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.DataSourceType;
45 import jalview.io.FileFormat;
46 import jalview.io.FileFormatI;
47 import jalview.io.FormatAdapter;
48 import jalview.util.MapList;
49 import jalview.util.MappingUtils;
51 import java.io.IOException;
52 import java.util.ArrayList;
53 import java.util.Arrays;
54 import java.util.LinkedHashMap;
55 import java.util.List;
57 import java.util.TreeMap;
59 import org.testng.annotations.Test;
61 public class AlignmentUtilsTests
63 public static Sequence ts = new Sequence("short",
64 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklm");
66 @Test(groups = { "Functional" })
67 public void testExpandContext()
69 AlignmentI al = new Alignment(new Sequence[] {});
70 for (int i = 4; i < 14; i += 2)
72 SequenceI s1 = ts.deriveSequence().getSubSequence(i, i + 7);
75 System.out.println(new AppletFormatAdapter().formatSequences(
78 for (int flnk = -1; flnk < 25; flnk++)
80 AlignmentI exp = AlignmentUtils.expandContext(al, flnk);
81 System.out.println("\nFlank size: " + flnk);
82 System.out.println(new AppletFormatAdapter().formatSequences(
83 FileFormat.Clustal, exp, true));
87 * Full expansion to complete sequences
89 for (SequenceI sq : exp.getSequences())
91 String ung = sq.getSequenceAsString().replaceAll("-+", "");
92 final String errorMsg = "Flanking sequence not the same as original dataset sequence.\n"
95 + sq.getDatasetSequence().getSequenceAsString();
96 assertTrue(errorMsg, ung.equalsIgnoreCase(sq.getDatasetSequence()
97 .getSequenceAsString()));
103 * Last sequence is fully expanded, others have leading gaps to match
105 assertTrue(exp.getSequenceAt(4).getSequenceAsString()
107 assertTrue(exp.getSequenceAt(3).getSequenceAsString()
108 .startsWith("--abc"));
109 assertTrue(exp.getSequenceAt(2).getSequenceAsString()
110 .startsWith("----abc"));
111 assertTrue(exp.getSequenceAt(1).getSequenceAsString()
112 .startsWith("------abc"));
113 assertTrue(exp.getSequenceAt(0).getSequenceAsString()
114 .startsWith("--------abc"));
120 * Test that annotations are correctly adjusted by expandContext
122 @Test(groups = { "Functional" })
123 public void testExpandContext_annotation()
125 AlignmentI al = new Alignment(new Sequence[] {});
126 SequenceI ds = new Sequence("Seq1", "ABCDEFGHI");
128 SequenceI seq1 = ds.deriveSequence().getSubSequence(3, 6);
129 al.addSequence(seq1);
132 * Annotate DEF with 4/5/6 respectively
134 Annotation[] anns = new Annotation[] { new Annotation(4),
135 new Annotation(5), new Annotation(6) };
136 AlignmentAnnotation ann = new AlignmentAnnotation("SS",
137 "secondary structure", anns);
138 seq1.addAlignmentAnnotation(ann);
141 * The annotations array should match aligned positions
143 assertEquals(3, ann.annotations.length);
144 assertEquals(4, ann.annotations[0].value, 0.001);
145 assertEquals(5, ann.annotations[1].value, 0.001);
146 assertEquals(6, ann.annotations[2].value, 0.001);
149 * Check annotation to sequence position mappings before expanding the
150 * sequence; these are set up in Sequence.addAlignmentAnnotation ->
151 * Annotation.setSequenceRef -> createSequenceMappings
153 assertNull(ann.getAnnotationForPosition(1));
154 assertNull(ann.getAnnotationForPosition(2));
155 assertNull(ann.getAnnotationForPosition(3));
156 assertEquals(4, ann.getAnnotationForPosition(4).value, 0.001);
157 assertEquals(5, ann.getAnnotationForPosition(5).value, 0.001);
158 assertEquals(6, ann.getAnnotationForPosition(6).value, 0.001);
159 assertNull(ann.getAnnotationForPosition(7));
160 assertNull(ann.getAnnotationForPosition(8));
161 assertNull(ann.getAnnotationForPosition(9));
164 * Expand the subsequence to the full sequence abcDEFghi
166 AlignmentI expanded = AlignmentUtils.expandContext(al, -1);
167 assertEquals("abcDEFghi", expanded.getSequenceAt(0)
168 .getSequenceAsString());
171 * Confirm the alignment and sequence have the same SS annotation,
172 * referencing the expanded sequence
174 ann = expanded.getSequenceAt(0).getAnnotation()[0];
175 assertSame(ann, expanded.getAlignmentAnnotation()[0]);
176 assertSame(expanded.getSequenceAt(0), ann.sequenceRef);
179 * The annotations array should have null values except for annotated
182 assertNull(ann.annotations[0]);
183 assertNull(ann.annotations[1]);
184 assertNull(ann.annotations[2]);
185 assertEquals(4, ann.annotations[3].value, 0.001);
186 assertEquals(5, ann.annotations[4].value, 0.001);
187 assertEquals(6, ann.annotations[5].value, 0.001);
188 assertNull(ann.annotations[6]);
189 assertNull(ann.annotations[7]);
190 assertNull(ann.annotations[8]);
193 * sequence position mappings should be unchanged
195 assertNull(ann.getAnnotationForPosition(1));
196 assertNull(ann.getAnnotationForPosition(2));
197 assertNull(ann.getAnnotationForPosition(3));
198 assertEquals(4, ann.getAnnotationForPosition(4).value, 0.001);
199 assertEquals(5, ann.getAnnotationForPosition(5).value, 0.001);
200 assertEquals(6, ann.getAnnotationForPosition(6).value, 0.001);
201 assertNull(ann.getAnnotationForPosition(7));
202 assertNull(ann.getAnnotationForPosition(8));
203 assertNull(ann.getAnnotationForPosition(9));
207 * Test method that returns a map of lists of sequences by sequence name.
209 * @throws IOException
211 @Test(groups = { "Functional" })
212 public void testGetSequencesByName() throws IOException
214 final String data = ">Seq1Name\nKQYL\n" + ">Seq2Name\nRFPW\n"
215 + ">Seq1Name\nABCD\n";
216 AlignmentI al = loadAlignment(data, FileFormat.Fasta);
217 Map<String, List<SequenceI>> map = AlignmentUtils
218 .getSequencesByName(al);
219 assertEquals(2, map.keySet().size());
220 assertEquals(2, map.get("Seq1Name").size());
221 assertEquals("KQYL", map.get("Seq1Name").get(0).getSequenceAsString());
222 assertEquals("ABCD", map.get("Seq1Name").get(1).getSequenceAsString());
223 assertEquals(1, map.get("Seq2Name").size());
224 assertEquals("RFPW", map.get("Seq2Name").get(0).getSequenceAsString());
228 * Helper method to load an alignment and ensure dataset sequences are set up.
234 * @throws IOException
236 protected AlignmentI loadAlignment(final String data, FileFormatI format)
239 AlignmentI a = new FormatAdapter().readFile(data,
240 DataSourceType.PASTE, format);
246 * Test mapping of protein to cDNA, for the case where we have no sequence
247 * cross-references, so mappings are made first-served 1-1 where sequences
250 * @throws IOException
252 @Test(groups = { "Functional" })
253 public void testMapProteinAlignmentToCdna_noXrefs() throws IOException
255 List<SequenceI> protseqs = new ArrayList<SequenceI>();
256 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
257 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
258 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
259 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
260 protein.setDataset(null);
262 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
263 dnaseqs.add(new Sequence("EMBL|A11111", "TCAGCACGC")); // = SAR
264 dnaseqs.add(new Sequence("EMBL|A22222", "GAGATACAA")); // = EIQ
265 dnaseqs.add(new Sequence("EMBL|A33333", "GAAATCCAG")); // = EIQ
266 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG")); // = EIQ
267 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[4]));
268 cdna.setDataset(null);
270 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
272 // 3 mappings made, each from 1 to 1 sequence
273 assertEquals(3, protein.getCodonFrames().size());
274 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
275 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
276 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
278 // V12345 mapped to A22222
279 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
281 assertEquals(1, acf.getdnaSeqs().length);
282 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
283 acf.getdnaSeqs()[0]);
284 Mapping[] protMappings = acf.getProtMappings();
285 assertEquals(1, protMappings.length);
286 MapList mapList = protMappings[0].getMap();
287 assertEquals(3, mapList.getFromRatio());
288 assertEquals(1, mapList.getToRatio());
289 assertTrue(Arrays.equals(new int[] { 1, 9 }, mapList.getFromRanges()
291 assertEquals(1, mapList.getFromRanges().size());
292 assertTrue(Arrays.equals(new int[] { 1, 3 },
293 mapList.getToRanges().get(0)));
294 assertEquals(1, mapList.getToRanges().size());
296 // V12346 mapped to A33333
297 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
298 assertEquals(1, acf.getdnaSeqs().length);
299 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
300 acf.getdnaSeqs()[0]);
302 // V12347 mapped to A11111
303 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
304 assertEquals(1, acf.getdnaSeqs().length);
305 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
306 acf.getdnaSeqs()[0]);
308 // no mapping involving the 'extra' A44444
309 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(3)).isEmpty());
313 * Test for the alignSequenceAs method that takes two sequences and a mapping.
315 @Test(groups = { "Functional" })
316 public void testAlignSequenceAs_withMapping_noIntrons()
318 MapList map = new MapList(new int[] { 1, 6 }, new int[] { 1, 2 }, 3, 1);
321 * No existing gaps in dna:
323 checkAlignSequenceAs("GGGAAA", "-A-L-", false, false, map,
327 * Now introduce gaps in dna but ignore them when realigning.
329 checkAlignSequenceAs("-G-G-G-A-A-A-", "-A-L-", false, false, map,
333 * Now include gaps in dna when realigning. First retaining 'mapped' gaps
334 * only, i.e. those within the exon region.
336 checkAlignSequenceAs("-G-G--G-A--A-A-", "-A-L-", true, false, map,
337 "---G-G--G---A--A-A");
340 * Include all gaps in dna when realigning (within and without the exon
341 * region). The leading gap, and the gaps between codons, are subsumed by
342 * the protein alignment gap.
344 checkAlignSequenceAs("-G-GG--AA-A---", "-A-L-", true, true, map,
345 "---G-GG---AA-A---");
348 * Include only unmapped gaps in dna when realigning (outside the exon
349 * region). The leading gap, and the gaps between codons, are subsumed by
350 * the protein alignment gap.
352 checkAlignSequenceAs("-G-GG--AA-A-", "-A-L-", false, true, map,
357 * Test for the alignSequenceAs method that takes two sequences and a mapping.
359 @Test(groups = { "Functional" })
360 public void testAlignSequenceAs_withMapping_withIntrons()
363 * Exons at codon 2 (AAA) and 4 (TTT)
365 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
366 new int[] { 1, 2 }, 3, 1);
369 * Simple case: no gaps in dna
371 checkAlignSequenceAs("GGGAAACCCTTTGGG", "--A-L-", false, false, map,
372 "GGG---AAACCCTTTGGG");
375 * Add gaps to dna - but ignore when realigning.
377 checkAlignSequenceAs("-G-G-G--A--A---AC-CC-T-TT-GG-G-", "--A-L-",
378 false, false, map, "GGG---AAACCCTTTGGG");
381 * Add gaps to dna - include within exons only when realigning.
383 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
384 true, false, map, "GGG---A--A---ACCCT-TTGGG");
387 * Include gaps outside exons only when realigning.
389 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
390 false, true, map, "-G-G-GAAAC-CCTTT-GG-G-");
393 * Include gaps following first intron if we are 'preserving mapped gaps'
395 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
396 true, true, map, "-G-G-G--A--A---A-C-CC-T-TT-GG-G-");
399 * Include all gaps in dna when realigning.
401 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
402 true, true, map, "-G-G-G--A--A---A-C-CC-T-TT-GG-G-");
406 * Test for the case where not all of the protein sequence is mapped to cDNA.
