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(), alignFrom.getDatasetSequence(), map);
444 AlignmentUtils.alignSequenceAs(alignMe, alignFrom, acf, "---", '-',
445 preserveMappedGaps, preserveUnmappedGaps);
446 assertEquals(expected, alignMe.getSequenceAsString());
450 * Test for the alignSequenceAs method where we preserve gaps in introns only.
452 @Test(groups = { "Functional" })
453 public void testAlignSequenceAs_keepIntronGapsOnly()
457 * Intron GGGAAA followed by exon CCCTTT
459 MapList map = new MapList(new int[] { 7, 12 }, new int[] { 1, 2 }, 3, 1);
461 checkAlignSequenceAs("GG-G-AA-A-C-CC-T-TT", "AL", false, true, map,
466 * Test the method that realigns protein to match mapped codon alignment.
468 @Test(groups = { "Functional" })
469 public void testAlignProteinAsDna()
471 // seq1 codons are [1,2,3] [4,5,6] [7,8,9] [10,11,12]
472 SequenceI dna1 = new Sequence("Seq1", "TGCCATTACCAG-");
473 // seq2 codons are [1,3,4] [5,6,7] [8,9,10] [11,12,13]
474 SequenceI dna2 = new Sequence("Seq2", "T-GCCATTACCAG");
475 // seq3 codons are [1,2,3] [4,5,7] [8,9,10] [11,12,13]
476 SequenceI dna3 = new Sequence("Seq3", "TGCCA-TTACCAG");
477 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
478 dna.setDataset(null);
480 // protein alignment will be realigned like dna
481 SequenceI prot1 = new Sequence("Seq1", "CHYQ");
482 SequenceI prot2 = new Sequence("Seq2", "CHYQ");
483 SequenceI prot3 = new Sequence("Seq3", "CHYQ");
484 SequenceI prot4 = new Sequence("Seq4", "R-QSV"); // unmapped, unchanged
485 AlignmentI protein = new Alignment(new SequenceI[] { prot1, prot2,
487 protein.setDataset(null);
489 MapList map = new MapList(new int[] { 1, 12 }, new int[] { 1, 4 }, 3, 1);
490 AlignedCodonFrame acf = new AlignedCodonFrame();
491 acf.addMap(dna1.getDatasetSequence(), prot1.getDatasetSequence(), map);
492 acf.addMap(dna2.getDatasetSequence(), prot2.getDatasetSequence(), map);
493 acf.addMap(dna3.getDatasetSequence(), prot3.getDatasetSequence(), map);
494 ArrayList<AlignedCodonFrame> acfs = new ArrayList<AlignedCodonFrame>();
496 protein.setCodonFrames(acfs);
499 * Translated codon order is [1,2,3] [1,3,4] [4,5,6] [4,5,7] [5,6,7] [7,8,9]
500 * [8,9,10] [10,11,12] [11,12,13]
502 AlignmentUtils.alignProteinAsDna(protein, dna);
503 assertEquals("C-H--Y-Q-", prot1.getSequenceAsString());
504 assertEquals("-C--H-Y-Q", prot2.getSequenceAsString());
505 assertEquals("C--H--Y-Q", prot3.getSequenceAsString());
506 assertEquals("R-QSV", prot4.getSequenceAsString());
510 * Test the method that tests whether a CDNA sequence translates to a protein
513 @Test(groups = { "Functional" })
514 public void testTranslatesAs()
516 // null arguments check
517 assertFalse(AlignmentUtils.translatesAs(null, 0, null));
518 assertFalse(AlignmentUtils.translatesAs(new char[] { 't' }, 0, null));
519 assertFalse(AlignmentUtils.translatesAs(null, 0, new char[] { 'a' }));
521 // straight translation
522 assertTrue(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(), 0,
523 "FPKG".toCharArray()));
524 // with extra start codon (not in protein)
525 assertTrue(AlignmentUtils.translatesAs("atgtttcccaaaggg".toCharArray(),
526 3, "FPKG".toCharArray()));
527 // with stop codon1 (not in protein)
528 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtaa".toCharArray(),
529 0, "FPKG".toCharArray()));
530 // with stop codon1 (in protein as *)
531 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtaa".toCharArray(),
532 0, "FPKG*".toCharArray()));
533 // with stop codon2 (not in protein)
534 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtag".toCharArray(),
535 0, "FPKG".toCharArray()));
536 // with stop codon3 (not in protein)
537 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtga".toCharArray(),
538 0, "FPKG".toCharArray()));
539 // with start and stop codon1
540 assertTrue(AlignmentUtils.translatesAs(
541 "atgtttcccaaagggtaa".toCharArray(), 3, "FPKG".toCharArray()));
542 // with start and stop codon1 (in protein as *)
543 assertTrue(AlignmentUtils.translatesAs(
544 "atgtttcccaaagggtaa".toCharArray(), 3, "FPKG*".toCharArray()));
545 // with start and stop codon2
546 assertTrue(AlignmentUtils.translatesAs(
547 "atgtttcccaaagggtag".toCharArray(), 3, "FPKG".toCharArray()));
548 // with start and stop codon3
549 assertTrue(AlignmentUtils.translatesAs(
550 "atgtttcccaaagggtga".toCharArray(), 3, "FPKG".toCharArray()));
552 // with embedded stop codons
553 assertTrue(AlignmentUtils.translatesAs(
554 "atgtttTAGcccaaaTAAgggtga".toCharArray(), 3,
555 "F*PK*G".toCharArray()));
558 assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(),
559 0, "FPMG".toCharArray()));
562 assertFalse(AlignmentUtils.translatesAs("tttcccaaagg".toCharArray(), 0,
563 "FPKG".toCharArray()));
566 assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(),
567 0, "FPK".toCharArray()));
569 // overlong dna (doesn't end in stop codon)
570 assertFalse(AlignmentUtils.translatesAs(
571 "tttcccaaagggttt".toCharArray(), 0, "FPKG".toCharArray()));
573 // dna + stop codon + more
574 assertFalse(AlignmentUtils.translatesAs(
575 "tttcccaaagggttaga".toCharArray(), 0, "FPKG".toCharArray()));
578 assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(),
579 0, "FPKGQ".toCharArray()));
583 * Test mapping of protein to cDNA, for cases where the cDNA has start and/or
584 * stop codons in addition to the protein coding sequence.
586 * @throws IOException
588 @Test(groups = { "Functional" })
589 public void testMapProteinAlignmentToCdna_withStartAndStopCodons()
592 List<SequenceI> protseqs = new ArrayList<SequenceI>();
593 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
594 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
595 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
596 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
597 protein.setDataset(null);
599 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
601 dnaseqs.add(new Sequence("EMBL|A11111", "ATGTCAGCACGC"));
603 dnaseqs.add(new Sequence("EMBL|A22222", "GAGATACAATAA"));
604 // = start +EIQ + stop
605 dnaseqs.add(new Sequence("EMBL|A33333", "ATGGAAATCCAGTAG"));
606 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG"));
607 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[4]));
608 cdna.setDataset(null);
610 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
612 // 3 mappings made, each from 1 to 1 sequence
613 assertEquals(3, protein.getCodonFrames().size());
614 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
615 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
616 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
618 // V12345 mapped from A22222
619 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
621 assertEquals(1, acf.getdnaSeqs().length);
622 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
623 acf.getdnaSeqs()[0]);
624 Mapping[] protMappings = acf.getProtMappings();
625 assertEquals(1, protMappings.length);
626 MapList mapList = protMappings[0].getMap();
627 assertEquals(3, mapList.getFromRatio());
628 assertEquals(1, mapList.getToRatio());
629 assertTrue(Arrays.equals(new int[] { 1, 9 }, mapList.getFromRanges()
631 assertEquals(1, mapList.getFromRanges().size());
632 assertTrue(Arrays.equals(new int[] { 1, 3 },
633 mapList.getToRanges().get(0)));
634 assertEquals(1, mapList.getToRanges().size());
636 // V12346 mapped from A33333 starting position 4
637 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
638 assertEquals(1, acf.getdnaSeqs().length);
639 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
640 acf.getdnaSeqs()[0]);
641 protMappings = acf.getProtMappings();
642 assertEquals(1, protMappings.length);
643 mapList = protMappings[0].getMap();
644 assertEquals(3, mapList.getFromRatio());
645 assertEquals(1, mapList.getToRatio());
646 assertTrue(Arrays.equals(new int[] { 4, 12 }, mapList.getFromRanges()
648 assertEquals(1, mapList.getFromRanges().size());
649 assertTrue(Arrays.equals(new int[] { 1, 3 },
650 mapList.getToRanges().get(0)));
651 assertEquals(1, mapList.getToRanges().size());
653 // V12347 mapped to A11111 starting position 4
654 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
655 assertEquals(1, acf.getdnaSeqs().length);
656 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
657 acf.getdnaSeqs()[0]);
658 protMappings = acf.getProtMappings();
659 assertEquals(1, protMappings.length);
660 mapList = protMappings[0].getMap();
661 assertEquals(3, mapList.getFromRatio());
662 assertEquals(1, mapList.getToRatio());
663 assertTrue(Arrays.equals(new int[] { 4, 12 }, mapList.getFromRanges()
665 assertEquals(1, mapList.getFromRanges().size());
666 assertTrue(Arrays.equals(new int[] { 1, 3 },
667 mapList.getToRanges().get(0)));
668 assertEquals(1, mapList.getToRanges().size());
670 // no mapping involving the 'extra' A44444
671 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(3)).isEmpty());
675 * Test mapping of protein to cDNA, for the case where we have some sequence
676 * cross-references. Verify that 1-to-many mappings are made where
677 * cross-references exist and sequences are mappable.
679 * @throws IOException
681 @Test(groups = { "Functional" })
682 public void testMapProteinAlignmentToCdna_withXrefs() throws IOException
684 List<SequenceI> protseqs = new ArrayList<SequenceI>();
685 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
686 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
687 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
688 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
689 protein.setDataset(null);
691 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
692 dnaseqs.add(new Sequence("EMBL|A11111", "TCAGCACGC")); // = SAR
693 dnaseqs.add(new Sequence("EMBL|A22222", "ATGGAGATACAA")); // = start + EIQ
694 dnaseqs.add(new Sequence("EMBL|A33333", "GAAATCCAG")); // = EIQ
695 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG")); // = EIQ
696 dnaseqs.add(new Sequence("EMBL|A55555", "GAGATTCAG")); // = EIQ
697 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[5]));
698 cdna.setDataset(null);
700 // Xref A22222 to V12345 (should get mapped)
701 dnaseqs.get(1).addDBRef(new DBRefEntry("UNIPROT", "1", "V12345"));
702 // Xref V12345 to A44444 (should get mapped)
703 protseqs.get(0).addDBRef(new DBRefEntry("EMBL", "1", "A44444"));
704 // Xref A33333 to V12347 (sequence mismatch - should not get mapped)
705 dnaseqs.get(2).addDBRef(new DBRefEntry("UNIPROT", "1", "V12347"));
706 // as V12345 is mapped to A22222 and A44444, this leaves V12346 unmapped.
