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.SearchResultMatchI;
39 import jalview.datamodel.SearchResultsI;
40 import jalview.datamodel.Sequence;
41 import jalview.datamodel.SequenceFeature;
42 import jalview.datamodel.SequenceI;
43 import jalview.datamodel.features.SequenceFeatures;
44 import jalview.gui.JvOptionPane;
45 import jalview.io.AppletFormatAdapter;
46 import jalview.io.DataSourceType;
47 import jalview.io.FileFormat;
48 import jalview.io.FileFormatI;
49 import jalview.io.FormatAdapter;
50 import jalview.util.MapList;
51 import jalview.util.MappingUtils;
53 import java.io.IOException;
54 import java.util.ArrayList;
55 import java.util.Arrays;
56 import java.util.LinkedHashMap;
57 import java.util.List;
59 import java.util.TreeMap;
61 import org.testng.annotations.BeforeClass;
62 import org.testng.annotations.Test;
64 public class AlignmentUtilsTests
67 @BeforeClass(alwaysRun = true)
68 public void setUpJvOptionPane()
70 JvOptionPane.setInteractiveMode(false);
71 JvOptionPane.setMockResponse(JvOptionPane.CANCEL_OPTION);
74 public static Sequence ts = new Sequence("short",
75 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklm");
77 @Test(groups = { "Functional" })
78 public void testExpandContext()
80 AlignmentI al = new Alignment(new Sequence[] {});
81 for (int i = 4; i < 14; i += 2)
83 SequenceI s1 = ts.deriveSequence().getSubSequence(i, i + 7);
86 System.out.println(new AppletFormatAdapter().formatSequences(
89 for (int flnk = -1; flnk < 25; flnk++)
91 AlignmentI exp = AlignmentUtils.expandContext(al, flnk);
92 System.out.println("\nFlank size: " + flnk);
93 System.out.println(new AppletFormatAdapter().formatSequences(
94 FileFormat.Clustal, exp, true));
98 * Full expansion to complete sequences
100 for (SequenceI sq : exp.getSequences())
102 String ung = sq.getSequenceAsString().replaceAll("-+", "");
103 final String errorMsg = "Flanking sequence not the same as original dataset sequence.\n"
106 + sq.getDatasetSequence().getSequenceAsString();
107 assertTrue(errorMsg, ung.equalsIgnoreCase(sq.getDatasetSequence()
108 .getSequenceAsString()));
114 * Last sequence is fully expanded, others have leading gaps to match
116 assertTrue(exp.getSequenceAt(4).getSequenceAsString()
118 assertTrue(exp.getSequenceAt(3).getSequenceAsString()
119 .startsWith("--abc"));
120 assertTrue(exp.getSequenceAt(2).getSequenceAsString()
121 .startsWith("----abc"));
122 assertTrue(exp.getSequenceAt(1).getSequenceAsString()
123 .startsWith("------abc"));
124 assertTrue(exp.getSequenceAt(0).getSequenceAsString()
125 .startsWith("--------abc"));
131 * Test that annotations are correctly adjusted by expandContext
133 @Test(groups = { "Functional" })
134 public void testExpandContext_annotation()
136 AlignmentI al = new Alignment(new Sequence[] {});
137 SequenceI ds = new Sequence("Seq1", "ABCDEFGHI");
139 SequenceI seq1 = ds.deriveSequence().getSubSequence(3, 6);
140 al.addSequence(seq1);
143 * Annotate DEF with 4/5/6 respectively
145 Annotation[] anns = new Annotation[] { new Annotation(4),
146 new Annotation(5), new Annotation(6) };
147 AlignmentAnnotation ann = new AlignmentAnnotation("SS",
148 "secondary structure", anns);
149 seq1.addAlignmentAnnotation(ann);
152 * The annotations array should match aligned positions
154 assertEquals(3, ann.annotations.length);
155 assertEquals(4, ann.annotations[0].value, 0.001);
156 assertEquals(5, ann.annotations[1].value, 0.001);
157 assertEquals(6, ann.annotations[2].value, 0.001);
160 * Check annotation to sequence position mappings before expanding the
161 * sequence; these are set up in Sequence.addAlignmentAnnotation ->
162 * Annotation.setSequenceRef -> createSequenceMappings
164 assertNull(ann.getAnnotationForPosition(1));
165 assertNull(ann.getAnnotationForPosition(2));
166 assertNull(ann.getAnnotationForPosition(3));
167 assertEquals(4, ann.getAnnotationForPosition(4).value, 0.001);
168 assertEquals(5, ann.getAnnotationForPosition(5).value, 0.001);
169 assertEquals(6, ann.getAnnotationForPosition(6).value, 0.001);
170 assertNull(ann.getAnnotationForPosition(7));
171 assertNull(ann.getAnnotationForPosition(8));
172 assertNull(ann.getAnnotationForPosition(9));
175 * Expand the subsequence to the full sequence abcDEFghi
177 AlignmentI expanded = AlignmentUtils.expandContext(al, -1);
178 assertEquals("abcDEFghi", expanded.getSequenceAt(0)
179 .getSequenceAsString());
182 * Confirm the alignment and sequence have the same SS annotation,
183 * referencing the expanded sequence
185 ann = expanded.getSequenceAt(0).getAnnotation()[0];
186 assertSame(ann, expanded.getAlignmentAnnotation()[0]);
187 assertSame(expanded.getSequenceAt(0), ann.sequenceRef);
190 * The annotations array should have null values except for annotated
193 assertNull(ann.annotations[0]);
194 assertNull(ann.annotations[1]);
195 assertNull(ann.annotations[2]);
196 assertEquals(4, ann.annotations[3].value, 0.001);
197 assertEquals(5, ann.annotations[4].value, 0.001);
198 assertEquals(6, ann.annotations[5].value, 0.001);
199 assertNull(ann.annotations[6]);
200 assertNull(ann.annotations[7]);
201 assertNull(ann.annotations[8]);
204 * sequence position mappings should be unchanged
206 assertNull(ann.getAnnotationForPosition(1));
207 assertNull(ann.getAnnotationForPosition(2));
208 assertNull(ann.getAnnotationForPosition(3));
209 assertEquals(4, ann.getAnnotationForPosition(4).value, 0.001);
210 assertEquals(5, ann.getAnnotationForPosition(5).value, 0.001);
211 assertEquals(6, ann.getAnnotationForPosition(6).value, 0.001);
212 assertNull(ann.getAnnotationForPosition(7));
213 assertNull(ann.getAnnotationForPosition(8));
214 assertNull(ann.getAnnotationForPosition(9));
218 * Test method that returns a map of lists of sequences by sequence name.
220 * @throws IOException
222 @Test(groups = { "Functional" })
223 public void testGetSequencesByName() throws IOException
225 final String data = ">Seq1Name\nKQYL\n" + ">Seq2Name\nRFPW\n"
226 + ">Seq1Name\nABCD\n";
227 AlignmentI al = loadAlignment(data, FileFormat.Fasta);
228 Map<String, List<SequenceI>> map = AlignmentUtils
229 .getSequencesByName(al);
230 assertEquals(2, map.keySet().size());
231 assertEquals(2, map.get("Seq1Name").size());
232 assertEquals("KQYL", map.get("Seq1Name").get(0).getSequenceAsString());
233 assertEquals("ABCD", map.get("Seq1Name").get(1).getSequenceAsString());
234 assertEquals(1, map.get("Seq2Name").size());
235 assertEquals("RFPW", map.get("Seq2Name").get(0).getSequenceAsString());
239 * Helper method to load an alignment and ensure dataset sequences are set up.
245 * @throws IOException
247 protected AlignmentI loadAlignment(final String data, FileFormatI format)
250 AlignmentI a = new FormatAdapter().readFile(data,
251 DataSourceType.PASTE, format);
257 * Test mapping of protein to cDNA, for the case where we have no sequence
258 * cross-references, so mappings are made first-served 1-1 where sequences
261 * @throws IOException
263 @Test(groups = { "Functional" })
264 public void testMapProteinAlignmentToCdna_noXrefs() throws IOException
266 List<SequenceI> protseqs = new ArrayList<SequenceI>();
267 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
268 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
269 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
270 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
271 protein.setDataset(null);
273 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
274 dnaseqs.add(new Sequence("EMBL|A11111", "TCAGCACGC")); // = SAR
275 dnaseqs.add(new Sequence("EMBL|A22222", "GAGATACAA")); // = EIQ
276 dnaseqs.add(new Sequence("EMBL|A33333", "GAAATCCAG")); // = EIQ
277 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG")); // = EIQ
278 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[4]));
279 cdna.setDataset(null);
281 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
283 // 3 mappings made, each from 1 to 1 sequence
284 assertEquals(3, protein.getCodonFrames().size());
285 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
286 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
287 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
289 // V12345 mapped to A22222
290 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
292 assertEquals(1, acf.getdnaSeqs().length);
293 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
294 acf.getdnaSeqs()[0]);
295 Mapping[] protMappings = acf.getProtMappings();
296 assertEquals(1, protMappings.length);
297 MapList mapList = protMappings[0].getMap();
298 assertEquals(3, mapList.getFromRatio());
299 assertEquals(1, mapList.getToRatio());
300 assertTrue(Arrays.equals(new int[] { 1, 9 }, mapList.getFromRanges()
302 assertEquals(1, mapList.getFromRanges().size());
303 assertTrue(Arrays.equals(new int[] { 1, 3 },
304 mapList.getToRanges().get(0)));
305 assertEquals(1, mapList.getToRanges().size());
307 // V12346 mapped to A33333
308 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
309 assertEquals(1, acf.getdnaSeqs().length);
310 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
311 acf.getdnaSeqs()[0]);
313 // V12347 mapped to A11111
314 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
315 assertEquals(1, acf.getdnaSeqs().length);
316 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
317 acf.getdnaSeqs()[0]);
319 // no mapping involving the 'extra' A44444
320 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(3)).isEmpty());
324 * Test for the alignSequenceAs method that takes two sequences and a mapping.
326 @Test(groups = { "Functional" })
327 public void testAlignSequenceAs_withMapping_noIntrons()
329 MapList map = new MapList(new int[] { 1, 6 }, new int[] { 1, 2 }, 3, 1);
332 * No existing gaps in dna:
334 checkAlignSequenceAs("GGGAAA", "-A-L-", false, false, map,
338 * Now introduce gaps in dna but ignore them when realigning.
340 checkAlignSequenceAs("-G-G-G-A-A-A-", "-A-L-", false, false, map,
344 * Now include gaps in dna when realigning. First retaining 'mapped' gaps
345 * only, i.e. those within the exon region.
347 checkAlignSequenceAs("-G-G--G-A--A-A-", "-A-L-", true, false, map,
348 "---G-G--G---A--A-A");
351 * Include all gaps in dna when realigning (within and without the exon
352 * region). The leading gap, and the gaps between codons, are subsumed by
353 * the protein alignment gap.
355 checkAlignSequenceAs("-G-GG--AA-A---", "-A-L-", true, true, map,
356 "---G-GG---AA-A---");
359 * Include only unmapped gaps in dna when realigning (outside the exon
360 * region). The leading gap, and the gaps between codons, are subsumed by
361 * the protein alignment gap.
363 checkAlignSequenceAs("-G-GG--AA-A-", "-A-L-", false, true, map,
368 * Test for the alignSequenceAs method that takes two sequences and a mapping.
370 @Test(groups = { "Functional" })
371 public void testAlignSequenceAs_withMapping_withIntrons()
374 * Exons at codon 2 (AAA) and 4 (TTT)
376 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
377 new int[] { 1, 2 }, 3, 1);
380 * Simple case: no gaps in dna
382 checkAlignSequenceAs("GGGAAACCCTTTGGG", "--A-L-", false, false, map,
383 "GGG---AAACCCTTTGGG");
386 * Add gaps to dna - but ignore when realigning.
388 checkAlignSequenceAs("-G-G-G--A--A---AC-CC-T-TT-GG-G-", "--A-L-",
389 false, false, map, "GGG---AAACCCTTTGGG");
392 * Add gaps to dna - include within exons only when realigning.
394 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
395 true, false, map, "GGG---A--A---ACCCT-TTGGG");
398 * Include gaps outside exons only when realigning.
400 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
401 false, true, map, "-G-G-GAAAC-CCTTT-GG-G-");
404 * Include gaps following first intron if we are 'preserving mapped gaps'
406 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
407 true, true, map, "-G-G-G--A--A---A-C-CC-T-TT-GG-G-");
410 * Include all gaps in dna when realigning.
412 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
413 true, true, map, "-G-G-G--A--A---A-C-CC-T-TT-GG-G-");
417 * Test for the case where not all of the protein sequence is mapped to cDNA.
