)[0]);
2208 assertEquals(13, ranges.get(0)[1]);
2209 assertEquals(9, ranges.get(1)[0]);
2210 assertEquals(5, ranges.get(1)[1]);
2213 @Test(groups = "Functional")
2214 public void testAlignAs_alternateTranscriptsUngapped()
2216 SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
2217 SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
2218 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
2219 ((Alignment) dna).createDatasetAlignment();
2220 SequenceI cds1 = new Sequence("cds1", "GGGTTT");
2221 SequenceI cds2 = new Sequence("cds2", "CCCAAA");
2222 AlignmentI cds = new Alignment(new SequenceI[] { cds1, cds2 });
2223 ((Alignment) cds).createDatasetAlignment();
2225 AlignedCodonFrame acf = new AlignedCodonFrame();
2226 MapList map = new MapList(new int[] { 4, 9 }, new int[] { 1, 6 }, 1, 1);
2227 acf.addMap(dna1.getDatasetSequence(), cds1.getDatasetSequence(), map);
2228 map = new MapList(new int[] { 1, 3, 10, 12 }, new int[] { 1, 6 }, 1, 1);
2229 acf.addMap(dna2.getDatasetSequence(), cds2.getDatasetSequence(), map);
2232 * verify CDS alignment is as:
2233 * cccGGGTTTaaa (cdna)
2234 * CCCgggtttAAA (cdna)
2236 * ---GGGTTT--- (cds)
2237 * CCC------AAA (cds)
2239 dna.addCodonFrame(acf);
2240 AlignmentUtils.alignAs(cds, dna);
2241 assertEquals("---GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
2242 assertEquals("CCC------AAA", cds.getSequenceAt(1).getSequenceAsString());
2245 @Test(groups = { "Functional" })
2246 public void testAddMappedPositions()
2248 SequenceI from = new Sequence("dna", "ggAA-ATcc-TT-g");
2249 SequenceI seq1 = new Sequence("cds", "AAATTT");
2250 from.createDatasetSequence();
2251 seq1.createDatasetSequence();
2252 Mapping mapping = new Mapping(seq1, new MapList(
2253 new int[] { 3, 6, 9, 10 }, new int[] { 1, 6 }, 1, 1));
2254 Map<Integer, Map<SequenceI, Character>> map = new TreeMap<Integer, Map<SequenceI, Character>>();
2255 AlignmentUtils.addMappedPositions(seq1, from, mapping, map);
2258 * verify map has seq1 residues in columns 3,4,6,7,11,12
2260 assertEquals(6, map.size());
2261 assertEquals('A', map.get(3).get(seq1).charValue());
2262 assertEquals('A', map.get(4).get(seq1).charValue());
2263 assertEquals('A', map.get(6).get(seq1).charValue());
2264 assertEquals('T', map.get(7).get(seq1).charValue());
2265 assertEquals('T', map.get(11).get(seq1).charValue());
2266 assertEquals('T', map.get(12).get(seq1).charValue());
2274 * Test case where the mapping 'from' range includes a stop codon which is
2275 * absent in the 'to' range
2277 @Test(groups = { "Functional" })
2278 public void testAddMappedPositions_withStopCodon()
2280 SequenceI from = new Sequence("dna", "ggAA-ATcc-TT-g");
2281 SequenceI seq1 = new Sequence("cds", "AAATTT");
2282 from.createDatasetSequence();
2283 seq1.createDatasetSequence();
2284 Mapping mapping = new Mapping(seq1, new MapList(
2285 new int[] { 3, 6, 9, 10 }, new int[] { 1, 6 }, 1, 1));
2286 Map<Integer, Map<SequenceI, Character>> map = new TreeMap<Integer, Map<SequenceI, Character>>();
2287 AlignmentUtils.addMappedPositions(seq1, from, mapping, map);
2290 * verify map has seq1 residues in columns 3,4,6,7,11,12
2292 assertEquals(6, map.size());
2293 assertEquals('A', map.get(3).get(seq1).charValue());
2294 assertEquals('A', map.get(4).get(seq1).charValue());
2295 assertEquals('A', map.get(6).get(seq1).charValue());
2296 assertEquals('T', map.get(7).get(seq1).charValue());
2297 assertEquals('T', map.get(11).get(seq1).charValue());
2298 assertEquals('T', map.get(12).get(seq1).charValue());
2302 * Test for the case where the products for which we want CDS are specified.
2303 * This is to represent the case where EMBL has CDS mappings to both Uniprot
2304 * and EMBLCDSPROTEIN. makeCdsAlignment() should only return the mappings for
2305 * the protein sequences specified.