707 // it should get paired up with the unmapped A33333
708 // A11111 should be mapped to V12347
709 // A55555 is spare and has no xref so is not mapped
711 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
713 // 4 protein mappings made for 3 proteins, 2 to V12345, 1 each to V12346/7
714 assertEquals(3, protein.getCodonFrames().size());
715 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
716 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
717 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
719 // one mapping for each of the first 4 cDNA sequences
720 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(0)).size());
721 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(1)).size());
722 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(2)).size());
723 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(3)).size());
725 // V12345 mapped to A22222 and A44444
726 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
728 assertEquals(2, acf.getdnaSeqs().length);
729 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
730 acf.getdnaSeqs()[0]);
731 assertEquals(cdna.getSequenceAt(3).getDatasetSequence(),
732 acf.getdnaSeqs()[1]);
734 // V12346 mapped to A33333
735 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
736 assertEquals(1, acf.getdnaSeqs().length);
737 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
738 acf.getdnaSeqs()[0]);
740 // V12347 mapped to A11111
741 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
742 assertEquals(1, acf.getdnaSeqs().length);
743 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
744 acf.getdnaSeqs()[0]);
746 // no mapping involving the 'extra' A55555
747 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(4)).isEmpty());
751 * Test mapping of protein to cDNA, for the case where we have some sequence
752 * cross-references. Verify that once we have made an xref mapping we don't
753 * also map un-xrefd sequeces.
755 * @throws IOException
757 @Test(groups = { "Functional" })
758 public void testMapProteinAlignmentToCdna_prioritiseXrefs()
761 List<SequenceI> protseqs = new ArrayList<SequenceI>();
762 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
763 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
764 AlignmentI protein = new Alignment(
765 protseqs.toArray(new SequenceI[protseqs.size()]));
766 protein.setDataset(null);
768 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
769 dnaseqs.add(new Sequence("EMBL|A11111", "GAAATCCAG")); // = EIQ
770 dnaseqs.add(new Sequence("EMBL|A22222", "GAAATTCAG")); // = EIQ
771 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[dnaseqs
773 cdna.setDataset(null);
775 // Xref A22222 to V12345 (should get mapped)
776 // A11111 should then be mapped to the unmapped V12346
777 dnaseqs.get(1).addDBRef(new DBRefEntry("UNIPROT", "1", "V12345"));
779 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
781 // 2 protein mappings made
782 assertEquals(2, protein.getCodonFrames().size());
783 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
784 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
786 // one mapping for each of the cDNA sequences
787 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(0)).size());
788 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(1)).size());
790 // V12345 mapped to A22222
791 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
793 assertEquals(1, acf.getdnaSeqs().length);
794 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
795 acf.getdnaSeqs()[0]);
797 // V12346 mapped to A11111
798 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
799 assertEquals(1, acf.getdnaSeqs().length);
800 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
801 acf.getdnaSeqs()[0]);
805 * Test the method that shows or hides sequence annotations by type(s) and
808 @Test(groups = { "Functional" })
809 public void testShowOrHideSequenceAnnotations()
811 SequenceI seq1 = new Sequence("Seq1", "AAA");
812 SequenceI seq2 = new Sequence("Seq2", "BBB");
813 SequenceI seq3 = new Sequence("Seq3", "CCC");
814 Annotation[] anns = new Annotation[] { new Annotation(2f) };
815 AlignmentAnnotation ann1 = new AlignmentAnnotation("Structure", "ann1",
817 ann1.setSequenceRef(seq1);
818 AlignmentAnnotation ann2 = new AlignmentAnnotation("Structure", "ann2",
820 ann2.setSequenceRef(seq2);
821 AlignmentAnnotation ann3 = new AlignmentAnnotation("Structure", "ann3",
823 AlignmentAnnotation ann4 = new AlignmentAnnotation("Temp", "ann4", anns);
824 ann4.setSequenceRef(seq1);
825 AlignmentAnnotation ann5 = new AlignmentAnnotation("Temp", "ann5", anns);
826 ann5.setSequenceRef(seq2);
827 AlignmentAnnotation ann6 = new AlignmentAnnotation("Temp", "ann6", anns);
828 AlignmentI al = new Alignment(new SequenceI[] { seq1, seq2, seq3 });
829 al.addAnnotation(ann1); // Structure for Seq1
830 al.addAnnotation(ann2); // Structure for Seq2
831 al.addAnnotation(ann3); // Structure for no sequence
832 al.addAnnotation(ann4); // Temp for seq1
833 al.addAnnotation(ann5); // Temp for seq2
834 al.addAnnotation(ann6); // Temp for no sequence
835 List<String> types = new ArrayList<String>();
836 List<SequenceI> scope = new ArrayList<SequenceI>();
839 * Set all sequence related Structure to hidden (ann1, ann2)
841 types.add("Structure");
842 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, false,
844 assertFalse(ann1.visible);
845 assertFalse(ann2.visible);
846 assertTrue(ann3.visible); // not sequence-related, not affected
847 assertTrue(ann4.visible); // not Structure, not affected
848 assertTrue(ann5.visible); // "
849 assertTrue(ann6.visible); // not sequence-related, not affected
852 * Set Temp in {seq1, seq3} to hidden
858 AlignmentUtils.showOrHideSequenceAnnotations(al, types, scope, false,
860 assertFalse(ann1.visible); // unchanged
861 assertFalse(ann2.visible); // unchanged
862 assertTrue(ann3.visible); // not sequence-related, not affected
863 assertFalse(ann4.visible); // Temp for seq1 hidden
864 assertTrue(ann5.visible); // not in scope, not affected
865 assertTrue(ann6.visible); // not sequence-related, not affected
868 * Set Temp in all sequences to hidden
874 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, false,
876 assertFalse(ann1.visible); // unchanged
877 assertFalse(ann2.visible); // unchanged
878 assertTrue(ann3.visible); // not sequence-related, not affected
879 assertFalse(ann4.visible); // Temp for seq1 hidden
880 assertFalse(ann5.visible); // Temp for seq2 hidden
881 assertTrue(ann6.visible); // not sequence-related, not affected
884 * Set all types in {seq1, seq3} to visible
890 AlignmentUtils.showOrHideSequenceAnnotations(al, types, scope, true,
892 assertTrue(ann1.visible); // Structure for seq1 set visible
893 assertFalse(ann2.visible); // not in scope, unchanged
894 assertTrue(ann3.visible); // not sequence-related, not affected
895 assertTrue(ann4.visible); // Temp for seq1 set visible
896 assertFalse(ann5.visible); // not in scope, unchanged
897 assertTrue(ann6.visible); // not sequence-related, not affected
900 * Set all types in all scope to hidden
902 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, true,
904 assertFalse(ann1.visible);
905 assertFalse(ann2.visible);
906 assertTrue(ann3.visible); // not sequence-related, not affected
907 assertFalse(ann4.visible);
908 assertFalse(ann5.visible);
909 assertTrue(ann6.visible); // not sequence-related, not affected
913 * Tests for the method that checks if one sequence cross-references another
915 @Test(groups = { "Functional" })
916 public void testHasCrossRef()
918 assertFalse(AlignmentUtils.hasCrossRef(null, null));
919 SequenceI seq1 = new Sequence("EMBL|A12345", "ABCDEF");
920 assertFalse(AlignmentUtils.hasCrossRef(seq1, null));
921 assertFalse(AlignmentUtils.hasCrossRef(null, seq1));
922 SequenceI seq2 = new Sequence("UNIPROT|V20192", "ABCDEF");
923 assertFalse(AlignmentUtils.hasCrossRef(seq1, seq2));
926 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "v20193"));
927 assertFalse(AlignmentUtils.hasCrossRef(seq1, seq2));
929 // case-insensitive; version number is ignored
930 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "v20192"));
931 assertTrue(AlignmentUtils.hasCrossRef(seq1, seq2));
934 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
935 assertTrue(AlignmentUtils.hasCrossRef(seq1, seq2));
936 // test is one-way only
937 assertFalse(AlignmentUtils.hasCrossRef(seq2, seq1));
941 * Tests for the method that checks if either sequence cross-references the
944 @Test(groups = { "Functional" })
945 public void testHaveCrossRef()
947 assertFalse(AlignmentUtils.hasCrossRef(null, null));
948 SequenceI seq1 = new Sequence("EMBL|A12345", "ABCDEF");
949 assertFalse(AlignmentUtils.haveCrossRef(seq1, null));
950 assertFalse(AlignmentUtils.haveCrossRef(null, seq1));
951 SequenceI seq2 = new Sequence("UNIPROT|V20192", "ABCDEF");
952 assertFalse(AlignmentUtils.haveCrossRef(seq1, seq2));
954 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
955 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
956 // next is true for haveCrossRef, false for hasCrossRef
957 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
959 // now the other way round
960 seq1.setDBRefs(null);
961 seq2.addDBRef(new DBRefEntry("EMBL", "1", "A12345"));
962 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
963 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
966 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
967 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
968 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
972 * Test the method that extracts the cds-only part of a dna alignment.