419 @Test(groups = { "Functional" })
420 public void testAlignSequenceAs_withMapping_withUnmappedProtein()
423 * Exons at codon 2 (AAA) and 4 (TTT) mapped to A and P
425 final MapList map = new MapList(new int[] { 4, 6, 10, 12 }, new int[] {
429 * -L- 'aligns' ccc------
431 checkAlignSequenceAs("gggAAAcccTTTggg", "-A-L-P-", false, false, map,
432 "gggAAAccc------TTTggg");
436 * Helper method that performs and verifies the method under test.
439 * the sequence to be realigned
441 * the sequence whose alignment is to be copied
442 * @param preserveMappedGaps
443 * @param preserveUnmappedGaps
447 protected void checkAlignSequenceAs(final String alignee,
448 final String alignModel, final boolean preserveMappedGaps,
449 final boolean preserveUnmappedGaps, MapList map,
450 final String expected)
452 SequenceI alignMe = new Sequence("Seq1", alignee);
453 alignMe.createDatasetSequence();
454 SequenceI alignFrom = new Sequence("Seq2", alignModel);
455 alignFrom.createDatasetSequence();
456 AlignedCodonFrame acf = new AlignedCodonFrame();
457 acf.addMap(alignMe.getDatasetSequence(),
458 alignFrom.getDatasetSequence(), map);
460 AlignmentUtils.alignSequenceAs(alignMe, alignFrom, acf, "---", '-',
461 preserveMappedGaps, preserveUnmappedGaps);
462 assertEquals(expected, alignMe.getSequenceAsString());
466 * Test for the alignSequenceAs method where we preserve gaps in introns only.
468 @Test(groups = { "Functional" })
469 public void testAlignSequenceAs_keepIntronGapsOnly()
473 * Intron GGGAAA followed by exon CCCTTT
475 MapList map = new MapList(new int[] { 7, 12 }, new int[] { 1, 2 }, 3, 1);
477 checkAlignSequenceAs("GG-G-AA-A-C-CC-T-TT", "AL", false, true, map,
482 * Test the method that realigns protein to match mapped codon alignment.
484 @Test(groups = { "Functional" })
485 public void testAlignProteinAsDna()
487 // seq1 codons are [1,2,3] [4,5,6] [7,8,9] [10,11,12]
488 SequenceI dna1 = new Sequence("Seq1", "TGCCATTACCAG-");
489 // seq2 codons are [1,3,4] [5,6,7] [8,9,10] [11,12,13]
490 SequenceI dna2 = new Sequence("Seq2", "T-GCCATTACCAG");
491 // seq3 codons are [1,2,3] [4,5,7] [8,9,10] [11,12,13]
492 SequenceI dna3 = new Sequence("Seq3", "TGCCA-TTACCAG");
493 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
494 dna.setDataset(null);
496 // protein alignment will be realigned like dna
497 SequenceI prot1 = new Sequence("Seq1", "CHYQ");
498 SequenceI prot2 = new Sequence("Seq2", "CHYQ");
499 SequenceI prot3 = new Sequence("Seq3", "CHYQ");
500 SequenceI prot4 = new Sequence("Seq4", "R-QSV"); // unmapped, unchanged
501 AlignmentI protein = new Alignment(new SequenceI[] { prot1, prot2,
503 protein.setDataset(null);
505 MapList map = new MapList(new int[] { 1, 12 }, new int[] { 1, 4 }, 3, 1);
506 AlignedCodonFrame acf = new AlignedCodonFrame();
507 acf.addMap(dna1.getDatasetSequence(), prot1.getDatasetSequence(), map);
508 acf.addMap(dna2.getDatasetSequence(), prot2.getDatasetSequence(), map);
509 acf.addMap(dna3.getDatasetSequence(), prot3.getDatasetSequence(), map);
510 ArrayList<AlignedCodonFrame> acfs = new ArrayList<AlignedCodonFrame>();
512 protein.setCodonFrames(acfs);
515 * Translated codon order is [1,2,3] [1,3,4] [4,5,6] [4,5,7] [5,6,7] [7,8,9]
516 * [8,9,10] [10,11,12] [11,12,13]
518 AlignmentUtils.alignProteinAsDna(protein, dna);
519 assertEquals("C-H--Y-Q-", prot1.getSequenceAsString());
520 assertEquals("-C--H-Y-Q", prot2.getSequenceAsString());
521 assertEquals("C--H--Y-Q", prot3.getSequenceAsString());
522 assertEquals("R-QSV", prot4.getSequenceAsString());
526 * Test the method that tests whether a CDNA sequence translates to a protein
529 @Test(groups = { "Functional" })
530 public void testTranslatesAs()
532 // null arguments check
533 assertFalse(AlignmentUtils.translatesAs(null, 0, null));
534 assertFalse(AlignmentUtils.translatesAs(new char[] { 't' }, 0, null));
535 assertFalse(AlignmentUtils.translatesAs(null, 0, new char[] { 'a' }));
537 // straight translation
538 assertTrue(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(), 0,
539 "FPKG".toCharArray()));
540 // with extra start codon (not in protein)
541 assertTrue(AlignmentUtils.translatesAs("atgtttcccaaaggg".toCharArray(),
542 3, "FPKG".toCharArray()));
543 // with stop codon1 (not in protein)
544 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtaa".toCharArray(),
545 0, "FPKG".toCharArray()));
546 // with stop codon1 (in protein as *)
547 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtaa".toCharArray(),
548 0, "FPKG*".toCharArray()));
549 // with stop codon2 (not in protein)
550 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtag".toCharArray(),
551 0, "FPKG".toCharArray()));
552 // with stop codon3 (not in protein)
553 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtga".toCharArray(),
554 0, "FPKG".toCharArray()));
555 // with start and stop codon1
556 assertTrue(AlignmentUtils.translatesAs(
557 "atgtttcccaaagggtaa".toCharArray(), 3, "FPKG".toCharArray()));
558 // with start and stop codon1 (in protein as *)
559 assertTrue(AlignmentUtils.translatesAs(
560 "atgtttcccaaagggtaa".toCharArray(), 3, "FPKG*".toCharArray()));
561 // with start and stop codon2
562 assertTrue(AlignmentUtils.translatesAs(
563 "atgtttcccaaagggtag".toCharArray(), 3, "FPKG".toCharArray()));
564 // with start and stop codon3
565 assertTrue(AlignmentUtils.translatesAs(
566 "atgtttcccaaagggtga".toCharArray(), 3, "FPKG".toCharArray()));
568 // with embedded stop codons
569 assertTrue(AlignmentUtils.translatesAs(
570 "atgtttTAGcccaaaTAAgggtga".toCharArray(), 3,
571 "F*PK*G".toCharArray()));
574 assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(),
575 0, "FPMG".toCharArray()));
578 assertFalse(AlignmentUtils.translatesAs("tttcccaaagg".toCharArray(), 0,
579 "FPKG".toCharArray()));
582 assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(),
583 0, "FPK".toCharArray()));
585 // overlong dna (doesn't end in stop codon)
586 assertFalse(AlignmentUtils.translatesAs(
587 "tttcccaaagggttt".toCharArray(), 0, "FPKG".toCharArray()));
589 // dna + stop codon + more
590 assertFalse(AlignmentUtils.translatesAs(
591 "tttcccaaagggttaga".toCharArray(), 0, "FPKG".toCharArray()));
594 assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(),
595 0, "FPKGQ".toCharArray()));
599 * Test mapping of protein to cDNA, for cases where the cDNA has start and/or
600 * stop codons in addition to the protein coding sequence.
602 * @throws IOException
604 @Test(groups = { "Functional" })
605 public void testMapProteinAlignmentToCdna_withStartAndStopCodons()
608 List<SequenceI> protseqs = new ArrayList<SequenceI>();
609 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
610 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
611 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
612 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
613 protein.setDataset(null);
615 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
617 dnaseqs.add(new Sequence("EMBL|A11111", "ATGTCAGCACGC"));
619 dnaseqs.add(new Sequence("EMBL|A22222", "GAGATACAATAA"));
620 // = start +EIQ + stop
621 dnaseqs.add(new Sequence("EMBL|A33333", "ATGGAAATCCAGTAG"));
622 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG"));
623 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[4]));
624 cdna.setDataset(null);
626 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
628 // 3 mappings made, each from 1 to 1 sequence
629 assertEquals(3, protein.getCodonFrames().size());
630 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
631 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
632 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
634 // V12345 mapped from A22222
635 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
637 assertEquals(1, acf.getdnaSeqs().length);
638 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
639 acf.getdnaSeqs()[0]);
640 Mapping[] protMappings = acf.getProtMappings();
641 assertEquals(1, protMappings.length);
642 MapList mapList = protMappings[0].getMap();
643 assertEquals(3, mapList.getFromRatio());
644 assertEquals(1, mapList.getToRatio());
645 assertTrue(Arrays.equals(new int[] { 1, 9 }, mapList.getFromRanges()
647 assertEquals(1, mapList.getFromRanges().size());
648 assertTrue(Arrays.equals(new int[] { 1, 3 },
649 mapList.getToRanges().get(0)));
650 assertEquals(1, mapList.getToRanges().size());
652 // V12346 mapped from A33333 starting position 4
653 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
654 assertEquals(1, acf.getdnaSeqs().length);
655 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
656 acf.getdnaSeqs()[0]);
657 protMappings = acf.getProtMappings();
658 assertEquals(1, protMappings.length);
659 mapList = protMappings[0].getMap();
660 assertEquals(3, mapList.getFromRatio());
661 assertEquals(1, mapList.getToRatio());
662 assertTrue(Arrays.equals(new int[] { 4, 12 }, mapList.getFromRanges()
664 assertEquals(1, mapList.getFromRanges().size());
665 assertTrue(Arrays.equals(new int[] { 1, 3 },
666 mapList.getToRanges().get(0)));
667 assertEquals(1, mapList.getToRanges().size());
669 // V12347 mapped to A11111 starting position 4
670 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
671 assertEquals(1, acf.getdnaSeqs().length);
672 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
673 acf.getdnaSeqs()[0]);
674 protMappings = acf.getProtMappings();
675 assertEquals(1, protMappings.length);
676 mapList = protMappings[0].getMap();
677 assertEquals(3, mapList.getFromRatio());
678 assertEquals(1, mapList.getToRatio());
679 assertTrue(Arrays.equals(new int[] { 4, 12 }, mapList.getFromRanges()
681 assertEquals(1, mapList.getFromRanges().size());
682 assertTrue(Arrays.equals(new int[] { 1, 3 },
683 mapList.getToRanges().get(0)));
684 assertEquals(1, mapList.getToRanges().size());
686 // no mapping involving the 'extra' A44444
687 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(3)).isEmpty());
691 * Test mapping of protein to cDNA, for the case where we have some sequence
692 * cross-references. Verify that 1-to-many mappings are made where
693 * cross-references exist and sequences are mappable.
695 * @throws IOException
697 @Test(groups = { "Functional" })
698 public void testMapProteinAlignmentToCdna_withXrefs() throws IOException
700 List<SequenceI> protseqs = new ArrayList<SequenceI>();
701 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
702 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
703 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
704 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
705 protein.setDataset(null);
707 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
708 dnaseqs.add(new Sequence("EMBL|A11111", "TCAGCACGC")); // = SAR
709 dnaseqs.add(new Sequence("EMBL|A22222", "ATGGAGATACAA")); // = start + EIQ
710 dnaseqs.add(new Sequence("EMBL|A33333", "GAAATCCAG")); // = EIQ
711 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG")); // = EIQ
712 dnaseqs.add(new Sequence("EMBL|A55555", "GAGATTCAG")); // = EIQ
713 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[5]));
714 cdna.setDataset(null);
716 // Xref A22222 to V12345 (should get mapped)
717 dnaseqs.get(1).addDBRef(new DBRefEntry("UNIPROT", "1", "V12345"));
718 // Xref V12345 to A44444 (should get mapped)
719 protseqs.get(0).addDBRef(new DBRefEntry("EMBL", "1", "A44444"));
720 // Xref A33333 to V12347 (sequence mismatch - should not get mapped)
721 dnaseqs.get(2).addDBRef(new DBRefEntry("UNIPROT", "1", "V12347"));
722 // as V12345 is mapped to A22222 and A44444, this leaves V12346 unmapped.
723 // it should get paired up with the unmapped A33333
724 // A11111 should be mapped to V12347
725 // A55555 is spare and has no xref so is not mapped
727 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
729 // 4 protein mappings made for 3 proteins, 2 to V12345, 1 each to V12346/7
730 assertEquals(3, protein.getCodonFrames().size());
731 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
732 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
733 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
735 // one mapping for each of the first 4 cDNA sequences
736 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(0)).size());
737 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(1)).size());
738 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(2)).size());
739 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(3)).size());
741 // V12345 mapped to A22222 and A44444
742 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
744 assertEquals(2, acf.getdnaSeqs().length);
745 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
746 acf.getdnaSeqs()[0]);
747 assertEquals(cdna.getSequenceAt(3).getDatasetSequence(),
748 acf.getdnaSeqs()[1]);
750 // V12346 mapped to A33333
751 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
752 assertEquals(1, acf.getdnaSeqs().length);
753 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
754 acf.getdnaSeqs()[0]);
756 // V12347 mapped to A11111
757 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
758 assertEquals(1, acf.getdnaSeqs().length);
759 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
760 acf.getdnaSeqs()[0]);
762 // no mapping involving the 'extra' A55555
763 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(4)).isEmpty());
767 * Test mapping of protein to cDNA, for the case where we have some sequence
768 * cross-references. Verify that once we have made an xref mapping we don't
769 * also map un-xrefd sequeces.
