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.gui.JvOptionPane;
44 import jalview.io.AppletFormatAdapter;
45 import jalview.io.FormatAdapter;
46 import jalview.util.MapList;
47 import jalview.util.MappingUtils;
49 import java.io.IOException;
50 import java.util.ArrayList;
51 import java.util.Arrays;
52 import java.util.LinkedHashMap;
53 import java.util.List;
55 import java.util.TreeMap;
57 import org.testng.annotations.BeforeClass;
58 import org.testng.annotations.Test;
60 public class AlignmentUtilsTests
63 @BeforeClass(alwaysRun = true)
64 public void setUpJvOptionPane()
66 JvOptionPane.setInteractiveMode(false);
67 JvOptionPane.setMockResponse(JvOptionPane.CANCEL_OPTION);
70 public static Sequence ts = new Sequence("short",
71 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklm");
73 @Test(groups = { "Functional" })
74 public void testExpandContext()
76 AlignmentI al = new Alignment(new Sequence[] {});
77 for (int i = 4; i < 14; i += 2)
79 SequenceI s1 = ts.deriveSequence().getSubSequence(i, i + 7);
82 System.out.println(new AppletFormatAdapter().formatSequences("Clustal",
84 for (int flnk = -1; flnk < 25; flnk++)
86 AlignmentI exp = AlignmentUtils.expandContext(al, flnk);
87 System.out.println("\nFlank size: " + flnk);
88 System.out.println(new AppletFormatAdapter().formatSequences(
89 "Clustal", exp, true));
93 * Full expansion to complete sequences
95 for (SequenceI sq : exp.getSequences())
97 String ung = sq.getSequenceAsString().replaceAll("-+", "");
98 final String errorMsg = "Flanking sequence not the same as original dataset sequence.\n"
101 + sq.getDatasetSequence().getSequenceAsString();
102 assertTrue(errorMsg, ung.equalsIgnoreCase(sq.getDatasetSequence()
103 .getSequenceAsString()));
109 * Last sequence is fully expanded, others have leading gaps to match
111 assertTrue(exp.getSequenceAt(4).getSequenceAsString()
113 assertTrue(exp.getSequenceAt(3).getSequenceAsString()
114 .startsWith("--abc"));
115 assertTrue(exp.getSequenceAt(2).getSequenceAsString()
116 .startsWith("----abc"));
117 assertTrue(exp.getSequenceAt(1).getSequenceAsString()
118 .startsWith("------abc"));
119 assertTrue(exp.getSequenceAt(0).getSequenceAsString()
120 .startsWith("--------abc"));
126 * Test that annotations are correctly adjusted by expandContext
128 @Test(groups = { "Functional" })
129 public void testExpandContext_annotation()
131 AlignmentI al = new Alignment(new Sequence[] {});
132 SequenceI ds = new Sequence("Seq1", "ABCDEFGHI");
134 SequenceI seq1 = ds.deriveSequence().getSubSequence(3, 6);
135 al.addSequence(seq1);
138 * Annotate DEF with 4/5/6 respectively
140 Annotation[] anns = new Annotation[] { new Annotation(4),
141 new Annotation(5), new Annotation(6) };
142 AlignmentAnnotation ann = new AlignmentAnnotation("SS",
143 "secondary structure", anns);
144 seq1.addAlignmentAnnotation(ann);
147 * The annotations array should match aligned positions
149 assertEquals(3, ann.annotations.length);
150 assertEquals(4, ann.annotations[0].value, 0.001);
151 assertEquals(5, ann.annotations[1].value, 0.001);
152 assertEquals(6, ann.annotations[2].value, 0.001);
155 * Check annotation to sequence position mappings before expanding the
156 * sequence; these are set up in Sequence.addAlignmentAnnotation ->
157 * Annotation.setSequenceRef -> createSequenceMappings
159 assertNull(ann.getAnnotationForPosition(1));
160 assertNull(ann.getAnnotationForPosition(2));
161 assertNull(ann.getAnnotationForPosition(3));
162 assertEquals(4, ann.getAnnotationForPosition(4).value, 0.001);
163 assertEquals(5, ann.getAnnotationForPosition(5).value, 0.001);
164 assertEquals(6, ann.getAnnotationForPosition(6).value, 0.001);
165 assertNull(ann.getAnnotationForPosition(7));
166 assertNull(ann.getAnnotationForPosition(8));
167 assertNull(ann.getAnnotationForPosition(9));
170 * Expand the subsequence to the full sequence abcDEFghi
172 AlignmentI expanded = AlignmentUtils.expandContext(al, -1);
173 assertEquals("abcDEFghi", expanded.getSequenceAt(0)
174 .getSequenceAsString());
177 * Confirm the alignment and sequence have the same SS annotation,
178 * referencing the expanded sequence
180 ann = expanded.getSequenceAt(0).getAnnotation()[0];
181 assertSame(ann, expanded.getAlignmentAnnotation()[0]);
182 assertSame(expanded.getSequenceAt(0), ann.sequenceRef);
185 * The annotations array should have null values except for annotated
188 assertNull(ann.annotations[0]);
189 assertNull(ann.annotations[1]);
190 assertNull(ann.annotations[2]);
191 assertEquals(4, ann.annotations[3].value, 0.001);
192 assertEquals(5, ann.annotations[4].value, 0.001);
193 assertEquals(6, ann.annotations[5].value, 0.001);
194 assertNull(ann.annotations[6]);
195 assertNull(ann.annotations[7]);
196 assertNull(ann.annotations[8]);
199 * sequence position mappings should be unchanged
201 assertNull(ann.getAnnotationForPosition(1));
202 assertNull(ann.getAnnotationForPosition(2));
203 assertNull(ann.getAnnotationForPosition(3));
204 assertEquals(4, ann.getAnnotationForPosition(4).value, 0.001);
205 assertEquals(5, ann.getAnnotationForPosition(5).value, 0.001);
206 assertEquals(6, ann.getAnnotationForPosition(6).value, 0.001);
207 assertNull(ann.getAnnotationForPosition(7));
208 assertNull(ann.getAnnotationForPosition(8));
209 assertNull(ann.getAnnotationForPosition(9));
213 * Test method that returns a map of lists of sequences by sequence name.
215 * @throws IOException
217 @Test(groups = { "Functional" })
218 public void testGetSequencesByName() throws IOException
220 final String data = ">Seq1Name\nKQYL\n" + ">Seq2Name\nRFPW\n"
221 + ">Seq1Name\nABCD\n";
222 AlignmentI al = loadAlignment(data, "FASTA");
223 Map<String, List<SequenceI>> map = AlignmentUtils
224 .getSequencesByName(al);
225 assertEquals(2, map.keySet().size());
226 assertEquals(2, map.get("Seq1Name").size());
227 assertEquals("KQYL", map.get("Seq1Name").get(0).getSequenceAsString());
228 assertEquals("ABCD", map.get("Seq1Name").get(1).getSequenceAsString());
229 assertEquals(1, map.get("Seq2Name").size());
230 assertEquals("RFPW", map.get("Seq2Name").get(0).getSequenceAsString());
234 * Helper method to load an alignment and ensure dataset sequences are set up.