974 @Test(groups = { "Functional" })
975 public void testMakeCdsAlignment()
979 * dna1 --> [4, 6] [10,12] --> pep1
980 * dna2 --> [1, 3] [7, 9] [13,15] --> pep1
982 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
983 SequenceI dna2 = new Sequence("dna2", "GGGcccTTTaaaCCC");
984 SequenceI pep1 = new Sequence("pep1", "GF");
985 SequenceI pep2 = new Sequence("pep2", "GFP");
986 pep1.addDBRef(new DBRefEntry("UNIPROT", "0", "pep1"));
987 pep2.addDBRef(new DBRefEntry("UNIPROT", "0", "pep2"));
988 dna1.createDatasetSequence();
989 dna2.createDatasetSequence();
990 pep1.createDatasetSequence();
991 pep2.createDatasetSequence();
992 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
993 dna.setDataset(null);
996 * need a sourceDbRef if we are to construct dbrefs to the CDS
997 * sequence from the dna contig sequences
999 DBRefEntry dbref = new DBRefEntry("ENSEMBL", "0", "dna1");
1000 dna1.getDatasetSequence().addDBRef(dbref);
1001 org.testng.Assert.assertEquals(dbref, dna1.getPrimaryDBRefs().get(0));
1002 dbref = new DBRefEntry("ENSEMBL", "0", "dna2");
1003 dna2.getDatasetSequence().addDBRef(dbref);
1004 org.testng.Assert.assertEquals(dbref, dna2.getPrimaryDBRefs().get(0));
1007 * CDS sequences are 'discovered' from dna-to-protein mappings on the alignment
1008 * dataset (e.g. added from dbrefs by CrossRef.findXrefSequences)
1010 MapList mapfordna1 = new MapList(new int[] { 4, 6, 10, 12 },
1011 new int[] { 1, 2 }, 3, 1);
1012 AlignedCodonFrame acf = new AlignedCodonFrame();
1013 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(),
1015 dna.addCodonFrame(acf);
1016 MapList mapfordna2 = new MapList(new int[] { 1, 3, 7, 9, 13, 15 },
1019 acf = new AlignedCodonFrame();
1020 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(),
1022 dna.addCodonFrame(acf);
1025 * 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
1027 DBRefEntry dna1xref = new DBRefEntry("UNIPROT", "ENSEMBL", "pep1",
1028 new Mapping(mapfordna1));
1029 dna1.getDatasetSequence().addDBRef(dna1xref);
1030 DBRefEntry dna2xref = new DBRefEntry("UNIPROT", "ENSEMBL", "pep2",
1031 new Mapping(mapfordna2));
1032 dna2.getDatasetSequence().addDBRef(dna2xref);
1035 * execute method under test:
1037 AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
1038 dna1, dna2 }, dna.getDataset(), null);
1041 * verify cds sequences
1043 assertEquals(2, cds.getSequences().size());
1044 assertEquals("GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
1045 assertEquals("GGGTTTCCC", cds.getSequenceAt(1).getSequenceAsString());
1048 * verify shared, extended alignment dataset
1050 assertSame(dna.getDataset(), cds.getDataset());
1051 SequenceI cds1Dss = cds.getSequenceAt(0).getDatasetSequence();
1052 SequenceI cds2Dss = cds.getSequenceAt(1).getDatasetSequence();
1053 assertTrue(dna.getDataset().getSequences().contains(cds1Dss));
1054 assertTrue(dna.getDataset().getSequences().contains(cds2Dss));
1057 * verify CDS has a dbref with mapping to peptide
1059 assertNotNull(cds1Dss.getDBRefs());
1060 assertEquals(2, cds1Dss.getDBRefs().length);
1061 dbref = cds1Dss.getDBRefs()[0];
1062 assertEquals(dna1xref.getSource(), dbref.getSource());
1063 // version is via ensembl's primary ref
1064 assertEquals(dna1xref.getVersion(), dbref.getVersion());
1065 assertEquals(dna1xref.getAccessionId(), dbref.getAccessionId());
1066 assertNotNull(dbref.getMap());
1067 assertSame(pep1.getDatasetSequence(), dbref.getMap().getTo());
1068 MapList cdsMapping = new MapList(new int[] { 1, 6 },
1069 new int[] { 1, 2 }, 3, 1);
1070 assertEquals(cdsMapping, dbref.getMap().getMap());
1073 * verify peptide has added a dbref with reverse mapping to CDS
1075 assertNotNull(pep1.getDBRefs());
1076 // FIXME pep1.getDBRefs() is 1 - is that the correct behaviour ?
1077 assertEquals(2, pep1.getDBRefs().length);
1078 dbref = pep1.getDBRefs()[1];
1079 assertEquals("ENSEMBL", dbref.getSource());
1080 assertEquals("0", dbref.getVersion());
1081 assertEquals("CDS|dna1", dbref.getAccessionId());
1082 assertNotNull(dbref.getMap());
1083 assertSame(cds1Dss, dbref.getMap().getTo());
1084 assertEquals(cdsMapping.getInverse(), dbref.getMap().getMap());
1087 * Verify mappings from CDS to peptide, cDNA to CDS, and cDNA to peptide
1088 * the mappings are on the shared alignment dataset
1089 * 6 mappings, 2*(DNA->CDS), 2*(DNA->Pep), 2*(CDS->Pep)
1091 List<AlignedCodonFrame> cdsMappings = cds.getDataset().getCodonFrames();
1092 assertEquals(6, cdsMappings.size());
1095 * verify that mapping sets for dna and cds alignments are different
1096 * [not current behaviour - all mappings are on the alignment dataset]
1098 // select -> subselect type to test.
1099 // Assert.assertNotSame(dna.getCodonFrames(), cds.getCodonFrames());
1100 // assertEquals(4, dna.getCodonFrames().size());
1101 // assertEquals(4, cds.getCodonFrames().size());
1104 * Two mappings involve pep1 (dna to pep1, cds to pep1)
1105 * Mapping from pep1 to GGGTTT in first new exon sequence
1107 List<AlignedCodonFrame> pep1Mappings = MappingUtils
1108 .findMappingsForSequence(pep1, cdsMappings);
1109 assertEquals(2, pep1Mappings.size());
1110 List<AlignedCodonFrame> mappings = MappingUtils
1111 .findMappingsForSequence(cds.getSequenceAt(0), pep1Mappings);
1112 assertEquals(1, mappings.size());
1115 SearchResults sr = MappingUtils.buildSearchResults(pep1, 1, mappings);
1116 assertEquals(1, sr.getResults().size());
1117 Match m = sr.getResults().get(0);
1118 assertSame(cds1Dss, m.getSequence());
1119 assertEquals(1, m.getStart());
1120 assertEquals(3, m.getEnd());
1122 sr = MappingUtils.buildSearchResults(pep1, 2, mappings);
1123 m = sr.getResults().get(0);
1124 assertSame(cds1Dss, m.getSequence());
1125 assertEquals(4, m.getStart());
1126 assertEquals(6, m.getEnd());
1129 * Two mappings involve pep2 (dna to pep2, cds to pep2)
1130 * Verify mapping from pep2 to GGGTTTCCC in second new exon sequence
1132 List<AlignedCodonFrame> pep2Mappings = MappingUtils
1133 .findMappingsForSequence(pep2, cdsMappings);
1134 assertEquals(2, pep2Mappings.size());
1135 mappings = MappingUtils.findMappingsForSequence(cds.getSequenceAt(1),
1137 assertEquals(1, mappings.size());
1139 sr = MappingUtils.buildSearchResults(pep2, 1, mappings);
1140 assertEquals(1, sr.getResults().size());
1141 m = sr.getResults().get(0);
1142 assertSame(cds2Dss, m.getSequence());
1143 assertEquals(1, m.getStart());
1144 assertEquals(3, m.getEnd());
1146 sr = MappingUtils.buildSearchResults(pep2, 2, mappings);
1147 m = sr.getResults().get(0);
1148 assertSame(cds2Dss, m.getSequence());
1149 assertEquals(4, m.getStart());
1150 assertEquals(6, m.getEnd());
1152 sr = MappingUtils.buildSearchResults(pep2, 3, mappings);
1153 m = sr.getResults().get(0);
1154 assertSame(cds2Dss, m.getSequence());
1155 assertEquals(7, m.getStart());
1156 assertEquals(9, m.getEnd());
1160 * Test the method that makes a cds-only alignment from a DNA sequence and its
1161 * product mappings, for the case where there are multiple exon mappings to
1162 * different protein products.
1164 @Test(groups = { "Functional" })
1165 public void testMakeCdsAlignment_multipleProteins()
1167 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
1168 SequenceI pep1 = new Sequence("pep1", "GF"); // GGGTTT
1169 SequenceI pep2 = new Sequence("pep2", "KP"); // aaaccc
1170 SequenceI pep3 = new Sequence("pep3", "KF"); // aaaTTT
1171 dna1.createDatasetSequence();
1172 pep1.createDatasetSequence();
1173 pep2.createDatasetSequence();
1174 pep3.createDatasetSequence();
1175 pep1.getDatasetSequence().addDBRef(
1176 new DBRefEntry("EMBLCDS", "2", "A12345"));
1177 pep2.getDatasetSequence().addDBRef(
1178 new DBRefEntry("EMBLCDS", "3", "A12346"));
1179 pep3.getDatasetSequence().addDBRef(
1180 new DBRefEntry("EMBLCDS", "4", "A12347"));
1183 * Create the CDS alignment
1185 AlignmentI dna = new Alignment(new SequenceI[] { dna1 });
1186 dna.setDataset(null);
1189 * Make the mappings from dna to protein
1191 // map ...GGG...TTT to GF
1192 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1193 new int[] { 1, 2 }, 3, 1);
1194 AlignedCodonFrame acf = new AlignedCodonFrame();
1195 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
1196 dna.addCodonFrame(acf);
1198 // map aaa...ccc to KP
1199 map = new MapList(new int[] { 1, 3, 7, 9 }, new int[] { 1, 2 }, 3, 1);
1200 acf = new AlignedCodonFrame();
1201 acf.addMap(dna1.getDatasetSequence(), pep2.getDatasetSequence(), map);
1202 dna.addCodonFrame(acf);
1204 // map aaa......TTT to KF
1205 map = new MapList(new int[] { 1, 3, 10, 12 }, new int[] { 1, 2 }, 3, 1);
1206 acf = new AlignedCodonFrame();
1207 acf.addMap(dna1.getDatasetSequence(), pep3.getDatasetSequence(), map);
1208 dna.addCodonFrame(acf);
1211 * execute method under test
1213 AlignmentI cdsal = AlignmentUtils.makeCdsAlignment(
1214 new SequenceI[] { dna1 }, dna.getDataset(), null);
1217 * Verify we have 3 cds sequences, mapped to pep1/2/3 respectively
1219 List<SequenceI> cds = cdsal.getSequences();
1220 assertEquals(3, cds.size());
1223 * verify shared, extended alignment dataset
1225 assertSame(cdsal.getDataset(), dna.getDataset());
1226 assertTrue(dna.getDataset().getSequences()
1227 .contains(cds.get(0).getDatasetSequence()));
1228 assertTrue(dna.getDataset().getSequences()
1229 .contains(cds.get(1).getDatasetSequence()));
1230 assertTrue(dna.getDataset().getSequences()
1231 .contains(cds.get(2).getDatasetSequence()));
1234 * verify aligned cds sequences and their xrefs
1236 SequenceI cdsSeq = cds.get(0);
1237 assertEquals("GGGTTT", cdsSeq.getSequenceAsString());
1238 // assertEquals("dna1|A12345", cdsSeq.getName());
1239 assertEquals("CDS|dna1", cdsSeq.getName());
1240 // assertEquals(1, cdsSeq.getDBRefs().length);
1241 // DBRefEntry cdsRef = cdsSeq.getDBRefs()[0];
1242 // assertEquals("EMBLCDS", cdsRef.getSource());
1243 // assertEquals("2", cdsRef.getVersion());
1244 // assertEquals("A12345", cdsRef.getAccessionId());
1246 cdsSeq = cds.get(1);
1247 assertEquals("aaaccc", cdsSeq.getSequenceAsString());
1248 // assertEquals("dna1|A12346", cdsSeq.getName());
1249 assertEquals("CDS|dna1", cdsSeq.getName());
1250 // assertEquals(1, cdsSeq.getDBRefs().length);
1251 // cdsRef = cdsSeq.getDBRefs()[0];