408 @Test(groups = { "Functional" })
409 public void testAlignSequenceAs_withMapping_withUnmappedProtein()
412 * Exons at codon 2 (AAA) and 4 (TTT) mapped to A and P
414 final MapList map = new MapList(new int[] { 4, 6, 10, 12 }, new int[] {
418 * -L- 'aligns' ccc------
420 checkAlignSequenceAs("gggAAAcccTTTggg", "-A-L-P-", false, false, map,
421 "gggAAAccc------TTTggg");
425 * Helper method that performs and verifies the method under test.
428 * the sequence to be realigned
430 * the sequence whose alignment is to be copied
431 * @param preserveMappedGaps
432 * @param preserveUnmappedGaps
436 protected void checkAlignSequenceAs(final String alignee,
437 final String alignModel, final boolean preserveMappedGaps,
438 final boolean preserveUnmappedGaps, MapList map,
439 final String expected)
441 SequenceI alignMe = new Sequence("Seq1", alignee);
442 alignMe.createDatasetSequence();
443 SequenceI alignFrom = new Sequence("Seq2", alignModel);
444 alignFrom.createDatasetSequence();
445 AlignedCodonFrame acf = new AlignedCodonFrame();
446 acf.addMap(alignMe.getDatasetSequence(), alignFrom.getDatasetSequence(), map);
448 AlignmentUtils.alignSequenceAs(alignMe, alignFrom, acf, "---", '-',
449 preserveMappedGaps, preserveUnmappedGaps);
450 assertEquals(expected, alignMe.getSequenceAsString());
454 * Test for the alignSequenceAs method where we preserve gaps in introns only.
456 @Test(groups = { "Functional" })
457 public void testAlignSequenceAs_keepIntronGapsOnly()
461 * Intron GGGAAA followed by exon CCCTTT
463 MapList map = new MapList(new int[] { 7, 12 }, new int[] { 1, 2 }, 3, 1);
465 checkAlignSequenceAs("GG-G-AA-A-C-CC-T-TT", "AL", false, true, map,
470 * Test the method that realigns protein to match mapped codon alignment.
472 @Test(groups = { "Functional" })
473 public void testAlignProteinAsDna()
475 // seq1 codons are [1,2,3] [4,5,6] [7,8,9] [10,11,12]
476 SequenceI dna1 = new Sequence("Seq1", "TGCCATTACCAG-");
477 // seq2 codons are [1,3,4] [5,6,7] [8,9,10] [11,12,13]
478 SequenceI dna2 = new Sequence("Seq2", "T-GCCATTACCAG");
479 // seq3 codons are [1,2,3] [4,5,7] [8,9,10] [11,12,13]
480 SequenceI dna3 = new Sequence("Seq3", "TGCCA-TTACCAG");
481 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
482 dna.setDataset(null);
484 // protein alignment will be realigned like dna
485 SequenceI prot1 = new Sequence("Seq1", "CHYQ");
486 SequenceI prot2 = new Sequence("Seq2", "CHYQ");
487 SequenceI prot3 = new Sequence("Seq3", "CHYQ");
488 SequenceI prot4 = new Sequence("Seq4", "R-QSV"); // unmapped, unchanged
489 AlignmentI protein = new Alignment(new SequenceI[] { prot1, prot2,
491 protein.setDataset(null);
493 MapList map = new MapList(new int[] { 1, 12 }, new int[] { 1, 4 }, 3, 1);
494 AlignedCodonFrame acf = new AlignedCodonFrame();
495 acf.addMap(dna1.getDatasetSequence(), prot1.getDatasetSequence(), map);
496 acf.addMap(dna2.getDatasetSequence(), prot2.getDatasetSequence(), map);
497 acf.addMap(dna3.getDatasetSequence(), prot3.getDatasetSequence(), map);
498 ArrayList<AlignedCodonFrame> acfs = new ArrayList<AlignedCodonFrame>();
500 protein.setCodonFrames(acfs);
503 * Translated codon order is [1,2,3] [1,3,4] [4,5,6] [4,5,7] [5,6,7] [7,8,9]
504 * [8,9,10] [10,11,12] [11,12,13]
506 AlignmentUtils.alignProteinAsDna(protein, dna);
507 assertEquals("C-H--Y-Q-", prot1.getSequenceAsString());
508 assertEquals("-C--H-Y-Q", prot2.getSequenceAsString());
509 assertEquals("C--H--Y-Q", prot3.getSequenceAsString());
510 assertEquals("R-QSV", prot4.getSequenceAsString());
514 * Test the method that tests whether a CDNA sequence translates to a protein
517 @Test(groups = { "Functional" })
518 public void testTranslatesAs()
520 // null arguments check
521 assertFalse(AlignmentUtils.translatesAs(null, 0, null));
522 assertFalse(AlignmentUtils.translatesAs(new char[] { 't' }, 0, null));
523 assertFalse(AlignmentUtils.translatesAs(null, 0, new char[] { 'a' }));
525 // straight translation
526 assertTrue(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(), 0,
527 "FPKG".toCharArray()));
528 // with extra start codon (not in protein)
529 assertTrue(AlignmentUtils.translatesAs("atgtttcccaaaggg".toCharArray(),
530 3, "FPKG".toCharArray()));
531 // with stop codon1 (not in protein)
532 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtaa".toCharArray(),
533 0, "FPKG".toCharArray()));
534 // with stop codon1 (in protein as *)
535 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtaa".toCharArray(),
536 0, "FPKG*".toCharArray()));
537 // with stop codon2 (not in protein)
538 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtag".toCharArray(),
539 0, "FPKG".toCharArray()));
540 // with stop codon3 (not in protein)
541 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtga".toCharArray(),
542 0, "FPKG".toCharArray()));
543 // with start and stop codon1
544 assertTrue(AlignmentUtils.translatesAs(
545 "atgtttcccaaagggtaa".toCharArray(), 3, "FPKG".toCharArray()));
546 // with start and stop codon1 (in protein as *)
547 assertTrue(AlignmentUtils.translatesAs(
548 "atgtttcccaaagggtaa".toCharArray(), 3, "FPKG*".toCharArray()));
549 // with start and stop codon2
550 assertTrue(AlignmentUtils.translatesAs(
551 "atgtttcccaaagggtag".toCharArray(), 3, "FPKG".toCharArray()));
552 // with start and stop codon3
553 assertTrue(AlignmentUtils.translatesAs(
554 "atgtttcccaaagggtga".toCharArray(), 3, "FPKG".toCharArray()));
556 // with embedded stop codons
557 assertTrue(AlignmentUtils.translatesAs(
558 "atgtttTAGcccaaaTAAgggtga".toCharArray(), 3,
559 "F*PK*G".toCharArray()));
562 assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(),
563 0, "FPMG".toCharArray()));
566 assertFalse(AlignmentUtils.translatesAs("tttcccaaagg".toCharArray(), 0,
567 "FPKG".toCharArray()));
570 assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(),
571 0, "FPK".toCharArray()));
573 // overlong dna (doesn't end in stop codon)
574 assertFalse(AlignmentUtils.translatesAs(
575 "tttcccaaagggttt".toCharArray(), 0, "FPKG".toCharArray()));
577 // dna + stop codon + more
578 assertFalse(AlignmentUtils.translatesAs(
579 "tttcccaaagggttaga".toCharArray(), 0, "FPKG".toCharArray()));
582 assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(),
583 0, "FPKGQ".toCharArray()));
587 * Test mapping of protein to cDNA, for cases where the cDNA has start and/or
588 * stop codons in addition to the protein coding sequence.
590 * @throws IOException
592 @Test(groups = { "Functional" })
593 public void testMapProteinAlignmentToCdna_withStartAndStopCodons()
596 List<SequenceI> protseqs = new ArrayList<SequenceI>();
597 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
598 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
599 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
600 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
601 protein.setDataset(null);
603 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
605 dnaseqs.add(new Sequence("EMBL|A11111", "ATGTCAGCACGC"));
607 dnaseqs.add(new Sequence("EMBL|A22222", "GAGATACAATAA"));
608 // = start +EIQ + stop
609 dnaseqs.add(new Sequence("EMBL|A33333", "ATGGAAATCCAGTAG"));
610 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG"));
611 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[4]));
612 cdna.setDataset(null);
614 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
616 // 3 mappings made, each from 1 to 1 sequence
617 assertEquals(3, protein.getCodonFrames().size());
618 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
619 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
620 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
622 // V12345 mapped from A22222
623 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
625 assertEquals(1, acf.getdnaSeqs().length);
626 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
627 acf.getdnaSeqs()[0]);
628 Mapping[] protMappings = acf.getProtMappings();
629 assertEquals(1, protMappings.length);
630 MapList mapList = protMappings[0].getMap();
631 assertEquals(3, mapList.getFromRatio());
632 assertEquals(1, mapList.getToRatio());
633 assertTrue(Arrays.equals(new int[] { 1, 9 }, mapList.getFromRanges()
635 assertEquals(1, mapList.getFromRanges().size());
636 assertTrue(Arrays.equals(new int[] { 1, 3 },
637 mapList.getToRanges().get(0)));
638 assertEquals(1, mapList.getToRanges().size());
640 // V12346 mapped from A33333 starting position 4
641 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
642 assertEquals(1, acf.getdnaSeqs().length);
643 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
644 acf.getdnaSeqs()[0]);
645 protMappings = acf.getProtMappings();
646 assertEquals(1, protMappings.length);
647 mapList = protMappings[0].getMap();
648 assertEquals(3, mapList.getFromRatio());
649 assertEquals(1, mapList.getToRatio());
650 assertTrue(Arrays.equals(new int[] { 4, 12 }, mapList.getFromRanges()
652 assertEquals(1, mapList.getFromRanges().size());
653 assertTrue(Arrays.equals(new int[] { 1, 3 },
654 mapList.getToRanges().get(0)));
655 assertEquals(1, mapList.getToRanges().size());
657 // V12347 mapped to A11111 starting position 4
658 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
659 assertEquals(1, acf.getdnaSeqs().length);
660 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
661 acf.getdnaSeqs()[0]);
662 protMappings = acf.getProtMappings();
663 assertEquals(1, protMappings.length);
664 mapList = protMappings[0].getMap();
665 assertEquals(3, mapList.getFromRatio());
666 assertEquals(1, mapList.getToRatio());
667 assertTrue(Arrays.equals(new int[] { 4, 12 }, mapList.getFromRanges()
669 assertEquals(1, mapList.getFromRanges().size());
670 assertTrue(Arrays.equals(new int[] { 1, 3 },
671 mapList.getToRanges().get(0)));
672 assertEquals(1, mapList.getToRanges().size());
674 // no mapping involving the 'extra' A44444
675 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(3)).isEmpty());
679 * Test mapping of protein to cDNA, for the case where we have some sequence
680 * cross-references. Verify that 1-to-many mappings are made where
681 * cross-references exist and sequences are mappable.
683 * @throws IOException
685 @Test(groups = { "Functional" })
686 public void testMapProteinAlignmentToCdna_withXrefs() throws IOException
688 List<SequenceI> protseqs = new ArrayList<SequenceI>();
689 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
690 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
691 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
692 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
693 protein.setDataset(null);
695 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
696 dnaseqs.add(new Sequence("EMBL|A11111", "TCAGCACGC")); // = SAR
697 dnaseqs.add(new Sequence("EMBL|A22222", "ATGGAGATACAA")); // = start + EIQ
698 dnaseqs.add(new Sequence("EMBL|A33333", "GAAATCCAG")); // = EIQ
699 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG")); // = EIQ
700 dnaseqs.add(new Sequence("EMBL|A55555", "GAGATTCAG")); // = EIQ
701 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[5]));
702 cdna.setDataset(null);
704 // Xref A22222 to V12345 (should get mapped)
705 dnaseqs.get(1).addDBRef(new DBRefEntry("UNIPROT", "1", "V12345"));
706 // Xref V12345 to A44444 (should get mapped)
707 protseqs.get(0).addDBRef(new DBRefEntry("EMBL", "1", "A44444"));
708 // Xref A33333 to V12347 (sequence mismatch - should not get mapped)
709 dnaseqs.get(2).addDBRef(new DBRefEntry("UNIPROT", "1", "V12347"));
710 // as V12345 is mapped to A22222 and A44444, this leaves V12346 unmapped.