240 * @throws IOException
242 protected AlignmentI loadAlignment(final String data, String format)
245 AlignmentI a = new FormatAdapter().readFile(data,
246 AppletFormatAdapter.PASTE, format);
252 * Test mapping of protein to cDNA, for the case where we have no sequence
253 * cross-references, so mappings are made first-served 1-1 where sequences
256 * @throws IOException
258 @Test(groups = { "Functional" })
259 public void testMapProteinAlignmentToCdna_noXrefs() throws IOException
261 List<SequenceI> protseqs = new ArrayList<SequenceI>();
262 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
263 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
264 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
265 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
266 protein.setDataset(null);
268 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
269 dnaseqs.add(new Sequence("EMBL|A11111", "TCAGCACGC")); // = SAR
270 dnaseqs.add(new Sequence("EMBL|A22222", "GAGATACAA")); // = EIQ
271 dnaseqs.add(new Sequence("EMBL|A33333", "GAAATCCAG")); // = EIQ
272 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG")); // = EIQ
273 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[4]));
274 cdna.setDataset(null);
276 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
278 // 3 mappings made, each from 1 to 1 sequence
279 assertEquals(3, protein.getCodonFrames().size());
280 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
281 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
282 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
284 // V12345 mapped to A22222
285 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
287 assertEquals(1, acf.getdnaSeqs().length);
288 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
289 acf.getdnaSeqs()[0]);
290 Mapping[] protMappings = acf.getProtMappings();
291 assertEquals(1, protMappings.length);
292 MapList mapList = protMappings[0].getMap();
293 assertEquals(3, mapList.getFromRatio());
294 assertEquals(1, mapList.getToRatio());
295 assertTrue(Arrays.equals(new int[] { 1, 9 }, mapList.getFromRanges()
297 assertEquals(1, mapList.getFromRanges().size());
298 assertTrue(Arrays.equals(new int[] { 1, 3 },
299 mapList.getToRanges().get(0)));
300 assertEquals(1, mapList.getToRanges().size());
302 // V12346 mapped to A33333
303 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
304 assertEquals(1, acf.getdnaSeqs().length);
305 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
306 acf.getdnaSeqs()[0]);
308 // V12347 mapped to A11111
309 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
310 assertEquals(1, acf.getdnaSeqs().length);
311 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
312 acf.getdnaSeqs()[0]);
314 // no mapping involving the 'extra' A44444
315 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(3)).isEmpty());
319 * Test for the alignSequenceAs method that takes two sequences and a mapping.
321 @Test(groups = { "Functional" })
322 public void testAlignSequenceAs_withMapping_noIntrons()
324 MapList map = new MapList(new int[] { 1, 6 }, new int[] { 1, 2 }, 3, 1);
327 * No existing gaps in dna:
329 checkAlignSequenceAs("GGGAAA", "-A-L-", false, false, map,
333 * Now introduce gaps in dna but ignore them when realigning.
335 checkAlignSequenceAs("-G-G-G-A-A-A-", "-A-L-", false, false, map,
339 * Now include gaps in dna when realigning. First retaining 'mapped' gaps
340 * only, i.e. those within the exon region.
342 checkAlignSequenceAs("-G-G--G-A--A-A-", "-A-L-", true, false, map,
343 "---G-G--G---A--A-A");
346 * Include all gaps in dna when realigning (within and without the exon
347 * region). The leading gap, and the gaps between codons, are subsumed by
348 * the protein alignment gap.
350 checkAlignSequenceAs("-G-GG--AA-A---", "-A-L-", true, true, map,
351 "---G-GG---AA-A---");
354 * Include only unmapped gaps in dna when realigning (outside the exon
355 * region). The leading gap, and the gaps between codons, are subsumed by
356 * the protein alignment gap.
358 checkAlignSequenceAs("-G-GG--AA-A-", "-A-L-", false, true, map,
363 * Test for the alignSequenceAs method that takes two sequences and a mapping.
365 @Test(groups = { "Functional" })
366 public void testAlignSequenceAs_withMapping_withIntrons()
369 * Exons at codon 2 (AAA) and 4 (TTT)
371 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
372 new int[] { 1, 2 }, 3, 1);
375 * Simple case: no gaps in dna
377 checkAlignSequenceAs("GGGAAACCCTTTGGG", "--A-L-", false, false, map,
378 "GGG---AAACCCTTTGGG");
381 * Add gaps to dna - but ignore when realigning.
383 checkAlignSequenceAs("-G-G-G--A--A---AC-CC-T-TT-GG-G-", "--A-L-",
384 false, false, map, "GGG---AAACCCTTTGGG");
387 * Add gaps to dna - include within exons only when realigning.
389 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
390 true, false, map, "GGG---A--A---ACCCT-TTGGG");
393 * Include gaps outside exons only when realigning.
395 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
396 false, true, map, "-G-G-GAAAC-CCTTT-GG-G-");
399 * Include gaps following first intron if we are 'preserving mapped gaps'
401 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
402 true, true, map, "-G-G-G--A--A---A-C-CC-T-TT-GG-G-");
405 * Include all gaps in dna when realigning.
407 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
408 true, true, map, "-G-G-G--A--A---A-C-CC-T-TT-GG-G-");
412 * Test for the case where not all of the protein sequence is mapped to cDNA.
414 @Test(groups = { "Functional" })
415 public void testAlignSequenceAs_withMapping_withUnmappedProtein()
418 * Exons at codon 2 (AAA) and 4 (TTT) mapped to A and P
420 final MapList map = new MapList(new int[] { 4, 6, 10, 12 }, new int[] {
424 * -L- 'aligns' ccc------
426 checkAlignSequenceAs("gggAAAcccTTTggg", "-A-L-P-", false, false, map,
427 "gggAAAccc------TTTggg");
431 * Helper method that performs and verifies the method under test.
434 * the sequence to be realigned
436 * the sequence whose alignment is to be copied
437 * @param preserveMappedGaps
438 * @param preserveUnmappedGaps
442 protected void checkAlignSequenceAs(final String alignee,
443 final String alignModel, final boolean preserveMappedGaps,
444 final boolean preserveUnmappedGaps, MapList map,
445 final String expected)
447 SequenceI alignMe = new Sequence("Seq1", alignee);
448 alignMe.createDatasetSequence();
449 SequenceI alignFrom = new Sequence("Seq2", alignModel);
450 alignFrom.createDatasetSequence();
451 AlignedCodonFrame acf = new AlignedCodonFrame();
452 acf.addMap(alignMe.getDatasetSequence(),
453 alignFrom.getDatasetSequence(), map);
455 AlignmentUtils.alignSequenceAs(alignMe, alignFrom, acf, "---", '-',
456 preserveMappedGaps, preserveUnmappedGaps);
457 assertEquals(expected, alignMe.getSequenceAsString());
461 * Test for the alignSequenceAs method where we preserve gaps in introns only.
463 @Test(groups = { "Functional" })
464 public void testAlignSequenceAs_keepIntronGapsOnly()
468 * Intron GGGAAA followed by exon CCCTTT
470 MapList map = new MapList(new int[] { 7, 12 }, new int[] { 1, 2 }, 3, 1);
472 checkAlignSequenceAs("GG-G-AA-A-C-CC-T-TT", "AL", false, true, map,
477 * Test the method that realigns protein to match mapped codon alignment.