1252 // assertEquals("EMBLCDS", cdsRef.getSource());
1253 // assertEquals("3", cdsRef.getVersion());
1254 // assertEquals("A12346", cdsRef.getAccessionId());
1256 cdsSeq = cds.get(2);
1257 assertEquals("aaaTTT", cdsSeq.getSequenceAsString());
1258 // assertEquals("dna1|A12347", cdsSeq.getName());
1259 assertEquals("CDS|dna1", cdsSeq.getName());
1260 // assertEquals(1, cdsSeq.getDBRefs().length);
1261 // cdsRef = cdsSeq.getDBRefs()[0];
1262 // assertEquals("EMBLCDS", cdsRef.getSource());
1263 // assertEquals("4", cdsRef.getVersion());
1264 // assertEquals("A12347", cdsRef.getAccessionId());
1267 * Verify there are mappings from each cds sequence to its protein product
1268 * and also to its dna source
1270 List<AlignedCodonFrame> newMappings = cdsal.getCodonFrames();
1273 * 6 mappings involve dna1 (to pep1/2/3, cds1/2/3)
1275 List<AlignedCodonFrame> dnaMappings = MappingUtils
1276 .findMappingsForSequence(dna1, newMappings);
1277 assertEquals(6, dnaMappings.size());
1282 List<AlignedCodonFrame> mappings = MappingUtils
1283 .findMappingsForSequence(pep1, dnaMappings);
1284 assertEquals(1, mappings.size());
1285 assertEquals(1, mappings.get(0).getMappings().size());
1286 assertSame(pep1.getDatasetSequence(), mappings.get(0).getMappings()
1287 .get(0).getMapping().getTo());
1292 List<AlignedCodonFrame> dnaToCds1Mappings = MappingUtils
1293 .findMappingsForSequence(cds.get(0), dnaMappings);
1294 Mapping mapping = dnaToCds1Mappings.get(0).getMappings().get(0)
1296 assertSame(cds.get(0).getDatasetSequence(), mapping
1298 assertEquals("G(1) in CDS should map to G(4) in DNA", 4, mapping
1299 .getMap().getToPosition(1));
1304 mappings = MappingUtils.findMappingsForSequence(pep2, dnaMappings);
1305 assertEquals(1, mappings.size());
1306 assertEquals(1, mappings.get(0).getMappings().size());
1307 assertSame(pep2.getDatasetSequence(), mappings.get(0).getMappings()
1308 .get(0).getMapping().getTo());
1313 List<AlignedCodonFrame> dnaToCds2Mappings = MappingUtils
1314 .findMappingsForSequence(cds.get(1), dnaMappings);
1315 mapping = dnaToCds2Mappings.get(0).getMappings().get(0).getMapping();
1316 assertSame(cds.get(1).getDatasetSequence(), mapping.getTo());
1317 assertEquals("c(4) in CDS should map to c(7) in DNA", 7, mapping
1318 .getMap().getToPosition(4));
1323 mappings = MappingUtils.findMappingsForSequence(pep3, dnaMappings);
1324 assertEquals(1, mappings.size());
1325 assertEquals(1, mappings.get(0).getMappings().size());
1326 assertSame(pep3.getDatasetSequence(), mappings.get(0).getMappings()
1327 .get(0).getMapping().getTo());
1332 List<AlignedCodonFrame> dnaToCds3Mappings = MappingUtils
1333 .findMappingsForSequence(cds.get(2), dnaMappings);
1334 mapping = dnaToCds3Mappings.get(0).getMappings().get(0).getMapping();
1335 assertSame(cds.get(2).getDatasetSequence(), mapping.getTo());
1336 assertEquals("T(4) in CDS should map to T(10) in DNA", 10, mapping
1337 .getMap().getToPosition(4));
1340 @Test(groups = { "Functional" })
1341 public void testIsMappable()
1343 SequenceI dna1 = new Sequence("dna1", "cgCAGtgGT");
1344 SequenceI aa1 = new Sequence("aa1", "RSG");
1345 AlignmentI al1 = new Alignment(new SequenceI[] { dna1 });
1346 AlignmentI al2 = new Alignment(new SequenceI[] { aa1 });
1348 assertFalse(AlignmentUtils.isMappable(null, null));
1349 assertFalse(AlignmentUtils.isMappable(al1, null));
1350 assertFalse(AlignmentUtils.isMappable(null, al1));
1351 assertFalse(AlignmentUtils.isMappable(al1, al1));
1352 assertFalse(AlignmentUtils.isMappable(al2, al2));
1354 assertTrue(AlignmentUtils.isMappable(al1, al2));
1355 assertTrue(AlignmentUtils.isMappable(al2, al1));
1359 * Test creating a mapping when the sequences involved do not start at residue
1362 * @throws IOException
1364 @Test(groups = { "Functional" })
1365 public void testMapCdnaToProtein_forSubsequence()
1368 SequenceI prot = new Sequence("UNIPROT|V12345", "E-I--Q", 10, 12);
1369 prot.createDatasetSequence();
1371 SequenceI dna = new Sequence("EMBL|A33333", "GAA--AT-C-CAG", 40, 48);
1372 dna.createDatasetSequence();
1374 MapList map = AlignmentUtils.mapCdnaToProtein(prot, dna);
1375 assertEquals(10, map.getToLowest());
1376 assertEquals(12, map.getToHighest());
1377 assertEquals(40, map.getFromLowest());
1378 assertEquals(48, map.getFromHighest());
1382 * Test for the alignSequenceAs method where we have protein mapped to protein
1384 @Test(groups = { "Functional" })
1385 public void testAlignSequenceAs_mappedProteinProtein()
1388 SequenceI alignMe = new Sequence("Match", "MGAASEV");
1389 alignMe.createDatasetSequence();
1390 SequenceI alignFrom = new Sequence("Query", "LQTGYMGAASEVMFSPTRR");
1391 alignFrom.createDatasetSequence();
1393 AlignedCodonFrame acf = new AlignedCodonFrame();
1394 // this is like a domain or motif match of part of a peptide sequence
1395 MapList map = new MapList(new int[] { 6, 12 }, new int[] { 1, 7 }, 1, 1);
1396 acf.addMap(alignFrom.getDatasetSequence(),
1397 alignMe.getDatasetSequence(), map);
1399 AlignmentUtils.alignSequenceAs(alignMe, alignFrom, acf, "-", '-', true,
1401 assertEquals("-----MGAASEV-------", alignMe.getSequenceAsString());
1405 * Test for the alignSequenceAs method where there are trailing unmapped
1406 * residues in the model sequence
1408 @Test(groups = { "Functional" })
1409 public void testAlignSequenceAs_withTrailingPeptide()
1411 // map first 3 codons to KPF; G is a trailing unmapped residue
1412 MapList map = new MapList(new int[] { 1, 9 }, new int[] { 1, 3 }, 3, 1);
1414 checkAlignSequenceAs("AAACCCTTT", "K-PFG", true, true, map,
1419 * Tests for transferring features between mapped sequences
1421 @Test(groups = { "Functional" })
1422 public void testTransferFeatures()
1424 SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
1425 SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
1428 dna.addSequenceFeature(new SequenceFeature("type1", "desc1", 1, 2, 1f,
1430 // partial overlap - to [1, 1]
1431 dna.addSequenceFeature(new SequenceFeature("type2", "desc2", 3, 4, 2f,
1433 // exact overlap - to [1, 3]
1434 dna.addSequenceFeature(new SequenceFeature("type3", "desc3", 4, 6, 3f,
1436 // spanning overlap - to [2, 5]
1437 dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
1439 // exactly overlaps whole mapped range [1, 6]
1440 dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
1442 // no overlap (internal)
1443 dna.addSequenceFeature(new SequenceFeature("type6", "desc6", 7, 9, 6f,
1445 // no overlap (3' end)
1446 dna.addSequenceFeature(new SequenceFeature("type7", "desc7", 13, 15,
1448 // overlap (3' end) - to [6, 6]
1449 dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
1451 // extended overlap - to [6, +]
1452 dna.addSequenceFeature(new SequenceFeature("type9", "desc9", 12, 13,
1455 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1456 new int[] { 1, 6 }, 1, 1);
1459 * transferFeatures() will build 'partial overlap' for regions
1460 * that partially overlap 5' or 3' (start or end) of target sequence
1462 AlignmentUtils.transferFeatures(dna, cds, map, null);
1463 SequenceFeature[] sfs = cds.getSequenceFeatures();
1464 assertEquals(6, sfs.length);
1466 SequenceFeature sf = sfs[0];
1467 assertEquals("type2", sf.getType());
1468 assertEquals("desc2", sf.getDescription());
1469 assertEquals(2f, sf.getScore());
1470 assertEquals(1, sf.getBegin());
1471 assertEquals(1, sf.getEnd());
1474 assertEquals("type3", sf.getType());
1475 assertEquals("desc3", sf.getDescription());
1476 assertEquals(3f, sf.getScore());
1477 assertEquals(1, sf.getBegin());
1478 assertEquals(3, sf.getEnd());
1481 assertEquals("type4", sf.getType());
1482 assertEquals(2, sf.getBegin());
1483 assertEquals(5, sf.getEnd());
1486 assertEquals("type5", sf.getType());
1487 assertEquals(1, sf.getBegin());
1488 assertEquals(6, sf.getEnd());
1491 assertEquals("type8", sf.getType());
1492 assertEquals(6, sf.getBegin());
1493 assertEquals(6, sf.getEnd());
1496 assertEquals("type9", sf.getType());
1497 assertEquals(6, sf.getBegin());
1498 assertEquals(6, sf.getEnd());
1502 * Tests for transferring features between mapped sequences
1504 @Test(groups = { "Functional" })
1505 public void testTransferFeatures_withOmit()
1507 SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
1508 SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
1510 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1511 new int[] { 1, 6 }, 1, 1);
1513 // [5, 11] maps to [2, 5]
1514 dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
1516 // [4, 12] maps to [1, 6]
1517 dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
1519 // [12, 12] maps to [6, 6]
1520 dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
1523 // desc4 and desc8 are the 'omit these' varargs
1524 AlignmentUtils.transferFeatures(dna, cds, map, null, "type4", "type8");
1525 SequenceFeature[] sfs = cds.getSequenceFeatures();
1526 assertEquals(1, sfs.length);
1528 SequenceFeature sf = sfs[0];
1529 assertEquals("type5", sf.getType());
1530 assertEquals(1, sf.getBegin());
1531 assertEquals(6, sf.getEnd());
1535 * Tests for transferring features between mapped sequences
1537 @Test(groups = { "Functional" })
1538 public void testTransferFeatures_withSelect()
1540 SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
1541 SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
1543 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1544 new int[] { 1, 6 }, 1, 1);
1546 // [5, 11] maps to [2, 5]
1547 dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
1549 // [4, 12] maps to [1, 6]
1550 dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
1552 // [12, 12] maps to [6, 6]
1553 dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
1556 // "type5" is the 'select this type' argument
1557 AlignmentUtils.transferFeatures(dna, cds, map, "type5");
1558 SequenceFeature[] sfs = cds.getSequenceFeatures();
1559 assertEquals(1, sfs.length);
1561 SequenceFeature sf = sfs[0];
1562 assertEquals("type5", sf.getType());
1563 assertEquals(1, sf.getBegin());
1564 assertEquals(6, sf.getEnd());
1568 * Test the method that extracts the cds-only part of a dna alignment, for the
1569 * case where the cds should be aligned to match its nucleotide sequence.