711 // it should get paired up with the unmapped A33333
712 // A11111 should be mapped to V12347
713 // A55555 is spare and has no xref so is not mapped
715 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
717 // 4 protein mappings made for 3 proteins, 2 to V12345, 1 each to V12346/7
718 assertEquals(3, protein.getCodonFrames().size());
719 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
720 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
721 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
723 // one mapping for each of the first 4 cDNA sequences
724 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(0)).size());
725 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(1)).size());
726 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(2)).size());
727 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(3)).size());
729 // V12345 mapped to A22222 and A44444
730 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
732 assertEquals(2, acf.getdnaSeqs().length);
733 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
734 acf.getdnaSeqs()[0]);
735 assertEquals(cdna.getSequenceAt(3).getDatasetSequence(),
736 acf.getdnaSeqs()[1]);
738 // V12346 mapped to A33333
739 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
740 assertEquals(1, acf.getdnaSeqs().length);
741 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
742 acf.getdnaSeqs()[0]);
744 // V12347 mapped to A11111
745 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
746 assertEquals(1, acf.getdnaSeqs().length);
747 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
748 acf.getdnaSeqs()[0]);
750 // no mapping involving the 'extra' A55555
751 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(4)).isEmpty());
755 * Test mapping of protein to cDNA, for the case where we have some sequence
756 * cross-references. Verify that once we have made an xref mapping we don't
757 * also map un-xrefd sequeces.
759 * @throws IOException
761 @Test(groups = { "Functional" })
762 public void testMapProteinAlignmentToCdna_prioritiseXrefs()
765 List<SequenceI> protseqs = new ArrayList<SequenceI>();
766 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
767 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
768 AlignmentI protein = new Alignment(
769 protseqs.toArray(new SequenceI[protseqs.size()]));
770 protein.setDataset(null);
772 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
773 dnaseqs.add(new Sequence("EMBL|A11111", "GAAATCCAG")); // = EIQ
774 dnaseqs.add(new Sequence("EMBL|A22222", "GAAATTCAG")); // = EIQ
775 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[dnaseqs
777 cdna.setDataset(null);
779 // Xref A22222 to V12345 (should get mapped)
780 // A11111 should then be mapped to the unmapped V12346
781 dnaseqs.get(1).addDBRef(new DBRefEntry("UNIPROT", "1", "V12345"));
783 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
785 // 2 protein mappings made
786 assertEquals(2, protein.getCodonFrames().size());
787 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
788 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
790 // one mapping for each of the cDNA sequences
791 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(0)).size());
792 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(1)).size());
794 // V12345 mapped to A22222
795 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
797 assertEquals(1, acf.getdnaSeqs().length);
798 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
799 acf.getdnaSeqs()[0]);
801 // V12346 mapped to A11111
802 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
803 assertEquals(1, acf.getdnaSeqs().length);
804 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
805 acf.getdnaSeqs()[0]);
809 * Test the method that shows or hides sequence annotations by type(s) and
812 @Test(groups = { "Functional" })
813 public void testShowOrHideSequenceAnnotations()
815 SequenceI seq1 = new Sequence("Seq1", "AAA");
816 SequenceI seq2 = new Sequence("Seq2", "BBB");
817 SequenceI seq3 = new Sequence("Seq3", "CCC");
818 Annotation[] anns = new Annotation[] { new Annotation(2f) };
819 AlignmentAnnotation ann1 = new AlignmentAnnotation("Structure", "ann1",
821 ann1.setSequenceRef(seq1);
822 AlignmentAnnotation ann2 = new AlignmentAnnotation("Structure", "ann2",
824 ann2.setSequenceRef(seq2);
825 AlignmentAnnotation ann3 = new AlignmentAnnotation("Structure", "ann3",
827 AlignmentAnnotation ann4 = new AlignmentAnnotation("Temp", "ann4", anns);
828 ann4.setSequenceRef(seq1);
829 AlignmentAnnotation ann5 = new AlignmentAnnotation("Temp", "ann5", anns);
830 ann5.setSequenceRef(seq2);
831 AlignmentAnnotation ann6 = new AlignmentAnnotation("Temp", "ann6", anns);
832 AlignmentI al = new Alignment(new SequenceI[] { seq1, seq2, seq3 });
833 al.addAnnotation(ann1); // Structure for Seq1
834 al.addAnnotation(ann2); // Structure for Seq2
835 al.addAnnotation(ann3); // Structure for no sequence
836 al.addAnnotation(ann4); // Temp for seq1
837 al.addAnnotation(ann5); // Temp for seq2
838 al.addAnnotation(ann6); // Temp for no sequence
839 List<String> types = new ArrayList<String>();
840 List<SequenceI> scope = new ArrayList<SequenceI>();
843 * Set all sequence related Structure to hidden (ann1, ann2)
845 types.add("Structure");
846 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, false,
848 assertFalse(ann1.visible);
849 assertFalse(ann2.visible);
850 assertTrue(ann3.visible); // not sequence-related, not affected
851 assertTrue(ann4.visible); // not Structure, not affected
852 assertTrue(ann5.visible); // "
853 assertTrue(ann6.visible); // not sequence-related, not affected
856 * Set Temp in {seq1, seq3} to hidden
862 AlignmentUtils.showOrHideSequenceAnnotations(al, types, scope, false,
864 assertFalse(ann1.visible); // unchanged
865 assertFalse(ann2.visible); // unchanged
866 assertTrue(ann3.visible); // not sequence-related, not affected
867 assertFalse(ann4.visible); // Temp for seq1 hidden
868 assertTrue(ann5.visible); // not in scope, not affected
869 assertTrue(ann6.visible); // not sequence-related, not affected
872 * Set Temp in all sequences to hidden
878 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, false,
880 assertFalse(ann1.visible); // unchanged
881 assertFalse(ann2.visible); // unchanged
882 assertTrue(ann3.visible); // not sequence-related, not affected
883 assertFalse(ann4.visible); // Temp for seq1 hidden
884 assertFalse(ann5.visible); // Temp for seq2 hidden
885 assertTrue(ann6.visible); // not sequence-related, not affected
888 * Set all types in {seq1, seq3} to visible
894 AlignmentUtils.showOrHideSequenceAnnotations(al, types, scope, true,
896 assertTrue(ann1.visible); // Structure for seq1 set visible
897 assertFalse(ann2.visible); // not in scope, unchanged
898 assertTrue(ann3.visible); // not sequence-related, not affected
899 assertTrue(ann4.visible); // Temp for seq1 set visible
900 assertFalse(ann5.visible); // not in scope, unchanged
901 assertTrue(ann6.visible); // not sequence-related, not affected
904 * Set all types in all scope to hidden
906 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, true,
908 assertFalse(ann1.visible);
909 assertFalse(ann2.visible);
910 assertTrue(ann3.visible); // not sequence-related, not affected
911 assertFalse(ann4.visible);
912 assertFalse(ann5.visible);
913 assertTrue(ann6.visible); // not sequence-related, not affected
917 * Tests for the method that checks if one sequence cross-references another
919 @Test(groups = { "Functional" })
920 public void testHasCrossRef()
922 assertFalse(AlignmentUtils.hasCrossRef(null, null));
923 SequenceI seq1 = new Sequence("EMBL|A12345", "ABCDEF");
924 assertFalse(AlignmentUtils.hasCrossRef(seq1, null));
925 assertFalse(AlignmentUtils.hasCrossRef(null, seq1));
926 SequenceI seq2 = new Sequence("UNIPROT|V20192", "ABCDEF");
927 assertFalse(AlignmentUtils.hasCrossRef(seq1, seq2));
930 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "v20193"));
931 assertFalse(AlignmentUtils.hasCrossRef(seq1, seq2));
933 // case-insensitive; version number is ignored
934 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "v20192"));
935 assertTrue(AlignmentUtils.hasCrossRef(seq1, seq2));
938 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
939 assertTrue(AlignmentUtils.hasCrossRef(seq1, seq2));
940 // test is one-way only
941 assertFalse(AlignmentUtils.hasCrossRef(seq2, seq1));
945 * Tests for the method that checks if either sequence cross-references the
948 @Test(groups = { "Functional" })
949 public void testHaveCrossRef()
951 assertFalse(AlignmentUtils.hasCrossRef(null, null));
952 SequenceI seq1 = new Sequence("EMBL|A12345", "ABCDEF");
953 assertFalse(AlignmentUtils.haveCrossRef(seq1, null));
954 assertFalse(AlignmentUtils.haveCrossRef(null, seq1));
955 SequenceI seq2 = new Sequence("UNIPROT|V20192", "ABCDEF");
956 assertFalse(AlignmentUtils.haveCrossRef(seq1, seq2));
958 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
959 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
960 // next is true for haveCrossRef, false for hasCrossRef
961 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
963 // now the other way round
964 seq1.setDBRefs(null);
965 seq2.addDBRef(new DBRefEntry("EMBL", "1", "A12345"));
966 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
967 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
970 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
971 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
972 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
976 * Test the method that extracts the cds-only part of a dna alignment.
978 @Test(groups = { "Functional" })
979 public void testMakeCdsAlignment()
983 * dna1 --> [4, 6] [10,12] --> pep1
984 * dna2 --> [1, 3] [7, 9] [13,15] --> pep1
986 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
987 SequenceI dna2 = new Sequence("dna2", "GGGcccTTTaaaCCC");
988 SequenceI pep1 = new Sequence("pep1", "GF");
989 SequenceI pep2 = new Sequence("pep2", "GFP");
990 pep1.addDBRef(new DBRefEntry("UNIPROT", "0", "pep1"));
991 pep2.addDBRef(new DBRefEntry("UNIPROT", "0", "pep2"));
992 dna1.createDatasetSequence();
993 dna2.createDatasetSequence();
994 pep1.createDatasetSequence();
995 pep2.createDatasetSequence();
996 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
997 dna.setDataset(null);
1000 * need a sourceDbRef if we are to construct dbrefs to the CDS
1003 DBRefEntry dbref = new DBRefEntry("ENSEMBL", "0", "dna1");
1004 dna1.getDatasetSequence().setSourceDBRef(dbref);
1005 dbref = new DBRefEntry("ENSEMBL", "0", "dna2");
1006 dna2.getDatasetSequence().setSourceDBRef(dbref);
1009 * CDS sequences are 'discovered' from dna-to-protein mappings on the alignment
1010 * dataset (e.g. added from dbrefs by CrossRef.findXrefSequences)
1012 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1013 new int[] { 1, 2 }, 3, 1);
1014 AlignedCodonFrame acf = new AlignedCodonFrame();
1015 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
1016 dna.addCodonFrame(acf);
1017 map = new MapList(new int[] { 1, 3, 7, 9, 13, 15 }, new int[] { 1, 3 },
1019 acf = new AlignedCodonFrame();
1020 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map);
1021 dna.addCodonFrame(acf);
1024 * execute method under test:
1026 AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
1027 dna1, dna2 }, dna.getDataset(), null);
1030 * verify cds sequences
1032 assertEquals(2, cds.getSequences().size());
1033 assertEquals("GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
1034 assertEquals("GGGTTTCCC", cds.getSequenceAt(1).getSequenceAsString());
1037 * verify shared, extended alignment dataset
1039 assertSame(dna.getDataset(), cds.getDataset());
1040 SequenceI cds1Dss = cds.getSequenceAt(0).getDatasetSequence();
1041 SequenceI cds2Dss = cds.getSequenceAt(1).getDatasetSequence();
1042 assertTrue(dna.getDataset().getSequences().contains(cds1Dss));
1043 assertTrue(dna.getDataset().getSequences().contains(cds2Dss));
1046 * verify CDS has a dbref with mapping to peptide
1048 assertNotNull(cds1Dss.getDBRefs());
1049 assertEquals(1, cds1Dss.getDBRefs().length);
1050 dbref = cds1Dss.getDBRefs()[0];
1051 assertEquals("UNIPROT", dbref.getSource());
1052 assertEquals("0", dbref.getVersion());
1053 assertEquals("pep1", dbref.getAccessionId());
1054 assertNotNull(dbref.getMap());
1055 assertSame(pep1.getDatasetSequence(), dbref.getMap().getTo());
1056 MapList cdsMapping = new MapList(new int[] { 1, 6 },
1057 new int[] { 1, 2 }, 3, 1);
1058 assertEquals(cdsMapping, dbref.getMap().getMap());
1061 * verify peptide has added a dbref with reverse mapping to CDS
1063 assertNotNull(pep1.getDBRefs());
1064 assertEquals(2, pep1.getDBRefs().length);
1065 dbref = pep1.getDBRefs()[1];
1066 assertEquals("ENSEMBL", dbref.getSource());
1067 assertEquals("0", dbref.getVersion());
1068 assertEquals("CDS|dna1", dbref.getAccessionId());
1069 assertNotNull(dbref.getMap());
1070 assertSame(cds1Dss, dbref.getMap().getTo());
1071 assertEquals(cdsMapping.getInverse(), dbref.getMap().getMap());
1074 * Verify mappings from CDS to peptide, cDNA to CDS, and cDNA to peptide
1075 * the mappings are on the shared alignment dataset
1076 * 6 mappings, 2*(DNA->CDS), 2*(DNA->Pep), 2*(CDS->Pep)
1078 List<AlignedCodonFrame> cdsMappings = cds.getDataset().getCodonFrames();
1079 assertEquals(6, cdsMappings.size());
1082 * verify that mapping sets for dna and cds alignments are different
1083 * [not current behaviour - all mappings are on the alignment dataset]
1085 // select -> subselect type to test.
