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] --> pep2
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 * put a variant feature on dna2 base 8
997 * - should transfer to cds2 base 5
999 dna2.addSequenceFeature(new SequenceFeature("variant", "hgmd", 8, 8,
1003 * need a sourceDbRef if we are to construct dbrefs to the CDS
1004 * sequence from the dna contig sequences
1006 DBRefEntry dbref = new DBRefEntry("ENSEMBL", "0", "dna1");
1007 dna1.getDatasetSequence().addDBRef(dbref);
1008 org.testng.Assert.assertEquals(dbref, dna1.getPrimaryDBRefs().get(0));
1009 dbref = new DBRefEntry("ENSEMBL", "0", "dna2");
1010 dna2.getDatasetSequence().addDBRef(dbref);
1011 org.testng.Assert.assertEquals(dbref, dna2.getPrimaryDBRefs().get(0));
1014 * CDS sequences are 'discovered' from dna-to-protein mappings on the alignment
1015 * dataset (e.g. added from dbrefs by CrossRef.findXrefSequences)
1017 MapList mapfordna1 = new MapList(new int[] { 4, 6, 10, 12 },
1018 new int[] { 1, 2 }, 3, 1);
1019 AlignedCodonFrame acf = new AlignedCodonFrame();
1020 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(),
1022 dna.addCodonFrame(acf);
1023 MapList mapfordna2 = new MapList(new int[] { 1, 3, 7, 9, 13, 15 },
1026 acf = new AlignedCodonFrame();
1027 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(),
1029 dna.addCodonFrame(acf);
1032 * In this case, mappings originally came from matching Uniprot accessions - so need an xref on dna involving those regions. These are normally constructed from CDS annotation
1034 DBRefEntry dna1xref = new DBRefEntry("UNIPROT", "ENSEMBL", "pep1",
1035 new Mapping(mapfordna1));
1036 dna1.getDatasetSequence().addDBRef(dna1xref);
1037 DBRefEntry dna2xref = new DBRefEntry("UNIPROT", "ENSEMBL", "pep2",
1038 new Mapping(mapfordna2));
1039 dna2.getDatasetSequence().addDBRef(dna2xref);
1042 * execute method under test:
1044 AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
1045 dna1, dna2 }, dna.getDataset(), null);
1048 * verify cds sequences
1050 assertEquals(2, cds.getSequences().size());
1051 assertEquals("GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
1052 assertEquals("GGGTTTCCC", cds.getSequenceAt(1).getSequenceAsString());
1055 * verify shared, extended alignment dataset
1057 assertSame(dna.getDataset(), cds.getDataset());
1058 SequenceI cds1Dss = cds.getSequenceAt(0).getDatasetSequence();
1059 SequenceI cds2Dss = cds.getSequenceAt(1).getDatasetSequence();
1060 assertTrue(dna.getDataset().getSequences().contains(cds1Dss));
1061 assertTrue(dna.getDataset().getSequences().contains(cds2Dss));
1064 * verify CDS has a dbref with mapping to peptide
1066 assertNotNull(cds1Dss.getDBRefs());
1067 assertEquals(2, cds1Dss.getDBRefs().length);
1068 dbref = cds1Dss.getDBRefs()[0];
1069 assertEquals(dna1xref.getSource(), dbref.getSource());
1070 // version is via ensembl's primary ref
1071 assertEquals(dna1xref.getVersion(), dbref.getVersion());
1072 assertEquals(dna1xref.getAccessionId(), dbref.getAccessionId());
1073 assertNotNull(dbref.getMap());
1074 assertSame(pep1.getDatasetSequence(), dbref.getMap().getTo());
1075 MapList cdsMapping = new MapList(new int[] { 1, 6 },
1076 new int[] { 1, 2 }, 3, 1);
1077 assertEquals(cdsMapping, dbref.getMap().getMap());
1080 * verify peptide has added a dbref with reverse mapping to CDS
1082 assertNotNull(pep1.getDBRefs());
1083 // FIXME pep1.getDBRefs() is 1 - is that the correct behaviour ?
1084 assertEquals(2, pep1.getDBRefs().length);
1085 dbref = pep1.getDBRefs()[1];
1086 assertEquals("ENSEMBL", dbref.getSource());
1087 assertEquals("0", dbref.getVersion());
1088 assertEquals("CDS|dna1", dbref.getAccessionId());
1089 assertNotNull(dbref.getMap());
1090 assertSame(cds1Dss, dbref.getMap().getTo());
1091 assertEquals(cdsMapping.getInverse(), dbref.getMap().getMap());
1094 * Verify mappings from CDS to peptide, cDNA to CDS, and cDNA to peptide
1095 * the mappings are on the shared alignment dataset
1096 * 6 mappings, 2*(DNA->CDS), 2*(DNA->Pep), 2*(CDS->Pep)
1098 List<AlignedCodonFrame> cdsMappings = cds.getDataset().getCodonFrames();
1099 assertEquals(6, cdsMappings.size());
1102 * verify that mapping sets for dna and cds alignments are different
1103 * [not current behaviour - all mappings are on the alignment dataset]
1105 // select -> subselect type to test.
1106 // Assert.assertNotSame(dna.getCodonFrames(), cds.getCodonFrames());
1107 // assertEquals(4, dna.getCodonFrames().size());
1108 // assertEquals(4, cds.getCodonFrames().size());
1111 * Two mappings involve pep1 (dna to pep1, cds to pep1)
1112 * Mapping from pep1 to GGGTTT in first new exon sequence
1114 List<AlignedCodonFrame> pep1Mappings = MappingUtils
1115 .findMappingsForSequence(pep1, cdsMappings);
1116 assertEquals(2, pep1Mappings.size());
1117 List<AlignedCodonFrame> mappings = MappingUtils
1118 .findMappingsForSequence(cds.getSequenceAt(0), pep1Mappings);
1119 assertEquals(1, mappings.size());
1122 SearchResults sr = MappingUtils.buildSearchResults(pep1, 1, mappings);
1123 assertEquals(1, sr.getResults().size());
1124 Match m = sr.getResults().get(0);
1125 assertSame(cds1Dss, m.getSequence());
1126 assertEquals(1, m.getStart());
1127 assertEquals(3, m.getEnd());
1129 sr = MappingUtils.buildSearchResults(pep1, 2, mappings);
1130 m = sr.getResults().get(0);
1131 assertSame(cds1Dss, m.getSequence());
1132 assertEquals(4, m.getStart());
1133 assertEquals(6, m.getEnd());
1136 * Two mappings involve pep2 (dna to pep2, cds to pep2)
1137 * Verify mapping from pep2 to GGGTTTCCC in second new exon sequence
1139 List<AlignedCodonFrame> pep2Mappings = MappingUtils
1140 .findMappingsForSequence(pep2, cdsMappings);
1141 assertEquals(2, pep2Mappings.size());
1142 mappings = MappingUtils.findMappingsForSequence(cds.getSequenceAt(1),
1144 assertEquals(1, mappings.size());
1146 sr = MappingUtils.buildSearchResults(pep2, 1, mappings);
1147 assertEquals(1, sr.getResults().size());
1148 m = sr.getResults().get(0);
1149 assertSame(cds2Dss, m.getSequence());
1150 assertEquals(1, m.getStart());
1151 assertEquals(3, m.getEnd());
1153 sr = MappingUtils.buildSearchResults(pep2, 2, mappings);
1154 m = sr.getResults().get(0);
1155 assertSame(cds2Dss, m.getSequence());
1156 assertEquals(4, m.getStart());
1157 assertEquals(6, m.getEnd());
1159 sr = MappingUtils.buildSearchResults(pep2, 3, mappings);
1160 m = sr.getResults().get(0);
1161 assertSame(cds2Dss, m.getSequence());
1162 assertEquals(7, m.getStart());
1163 assertEquals(9, m.getEnd());
1166 * check cds2 acquired a variant feature in position 5
1168 SequenceFeature[] sfs = cds2Dss.getSequenceFeatures();
1170 assertEquals(1, sfs.length);
1171 assertEquals("variant", sfs[0].type);
1172 assertEquals(5, sfs[0].begin);
1173 assertEquals(5, sfs[0].end);
1177 * Test the method that makes a cds-only alignment from a DNA sequence and its
1178 * product mappings, for the case where there are multiple exon mappings to
1179 * different protein products.