479 @Test(groups = { "Functional" })
480 public void testAlignProteinAsDna()
482 // seq1 codons are [1,2,3] [4,5,6] [7,8,9] [10,11,12]
483 SequenceI dna1 = new Sequence("Seq1", "TGCCATTACCAG-");
484 // seq2 codons are [1,3,4] [5,6,7] [8,9,10] [11,12,13]
485 SequenceI dna2 = new Sequence("Seq2", "T-GCCATTACCAG");
486 // seq3 codons are [1,2,3] [4,5,7] [8,9,10] [11,12,13]
487 SequenceI dna3 = new Sequence("Seq3", "TGCCA-TTACCAG");
488 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
489 dna.setDataset(null);
491 // protein alignment will be realigned like dna
492 SequenceI prot1 = new Sequence("Seq1", "CHYQ");
493 SequenceI prot2 = new Sequence("Seq2", "CHYQ");
494 SequenceI prot3 = new Sequence("Seq3", "CHYQ");
495 SequenceI prot4 = new Sequence("Seq4", "R-QSV"); // unmapped, unchanged
496 AlignmentI protein = new Alignment(new SequenceI[] { prot1, prot2,
498 protein.setDataset(null);
500 MapList map = new MapList(new int[] { 1, 12 }, new int[] { 1, 4 }, 3, 1);
501 AlignedCodonFrame acf = new AlignedCodonFrame();
502 acf.addMap(dna1.getDatasetSequence(), prot1.getDatasetSequence(), map);
503 acf.addMap(dna2.getDatasetSequence(), prot2.getDatasetSequence(), map);
504 acf.addMap(dna3.getDatasetSequence(), prot3.getDatasetSequence(), map);
505 ArrayList<AlignedCodonFrame> acfs = new ArrayList<AlignedCodonFrame>();
507 protein.setCodonFrames(acfs);
510 * Translated codon order is [1,2,3] [1,3,4] [4,5,6] [4,5,7] [5,6,7] [7,8,9]
511 * [8,9,10] [10,11,12] [11,12,13]
513 AlignmentUtils.alignProteinAsDna(protein, dna);
514 assertEquals("C-H--Y-Q-", prot1.getSequenceAsString());
515 assertEquals("-C--H-Y-Q", prot2.getSequenceAsString());
516 assertEquals("C--H--Y-Q", prot3.getSequenceAsString());
517 assertEquals("R-QSV", prot4.getSequenceAsString());
521 * Test the method that tests whether a CDNA sequence translates to a protein
524 @Test(groups = { "Functional" })
525 public void testTranslatesAs()
527 // null arguments check
528 assertFalse(AlignmentUtils.translatesAs(null, 0, null));
529 assertFalse(AlignmentUtils.translatesAs(new char[] { 't' }, 0, null));
530 assertFalse(AlignmentUtils.translatesAs(null, 0, new char[] { 'a' }));
532 // straight translation
533 assertTrue(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(), 0,
534 "FPKG".toCharArray()));
535 // with extra start codon (not in protein)
536 assertTrue(AlignmentUtils.translatesAs("atgtttcccaaaggg".toCharArray(),
537 3, "FPKG".toCharArray()));
538 // with stop codon1 (not in protein)
539 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtaa".toCharArray(),
540 0, "FPKG".toCharArray()));
541 // with stop codon1 (in protein as *)
542 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtaa".toCharArray(),
543 0, "FPKG*".toCharArray()));
544 // with stop codon2 (not in protein)
545 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtag".toCharArray(),
546 0, "FPKG".toCharArray()));
547 // with stop codon3 (not in protein)
548 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtga".toCharArray(),
549 0, "FPKG".toCharArray()));
550 // with start and stop codon1
551 assertTrue(AlignmentUtils.translatesAs(
552 "atgtttcccaaagggtaa".toCharArray(), 3, "FPKG".toCharArray()));
553 // with start and stop codon1 (in protein as *)
554 assertTrue(AlignmentUtils.translatesAs(
555 "atgtttcccaaagggtaa".toCharArray(), 3, "FPKG*".toCharArray()));
556 // with start and stop codon2
557 assertTrue(AlignmentUtils.translatesAs(
558 "atgtttcccaaagggtag".toCharArray(), 3, "FPKG".toCharArray()));
559 // with start and stop codon3
560 assertTrue(AlignmentUtils.translatesAs(
561 "atgtttcccaaagggtga".toCharArray(), 3, "FPKG".toCharArray()));
563 // with embedded stop codons
564 assertTrue(AlignmentUtils.translatesAs(
565 "atgtttTAGcccaaaTAAgggtga".toCharArray(), 3,
566 "F*PK*G".toCharArray()));
569 assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(),
570 0, "FPMG".toCharArray()));
573 assertFalse(AlignmentUtils.translatesAs("tttcccaaagg".toCharArray(), 0,
574 "FPKG".toCharArray()));
577 assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(),
578 0, "FPK".toCharArray()));
580 // overlong dna (doesn't end in stop codon)
581 assertFalse(AlignmentUtils.translatesAs(
582 "tttcccaaagggttt".toCharArray(), 0, "FPKG".toCharArray()));
584 // dna + stop codon + more
585 assertFalse(AlignmentUtils.translatesAs(
586 "tttcccaaagggttaga".toCharArray(), 0, "FPKG".toCharArray()));
589 assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(),
590 0, "FPKGQ".toCharArray()));
594 * Test mapping of protein to cDNA, for cases where the cDNA has start and/or
595 * stop codons in addition to the protein coding sequence.