1086 // Assert.assertNotSame(dna.getCodonFrames(), cds.getCodonFrames());
1087 // assertEquals(4, dna.getCodonFrames().size());
1088 // assertEquals(4, cds.getCodonFrames().size());
1091 * Two mappings involve pep1 (dna to pep1, cds to pep1)
1092 * Mapping from pep1 to GGGTTT in first new exon sequence
1094 List<AlignedCodonFrame> pep1Mappings = MappingUtils
1095 .findMappingsForSequence(pep1, cdsMappings);
1096 assertEquals(2, pep1Mappings.size());
1097 List<AlignedCodonFrame> mappings = MappingUtils
1098 .findMappingsForSequence(cds.getSequenceAt(0), pep1Mappings);
1099 assertEquals(1, mappings.size());
1102 SearchResults sr = MappingUtils.buildSearchResults(pep1, 1, mappings);
1103 assertEquals(1, sr.getResults().size());
1104 Match m = sr.getResults().get(0);
1105 assertSame(cds1Dss, m.getSequence());
1106 assertEquals(1, m.getStart());
1107 assertEquals(3, m.getEnd());
1109 sr = MappingUtils.buildSearchResults(pep1, 2, mappings);
1110 m = sr.getResults().get(0);
1111 assertSame(cds1Dss, m.getSequence());
1112 assertEquals(4, m.getStart());
1113 assertEquals(6, m.getEnd());
1116 * Two mappings involve pep2 (dna to pep2, cds to pep2)
1117 * Verify mapping from pep2 to GGGTTTCCC in second new exon sequence
1119 List<AlignedCodonFrame> pep2Mappings = MappingUtils
1120 .findMappingsForSequence(pep2, cdsMappings);
1121 assertEquals(2, pep2Mappings.size());
1122 mappings = MappingUtils.findMappingsForSequence(cds.getSequenceAt(1),
1124 assertEquals(1, mappings.size());
1126 sr = MappingUtils.buildSearchResults(pep2, 1, mappings);
1127 assertEquals(1, sr.getResults().size());
1128 m = sr.getResults().get(0);
1129 assertSame(cds2Dss, m.getSequence());
1130 assertEquals(1, m.getStart());
1131 assertEquals(3, m.getEnd());
1133 sr = MappingUtils.buildSearchResults(pep2, 2, mappings);
1134 m = sr.getResults().get(0);
1135 assertSame(cds2Dss, m.getSequence());
1136 assertEquals(4, m.getStart());
1137 assertEquals(6, m.getEnd());
1139 sr = MappingUtils.buildSearchResults(pep2, 3, mappings);
1140 m = sr.getResults().get(0);
1141 assertSame(cds2Dss, m.getSequence());
1142 assertEquals(7, m.getStart());
1143 assertEquals(9, m.getEnd());
1147 * Test the method that makes a cds-only alignment from a DNA sequence and its
1148 * product mappings, for the case where there are multiple exon mappings to
1149 * different protein products.
1151 @Test(groups = { "Functional" })
1152 public void testMakeCdsAlignment_multipleProteins()
1154 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
1155 SequenceI pep1 = new Sequence("pep1", "GF"); // GGGTTT
1156 SequenceI pep2 = new Sequence("pep2", "KP"); // aaaccc
1157 SequenceI pep3 = new Sequence("pep3", "KF"); // aaaTTT
1158 dna1.createDatasetSequence();
1159 pep1.createDatasetSequence();
1160 pep2.createDatasetSequence();
1161 pep3.createDatasetSequence();
1162 pep1.getDatasetSequence().addDBRef(
1163 new DBRefEntry("EMBLCDS", "2", "A12345"));
1164 pep2.getDatasetSequence().addDBRef(
1165 new DBRefEntry("EMBLCDS", "3", "A12346"));
1166 pep3.getDatasetSequence().addDBRef(
1167 new DBRefEntry("EMBLCDS", "4", "A12347"));
1170 * Create the CDS alignment
1172 AlignmentI dna = new Alignment(new SequenceI[] { dna1 });
1173 dna.setDataset(null);
1176 * Make the mappings from dna to protein
1178 // map ...GGG...TTT to GF
1179 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1180 new int[] { 1, 2 }, 3, 1);
1181 AlignedCodonFrame acf = new AlignedCodonFrame();
1182 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
1183 dna.addCodonFrame(acf);
1185 // map aaa...ccc to KP
1186 map = new MapList(new int[] { 1, 3, 7, 9 }, new int[] { 1, 2 }, 3, 1);
1187 acf = new AlignedCodonFrame();
1188 acf.addMap(dna1.getDatasetSequence(), pep2.getDatasetSequence(), map);
1189 dna.addCodonFrame(acf);
1191 // map aaa......TTT to KF
1192 map = new MapList(new int[] { 1, 3, 10, 12 }, new int[] { 1, 2 }, 3, 1);
1193 acf = new AlignedCodonFrame();
1194 acf.addMap(dna1.getDatasetSequence(), pep3.getDatasetSequence(), map);
1195 dna.addCodonFrame(acf);
1198 * execute method under test
1200 AlignmentI cdsal = AlignmentUtils.makeCdsAlignment(
1201 new SequenceI[] { dna1 }, dna.getDataset(), null);
1204 * Verify we have 3 cds sequences, mapped to pep1/2/3 respectively
1206 List<SequenceI> cds = cdsal.getSequences();
1207 assertEquals(3, cds.size());
1210 * verify shared, extended alignment dataset
1212 assertSame(cdsal.getDataset(), dna.getDataset());
1213 assertTrue(dna.getDataset().getSequences()
1214 .contains(cds.get(0).getDatasetSequence()));
1215 assertTrue(dna.getDataset().getSequences()
1216 .contains(cds.get(1).getDatasetSequence()));
1217 assertTrue(dna.getDataset().getSequences()
1218 .contains(cds.get(2).getDatasetSequence()));
1221 * verify aligned cds sequences and their xrefs
1223 SequenceI cdsSeq = cds.get(0);
1224 assertEquals("GGGTTT", cdsSeq.getSequenceAsString());
1225 // assertEquals("dna1|A12345", cdsSeq.getName());
1226 assertEquals("CDS|dna1", cdsSeq.getName());
1227 // assertEquals(1, cdsSeq.getDBRefs().length);
1228 // DBRefEntry cdsRef = cdsSeq.getDBRefs()[0];
1229 // assertEquals("EMBLCDS", cdsRef.getSource());
1230 // assertEquals("2", cdsRef.getVersion());
1231 // assertEquals("A12345", cdsRef.getAccessionId());
1233 cdsSeq = cds.get(1);
1234 assertEquals("aaaccc", cdsSeq.getSequenceAsString());
1235 // assertEquals("dna1|A12346", cdsSeq.getName());
1236 assertEquals("CDS|dna1", cdsSeq.getName());
1237 // assertEquals(1, cdsSeq.getDBRefs().length);
1238 // cdsRef = cdsSeq.getDBRefs()[0];
1239 // assertEquals("EMBLCDS", cdsRef.getSource());
1240 // assertEquals("3", cdsRef.getVersion());
1241 // assertEquals("A12346", cdsRef.getAccessionId());
1243 cdsSeq = cds.get(2);
1244 assertEquals("aaaTTT", cdsSeq.getSequenceAsString());
1245 // assertEquals("dna1|A12347", cdsSeq.getName());
1246 assertEquals("CDS|dna1", cdsSeq.getName());
1247 // assertEquals(1, cdsSeq.getDBRefs().length);
1248 // cdsRef = cdsSeq.getDBRefs()[0];
1249 // assertEquals("EMBLCDS", cdsRef.getSource());
1250 // assertEquals("4", cdsRef.getVersion());
1251 // assertEquals("A12347", cdsRef.getAccessionId());
1254 * Verify there are mappings from each cds sequence to its protein product
1255 * and also to its dna source
1257 List<AlignedCodonFrame> newMappings = cdsal.getCodonFrames();
1260 * 6 mappings involve dna1 (to pep1/2/3, cds1/2/3)
1262 List<AlignedCodonFrame> dnaMappings = MappingUtils
1263 .findMappingsForSequence(dna1, newMappings);
1264 assertEquals(6, dnaMappings.size());
1269 List<AlignedCodonFrame> mappings = MappingUtils
1270 .findMappingsForSequence(pep1, dnaMappings);
1271 assertEquals(1, mappings.size());
1272 assertEquals(1, mappings.get(0).getMappings().size());
1273 assertSame(pep1.getDatasetSequence(), mappings.get(0).getMappings()
1274 .get(0).getMapping().getTo());
1279 List<AlignedCodonFrame> dnaToCds1Mappings = MappingUtils
1280 .findMappingsForSequence(cds.get(0), dnaMappings);
1281 Mapping mapping = dnaToCds1Mappings.get(0).getMappings().get(0)
1283 assertSame(cds.get(0).getDatasetSequence(), mapping
1285 assertEquals("G(1) in CDS should map to G(4) in DNA", 4, mapping
1286 .getMap().getToPosition(1));
1291 mappings = MappingUtils.findMappingsForSequence(pep2, dnaMappings);
1292 assertEquals(1, mappings.size());
1293 assertEquals(1, mappings.get(0).getMappings().size());
1294 assertSame(pep2.getDatasetSequence(), mappings.get(0).getMappings()
1295 .get(0).getMapping().getTo());
1300 List<AlignedCodonFrame> dnaToCds2Mappings = MappingUtils
1301 .findMappingsForSequence(cds.get(1), dnaMappings);
1302 mapping = dnaToCds2Mappings.get(0).getMappings().get(0).getMapping();
1303 assertSame(cds.get(1).getDatasetSequence(), mapping.getTo());
1304 assertEquals("c(4) in CDS should map to c(7) in DNA", 7, mapping
1305 .getMap().getToPosition(4));
1310 mappings = MappingUtils.findMappingsForSequence(pep3, dnaMappings);
1311 assertEquals(1, mappings.size());
1312 assertEquals(1, mappings.get(0).getMappings().size());
1313 assertSame(pep3.getDatasetSequence(), mappings.get(0).getMappings()
1314 .get(0).getMapping().getTo());
1319 List<AlignedCodonFrame> dnaToCds3Mappings = MappingUtils
1320 .findMappingsForSequence(cds.get(2), dnaMappings);
1321 mapping = dnaToCds3Mappings.get(0).getMappings().get(0).getMapping();
1322 assertSame(cds.get(2).getDatasetSequence(), mapping.getTo());
1323 assertEquals("T(4) in CDS should map to T(10) in DNA", 10, mapping
1324 .getMap().getToPosition(4));
1327 @Test(groups = { "Functional" })
1328 public void testIsMappable()
1330 SequenceI dna1 = new Sequence("dna1", "cgCAGtgGT");
1331 SequenceI aa1 = new Sequence("aa1", "RSG");
1332 AlignmentI al1 = new Alignment(new SequenceI[] { dna1 });
1333 AlignmentI al2 = new Alignment(new SequenceI[] { aa1 });
1335 assertFalse(AlignmentUtils.isMappable(null, null));
1336 assertFalse(AlignmentUtils.isMappable(al1, null));
1337 assertFalse(AlignmentUtils.isMappable(null, al1));
1338 assertFalse(AlignmentUtils.isMappable(al1, al1));
1339 assertFalse(AlignmentUtils.isMappable(al2, al2));
1341 assertTrue(AlignmentUtils.isMappable(al1, al2));
1342 assertTrue(AlignmentUtils.isMappable(al2, al1));
1346 * Test creating a mapping when the sequences involved do not start at residue
1349 * @throws IOException
1351 @Test(groups = { "Functional" })
1352 public void testMapCdnaToProtein_forSubsequence()
1355 SequenceI prot = new Sequence("UNIPROT|V12345", "E-I--Q", 10, 12);
1356 prot.createDatasetSequence();
1358 SequenceI dna = new Sequence("EMBL|A33333", "GAA--AT-C-CAG", 40, 48);
1359 dna.createDatasetSequence();
1361 MapList map = AlignmentUtils.mapCdnaToProtein(prot, dna);
1362 assertEquals(10, map.getToLowest());
1363 assertEquals(12, map.getToHighest());
1364 assertEquals(40, map.getFromLowest());
1365 assertEquals(48, map.getFromHighest());
1369 * Test for the alignSequenceAs method where we have protein mapped to protein
1371 @Test(groups = { "Functional" })
1372 public void testAlignSequenceAs_mappedProteinProtein()
1375 SequenceI alignMe = new Sequence("Match", "MGAASEV");
1376 alignMe.createDatasetSequence();