1181 @Test(groups = { "Functional" })
1182 public void testMakeCdsAlignment_multipleProteins()
1184 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
1185 SequenceI pep1 = new Sequence("pep1", "GF"); // GGGTTT
1186 SequenceI pep2 = new Sequence("pep2", "KP"); // aaaccc
1187 SequenceI pep3 = new Sequence("pep3", "KF"); // aaaTTT
1188 dna1.createDatasetSequence();
1189 pep1.createDatasetSequence();
1190 pep2.createDatasetSequence();
1191 pep3.createDatasetSequence();
1192 pep1.getDatasetSequence().addDBRef(
1193 new DBRefEntry("EMBLCDS", "2", "A12345"));
1194 pep2.getDatasetSequence().addDBRef(
1195 new DBRefEntry("EMBLCDS", "3", "A12346"));
1196 pep3.getDatasetSequence().addDBRef(
1197 new DBRefEntry("EMBLCDS", "4", "A12347"));
1200 * Create the CDS alignment
1202 AlignmentI dna = new Alignment(new SequenceI[] { dna1 });
1203 dna.setDataset(null);
1206 * Make the mappings from dna to protein
1208 // map ...GGG...TTT to GF
1209 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1210 new int[] { 1, 2 }, 3, 1);
1211 AlignedCodonFrame acf = new AlignedCodonFrame();
1212 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
1213 dna.addCodonFrame(acf);
1215 // map aaa...ccc to KP
1216 map = new MapList(new int[] { 1, 3, 7, 9 }, new int[] { 1, 2 }, 3, 1);
1217 acf = new AlignedCodonFrame();
1218 acf.addMap(dna1.getDatasetSequence(), pep2.getDatasetSequence(), map);
1219 dna.addCodonFrame(acf);
1221 // map aaa......TTT to KF
1222 map = new MapList(new int[] { 1, 3, 10, 12 }, new int[] { 1, 2 }, 3, 1);
1223 acf = new AlignedCodonFrame();
1224 acf.addMap(dna1.getDatasetSequence(), pep3.getDatasetSequence(), map);
1225 dna.addCodonFrame(acf);
1228 * execute method under test
1230 AlignmentI cdsal = AlignmentUtils.makeCdsAlignment(
1231 new SequenceI[] { dna1 }, dna.getDataset(), null);
1234 * Verify we have 3 cds sequences, mapped to pep1/2/3 respectively
1236 List<SequenceI> cds = cdsal.getSequences();
1237 assertEquals(3, cds.size());
1240 * verify shared, extended alignment dataset
1242 assertSame(cdsal.getDataset(), dna.getDataset());
1243 assertTrue(dna.getDataset().getSequences()
1244 .contains(cds.get(0).getDatasetSequence()));
1245 assertTrue(dna.getDataset().getSequences()
1246 .contains(cds.get(1).getDatasetSequence()));
1247 assertTrue(dna.getDataset().getSequences()
1248 .contains(cds.get(2).getDatasetSequence()));
1251 * verify aligned cds sequences and their xrefs
1253 SequenceI cdsSeq = cds.get(0);
1254 assertEquals("GGGTTT", cdsSeq.getSequenceAsString());
1255 // assertEquals("dna1|A12345", cdsSeq.getName());
1256 assertEquals("CDS|dna1", cdsSeq.getName());
1257 // assertEquals(1, cdsSeq.getDBRefs().length);
1258 // DBRefEntry cdsRef = cdsSeq.getDBRefs()[0];
1259 // assertEquals("EMBLCDS", cdsRef.getSource());
1260 // assertEquals("2", cdsRef.getVersion());
1261 // assertEquals("A12345", cdsRef.getAccessionId());
1263 cdsSeq = cds.get(1);
1264 assertEquals("aaaccc", cdsSeq.getSequenceAsString());
1265 // assertEquals("dna1|A12346", cdsSeq.getName());
1266 assertEquals("CDS|dna1", cdsSeq.getName());
1267 // assertEquals(1, cdsSeq.getDBRefs().length);
1268 // cdsRef = cdsSeq.getDBRefs()[0];
1269 // assertEquals("EMBLCDS", cdsRef.getSource());
1270 // assertEquals("3", cdsRef.getVersion());
1271 // assertEquals("A12346", cdsRef.getAccessionId());
1273 cdsSeq = cds.get(2);
1274 assertEquals("aaaTTT", cdsSeq.getSequenceAsString());
1275 // assertEquals("dna1|A12347", cdsSeq.getName());
1276 assertEquals("CDS|dna1", cdsSeq.getName());
1277 // assertEquals(1, cdsSeq.getDBRefs().length);
1278 // cdsRef = cdsSeq.getDBRefs()[0];
1279 // assertEquals("EMBLCDS", cdsRef.getSource());
1280 // assertEquals("4", cdsRef.getVersion());
1281 // assertEquals("A12347", cdsRef.getAccessionId());
1284 * Verify there are mappings from each cds sequence to its protein product
1285 * and also to its dna source
1287 List<AlignedCodonFrame> newMappings = cdsal.getCodonFrames();
1290 * 6 mappings involve dna1 (to pep1/2/3, cds1/2/3)
1292 List<AlignedCodonFrame> dnaMappings = MappingUtils
1293 .findMappingsForSequence(dna1, newMappings);
1294 assertEquals(6, dnaMappings.size());
1299 List<AlignedCodonFrame> mappings = MappingUtils
1300 .findMappingsForSequence(pep1, dnaMappings);
1301 assertEquals(1, mappings.size());
1302 assertEquals(1, mappings.get(0).getMappings().size());
1303 assertSame(pep1.getDatasetSequence(), mappings.get(0).getMappings()
1304 .get(0).getMapping().getTo());
1309 List<AlignedCodonFrame> dnaToCds1Mappings = MappingUtils
1310 .findMappingsForSequence(cds.get(0), dnaMappings);
1311 Mapping mapping = dnaToCds1Mappings.get(0).getMappings().get(0)
1313 assertSame(cds.get(0).getDatasetSequence(), mapping
1315 assertEquals("G(1) in CDS should map to G(4) in DNA", 4, mapping
1316 .getMap().getToPosition(1));
1321 mappings = MappingUtils.findMappingsForSequence(pep2, dnaMappings);
1322 assertEquals(1, mappings.size());
1323 assertEquals(1, mappings.get(0).getMappings().size());
1324 assertSame(pep2.getDatasetSequence(), mappings.get(0).getMappings()
1325 .get(0).getMapping().getTo());
1330 List<AlignedCodonFrame> dnaToCds2Mappings = MappingUtils
1331 .findMappingsForSequence(cds.get(1), dnaMappings);
1332 mapping = dnaToCds2Mappings.get(0).getMappings().get(0).getMapping();
1333 assertSame(cds.get(1).getDatasetSequence(), mapping.getTo());
1334 assertEquals("c(4) in CDS should map to c(7) in DNA", 7, mapping
1335 .getMap().getToPosition(4));
1340 mappings = MappingUtils.findMappingsForSequence(pep3, dnaMappings);
1341 assertEquals(1, mappings.size());
1342 assertEquals(1, mappings.get(0).getMappings().size());
1343 assertSame(pep3.getDatasetSequence(), mappings.get(0).getMappings()
1344 .get(0).getMapping().getTo());
1349 List<AlignedCodonFrame> dnaToCds3Mappings = MappingUtils
1350 .findMappingsForSequence(cds.get(2), dnaMappings);
1351 mapping = dnaToCds3Mappings.get(0).getMappings().get(0).getMapping();
1352 assertSame(cds.get(2).getDatasetSequence(), mapping.getTo());
1353 assertEquals("T(4) in CDS should map to T(10) in DNA", 10, mapping
1354 .getMap().getToPosition(4));
1357 @Test(groups = { "Functional" })
1358 public void testIsMappable()
1360 SequenceI dna1 = new Sequence("dna1", "cgCAGtgGT");
1361 SequenceI aa1 = new Sequence("aa1", "RSG");
1362 AlignmentI al1 = new Alignment(new SequenceI[] { dna1 });
1363 AlignmentI al2 = new Alignment(new SequenceI[] { aa1 });
1365 assertFalse(AlignmentUtils.isMappable(null, null));
1366 assertFalse(AlignmentUtils.isMappable(al1, null));
1367 assertFalse(AlignmentUtils.isMappable(null, al1));
1368 assertFalse(AlignmentUtils.isMappable(al1, al1));
1369 assertFalse(AlignmentUtils.isMappable(al2, al2));