1571 @Test(groups = { "Functional" })
1572 public void testMakeCdsAlignment_alternativeTranscripts()
1574 SequenceI dna1 = new Sequence("dna1", "aaaGGGCC-----CTTTaaaGGG");
1575 // alternative transcript of same dna skips CCC codon
1576 SequenceI dna2 = new Sequence("dna2", "aaaGGGCC-----cttTaaaGGG");
1577 // dna3 has no mapping (protein product) so should be ignored here
1578 SequenceI dna3 = new Sequence("dna3", "aaaGGGCCCCCGGGcttTaaaGGG");
1579 SequenceI pep1 = new Sequence("pep1", "GPFG");
1580 SequenceI pep2 = new Sequence("pep2", "GPG");
1581 dna1.createDatasetSequence();
1582 dna2.createDatasetSequence();
1583 dna3.createDatasetSequence();
1584 pep1.createDatasetSequence();
1585 pep2.createDatasetSequence();
1587 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
1588 dna.setDataset(null);
1590 MapList map = new MapList(new int[] { 4, 12, 16, 18 },
1591 new int[] { 1, 4 }, 3, 1);
1592 AlignedCodonFrame acf = new AlignedCodonFrame();
1593 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
1594 dna.addCodonFrame(acf);
1595 map = new MapList(new int[] { 4, 8, 12, 12, 16, 18 },
1598 acf = new AlignedCodonFrame();
1599 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map);
1600 dna.addCodonFrame(acf);
1602 AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
1603 dna1, dna2, dna3 }, dna.getDataset(), null);
1604 List<SequenceI> cdsSeqs = cds.getSequences();
1605 assertEquals(2, cdsSeqs.size());
1606 assertEquals("GGGCCCTTTGGG", cdsSeqs.get(0).getSequenceAsString());
1607 assertEquals("GGGCCTGGG", cdsSeqs.get(1).getSequenceAsString());
1610 * verify shared, extended alignment dataset
1612 assertSame(dna.getDataset(), cds.getDataset());
1613 assertTrue(dna.getDataset().getSequences()
1614 .contains(cdsSeqs.get(0).getDatasetSequence()));
1615 assertTrue(dna.getDataset().getSequences()
1616 .contains(cdsSeqs.get(1).getDatasetSequence()));
1619 * Verify 6 mappings: dna1 to cds1, cds1 to pep1, dna1 to pep1
1620 * and the same for dna2/cds2/pep2
1622 List<AlignedCodonFrame> mappings = cds.getCodonFrames();
1623 assertEquals(6, mappings.size());
1626 * 2 mappings involve pep1
1628 List<AlignedCodonFrame> pep1Mappings = MappingUtils
1629 .findMappingsForSequence(pep1, mappings);
1630 assertEquals(2, pep1Mappings.size());
1633 * Get mapping of pep1 to cds1 and verify it
1634 * maps GPFG to 1-3,4-6,7-9,10-12
1636 List<AlignedCodonFrame> pep1CdsMappings = MappingUtils
1637 .findMappingsForSequence(cds.getSequenceAt(0), pep1Mappings);
1638 assertEquals(1, pep1CdsMappings.size());
1639 SearchResults sr = MappingUtils.buildSearchResults(pep1, 1,
1641 assertEquals(1, sr.getResults().size());
1642 Match m = sr.getResults().get(0);
1643 assertEquals(cds.getSequenceAt(0).getDatasetSequence(),
1645 assertEquals(1, m.getStart());
1646 assertEquals(3, m.getEnd());
1647 sr = MappingUtils.buildSearchResults(pep1, 2, pep1CdsMappings);
1648 m = sr.getResults().get(0);
1649 assertEquals(4, m.getStart());
1650 assertEquals(6, m.getEnd());
1651 sr = MappingUtils.buildSearchResults(pep1, 3, pep1CdsMappings);
1652 m = sr.getResults().get(0);
1653 assertEquals(7, m.getStart());
1654 assertEquals(9, m.getEnd());
1655 sr = MappingUtils.buildSearchResults(pep1, 4, pep1CdsMappings);
1656 m = sr.getResults().get(0);
1657 assertEquals(10, m.getStart());
1658 assertEquals(12, m.getEnd());
1661 * Get mapping of pep2 to cds2 and verify it
1662 * maps GPG in pep2 to 1-3,4-6,7-9 in second CDS sequence
1664 List<AlignedCodonFrame> pep2Mappings = MappingUtils
1665 .findMappingsForSequence(pep2, mappings);
1666 assertEquals(2, pep2Mappings.size());
1667 List<AlignedCodonFrame> pep2CdsMappings = MappingUtils
1668 .findMappingsForSequence(cds.getSequenceAt(1), pep2Mappings);
1669 assertEquals(1, pep2CdsMappings.size());
1670 sr = MappingUtils.buildSearchResults(pep2, 1, pep2CdsMappings);
1671 assertEquals(1, sr.getResults().size());
1672 m = sr.getResults().get(0);
1673 assertEquals(cds.getSequenceAt(1).getDatasetSequence(),
1675 assertEquals(1, m.getStart());
1676 assertEquals(3, m.getEnd());
1677 sr = MappingUtils.buildSearchResults(pep2, 2, pep2CdsMappings);
1678 m = sr.getResults().get(0);
1679 assertEquals(4, m.getStart());
1680 assertEquals(6, m.getEnd());
1681 sr = MappingUtils.buildSearchResults(pep2, 3, pep2CdsMappings);
1682 m = sr.getResults().get(0);
1683 assertEquals(7, m.getStart());
1684 assertEquals(9, m.getEnd());
1688 * Test the method that realigns protein to match mapped codon alignment.
1690 @Test(groups = { "Functional" })
1691 public void testAlignProteinAsDna_incompleteStartCodon()
1693 // seq1: incomplete start codon (not mapped), then [3, 11]
1694 SequenceI dna1 = new Sequence("Seq1", "ccAAA-TTT-GGG-");
1695 // seq2 codons are [4, 5], [8, 11]
1696 SequenceI dna2 = new Sequence("Seq2", "ccaAA-ttT-GGG-");
1697 // seq3 incomplete start codon at 'tt'
1698 SequenceI dna3 = new Sequence("Seq3", "ccaaa-ttt-GGG-");
1699 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
1700 dna.setDataset(null);
1702 // prot1 has 'X' for incomplete start codon (not mapped)
1703 SequenceI prot1 = new Sequence("Seq1", "XKFG"); // X for incomplete start
1704 SequenceI prot2 = new Sequence("Seq2", "NG");
1705 SequenceI prot3 = new Sequence("Seq3", "XG"); // X for incomplete start
1706 AlignmentI protein = new Alignment(new SequenceI[] { prot1, prot2,
1708 protein.setDataset(null);
1710 // map dna1 [3, 11] to prot1 [2, 4] KFG
1711 MapList map = new MapList(new int[] { 3, 11 }, new int[] { 2, 4 }, 3, 1);
1712 AlignedCodonFrame acf = new AlignedCodonFrame();
1713 acf.addMap(dna1.getDatasetSequence(), prot1.getDatasetSequence(), map);
1715 // map dna2 [4, 5] [8, 11] to prot2 [1, 2] NG
1716 map = new MapList(new int[] { 4, 5, 8, 11 }, new int[] { 1, 2 }, 3, 1);
1717 acf.addMap(dna2.getDatasetSequence(), prot2.getDatasetSequence(), map);
1719 // map dna3 [9, 11] to prot3 [2, 2] G
1720 map = new MapList(new int[] { 9, 11 }, new int[] { 2, 2 }, 3, 1);
1721 acf.addMap(dna3.getDatasetSequence(), prot3.getDatasetSequence(), map);
1723 ArrayList<AlignedCodonFrame> acfs = new ArrayList<AlignedCodonFrame>();
1725 protein.setCodonFrames(acfs);
1728 * verify X is included in the aligned proteins, and placed just
1729 * before the first mapped residue
1730 * CCT is between CCC and TTT
1732 AlignmentUtils.alignProteinAsDna(protein, dna);
1733 assertEquals("XK-FG", prot1.getSequenceAsString());
1734 assertEquals("--N-G", prot2.getSequenceAsString());
1735 assertEquals("---XG", prot3.getSequenceAsString());
1739 * Tests for the method that maps the subset of a dna sequence that has CDS
1740 * (or subtype) feature - case where the start codon is incomplete.
1742 @Test(groups = "Functional")
1743 public void testFindCdsPositions_fivePrimeIncomplete()
1745 SequenceI dnaSeq = new Sequence("dna", "aaagGGCCCaaaTTTttt");
1746 dnaSeq.createDatasetSequence();
1747 SequenceI ds = dnaSeq.getDatasetSequence();
1749 // CDS for dna 5-6 (incomplete codon), 7-9
1750 SequenceFeature sf = new SequenceFeature("CDS", "", 5, 9, 0f, null);
1751 sf.setPhase("2"); // skip 2 bases to start of next codon
1752 ds.addSequenceFeature(sf);
1753 // CDS for dna 13-15
1754 sf = new SequenceFeature("CDS_predicted", "", 13, 15, 0f, null);
1755 ds.addSequenceFeature(sf);
1757 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
1760 * check the mapping starts with the first complete codon
1762 assertEquals(6, MappingUtils.getLength(ranges));
1763 assertEquals(2, ranges.size());
1764 assertEquals(7, ranges.get(0)[0]);
1765 assertEquals(9, ranges.get(0)[1]);
1766 assertEquals(13, ranges.get(1)[0]);
1767 assertEquals(15, ranges.get(1)[1]);
1771 * Tests for the method that maps the subset of a dna sequence that has CDS
1772 * (or subtype) feature.