1377 SequenceI alignFrom = new Sequence("Query", "LQTGYMGAASEVMFSPTRR");
1378 alignFrom.createDatasetSequence();
1380 AlignedCodonFrame acf = new AlignedCodonFrame();
1381 // this is like a domain or motif match of part of a peptide sequence
1382 MapList map = new MapList(new int[] { 6, 12 }, new int[] { 1, 7 }, 1, 1);
1383 acf.addMap(alignFrom.getDatasetSequence(),
1384 alignMe.getDatasetSequence(), map);
1386 AlignmentUtils.alignSequenceAs(alignMe, alignFrom, acf, "-", '-', true,
1388 assertEquals("-----MGAASEV-------", alignMe.getSequenceAsString());
1392 * Test for the alignSequenceAs method where there are trailing unmapped
1393 * residues in the model sequence
1395 @Test(groups = { "Functional" })
1396 public void testAlignSequenceAs_withTrailingPeptide()
1398 // map first 3 codons to KPF; G is a trailing unmapped residue
1399 MapList map = new MapList(new int[] { 1, 9 }, new int[] { 1, 3 }, 3, 1);
1401 checkAlignSequenceAs("AAACCCTTT", "K-PFG", true, true, map,
1406 * Tests for transferring features between mapped sequences
1408 @Test(groups = { "Functional" })
1409 public void testTransferFeatures()
1411 SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
1412 SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
1415 dna.addSequenceFeature(new SequenceFeature("type1", "desc1", 1, 2, 1f,
1417 // partial overlap - to [1, 1]
1418 dna.addSequenceFeature(new SequenceFeature("type2", "desc2", 3, 4, 2f,
1420 // exact overlap - to [1, 3]
1421 dna.addSequenceFeature(new SequenceFeature("type3", "desc3", 4, 6, 3f,
1423 // spanning overlap - to [2, 5]
1424 dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
1426 // exactly overlaps whole mapped range [1, 6]
1427 dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
1429 // no overlap (internal)
1430 dna.addSequenceFeature(new SequenceFeature("type6", "desc6", 7, 9, 6f,
1432 // no overlap (3' end)
1433 dna.addSequenceFeature(new SequenceFeature("type7", "desc7", 13, 15,
1435 // overlap (3' end) - to [6, 6]
1436 dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
1438 // extended overlap - to [6, +]
1439 dna.addSequenceFeature(new SequenceFeature("type9", "desc9", 12, 13,
1442 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1443 new int[] { 1, 6 }, 1, 1);
1446 * transferFeatures() will build 'partial overlap' for regions
1447 * that partially overlap 5' or 3' (start or end) of target sequence
1449 AlignmentUtils.transferFeatures(dna, cds, map, null);
1450 SequenceFeature[] sfs = cds.getSequenceFeatures();
1451 assertEquals(6, sfs.length);
1453 SequenceFeature sf = sfs[0];
1454 assertEquals("type2", sf.getType());
1455 assertEquals("desc2", sf.getDescription());
1456 assertEquals(2f, sf.getScore());
1457 assertEquals(1, sf.getBegin());
1458 assertEquals(1, sf.getEnd());
1461 assertEquals("type3", sf.getType());
1462 assertEquals("desc3", sf.getDescription());
1463 assertEquals(3f, sf.getScore());
1464 assertEquals(1, sf.getBegin());
1465 assertEquals(3, sf.getEnd());
1468 assertEquals("type4", sf.getType());
1469 assertEquals(2, sf.getBegin());
1470 assertEquals(5, sf.getEnd());
1473 assertEquals("type5", sf.getType());
1474 assertEquals(1, sf.getBegin());
1475 assertEquals(6, sf.getEnd());
1478 assertEquals("type8", sf.getType());
1479 assertEquals(6, sf.getBegin());
1480 assertEquals(6, sf.getEnd());
1483 assertEquals("type9", sf.getType());
1484 assertEquals(6, sf.getBegin());
1485 assertEquals(6, sf.getEnd());
1489 * Tests for transferring features between mapped sequences
1491 @Test(groups = { "Functional" })
1492 public void testTransferFeatures_withOmit()
1494 SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
1495 SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
1497 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1498 new int[] { 1, 6 }, 1, 1);
1500 // [5, 11] maps to [2, 5]
1501 dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
1503 // [4, 12] maps to [1, 6]
1504 dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
1506 // [12, 12] maps to [6, 6]
1507 dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
1510 // desc4 and desc8 are the 'omit these' varargs
1511 AlignmentUtils.transferFeatures(dna, cds, map, null, "type4", "type8");
1512 SequenceFeature[] sfs = cds.getSequenceFeatures();
1513 assertEquals(1, sfs.length);
1515 SequenceFeature sf = sfs[0];
1516 assertEquals("type5", sf.getType());
1517 assertEquals(1, sf.getBegin());
1518 assertEquals(6, sf.getEnd());
1522 * Tests for transferring features between mapped sequences
1524 @Test(groups = { "Functional" })
1525 public void testTransferFeatures_withSelect()
1527 SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
1528 SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
1530 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1531 new int[] { 1, 6 }, 1, 1);
1533 // [5, 11] maps to [2, 5]
1534 dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
1536 // [4, 12] maps to [1, 6]
1537 dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
1539 // [12, 12] maps to [6, 6]
1540 dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
1543 // "type5" is the 'select this type' argument
1544 AlignmentUtils.transferFeatures(dna, cds, map, "type5");
1545 SequenceFeature[] sfs = cds.getSequenceFeatures();
1546 assertEquals(1, sfs.length);
1548 SequenceFeature sf = sfs[0];
1549 assertEquals("type5", sf.getType());
1550 assertEquals(1, sf.getBegin());
1551 assertEquals(6, sf.getEnd());
1555 * Test the method that extracts the cds-only part of a dna alignment, for the
1556 * case where the cds should be aligned to match its nucleotide sequence.
1558 @Test(groups = { "Functional" })
1559 public void testMakeCdsAlignment_alternativeTranscripts()
1561 SequenceI dna1 = new Sequence("dna1", "aaaGGGCC-----CTTTaaaGGG");
1562 // alternative transcript of same dna skips CCC codon
1563 SequenceI dna2 = new Sequence("dna2", "aaaGGGCC-----cttTaaaGGG");
1564 // dna3 has no mapping (protein product) so should be ignored here
1565 SequenceI dna3 = new Sequence("dna3", "aaaGGGCCCCCGGGcttTaaaGGG");
1566 SequenceI pep1 = new Sequence("pep1", "GPFG");
1567 SequenceI pep2 = new Sequence("pep2", "GPG");
1568 dna1.createDatasetSequence();
1569 dna2.createDatasetSequence();
1570 dna3.createDatasetSequence();
1571 pep1.createDatasetSequence();
1572 pep2.createDatasetSequence();
1574 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
1575 dna.setDataset(null);
1577 MapList map = new MapList(new int[] { 4, 12, 16, 18 },
1578 new int[] { 1, 4 }, 3, 1);
1579 AlignedCodonFrame acf = new AlignedCodonFrame();
1580 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
1581 dna.addCodonFrame(acf);
1582 map = new MapList(new int[] { 4, 8, 12, 12, 16, 18 },
1585 acf = new AlignedCodonFrame();
1586 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map);
1587 dna.addCodonFrame(acf);
1589 AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
1590 dna1, dna2, dna3 }, dna.getDataset(), null);
1591 List<SequenceI> cdsSeqs = cds.getSequences();
1592 assertEquals(2, cdsSeqs.size());
1593 assertEquals("GGGCCCTTTGGG", cdsSeqs.get(0).getSequenceAsString());
1594 assertEquals("GGGCCTGGG", cdsSeqs.get(1).getSequenceAsString());
1597 * verify shared, extended alignment dataset
1599 assertSame(dna.getDataset(), cds.getDataset());
1600 assertTrue(dna.getDataset().getSequences()
1601 .contains(cdsSeqs.get(0).getDatasetSequence()));
1602 assertTrue(dna.getDataset().getSequences()
1603 .contains(cdsSeqs.get(1).getDatasetSequence()));
1606 * Verify 6 mappings: dna1 to cds1, cds1 to pep1, dna1 to pep1
1607 * and the same for dna2/cds2/pep2
1609 List<AlignedCodonFrame> mappings = cds.getCodonFrames();
1610 assertEquals(6, mappings.size());
1613 * 2 mappings involve pep1
1615 List<AlignedCodonFrame> pep1Mappings = MappingUtils
1616 .findMappingsForSequence(pep1, mappings);
1617 assertEquals(2, pep1Mappings.size());
1620 * Get mapping of pep1 to cds1 and verify it
1621 * maps GPFG to 1-3,4-6,7-9,10-12
1623 List<AlignedCodonFrame> pep1CdsMappings = MappingUtils
1624 .findMappingsForSequence(cds.getSequenceAt(0), pep1Mappings);
1625 assertEquals(1, pep1CdsMappings.size());
1626 SearchResults sr = MappingUtils.buildSearchResults(pep1, 1,
1628 assertEquals(1, sr.getResults().size());
1629 Match m = sr.getResults().get(0);
1630 assertEquals(cds.getSequenceAt(0).getDatasetSequence(),
1632 assertEquals(1, m.getStart());
1633 assertEquals(3, m.getEnd());
1634 sr = MappingUtils.buildSearchResults(pep1, 2, pep1CdsMappings);
1635 m = sr.getResults().get(0);
1636 assertEquals(4, m.getStart());
1637 assertEquals(6, m.getEnd());
1638 sr = MappingUtils.buildSearchResults(pep1, 3, pep1CdsMappings);
1639 m = sr.getResults().get(0);
1640 assertEquals(7, m.getStart());
1641 assertEquals(9, m.getEnd());
1642 sr = MappingUtils.buildSearchResults(pep1, 4, pep1CdsMappings);
1643 m = sr.getResults().get(0);
1644 assertEquals(10, m.getStart());
1645 assertEquals(12, m.getEnd());
1648 * Get mapping of pep2 to cds2 and verify it
1649 * maps GPG in pep2 to 1-3,4-6,7-9 in second CDS sequence
1651 List<AlignedCodonFrame> pep2Mappings = MappingUtils
1652 .findMappingsForSequence(pep2, mappings);
1653 assertEquals(2, pep2Mappings.size());
1654 List<AlignedCodonFrame> pep2CdsMappings = MappingUtils
1655 .findMappingsForSequence(cds.getSequenceAt(1), pep2Mappings);
1656 assertEquals(1, pep2CdsMappings.size());
1657 sr = MappingUtils.buildSearchResults(pep2, 1, pep2CdsMappings);
1658 assertEquals(1, sr.getResults().size());
1659 m = sr.getResults().get(0);
1660 assertEquals(cds.getSequenceAt(1).getDatasetSequence(),
1662 assertEquals(1, m.getStart());
1663 assertEquals(3, m.getEnd());
1664 sr = MappingUtils.buildSearchResults(pep2, 2, pep2CdsMappings);
1665 m = sr.getResults().get(0);
1666 assertEquals(4, m.getStart());
1667 assertEquals(6, m.getEnd());
1668 sr = MappingUtils.buildSearchResults(pep2, 3, pep2CdsMappings);
1669 m = sr.getResults().get(0);
1670 assertEquals(7, m.getStart());
1671 assertEquals(9, m.getEnd());
1675 * Test the method that realigns protein to match mapped codon alignment.