597 * @throws IOException
599 @Test(groups = { "Functional" })
600 public void testMapProteinAlignmentToCdna_withStartAndStopCodons()
603 List<SequenceI> protseqs = new ArrayList<SequenceI>();
604 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
605 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
606 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
607 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
608 protein.setDataset(null);
610 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
612 dnaseqs.add(new Sequence("EMBL|A11111", "ATGTCAGCACGC"));
614 dnaseqs.add(new Sequence("EMBL|A22222", "GAGATACAATAA"));
615 // = start +EIQ + stop
616 dnaseqs.add(new Sequence("EMBL|A33333", "ATGGAAATCCAGTAG"));
617 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG"));
618 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[4]));
619 cdna.setDataset(null);
621 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
623 // 3 mappings made, each from 1 to 1 sequence
624 assertEquals(3, protein.getCodonFrames().size());
625 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
626 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
627 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
629 // V12345 mapped from A22222
630 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
632 assertEquals(1, acf.getdnaSeqs().length);
633 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
634 acf.getdnaSeqs()[0]);
635 Mapping[] protMappings = acf.getProtMappings();
636 assertEquals(1, protMappings.length);
637 MapList mapList = protMappings[0].getMap();
638 assertEquals(3, mapList.getFromRatio());
639 assertEquals(1, mapList.getToRatio());
640 assertTrue(Arrays.equals(new int[] { 1, 9 }, mapList.getFromRanges()
642 assertEquals(1, mapList.getFromRanges().size());
643 assertTrue(Arrays.equals(new int[] { 1, 3 },
644 mapList.getToRanges().get(0)));
645 assertEquals(1, mapList.getToRanges().size());
647 // V12346 mapped from A33333 starting position 4
648 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
649 assertEquals(1, acf.getdnaSeqs().length);
650 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
651 acf.getdnaSeqs()[0]);
652 protMappings = acf.getProtMappings();
653 assertEquals(1, protMappings.length);
654 mapList = protMappings[0].getMap();
655 assertEquals(3, mapList.getFromRatio());
656 assertEquals(1, mapList.getToRatio());
657 assertTrue(Arrays.equals(new int[] { 4, 12 }, mapList.getFromRanges()
659 assertEquals(1, mapList.getFromRanges().size());
660 assertTrue(Arrays.equals(new int[] { 1, 3 },
661 mapList.getToRanges().get(0)));
662 assertEquals(1, mapList.getToRanges().size());
664 // V12347 mapped to A11111 starting position 4
665 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
666 assertEquals(1, acf.getdnaSeqs().length);
667 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
668 acf.getdnaSeqs()[0]);
669 protMappings = acf.getProtMappings();
670 assertEquals(1, protMappings.length);
671 mapList = protMappings[0].getMap();
672 assertEquals(3, mapList.getFromRatio());
673 assertEquals(1, mapList.getToRatio());
674 assertTrue(Arrays.equals(new int[] { 4, 12 }, mapList.getFromRanges()
676 assertEquals(1, mapList.getFromRanges().size());
677 assertTrue(Arrays.equals(new int[] { 1, 3 },
678 mapList.getToRanges().get(0)));
679 assertEquals(1, mapList.getToRanges().size());
681 // no mapping involving the 'extra' A44444
682 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(3)).isEmpty());
686 * Test mapping of protein to cDNA, for the case where we have some sequence
687 * cross-references. Verify that 1-to-many mappings are made where
688 * cross-references exist and sequences are mappable.
690 * @throws IOException
692 @Test(groups = { "Functional" })
693 public void testMapProteinAlignmentToCdna_withXrefs() throws IOException
695 List<SequenceI> protseqs = new ArrayList<SequenceI>();
696 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
697 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
698 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
699 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
700 protein.setDataset(null);
702 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
703 dnaseqs.add(new Sequence("EMBL|A11111", "TCAGCACGC")); // = SAR
704 dnaseqs.add(new Sequence("EMBL|A22222", "ATGGAGATACAA")); // = start + EIQ
705 dnaseqs.add(new Sequence("EMBL|A33333", "GAAATCCAG")); // = EIQ
706 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG")); // = EIQ
707 dnaseqs.add(new Sequence("EMBL|A55555", "GAGATTCAG")); // = EIQ
708 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[5]));
709 cdna.setDataset(null);
711 // Xref A22222 to V12345 (should get mapped)
712 dnaseqs.get(1).addDBRef(new DBRefEntry("UNIPROT", "1", "V12345"));
713 // Xref V12345 to A44444 (should get mapped)
714 protseqs.get(0).addDBRef(new DBRefEntry("EMBL", "1", "A44444"));
715 // Xref A33333 to V12347 (sequence mismatch - should not get mapped)
716 dnaseqs.get(2).addDBRef(new DBRefEntry("UNIPROT", "1", "V12347"));
717 // as V12345 is mapped to A22222 and A44444, this leaves V12346 unmapped.
718 // it should get paired up with the unmapped A33333
719 // A11111 should be mapped to V12347
720 // A55555 is spare and has no xref so is not mapped
722 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
724 // 4 protein mappings made for 3 proteins, 2 to V12345, 1 each to V12346/7
725 assertEquals(3, protein.getCodonFrames().size());
726 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
727 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
728 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
730 // one mapping for each of the first 4 cDNA sequences
731 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(0)).size());
732 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(1)).size());
733 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(2)).size());
734 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(3)).size());
736 // V12345 mapped to A22222 and A44444
737 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
739 assertEquals(2, acf.getdnaSeqs().length);
740 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
741 acf.getdnaSeqs()[0]);
742 assertEquals(cdna.getSequenceAt(3).getDatasetSequence(),
743 acf.getdnaSeqs()[1]);
745 // V12346 mapped to A33333
746 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
747 assertEquals(1, acf.getdnaSeqs().length);
748 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
749 acf.getdnaSeqs()[0]);
751 // V12347 mapped to A11111
752 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
753 assertEquals(1, acf.getdnaSeqs().length);
754 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
755 acf.getdnaSeqs()[0]);
757 // no mapping involving the 'extra' A55555
758 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(4)).isEmpty());
762 * Test mapping of protein to cDNA, for the case where we have some sequence
763 * cross-references. Verify that once we have made an xref mapping we don't
764 * also map un-xrefd sequeces.
766 * @throws IOException
768 @Test(groups = { "Functional" })
769 public void testMapProteinAlignmentToCdna_prioritiseXrefs()
772 List<SequenceI> protseqs = new ArrayList<SequenceI>();
773 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
774 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