1371 assertTrue(AlignmentUtils.isMappable(al1, al2));
1372 assertTrue(AlignmentUtils.isMappable(al2, al1));
1376 * Test creating a mapping when the sequences involved do not start at residue
1379 * @throws IOException
1381 @Test(groups = { "Functional" })
1382 public void testMapCdnaToProtein_forSubsequence()
1385 SequenceI prot = new Sequence("UNIPROT|V12345", "E-I--Q", 10, 12);
1386 prot.createDatasetSequence();
1388 SequenceI dna = new Sequence("EMBL|A33333", "GAA--AT-C-CAG", 40, 48);
1389 dna.createDatasetSequence();
1391 MapList map = AlignmentUtils.mapCdnaToProtein(prot, dna);
1392 assertEquals(10, map.getToLowest());
1393 assertEquals(12, map.getToHighest());
1394 assertEquals(40, map.getFromLowest());
1395 assertEquals(48, map.getFromHighest());
1399 * Test for the alignSequenceAs method where we have protein mapped to protein
1401 @Test(groups = { "Functional" })
1402 public void testAlignSequenceAs_mappedProteinProtein()
1405 SequenceI alignMe = new Sequence("Match", "MGAASEV");
1406 alignMe.createDatasetSequence();
1407 SequenceI alignFrom = new Sequence("Query", "LQTGYMGAASEVMFSPTRR");
1408 alignFrom.createDatasetSequence();
1410 AlignedCodonFrame acf = new AlignedCodonFrame();
1411 // this is like a domain or motif match of part of a peptide sequence
1412 MapList map = new MapList(new int[] { 6, 12 }, new int[] { 1, 7 }, 1, 1);
1413 acf.addMap(alignFrom.getDatasetSequence(),
1414 alignMe.getDatasetSequence(), map);
1416 AlignmentUtils.alignSequenceAs(alignMe, alignFrom, acf, "-", '-', true,
1418 assertEquals("-----MGAASEV-------", alignMe.getSequenceAsString());
1422 * Test for the alignSequenceAs method where there are trailing unmapped
1423 * residues in the model sequence
1425 @Test(groups = { "Functional" })
1426 public void testAlignSequenceAs_withTrailingPeptide()
1428 // map first 3 codons to KPF; G is a trailing unmapped residue
1429 MapList map = new MapList(new int[] { 1, 9 }, new int[] { 1, 3 }, 3, 1);
1431 checkAlignSequenceAs("AAACCCTTT", "K-PFG", true, true, map,
1436 * Tests for transferring features between mapped sequences
1438 @Test(groups = { "Functional" })
1439 public void testTransferFeatures()
1441 SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
1442 SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
1445 dna.addSequenceFeature(new SequenceFeature("type1", "desc1", 1, 2, 1f,
1447 // partial overlap - to [1, 1]
1448 dna.addSequenceFeature(new SequenceFeature("type2", "desc2", 3, 4, 2f,
1450 // exact overlap - to [1, 3]
1451 dna.addSequenceFeature(new SequenceFeature("type3", "desc3", 4, 6, 3f,
1453 // spanning overlap - to [2, 5]
1454 dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
1456 // exactly overlaps whole mapped range [1, 6]
1457 dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
1459 // no overlap (internal)
1460 dna.addSequenceFeature(new SequenceFeature("type6", "desc6", 7, 9, 6f,
1462 // no overlap (3' end)
1463 dna.addSequenceFeature(new SequenceFeature("type7", "desc7", 13, 15,
1465 // overlap (3' end) - to [6, 6]
1466 dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
1468 // extended overlap - to [6, +]
1469 dna.addSequenceFeature(new SequenceFeature("type9", "desc9", 12, 13,
1472 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1473 new int[] { 1, 6 }, 1, 1);
1476 * transferFeatures() will build 'partial overlap' for regions
1477 * that partially overlap 5' or 3' (start or end) of target sequence
1479 AlignmentUtils.transferFeatures(dna, cds, map, null);
1480 SequenceFeature[] sfs = cds.getSequenceFeatures();
1481 assertEquals(6, sfs.length);
1483 SequenceFeature sf = sfs[0];
1484 assertEquals("type2", sf.getType());
1485 assertEquals("desc2", sf.getDescription());
1486 assertEquals(2f, sf.getScore());
1487 assertEquals(1, sf.getBegin());
1488 assertEquals(1, sf.getEnd());
1491 assertEquals("type3", sf.getType());
1492 assertEquals("desc3", sf.getDescription());
1493 assertEquals(3f, sf.getScore());
1494 assertEquals(1, sf.getBegin());
1495 assertEquals(3, sf.getEnd());
1498 assertEquals("type4", sf.getType());
1499 assertEquals(2, sf.getBegin());
1500 assertEquals(5, sf.getEnd());
1503 assertEquals("type5", sf.getType());
1504 assertEquals(1, sf.getBegin());
1505 assertEquals(6, sf.getEnd());
1508 assertEquals("type8", sf.getType());
1509 assertEquals(6, sf.getBegin());
1510 assertEquals(6, sf.getEnd());
1513 assertEquals("type9", sf.getType());
1514 assertEquals(6, sf.getBegin());
1515 assertEquals(6, sf.getEnd());
1519 * Tests for transferring features between mapped sequences
1521 @Test(groups = { "Functional" })
1522 public void testTransferFeatures_withOmit()
1524 SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
1525 SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
1527 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1528 new int[] { 1, 6 }, 1, 1);
1530 // [5, 11] maps to [2, 5]
1531 dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
1533 // [4, 12] maps to [1, 6]
1534 dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
1536 // [12, 12] maps to [6, 6]
1537 dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
1540 // desc4 and desc8 are the 'omit these' varargs
1541 AlignmentUtils.transferFeatures(dna, cds, map, null, "type4", "type8");
1542 SequenceFeature[] sfs = cds.getSequenceFeatures();
1543 assertEquals(1, sfs.length);
1545 SequenceFeature sf = sfs[0];
1546 assertEquals("type5", sf.getType());
1547 assertEquals(1, sf.getBegin());
1548 assertEquals(6, sf.getEnd());
1552 * Tests for transferring features between mapped sequences
1554 @Test(groups = { "Functional" })
1555 public void testTransferFeatures_withSelect()
1557 SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
1558 SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
1560 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1561 new int[] { 1, 6 }, 1, 1);
1563 // [5, 11] maps to [2, 5]
1564 dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
1566 // [4, 12] maps to [1, 6]
1567 dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
1569 // [12, 12] maps to [6, 6]
1570 dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
1573 // "type5" is the 'select this type' argument
1574 AlignmentUtils.transferFeatures(dna, cds, map, "type5");
1575 SequenceFeature[] sfs = cds.getSequenceFeatures();
1576 assertEquals(1, sfs.length);
1578 SequenceFeature sf = sfs[0];
1579 assertEquals("type5", sf.getType());
1580 assertEquals(1, sf.getBegin());
1581 assertEquals(6, sf.getEnd());
1585 * Test the method that extracts the cds-only part of a dna alignment, for the
1586 * case where the cds should be aligned to match its nucleotide sequence.