771 * @throws IOException
773 @Test(groups = { "Functional" })
774 public void testMapProteinAlignmentToCdna_prioritiseXrefs()
777 List<SequenceI> protseqs = new ArrayList<SequenceI>();
778 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
779 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
780 AlignmentI protein = new Alignment(
781 protseqs.toArray(new SequenceI[protseqs.size()]));
782 protein.setDataset(null);
784 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
785 dnaseqs.add(new Sequence("EMBL|A11111", "GAAATCCAG")); // = EIQ
786 dnaseqs.add(new Sequence("EMBL|A22222", "GAAATTCAG")); // = EIQ
787 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[dnaseqs
789 cdna.setDataset(null);
791 // Xref A22222 to V12345 (should get mapped)
792 // A11111 should then be mapped to the unmapped V12346
793 dnaseqs.get(1).addDBRef(new DBRefEntry("UNIPROT", "1", "V12345"));
795 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
797 // 2 protein mappings made
798 assertEquals(2, protein.getCodonFrames().size());
799 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
800 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
802 // one mapping for each of the cDNA sequences
803 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(0)).size());
804 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(1)).size());
806 // V12345 mapped to A22222
807 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
809 assertEquals(1, acf.getdnaSeqs().length);
810 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
811 acf.getdnaSeqs()[0]);
813 // V12346 mapped to A11111
814 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
815 assertEquals(1, acf.getdnaSeqs().length);
816 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
817 acf.getdnaSeqs()[0]);
821 * Test the method that shows or hides sequence annotations by type(s) and
824 @Test(groups = { "Functional" })
825 public void testShowOrHideSequenceAnnotations()
827 SequenceI seq1 = new Sequence("Seq1", "AAA");
828 SequenceI seq2 = new Sequence("Seq2", "BBB");
829 SequenceI seq3 = new Sequence("Seq3", "CCC");
830 Annotation[] anns = new Annotation[] { new Annotation(2f) };
831 AlignmentAnnotation ann1 = new AlignmentAnnotation("Structure", "ann1",
833 ann1.setSequenceRef(seq1);
834 AlignmentAnnotation ann2 = new AlignmentAnnotation("Structure", "ann2",
836 ann2.setSequenceRef(seq2);
837 AlignmentAnnotation ann3 = new AlignmentAnnotation("Structure", "ann3",
839 AlignmentAnnotation ann4 = new AlignmentAnnotation("Temp", "ann4", anns);
840 ann4.setSequenceRef(seq1);
841 AlignmentAnnotation ann5 = new AlignmentAnnotation("Temp", "ann5", anns);
842 ann5.setSequenceRef(seq2);
843 AlignmentAnnotation ann6 = new AlignmentAnnotation("Temp", "ann6", anns);
844 AlignmentI al = new Alignment(new SequenceI[] { seq1, seq2, seq3 });
845 al.addAnnotation(ann1); // Structure for Seq1
846 al.addAnnotation(ann2); // Structure for Seq2
847 al.addAnnotation(ann3); // Structure for no sequence
848 al.addAnnotation(ann4); // Temp for seq1
849 al.addAnnotation(ann5); // Temp for seq2
850 al.addAnnotation(ann6); // Temp for no sequence
851 List<String> types = new ArrayList<String>();
852 List<SequenceI> scope = new ArrayList<SequenceI>();
855 * Set all sequence related Structure to hidden (ann1, ann2)
857 types.add("Structure");
858 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, false,
860 assertFalse(ann1.visible);
861 assertFalse(ann2.visible);
862 assertTrue(ann3.visible); // not sequence-related, not affected
863 assertTrue(ann4.visible); // not Structure, not affected
864 assertTrue(ann5.visible); // "
865 assertTrue(ann6.visible); // not sequence-related, not affected
868 * Set Temp in {seq1, seq3} to hidden
874 AlignmentUtils.showOrHideSequenceAnnotations(al, types, scope, 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 assertTrue(ann5.visible); // not in scope, not affected
881 assertTrue(ann6.visible); // not sequence-related, not affected
884 * Set Temp in all sequences to hidden
890 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, false,
892 assertFalse(ann1.visible); // unchanged
893 assertFalse(ann2.visible); // unchanged
894 assertTrue(ann3.visible); // not sequence-related, not affected
895 assertFalse(ann4.visible); // Temp for seq1 hidden
896 assertFalse(ann5.visible); // Temp for seq2 hidden
897 assertTrue(ann6.visible); // not sequence-related, not affected
900 * Set all types in {seq1, seq3} to visible
906 AlignmentUtils.showOrHideSequenceAnnotations(al, types, scope, true,
908 assertTrue(ann1.visible); // Structure for seq1 set visible
909 assertFalse(ann2.visible); // not in scope, unchanged
910 assertTrue(ann3.visible); // not sequence-related, not affected
911 assertTrue(ann4.visible); // Temp for seq1 set visible
912 assertFalse(ann5.visible); // not in scope, unchanged
913 assertTrue(ann6.visible); // not sequence-related, not affected
916 * Set all types in all scope to hidden
918 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, true,
920 assertFalse(ann1.visible);
921 assertFalse(ann2.visible);
922 assertTrue(ann3.visible); // not sequence-related, not affected
923 assertFalse(ann4.visible);
924 assertFalse(ann5.visible);
925 assertTrue(ann6.visible); // not sequence-related, not affected
929 * Tests for the method that checks if one sequence cross-references another
931 @Test(groups = { "Functional" })
932 public void testHasCrossRef()
934 assertFalse(AlignmentUtils.hasCrossRef(null, null));
935 SequenceI seq1 = new Sequence("EMBL|A12345", "ABCDEF");
936 assertFalse(AlignmentUtils.hasCrossRef(seq1, null));
937 assertFalse(AlignmentUtils.hasCrossRef(null, seq1));
938 SequenceI seq2 = new Sequence("UNIPROT|V20192", "ABCDEF");
939 assertFalse(AlignmentUtils.hasCrossRef(seq1, seq2));
942 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "v20193"));
943 assertFalse(AlignmentUtils.hasCrossRef(seq1, seq2));
945 // case-insensitive; version number is ignored
946 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "v20192"));
947 assertTrue(AlignmentUtils.hasCrossRef(seq1, seq2));
950 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
951 assertTrue(AlignmentUtils.hasCrossRef(seq1, seq2));
952 // test is one-way only
953 assertFalse(AlignmentUtils.hasCrossRef(seq2, seq1));
957 * Tests for the method that checks if either sequence cross-references the
960 @Test(groups = { "Functional" })
961 public void testHaveCrossRef()
963 assertFalse(AlignmentUtils.hasCrossRef(null, null));
964 SequenceI seq1 = new Sequence("EMBL|A12345", "ABCDEF");
965 assertFalse(AlignmentUtils.haveCrossRef(seq1, null));
966 assertFalse(AlignmentUtils.haveCrossRef(null, seq1));
967 SequenceI seq2 = new Sequence("UNIPROT|V20192", "ABCDEF");
968 assertFalse(AlignmentUtils.haveCrossRef(seq1, seq2));
970 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
971 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
972 // next is true for haveCrossRef, false for hasCrossRef
973 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
975 // now the other way round
976 seq1.setDBRefs(null);
977 seq2.addDBRef(new DBRefEntry("EMBL", "1", "A12345"));
978 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
979 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
982 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
983 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
984 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
988 * Test the method that extracts the cds-only part of a dna alignment.
990 @Test(groups = { "Functional" })
991 public void testMakeCdsAlignment()
995 * dna1 --> [4, 6] [10,12] --> pep1
996 * dna2 --> [1, 3] [7, 9] [13,15] --> pep2
998 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
999 SequenceI dna2 = new Sequence("dna2", "GGGcccTTTaaaCCC");
1000 SequenceI pep1 = new Sequence("pep1", "GF");
1001 SequenceI pep2 = new Sequence("pep2", "GFP");
1002 pep1.addDBRef(new DBRefEntry("UNIPROT", "0", "pep1"));
1003 pep2.addDBRef(new DBRefEntry("UNIPROT", "0", "pep2"));
1004 dna1.createDatasetSequence();
1005 dna2.createDatasetSequence();
1006 pep1.createDatasetSequence();
1007 pep2.createDatasetSequence();
1008 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
1009 dna.setDataset(null);
1012 * put a variant feature on dna2 base 8
1013 * - should transfer to cds2 base 5
1015 dna2.addSequenceFeature(new SequenceFeature("variant", "hgmd", 8, 8,
1019 * need a sourceDbRef if we are to construct dbrefs to the CDS
1020 * sequence from the dna contig sequences
1022 DBRefEntry dbref = new DBRefEntry("ENSEMBL", "0", "dna1");
1023 dna1.getDatasetSequence().addDBRef(dbref);
1024 org.testng.Assert.assertEquals(dbref, dna1.getPrimaryDBRefs().get(0));
1025 dbref = new DBRefEntry("ENSEMBL", "0", "dna2");
1026 dna2.getDatasetSequence().addDBRef(dbref);
1027 org.testng.Assert.assertEquals(dbref, dna2.getPrimaryDBRefs().get(0));
1030 * CDS sequences are 'discovered' from dna-to-protein mappings on the alignment
1031 * dataset (e.g. added from dbrefs by CrossRef.findXrefSequences)
1033 MapList mapfordna1 = new MapList(new int[] { 4, 6, 10, 12 }, new int[] {
1035 AlignedCodonFrame acf = new AlignedCodonFrame();
1036 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(),
1038 dna.addCodonFrame(acf);
1039 MapList mapfordna2 = new MapList(new int[] { 1, 3, 7, 9, 13, 15 },
1040 new int[] { 1, 3 }, 3, 1);
1041 acf = new AlignedCodonFrame();
1042 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(),
1044 dna.addCodonFrame(acf);
1047 * 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
1049 DBRefEntry dna1xref = new DBRefEntry("UNIPROT", "ENSEMBL", "pep1",
1050 new Mapping(mapfordna1));
1051 dna1.getDatasetSequence().addDBRef(dna1xref);
1052 DBRefEntry dna2xref = new DBRefEntry("UNIPROT", "ENSEMBL", "pep2",
1053 new Mapping(mapfordna2));
1054 dna2.getDatasetSequence().addDBRef(dna2xref);
1057 * execute method under test:
1059 AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
1060 dna1, dna2 }, dna.getDataset(), null);
1063 * verify cds sequences
1065 assertEquals(2, cds.getSequences().size());
1066 assertEquals("GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
1067 assertEquals("GGGTTTCCC", cds.getSequenceAt(1).getSequenceAsString());
1070 * verify shared, extended alignment dataset
1072 assertSame(dna.getDataset(), cds.getDataset());
1073 SequenceI cds1Dss = cds.getSequenceAt(0).getDatasetSequence();
1074 SequenceI cds2Dss = cds.getSequenceAt(1).getDatasetSequence();
1075 assertTrue(dna.getDataset().getSequences().contains(cds1Dss));
1076 assertTrue(dna.getDataset().getSequences().contains(cds2Dss));
1079 * verify CDS has a dbref with mapping to peptide
1081 assertNotNull(cds1Dss.getDBRefs());
1082 assertEquals(2, cds1Dss.getDBRefs().length);
1083 dbref = cds1Dss.getDBRefs()[0];
1084 assertEquals(dna1xref.getSource(), dbref.getSource());
1085 // version is via ensembl's primary ref
1086 assertEquals(dna1xref.getVersion(), dbref.getVersion());
1087 assertEquals(dna1xref.getAccessionId(), dbref.getAccessionId());
1088 assertNotNull(dbref.getMap());
1089 assertSame(pep1.getDatasetSequence(), dbref.getMap().getTo());
1090 MapList cdsMapping = new MapList(new int[] { 1, 6 },
1091 new int[] { 1, 2 }, 3, 1);
1092 assertEquals(cdsMapping, dbref.getMap().getMap());
1095 * verify peptide has added a dbref with reverse mapping to CDS
1097 assertNotNull(pep1.getDBRefs());
1098 // FIXME pep1.getDBRefs() is 1 - is that the correct behaviour ?