775 AlignmentI protein = new Alignment(
776 protseqs.toArray(new SequenceI[protseqs.size()]));
777 protein.setDataset(null);
779 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
780 dnaseqs.add(new Sequence("EMBL|A11111", "GAAATCCAG")); // = EIQ
781 dnaseqs.add(new Sequence("EMBL|A22222", "GAAATTCAG")); // = EIQ
782 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[dnaseqs
784 cdna.setDataset(null);
786 // Xref A22222 to V12345 (should get mapped)
787 // A11111 should then be mapped to the unmapped V12346
788 dnaseqs.get(1).addDBRef(new DBRefEntry("UNIPROT", "1", "V12345"));
790 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
792 // 2 protein mappings made
793 assertEquals(2, protein.getCodonFrames().size());
794 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
795 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
797 // one mapping for each of the cDNA sequences
798 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(0)).size());
799 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(1)).size());
801 // V12345 mapped to A22222
802 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
804 assertEquals(1, acf.getdnaSeqs().length);
805 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
806 acf.getdnaSeqs()[0]);
808 // V12346 mapped to A11111
809 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
810 assertEquals(1, acf.getdnaSeqs().length);
811 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
812 acf.getdnaSeqs()[0]);
816 * Test the method that shows or hides sequence annotations by type(s) and
819 @Test(groups = { "Functional" })
820 public void testShowOrHideSequenceAnnotations()
822 SequenceI seq1 = new Sequence("Seq1", "AAA");
823 SequenceI seq2 = new Sequence("Seq2", "BBB");
824 SequenceI seq3 = new Sequence("Seq3", "CCC");
825 Annotation[] anns = new Annotation[] { new Annotation(2f) };
826 AlignmentAnnotation ann1 = new AlignmentAnnotation("Structure", "ann1",
828 ann1.setSequenceRef(seq1);
829 AlignmentAnnotation ann2 = new AlignmentAnnotation("Structure", "ann2",
831 ann2.setSequenceRef(seq2);
832 AlignmentAnnotation ann3 = new AlignmentAnnotation("Structure", "ann3",
834 AlignmentAnnotation ann4 = new AlignmentAnnotation("Temp", "ann4", anns);
835 ann4.setSequenceRef(seq1);
836 AlignmentAnnotation ann5 = new AlignmentAnnotation("Temp", "ann5", anns);
837 ann5.setSequenceRef(seq2);
838 AlignmentAnnotation ann6 = new AlignmentAnnotation("Temp", "ann6", anns);
839 AlignmentI al = new Alignment(new SequenceI[] { seq1, seq2, seq3 });
840 al.addAnnotation(ann1); // Structure for Seq1
841 al.addAnnotation(ann2); // Structure for Seq2
842 al.addAnnotation(ann3); // Structure for no sequence
843 al.addAnnotation(ann4); // Temp for seq1
844 al.addAnnotation(ann5); // Temp for seq2
845 al.addAnnotation(ann6); // Temp for no sequence
846 List<String> types = new ArrayList<String>();
847 List<SequenceI> scope = new ArrayList<SequenceI>();
850 * Set all sequence related Structure to hidden (ann1, ann2)
852 types.add("Structure");
853 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, false,
855 assertFalse(ann1.visible);
856 assertFalse(ann2.visible);
857 assertTrue(ann3.visible); // not sequence-related, not affected
858 assertTrue(ann4.visible); // not Structure, not affected
859 assertTrue(ann5.visible); // "
860 assertTrue(ann6.visible); // not sequence-related, not affected
863 * Set Temp in {seq1, seq3} to hidden
869 AlignmentUtils.showOrHideSequenceAnnotations(al, types, scope, false,
871 assertFalse(ann1.visible); // unchanged
872 assertFalse(ann2.visible); // unchanged
873 assertTrue(ann3.visible); // not sequence-related, not affected
874 assertFalse(ann4.visible); // Temp for seq1 hidden
875 assertTrue(ann5.visible); // not in scope, not affected
876 assertTrue(ann6.visible); // not sequence-related, not affected
879 * Set Temp in all sequences to hidden
885 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, false,
887 assertFalse(ann1.visible); // unchanged
888 assertFalse(ann2.visible); // unchanged
889 assertTrue(ann3.visible); // not sequence-related, not affected
890 assertFalse(ann4.visible); // Temp for seq1 hidden
891 assertFalse(ann5.visible); // Temp for seq2 hidden
892 assertTrue(ann6.visible); // not sequence-related, not affected
895 * Set all types in {seq1, seq3} to visible
901 AlignmentUtils.showOrHideSequenceAnnotations(al, types, scope, true,
903 assertTrue(ann1.visible); // Structure for seq1 set visible
904 assertFalse(ann2.visible); // not in scope, unchanged
905 assertTrue(ann3.visible); // not sequence-related, not affected
906 assertTrue(ann4.visible); // Temp for seq1 set visible
907 assertFalse(ann5.visible); // not in scope, unchanged
908 assertTrue(ann6.visible); // not sequence-related, not affected
911 * Set all types in all scope to hidden
913 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, true,
915 assertFalse(ann1.visible);
916 assertFalse(ann2.visible);
917 assertTrue(ann3.visible); // not sequence-related, not affected
918 assertFalse(ann4.visible);
919 assertFalse(ann5.visible);
920 assertTrue(ann6.visible); // not sequence-related, not affected
924 * Tests for the method that checks if one sequence cross-references another
926 @Test(groups = { "Functional" })
927 public void testHasCrossRef()
929 assertFalse(AlignmentUtils.hasCrossRef(null, null));
930 SequenceI seq1 = new Sequence("EMBL|A12345", "ABCDEF");
931 assertFalse(AlignmentUtils.hasCrossRef(seq1, null));
932 assertFalse(AlignmentUtils.hasCrossRef(null, seq1));
933 SequenceI seq2 = new Sequence("UNIPROT|V20192", "ABCDEF");
934 assertFalse(AlignmentUtils.hasCrossRef(seq1, seq2));
937 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "v20193"));
938 assertFalse(AlignmentUtils.hasCrossRef(seq1, seq2));
940 // case-insensitive; version number is ignored
941 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "v20192"));
942 assertTrue(AlignmentUtils.hasCrossRef(seq1, seq2));
945 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
946 assertTrue(AlignmentUtils.hasCrossRef(seq1, seq2));
947 // test is one-way only
948 assertFalse(AlignmentUtils.hasCrossRef(seq2, seq1));
952 * Tests for the method that checks if either sequence cross-references the
955 @Test(groups = { "Functional" })
956 public void testHaveCrossRef()
958 assertFalse(AlignmentUtils.hasCrossRef(null, null));
959 SequenceI seq1 = new Sequence("EMBL|A12345", "ABCDEF");
960 assertFalse(AlignmentUtils.haveCrossRef(seq1, null));
961 assertFalse(AlignmentUtils.haveCrossRef(null, seq1));
962 SequenceI seq2 = new Sequence("UNIPROT|V20192", "ABCDEF");
963 assertFalse(AlignmentUtils.haveCrossRef(seq1, seq2));
965 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
966 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
967 // next is true for haveCrossRef, false for hasCrossRef
968 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
970 // now the other way round
971 seq1.setDBRefs(null);
972 seq2.addDBRef(new DBRefEntry("EMBL", "1", "A12345"));
973 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
974 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
977 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
978 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
979 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
983 * Test the method that extracts the cds-only part of a dna alignment.