1677 @Test(groups = { "Functional" })
1678 public void testAlignProteinAsDna_incompleteStartCodon()
1680 // seq1: incomplete start codon (not mapped), then [3, 11]
1681 SequenceI dna1 = new Sequence("Seq1", "ccAAA-TTT-GGG-");
1682 // seq2 codons are [4, 5], [8, 11]
1683 SequenceI dna2 = new Sequence("Seq2", "ccaAA-ttT-GGG-");
1684 // seq3 incomplete start codon at 'tt'
1685 SequenceI dna3 = new Sequence("Seq3", "ccaaa-ttt-GGG-");
1686 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
1687 dna.setDataset(null);
1689 // prot1 has 'X' for incomplete start codon (not mapped)
1690 SequenceI prot1 = new Sequence("Seq1", "XKFG"); // X for incomplete start
1691 SequenceI prot2 = new Sequence("Seq2", "NG");
1692 SequenceI prot3 = new Sequence("Seq3", "XG"); // X for incomplete start
1693 AlignmentI protein = new Alignment(new SequenceI[] { prot1, prot2,
1695 protein.setDataset(null);
1697 // map dna1 [3, 11] to prot1 [2, 4] KFG
1698 MapList map = new MapList(new int[] { 3, 11 }, new int[] { 2, 4 }, 3, 1);
1699 AlignedCodonFrame acf = new AlignedCodonFrame();
1700 acf.addMap(dna1.getDatasetSequence(), prot1.getDatasetSequence(), map);
1702 // map dna2 [4, 5] [8, 11] to prot2 [1, 2] NG
1703 map = new MapList(new int[] { 4, 5, 8, 11 }, new int[] { 1, 2 }, 3, 1);
1704 acf.addMap(dna2.getDatasetSequence(), prot2.getDatasetSequence(), map);
1706 // map dna3 [9, 11] to prot3 [2, 2] G
1707 map = new MapList(new int[] { 9, 11 }, new int[] { 2, 2 }, 3, 1);
1708 acf.addMap(dna3.getDatasetSequence(), prot3.getDatasetSequence(), map);
1710 ArrayList<AlignedCodonFrame> acfs = new ArrayList<AlignedCodonFrame>();
1712 protein.setCodonFrames(acfs);
1715 * verify X is included in the aligned proteins, and placed just
1716 * before the first mapped residue
1717 * CCT is between CCC and TTT
1719 AlignmentUtils.alignProteinAsDna(protein, dna);
1720 assertEquals("XK-FG", prot1.getSequenceAsString());
1721 assertEquals("--N-G", prot2.getSequenceAsString());
1722 assertEquals("---XG", prot3.getSequenceAsString());
1726 * Tests for the method that maps the subset of a dna sequence that has CDS
1727 * (or subtype) feature - case where the start codon is incomplete.
1729 @Test(groups = "Functional")
1730 public void testFindCdsPositions_fivePrimeIncomplete()
1732 SequenceI dnaSeq = new Sequence("dna", "aaagGGCCCaaaTTTttt");
1733 dnaSeq.createDatasetSequence();
1734 SequenceI ds = dnaSeq.getDatasetSequence();
1736 // CDS for dna 5-6 (incomplete codon), 7-9
1737 SequenceFeature sf = new SequenceFeature("CDS", "", 5, 9, 0f, null);
1738 sf.setPhase("2"); // skip 2 bases to start of next codon
1739 ds.addSequenceFeature(sf);
1740 // CDS for dna 13-15
1741 sf = new SequenceFeature("CDS_predicted", "", 13, 15, 0f, null);
1742 ds.addSequenceFeature(sf);
1744 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
1747 * check the mapping starts with the first complete codon
1749 assertEquals(6, MappingUtils.getLength(ranges));
1750 assertEquals(2, ranges.size());
1751 assertEquals(7, ranges.get(0)[0]);
1752 assertEquals(9, ranges.get(0)[1]);
1753 assertEquals(13, ranges.get(1)[0]);
1754 assertEquals(15, ranges.get(1)[1]);
1758 * Tests for the method that maps the subset of a dna sequence that has CDS
1759 * (or subtype) feature.
1761 @Test(groups = "Functional")
1762 public void testFindCdsPositions()
1764 SequenceI dnaSeq = new Sequence("dna", "aaaGGGcccAAATTTttt");
1765 dnaSeq.createDatasetSequence();
1766 SequenceI ds = dnaSeq.getDatasetSequence();
1768 // CDS for dna 10-12
1769 SequenceFeature sf = new SequenceFeature("CDS_predicted", "", 10, 12,
1772 ds.addSequenceFeature(sf);
1774 sf = new SequenceFeature("CDS", "", 4, 6, 0f, null);
1776 ds.addSequenceFeature(sf);
1777 // exon feature should be ignored here
1778 sf = new SequenceFeature("exon", "", 7, 9, 0f, null);
1779 ds.addSequenceFeature(sf);
1781 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
1783 * verify ranges { [4-6], [12-10] }
1784 * note CDS ranges are ordered ascending even if the CDS
1787 assertEquals(6, MappingUtils.getLength(ranges));
1788 assertEquals(2, ranges.size());
1789 assertEquals(4, ranges.get(0)[0]);
1790 assertEquals(6, ranges.get(0)[1]);
1791 assertEquals(10, ranges.get(1)[0]);
1792 assertEquals(12, ranges.get(1)[1]);
1796 * Test the method that computes a map of codon variants for each protein
1797 * position from "sequence_variant" features on dna
1799 @Test(groups = "Functional")
1800 public void testBuildDnaVariantsMap()
1802 SequenceI dna = new Sequence("dna", "atgAAATTTGGGCCCtag");
1803 MapList map = new MapList(new int[] { 1, 18 }, new int[] { 1, 5 }, 3, 1);
1806 * first with no variants on dna
1808 LinkedHashMap<Integer, List<DnaVariant>[]> variantsMap = AlignmentUtils
1809 .buildDnaVariantsMap(dna, map);
1810 assertTrue(variantsMap.isEmpty());
1813 * single allele codon 1, on base 1
1815 SequenceFeature sf1 = new SequenceFeature("sequence_variant", "", 1, 1,
1817 sf1.setValue("alleles", "T");
1818 sf1.setValue("ID", "sequence_variant:rs758803211");
1819 dna.addSequenceFeature(sf1);
1822 * two alleles codon 2, on bases 2 and 3 (distinct variants)
1824 SequenceFeature sf2 = new SequenceFeature("sequence_variant", "", 5, 5,
1826 sf2.setValue("alleles", "T");
1827 sf2.setValue("ID", "sequence_variant:rs758803212");
1828 dna.addSequenceFeature(sf2);
1829 SequenceFeature sf3 = new SequenceFeature("sequence_variant", "", 6, 6,
1831 sf3.setValue("alleles", "G");
1832 sf3.setValue("ID", "sequence_variant:rs758803213");
1833 dna.addSequenceFeature(sf3);
1836 * two alleles codon 3, both on base 2 (one variant)
1838 SequenceFeature sf4 = new SequenceFeature("sequence_variant", "", 8, 8,
1840 sf4.setValue("alleles", "C, G");
1841 sf4.setValue("ID", "sequence_variant:rs758803214");
1842 dna.addSequenceFeature(sf4);
1844 // no alleles on codon 4
1847 * alleles on codon 5 on all 3 bases (distinct variants)
1849 SequenceFeature sf5 = new SequenceFeature("sequence_variant", "", 13,
1851 sf5.setValue("alleles", "C, G"); // (C duplicates given base value)
1852 sf5.setValue("ID", "sequence_variant:rs758803215");
1853 dna.addSequenceFeature(sf5);
1854 SequenceFeature sf6 = new SequenceFeature("sequence_variant", "", 14,
1856 sf6.setValue("alleles", "g, a"); // should force to upper-case
1857 sf6.setValue("ID", "sequence_variant:rs758803216");
1858 dna.addSequenceFeature(sf6);
1859 SequenceFeature sf7 = new SequenceFeature("sequence_variant", "", 15,
1861 sf7.setValue("alleles", "A, T");
1862 sf7.setValue("ID", "sequence_variant:rs758803217");
1863 dna.addSequenceFeature(sf7);
1866 * build map - expect variants on positions 1, 2, 3, 5
1868 variantsMap = AlignmentUtils.buildDnaVariantsMap(dna, map);
1869 assertEquals(4, variantsMap.size());
1872 * protein residue 1: variant on codon (ATG) base 1, not on 2 or 3
1874 List<DnaVariant>[] pep1Variants = variantsMap.get(1);
1875 assertEquals(3, pep1Variants.length);
1876 assertEquals(1, pep1Variants[0].size());
1877 assertEquals("A", pep1Variants[0].get(0).base); // codon[1] base
1878 assertSame(sf1, pep1Variants[0].get(0).variant); // codon[1] variant
1879 assertEquals(1, pep1Variants[1].size());
1880 assertEquals("T", pep1Variants[1].get(0).base); // codon[2] base
1881 assertNull(pep1Variants[1].get(0).variant); // no variant here
1882 assertEquals(1, pep1Variants[2].size());
1883 assertEquals("G", pep1Variants[2].get(0).base); // codon[3] base
1884 assertNull(pep1Variants[2].get(0).variant); // no variant here
1887 * protein residue 2: variants on codon (AAA) bases 2 and 3
1889 List<DnaVariant>[] pep2Variants = variantsMap.get(2);
1890 assertEquals(3, pep2Variants.length);
1891 assertEquals(1, pep2Variants[0].size());
1892 // codon[1] base recorded while processing variant on codon[2]
1893 assertEquals("A", pep2Variants[0].get(0).base);
1894 assertNull(pep2Variants[0].get(0).variant); // no variant here
1895 // codon[2] base and variant:
1896 assertEquals(1, pep2Variants[1].size());
1897 assertEquals("A", pep2Variants[1].get(0).base);
1898 assertSame(sf2, pep2Variants[1].get(0).variant);
1899 // codon[3] base was recorded when processing codon[2] variant
1900 // and then the variant for codon[3] added to it
1901 assertEquals(1, pep2Variants[2].size());
1902 assertEquals("A", pep2Variants[2].get(0).base);
1903 assertSame(sf3, pep2Variants[2].get(0).variant);
1906 * protein residue 3: variants on codon (TTT) base 2 only
1908 List<DnaVariant>[] pep3Variants = variantsMap.get(3);
1909 assertEquals(3, pep3Variants.length);
1910 assertEquals(1, pep3Variants[0].size());
1911 assertEquals("T", pep3Variants[0].get(0).base); // codon[1] base
1912 assertNull(pep3Variants[0].get(0).variant); // no variant here
1913 assertEquals(1, pep3Variants[1].size());
1914 assertEquals("T", pep3Variants[1].get(0).base); // codon[2] base
1915 assertSame(sf4, pep3Variants[1].get(0).variant); // codon[2] variant
1916 assertEquals(1, pep3Variants[2].size());
1917 assertEquals("T", pep3Variants[2].get(0).base); // codon[3] base
1918 assertNull(pep3Variants[2].get(0).variant); // no variant here
1921 * three variants on protein position 5
1923 List<DnaVariant>[] pep5Variants = variantsMap.get(5);
1924 assertEquals(3, pep5Variants.length);
1925 assertEquals(1, pep5Variants[0].size());
1926 assertEquals("C", pep5Variants[0].get(0).base); // codon[1] base
1927 assertSame(sf5, pep5Variants[0].get(0).variant); // codon[1] variant
1928 assertEquals(1, pep5Variants[1].size());
1929 assertEquals("C", pep5Variants[1].get(0).base); // codon[2] base
1930 assertSame(sf6, pep5Variants[1].get(0).variant); // codon[2] variant
1931 assertEquals(1, pep5Variants[2].size());
1932 assertEquals("C", pep5Variants[2].get(0).base); // codon[3] base
1933 assertSame(sf7, pep5Variants[2].get(0).variant); // codon[3] variant
1937 * Tests for the method that computes all peptide variants given codon
1940 @Test(groups = "Functional")
1941 public void testComputePeptideVariants()
1944 * scenario: AAATTTCCC codes for KFP
1946 * GAA -> E source: Ensembl
1947 * CAA -> Q source: dbSNP
1948 * AAG synonymous source: COSMIC
1949 * AAT -> N source: Ensembl
1950 * ...TTC synonymous source: dbSNP
1951 * ......CAC,CGC -> H,R source: COSMIC
1952 * (one variant with two alleles)
1954 SequenceI peptide = new Sequence("pep/10-12", "KFP");
1957 * two distinct variants for codon 1 position 1
1958 * second one has clinical significance
1960 String ensembl = "Ensembl";
1961 String dbSnp = "dbSNP";
1962 String cosmic = "COSMIC";
1963 SequenceFeature sf1 = new SequenceFeature("sequence_variant", "", 1, 1,
1965 sf1.setValue("alleles", "A,G"); // GAA -> E
1966 sf1.setValue("ID", "var1.125A>G");
1967 SequenceFeature sf2 = new SequenceFeature("sequence_variant", "", 1, 1,
1969 sf2.setValue("alleles", "A,C"); // CAA -> Q
1970 sf2.setValue("ID", "var2");
1971 sf2.setValue("clinical_significance", "Dodgy");
1972 SequenceFeature sf3 = new SequenceFeature("sequence_variant", "", 3, 3,