1099 assertEquals(2, pep1.getDBRefs().length);
1100 dbref = pep1.getDBRefs()[1];
1101 assertEquals("ENSEMBL", dbref.getSource());
1102 assertEquals("0", dbref.getVersion());
1103 assertEquals("CDS|dna1", dbref.getAccessionId());
1104 assertNotNull(dbref.getMap());
1105 assertSame(cds1Dss, dbref.getMap().getTo());
1106 assertEquals(cdsMapping.getInverse(), dbref.getMap().getMap());
1109 * Verify mappings from CDS to peptide, cDNA to CDS, and cDNA to peptide
1110 * the mappings are on the shared alignment dataset
1111 * 6 mappings, 2*(DNA->CDS), 2*(DNA->Pep), 2*(CDS->Pep)
1113 List<AlignedCodonFrame> cdsMappings = cds.getDataset().getCodonFrames();
1114 assertEquals(6, cdsMappings.size());
1117 * verify that mapping sets for dna and cds alignments are different
1118 * [not current behaviour - all mappings are on the alignment dataset]
1120 // select -> subselect type to test.
1121 // Assert.assertNotSame(dna.getCodonFrames(), cds.getCodonFrames());
1122 // assertEquals(4, dna.getCodonFrames().size());
1123 // assertEquals(4, cds.getCodonFrames().size());
1126 * Two mappings involve pep1 (dna to pep1, cds to pep1)
1127 * Mapping from pep1 to GGGTTT in first new exon sequence
1129 List<AlignedCodonFrame> pep1Mappings = MappingUtils
1130 .findMappingsForSequence(pep1, cdsMappings);
1131 assertEquals(2, pep1Mappings.size());
1132 List<AlignedCodonFrame> mappings = MappingUtils
1133 .findMappingsForSequence(cds.getSequenceAt(0), pep1Mappings);
1134 assertEquals(1, mappings.size());
1137 SearchResultsI sr = MappingUtils.buildSearchResults(pep1, 1, mappings);
1138 assertEquals(1, sr.getResults().size());
1139 SearchResultMatchI m = sr.getResults().get(0);
1140 assertSame(cds1Dss, m.getSequence());
1141 assertEquals(1, m.getStart());
1142 assertEquals(3, m.getEnd());
1144 sr = MappingUtils.buildSearchResults(pep1, 2, mappings);
1145 m = sr.getResults().get(0);
1146 assertSame(cds1Dss, m.getSequence());
1147 assertEquals(4, m.getStart());
1148 assertEquals(6, m.getEnd());
1151 * Two mappings involve pep2 (dna to pep2, cds to pep2)
1152 * Verify mapping from pep2 to GGGTTTCCC in second new exon sequence
1154 List<AlignedCodonFrame> pep2Mappings = MappingUtils
1155 .findMappingsForSequence(pep2, cdsMappings);
1156 assertEquals(2, pep2Mappings.size());
1157 mappings = MappingUtils.findMappingsForSequence(cds.getSequenceAt(1),
1159 assertEquals(1, mappings.size());
1161 sr = MappingUtils.buildSearchResults(pep2, 1, mappings);
1162 assertEquals(1, sr.getResults().size());
1163 m = sr.getResults().get(0);
1164 assertSame(cds2Dss, m.getSequence());
1165 assertEquals(1, m.getStart());
1166 assertEquals(3, m.getEnd());
1168 sr = MappingUtils.buildSearchResults(pep2, 2, mappings);
1169 m = sr.getResults().get(0);
1170 assertSame(cds2Dss, m.getSequence());
1171 assertEquals(4, m.getStart());
1172 assertEquals(6, m.getEnd());
1174 sr = MappingUtils.buildSearchResults(pep2, 3, mappings);
1175 m = sr.getResults().get(0);
1176 assertSame(cds2Dss, m.getSequence());
1177 assertEquals(7, m.getStart());
1178 assertEquals(9, m.getEnd());
1181 * check cds2 acquired a variant feature in position 5
1183 List<SequenceFeature> sfs = cds2Dss.getSequenceFeatures();
1185 assertEquals(1, sfs.size());
1186 assertEquals("variant", sfs.get(0).type);
1187 assertEquals(5, sfs.get(0).begin);
1188 assertEquals(5, sfs.get(0).end);
1192 * Test the method that makes a cds-only alignment from a DNA sequence and its
1193 * product mappings, for the case where there are multiple exon mappings to
1194 * different protein products.
1196 @Test(groups = { "Functional" })
1197 public void testMakeCdsAlignment_multipleProteins()
1199 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
1200 SequenceI pep1 = new Sequence("pep1", "GF"); // GGGTTT
1201 SequenceI pep2 = new Sequence("pep2", "KP"); // aaaccc
1202 SequenceI pep3 = new Sequence("pep3", "KF"); // aaaTTT
1203 dna1.createDatasetSequence();
1204 pep1.createDatasetSequence();
1205 pep2.createDatasetSequence();
1206 pep3.createDatasetSequence();
1207 pep1.getDatasetSequence().addDBRef(
1208 new DBRefEntry("EMBLCDS", "2", "A12345"));
1209 pep2.getDatasetSequence().addDBRef(
1210 new DBRefEntry("EMBLCDS", "3", "A12346"));
1211 pep3.getDatasetSequence().addDBRef(
1212 new DBRefEntry("EMBLCDS", "4", "A12347"));
1215 * Create the CDS alignment
1217 AlignmentI dna = new Alignment(new SequenceI[] { dna1 });
1218 dna.setDataset(null);
1221 * Make the mappings from dna to protein
1223 // map ...GGG...TTT to GF
1224 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1225 new int[] { 1, 2 }, 3, 1);
1226 AlignedCodonFrame acf = new AlignedCodonFrame();
1227 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
1228 dna.addCodonFrame(acf);
1230 // map aaa...ccc to KP
1231 map = new MapList(new int[] { 1, 3, 7, 9 }, new int[] { 1, 2 }, 3, 1);
1232 acf = new AlignedCodonFrame();
1233 acf.addMap(dna1.getDatasetSequence(), pep2.getDatasetSequence(), map);
1234 dna.addCodonFrame(acf);
1236 // map aaa......TTT to KF
1237 map = new MapList(new int[] { 1, 3, 10, 12 }, new int[] { 1, 2 }, 3, 1);
1238 acf = new AlignedCodonFrame();
1239 acf.addMap(dna1.getDatasetSequence(), pep3.getDatasetSequence(), map);
1240 dna.addCodonFrame(acf);
1243 * execute method under test
1245 AlignmentI cdsal = AlignmentUtils.makeCdsAlignment(
1246 new SequenceI[] { dna1 }, dna.getDataset(), null);
1249 * Verify we have 3 cds sequences, mapped to pep1/2/3 respectively
1251 List<SequenceI> cds = cdsal.getSequences();
1252 assertEquals(3, cds.size());
1255 * verify shared, extended alignment dataset
1257 assertSame(cdsal.getDataset(), dna.getDataset());
1258 assertTrue(dna.getDataset().getSequences()
1259 .contains(cds.get(0).getDatasetSequence()));
1260 assertTrue(dna.getDataset().getSequences()
1261 .contains(cds.get(1).getDatasetSequence()));
1262 assertTrue(dna.getDataset().getSequences()
1263 .contains(cds.get(2).getDatasetSequence()));
1266 * verify aligned cds sequences and their xrefs
1268 SequenceI cdsSeq = cds.get(0);
1269 assertEquals("GGGTTT", cdsSeq.getSequenceAsString());
1270 // assertEquals("dna1|A12345", cdsSeq.getName());
1271 assertEquals("CDS|dna1", cdsSeq.getName());
1272 // assertEquals(1, cdsSeq.getDBRefs().length);
1273 // DBRefEntry cdsRef = cdsSeq.getDBRefs()[0];
1274 // assertEquals("EMBLCDS", cdsRef.getSource());
1275 // assertEquals("2", cdsRef.getVersion());
1276 // assertEquals("A12345", cdsRef.getAccessionId());
1278 cdsSeq = cds.get(1);
1279 assertEquals("aaaccc", cdsSeq.getSequenceAsString());
1280 // assertEquals("dna1|A12346", cdsSeq.getName());
1281 assertEquals("CDS|dna1", cdsSeq.getName());
1282 // assertEquals(1, cdsSeq.getDBRefs().length);
1283 // cdsRef = cdsSeq.getDBRefs()[0];
1284 // assertEquals("EMBLCDS", cdsRef.getSource());
1285 // assertEquals("3", cdsRef.getVersion());
1286 // assertEquals("A12346", cdsRef.getAccessionId());
1288 cdsSeq = cds.get(2);
1289 assertEquals("aaaTTT", cdsSeq.getSequenceAsString());
1290 // assertEquals("dna1|A12347", cdsSeq.getName());
1291 assertEquals("CDS|dna1", cdsSeq.getName());
1292 // assertEquals(1, cdsSeq.getDBRefs().length);
1293 // cdsRef = cdsSeq.getDBRefs()[0];
1294 // assertEquals("EMBLCDS", cdsRef.getSource());
1295 // assertEquals("4", cdsRef.getVersion());
1296 // assertEquals("A12347", cdsRef.getAccessionId());
1299 * Verify there are mappings from each cds sequence to its protein product
1300 * and also to its dna source
1302 List<AlignedCodonFrame> newMappings = cdsal.getCodonFrames();
1305 * 6 mappings involve dna1 (to pep1/2/3, cds1/2/3)
1307 List<AlignedCodonFrame> dnaMappings = MappingUtils
1308 .findMappingsForSequence(dna1, newMappings);
1309 assertEquals(6, dnaMappings.size());
1314 List<AlignedCodonFrame> mappings = MappingUtils
1315 .findMappingsForSequence(pep1, dnaMappings);
1316 assertEquals(1, mappings.size());
1317 assertEquals(1, mappings.get(0).getMappings().size());
1318 assertSame(pep1.getDatasetSequence(), mappings.get(0).getMappings()
1319 .get(0).getMapping().getTo());
1324 List<AlignedCodonFrame> dnaToCds1Mappings = MappingUtils
1325 .findMappingsForSequence(cds.get(0), dnaMappings);
1326 Mapping mapping = dnaToCds1Mappings.get(0).getMappings().get(0)
1328 assertSame(cds.get(0).getDatasetSequence(), mapping.getTo());
1329 assertEquals("G(1) in CDS should map to G(4) in DNA", 4, mapping
1330 .getMap().getToPosition(1));
1335 mappings = MappingUtils.findMappingsForSequence(pep2, dnaMappings);
1336 assertEquals(1, mappings.size());
1337 assertEquals(1, mappings.get(0).getMappings().size());
1338 assertSame(pep2.getDatasetSequence(), mappings.get(0).getMappings()
1339 .get(0).getMapping().getTo());
1344 List<AlignedCodonFrame> dnaToCds2Mappings = MappingUtils
1345 .findMappingsForSequence(cds.get(1), dnaMappings);
1346 mapping = dnaToCds2Mappings.get(0).getMappings().get(0).getMapping();
1347 assertSame(cds.get(1).getDatasetSequence(), mapping.getTo());
1348 assertEquals("c(4) in CDS should map to c(7) in DNA", 7, mapping
1349 .getMap().getToPosition(4));
1354 mappings = MappingUtils.findMappingsForSequence(pep3, dnaMappings);
1355 assertEquals(1, mappings.size());
1356 assertEquals(1, mappings.get(0).getMappings().size());
1357 assertSame(pep3.getDatasetSequence(), mappings.get(0).getMappings()
1358 .get(0).getMapping().getTo());
1363 List<AlignedCodonFrame> dnaToCds3Mappings = MappingUtils
1364 .findMappingsForSequence(cds.get(2), dnaMappings);
1365 mapping = dnaToCds3Mappings.get(0).getMappings().get(0).getMapping();
1366 assertSame(cds.get(2).getDatasetSequence(), mapping.getTo());
1367 assertEquals("T(4) in CDS should map to T(10) in DNA", 10, mapping
1368 .getMap().getToPosition(4));
1371 @Test(groups = { "Functional" })
1372 public void testIsMappable()
1374 SequenceI dna1 = new Sequence("dna1", "cgCAGtgGT");
1375 SequenceI aa1 = new Sequence("aa1", "RSG");
1376 AlignmentI al1 = new Alignment(new SequenceI[] { dna1 });
1377 AlignmentI al2 = new Alignment(new SequenceI[] { aa1 });
1379 assertFalse(AlignmentUtils.isMappable(null, null));
1380 assertFalse(AlignmentUtils.isMappable(al1, null));
1381 assertFalse(AlignmentUtils.isMappable(null, al1));
1382 assertFalse(AlignmentUtils.isMappable(al1, al1));
1383 assertFalse(AlignmentUtils.isMappable(al2, al2));
1385 assertTrue(AlignmentUtils.isMappable(al1, al2));
1386 assertTrue(AlignmentUtils.isMappable(al2, al1));
1390 * Test creating a mapping when the sequences involved do not start at residue
1393 * @throws IOException
1395 @Test(groups = { "Functional" })
1396 public void testMapCdnaToProtein_forSubsequence() throws IOException
1398 SequenceI prot = new Sequence("UNIPROT|V12345", "E-I--Q", 10, 12);
1399 prot.createDatasetSequence();
1401 SequenceI dna = new Sequence("EMBL|A33333", "GAA--AT-C-CAG", 40, 48);