1774 @Test(groups = "Functional")
1775 public void testFindCdsPositions()
1777 SequenceI dnaSeq = new Sequence("dna", "aaaGGGcccAAATTTttt");
1778 dnaSeq.createDatasetSequence();
1779 SequenceI ds = dnaSeq.getDatasetSequence();
1781 // CDS for dna 10-12
1782 SequenceFeature sf = new SequenceFeature("CDS_predicted", "", 10, 12,
1785 ds.addSequenceFeature(sf);
1787 sf = new SequenceFeature("CDS", "", 4, 6, 0f, null);
1789 ds.addSequenceFeature(sf);
1790 // exon feature should be ignored here
1791 sf = new SequenceFeature("exon", "", 7, 9, 0f, null);
1792 ds.addSequenceFeature(sf);
1794 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
1796 * verify ranges { [4-6], [12-10] }
1797 * note CDS ranges are ordered ascending even if the CDS
1800 assertEquals(6, MappingUtils.getLength(ranges));
1801 assertEquals(2, ranges.size());
1802 assertEquals(4, ranges.get(0)[0]);
1803 assertEquals(6, ranges.get(0)[1]);
1804 assertEquals(10, ranges.get(1)[0]);
1805 assertEquals(12, ranges.get(1)[1]);
1809 * Test the method that computes a map of codon variants for each protein
1810 * position from "sequence_variant" features on dna
1812 @Test(groups = "Functional")
1813 public void testBuildDnaVariantsMap()
1815 SequenceI dna = new Sequence("dna", "atgAAATTTGGGCCCtag");
1816 MapList map = new MapList(new int[] { 1, 18 }, new int[] { 1, 5 }, 3, 1);
1819 * first with no variants on dna
1821 LinkedHashMap<Integer, List<DnaVariant>[]> variantsMap = AlignmentUtils
1822 .buildDnaVariantsMap(dna, map);
1823 assertTrue(variantsMap.isEmpty());
1826 * single allele codon 1, on base 1
1828 SequenceFeature sf1 = new SequenceFeature("sequence_variant", "", 1, 1,
1830 sf1.setValue("alleles", "T");
1831 sf1.setValue("ID", "sequence_variant:rs758803211");
1832 dna.addSequenceFeature(sf1);
1835 * two alleles codon 2, on bases 2 and 3 (distinct variants)
1837 SequenceFeature sf2 = new SequenceFeature("sequence_variant", "", 5, 5,
1839 sf2.setValue("alleles", "T");
1840 sf2.setValue("ID", "sequence_variant:rs758803212");
1841 dna.addSequenceFeature(sf2);
1842 SequenceFeature sf3 = new SequenceFeature("sequence_variant", "", 6, 6,
1844 sf3.setValue("alleles", "G");
1845 sf3.setValue("ID", "sequence_variant:rs758803213");
1846 dna.addSequenceFeature(sf3);
1849 * two alleles codon 3, both on base 2 (one variant)
1851 SequenceFeature sf4 = new SequenceFeature("sequence_variant", "", 8, 8,
1853 sf4.setValue("alleles", "C, G");
1854 sf4.setValue("ID", "sequence_variant:rs758803214");
1855 dna.addSequenceFeature(sf4);
1857 // no alleles on codon 4
1860 * alleles on codon 5 on all 3 bases (distinct variants)
1862 SequenceFeature sf5 = new SequenceFeature("sequence_variant", "", 13,
1864 sf5.setValue("alleles", "C, G"); // (C duplicates given base value)
1865 sf5.setValue("ID", "sequence_variant:rs758803215");
1866 dna.addSequenceFeature(sf5);
1867 SequenceFeature sf6 = new SequenceFeature("sequence_variant", "", 14,
1869 sf6.setValue("alleles", "g, a"); // should force to upper-case
1870 sf6.setValue("ID", "sequence_variant:rs758803216");
1871 dna.addSequenceFeature(sf6);
1872 SequenceFeature sf7 = new SequenceFeature("sequence_variant", "", 15,
1874 sf7.setValue("alleles", "A, T");
1875 sf7.setValue("ID", "sequence_variant:rs758803217");
1876 dna.addSequenceFeature(sf7);
1879 * build map - expect variants on positions 1, 2, 3, 5
1881 variantsMap = AlignmentUtils.buildDnaVariantsMap(dna, map);
1882 assertEquals(4, variantsMap.size());
1885 * protein residue 1: variant on codon (ATG) base 1, not on 2 or 3
1887 List<DnaVariant>[] pep1Variants = variantsMap.get(1);
1888 assertEquals(3, pep1Variants.length);
1889 assertEquals(1, pep1Variants[0].size());
1890 assertEquals("A", pep1Variants[0].get(0).base); // codon[1] base
1891 assertSame(sf1, pep1Variants[0].get(0).variant); // codon[1] variant
1892 assertEquals(1, pep1Variants[1].size());
1893 assertEquals("T", pep1Variants[1].get(0).base); // codon[2] base
1894 assertNull(pep1Variants[1].get(0).variant); // no variant here
1895 assertEquals(1, pep1Variants[2].size());
1896 assertEquals("G", pep1Variants[2].get(0).base); // codon[3] base
1897 assertNull(pep1Variants[2].get(0).variant); // no variant here
1900 * protein residue 2: variants on codon (AAA) bases 2 and 3
1902 List<DnaVariant>[] pep2Variants = variantsMap.get(2);
1903 assertEquals(3, pep2Variants.length);
1904 assertEquals(1, pep2Variants[0].size());
1905 // codon[1] base recorded while processing variant on codon[2]
1906 assertEquals("A", pep2Variants[0].get(0).base);
1907 assertNull(pep2Variants[0].get(0).variant); // no variant here
1908 // codon[2] base and variant:
1909 assertEquals(1, pep2Variants[1].size());
1910 assertEquals("A", pep2Variants[1].get(0).base);
1911 assertSame(sf2, pep2Variants[1].get(0).variant);
1912 // codon[3] base was recorded when processing codon[2] variant
1913 // and then the variant for codon[3] added to it
1914 assertEquals(1, pep2Variants[2].size());
1915 assertEquals("A", pep2Variants[2].get(0).base);
1916 assertSame(sf3, pep2Variants[2].get(0).variant);
1919 * protein residue 3: variants on codon (TTT) base 2 only
1921 List<DnaVariant>[] pep3Variants = variantsMap.get(3);
1922 assertEquals(3, pep3Variants.length);
1923 assertEquals(1, pep3Variants[0].size());
1924 assertEquals("T", pep3Variants[0].get(0).base); // codon[1] base
1925 assertNull(pep3Variants[0].get(0).variant); // no variant here
1926 assertEquals(1, pep3Variants[1].size());
1927 assertEquals("T", pep3Variants[1].get(0).base); // codon[2] base
1928 assertSame(sf4, pep3Variants[1].get(0).variant); // codon[2] variant
1929 assertEquals(1, pep3Variants[2].size());
1930 assertEquals("T", pep3Variants[2].get(0).base); // codon[3] base
1931 assertNull(pep3Variants[2].get(0).variant); // no variant here
1934 * three variants on protein position 5
1936 List<DnaVariant>[] pep5Variants = variantsMap.get(5);
1937 assertEquals(3, pep5Variants.length);
1938 assertEquals(1, pep5Variants[0].size());
1939 assertEquals("C", pep5Variants[0].get(0).base); // codon[1] base
1940 assertSame(sf5, pep5Variants[0].get(0).variant); // codon[1] variant
1941 assertEquals(1, pep5Variants[1].size());
1942 assertEquals("C", pep5Variants[1].get(0).base); // codon[2] base
1943 assertSame(sf6, pep5Variants[1].get(0).variant); // codon[2] variant
1944 assertEquals(1, pep5Variants[2].size());
1945 assertEquals("C", pep5Variants[2].get(0).base); // codon[3] base
1946 assertSame(sf7, pep5Variants[2].get(0).variant); // codon[3] variant
1950 * Tests for the method that computes all peptide variants given codon
1953 @Test(groups = "Functional")
1954 public void testComputePeptideVariants()
1957 * scenario: AAATTTCCC codes for KFP
1959 * GAA -> E source: Ensembl
1960 * CAA -> Q source: dbSNP
1961 * AAG synonymous source: COSMIC
1962 * AAT -> N source: Ensembl
1963 * ...TTC synonymous source: dbSNP
1964 * ......CAC,CGC -> H,R source: COSMIC
1965 * (one variant with two alleles)
1967 SequenceI peptide = new Sequence("pep/10-12", "KFP");
1970 * two distinct variants for codon 1 position 1
1971 * second one has clinical significance
1973 String ensembl = "Ensembl";
1974 String dbSnp = "dbSNP";
1975 String cosmic = "COSMIC";
1976 SequenceFeature sf1 = new SequenceFeature("sequence_variant", "", 1, 1,
1978 sf1.setValue("alleles", "A,G"); // GAA -> E
1979 sf1.setValue("ID", "var1.125A>G");
1980 SequenceFeature sf2 = new SequenceFeature("sequence_variant", "", 1, 1,
1982 sf2.setValue("alleles", "A,C"); // CAA -> Q
1983 sf2.setValue("ID", "var2");
1984 sf2.setValue("clinical_significance", "Dodgy");
1985 SequenceFeature sf3 = new SequenceFeature("sequence_variant", "", 3, 3,
1987 sf3.setValue("alleles", "A,G"); // synonymous
1988 sf3.setValue("ID", "var3");
1989 sf3.setValue("clinical_significance", "None");
1990 SequenceFeature sf4 = new SequenceFeature("sequence_variant", "", 3, 3,
1992 sf4.setValue("alleles", "A,T"); // AAT -> N
1993 sf4.setValue("ID", "sequence_variant:var4"); // prefix gets stripped off
1994 sf4.setValue("clinical_significance", "Benign");
1995 SequenceFeature sf5 = new SequenceFeature("sequence_variant", "", 6, 6,
1997 sf5.setValue("alleles", "T,C"); // synonymous
1998 sf5.setValue("ID", "var5");
1999 sf5.setValue("clinical_significance", "Bad");
2000 SequenceFeature sf6 = new SequenceFeature("sequence_variant", "", 8, 8,
2002 sf6.setValue("alleles", "C,A,G"); // CAC,CGC -> H,R
2003 sf6.setValue("ID", "var6");
2004 sf6.setValue("clinical_significance", "Good");
2006 List<DnaVariant> codon1Variants = new ArrayList<DnaVariant>();
2007 List<DnaVariant> codon2Variants = new ArrayList<DnaVariant>();
2008 List<DnaVariant> codon3Variants = new ArrayList<DnaVariant>();
2009 List<DnaVariant> codonVariants[] = new ArrayList[3];
2010 codonVariants[0] = codon1Variants;
2011 codonVariants[1] = codon2Variants;
2012 codonVariants[2] = codon3Variants;
2015 * compute variants for protein position 1
2017 codon1Variants.add(new DnaVariant("A", sf1));
2018 codon1Variants.add(new DnaVariant("A", sf2));
2019 codon2Variants.add(new DnaVariant("A"));
2020 codon2Variants.add(new DnaVariant("A"));
2021 codon3Variants.add(new DnaVariant("A", sf3));
2022 codon3Variants.add(new DnaVariant("A", sf4));
2023 AlignmentUtils.computePeptideVariants(peptide, 1, codonVariants);
2026 * compute variants for protein position 2
2028 codon1Variants.clear();
2029 codon2Variants.clear();