1588 @Test(groups = { "Functional" })
1589 public void testMakeCdsAlignment_alternativeTranscripts()
1591 SequenceI dna1 = new Sequence("dna1", "aaaGGGCC-----CTTTaaaGGG");
1592 // alternative transcript of same dna skips CCC codon
1593 SequenceI dna2 = new Sequence("dna2", "aaaGGGCC-----cttTaaaGGG");
1594 // dna3 has no mapping (protein product) so should be ignored here
1595 SequenceI dna3 = new Sequence("dna3", "aaaGGGCCCCCGGGcttTaaaGGG");
1596 SequenceI pep1 = new Sequence("pep1", "GPFG");
1597 SequenceI pep2 = new Sequence("pep2", "GPG");
1598 dna1.createDatasetSequence();
1599 dna2.createDatasetSequence();
1600 dna3.createDatasetSequence();
1601 pep1.createDatasetSequence();
1602 pep2.createDatasetSequence();
1604 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
1605 dna.setDataset(null);
1607 MapList map = new MapList(new int[] { 4, 12, 16, 18 },
1608 new int[] { 1, 4 }, 3, 1);
1609 AlignedCodonFrame acf = new AlignedCodonFrame();
1610 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
1611 dna.addCodonFrame(acf);
1612 map = new MapList(new int[] { 4, 8, 12, 12, 16, 18 },
1615 acf = new AlignedCodonFrame();
1616 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map);
1617 dna.addCodonFrame(acf);
1619 AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
1620 dna1, dna2, dna3 }, dna.getDataset(), null);
1621 List<SequenceI> cdsSeqs = cds.getSequences();
1622 assertEquals(2, cdsSeqs.size());
1623 assertEquals("GGGCCCTTTGGG", cdsSeqs.get(0).getSequenceAsString());
1624 assertEquals("GGGCCTGGG", cdsSeqs.get(1).getSequenceAsString());
1627 * verify shared, extended alignment dataset
1629 assertSame(dna.getDataset(), cds.getDataset());
1630 assertTrue(dna.getDataset().getSequences()
1631 .contains(cdsSeqs.get(0).getDatasetSequence()));
1632 assertTrue(dna.getDataset().getSequences()
1633 .contains(cdsSeqs.get(1).getDatasetSequence()));
1636 * Verify 6 mappings: dna1 to cds1, cds1 to pep1, dna1 to pep1
1637 * and the same for dna2/cds2/pep2
1639 List<AlignedCodonFrame> mappings = cds.getCodonFrames();
1640 assertEquals(6, mappings.size());
1643 * 2 mappings involve pep1
1645 List<AlignedCodonFrame> pep1Mappings = MappingUtils
1646 .findMappingsForSequence(pep1, mappings);
1647 assertEquals(2, pep1Mappings.size());
1650 * Get mapping of pep1 to cds1 and verify it
1651 * maps GPFG to 1-3,4-6,7-9,10-12
1653 List<AlignedCodonFrame> pep1CdsMappings = MappingUtils
1654 .findMappingsForSequence(cds.getSequenceAt(0), pep1Mappings);
1655 assertEquals(1, pep1CdsMappings.size());
1656 SearchResults sr = MappingUtils.buildSearchResults(pep1, 1,
1658 assertEquals(1, sr.getResults().size());
1659 Match m = sr.getResults().get(0);
1660 assertEquals(cds.getSequenceAt(0).getDatasetSequence(),
1662 assertEquals(1, m.getStart());
1663 assertEquals(3, m.getEnd());
1664 sr = MappingUtils.buildSearchResults(pep1, 2, pep1CdsMappings);
1665 m = sr.getResults().get(0);
1666 assertEquals(4, m.getStart());
1667 assertEquals(6, m.getEnd());
1668 sr = MappingUtils.buildSearchResults(pep1, 3, pep1CdsMappings);
1669 m = sr.getResults().get(0);
1670 assertEquals(7, m.getStart());
1671 assertEquals(9, m.getEnd());
1672 sr = MappingUtils.buildSearchResults(pep1, 4, pep1CdsMappings);
1673 m = sr.getResults().get(0);
1674 assertEquals(10, m.getStart());
1675 assertEquals(12, m.getEnd());
1678 * Get mapping of pep2 to cds2 and verify it
1679 * maps GPG in pep2 to 1-3,4-6,7-9 in second CDS sequence
1681 List<AlignedCodonFrame> pep2Mappings = MappingUtils
1682 .findMappingsForSequence(pep2, mappings);
1683 assertEquals(2, pep2Mappings.size());
1684 List<AlignedCodonFrame> pep2CdsMappings = MappingUtils
1685 .findMappingsForSequence(cds.getSequenceAt(1), pep2Mappings);
1686 assertEquals(1, pep2CdsMappings.size());
1687 sr = MappingUtils.buildSearchResults(pep2, 1, pep2CdsMappings);
1688 assertEquals(1, sr.getResults().size());
1689 m = sr.getResults().get(0);
1690 assertEquals(cds.getSequenceAt(1).getDatasetSequence(),
1692 assertEquals(1, m.getStart());
1693 assertEquals(3, m.getEnd());
1694 sr = MappingUtils.buildSearchResults(pep2, 2, pep2CdsMappings);
1695 m = sr.getResults().get(0);
1696 assertEquals(4, m.getStart());
1697 assertEquals(6, m.getEnd());
1698 sr = MappingUtils.buildSearchResults(pep2, 3, pep2CdsMappings);
1699 m = sr.getResults().get(0);
1700 assertEquals(7, m.getStart());
1701 assertEquals(9, m.getEnd());
1705 * Test the method that realigns protein to match mapped codon alignment.
1707 @Test(groups = { "Functional" })
1708 public void testAlignProteinAsDna_incompleteStartCodon()
1710 // seq1: incomplete start codon (not mapped), then [3, 11]
1711 SequenceI dna1 = new Sequence("Seq1", "ccAAA-TTT-GGG-");
1712 // seq2 codons are [4, 5], [8, 11]
1713 SequenceI dna2 = new Sequence("Seq2", "ccaAA-ttT-GGG-");
1714 // seq3 incomplete start codon at 'tt'
1715 SequenceI dna3 = new Sequence("Seq3", "ccaaa-ttt-GGG-");
1716 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
1717 dna.setDataset(null);
1719 // prot1 has 'X' for incomplete start codon (not mapped)
1720 SequenceI prot1 = new Sequence("Seq1", "XKFG"); // X for incomplete start
1721 SequenceI prot2 = new Sequence("Seq2", "NG");
1722 SequenceI prot3 = new Sequence("Seq3", "XG"); // X for incomplete start
1723 AlignmentI protein = new Alignment(new SequenceI[] { prot1, prot2,
1725 protein.setDataset(null);
1727 // map dna1 [3, 11] to prot1 [2, 4] KFG
1728 MapList map = new MapList(new int[] { 3, 11 }, new int[] { 2, 4 }, 3, 1);
1729 AlignedCodonFrame acf = new AlignedCodonFrame();
1730 acf.addMap(dna1.getDatasetSequence(), prot1.getDatasetSequence(), map);
1732 // map dna2 [4, 5] [8, 11] to prot2 [1, 2] NG
1733 map = new MapList(new int[] { 4, 5, 8, 11 }, new int[] { 1, 2 }, 3, 1);
1734 acf.addMap(dna2.getDatasetSequence(), prot2.getDatasetSequence(), map);
1736 // map dna3 [9, 11] to prot3 [2, 2] G
1737 map = new MapList(new int[] { 9, 11 }, new int[] { 2, 2 }, 3, 1);
1738 acf.addMap(dna3.getDatasetSequence(), prot3.getDatasetSequence(), map);
1740 ArrayList<AlignedCodonFrame> acfs = new ArrayList<AlignedCodonFrame>();
1742 protein.setCodonFrames(acfs);
1745 * verify X is included in the aligned proteins, and placed just
1746 * before the first mapped residue
1747 * CCT is between CCC and TTT
1749 AlignmentUtils.alignProteinAsDna(protein, dna);
1750 assertEquals("XK-FG", prot1.getSequenceAsString());
1751 assertEquals("--N-G", prot2.getSequenceAsString());
1752 assertEquals("---XG", prot3.getSequenceAsString());
1756 * Tests for the method that maps the subset of a dna sequence that has CDS
1757 * (or subtype) feature - case where the start codon is incomplete.
1759 @Test(groups = "Functional")
1760 public void testFindCdsPositions_fivePrimeIncomplete()
1762 SequenceI dnaSeq = new Sequence("dna", "aaagGGCCCaaaTTTttt");
1763 dnaSeq.createDatasetSequence();
1764 SequenceI ds = dnaSeq.getDatasetSequence();
1766 // CDS for dna 5-6 (incomplete codon), 7-9
1767 SequenceFeature sf = new SequenceFeature("CDS", "", 5, 9, 0f, null);
1768 sf.setPhase("2"); // skip 2 bases to start of next codon
1769 ds.addSequenceFeature(sf);
1770 // CDS for dna 13-15
1771 sf = new SequenceFeature("CDS_predicted", "", 13, 15, 0f, null);
1772 ds.addSequenceFeature(sf);
1774 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
1777 * check the mapping starts with the first complete codon
1779 assertEquals(6, MappingUtils.getLength(ranges));
1780 assertEquals(2, ranges.size());
1781 assertEquals(7, ranges.get(0)[0]);
1782 assertEquals(9, ranges.get(0)[1]);
1783 assertEquals(13, ranges.get(1)[0]);
1784 assertEquals(15, ranges.get(1)[1]);
1788 * Tests for the method that maps the subset of a dna sequence that has CDS
1789 * (or subtype) feature.