985 @Test(groups = { "Functional" })
986 public void testMakeCdsAlignment()
990 * dna1 --> [4, 6] [10,12] --> pep1
991 * dna2 --> [1, 3] [7, 9] [13,15] --> pep2
993 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
994 SequenceI dna2 = new Sequence("dna2", "GGGcccTTTaaaCCC");
995 SequenceI pep1 = new Sequence("pep1", "GF");
996 SequenceI pep2 = new Sequence("pep2", "GFP");
997 pep1.addDBRef(new DBRefEntry("UNIPROT", "0", "pep1"));
998 pep2.addDBRef(new DBRefEntry("UNIPROT", "0", "pep2"));
999 dna1.createDatasetSequence();
1000 dna2.createDatasetSequence();
1001 pep1.createDatasetSequence();
1002 pep2.createDatasetSequence();
1003 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
1004 dna.setDataset(null);
1007 * put a variant feature on dna2 base 8
1008 * - should transfer to cds2 base 5
1010 dna2.addSequenceFeature(new SequenceFeature("variant", "hgmd", 8, 8,
1014 * need a sourceDbRef if we are to construct dbrefs to the CDS
1015 * sequence from the dna contig sequences
1017 DBRefEntry dbref = new DBRefEntry("ENSEMBL", "0", "dna1");
1018 dna1.getDatasetSequence().addDBRef(dbref);
1019 org.testng.Assert.assertEquals(dbref, dna1.getPrimaryDBRefs().get(0));
1020 dbref = new DBRefEntry("ENSEMBL", "0", "dna2");
1021 dna2.getDatasetSequence().addDBRef(dbref);
1022 org.testng.Assert.assertEquals(dbref, dna2.getPrimaryDBRefs().get(0));
1025 * CDS sequences are 'discovered' from dna-to-protein mappings on the alignment
1026 * dataset (e.g. added from dbrefs by CrossRef.findXrefSequences)
1028 MapList mapfordna1 = new MapList(new int[] { 4, 6, 10, 12 }, new int[] {
1030 AlignedCodonFrame acf = new AlignedCodonFrame();
1031 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(),
1033 dna.addCodonFrame(acf);
1034 MapList mapfordna2 = new MapList(new int[] { 1, 3, 7, 9, 13, 15 },
1035 new int[] { 1, 3 }, 3, 1);
1036 acf = new AlignedCodonFrame();
1037 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(),
1039 dna.addCodonFrame(acf);
1042 * 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
1044 DBRefEntry dna1xref = new DBRefEntry("UNIPROT", "ENSEMBL", "pep1",
1045 new Mapping(mapfordna1));
1046 dna1.getDatasetSequence().addDBRef(dna1xref);
1047 DBRefEntry dna2xref = new DBRefEntry("UNIPROT", "ENSEMBL", "pep2",
1048 new Mapping(mapfordna2));
1049 dna2.getDatasetSequence().addDBRef(dna2xref);
1052 * execute method under test:
1054 AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
1055 dna1, dna2 }, dna.getDataset(), null);
1058 * verify cds sequences
1060 assertEquals(2, cds.getSequences().size());
1061 assertEquals("GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
1062 assertEquals("GGGTTTCCC", cds.getSequenceAt(1).getSequenceAsString());
1065 * verify shared, extended alignment dataset
1067 assertSame(dna.getDataset(), cds.getDataset());
1068 SequenceI cds1Dss = cds.getSequenceAt(0).getDatasetSequence();
1069 SequenceI cds2Dss = cds.getSequenceAt(1).getDatasetSequence();
1070 assertTrue(dna.getDataset().getSequences().contains(cds1Dss));
1071 assertTrue(dna.getDataset().getSequences().contains(cds2Dss));
1074 * verify CDS has a dbref with mapping to peptide
1076 assertNotNull(cds1Dss.getDBRefs());
1077 assertEquals(2, cds1Dss.getDBRefs().length);
1078 dbref = cds1Dss.getDBRefs()[0];
1079 assertEquals(dna1xref.getSource(), dbref.getSource());
1080 // version is via ensembl's primary ref
1081 assertEquals(dna1xref.getVersion(), dbref.getVersion());
1082 assertEquals(dna1xref.getAccessionId(), dbref.getAccessionId());
1083 assertNotNull(dbref.getMap());
1084 assertSame(pep1.getDatasetSequence(), dbref.getMap().getTo());
1085 MapList cdsMapping = new MapList(new int[] { 1, 6 },
1086 new int[] { 1, 2 }, 3, 1);
1087 assertEquals(cdsMapping, dbref.getMap().getMap());
1090 * verify peptide has added a dbref with reverse mapping to CDS
1092 assertNotNull(pep1.getDBRefs());
1093 // FIXME pep1.getDBRefs() is 1 - is that the correct behaviour ?
1094 assertEquals(2, pep1.getDBRefs().length);
1095 dbref = pep1.getDBRefs()[1];
1096 assertEquals("ENSEMBL", dbref.getSource());
1097 assertEquals("0", dbref.getVersion());
1098 assertEquals("CDS|dna1", dbref.getAccessionId());
1099 assertNotNull(dbref.getMap());
1100 assertSame(cds1Dss, dbref.getMap().getTo());
1101 assertEquals(cdsMapping.getInverse(), dbref.getMap().getMap());
1104 * Verify mappings from CDS to peptide, cDNA to CDS, and cDNA to peptide
1105 * the mappings are on the shared alignment dataset
1106 * 6 mappings, 2*(DNA->CDS), 2*(DNA->Pep), 2*(CDS->Pep)
1108 List<AlignedCodonFrame> cdsMappings = cds.getDataset().getCodonFrames();
1109 assertEquals(6, cdsMappings.size());
1112 * verify that mapping sets for dna and cds alignments are different
1113 * [not current behaviour - all mappings are on the alignment dataset]
1115 // select -> subselect type to test.
1116 // Assert.assertNotSame(dna.getCodonFrames(), cds.getCodonFrames());
1117 // assertEquals(4, dna.getCodonFrames().size());
1118 // assertEquals(4, cds.getCodonFrames().size());
1121 * Two mappings involve pep1 (dna to pep1, cds to pep1)
1122 * Mapping from pep1 to GGGTTT in first new exon sequence
1124 List<AlignedCodonFrame> pep1Mappings = MappingUtils
1125 .findMappingsForSequence(pep1, cdsMappings);
1126 assertEquals(2, pep1Mappings.size());
1127 List<AlignedCodonFrame> mappings = MappingUtils
1128 .findMappingsForSequence(cds.getSequenceAt(0), pep1Mappings);
1129 assertEquals(1, mappings.size());
1132 SearchResultsI sr = MappingUtils.buildSearchResults(pep1, 1, mappings);
1133 assertEquals(1, sr.getResults().size());
1134 SearchResultMatchI m = sr.getResults().get(0);
1135 assertSame(cds1Dss, m.getSequence());
1136 assertEquals(1, m.getStart());
1137 assertEquals(3, m.getEnd());
1139 sr = MappingUtils.buildSearchResults(pep1, 2, mappings);
1140 m = sr.getResults().get(0);
1141 assertSame(cds1Dss, m.getSequence());
1142 assertEquals(4, m.getStart());
1143 assertEquals(6, m.getEnd());
1146 * Two mappings involve pep2 (dna to pep2, cds to pep2)
1147 * Verify mapping from pep2 to GGGTTTCCC in second new exon sequence
1149 List<AlignedCodonFrame> pep2Mappings = MappingUtils
1150 .findMappingsForSequence(pep2, cdsMappings);
1151 assertEquals(2, pep2Mappings.size());
1152 mappings = MappingUtils.findMappingsForSequence(cds.getSequenceAt(1),
1154 assertEquals(1, mappings.size());
1156 sr = MappingUtils.buildSearchResults(pep2, 1, mappings);
1157 assertEquals(1, sr.getResults().size());
1158 m = sr.getResults().get(0);
1159 assertSame(cds2Dss, m.getSequence());
1160 assertEquals(1, m.getStart());
1161 assertEquals(3, m.getEnd());
1163 sr = MappingUtils.buildSearchResults(pep2, 2, mappings);
1164 m = sr.getResults().get(0);
1165 assertSame(cds2Dss, m.getSequence());
1166 assertEquals(4, m.getStart());
1167 assertEquals(6, m.getEnd());
1169 sr = MappingUtils.buildSearchResults(pep2, 3, mappings);
1170 m = sr.getResults().get(0);
1171 assertSame(cds2Dss, m.getSequence());
1172 assertEquals(7, m.getStart());
1173 assertEquals(9, m.getEnd());
1176 * check cds2 acquired a variant feature in position 5
1178 SequenceFeature[] sfs = cds2Dss.getSequenceFeatures();
1180 assertEquals(1, sfs.length);
1181 assertEquals("variant", sfs[0].type);
1182 assertEquals(5, sfs[0].begin);
1183 assertEquals(5, sfs[0].end);
1187 * Test the method that makes a cds-only alignment from a DNA sequence and its
1188 * product mappings, for the case where there are multiple exon mappings to
1189 * different protein products.