1974 sf3.setValue("alleles", "A,G"); // synonymous
1975 sf3.setValue("ID", "var3");
1976 sf3.setValue("clinical_significance", "None");
1977 SequenceFeature sf4 = new SequenceFeature("sequence_variant", "", 3, 3,
1979 sf4.setValue("alleles", "A,T"); // AAT -> N
1980 sf4.setValue("ID", "sequence_variant:var4"); // prefix gets stripped off
1981 sf4.setValue("clinical_significance", "Benign");
1982 SequenceFeature sf5 = new SequenceFeature("sequence_variant", "", 6, 6,
1984 sf5.setValue("alleles", "T,C"); // synonymous
1985 sf5.setValue("ID", "var5");
1986 sf5.setValue("clinical_significance", "Bad");
1987 SequenceFeature sf6 = new SequenceFeature("sequence_variant", "", 8, 8,
1989 sf6.setValue("alleles", "C,A,G"); // CAC,CGC -> H,R
1990 sf6.setValue("ID", "var6");
1991 sf6.setValue("clinical_significance", "Good");
1993 List<DnaVariant> codon1Variants = new ArrayList<DnaVariant>();
1994 List<DnaVariant> codon2Variants = new ArrayList<DnaVariant>();
1995 List<DnaVariant> codon3Variants = new ArrayList<DnaVariant>();
1996 List<DnaVariant> codonVariants[] = new ArrayList[3];
1997 codonVariants[0] = codon1Variants;
1998 codonVariants[1] = codon2Variants;
1999 codonVariants[2] = codon3Variants;
2002 * compute variants for protein position 1
2004 codon1Variants.add(new DnaVariant("A", sf1));
2005 codon1Variants.add(new DnaVariant("A", sf2));
2006 codon2Variants.add(new DnaVariant("A"));
2007 codon2Variants.add(new DnaVariant("A"));
2008 codon3Variants.add(new DnaVariant("A", sf3));
2009 codon3Variants.add(new DnaVariant("A", sf4));
2010 AlignmentUtils.computePeptideVariants(peptide, 1, codonVariants);
2013 * compute variants for protein position 2
2015 codon1Variants.clear();
2016 codon2Variants.clear();
2017 codon3Variants.clear();
2018 codon1Variants.add(new DnaVariant("T"));
2019 codon2Variants.add(new DnaVariant("T"));
2020 codon3Variants.add(new DnaVariant("T", sf5));
2021 AlignmentUtils.computePeptideVariants(peptide, 2, codonVariants);
2024 * compute variants for protein position 3
2026 codon1Variants.clear();
2027 codon2Variants.clear();
2028 codon3Variants.clear();
2029 codon1Variants.add(new DnaVariant("C"));
2030 codon2Variants.add(new DnaVariant("C", sf6));
2031 codon3Variants.add(new DnaVariant("C"));
2032 AlignmentUtils.computePeptideVariants(peptide, 3, codonVariants);
2035 * verify added sequence features for
2036 * var1 K -> E Ensembl
2038 * var4 K -> N Ensembl
2039 * var6 P -> H COSMIC
2040 * var6 P -> R COSMIC
2042 SequenceFeature[] sfs = peptide.getSequenceFeatures();
2043 assertEquals(5, sfs.length);
2045 SequenceFeature sf = sfs[0];
2046 assertEquals(1, sf.getBegin());
2047 assertEquals(1, sf.getEnd());
2048 assertEquals("p.Lys1Glu", sf.getDescription());
2049 assertEquals("var1.125A>G", sf.getValue("ID"));
2050 assertNull(sf.getValue("clinical_significance"));
2051 assertEquals("ID=var1.125A>G", sf.getAttributes());
2052 assertEquals(1, sf.links.size());
2053 // link to variation is urlencoded
2055 "p.Lys1Glu var1.125A>G|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var1.125A%3EG",
2057 assertEquals(ensembl, sf.getFeatureGroup());
2060 assertEquals(1, sf.getBegin());
2061 assertEquals(1, sf.getEnd());
2062 assertEquals("p.Lys1Gln", sf.getDescription());
2063 assertEquals("var2", sf.getValue("ID"));
2064 assertEquals("Dodgy", sf.getValue("clinical_significance"));
2065 assertEquals("ID=var2;clinical_significance=Dodgy", sf.getAttributes());
2066 assertEquals(1, sf.links.size());
2068 "p.Lys1Gln var2|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var2",
2070 assertEquals(dbSnp, sf.getFeatureGroup());
2073 assertEquals(1, sf.getBegin());
2074 assertEquals(1, sf.getEnd());
2075 assertEquals("p.Lys1Asn", sf.getDescription());
2076 assertEquals("var4", sf.getValue("ID"));
2077 assertEquals("Benign", sf.getValue("clinical_significance"));
2078 assertEquals("ID=var4;clinical_significance=Benign", sf.getAttributes());
2079 assertEquals(1, sf.links.size());
2081 "p.Lys1Asn var4|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var4",
2083 assertEquals(ensembl, sf.getFeatureGroup());
2085 // var5 generates two distinct protein variant features
2087 assertEquals(3, sf.getBegin());
2088 assertEquals(3, sf.getEnd());
2089 assertEquals("p.Pro3His", sf.getDescription());
2090 assertEquals("var6", sf.getValue("ID"));
2091 assertEquals("Good", sf.getValue("clinical_significance"));
2092 assertEquals("ID=var6;clinical_significance=Good", sf.getAttributes());
2093 assertEquals(1, sf.links.size());
2095 "p.Pro3His var6|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var6",
2097 assertEquals(cosmic, sf.getFeatureGroup());
2100 assertEquals(3, sf.getBegin());
2101 assertEquals(3, sf.getEnd());
2102 assertEquals("p.Pro3Arg", 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.Pro3Arg var6|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var6",
2110 assertEquals(cosmic, sf.getFeatureGroup());
2114 * Tests for the method that maps the subset of a dna sequence that has CDS
2115 * (or subtype) feature, with CDS strand = '-' (reverse)
2117 // test turned off as currently findCdsPositions is not strand-dependent
2118 // left in case it comes around again...
2119 @Test(groups = "Functional", enabled = false)
2120 public void testFindCdsPositions_reverseStrand()
2122 SequenceI dnaSeq = new Sequence("dna", "aaaGGGcccAAATTTttt");
2123 dnaSeq.createDatasetSequence();
2124 SequenceI ds = dnaSeq.getDatasetSequence();
2127 SequenceFeature sf = new SequenceFeature("CDS", "", 4, 6, 0f, null);
2129 ds.addSequenceFeature(sf);
2130 // exon feature should be ignored here
2131 sf = new SequenceFeature("exon", "", 7, 9, 0f, null);
2132 ds.addSequenceFeature(sf);
2133 // CDS for dna 10-12
2134 sf = new SequenceFeature("CDS_predicted", "", 10, 12, 0f, null);
2136 ds.addSequenceFeature(sf);
2138 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
2140 * verify ranges { [12-10], [6-4] }
2142 assertEquals(6, MappingUtils.getLength(ranges));
2143 assertEquals(2, ranges.size());
2144 assertEquals(12, ranges.get(0)[0]);
2145 assertEquals(10, ranges.get(0)[1]);
2146 assertEquals(6, ranges.get(1)[0]);
2147 assertEquals(4, ranges.get(1)[1]);
2151 * Tests for the method that maps the subset of a dna sequence that has CDS
2152 * (or subtype) feature - reverse strand case where the start codon is
2155 @Test(groups = "Functional", enabled = false)
2156 // test turned off as currently findCdsPositions is not strand-dependent
2157 // left in case it comes around again...
2158 public void testFindCdsPositions_reverseStrandThreePrimeIncomplete()
2160 SequenceI dnaSeq = new Sequence("dna", "aaagGGCCCaaaTTTttt");
2161 dnaSeq.createDatasetSequence();
2162 SequenceI ds = dnaSeq.getDatasetSequence();
2165 SequenceFeature sf = new SequenceFeature("CDS", "", 5, 9, 0f, null);
2167 ds.addSequenceFeature(sf);
2168 // CDS for dna 13-15
2169 sf = new SequenceFeature("CDS_predicted", "", 13, 15, 0f, null);
2171 sf.setPhase("2"); // skip 2 bases to start of next codon
2172 ds.addSequenceFeature(sf);
2174 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
2177 * check the mapping starts with the first complete codon
2178 * expect ranges [13, 13], [9, 5]
2180 assertEquals(6, MappingUtils.getLength(ranges));
2181 assertEquals(2, ranges.size());
2182 assertEquals(13, ranges.get(0)[0]);
2183 assertEquals(13, ranges.get(0)[1]);
2184 assertEquals(9, ranges.get(1)[0]);
2185 assertEquals(5, ranges.get(1)[1]);
2188 @Test(groups = "Functional")
2189 public void testAlignAs_alternateTranscriptsUngapped()
2191 SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
2192 SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
2193 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
2194 ((Alignment) dna).createDatasetAlignment();
2195 SequenceI cds1 = new Sequence("cds1", "GGGTTT");
2196 SequenceI cds2 = new Sequence("cds2", "CCCAAA");
2197 AlignmentI cds = new Alignment(new SequenceI[] { cds1, cds2 });
2198 ((Alignment) cds).createDatasetAlignment();
2200 AlignedCodonFrame acf = new AlignedCodonFrame();
2201 MapList map = new MapList(new int[] { 4, 9 }, new int[] { 1, 6 }, 1, 1);
2202 acf.addMap(dna1.getDatasetSequence(), cds1.getDatasetSequence(), map);
2203 map = new MapList(new int[] { 1, 3, 10, 12 }, new int[] { 1, 6 }, 1, 1);
2204 acf.addMap(dna2.getDatasetSequence(), cds2.getDatasetSequence(), map);
2207 * verify CDS alignment is as:
2208 * cccGGGTTTaaa (cdna)
2209 * CCCgggtttAAA (cdna)
2211 * ---GGGTTT--- (cds)
2212 * CCC------AAA (cds)
2214 dna.addCodonFrame(acf);
2215 AlignmentUtils.alignAs(cds, dna);
2216 assertEquals("---GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
2217 assertEquals("CCC------AAA", cds.getSequenceAt(1).getSequenceAsString());
2220 @Test(groups = { "Functional" })
2221 public void testAddMappedPositions()
2223 SequenceI from = new Sequence("dna", "ggAA-ATcc-TT-g");
2224 SequenceI seq1 = new Sequence("cds", "AAATTT");
2225 from.createDatasetSequence();
2226 seq1.createDatasetSequence();
2227 Mapping mapping = new Mapping(seq1, new MapList(
2228 new int[] { 3, 6, 9, 10 },
2229 new int[] { 1, 6 }, 1, 1));
2230 Map<Integer, Map<SequenceI, Character>> map = new TreeMap<Integer, Map<SequenceI, Character>>();
2231 AlignmentUtils.addMappedPositions(seq1, from, mapping, map);
2234 * verify map has seq1 residues in columns 3,4,6,7,11,12
2236 assertEquals(6, map.size());
2237 assertEquals('A', map.get(3).get(seq1).charValue());
2238 assertEquals('A', map.get(4).get(seq1).charValue());
2239 assertEquals('A', map.get(6).get(seq1).charValue());
2240 assertEquals('T', map.get(7).get(seq1).charValue());
2241 assertEquals('T', map.get(11).get(seq1).charValue());
2242 assertEquals('T', map.get(12).get(seq1).charValue());
2250 * Test case where the mapping 'from' range includes a stop codon which is
2251 * absent in the 'to' range
2253 @Test(groups = { "Functional" })
2254 public void testAddMappedPositions_withStopCodon()
2256 SequenceI from = new Sequence("dna", "ggAA-ATcc-TT-g");
2257 SequenceI seq1 = new Sequence("cds", "AAATTT");
2258 from.createDatasetSequence();
2259 seq1.createDatasetSequence();
2260 Mapping mapping = new Mapping(seq1, new MapList(
2261 new int[] { 3, 6, 9, 10 },
2262 new int[] { 1, 6 }, 1, 1));
2263 Map<Integer, Map<SequenceI, Character>> map = new TreeMap<Integer, Map<SequenceI, Character>>();
2264 AlignmentUtils.addMappedPositions(seq1, from, mapping, map);
2267 * verify map has seq1 residues in columns 3,4,6,7,11,12
2269 assertEquals(6, map.size());
2270 assertEquals('A', map.get(3).get(seq1).charValue());
2271 assertEquals('A', map.get(4).get(seq1).charValue());
2272 assertEquals('A', map.get(6).get(seq1).charValue());
2273 assertEquals('T', map.get(7).get(seq1).charValue());
2274 assertEquals('T', map.get(11).get(seq1).charValue());
2275 assertEquals('T', map.get(12).get(seq1).charValue());
2279 * Test for the case where the products for which we want CDS are specified.