1402 dna.createDatasetSequence();
1404 MapList map = AlignmentUtils.mapCdnaToProtein(prot, dna);
1405 assertEquals(10, map.getToLowest());
1406 assertEquals(12, map.getToHighest());
1407 assertEquals(40, map.getFromLowest());
1408 assertEquals(48, map.getFromHighest());
1412 * Test for the alignSequenceAs method where we have protein mapped to protein
1414 @Test(groups = { "Functional" })
1415 public void testAlignSequenceAs_mappedProteinProtein()
1418 SequenceI alignMe = new Sequence("Match", "MGAASEV");
1419 alignMe.createDatasetSequence();
1420 SequenceI alignFrom = new Sequence("Query", "LQTGYMGAASEVMFSPTRR");
1421 alignFrom.createDatasetSequence();
1423 AlignedCodonFrame acf = new AlignedCodonFrame();
1424 // this is like a domain or motif match of part of a peptide sequence
1425 MapList map = new MapList(new int[] { 6, 12 }, new int[] { 1, 7 }, 1, 1);
1426 acf.addMap(alignFrom.getDatasetSequence(),
1427 alignMe.getDatasetSequence(), map);
1429 AlignmentUtils.alignSequenceAs(alignMe, alignFrom, acf, "-", '-', true,
1431 assertEquals("-----MGAASEV-------", alignMe.getSequenceAsString());
1435 * Test for the alignSequenceAs method where there are trailing unmapped
1436 * residues in the model sequence
1438 @Test(groups = { "Functional" })
1439 public void testAlignSequenceAs_withTrailingPeptide()
1441 // map first 3 codons to KPF; G is a trailing unmapped residue
1442 MapList map = new MapList(new int[] { 1, 9 }, new int[] { 1, 3 }, 3, 1);
1444 checkAlignSequenceAs("AAACCCTTT", "K-PFG", true, true, map,
1449 * Tests for transferring features between mapped sequences
1451 @Test(groups = { "Functional" })
1452 public void testTransferFeatures()
1454 SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
1455 SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
1458 dna.addSequenceFeature(new SequenceFeature("type1", "desc1", 1, 2, 1f,
1460 // partial overlap - to [1, 1]
1461 dna.addSequenceFeature(new SequenceFeature("type2", "desc2", 3, 4, 2f,
1463 // exact overlap - to [1, 3]
1464 dna.addSequenceFeature(new SequenceFeature("type3", "desc3", 4, 6, 3f,
1466 // spanning overlap - to [2, 5]
1467 dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
1469 // exactly overlaps whole mapped range [1, 6]
1470 dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
1472 // no overlap (internal)
1473 dna.addSequenceFeature(new SequenceFeature("type6", "desc6", 7, 9, 6f,
1475 // no overlap (3' end)
1476 dna.addSequenceFeature(new SequenceFeature("type7", "desc7", 13, 15,
1478 // overlap (3' end) - to [6, 6]
1479 dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
1481 // extended overlap - to [6, +]
1482 dna.addSequenceFeature(new SequenceFeature("type9", "desc9", 12, 13,
1485 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1486 new int[] { 1, 6 }, 1, 1);
1489 * transferFeatures() will build 'partial overlap' for regions
1490 * that partially overlap 5' or 3' (start or end) of target sequence
1492 AlignmentUtils.transferFeatures(dna, cds, map, null);
1493 List<SequenceFeature> sfs = cds.getSequenceFeatures();
1494 assertEquals(6, sfs.size());
1496 SequenceFeature sf = sfs.get(0);
1497 assertEquals("type2", sf.getType());
1498 assertEquals("desc2", sf.getDescription());
1499 assertEquals(2f, sf.getScore());
1500 assertEquals(1, sf.getBegin());
1501 assertEquals(1, sf.getEnd());
1504 assertEquals("type3", sf.getType());
1505 assertEquals("desc3", sf.getDescription());
1506 assertEquals(3f, sf.getScore());
1507 assertEquals(1, sf.getBegin());
1508 assertEquals(3, sf.getEnd());
1511 assertEquals("type4", sf.getType());
1512 assertEquals(2, sf.getBegin());
1513 assertEquals(5, sf.getEnd());
1516 assertEquals("type5", sf.getType());
1517 assertEquals(1, sf.getBegin());
1518 assertEquals(6, sf.getEnd());
1521 assertEquals("type8", sf.getType());
1522 assertEquals(6, sf.getBegin());
1523 assertEquals(6, sf.getEnd());
1526 assertEquals("type9", sf.getType());
1527 assertEquals(6, sf.getBegin());
1528 assertEquals(6, sf.getEnd());
1532 * Tests for transferring features between mapped sequences
1534 @Test(groups = { "Functional" })
1535 public void testTransferFeatures_withOmit()
1537 SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
1538 SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
1540 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1541 new int[] { 1, 6 }, 1, 1);
1543 // [5, 11] maps to [2, 5]
1544 dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
1546 // [4, 12] maps to [1, 6]
1547 dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
1549 // [12, 12] maps to [6, 6]
1550 dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
1553 // desc4 and desc8 are the 'omit these' varargs
1554 AlignmentUtils.transferFeatures(dna, cds, map, null, "type4", "type8");
1555 List<SequenceFeature> sfs = cds.getSequenceFeatures();
1556 assertEquals(1, sfs.size());
1558 SequenceFeature sf = sfs.get(0);
1559 assertEquals("type5", sf.getType());
1560 assertEquals(1, sf.getBegin());
1561 assertEquals(6, sf.getEnd());
1565 * Tests for transferring features between mapped sequences
1567 @Test(groups = { "Functional" })
1568 public void testTransferFeatures_withSelect()
1570 SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
1571 SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
1573 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1574 new int[] { 1, 6 }, 1, 1);
1576 // [5, 11] maps to [2, 5]
1577 dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
1579 // [4, 12] maps to [1, 6]
1580 dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
1582 // [12, 12] maps to [6, 6]
1583 dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
1586 // "type5" is the 'select this type' argument
1587 AlignmentUtils.transferFeatures(dna, cds, map, "type5");
1588 List<SequenceFeature> sfs = cds.getSequenceFeatures();
1589 assertEquals(1, sfs.size());
1591 SequenceFeature sf = sfs.get(0);
1592 assertEquals("type5", sf.getType());
1593 assertEquals(1, sf.getBegin());
1594 assertEquals(6, sf.getEnd());
1598 * Test the method that extracts the cds-only part of a dna alignment, for the
1599 * case where the cds should be aligned to match its nucleotide sequence.
1601 @Test(groups = { "Functional" })
1602 public void testMakeCdsAlignment_alternativeTranscripts()
1604 SequenceI dna1 = new Sequence("dna1", "aaaGGGCC-----CTTTaaaGGG");
1605 // alternative transcript of same dna skips CCC codon
1606 SequenceI dna2 = new Sequence("dna2", "aaaGGGCC-----cttTaaaGGG");
1607 // dna3 has no mapping (protein product) so should be ignored here
1608 SequenceI dna3 = new Sequence("dna3", "aaaGGGCCCCCGGGcttTaaaGGG");
1609 SequenceI pep1 = new Sequence("pep1", "GPFG");
1610 SequenceI pep2 = new Sequence("pep2", "GPG");
1611 dna1.createDatasetSequence();
1612 dna2.createDatasetSequence();
1613 dna3.createDatasetSequence();
1614 pep1.createDatasetSequence();
1615 pep2.createDatasetSequence();
1617 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
1618 dna.setDataset(null);
1620 MapList map = new MapList(new int[] { 4, 12, 16, 18 },
1621 new int[] { 1, 4 }, 3, 1);
1622 AlignedCodonFrame acf = new AlignedCodonFrame();
1623 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
1624 dna.addCodonFrame(acf);
1625 map = new MapList(new int[] { 4, 8, 12, 12, 16, 18 },
1626 new int[] { 1, 3 }, 3, 1);
1627 acf = new AlignedCodonFrame();
1628 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map);
1629 dna.addCodonFrame(acf);
1631 AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
1632 dna1, dna2, dna3 }, dna.getDataset(), null);
1633 List<SequenceI> cdsSeqs = cds.getSequences();
1634 assertEquals(2, cdsSeqs.size());
1635 assertEquals("GGGCCCTTTGGG", cdsSeqs.get(0).getSequenceAsString());
1636 assertEquals("GGGCCTGGG", cdsSeqs.get(1).getSequenceAsString());
1639 * verify shared, extended alignment dataset
1641 assertSame(dna.getDataset(), cds.getDataset());
1642 assertTrue(dna.getDataset().getSequences()
1643 .contains(cdsSeqs.get(0).getDatasetSequence()));
1644 assertTrue(dna.getDataset().getSequences()
1645 .contains(cdsSeqs.get(1).getDatasetSequence()));
1648 * Verify 6 mappings: dna1 to cds1, cds1 to pep1, dna1 to pep1
1649 * and the same for dna2/cds2/pep2
1651 List<AlignedCodonFrame> mappings = cds.getCodonFrames();
1652 assertEquals(6, mappings.size());
1655 * 2 mappings involve pep1
1657 List<AlignedCodonFrame> pep1Mappings = MappingUtils
1658 .findMappingsForSequence(pep1, mappings);
1659 assertEquals(2, pep1Mappings.size());
1662 * Get mapping of pep1 to cds1 and verify it
1663 * maps GPFG to 1-3,4-6,7-9,10-12
1665 List<AlignedCodonFrame> pep1CdsMappings = MappingUtils
1666 .findMappingsForSequence(cds.getSequenceAt(0), pep1Mappings);
1667 assertEquals(1, pep1CdsMappings.size());
1668 SearchResultsI sr = MappingUtils.buildSearchResults(pep1, 1,
1670 assertEquals(1, sr.getResults().size());
1671 SearchResultMatchI m = sr.getResults().get(0);
1672 assertEquals(cds.getSequenceAt(0).getDatasetSequence(), m.getSequence());
1673 assertEquals(1, m.getStart());
1674 assertEquals(3, m.getEnd());
1675 sr = MappingUtils.buildSearchResults(pep1, 2, pep1CdsMappings);
1676 m = sr.getResults().get(0);
1677 assertEquals(4, m.getStart());
1678 assertEquals(6, m.getEnd());
1679 sr = MappingUtils.buildSearchResults(pep1, 3, pep1CdsMappings);
1680 m = sr.getResults().get(0);
1681 assertEquals(7, m.getStart());
1682 assertEquals(9, m.getEnd());
1683 sr = MappingUtils.buildSearchResults(pep1, 4, pep1CdsMappings);
1684 m = sr.getResults().get(0);
1685 assertEquals(10, m.getStart());
1686 assertEquals(12, m.getEnd());
1689 * Get mapping of pep2 to cds2 and verify it
1690 * maps GPG in pep2 to 1-3,4-6,7-9 in second CDS sequence
1692 List<AlignedCodonFrame> pep2Mappings = MappingUtils
1693 .findMappingsForSequence(pep2, mappings);
1694 assertEquals(2, pep2Mappings.size());
1695 List<AlignedCodonFrame> pep2CdsMappings = MappingUtils
1696 .findMappingsForSequence(cds.getSequenceAt(1), pep2Mappings);
1697 assertEquals(1, pep2CdsMappings.size());
1698 sr = MappingUtils.buildSearchResults(pep2, 1, pep2CdsMappings);
1699 assertEquals(1, sr.getResults().size());
1700 m = sr.getResults().get(0);
1701 assertEquals(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
1702 assertEquals(1, m.getStart());
1703 assertEquals(3, m.getEnd());
1704 sr = MappingUtils.buildSearchResults(pep2, 2, pep2CdsMappings);
1705 m = sr.getResults().get(0);
1706 assertEquals(4, m.getStart());
1707 assertEquals(6, m.getEnd());
1708 sr = MappingUtils.buildSearchResults(pep2, 3, pep2CdsMappings);
1709 m = sr.getResults().get(0);
1710 assertEquals(7, m.getStart());
1711 assertEquals(9, m.getEnd());
1715 * Test the method that realigns protein to match mapped codon alignment.