2307 @Test(groups = { "Functional" })
2308 public void testMakeCdsAlignment_filterProducts()
2310 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
2311 SequenceI dna2 = new Sequence("dna2", "GGGcccTTTaaaCCC");
2312 SequenceI pep1 = new Sequence("Uniprot|pep1", "GF");
2313 SequenceI pep2 = new Sequence("Uniprot|pep2", "GFP");
2314 SequenceI pep3 = new Sequence("EMBL|pep3", "GF");
2315 SequenceI pep4 = new Sequence("EMBL|pep4", "GFP");
2316 dna1.createDatasetSequence();
2317 dna2.createDatasetSequence();
2318 pep1.createDatasetSequence();
2319 pep2.createDatasetSequence();
2320 pep3.createDatasetSequence();
2321 pep4.createDatasetSequence();
2322 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
2323 dna.setDataset(null);
2324 AlignmentI emblPeptides = new Alignment(new SequenceI[] { pep3, pep4 });
2325 emblPeptides.setDataset(null);
2327 AlignedCodonFrame acf = new AlignedCodonFrame();
2328 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
2329 new int[] { 1, 2 }, 3, 1);
2330 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
2331 acf.addMap(dna1.getDatasetSequence(), pep3.getDatasetSequence(), map);
2332 dna.addCodonFrame(acf);
2334 acf = new AlignedCodonFrame();
2335 map = new MapList(new int[] { 1, 3, 7, 9, 13, 15 }, new int[] { 1, 3 },
2337 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map);
2338 acf.addMap(dna2.getDatasetSequence(), pep4.getDatasetSequence(), map);
2339 dna.addCodonFrame(acf);
2342 * execute method under test to find CDS for EMBL peptides only
2344 AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
2345 dna1, dna2 }, dna.getDataset(), emblPeptides.getSequencesArray());
2347 assertEquals(2, cds.getSequences().size());
2348 assertEquals("GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
2349 assertEquals("GGGTTTCCC", cds.getSequenceAt(1).getSequenceAsString());
2352 * verify shared, extended alignment dataset
2354 assertSame(dna.getDataset(), cds.getDataset());
2355 assertTrue(dna.getDataset().getSequences()
2356 .contains(cds.getSequenceAt(0).getDatasetSequence()));
2357 assertTrue(dna.getDataset().getSequences()
2358 .contains(cds.getSequenceAt(1).getDatasetSequence()));
2361 * Verify mappings from CDS to peptide, cDNA to CDS, and cDNA to peptide
2362 * the mappings are on the shared alignment dataset
2364 List<AlignedCodonFrame> cdsMappings = cds.getDataset().getCodonFrames();
2366 * 6 mappings, 2*(DNA->CDS), 2*(DNA->Pep), 2*(CDS->Pep)
2368 assertEquals(6, cdsMappings.size());
2371 * verify that mapping sets for dna and cds alignments are different
2372 * [not current behaviour - all mappings are on the alignment dataset]
2374 // select -> subselect type to test.
2375 // Assert.assertNotSame(dna.getCodonFrames(), cds.getCodonFrames());
2376 // assertEquals(4, dna.getCodonFrames().size());
2377 // assertEquals(4, cds.getCodonFrames().size());
2380 * Two mappings involve pep3 (dna to pep3, cds to pep3)
2381 * Mapping from pep3 to GGGTTT in first new exon sequence
2383 List<AlignedCodonFrame> pep3Mappings = MappingUtils
2384 .findMappingsForSequence(pep3, cdsMappings);
2385 assertEquals(2, pep3Mappings.size());
2386 List<AlignedCodonFrame> mappings = MappingUtils
2387 .findMappingsForSequence(cds.getSequenceAt(0), pep3Mappings);
2388 assertEquals(1, mappings.size());
2391 SearchResults sr = MappingUtils.buildSearchResults(pep3, 1, mappings);
2392 assertEquals(1, sr.getResults().size());
2393 Match m = sr.getResults().get(0);
2394 assertSame(cds.getSequenceAt(0).getDatasetSequence(), m.getSequence());
2395 assertEquals(1, m.getStart());
2396 assertEquals(3, m.getEnd());
2398 sr = MappingUtils.buildSearchResults(pep3, 2, mappings);
2399 m = sr.getResults().get(0);
2400 assertSame(cds.getSequenceAt(0).getDatasetSequence(), m.getSequence());
2401 assertEquals(4, m.getStart());
2402 assertEquals(6, m.getEnd());
2405 * Two mappings involve pep4 (dna to pep4, cds to pep4)
2406 * Verify mapping from pep4 to GGGTTTCCC in second new exon sequence
2408 List<AlignedCodonFrame> pep4Mappings = MappingUtils
2409 .findMappingsForSequence(pep4, cdsMappings);
2410 assertEquals(2, pep4Mappings.size());