1791 @Test(groups = "Functional")
1792 public void testFindCdsPositions()
1794 SequenceI dnaSeq = new Sequence("dna", "aaaGGGcccAAATTTttt");
1795 dnaSeq.createDatasetSequence();
1796 SequenceI ds = dnaSeq.getDatasetSequence();
1798 // CDS for dna 10-12
1799 SequenceFeature sf = new SequenceFeature("CDS_predicted", "", 10, 12,
1802 ds.addSequenceFeature(sf);
1804 sf = new SequenceFeature("CDS", "", 4, 6, 0f, null);
1806 ds.addSequenceFeature(sf);
1807 // exon feature should be ignored here
1808 sf = new SequenceFeature("exon", "", 7, 9, 0f, null);
1809 ds.addSequenceFeature(sf);
1811 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
1813 * verify ranges { [4-6], [12-10] }
1814 * note CDS ranges are ordered ascending even if the CDS
1817 assertEquals(6, MappingUtils.getLength(ranges));
1818 assertEquals(2, ranges.size());
1819 assertEquals(4, ranges.get(0)[0]);
1820 assertEquals(6, ranges.get(0)[1]);
1821 assertEquals(10, ranges.get(1)[0]);
1822 assertEquals(12, ranges.get(1)[1]);
1826 * Test the method that computes a map of codon variants for each protein
1827 * position from "sequence_variant" features on dna
1829 @Test(groups = "Functional")
1830 public void testBuildDnaVariantsMap()
1832 SequenceI dna = new Sequence("dna", "atgAAATTTGGGCCCtag");
1833 MapList map = new MapList(new int[] { 1, 18 }, new int[] { 1, 5 }, 3, 1);
1836 * first with no variants on dna
1838 LinkedHashMap<Integer, List<DnaVariant>[]> variantsMap = AlignmentUtils
1839 .buildDnaVariantsMap(dna, map);
1840 assertTrue(variantsMap.isEmpty());
1843 * single allele codon 1, on base 1
1845 SequenceFeature sf1 = new SequenceFeature("sequence_variant", "", 1, 1,
1847 sf1.setValue("alleles", "T");
1848 sf1.setValue("ID", "sequence_variant:rs758803211");
1849 dna.addSequenceFeature(sf1);
1852 * two alleles codon 2, on bases 2 and 3 (distinct variants)
1854 SequenceFeature sf2 = new SequenceFeature("sequence_variant", "", 5, 5,
1856 sf2.setValue("alleles", "T");
1857 sf2.setValue("ID", "sequence_variant:rs758803212");
1858 dna.addSequenceFeature(sf2);
1859 SequenceFeature sf3 = new SequenceFeature("sequence_variant", "", 6, 6,
1861 sf3.setValue("alleles", "G");
1862 sf3.setValue("ID", "sequence_variant:rs758803213");
1863 dna.addSequenceFeature(sf3);
1866 * two alleles codon 3, both on base 2 (one variant)
1868 SequenceFeature sf4 = new SequenceFeature("sequence_variant", "", 8, 8,
1870 sf4.setValue("alleles", "C, G");
1871 sf4.setValue("ID", "sequence_variant:rs758803214");
1872 dna.addSequenceFeature(sf4);
1874 // no alleles on codon 4
1877 * alleles on codon 5 on all 3 bases (distinct variants)
1879 SequenceFeature sf5 = new SequenceFeature("sequence_variant", "", 13,
1881 sf5.setValue("alleles", "C, G"); // (C duplicates given base value)
1882 sf5.setValue("ID", "sequence_variant:rs758803215");
1883 dna.addSequenceFeature(sf5);
1884 SequenceFeature sf6 = new SequenceFeature("sequence_variant", "", 14,
1886 sf6.setValue("alleles", "g, a"); // should force to upper-case
1887 sf6.setValue("ID", "sequence_variant:rs758803216");
1888 dna.addSequenceFeature(sf6);
1889 SequenceFeature sf7 = new SequenceFeature("sequence_variant", "", 15,
1891 sf7.setValue("alleles", "A, T");
1892 sf7.setValue("ID", "sequence_variant:rs758803217");
1893 dna.addSequenceFeature(sf7);
1896 * build map - expect variants on positions 1, 2, 3, 5
1898 variantsMap = AlignmentUtils.buildDnaVariantsMap(dna, map);
1899 assertEquals(4, variantsMap.size());
1902 * protein residue 1: variant on codon (ATG) base 1, not on 2 or 3
1904 List<DnaVariant>[] pep1Variants = variantsMap.get(1);
1905 assertEquals(3, pep1Variants.length);
1906 assertEquals(1, pep1Variants[0].size());
1907 assertEquals("A", pep1Variants[0].get(0).base); // codon[1] base
1908 assertSame(sf1, pep1Variants[0].get(0).variant); // codon[1] variant
1909 assertEquals(1, pep1Variants[1].size());
1910 assertEquals("T", pep1Variants[1].get(0).base); // codon[2] base
1911 assertNull(pep1Variants[1].get(0).variant); // no variant here
1912 assertEquals(1, pep1Variants[2].size());
1913 assertEquals("G", pep1Variants[2].get(0).base); // codon[3] base
1914 assertNull(pep1Variants[2].get(0).variant); // no variant here
1917 * protein residue 2: variants on codon (AAA) bases 2 and 3
1919 List<DnaVariant>[] pep2Variants = variantsMap.get(2);
1920 assertEquals(3, pep2Variants.length);
1921 assertEquals(1, pep2Variants[0].size());
1922 // codon[1] base recorded while processing variant on codon[2]
1923 assertEquals("A", pep2Variants[0].get(0).base);
1924 assertNull(pep2Variants[0].get(0).variant); // no variant here
1925 // codon[2] base and variant:
1926 assertEquals(1, pep2Variants[1].size());
1927 assertEquals("A", pep2Variants[1].get(0).base);
1928 assertSame(sf2, pep2Variants[1].get(0).variant);
1929 // codon[3] base was recorded when processing codon[2] variant
1930 // and then the variant for codon[3] added to it
1931 assertEquals(1, pep2Variants[2].size());
1932 assertEquals("A", pep2Variants[2].get(0).base);
1933 assertSame(sf3, pep2Variants[2].get(0).variant);
1936 * protein residue 3: variants on codon (TTT) base 2 only
1938 List<DnaVariant>[] pep3Variants = variantsMap.get(3);
1939 assertEquals(3, pep3Variants.length);
1940 assertEquals(1, pep3Variants[0].size());
1941 assertEquals("T", pep3Variants[0].get(0).base); // codon[1] base
1942 assertNull(pep3Variants[0].get(0).variant); // no variant here
1943 assertEquals(1, pep3Variants[1].size());
1944 assertEquals("T", pep3Variants[1].get(0).base); // codon[2] base
1945 assertSame(sf4, pep3Variants[1].get(0).variant); // codon[2] variant
1946 assertEquals(1, pep3Variants[2].size());
1947 assertEquals("T", pep3Variants[2].get(0).base); // codon[3] base
1948 assertNull(pep3Variants[2].get(0).variant); // no variant here
1951 * three variants on protein position 5
1953 List<DnaVariant>[] pep5Variants = variantsMap.get(5);
1954 assertEquals(3, pep5Variants.length);
1955 assertEquals(1, pep5Variants[0].size());
1956 assertEquals("C", pep5Variants[0].get(0).base); // codon[1] base
1957 assertSame(sf5, pep5Variants[0].get(0).variant); // codon[1] variant
1958 assertEquals(1, pep5Variants[1].size());
1959 assertEquals("C", pep5Variants[1].get(0).base); // codon[2] base
1960 assertSame(sf6, pep5Variants[1].get(0).variant); // codon[2] variant
1961 assertEquals(1, pep5Variants[2].size());
1962 assertEquals("C", pep5Variants[2].get(0).base); // codon[3] base
1963 assertSame(sf7, pep5Variants[2].get(0).variant); // codon[3] variant