1191 @Test(groups = { "Functional" })
1192 public void testMakeCdsAlignment_multipleProteins()
1194 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
1195 SequenceI pep1 = new Sequence("pep1", "GF"); // GGGTTT
1196 SequenceI pep2 = new Sequence("pep2", "KP"); // aaaccc
1197 SequenceI pep3 = new Sequence("pep3", "KF"); // aaaTTT
1198 dna1.createDatasetSequence();
1199 pep1.createDatasetSequence();
1200 pep2.createDatasetSequence();
1201 pep3.createDatasetSequence();
1202 pep1.getDatasetSequence().addDBRef(
1203 new DBRefEntry("EMBLCDS", "2", "A12345"));
1204 pep2.getDatasetSequence().addDBRef(
1205 new DBRefEntry("EMBLCDS", "3", "A12346"));
1206 pep3.getDatasetSequence().addDBRef(
1207 new DBRefEntry("EMBLCDS", "4", "A12347"));
1210 * Create the CDS alignment
1212 AlignmentI dna = new Alignment(new SequenceI[] { dna1 });
1213 dna.setDataset(null);
1216 * Make the mappings from dna to protein
1218 // map ...GGG...TTT to GF
1219 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1220 new int[] { 1, 2 }, 3, 1);
1221 AlignedCodonFrame acf = new AlignedCodonFrame();
1222 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
1223 dna.addCodonFrame(acf);
1225 // map aaa...ccc to KP
1226 map = new MapList(new int[] { 1, 3, 7, 9 }, new int[] { 1, 2 }, 3, 1);
1227 acf = new AlignedCodonFrame();
1228 acf.addMap(dna1.getDatasetSequence(), pep2.getDatasetSequence(), map);
1229 dna.addCodonFrame(acf);
1231 // map aaa......TTT to KF
1232 map = new MapList(new int[] { 1, 3, 10, 12 }, new int[] { 1, 2 }, 3, 1);
1233 acf = new AlignedCodonFrame();
1234 acf.addMap(dna1.getDatasetSequence(), pep3.getDatasetSequence(), map);
1235 dna.addCodonFrame(acf);
1238 * execute method under test
1240 AlignmentI cdsal = AlignmentUtils.makeCdsAlignment(
1241 new SequenceI[] { dna1 }, dna.getDataset(), null);
1244 * Verify we have 3 cds sequences, mapped to pep1/2/3 respectively
1246 List<SequenceI> cds = cdsal.getSequences();
1247 assertEquals(3, cds.size());
1250 * verify shared, extended alignment dataset
1252 assertSame(cdsal.getDataset(), dna.getDataset());
1253 assertTrue(dna.getDataset().getSequences()
1254 .contains(cds.get(0).getDatasetSequence()));
1255 assertTrue(dna.getDataset().getSequences()
1256 .contains(cds.get(1).getDatasetSequence()));
1257 assertTrue(dna.getDataset().getSequences()
1258 .contains(cds.get(2).getDatasetSequence()));
1261 * verify aligned cds sequences and their xrefs
1263 SequenceI cdsSeq = cds.get(0);
1264 assertEquals("GGGTTT", cdsSeq.getSequenceAsString());
1265 // assertEquals("dna1|A12345", cdsSeq.getName());
1266 assertEquals("CDS|dna1", cdsSeq.getName());
1267 // assertEquals(1, cdsSeq.getDBRefs().length);
1268 // DBRefEntry cdsRef = cdsSeq.getDBRefs()[0];
1269 // assertEquals("EMBLCDS", cdsRef.getSource());
1270 // assertEquals("2", cdsRef.getVersion());
1271 // assertEquals("A12345", cdsRef.getAccessionId());
1273 cdsSeq = cds.get(1);
1274 assertEquals("aaaccc", cdsSeq.getSequenceAsString());
1275 // assertEquals("dna1|A12346", cdsSeq.getName());
1276 assertEquals("CDS|dna1", cdsSeq.getName());
1277 // assertEquals(1, cdsSeq.getDBRefs().length);
1278 // cdsRef = cdsSeq.getDBRefs()[0];
1279 // assertEquals("EMBLCDS", cdsRef.getSource());
1280 // assertEquals("3", cdsRef.getVersion());
1281 // assertEquals("A12346", cdsRef.getAccessionId());
1283 cdsSeq = cds.get(2);
1284 assertEquals("aaaTTT", cdsSeq.getSequenceAsString());
1285 // assertEquals("dna1|A12347", cdsSeq.getName());
1286 assertEquals("CDS|dna1", cdsSeq.getName());
1287 // assertEquals(1, cdsSeq.getDBRefs().length);
1288 // cdsRef = cdsSeq.getDBRefs()[0];
1289 // assertEquals("EMBLCDS", cdsRef.getSource());
1290 // assertEquals("4", cdsRef.getVersion());
1291 // assertEquals("A12347", cdsRef.getAccessionId());
1294 * Verify there are mappings from each cds sequence to its protein product
1295 * and also to its dna source
1297 List<AlignedCodonFrame> newMappings = cdsal.getCodonFrames();
1300 * 6 mappings involve dna1 (to pep1/2/3, cds1/2/3)
1302 List<AlignedCodonFrame> dnaMappings = MappingUtils
1303 .findMappingsForSequence(dna1, newMappings);
1304 assertEquals(6, dnaMappings.size());
1309 List<AlignedCodonFrame> mappings = MappingUtils
1310 .findMappingsForSequence(pep1, dnaMappings);
1311 assertEquals(1, mappings.size());
1312 assertEquals(1, mappings.get(0).getMappings().size());
1313 assertSame(pep1.getDatasetSequence(), mappings.get(0).getMappings()
1314 .get(0).getMapping().getTo());
1319 List<AlignedCodonFrame> dnaToCds1Mappings = MappingUtils
1320 .findMappingsForSequence(cds.get(0), dnaMappings);
1321 Mapping mapping = dnaToCds1Mappings.get(0).getMappings().get(0)
1323 assertSame(cds.get(0).getDatasetSequence(), mapping.getTo());
1324 assertEquals("G(1) in CDS should map to G(4) in DNA", 4, mapping
1325 .getMap().getToPosition(1));
1330 mappings = MappingUtils.findMappingsForSequence(pep2, dnaMappings);
1331 assertEquals(1, mappings.size());
1332 assertEquals(1, mappings.get(0).getMappings().size());
1333 assertSame(pep2.getDatasetSequence(), mappings.get(0).getMappings()
1334 .get(0).getMapping().getTo());
1339 List<AlignedCodonFrame> dnaToCds2Mappings = MappingUtils
1340 .findMappingsForSequence(cds.get(1), dnaMappings);
1341 mapping = dnaToCds2Mappings.get(0).getMappings().get(0).getMapping();
1342 assertSame(cds.get(1).getDatasetSequence(), mapping.getTo());
1343 assertEquals("c(4) in CDS should map to c(7) in DNA", 7, mapping
1344 .getMap().getToPosition(4));
1349 mappings = MappingUtils.findMappingsForSequence(pep3, dnaMappings);
1350 assertEquals(1, mappings.size());
1351 assertEquals(1, mappings.get(0).getMappings().size());
1352 assertSame(pep3.getDatasetSequence(), mappings.get(0).getMappings()
1353 .get(0).getMapping().getTo());
1358 List<AlignedCodonFrame> dnaToCds3Mappings = MappingUtils
1359 .findMappingsForSequence(cds.get(2), dnaMappings);
1360 mapping = dnaToCds3Mappings.get(0).getMappings().get(0).getMapping();
1361 assertSame(cds.get(2).getDatasetSequence(), mapping.getTo());
1362 assertEquals("T(4) in CDS should map to T(10) in DNA", 10, mapping
1363 .getMap().getToPosition(4));