2030 codon3Variants.clear();
2031 codon1Variants.add(new DnaVariant("T"));
2032 codon2Variants.add(new DnaVariant("T"));
2033 codon3Variants.add(new DnaVariant("T", sf5));
2034 AlignmentUtils.computePeptideVariants(peptide, 2, codonVariants);
2037 * compute variants for protein position 3
2039 codon1Variants.clear();
2040 codon2Variants.clear();
2041 codon3Variants.clear();
2042 codon1Variants.add(new DnaVariant("C"));
2043 codon2Variants.add(new DnaVariant("C", sf6));
2044 codon3Variants.add(new DnaVariant("C"));
2045 AlignmentUtils.computePeptideVariants(peptide, 3, codonVariants);
2048 * verify added sequence features for
2049 * var1 K -> E Ensembl
2051 * var4 K -> N Ensembl
2052 * var6 P -> H COSMIC
2053 * var6 P -> R COSMIC
2055 SequenceFeature[] sfs = peptide.getSequenceFeatures();
2056 assertEquals(5, sfs.length);
2058 SequenceFeature sf = sfs[0];
2059 assertEquals(1, sf.getBegin());
2060 assertEquals(1, sf.getEnd());
2061 assertEquals("p.Lys1Glu", sf.getDescription());
2062 assertEquals("var1.125A>G", sf.getValue("ID"));
2063 assertNull(sf.getValue("clinical_significance"));
2064 assertEquals("ID=var1.125A>G", sf.getAttributes());
2065 assertEquals(1, sf.links.size());
2066 // link to variation is urlencoded
2068 "p.Lys1Glu var1.125A>G|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var1.125A%3EG",
2070 assertEquals(ensembl, sf.getFeatureGroup());
2073 assertEquals(1, sf.getBegin());
2074 assertEquals(1, sf.getEnd());
2075 assertEquals("p.Lys1Gln", sf.getDescription());
2076 assertEquals("var2", sf.getValue("ID"));
2077 assertEquals("Dodgy", sf.getValue("clinical_significance"));
2078 assertEquals("ID=var2;clinical_significance=Dodgy", sf.getAttributes());
2079 assertEquals(1, sf.links.size());
2081 "p.Lys1Gln var2|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var2",
2083 assertEquals(dbSnp, sf.getFeatureGroup());
2086 assertEquals(1, sf.getBegin());
2087 assertEquals(1, sf.getEnd());
2088 assertEquals("p.Lys1Asn", sf.getDescription());
2089 assertEquals("var4", sf.getValue("ID"));
2090 assertEquals("Benign", sf.getValue("clinical_significance"));
2091 assertEquals("ID=var4;clinical_significance=Benign", sf.getAttributes());
2092 assertEquals(1, sf.links.size());
2094 "p.Lys1Asn var4|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var4",
2096 assertEquals(ensembl, sf.getFeatureGroup());
2098 // var5 generates two distinct protein variant features
2100 assertEquals(3, sf.getBegin());
2101 assertEquals(3, sf.getEnd());
2102 assertEquals("p.Pro3His", sf.getDescription());
2103 assertEquals("var6", sf.getValue("ID"));
2104 assertEquals("Good", sf.getValue("clinical_significance"));
2105 assertEquals("ID=var6;clinical_significance=Good", sf.getAttributes());
2106 assertEquals(1, sf.links.size());
2108 "p.Pro3His var6|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var6",
2110 assertEquals(cosmic, sf.getFeatureGroup());
2113 assertEquals(3, sf.getBegin());
2114 assertEquals(3, sf.getEnd());
2115 assertEquals("p.Pro3Arg", sf.getDescription());
2116 assertEquals("var6", sf.getValue("ID"));
2117 assertEquals("Good", sf.getValue("clinical_significance"));
2118 assertEquals("ID=var6;clinical_significance=Good", sf.getAttributes());
2119 assertEquals(1, sf.links.size());
2121 "p.Pro3Arg var6|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var6",
2123 assertEquals(cosmic, sf.getFeatureGroup());
2127 * Tests for the method that maps the subset of a dna sequence that has CDS
2128 * (or subtype) feature, with CDS strand = '-' (reverse)
2130 // test turned off as currently findCdsPositions is not strand-dependent
2131 // left in case it comes around again...
2132 @Test(groups = "Functional", enabled = false)
2133 public void testFindCdsPositions_reverseStrand()
2135 SequenceI dnaSeq = new Sequence("dna", "aaaGGGcccAAATTTttt");
2136 dnaSeq.createDatasetSequence();
2137 SequenceI ds = dnaSeq.getDatasetSequence();
2140 SequenceFeature sf = new SequenceFeature("CDS", "", 4, 6, 0f, null);
2142 ds.addSequenceFeature(sf);
2143 // exon feature should be ignored here
2144 sf = new SequenceFeature("exon", "", 7, 9, 0f, null);
2145 ds.addSequenceFeature(sf);
2146 // CDS for dna 10-12
2147 sf = new SequenceFeature("CDS_predicted", "", 10, 12, 0f, null);
2149 ds.addSequenceFeature(sf);
2151 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
2153 * verify ranges { [12-10], [6-4] }
2155 assertEquals(6, MappingUtils.getLength(ranges));
2156 assertEquals(2, ranges.size());
2157 assertEquals(12, ranges.get(0)[0]);
2158 assertEquals(10, ranges.get(0)[1]);
2159 assertEquals(6, ranges.get(1)[0]);
2160 assertEquals(4, ranges.get(1)[1]);
2164 * Tests for the method that maps the subset of a dna sequence that has CDS
2165 * (or subtype) feature - reverse strand case where the start codon is
2168 @Test(groups = "Functional", enabled = false)
2169 // test turned off as currently findCdsPositions is not strand-dependent
2170 // left in case it comes around again...
2171 public void testFindCdsPositions_reverseStrandThreePrimeIncomplete()
2173 SequenceI dnaSeq = new Sequence("dna", "aaagGGCCCaaaTTTttt");
2174 dnaSeq.createDatasetSequence();
2175 SequenceI ds = dnaSeq.getDatasetSequence();
2178 SequenceFeature sf = new SequenceFeature("CDS", "", 5, 9, 0f, null);
2180 ds.addSequenceFeature(sf);
2181 // CDS for dna 13-15
2182 sf = new SequenceFeature("CDS_predicted", "", 13, 15, 0f, null);
2184 sf.setPhase("2"); // skip 2 bases to start of next codon
2185 ds.addSequenceFeature(sf);
2187 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
2190 * check the mapping starts with the first complete codon
2191 * expect ranges [13, 13], [9, 5]
2193 assertEquals(6, MappingUtils.getLength(ranges));
2194 assertEquals(2, ranges.size());
2195 assertEquals(13, ranges.get(0)[0]);
2196 assertEquals(13, ranges.get(0)[1]);
2197 assertEquals(9, ranges.get(1)[0]);
2198 assertEquals(5, ranges.get(1)[1]);
2201 @Test(groups = "Functional")
2202 public void testAlignAs_alternateTranscriptsUngapped()
2204 SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
2205 SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
2206 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
2207 ((Alignment) dna).createDatasetAlignment();
2208 SequenceI cds1 = new Sequence("cds1", "GGGTTT");
2209 SequenceI cds2 = new Sequence("cds2", "CCCAAA");
2210 AlignmentI cds = new Alignment(new SequenceI[] { cds1, cds2 });
2211 ((Alignment) cds).createDatasetAlignment();
2213 AlignedCodonFrame acf = new AlignedCodonFrame();
2214 MapList map = new MapList(new int[] { 4, 9 }, new int[] { 1, 6 }, 1, 1);
2215 acf.addMap(dna1.getDatasetSequence(), cds1.getDatasetSequence(), map);
2216 map = new MapList(new int[] { 1, 3, 10, 12 }, new int[] { 1, 6 }, 1, 1);
2217 acf.addMap(dna2.getDatasetSequence(), cds2.getDatasetSequence(), map);
2220 * verify CDS alignment is as:
2221 * cccGGGTTTaaa (cdna)
2222 * CCCgggtttAAA (cdna)
2224 * ---GGGTTT--- (cds)
2225 * CCC------AAA (cds)
2227 dna.addCodonFrame(acf);
2228 AlignmentUtils.alignAs(cds, dna);
2229 assertEquals("---GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
2230 assertEquals("CCC------AAA", cds.getSequenceAt(1).getSequenceAsString());
2233 @Test(groups = { "Functional" })
2234 public void testAddMappedPositions()
2236 SequenceI from = new Sequence("dna", "ggAA-ATcc-TT-g");
2237 SequenceI seq1 = new Sequence("cds", "AAATTT");
2238 from.createDatasetSequence();
2239 seq1.createDatasetSequence();
2240 Mapping mapping = new Mapping(seq1, new MapList(
2241 new int[] { 3, 6, 9, 10 },
2242 new int[] { 1, 6 }, 1, 1));
2243 Map<Integer, Map<SequenceI, Character>> map = new TreeMap<Integer, Map<SequenceI, Character>>();
2244 AlignmentUtils.addMappedPositions(seq1, from, mapping, map);
2247 * verify map has seq1 residues in columns 3,4,6,7,11,12
2249 assertEquals(6, map.size());
2250 assertEquals('A', map.get(3).get(seq1).charValue());
2251 assertEquals('A', map.get(4).get(seq1).charValue());
2252 assertEquals('A', map.get(6).get(seq1).charValue());
2253 assertEquals('T', map.get(7).get(seq1).charValue());
2254 assertEquals('T', map.get(11).get(seq1).charValue());
2255 assertEquals('T', map.get(12).get(seq1).charValue());
2263 * Test case where the mapping 'from' range includes a stop codon which is
2264 * absent in the 'to' range
2266 @Test(groups = { "Functional" })
2267 public void testAddMappedPositions_withStopCodon()
2269 SequenceI from = new Sequence("dna", "ggAA-ATcc-TT-g");
2270 SequenceI seq1 = new Sequence("cds", "AAATTT");
2271 from.createDatasetSequence();
2272 seq1.createDatasetSequence();
2273 Mapping mapping = new Mapping(seq1, new MapList(
2274 new int[] { 3, 6, 9, 10 },
2275 new int[] { 1, 6 }, 1, 1));
2276 Map<Integer, Map<SequenceI, Character>> map = new TreeMap<Integer, Map<SequenceI, Character>>();
2277 AlignmentUtils.addMappedPositions(seq1, from, mapping, map);
2280 * verify map has seq1 residues in columns 3,4,6,7,11,12
2282 assertEquals(6, map.size());
2283 assertEquals('A', map.get(3).get(seq1).charValue());
2284 assertEquals('A', map.get(4).get(seq1).charValue());
2285 assertEquals('A', map.get(6).get(seq1).charValue());
2286 assertEquals('T', map.get(7).get(seq1).charValue());
2287 assertEquals('T', map.get(11).get(seq1).charValue());
2288 assertEquals('T', map.get(12).get(seq1).charValue());
2292 * Test for the case where the products for which we want CDS are specified.