1717 @Test(groups = { "Functional" })
1718 public void testAlignProteinAsDna_incompleteStartCodon()
1720 // seq1: incomplete start codon (not mapped), then [3, 11]
1721 SequenceI dna1 = new Sequence("Seq1", "ccAAA-TTT-GGG-");
1722 // seq2 codons are [4, 5], [8, 11]
1723 SequenceI dna2 = new Sequence("Seq2", "ccaAA-ttT-GGG-");
1724 // seq3 incomplete start codon at 'tt'
1725 SequenceI dna3 = new Sequence("Seq3", "ccaaa-ttt-GGG-");
1726 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
1727 dna.setDataset(null);
1729 // prot1 has 'X' for incomplete start codon (not mapped)
1730 SequenceI prot1 = new Sequence("Seq1", "XKFG"); // X for incomplete start
1731 SequenceI prot2 = new Sequence("Seq2", "NG");
1732 SequenceI prot3 = new Sequence("Seq3", "XG"); // X for incomplete start
1733 AlignmentI protein = new Alignment(new SequenceI[] { prot1, prot2,
1735 protein.setDataset(null);
1737 // map dna1 [3, 11] to prot1 [2, 4] KFG
1738 MapList map = new MapList(new int[] { 3, 11 }, new int[] { 2, 4 }, 3, 1);
1739 AlignedCodonFrame acf = new AlignedCodonFrame();
1740 acf.addMap(dna1.getDatasetSequence(), prot1.getDatasetSequence(), map);
1742 // map dna2 [4, 5] [8, 11] to prot2 [1, 2] NG
1743 map = new MapList(new int[] { 4, 5, 8, 11 }, new int[] { 1, 2 }, 3, 1);
1744 acf.addMap(dna2.getDatasetSequence(), prot2.getDatasetSequence(), map);
1746 // map dna3 [9, 11] to prot3 [2, 2] G
1747 map = new MapList(new int[] { 9, 11 }, new int[] { 2, 2 }, 3, 1);
1748 acf.addMap(dna3.getDatasetSequence(), prot3.getDatasetSequence(), map);
1750 ArrayList<AlignedCodonFrame> acfs = new ArrayList<AlignedCodonFrame>();
1752 protein.setCodonFrames(acfs);
1755 * verify X is included in the aligned proteins, and placed just
1756 * before the first mapped residue
1757 * CCT is between CCC and TTT
1759 AlignmentUtils.alignProteinAsDna(protein, dna);
1760 assertEquals("XK-FG", prot1.getSequenceAsString());
1761 assertEquals("--N-G", prot2.getSequenceAsString());
1762 assertEquals("---XG", prot3.getSequenceAsString());
1766 * Tests for the method that maps the subset of a dna sequence that has CDS
1767 * (or subtype) feature - case where the start codon is incomplete.
1769 @Test(groups = "Functional")
1770 public void testFindCdsPositions_fivePrimeIncomplete()
1772 SequenceI dnaSeq = new Sequence("dna", "aaagGGCCCaaaTTTttt");
1773 dnaSeq.createDatasetSequence();
1774 SequenceI ds = dnaSeq.getDatasetSequence();
1776 // CDS for dna 5-6 (incomplete codon), 7-9
1777 SequenceFeature sf = new SequenceFeature("CDS", "", 5, 9, 0f, null);
1778 sf.setPhase("2"); // skip 2 bases to start of next codon
1779 ds.addSequenceFeature(sf);
1780 // CDS for dna 13-15
1781 sf = new SequenceFeature("CDS_predicted", "", 13, 15, 0f, null);
1782 ds.addSequenceFeature(sf);
1784 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
1787 * check the mapping starts with the first complete codon
1789 assertEquals(6, MappingUtils.getLength(ranges));
1790 assertEquals(2, ranges.size());
1791 assertEquals(7, ranges.get(0)[0]);
1792 assertEquals(9, ranges.get(0)[1]);
1793 assertEquals(13, ranges.get(1)[0]);
1794 assertEquals(15, ranges.get(1)[1]);
1798 * Tests for the method that maps the subset of a dna sequence that has CDS
1799 * (or subtype) feature.
1801 @Test(groups = "Functional")
1802 public void testFindCdsPositions()
1804 SequenceI dnaSeq = new Sequence("dna", "aaaGGGcccAAATTTttt");
1805 dnaSeq.createDatasetSequence();
1806 SequenceI ds = dnaSeq.getDatasetSequence();
1808 // CDS for dna 10-12
1809 SequenceFeature sf = new SequenceFeature("CDS_predicted", "", 10, 12,
1812 ds.addSequenceFeature(sf);
1814 sf = new SequenceFeature("CDS", "", 4, 6, 0f, null);
1816 ds.addSequenceFeature(sf);
1817 // exon feature should be ignored here
1818 sf = new SequenceFeature("exon", "", 7, 9, 0f, null);
1819 ds.addSequenceFeature(sf);
1821 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
1823 * verify ranges { [4-6], [12-10] }
1824 * note CDS ranges are ordered ascending even if the CDS
1827 assertEquals(6, MappingUtils.getLength(ranges));
1828 assertEquals(2, ranges.size());
1829 assertEquals(4, ranges.get(0)[0]);
1830 assertEquals(6, ranges.get(0)[1]);
1831 assertEquals(10, ranges.get(1)[0]);
1832 assertEquals(12, ranges.get(1)[1]);
1836 * Test the method that computes a map of codon variants for each protein
1837 * position from "sequence_variant" features on dna
1839 @Test(groups = "Functional")
1840 public void testBuildDnaVariantsMap()
1842 SequenceI dna = new Sequence("dna", "atgAAATTTGGGCCCtag");
1843 MapList map = new MapList(new int[] { 1, 18 }, new int[] { 1, 5 }, 3, 1);
1846 * first with no variants on dna
1848 LinkedHashMap<Integer, List<DnaVariant>[]> variantsMap = AlignmentUtils
1849 .buildDnaVariantsMap(dna, map);
1850 assertTrue(variantsMap.isEmpty());
1853 * single allele codon 1, on base 1
1855 SequenceFeature sf1 = new SequenceFeature("sequence_variant", "", 1, 1,
1857 sf1.setValue("alleles", "T");
1858 sf1.setValue("ID", "sequence_variant:rs758803211");
1859 dna.addSequenceFeature(sf1);
1862 * two alleles codon 2, on bases 2 and 3 (distinct variants)
1864 SequenceFeature sf2 = new SequenceFeature("sequence_variant", "", 5, 5,
1866 sf2.setValue("alleles", "T");
1867 sf2.setValue("ID", "sequence_variant:rs758803212");
1868 dna.addSequenceFeature(sf2);
1869 SequenceFeature sf3 = new SequenceFeature("sequence_variant", "", 6, 6,
1871 sf3.setValue("alleles", "G");
1872 sf3.setValue("ID", "sequence_variant:rs758803213");
1873 dna.addSequenceFeature(sf3);
1876 * two alleles codon 3, both on base 2 (one variant)
1878 SequenceFeature sf4 = new SequenceFeature("sequence_variant", "", 8, 8,
1880 sf4.setValue("alleles", "C, G");
1881 sf4.setValue("ID", "sequence_variant:rs758803214");
1882 dna.addSequenceFeature(sf4);
1884 // no alleles on codon 4
1887 * alleles on codon 5 on all 3 bases (distinct variants)
1889 SequenceFeature sf5 = new SequenceFeature("sequence_variant", "", 13,
1891 sf5.setValue("alleles", "C, G"); // (C duplicates given base value)
1892 sf5.setValue("ID", "sequence_variant:rs758803215");
1893 dna.addSequenceFeature(sf5);
1894 SequenceFeature sf6 = new SequenceFeature("sequence_variant", "", 14,
1896 sf6.setValue("alleles", "g, a"); // should force to upper-case
1897 sf6.setValue("ID", "sequence_variant:rs758803216");
1898 dna.addSequenceFeature(sf6);
1899 SequenceFeature sf7 = new SequenceFeature("sequence_variant", "", 15,
1901 sf7.setValue("alleles", "A, T");
1902 sf7.setValue("ID", "sequence_variant:rs758803217");
1903 dna.addSequenceFeature(sf7);
1906 * build map - expect variants on positions 1, 2, 3, 5
1908 variantsMap = AlignmentUtils.buildDnaVariantsMap(dna, map);
1909 assertEquals(4, variantsMap.size());
1912 * protein residue 1: variant on codon (ATG) base 1, not on 2 or 3
1914 List<DnaVariant>[] pep1Variants = variantsMap.get(1);
1915 assertEquals(3, pep1Variants.length);
1916 assertEquals(1, pep1Variants[0].size());
1917 assertEquals("A", pep1Variants[0].get(0).base); // codon[1] base
1918 assertSame(sf1, pep1Variants[0].get(0).variant); // codon[1] variant
1919 assertEquals(1, pep1Variants[1].size());
1920 assertEquals("T", pep1Variants[1].get(0).base); // codon[2] base
1921 assertNull(pep1Variants[1].get(0).variant); // no variant here
1922 assertEquals(1, pep1Variants[2].size());
1923 assertEquals("G", pep1Variants[2].get(0).base); // codon[3] base
1924 assertNull(pep1Variants[2].get(0).variant); // no variant here
1927 * protein residue 2: variants on codon (AAA) bases 2 and 3
1929 List<DnaVariant>[] pep2Variants = variantsMap.get(2);
1930 assertEquals(3, pep2Variants.length);
1931 assertEquals(1, pep2Variants[0].size());
1932 // codon[1] base recorded while processing variant on codon[2]
1933 assertEquals("A", pep2Variants[0].get(0).base);
1934 assertNull(pep2Variants[0].get(0).variant); // no variant here
1935 // codon[2] base and variant:
1936 assertEquals(1, pep2Variants[1].size());
1937 assertEquals("A", pep2Variants[1].get(0).base);
1938 assertSame(sf2, pep2Variants[1].get(0).variant);
1939 // codon[3] base was recorded when processing codon[2] variant
1940 // and then the variant for codon[3] added to it
1941 assertEquals(1, pep2Variants[2].size());
1942 assertEquals("A", pep2Variants[2].get(0).base);
1943 assertSame(sf3, pep2Variants[2].get(0).variant);
1946 * protein residue 3: variants on codon (TTT) base 2 only
1948 List<DnaVariant>[] pep3Variants = variantsMap.get(3);
1949 assertEquals(3, pep3Variants.length);
1950 assertEquals(1, pep3Variants[0].size());
1951 assertEquals("T", pep3Variants[0].get(0).base); // codon[1] base
1952 assertNull(pep3Variants[0].get(0).variant); // no variant here
1953 assertEquals(1, pep3Variants[1].size());
1954 assertEquals("T", pep3Variants[1].get(0).base); // codon[2] base
1955 assertSame(sf4, pep3Variants[1].get(0).variant); // codon[2] variant
1956 assertEquals(1, pep3Variants[2].size());
1957 assertEquals("T", pep3Variants[2].get(0).base); // codon[3] base
1958 assertNull(pep3Variants[2].get(0).variant); // no variant here
1961 * three variants on protein position 5
1963 List<DnaVariant>[] pep5Variants = variantsMap.get(5);
1964 assertEquals(3, pep5Variants.length);
1965 assertEquals(1, pep5Variants[0].size());
1966 assertEquals("C", pep5Variants[0].get(0).base); // codon[1] base
1967 assertSame(sf5, pep5Variants[0].get(0).variant); // codon[1] variant
1968 assertEquals(1, pep5Variants[1].size());
1969 assertEquals("C", pep5Variants[1].get(0).base); // codon[2] base
1970 assertSame(sf6, pep5Variants[1].get(0).variant); // codon[2] variant
1971 assertEquals(1, pep5Variants[2].size());
1972 assertEquals("C", pep5Variants[2].get(0).base); // codon[3] base
1973 assertSame(sf7, pep5Variants[2].get(0).variant); // codon[3] variant
1977 * Tests for the method that computes all peptide variants given codon
1980 @Test(groups = "Functional")
1981 public void testComputePeptideVariants()
1984 * scenario: AAATTTCCC codes for KFP
1986 * GAA -> E source: Ensembl
1987 * CAA -> Q source: dbSNP
1988 * AAG synonymous source: COSMIC
1989 * AAT -> N source: Ensembl
1990 * ...TTC synonymous source: dbSNP
1991 * ......CAC,CGC -> H,R source: COSMIC
1992 * (one variant with two alleles)
1994 SequenceI peptide = new Sequence("pep/10-12", "KFP");
1997 * two distinct variants for codon 1 position 1
1998 * second one has clinical significance
2000 String ensembl = "Ensembl";
2001 String dbSnp = "dbSNP";
2002 String cosmic = "COSMIC";
2003 SequenceFeature sf1 = new SequenceFeature("sequence_variant", "", 1, 1,
2005 sf1.setValue("alleles", "A,G"); // GAA -> E
2006 sf1.setValue("ID", "var1.125A>G");
2007 SequenceFeature sf2 = new SequenceFeature("sequence_variant", "", 1, 1,
2009 sf2.setValue("alleles", "A,C"); // CAA -> Q
2010 sf2.setValue("ID", "var2");
2011 sf2.setValue("clinical_significance", "Dodgy");
2012 SequenceFeature sf3 = new SequenceFeature("sequence_variant", "", 3, 3,
2014 sf3.setValue("alleles", "A,G"); // synonymous
2015 sf3.setValue("ID", "var3");
2016 sf3.setValue("clinical_significance", "None");
2017 SequenceFeature sf4 = new SequenceFeature("sequence_variant", "", 3, 3,