1967 * Tests for the method that computes all peptide variants given codon
1970 @Test(groups = "Functional")
1971 public void testComputePeptideVariants()
1974 * scenario: AAATTTCCC codes for KFP
1976 * GAA -> E source: Ensembl
1977 * CAA -> Q source: dbSNP
1978 * AAG synonymous source: COSMIC
1979 * AAT -> N source: Ensembl
1980 * ...TTC synonymous source: dbSNP
1981 * ......CAC,CGC -> H,R source: COSMIC
1982 * (one variant with two alleles)
1984 SequenceI peptide = new Sequence("pep/10-12", "KFP");
1987 * two distinct variants for codon 1 position 1
1988 * second one has clinical significance
1990 String ensembl = "Ensembl";
1991 String dbSnp = "dbSNP";
1992 String cosmic = "COSMIC";
1993 SequenceFeature sf1 = new SequenceFeature("sequence_variant", "", 1, 1,
1995 sf1.setValue("alleles", "A,G"); // GAA -> E
1996 sf1.setValue("ID", "var1.125A>G");
1997 SequenceFeature sf2 = new SequenceFeature("sequence_variant", "", 1, 1,
1999 sf2.setValue("alleles", "A,C"); // CAA -> Q
2000 sf2.setValue("ID", "var2");
2001 sf2.setValue("clinical_significance", "Dodgy");
2002 SequenceFeature sf3 = new SequenceFeature("sequence_variant", "", 3, 3,
2004 sf3.setValue("alleles", "A,G"); // synonymous
2005 sf3.setValue("ID", "var3");
2006 sf3.setValue("clinical_significance", "None");
2007 SequenceFeature sf4 = new SequenceFeature("sequence_variant", "", 3, 3,
2009 sf4.setValue("alleles", "A,T"); // AAT -> N
2010 sf4.setValue("ID", "sequence_variant:var4"); // prefix gets stripped off
2011 sf4.setValue("clinical_significance", "Benign");
2012 SequenceFeature sf5 = new SequenceFeature("sequence_variant", "", 6, 6,
2014 sf5.setValue("alleles", "T,C"); // synonymous
2015 sf5.setValue("ID", "var5");
2016 sf5.setValue("clinical_significance", "Bad");
2017 SequenceFeature sf6 = new SequenceFeature("sequence_variant", "", 8, 8,
2019 sf6.setValue("alleles", "C,A,G"); // CAC,CGC -> H,R
2020 sf6.setValue("ID", "var6");
2021 sf6.setValue("clinical_significance", "Good");
2023 List<DnaVariant> codon1Variants = new ArrayList<DnaVariant>();
2024 List<DnaVariant> codon2Variants = new ArrayList<DnaVariant>();
2025 List<DnaVariant> codon3Variants = new ArrayList<DnaVariant>();
2026 List<DnaVariant> codonVariants[] = new ArrayList[3];
2027 codonVariants[0] = codon1Variants;
2028 codonVariants[1] = codon2Variants;
2029 codonVariants[2] = codon3Variants;
2032 * compute variants for protein position 1
2034 codon1Variants.add(new DnaVariant("A", sf1));
2035 codon1Variants.add(new DnaVariant("A", sf2));
2036 codon2Variants.add(new DnaVariant("A"));
2037 codon2Variants.add(new DnaVariant("A"));
2038 codon3Variants.add(new DnaVariant("A", sf3));
2039 codon3Variants.add(new DnaVariant("A", sf4));
2040 AlignmentUtils.computePeptideVariants(peptide, 1, codonVariants);
2043 * compute variants for protein position 2
2045 codon1Variants.clear();
2046 codon2Variants.clear();
2047 codon3Variants.clear();
2048 codon1Variants.add(new DnaVariant("T"));
2049 codon2Variants.add(new DnaVariant("T"));
2050 codon3Variants.add(new DnaVariant("T", sf5));
2051 AlignmentUtils.computePeptideVariants(peptide, 2, codonVariants);
2054 * compute variants for protein position 3
2056 codon1Variants.clear();
2057 codon2Variants.clear();
2058 codon3Variants.clear();
2059 codon1Variants.add(new DnaVariant("C"));
2060 codon2Variants.add(new DnaVariant("C", sf6));
2061 codon3Variants.add(new DnaVariant("C"));
2062 AlignmentUtils.computePeptideVariants(peptide, 3, codonVariants);
2065 * verify added sequence features for
2066 * var1 K -> E Ensembl
2068 * var4 K -> N Ensembl
2069 * var6 P -> H COSMIC
2070 * var6 P -> R COSMIC
2072 SequenceFeature[] sfs = peptide.getSequenceFeatures();
2073 assertEquals(5, sfs.length);
2075 SequenceFeature sf = sfs[0];
2076 assertEquals(1, sf.getBegin());
2077 assertEquals(1, sf.getEnd());
2078 assertEquals("p.Lys1Glu", sf.getDescription());
2079 assertEquals("var1.125A>G", sf.getValue("ID"));
2080 assertNull(sf.getValue("clinical_significance"));
2081 assertEquals("ID=var1.125A>G", sf.getAttributes());
2082 assertEquals(1, sf.links.size());
2083 // link to variation is urlencoded
2085 "p.Lys1Glu var1.125A>G|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var1.125A%3EG",
2087 assertEquals(ensembl, sf.getFeatureGroup());
2090 assertEquals(1, sf.getBegin());
2091 assertEquals(1, sf.getEnd());
2092 assertEquals("p.Lys1Gln", sf.getDescription());
2093 assertEquals("var2", sf.getValue("ID"));
2094 assertEquals("Dodgy", sf.getValue("clinical_significance"));
2095 assertEquals("ID=var2;clinical_significance=Dodgy", sf.getAttributes());
2096 assertEquals(1, sf.links.size());
2098 "p.Lys1Gln var2|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var2",
2100 assertEquals(dbSnp, sf.getFeatureGroup());
2103 assertEquals(1, sf.getBegin());
2104 assertEquals(1, sf.getEnd());
2105 assertEquals("p.Lys1Asn", sf.getDescription());
2106 assertEquals("var4", sf.getValue("ID"));
2107 assertEquals("Benign", sf.getValue("clinical_significance"));
2108 assertEquals("ID=var4;clinical_significance=Benign", sf.getAttributes());
2109 assertEquals(1, sf.links.size());
2111 "p.Lys1Asn var4|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var4",
2113 assertEquals(ensembl, sf.getFeatureGroup());
2115 // var5 generates two distinct protein variant features
2117 assertEquals(3, sf.getBegin());
2118 assertEquals(3, sf.getEnd());
2119 assertEquals("p.Pro3His", sf.getDescription());
2120 assertEquals("var6", sf.getValue("ID"));
2121 assertEquals("Good", sf.getValue("clinical_significance"));
2122 assertEquals("ID=var6;clinical_significance=Good", sf.getAttributes());
2123 assertEquals(1, sf.links.size());
2125 "p.Pro3His var6|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var6",
2127 assertEquals(cosmic, sf.getFeatureGroup());
2130 assertEquals(3, sf.getBegin());
2131 assertEquals(3, sf.getEnd());
2132 assertEquals("p.Pro3Arg", sf.getDescription());
2133 assertEquals("var6", sf.getValue("ID"));
2134 assertEquals("Good", sf.getValue("clinical_significance"));
2135 assertEquals("ID=var6;clinical_significance=Good", sf.getAttributes());
2136 assertEquals(1, sf.links.size());
2138 "p.Pro3Arg var6|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var6",
2140 assertEquals(cosmic, sf.getFeatureGroup());
2144 * Tests for the method that maps the subset of a dna sequence that has CDS
2145 * (or subtype) feature, with CDS strand = '-' (reverse)
2147 // test turned off as currently findCdsPositions is not strand-dependent
2148 // left in case it comes around again...