2293 * This is to represent the case where EMBL has CDS mappings to both Uniprot
2294 * and EMBLCDSPROTEIN. makeCdsAlignment() should only return the mappings for
2295 * the protein sequences specified.
2297 @Test(groups = { "Functional" })
2298 public void testMakeCdsAlignment_filterProducts()
2300 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
2301 SequenceI dna2 = new Sequence("dna2", "GGGcccTTTaaaCCC");
2302 SequenceI pep1 = new Sequence("Uniprot|pep1", "GF");
2303 SequenceI pep2 = new Sequence("Uniprot|pep2", "GFP");
2304 SequenceI pep3 = new Sequence("EMBL|pep3", "GF");
2305 SequenceI pep4 = new Sequence("EMBL|pep4", "GFP");
2306 dna1.createDatasetSequence();
2307 dna2.createDatasetSequence();
2308 pep1.createDatasetSequence();
2309 pep2.createDatasetSequence();
2310 pep3.createDatasetSequence();
2311 pep4.createDatasetSequence();
2312 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
2313 dna.setDataset(null);
2314 AlignmentI emblPeptides = new Alignment(new SequenceI[] { pep3, pep4 });
2315 emblPeptides.setDataset(null);
2317 AlignedCodonFrame acf = new AlignedCodonFrame();
2318 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
2319 new int[] { 1, 2 }, 3, 1);
2320 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
2321 acf.addMap(dna1.getDatasetSequence(), pep3.getDatasetSequence(), map);
2322 dna.addCodonFrame(acf);
2324 acf = new AlignedCodonFrame();
2325 map = new MapList(new int[] { 1, 3, 7, 9, 13, 15 }, new int[] { 1, 3 },
2327 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map);
2328 acf.addMap(dna2.getDatasetSequence(), pep4.getDatasetSequence(), map);
2329 dna.addCodonFrame(acf);
2332 * execute method under test to find CDS for EMBL peptides only
2334 AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
2335 dna1, dna2 }, dna.getDataset(), emblPeptides.getSequencesArray());
2337 assertEquals(2, cds.getSequences().size());
2338 assertEquals("GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
2339 assertEquals("GGGTTTCCC", cds.getSequenceAt(1).getSequenceAsString());
2342 * verify shared, extended alignment dataset
2344 assertSame(dna.getDataset(), cds.getDataset());
2345 assertTrue(dna.getDataset().getSequences()
2346 .contains(cds.getSequenceAt(0).getDatasetSequence()));
2347 assertTrue(dna.getDataset().getSequences()
2348 .contains(cds.getSequenceAt(1).getDatasetSequence()));
2351 * Verify mappings from CDS to peptide, cDNA to CDS, and cDNA to peptide
2352 * the mappings are on the shared alignment dataset
2354 List<AlignedCodonFrame> cdsMappings = cds.getDataset().getCodonFrames();
2356 * 6 mappings, 2*(DNA->CDS), 2*(DNA->Pep), 2*(CDS->Pep)
2358 assertEquals(6, cdsMappings.size());
2361 * verify that mapping sets for dna and cds alignments are different
2362 * [not current behaviour - all mappings are on the alignment dataset]
2364 // select -> subselect type to test.
2365 // Assert.assertNotSame(dna.getCodonFrames(), cds.getCodonFrames());
2366 // assertEquals(4, dna.getCodonFrames().size());
2367 // assertEquals(4, cds.getCodonFrames().size());
2370 * Two mappings involve pep3 (dna to pep3, cds to pep3)
2371 * Mapping from pep3 to GGGTTT in first new exon sequence
2373 List<AlignedCodonFrame> pep3Mappings = MappingUtils
2374 .findMappingsForSequence(pep3, cdsMappings);
2375 assertEquals(2, pep3Mappings.size());
2376 List<AlignedCodonFrame> mappings = MappingUtils
2377 .findMappingsForSequence(cds.getSequenceAt(0), pep3Mappings);
2378 assertEquals(1, mappings.size());
2381 SearchResults sr = MappingUtils.buildSearchResults(pep3, 1, mappings);
2382 assertEquals(1, sr.getResults().size());
2383 Match m = sr.getResults().get(0);
2384 assertSame(cds.getSequenceAt(0).getDatasetSequence(), m.getSequence());
2385 assertEquals(1, m.getStart());
2386 assertEquals(3, m.getEnd());
2388 sr = MappingUtils.buildSearchResults(pep3, 2, mappings);
2389 m = sr.getResults().get(0);
2390 assertSame(cds.getSequenceAt(0).getDatasetSequence(), m.getSequence());
2391 assertEquals(4, m.getStart());
2392 assertEquals(6, m.getEnd());
2395 * Two mappings involve pep4 (dna to pep4, cds to pep4)
2396 * Verify mapping from pep4 to GGGTTTCCC in second new exon sequence
2398 List<AlignedCodonFrame> pep4Mappings = MappingUtils
2399 .findMappingsForSequence(pep4, cdsMappings);
2400 assertEquals(2, pep4Mappings.size());
2401 mappings = MappingUtils.findMappingsForSequence(cds.getSequenceAt(1),
2403 assertEquals(1, mappings.size());
2405 sr = MappingUtils.buildSearchResults(pep4, 1, mappings);
2406 assertEquals(1, sr.getResults().size());
2407 m = sr.getResults().get(0);
2408 assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
2409 assertEquals(1, m.getStart());
2410 assertEquals(3, m.getEnd());
2412 sr = MappingUtils.buildSearchResults(pep4, 2, mappings);
2413 m = sr.getResults().get(0);
2414 assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
2415 assertEquals(4, m.getStart());
2416 assertEquals(6, m.getEnd());
2418 sr = MappingUtils.buildSearchResults(pep4, 3, mappings);
2419 m = sr.getResults().get(0);
2420 assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
2421 assertEquals(7, m.getStart());
2422 assertEquals(9, m.getEnd());
2426 * Test the method that just copies aligned sequences, provided all sequences
2427 * to be aligned share the aligned sequence's dataset
2429 @Test(groups = "Functional")
2430 public void testAlignAsSameSequences()
2432 SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
2433 SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
2434 AlignmentI al1 = new Alignment(new SequenceI[] { dna1, dna2 });
2435 ((Alignment) al1).createDatasetAlignment();
2437 SequenceI dna3 = new Sequence(dna1);
2438 SequenceI dna4 = new Sequence(dna2);
2439 assertSame(dna3.getDatasetSequence(), dna1.getDatasetSequence());
2440 assertSame(dna4.getDatasetSequence(), dna2.getDatasetSequence());
2441 String seq1 = "-cc-GG-GT-TT--aaa";
2442 dna3.setSequence(seq1);
2443 String seq2 = "C--C-Cgg--gtt-tAA-A-";
2444 dna4.setSequence(seq2);
2445 AlignmentI al2 = new Alignment(new SequenceI[] { dna3, dna4 });
2446 ((Alignment) al2).createDatasetAlignment();
2448 assertTrue(AlignmentUtils.alignAsSameSequences(al1, al2));
2449 assertEquals(seq1, al1.getSequenceAt(0).getSequenceAsString());
2450 assertEquals(seq2, al1.getSequenceAt(1).getSequenceAsString());
2453 * add another sequence to 'aligned' - should still succeed, since
2454 * unaligned sequences still share a dataset with aligned sequences
2456 SequenceI dna5 = new Sequence("dna5", "CCCgggtttAAA");
2457 dna5.createDatasetSequence();
2458 al2.addSequence(dna5);
2459 assertTrue(AlignmentUtils.alignAsSameSequences(al1, al2));
2460 assertEquals(seq1, al1.getSequenceAt(0).getSequenceAsString());
2461 assertEquals(seq2, al1.getSequenceAt(1).getSequenceAsString());
2464 * add another sequence to 'unaligned' - should fail, since now not
2465 * all unaligned sequences share a dataset with aligned sequences
2467 SequenceI dna6 = new Sequence("dna6", "CCCgggtttAAA");
2468 dna6.createDatasetSequence();
2469 al1.addSequence(dna6);
2470 // JAL-2110 JBP Comment: what's the use case for this behaviour ?
2471 assertFalse(AlignmentUtils.alignAsSameSequences(al1, al2));
2474 @Test(groups = "Functional")
2475 public void testAlignAsSameSequencesMultipleSubSeq()
2477 SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
2478 SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
2479 SequenceI as1 = dna1.deriveSequence();
2480 SequenceI as2 = dna1.deriveSequence().getSubSequence(3, 7);
2481 SequenceI as3 = dna2.deriveSequence();
2482 as1.insertCharAt(6, 5, '-');
2483 String s_as1 = as1.getSequenceAsString();
2484 as2.insertCharAt(6, 5, '-');
2485 String s_as2 = as2.getSequenceAsString();
2486 as3.insertCharAt(6, 5, '-');
2487 String s_as3 = as3.getSequenceAsString();
2488 AlignmentI aligned = new Alignment(new SequenceI[] { as1, as2, as3 });
2490 // why do we need to cast this still ?
2491 ((Alignment) aligned).createDatasetAlignment();
2492 SequenceI uas1 = dna1.deriveSequence();
2493 SequenceI uas2 = dna1.deriveSequence().getSubSequence(3, 7);
2494 SequenceI uas3 = dna2.deriveSequence();
2495 AlignmentI tobealigned = new Alignment(new SequenceI[] { uas1, uas2,
2497 ((Alignment) tobealigned).createDatasetAlignment();
2499 assertTrue(AlignmentUtils.alignAsSameSequences(tobealigned, aligned));
2500 assertEquals(s_as1, uas1.getSequenceAsString());
2501 assertEquals(s_as2, uas2.getSequenceAsString());
2502 assertEquals(s_as3, uas3.getSequenceAsString());