2411 mappings = MappingUtils.findMappingsForSequence(cds.getSequenceAt(1),
2413 assertEquals(1, mappings.size());
2415 sr = MappingUtils.buildSearchResults(pep4, 1, mappings);
2416 assertEquals(1, sr.getResults().size());
2417 m = sr.getResults().get(0);
2418 assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
2419 assertEquals(1, m.getStart());
2420 assertEquals(3, m.getEnd());
2422 sr = MappingUtils.buildSearchResults(pep4, 2, mappings);
2423 m = sr.getResults().get(0);
2424 assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
2425 assertEquals(4, m.getStart());
2426 assertEquals(6, m.getEnd());
2428 sr = MappingUtils.buildSearchResults(pep4, 3, mappings);
2429 m = sr.getResults().get(0);
2430 assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
2431 assertEquals(7, m.getStart());
2432 assertEquals(9, m.getEnd());
2436 * Test the method that just copies aligned sequences, provided all sequences
2437 * to be aligned share the aligned sequence's dataset
2439 @Test(groups = "Functional")
2440 public void testAlignAsSameSequences()
2442 SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
2443 SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
2444 AlignmentI al1 = new Alignment(new SequenceI[] { dna1, dna2 });
2445 ((Alignment) al1).createDatasetAlignment();
2447 SequenceI dna3 = new Sequence(dna1);
2448 SequenceI dna4 = new Sequence(dna2);
2449 assertSame(dna3.getDatasetSequence(), dna1.getDatasetSequence());
2450 assertSame(dna4.getDatasetSequence(), dna2.getDatasetSequence());
2451 String seq1 = "-cc-GG-GT-TT--aaa";
2452 dna3.setSequence(seq1);
2453 String seq2 = "C--C-Cgg--gtt-tAA-A-";
2454 dna4.setSequence(seq2);
2455 AlignmentI al2 = new Alignment(new SequenceI[] { dna3, dna4 });
2456 ((Alignment) al2).createDatasetAlignment();
2458 assertTrue(AlignmentUtils.alignAsSameSequences(al1, al2));
2459 assertEquals(seq1, al1.getSequenceAt(0).getSequenceAsString());
2460 assertEquals(seq2, al1.getSequenceAt(1).getSequenceAsString());
2463 * add another sequence to 'aligned' - should still succeed, since
2464 * unaligned sequences still share a dataset with aligned sequences
2466 SequenceI dna5 = new Sequence("dna5", "CCCgggtttAAA");
2467 dna5.createDatasetSequence();
2468 al2.addSequence(dna5);
2469 assertTrue(AlignmentUtils.alignAsSameSequences(al1, al2));
2470 assertEquals(seq1, al1.getSequenceAt(0).getSequenceAsString());
2471 assertEquals(seq2, al1.getSequenceAt(1).getSequenceAsString());
2474 * add another sequence to 'unaligned' - should fail, since now not
2475 * all unaligned sequences share a dataset with aligned sequences
2477 SequenceI dna6 = new Sequence("dna6", "CCCgggtttAAA");
2478 dna6.createDatasetSequence();
2479 al1.addSequence(dna6);
2480 // JAL-2110 JBP Comment: what's the use case for this behaviour ?
2481 assertFalse(AlignmentUtils.alignAsSameSequences(al1, al2));
2484 @Test(groups = "Functional")
2485 public void testAlignAsSameSequencesMultipleSubSeq()
2487 SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
2488 SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
2489 SequenceI as1 = dna1.deriveSequence();
2490 SequenceI as2 = dna1.deriveSequence().getSubSequence(3, 7);
2491 SequenceI as3 = dna2.deriveSequence();
2492 as1.insertCharAt(6, 5, '-');
2493 String s_as1 = as1.getSequenceAsString();
2494 as2.insertCharAt(6, 5, '-');
2495 String s_as2 = as2.getSequenceAsString();
2496 as3.insertCharAt(6, 5, '-');
2497 String s_as3 = as3.getSequenceAsString();
2498 AlignmentI aligned = new Alignment(new SequenceI[] { as1, as2, as3 });
2500 // why do we need to cast this still ?
2501 ((Alignment) aligned).createDatasetAlignment();
2502 SequenceI uas1 = dna1.deriveSequence();
2503 SequenceI uas2 = dna1.deriveSequence().getSubSequence(3, 7);
2504 SequenceI uas3 = dna2.deriveSequence();
2505 AlignmentI tobealigned = new Alignment(new SequenceI[] { uas1, uas2,
2507 ((Alignment) tobealigned).createDatasetAlignment();
2509 assertTrue(AlignmentUtils.alignAsSameSequences(tobealigned, aligned));
2510 assertEquals(s_as1, uas1.getSequenceAsString());
2511 assertEquals(s_as2, uas2.getSequenceAsString());
2512 assertEquals(s_as3, uas3.getSequenceAsString());