2149 @Test(groups = "Functional", enabled = false)
2150 public void testFindCdsPositions_reverseStrand()
2152 SequenceI dnaSeq = new Sequence("dna", "aaaGGGcccAAATTTttt");
2153 dnaSeq.createDatasetSequence();
2154 SequenceI ds = dnaSeq.getDatasetSequence();
2157 SequenceFeature sf = new SequenceFeature("CDS", "", 4, 6, 0f, null);
2159 ds.addSequenceFeature(sf);
2160 // exon feature should be ignored here
2161 sf = new SequenceFeature("exon", "", 7, 9, 0f, null);
2162 ds.addSequenceFeature(sf);
2163 // CDS for dna 10-12
2164 sf = new SequenceFeature("CDS_predicted", "", 10, 12, 0f, null);
2166 ds.addSequenceFeature(sf);
2168 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
2170 * verify ranges { [12-10], [6-4] }
2172 assertEquals(6, MappingUtils.getLength(ranges));
2173 assertEquals(2, ranges.size());
2174 assertEquals(12, ranges.get(0)[0]);
2175 assertEquals(10, ranges.get(0)[1]);
2176 assertEquals(6, ranges.get(1)[0]);
2177 assertEquals(4, ranges.get(1)[1]);
2181 * Tests for the method that maps the subset of a dna sequence that has CDS
2182 * (or subtype) feature - reverse strand case where the start codon is
2185 @Test(groups = "Functional", enabled = false)
2186 // test turned off as currently findCdsPositions is not strand-dependent
2187 // left in case it comes around again...
2188 public void testFindCdsPositions_reverseStrandThreePrimeIncomplete()
2190 SequenceI dnaSeq = new Sequence("dna", "aaagGGCCCaaaTTTttt");
2191 dnaSeq.createDatasetSequence();
2192 SequenceI ds = dnaSeq.getDatasetSequence();
2195 SequenceFeature sf = new SequenceFeature("CDS", "", 5, 9, 0f, null);
2197 ds.addSequenceFeature(sf);
2198 // CDS for dna 13-15
2199 sf = new SequenceFeature("CDS_predicted", "", 13, 15, 0f, null);
2201 sf.setPhase("2"); // skip 2 bases to start of next codon
2202 ds.addSequenceFeature(sf);
2204 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
2207 * check the mapping starts with the first complete codon
2208 * expect ranges [13, 13], [9, 5]
2210 assertEquals(6, MappingUtils.getLength(ranges));
2211 assertEquals(2, ranges.size());
2212 assertEquals(13, ranges.get(0)[0]);
2213 assertEquals(13, ranges.get(0)[1]);
2214 assertEquals(9, ranges.get(1)[0]);
2215 assertEquals(5, ranges.get(1)[1]);
2218 @Test(groups = "Functional")
2219 public void testAlignAs_alternateTranscriptsUngapped()
2221 SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
2222 SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
2223 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
2224 ((Alignment) dna).createDatasetAlignment();
2225 SequenceI cds1 = new Sequence("cds1", "GGGTTT");
2226 SequenceI cds2 = new Sequence("cds2", "CCCAAA");
2227 AlignmentI cds = new Alignment(new SequenceI[] { cds1, cds2 });
2228 ((Alignment) cds).createDatasetAlignment();
2230 AlignedCodonFrame acf = new AlignedCodonFrame();
2231 MapList map = new MapList(new int[] { 4, 9 }, new int[] { 1, 6 }, 1, 1);
2232 acf.addMap(dna1.getDatasetSequence(), cds1.getDatasetSequence(), map);
2233 map = new MapList(new int[] { 1, 3, 10, 12 }, new int[] { 1, 6 }, 1, 1);
2234 acf.addMap(dna2.getDatasetSequence(), cds2.getDatasetSequence(), map);
2237 * verify CDS alignment is as:
2238 * cccGGGTTTaaa (cdna)
2239 * CCCgggtttAAA (cdna)
2241 * ---GGGTTT--- (cds)
2242 * CCC------AAA (cds)
2244 dna.addCodonFrame(acf);
2245 AlignmentUtils.alignAs(cds, dna);
2246 assertEquals("---GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
2247 assertEquals("CCC------AAA", cds.getSequenceAt(1).getSequenceAsString());
2250 @Test(groups = { "Functional" })
2251 public void testAddMappedPositions()
2253 SequenceI from = new Sequence("dna", "ggAA-ATcc-TT-g");
2254 SequenceI seq1 = new Sequence("cds", "AAATTT");
2255 from.createDatasetSequence();
2256 seq1.createDatasetSequence();
2257 Mapping mapping = new Mapping(seq1, new MapList(
2258 new int[] { 3, 6, 9, 10 },
2259 new int[] { 1, 6 }, 1, 1));
2260 Map<Integer, Map<SequenceI, Character>> map = new TreeMap<Integer, Map<SequenceI, Character>>();
2261 AlignmentUtils.addMappedPositions(seq1, from, mapping, map);
2264 * verify map has seq1 residues in columns 3,4,6,7,11,12
2266 assertEquals(6, map.size());
2267 assertEquals('A', map.get(3).get(seq1).charValue());
2268 assertEquals('A', map.get(4).get(seq1).charValue());
2269 assertEquals('A', map.get(6).get(seq1).charValue());
2270 assertEquals('T', map.get(7).get(seq1).charValue());
2271 assertEquals('T', map.get(11).get(seq1).charValue());
2272 assertEquals('T', map.get(12).get(seq1).charValue());
2280 * Test case where the mapping 'from' range includes a stop codon which is
2281 * absent in the 'to' range
2283 @Test(groups = { "Functional" })
2284 public void testAddMappedPositions_withStopCodon()
2286 SequenceI from = new Sequence("dna", "ggAA-ATcc-TT-g");
2287 SequenceI seq1 = new Sequence("cds", "AAATTT");
2288 from.createDatasetSequence();
2289 seq1.createDatasetSequence();
2290 Mapping mapping = new Mapping(seq1, new MapList(
2291 new int[] { 3, 6, 9, 10 },
2292 new int[] { 1, 6 }, 1, 1));
2293 Map<Integer, Map<SequenceI, Character>> map = new TreeMap<Integer, Map<SequenceI, Character>>();
2294 AlignmentUtils.addMappedPositions(seq1, from, mapping, map);
2297 * verify map has seq1 residues in columns 3,4,6,7,11,12
2299 assertEquals(6, map.size());
2300 assertEquals('A', map.get(3).get(seq1).charValue());
2301 assertEquals('A', map.get(4).get(seq1).charValue());
2302 assertEquals('A', map.get(6).get(seq1).charValue());
2303 assertEquals('T', map.get(7).get(seq1).charValue());
2304 assertEquals('T', map.get(11).get(seq1).charValue());
2305 assertEquals('T', map.get(12).get(seq1).charValue());
2309 * Test for the case where the products for which we want CDS are specified.
2310 * This is to represent the case where EMBL has CDS mappings to both Uniprot
2311 * and EMBLCDSPROTEIN. makeCdsAlignment() should only return the mappings for
2312 * the protein sequences specified.