2019 sf4.setValue("alleles", "A,T"); // AAT -> N
2020 sf4.setValue("ID", "sequence_variant:var4"); // prefix gets stripped off
2021 sf4.setValue("clinical_significance", "Benign");
2022 SequenceFeature sf5 = new SequenceFeature("sequence_variant", "", 6, 6,
2024 sf5.setValue("alleles", "T,C"); // synonymous
2025 sf5.setValue("ID", "var5");
2026 sf5.setValue("clinical_significance", "Bad");
2027 SequenceFeature sf6 = new SequenceFeature("sequence_variant", "", 8, 8,
2029 sf6.setValue("alleles", "C,A,G"); // CAC,CGC -> H,R
2030 sf6.setValue("ID", "var6");
2031 sf6.setValue("clinical_significance", "Good");
2033 List<DnaVariant> codon1Variants = new ArrayList<DnaVariant>();
2034 List<DnaVariant> codon2Variants = new ArrayList<DnaVariant>();
2035 List<DnaVariant> codon3Variants = new ArrayList<DnaVariant>();
2036 List<DnaVariant> codonVariants[] = new ArrayList[3];
2037 codonVariants[0] = codon1Variants;
2038 codonVariants[1] = codon2Variants;
2039 codonVariants[2] = codon3Variants;
2042 * compute variants for protein position 1
2044 codon1Variants.add(new DnaVariant("A", sf1));
2045 codon1Variants.add(new DnaVariant("A", sf2));
2046 codon2Variants.add(new DnaVariant("A"));
2047 codon2Variants.add(new DnaVariant("A"));
2048 codon3Variants.add(new DnaVariant("A", sf3));
2049 codon3Variants.add(new DnaVariant("A", sf4));
2050 AlignmentUtils.computePeptideVariants(peptide, 1, codonVariants);
2053 * compute variants for protein position 2
2055 codon1Variants.clear();
2056 codon2Variants.clear();
2057 codon3Variants.clear();
2058 codon1Variants.add(new DnaVariant("T"));
2059 codon2Variants.add(new DnaVariant("T"));
2060 codon3Variants.add(new DnaVariant("T", sf5));
2061 AlignmentUtils.computePeptideVariants(peptide, 2, codonVariants);
2064 * compute variants for protein position 3
2066 codon1Variants.clear();
2067 codon2Variants.clear();
2068 codon3Variants.clear();
2069 codon1Variants.add(new DnaVariant("C"));
2070 codon2Variants.add(new DnaVariant("C", sf6));
2071 codon3Variants.add(new DnaVariant("C"));
2072 AlignmentUtils.computePeptideVariants(peptide, 3, codonVariants);
2075 * verify added sequence features for
2076 * var1 K -> E Ensembl
2078 * var4 K -> N Ensembl
2079 * var6 P -> H COSMIC
2080 * var6 P -> R COSMIC
2082 List<SequenceFeature> sfs = peptide.getSequenceFeatures();
2083 SequenceFeatures.sortFeatures(sfs, true);
2084 assertEquals(5, sfs.size());
2087 * features are sorted by start position ascending, but in no
2088 * particular order where start positions match; asserts here
2089 * simply match the data returned (the order is not important)
2091 SequenceFeature sf = sfs.get(0);
2092 assertEquals(1, sf.getBegin());
2093 assertEquals(1, sf.getEnd());
2094 assertEquals("p.Lys1Asn", sf.getDescription());
2095 assertEquals("var4", sf.getValue("ID"));
2096 assertEquals("Benign", sf.getValue("clinical_significance"));
2097 assertEquals("ID=var4;clinical_significance=Benign", sf.getAttributes());
2098 assertEquals(1, sf.links.size());
2100 "p.Lys1Asn var4|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var4",
2102 assertEquals(ensembl, sf.getFeatureGroup());
2105 assertEquals(1, sf.getBegin());
2106 assertEquals(1, sf.getEnd());
2107 assertEquals("p.Lys1Gln", sf.getDescription());
2108 assertEquals("var2", sf.getValue("ID"));
2109 assertEquals("Dodgy", sf.getValue("clinical_significance"));
2110 assertEquals("ID=var2;clinical_significance=Dodgy", sf.getAttributes());
2111 assertEquals(1, sf.links.size());
2113 "p.Lys1Gln var2|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var2",
2115 assertEquals(dbSnp, sf.getFeatureGroup());
2118 assertEquals(1, sf.getBegin());
2119 assertEquals(1, sf.getEnd());
2120 assertEquals("p.Lys1Glu", sf.getDescription());
2121 assertEquals("var1.125A>G", sf.getValue("ID"));
2122 assertNull(sf.getValue("clinical_significance"));
2123 assertEquals("ID=var1.125A>G", sf.getAttributes());
2124 assertEquals(1, sf.links.size());
2125 // link to variation is urlencoded
2127 "p.Lys1Glu var1.125A>G|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var1.125A%3EG",
2129 assertEquals(ensembl, sf.getFeatureGroup());
2132 assertEquals(3, sf.getBegin());
2133 assertEquals(3, sf.getEnd());
2134 assertEquals("p.Pro3Arg", sf.getDescription());
2135 assertEquals("var6", sf.getValue("ID"));
2136 assertEquals("Good", sf.getValue("clinical_significance"));
2137 assertEquals("ID=var6;clinical_significance=Good", sf.getAttributes());
2138 assertEquals(1, sf.links.size());
2140 "p.Pro3Arg var6|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var6",
2142 assertEquals(cosmic, sf.getFeatureGroup());
2144 // var5 generates two distinct protein variant features
2146 assertEquals(3, sf.getBegin());
2147 assertEquals(3, sf.getEnd());
2148 assertEquals("p.Pro3His", sf.getDescription());
2149 assertEquals("var6", sf.getValue("ID"));
2150 assertEquals("Good", sf.getValue("clinical_significance"));
2151 assertEquals("ID=var6;clinical_significance=Good", sf.getAttributes());
2152 assertEquals(1, sf.links.size());
2154 "p.Pro3His var6|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var6",
2156 assertEquals(cosmic, sf.getFeatureGroup());
2160 * Tests for the method that maps the subset of a dna sequence that has CDS
2161 * (or subtype) feature, with CDS strand = '-' (reverse)
2163 // test turned off as currently findCdsPositions is not strand-dependent
2164 // left in case it comes around again...
2165 @Test(groups = "Functional", enabled = false)
2166 public void testFindCdsPositions_reverseStrand()
2168 SequenceI dnaSeq = new Sequence("dna", "aaaGGGcccAAATTTttt");
2169 dnaSeq.createDatasetSequence();
2170 SequenceI ds = dnaSeq.getDatasetSequence();
2173 SequenceFeature sf = new SequenceFeature("CDS", "", 4, 6, 0f, null);
2175 ds.addSequenceFeature(sf);
2176 // exon feature should be ignored here
2177 sf = new SequenceFeature("exon", "", 7, 9, 0f, null);
2178 ds.addSequenceFeature(sf);
2179 // CDS for dna 10-12
2180 sf = new SequenceFeature("CDS_predicted", "", 10, 12, 0f, null);
2182 ds.addSequenceFeature(sf);
2184 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
2186 * verify ranges { [12-10], [6-4] }
2188 assertEquals(6, MappingUtils.getLength(ranges));
2189 assertEquals(2, ranges.size());
2190 assertEquals(12, ranges.get(0)[0]);
2191 assertEquals(10, ranges.get(0)[1]);
2192 assertEquals(6, ranges.get(1)[0]);
2193 assertEquals(4, ranges.get(1)[1]);
2197 * Tests for the method that maps the subset of a dna sequence that has CDS
2198 * (or subtype) feature - reverse strand case where the start codon is
2201 @Test(groups = "Functional", enabled = false)
2202 // test turned off as currently findCdsPositions is not strand-dependent
2203 // left in case it comes around again...
2204 public void testFindCdsPositions_reverseStrandThreePrimeIncomplete()
2206 SequenceI dnaSeq = new Sequence("dna", "aaagGGCCCaaaTTTttt");
2207 dnaSeq.createDatasetSequence();
2208 SequenceI ds = dnaSeq.getDatasetSequence();
2211 SequenceFeature sf = new SequenceFeature("CDS", "", 5, 9, 0f, null);
2213 ds.addSequenceFeature(sf);
2214 // CDS for dna 13-15
2215 sf = new SequenceFeature("CDS_predicted", "", 13, 15, 0f, null);
2217 sf.setPhase("2"); // skip 2 bases to start of next codon
2218 ds.addSequenceFeature(sf);
2220 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
2223 * check the mapping starts with the first complete codon
2224 * expect ranges [13, 13], [9, 5]
2226 assertEquals(6, MappingUtils.getLength(ranges));
2227 assertEquals(2, ranges.size());
2228 assertEquals(13, ranges.get(0)[0]);
2229 assertEquals(13, ranges.get(0)[1]);
2230 assertEquals(9, ranges.get(1)[0]);
2231 assertEquals(5, ranges.get(1)[1]);
2234 @Test(groups = "Functional")
2235 public void testAlignAs_alternateTranscriptsUngapped()
2237 SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
2238 SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
2239 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
2240 ((Alignment) dna).createDatasetAlignment();
2241 SequenceI cds1 = new Sequence("cds1", "GGGTTT");
2242 SequenceI cds2 = new Sequence("cds2", "CCCAAA");
2243 AlignmentI cds = new Alignment(new SequenceI[] { cds1, cds2 });
2244 ((Alignment) cds).createDatasetAlignment();
2246 AlignedCodonFrame acf = new AlignedCodonFrame();
2247 MapList map = new MapList(new int[] { 4, 9 }, new int[] { 1, 6 }, 1, 1);
2248 acf.addMap(dna1.getDatasetSequence(), cds1.getDatasetSequence(), map);
2249 map = new MapList(new int[] { 1, 3, 10, 12 }, new int[] { 1, 6 }, 1, 1);
2250 acf.addMap(dna2.getDatasetSequence(), cds2.getDatasetSequence(), map);
2253 * verify CDS alignment is as:
2254 * cccGGGTTTaaa (cdna)
2255 * CCCgggtttAAA (cdna)
2257 * ---GGGTTT--- (cds)
2258 * CCC------AAA (cds)
2260 dna.addCodonFrame(acf);
2261 AlignmentUtils.alignAs(cds, dna);
2262 assertEquals("---GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
2263 assertEquals("CCC------AAA", cds.getSequenceAt(1).getSequenceAsString());
2266 @Test(groups = { "Functional" })
2267 public void testAddMappedPositions()
2269 SequenceI from = new Sequence("dna", "ggAA-ATcc-TT-g");
2270 SequenceI seq1 = new Sequence("cds", "AAATTT");
2271 from.createDatasetSequence();
2272 seq1.createDatasetSequence();
2273 Mapping mapping = new Mapping(seq1, new MapList(
2274 new int[] { 3, 6, 9, 10 }, new int[] { 1, 6 }, 1, 1));
2275 Map<Integer, Map<SequenceI, Character>> map = new TreeMap<Integer, Map<SequenceI, Character>>();
2276 AlignmentUtils.addMappedPositions(seq1, from, mapping, map);
2279 * verify map has seq1 residues in columns 3,4,6,7,11,12
2281 assertEquals(6, map.size());
2282 assertEquals('A', map.get(3).get(seq1).charValue());
2283 assertEquals('A', map.get(4).get(seq1).charValue());
2284 assertEquals('A', map.get(6).get(seq1).charValue());
2285 assertEquals('T', map.get(7).get(seq1).charValue());
2286 assertEquals('T', map.get(11).get(seq1).charValue());
2287 assertEquals('T', map.get(12).get(seq1).charValue());
2295 * Test case where the mapping 'from' range includes a stop codon which is
2296 * absent in the 'to' range
2298 @Test(groups = { "Functional" })
2299 public void testAddMappedPositions_withStopCodon()
2301 SequenceI from = new Sequence("dna", "ggAA-ATcc-TT-g");
2302 SequenceI seq1 = new Sequence("cds", "AAATTT");
2303 from.createDatasetSequence();
2304 seq1.createDatasetSequence();
2305 Mapping mapping = new Mapping(seq1, new MapList(
2306 new int[] { 3, 6, 9, 10 }, new int[] { 1, 6 }, 1, 1));
2307 Map<Integer, Map<SequenceI, Character>> map = new TreeMap<Integer, Map<SequenceI, Character>>();
2308 AlignmentUtils.addMappedPositions(seq1, from, mapping, map);
2311 * verify map has seq1 residues in columns 3,4,6,7,11,12
2313 assertEquals(6, map.size());
2314 assertEquals('A', map.get(3).get(seq1).charValue());
2315 assertEquals('A', map.get(4).get(seq1).charValue());
2316 assertEquals('A', map.get(6).get(seq1).charValue());
2317 assertEquals('T', map.get(7).get(seq1).charValue());
2318 assertEquals('T', map.get(11).get(seq1).charValue());
2319 assertEquals('T', map.get(12).get(seq1).charValue());
2323 * Test for the case where the products for which we want CDS are specified.