2280 * This is to represent the case where EMBL has CDS mappings to both Uniprot
2281 * and EMBLCDSPROTEIN. makeCdsAlignment() should only return the mappings for
2282 * the protein sequences specified.
2284 @Test(groups = { "Functional" })
2285 public void testMakeCdsAlignment_filterProducts()
2287 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
2288 SequenceI dna2 = new Sequence("dna2", "GGGcccTTTaaaCCC");
2289 SequenceI pep1 = new Sequence("Uniprot|pep1", "GF");
2290 SequenceI pep2 = new Sequence("Uniprot|pep2", "GFP");
2291 SequenceI pep3 = new Sequence("EMBL|pep3", "GF");
2292 SequenceI pep4 = new Sequence("EMBL|pep4", "GFP");
2293 dna1.createDatasetSequence();
2294 dna2.createDatasetSequence();
2295 pep1.createDatasetSequence();
2296 pep2.createDatasetSequence();
2297 pep3.createDatasetSequence();
2298 pep4.createDatasetSequence();
2299 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
2300 dna.setDataset(null);
2301 AlignmentI emblPeptides = new Alignment(new SequenceI[] { pep3, pep4 });
2302 emblPeptides.setDataset(null);
2304 AlignedCodonFrame acf = new AlignedCodonFrame();
2305 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
2306 new int[] { 1, 2 }, 3, 1);
2307 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
2308 acf.addMap(dna1.getDatasetSequence(), pep3.getDatasetSequence(), map);
2309 dna.addCodonFrame(acf);
2311 acf = new AlignedCodonFrame();
2312 map = new MapList(new int[] { 1, 3, 7, 9, 13, 15 }, new int[] { 1, 3 },
2314 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map);
2315 acf.addMap(dna2.getDatasetSequence(), pep4.getDatasetSequence(), map);
2316 dna.addCodonFrame(acf);
2319 * execute method under test to find CDS for EMBL peptides only
2321 AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
2322 dna1, dna2 }, dna.getDataset(), emblPeptides.getSequencesArray());
2324 assertEquals(2, cds.getSequences().size());
2325 assertEquals("GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
2326 assertEquals("GGGTTTCCC", cds.getSequenceAt(1).getSequenceAsString());
2329 * verify shared, extended alignment dataset
2331 assertSame(dna.getDataset(), cds.getDataset());
2332 assertTrue(dna.getDataset().getSequences()
2333 .contains(cds.getSequenceAt(0).getDatasetSequence()));
2334 assertTrue(dna.getDataset().getSequences()
2335 .contains(cds.getSequenceAt(1).getDatasetSequence()));
2338 * Verify mappings from CDS to peptide, cDNA to CDS, and cDNA to peptide
2339 * the mappings are on the shared alignment dataset
2341 List<AlignedCodonFrame> cdsMappings = cds.getDataset().getCodonFrames();
2343 * 6 mappings, 2*(DNA->CDS), 2*(DNA->Pep), 2*(CDS->Pep)
2345 assertEquals(6, cdsMappings.size());
2348 * verify that mapping sets for dna and cds alignments are different
2349 * [not current behaviour - all mappings are on the alignment dataset]
2351 // select -> subselect type to test.
2352 // Assert.assertNotSame(dna.getCodonFrames(), cds.getCodonFrames());
2353 // assertEquals(4, dna.getCodonFrames().size());
2354 // assertEquals(4, cds.getCodonFrames().size());
2357 * Two mappings involve pep3 (dna to pep3, cds to pep3)
2358 * Mapping from pep3 to GGGTTT in first new exon sequence
2360 List<AlignedCodonFrame> pep3Mappings = MappingUtils
2361 .findMappingsForSequence(pep3, cdsMappings);
2362 assertEquals(2, pep3Mappings.size());
2363 List<AlignedCodonFrame> mappings = MappingUtils
2364 .findMappingsForSequence(cds.getSequenceAt(0), pep3Mappings);
2365 assertEquals(1, mappings.size());
2368 SearchResults sr = MappingUtils.buildSearchResults(pep3, 1, mappings);
2369 assertEquals(1, sr.getResults().size());
2370 Match m = sr.getResults().get(0);
2371 assertSame(cds.getSequenceAt(0).getDatasetSequence(), m.getSequence());
2372 assertEquals(1, m.getStart());
2373 assertEquals(3, m.getEnd());
2375 sr = MappingUtils.buildSearchResults(pep3, 2, mappings);
2376 m = sr.getResults().get(0);
2377 assertSame(cds.getSequenceAt(0).getDatasetSequence(), m.getSequence());
2378 assertEquals(4, m.getStart());
2379 assertEquals(6, m.getEnd());
2382 * Two mappings involve pep4 (dna to pep4, cds to pep4)
2383 * Verify mapping from pep4 to GGGTTTCCC in second new exon sequence
2385 List<AlignedCodonFrame> pep4Mappings = MappingUtils
2386 .findMappingsForSequence(pep4, cdsMappings);
2387 assertEquals(2, pep4Mappings.size());
2388 mappings = MappingUtils.findMappingsForSequence(cds.getSequenceAt(1),
2390 assertEquals(1, mappings.size());
2392 sr = MappingUtils.buildSearchResults(pep4, 1, mappings);
2393 assertEquals(1, sr.getResults().size());
2394 m = sr.getResults().get(0);
2395 assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
2396 assertEquals(1, m.getStart());
2397 assertEquals(3, m.getEnd());
2399 sr = MappingUtils.buildSearchResults(pep4, 2, mappings);
2400 m = sr.getResults().get(0);
2401 assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
2402 assertEquals(4, m.getStart());
2403 assertEquals(6, m.getEnd());
2405 sr = MappingUtils.buildSearchResults(pep4, 3, mappings);
2406 m = sr.getResults().get(0);
2407 assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
2408 assertEquals(7, m.getStart());
2409 assertEquals(9, m.getEnd());
2413 * Test the method that just copies aligned sequences, provided all sequences
2414 * to be aligned share the aligned sequence's dataset
2416 @Test(groups = "Functional")
2417 public void testAlignAsSameSequences()
2419 SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
2420 SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
2421 AlignmentI al1 = new Alignment(new SequenceI[] { dna1, dna2 });
2422 ((Alignment) al1).createDatasetAlignment();
2424 SequenceI dna3 = new Sequence(dna1);
2425 SequenceI dna4 = new Sequence(dna2);
2426 assertSame(dna3.getDatasetSequence(), dna1.getDatasetSequence());
2427 assertSame(dna4.getDatasetSequence(), dna2.getDatasetSequence());
2428 String seq1 = "-cc-GG-GT-TT--aaa";
2429 dna3.setSequence(seq1);
2430 String seq2 = "C--C-Cgg--gtt-tAA-A-";
2431 dna4.setSequence(seq2);
2432 AlignmentI al2 = new Alignment(new SequenceI[] { dna3, dna4 });
2433 ((Alignment) al2).createDatasetAlignment();
2435 assertTrue(AlignmentUtils.alignAsSameSequences(al1, al2));
2436 assertEquals(seq1, al1.getSequenceAt(0).getSequenceAsString());
2437 assertEquals(seq2, al1.getSequenceAt(1).getSequenceAsString());
2440 * add another sequence to 'aligned' - should still succeed, since
2441 * unaligned sequences still share a dataset with aligned sequences
2443 SequenceI dna5 = new Sequence("dna5", "CCCgggtttAAA");
2444 dna5.createDatasetSequence();
2445 al2.addSequence(dna5);
2446 assertTrue(AlignmentUtils.alignAsSameSequences(al1, al2));
2447 assertEquals(seq1, al1.getSequenceAt(0).getSequenceAsString());
2448 assertEquals(seq2, al1.getSequenceAt(1).getSequenceAsString());
2451 * add another sequence to 'unaligned' - should fail, since now not
2452 * all unaligned sequences share a dataset with aligned sequences
2454 SequenceI dna6 = new Sequence("dna6", "CCCgggtttAAA");
2455 dna6.createDatasetSequence();
2456 al1.addSequence(dna6);
2457 // JAL-2110 JBP Comment: what's the use case for this behaviour ?
2458 assertFalse(AlignmentUtils.alignAsSameSequences(al1, al2));
2461 @Test(groups = "Functional")
2462 public void testAlignAsSameSequencesMultipleSubSeq()
2464 SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
2465 SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
2466 SequenceI as1 = dna1.deriveSequence();
2467 SequenceI as2 = dna1.deriveSequence().getSubSequence(3, 7);
2468 SequenceI as3 = dna2.deriveSequence();
2469 as1.insertCharAt(6, 5, '-');
2470 String s_as1 = as1.getSequenceAsString();
2471 as2.insertCharAt(6, 5, '-');
2472 String s_as2 = as2.getSequenceAsString();
2473 as3.insertCharAt(6, 5, '-');
2474 String s_as3 = as3.getSequenceAsString();
2475 AlignmentI aligned = new Alignment(new SequenceI[] { as1, as2, as3 });
2477 // why do we need to cast this still ?
2478 ((Alignment) aligned).createDatasetAlignment();
2479 SequenceI uas1 = dna1.deriveSequence();
2480 SequenceI uas2 = dna1.deriveSequence().getSubSequence(3, 7);
2481 SequenceI uas3 = dna2.deriveSequence();
2482 AlignmentI tobealigned = new Alignment(new SequenceI[] { uas1, uas2,
2484 ((Alignment) tobealigned).createDatasetAlignment();
2486 assertTrue(AlignmentUtils.alignAsSameSequences(tobealigned, aligned));
2487 assertEquals(s_as1, uas1.getSequenceAsString());
2488 assertEquals(s_as2, uas2.getSequenceAsString());
2489 assertEquals(s_as3, uas3.getSequenceAsString());