2314 @Test(groups = { "Functional" })
2315 public void testMakeCdsAlignment_filterProducts()
2317 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
2318 SequenceI dna2 = new Sequence("dna2", "GGGcccTTTaaaCCC");
2319 SequenceI pep1 = new Sequence("Uniprot|pep1", "GF");
2320 SequenceI pep2 = new Sequence("Uniprot|pep2", "GFP");
2321 SequenceI pep3 = new Sequence("EMBL|pep3", "GF");
2322 SequenceI pep4 = new Sequence("EMBL|pep4", "GFP");
2323 dna1.createDatasetSequence();
2324 dna2.createDatasetSequence();
2325 pep1.createDatasetSequence();
2326 pep2.createDatasetSequence();
2327 pep3.createDatasetSequence();
2328 pep4.createDatasetSequence();
2329 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
2330 dna.setDataset(null);
2331 AlignmentI emblPeptides = new Alignment(new SequenceI[] { pep3, pep4 });
2332 emblPeptides.setDataset(null);
2334 AlignedCodonFrame acf = new AlignedCodonFrame();
2335 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
2336 new int[] { 1, 2 }, 3, 1);
2337 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
2338 acf.addMap(dna1.getDatasetSequence(), pep3.getDatasetSequence(), map);
2339 dna.addCodonFrame(acf);
2341 acf = new AlignedCodonFrame();
2342 map = new MapList(new int[] { 1, 3, 7, 9, 13, 15 }, new int[] { 1, 3 },
2344 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map);
2345 acf.addMap(dna2.getDatasetSequence(), pep4.getDatasetSequence(), map);
2346 dna.addCodonFrame(acf);
2349 * execute method under test to find CDS for EMBL peptides only
2351 AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
2352 dna1, dna2 }, dna.getDataset(), emblPeptides.getSequencesArray());
2354 assertEquals(2, cds.getSequences().size());
2355 assertEquals("GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
2356 assertEquals("GGGTTTCCC", cds.getSequenceAt(1).getSequenceAsString());
2359 * verify shared, extended alignment dataset
2361 assertSame(dna.getDataset(), cds.getDataset());
2362 assertTrue(dna.getDataset().getSequences()
2363 .contains(cds.getSequenceAt(0).getDatasetSequence()));
2364 assertTrue(dna.getDataset().getSequences()
2365 .contains(cds.getSequenceAt(1).getDatasetSequence()));
2368 * Verify mappings from CDS to peptide, cDNA to CDS, and cDNA to peptide
2369 * the mappings are on the shared alignment dataset
2371 List<AlignedCodonFrame> cdsMappings = cds.getDataset().getCodonFrames();
2373 * 6 mappings, 2*(DNA->CDS), 2*(DNA->Pep), 2*(CDS->Pep)
2375 assertEquals(6, cdsMappings.size());
2378 * verify that mapping sets for dna and cds alignments are different
2379 * [not current behaviour - all mappings are on the alignment dataset]
2381 // select -> subselect type to test.
2382 // Assert.assertNotSame(dna.getCodonFrames(), cds.getCodonFrames());
2383 // assertEquals(4, dna.getCodonFrames().size());
2384 // assertEquals(4, cds.getCodonFrames().size());
2387 * Two mappings involve pep3 (dna to pep3, cds to pep3)
2388 * Mapping from pep3 to GGGTTT in first new exon sequence
2390 List<AlignedCodonFrame> pep3Mappings = MappingUtils
2391 .findMappingsForSequence(pep3, cdsMappings);
2392 assertEquals(2, pep3Mappings.size());
2393 List<AlignedCodonFrame> mappings = MappingUtils
2394 .findMappingsForSequence(cds.getSequenceAt(0), pep3Mappings);
2395 assertEquals(1, mappings.size());
2398 SearchResults sr = MappingUtils.buildSearchResults(pep3, 1, mappings);
2399 assertEquals(1, sr.getResults().size());
2400 Match m = sr.getResults().get(0);
2401 assertSame(cds.getSequenceAt(0).getDatasetSequence(), m.getSequence());
2402 assertEquals(1, m.getStart());
2403 assertEquals(3, m.getEnd());
2405 sr = MappingUtils.buildSearchResults(pep3, 2, mappings);
2406 m = sr.getResults().get(0);
2407 assertSame(cds.getSequenceAt(0).getDatasetSequence(), m.getSequence());
2408 assertEquals(4, m.getStart());
2409 assertEquals(6, m.getEnd());
2412 * Two mappings involve pep4 (dna to pep4, cds to pep4)
2413 * Verify mapping from pep4 to GGGTTTCCC in second new exon sequence
2415 List<AlignedCodonFrame> pep4Mappings = MappingUtils
2416 .findMappingsForSequence(pep4, cdsMappings);
2417 assertEquals(2, pep4Mappings.size());
2418 mappings = MappingUtils.findMappingsForSequence(cds.getSequenceAt(1),
2420 assertEquals(1, mappings.size());
2422 sr = MappingUtils.buildSearchResults(pep4, 1, mappings);
2423 assertEquals(1, sr.getResults().size());
2424 m = sr.getResults().get(0);
2425 assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
2426 assertEquals(1, m.getStart());
2427 assertEquals(3, m.getEnd());
2429 sr = MappingUtils.buildSearchResults(pep4, 2, mappings);
2430 m = sr.getResults().get(0);
2431 assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
2432 assertEquals(4, m.getStart());
2433 assertEquals(6, m.getEnd());
2435 sr = MappingUtils.buildSearchResults(pep4, 3, mappings);
2436 m = sr.getResults().get(0);
2437 assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
2438 assertEquals(7, m.getStart());
2439 assertEquals(9, m.getEnd());
2443 * Test the method that just copies aligned sequences, provided all sequences
2444 * to be aligned share the aligned sequence's dataset
2446 @Test(groups = "Functional")
2447 public void testAlignAsSameSequences()
2449 SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
2450 SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
2451 AlignmentI al1 = new Alignment(new SequenceI[] { dna1, dna2 });
2452 ((Alignment) al1).createDatasetAlignment();
2454 SequenceI dna3 = new Sequence(dna1);
2455 SequenceI dna4 = new Sequence(dna2);
2456 assertSame(dna3.getDatasetSequence(), dna1.getDatasetSequence());
2457 assertSame(dna4.getDatasetSequence(), dna2.getDatasetSequence());
2458 String seq1 = "-cc-GG-GT-TT--aaa";
2459 dna3.setSequence(seq1);
2460 String seq2 = "C--C-Cgg--gtt-tAA-A-";
2461 dna4.setSequence(seq2);
2462 AlignmentI al2 = new Alignment(new SequenceI[] { dna3, dna4 });
2463 ((Alignment) al2).createDatasetAlignment();
2465 assertTrue(AlignmentUtils.alignAsSameSequences(al1, al2));
2466 assertEquals(seq1, al1.getSequenceAt(0).getSequenceAsString());
2467 assertEquals(seq2, al1.getSequenceAt(1).getSequenceAsString());
2470 * add another sequence to 'aligned' - should still succeed, since
2471 * unaligned sequences still share a dataset with aligned sequences
2473 SequenceI dna5 = new Sequence("dna5", "CCCgggtttAAA");
2474 dna5.createDatasetSequence();
2475 al2.addSequence(dna5);
2476 assertTrue(AlignmentUtils.alignAsSameSequences(al1, al2));
2477 assertEquals(seq1, al1.getSequenceAt(0).getSequenceAsString());
2478 assertEquals(seq2, al1.getSequenceAt(1).getSequenceAsString());
2481 * add another sequence to 'unaligned' - should fail, since now not
2482 * all unaligned sequences share a dataset with aligned sequences
2484 SequenceI dna6 = new Sequence("dna6", "CCCgggtttAAA");
2485 dna6.createDatasetSequence();
2486 al1.addSequence(dna6);
2487 // JAL-2110 JBP Comment: what's the use case for this behaviour ?
2488 assertFalse(AlignmentUtils.alignAsSameSequences(al1, al2));
2491 @Test(groups = "Functional")
2492 public void testAlignAsSameSequencesMultipleSubSeq()
2494 SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
2495 SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
2496 SequenceI as1 = dna1.deriveSequence();
2497 SequenceI as2 = dna1.deriveSequence().getSubSequence(3, 7);
2498 SequenceI as3 = dna2.deriveSequence();
2499 as1.insertCharAt(6, 5, '-');
2500 String s_as1 = as1.getSequenceAsString();
2501 as2.insertCharAt(6, 5, '-');
2502 String s_as2 = as2.getSequenceAsString();
2503 as3.insertCharAt(6, 5, '-');
2504 String s_as3 = as3.getSequenceAsString();
2505 AlignmentI aligned = new Alignment(new SequenceI[] { as1, as2, as3 });
2507 // why do we need to cast this still ?
2508 ((Alignment) aligned).createDatasetAlignment();
2509 SequenceI uas1 = dna1.deriveSequence();
2510 SequenceI uas2 = dna1.deriveSequence().getSubSequence(3, 7);
2511 SequenceI uas3 = dna2.deriveSequence();
2512 AlignmentI tobealigned = new Alignment(new SequenceI[] { uas1, uas2,
2514 ((Alignment) tobealigned).createDatasetAlignment();
2516 assertTrue(AlignmentUtils.alignAsSameSequences(tobealigned, aligned));
2517 assertEquals(s_as1, uas1.getSequenceAsString());
2518 assertEquals(s_as2, uas2.getSequenceAsString());
2519 assertEquals(s_as3, uas3.getSequenceAsString());