2324 * This is to represent the case where EMBL has CDS mappings to both Uniprot
2325 * and EMBLCDSPROTEIN. makeCdsAlignment() should only return the mappings for
2326 * the protein sequences specified.
2328 @Test(groups = { "Functional" })
2329 public void testMakeCdsAlignment_filterProducts()
2331 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
2332 SequenceI dna2 = new Sequence("dna2", "GGGcccTTTaaaCCC");
2333 SequenceI pep1 = new Sequence("Uniprot|pep1", "GF");
2334 SequenceI pep2 = new Sequence("Uniprot|pep2", "GFP");
2335 SequenceI pep3 = new Sequence("EMBL|pep3", "GF");
2336 SequenceI pep4 = new Sequence("EMBL|pep4", "GFP");
2337 dna1.createDatasetSequence();
2338 dna2.createDatasetSequence();
2339 pep1.createDatasetSequence();
2340 pep2.createDatasetSequence();
2341 pep3.createDatasetSequence();
2342 pep4.createDatasetSequence();
2343 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
2344 dna.setDataset(null);
2345 AlignmentI emblPeptides = new Alignment(new SequenceI[] { pep3, pep4 });
2346 emblPeptides.setDataset(null);
2348 AlignedCodonFrame acf = new AlignedCodonFrame();
2349 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
2350 new int[] { 1, 2 }, 3, 1);
2351 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
2352 acf.addMap(dna1.getDatasetSequence(), pep3.getDatasetSequence(), map);
2353 dna.addCodonFrame(acf);
2355 acf = new AlignedCodonFrame();
2356 map = new MapList(new int[] { 1, 3, 7, 9, 13, 15 }, new int[] { 1, 3 },
2358 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map);
2359 acf.addMap(dna2.getDatasetSequence(), pep4.getDatasetSequence(), map);
2360 dna.addCodonFrame(acf);
2363 * execute method under test to find CDS for EMBL peptides only
2365 AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
2366 dna1, dna2 }, dna.getDataset(), emblPeptides.getSequencesArray());
2368 assertEquals(2, cds.getSequences().size());
2369 assertEquals("GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
2370 assertEquals("GGGTTTCCC", cds.getSequenceAt(1).getSequenceAsString());
2373 * verify shared, extended alignment dataset
2375 assertSame(dna.getDataset(), cds.getDataset());
2376 assertTrue(dna.getDataset().getSequences()
2377 .contains(cds.getSequenceAt(0).getDatasetSequence()));
2378 assertTrue(dna.getDataset().getSequences()
2379 .contains(cds.getSequenceAt(1).getDatasetSequence()));
2382 * Verify mappings from CDS to peptide, cDNA to CDS, and cDNA to peptide
2383 * the mappings are on the shared alignment dataset
2385 List<AlignedCodonFrame> cdsMappings = cds.getDataset().getCodonFrames();
2387 * 6 mappings, 2*(DNA->CDS), 2*(DNA->Pep), 2*(CDS->Pep)
2389 assertEquals(6, cdsMappings.size());
2392 * verify that mapping sets for dna and cds alignments are different
2393 * [not current behaviour - all mappings are on the alignment dataset]
2395 // select -> subselect type to test.
2396 // Assert.assertNotSame(dna.getCodonFrames(), cds.getCodonFrames());
2397 // assertEquals(4, dna.getCodonFrames().size());
2398 // assertEquals(4, cds.getCodonFrames().size());
2401 * Two mappings involve pep3 (dna to pep3, cds to pep3)
2402 * Mapping from pep3 to GGGTTT in first new exon sequence
2404 List<AlignedCodonFrame> pep3Mappings = MappingUtils
2405 .findMappingsForSequence(pep3, cdsMappings);
2406 assertEquals(2, pep3Mappings.size());
2407 List<AlignedCodonFrame> mappings = MappingUtils
2408 .findMappingsForSequence(cds.getSequenceAt(0), pep3Mappings);
2409 assertEquals(1, mappings.size());
2412 SearchResultsI sr = MappingUtils.buildSearchResults(pep3, 1, mappings);
2413 assertEquals(1, sr.getResults().size());
2414 SearchResultMatchI m = sr.getResults().get(0);
2415 assertSame(cds.getSequenceAt(0).getDatasetSequence(), m.getSequence());
2416 assertEquals(1, m.getStart());
2417 assertEquals(3, m.getEnd());
2419 sr = MappingUtils.buildSearchResults(pep3, 2, mappings);
2420 m = sr.getResults().get(0);
2421 assertSame(cds.getSequenceAt(0).getDatasetSequence(), m.getSequence());
2422 assertEquals(4, m.getStart());
2423 assertEquals(6, m.getEnd());
2426 * Two mappings involve pep4 (dna to pep4, cds to pep4)
2427 * Verify mapping from pep4 to GGGTTTCCC in second new exon sequence
2429 List<AlignedCodonFrame> pep4Mappings = MappingUtils
2430 .findMappingsForSequence(pep4, cdsMappings);
2431 assertEquals(2, pep4Mappings.size());
2432 mappings = MappingUtils.findMappingsForSequence(cds.getSequenceAt(1),
2434 assertEquals(1, mappings.size());
2436 sr = MappingUtils.buildSearchResults(pep4, 1, mappings);
2437 assertEquals(1, sr.getResults().size());
2438 m = sr.getResults().get(0);
2439 assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
2440 assertEquals(1, m.getStart());
2441 assertEquals(3, m.getEnd());
2443 sr = MappingUtils.buildSearchResults(pep4, 2, mappings);
2444 m = sr.getResults().get(0);
2445 assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
2446 assertEquals(4, m.getStart());
2447 assertEquals(6, m.getEnd());
2449 sr = MappingUtils.buildSearchResults(pep4, 3, mappings);
2450 m = sr.getResults().get(0);
2451 assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
2452 assertEquals(7, m.getStart());
2453 assertEquals(9, m.getEnd());
2457 * Test the method that just copies aligned sequences, provided all sequences
2458 * to be aligned share the aligned sequence's dataset
2460 @Test(groups = "Functional")
2461 public void testAlignAsSameSequences()
2463 SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
2464 SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
2465 AlignmentI al1 = new Alignment(new SequenceI[] { dna1, dna2 });
2466 ((Alignment) al1).createDatasetAlignment();
2468 SequenceI dna3 = new Sequence(dna1);
2469 SequenceI dna4 = new Sequence(dna2);
2470 assertSame(dna3.getDatasetSequence(), dna1.getDatasetSequence());
2471 assertSame(dna4.getDatasetSequence(), dna2.getDatasetSequence());
2472 String seq1 = "-cc-GG-GT-TT--aaa";
2473 dna3.setSequence(seq1);
2474 String seq2 = "C--C-Cgg--gtt-tAA-A-";
2475 dna4.setSequence(seq2);
2476 AlignmentI al2 = new Alignment(new SequenceI[] { dna3, dna4 });
2477 ((Alignment) al2).createDatasetAlignment();
2479 assertTrue(AlignmentUtils.alignAsSameSequences(al1, al2));
2480 assertEquals(seq1, al1.getSequenceAt(0).getSequenceAsString());
2481 assertEquals(seq2, al1.getSequenceAt(1).getSequenceAsString());
2484 * add another sequence to 'aligned' - should still succeed, since
2485 * unaligned sequences still share a dataset with aligned sequences
2487 SequenceI dna5 = new Sequence("dna5", "CCCgggtttAAA");
2488 dna5.createDatasetSequence();
2489 al2.addSequence(dna5);
2490 assertTrue(AlignmentUtils.alignAsSameSequences(al1, al2));
2491 assertEquals(seq1, al1.getSequenceAt(0).getSequenceAsString());
2492 assertEquals(seq2, al1.getSequenceAt(1).getSequenceAsString());
2495 * add another sequence to 'unaligned' - should fail, since now not
2496 * all unaligned sequences share a dataset with aligned sequences
2498 SequenceI dna6 = new Sequence("dna6", "CCCgggtttAAA");
2499 dna6.createDatasetSequence();
2500 al1.addSequence(dna6);
2501 // JAL-2110 JBP Comment: what's the use case for this behaviour ?
2502 assertFalse(AlignmentUtils.alignAsSameSequences(al1, al2));
2505 @Test(groups = "Functional")
2506 public void testAlignAsSameSequencesMultipleSubSeq()
2508 SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
2509 SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
2510 SequenceI as1 = dna1.deriveSequence();
2511 SequenceI as2 = dna1.deriveSequence().getSubSequence(3, 7);
2512 SequenceI as3 = dna2.deriveSequence();
2513 as1.insertCharAt(6, 5, '-');
2514 String s_as1 = as1.getSequenceAsString();
2515 as2.insertCharAt(6, 5, '-');
2516 String s_as2 = as2.getSequenceAsString();
2517 as3.insertCharAt(6, 5, '-');
2518 String s_as3 = as3.getSequenceAsString();
2519 AlignmentI aligned = new Alignment(new SequenceI[] { as1, as2, as3 });
2521 // why do we need to cast this still ?
2522 ((Alignment) aligned).createDatasetAlignment();
2523 SequenceI uas1 = dna1.deriveSequence();
2524 SequenceI uas2 = dna1.deriveSequence().getSubSequence(3, 7);
2525 SequenceI uas3 = dna2.deriveSequence();
2526 AlignmentI tobealigned = new Alignment(new SequenceI[] { uas1, uas2,
2528 ((Alignment) tobealigned).createDatasetAlignment();
2530 assertTrue(AlignmentUtils.alignAsSameSequences(tobealigned, aligned));
2531 assertEquals(s_as1, uas1.getSequenceAsString());
2532 assertEquals(s_as2, uas2.getSequenceAsString());
2533 assertEquals(s_as3, uas3.getSequenceAsString());
2537 * Tests for the method that maps nucleotide to protein based on CDS features
2539 @Test(groups = "Functional")
2540 public void testMapCdsToProtein()
2542 SequenceI peptide = new Sequence("pep", "KLQ");
2545 * Case 1: CDS 3 times length of peptide
2546 * NB method only checks lengths match, not translation
2548 SequenceI dna = new Sequence("dna", "AACGacgtCTCCT");
2549 dna.createDatasetSequence();
2550 dna.addSequenceFeature(new SequenceFeature("CDS", "", 1, 4, null));
2551 dna.addSequenceFeature(new SequenceFeature("CDS", "", 9, 13, null));
2552 MapList ml = AlignmentUtils.mapCdsToProtein(dna, peptide);
2553 assertEquals(3, ml.getFromRatio());
2554 assertEquals(1, ml.getToRatio());
2555 assertEquals("[[1, 3]]",
2556 Arrays.deepToString(ml.getToRanges().toArray()));
2557 assertEquals("[[1, 4], [9, 13]]",
2558 Arrays.deepToString(ml.getFromRanges().toArray()));
2561 * Case 2: CDS 3 times length of peptide + stop codon
2562 * (note code does not currently check trailing codon is a stop codon)
2564 dna = new Sequence("dna", "AACGacgtCTCCTTGA");
2565 dna.createDatasetSequence();
2566 dna.addSequenceFeature(new SequenceFeature("CDS", "", 1, 4, null));
2567 dna.addSequenceFeature(new SequenceFeature("CDS", "", 9, 16, null));
2568 ml = AlignmentUtils.mapCdsToProtein(dna, peptide);
2569 assertEquals(3, ml.getFromRatio());
2570 assertEquals(1, ml.getToRatio());
2571 assertEquals("[[1, 3]]",
2572 Arrays.deepToString(ml.getToRanges().toArray()));
2573 assertEquals("[[1, 4], [9, 13]]",
2574 Arrays.deepToString(ml.getFromRanges().toArray()));
2577 * Case 3: CDS not 3 times length of peptide - no mapping is made
2579 dna = new Sequence("dna", "AACGacgtCTCCTTG");
2580 dna.createDatasetSequence();
2581 dna.addSequenceFeature(new SequenceFeature("CDS", "", 1, 4, null));
2582 dna.addSequenceFeature(new SequenceFeature("CDS", "", 9, 15, null));
2583 ml = AlignmentUtils.mapCdsToProtein(dna, peptide);
2587 * Case 4: incomplete start codon corresponding to X in peptide
2589 dna = new Sequence("dna", "ACGacgtCTCCTTGG");
2590 dna.createDatasetSequence();
2591 SequenceFeature sf = new SequenceFeature("CDS", "", 1, 3, null);
2592 sf.setPhase("2"); // skip 2 positions (AC) to start of next codon (GCT)
2593 dna.addSequenceFeature(sf);
2594 dna.addSequenceFeature(new SequenceFeature("CDS", "", 8, 15, null));
2595 peptide = new Sequence("pep", "XLQ");
2596 ml = AlignmentUtils.mapCdsToProtein(dna, peptide);
2597 assertEquals("[[2, 3]]",
2598 Arrays.deepToString(ml.getToRanges().toArray()));
2599 assertEquals("[[3, 3], [8, 12]]",
2600 Arrays.deepToString(ml.getFromRanges().toArray()));