1366 @Test(groups = { "Functional" })
1367 public void testIsMappable()
1369 SequenceI dna1 = new Sequence("dna1", "cgCAGtgGT");
1370 SequenceI aa1 = new Sequence("aa1", "RSG");
1371 AlignmentI al1 = new Alignment(new SequenceI[] { dna1 });
1372 AlignmentI al2 = new Alignment(new SequenceI[] { aa1 });
1374 assertFalse(AlignmentUtils.isMappable(null, null));
1375 assertFalse(AlignmentUtils.isMappable(al1, null));
1376 assertFalse(AlignmentUtils.isMappable(null, al1));
1377 assertFalse(AlignmentUtils.isMappable(al1, al1));
1378 assertFalse(AlignmentUtils.isMappable(al2, al2));
1380 assertTrue(AlignmentUtils.isMappable(al1, al2));
1381 assertTrue(AlignmentUtils.isMappable(al2, al1));
1385 * Test creating a mapping when the sequences involved do not start at residue
1388 * @throws IOException
1390 @Test(groups = { "Functional" })
1391 public void testMapCdnaToProtein_forSubsequence() throws IOException
1393 SequenceI prot = new Sequence("UNIPROT|V12345", "E-I--Q", 10, 12);
1394 prot.createDatasetSequence();
1396 SequenceI dna = new Sequence("EMBL|A33333", "GAA--AT-C-CAG", 40, 48);
1397 dna.createDatasetSequence();
1399 MapList map = AlignmentUtils.mapCdnaToProtein(prot, dna);
1400 assertEquals(10, map.getToLowest());
1401 assertEquals(12, map.getToHighest());
1402 assertEquals(40, map.getFromLowest());
1403 assertEquals(48, map.getFromHighest());
1407 * Test for the alignSequenceAs method where we have protein mapped to protein
1409 @Test(groups = { "Functional" })
1410 public void testAlignSequenceAs_mappedProteinProtein()
1413 SequenceI alignMe = new Sequence("Match", "MGAASEV");
1414 alignMe.createDatasetSequence();
1415 SequenceI alignFrom = new Sequence("Query", "LQTGYMGAASEVMFSPTRR");
1416 alignFrom.createDatasetSequence();
1418 AlignedCodonFrame acf = new AlignedCodonFrame();
1419 // this is like a domain or motif match of part of a peptide sequence
1420 MapList map = new MapList(new int[] { 6, 12 }, new int[] { 1, 7 }, 1, 1);
1421 acf.addMap(alignFrom.getDatasetSequence(),
1422 alignMe.getDatasetSequence(), map);
1424 AlignmentUtils.alignSequenceAs(alignMe, alignFrom, acf, "-", '-', true,
1426 assertEquals("-----MGAASEV-------", alignMe.getSequenceAsString());
1430 * Test for the alignSequenceAs method where there are trailing unmapped
1431 * residues in the model sequence
1433 @Test(groups = { "Functional" })
1434 public void testAlignSequenceAs_withTrailingPeptide()
1436 // map first 3 codons to KPF; G is a trailing unmapped residue
1437 MapList map = new MapList(new int[] { 1, 9 }, new int[] { 1, 3 }, 3, 1);
1439 checkAlignSequenceAs("AAACCCTTT", "K-PFG", true, true, map,
1444 * Tests for transferring features between mapped sequences
1446 @Test(groups = { "Functional" })
1447 public void testTransferFeatures()
1449 SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
1450 SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
1453 dna.addSequenceFeature(new SequenceFeature("type1", "desc1", 1, 2, 1f,
1455 // partial overlap - to [1, 1]
1456 dna.addSequenceFeature(new SequenceFeature("type2", "desc2", 3, 4, 2f,
1458 // exact overlap - to [1, 3]
1459 dna.addSequenceFeature(new SequenceFeature("type3", "desc3", 4, 6, 3f,
1461 // spanning overlap - to [2, 5]
1462 dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
1464 // exactly overlaps whole mapped range [1, 6]
1465 dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
1467 // no overlap (internal)
1468 dna.addSequenceFeature(new SequenceFeature("type6", "desc6", 7, 9, 6f,
1470 // no overlap (3' end)
1471 dna.addSequenceFeature(new SequenceFeature("type7", "desc7", 13, 15,
1473 // overlap (3' end) - to [6, 6]
1474 dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
1476 // extended overlap - to [6, +]
1477 dna.addSequenceFeature(new SequenceFeature("type9", "desc9", 12, 13,
1480 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1481 new int[] { 1, 6 }, 1, 1);
1484 * transferFeatures() will build 'partial overlap' for regions
1485 * that partially overlap 5' or 3' (start or end) of target sequence
1487 AlignmentUtils.transferFeatures(dna, cds, map, null);
1488 SequenceFeature[] sfs = cds.getSequenceFeatures();
1489 assertEquals(6, sfs.length);
1491 SequenceFeature sf = sfs[0];
1492 assertEquals("type2", sf.getType());
1493 assertEquals("desc2", sf.getDescription());
1494 assertEquals(2f, sf.getScore());
1495 assertEquals(1, sf.getBegin());
1496 assertEquals(1, sf.getEnd());
1499 assertEquals("type3", sf.getType());
1500 assertEquals("desc3", sf.getDescription());
1501 assertEquals(3f, sf.getScore());
1502 assertEquals(1, sf.getBegin());
1503 assertEquals(3, sf.getEnd());
1506 assertEquals("type4", sf.getType());
1507 assertEquals(2, sf.getBegin());
1508 assertEquals(5, sf.getEnd());
1511 assertEquals("type5", sf.getType());
1512 assertEquals(1, sf.getBegin());
1513 assertEquals(6, sf.getEnd());
1516 assertEquals("type8", sf.getType());
1517 assertEquals(6, sf.getBegin());
1518 assertEquals(6, sf.getEnd());
1521 assertEquals("type9", sf.getType());
1522 assertEquals(6, sf.getBegin());
1523 assertEquals(6, sf.getEnd());
1527 * Tests for transferring features between mapped sequences
1529 @Test(groups = { "Functional" })
1530 public void testTransferFeatures_withOmit()
1532 SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
1533 SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
1535 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1536 new int[] { 1, 6 }, 1, 1);
1538 // [5, 11] maps to [2, 5]
1539 dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
1541 // [4, 12] maps to [1, 6]
1542 dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
1544 // [12, 12] maps to [6, 6]
1545 dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
1548 // desc4 and desc8 are the 'omit these' varargs
1549 AlignmentUtils.transferFeatures(dna, cds, map, null, "type4", "type8");
1550 SequenceFeature[] sfs = cds.getSequenceFeatures();
1551 assertEquals(1, sfs.length);
1553 SequenceFeature sf = sfs[0];
1554 assertEquals("type5", sf.getType());
1555 assertEquals(1, sf.getBegin());
1556 assertEquals(6, sf.getEnd());
1560 * Tests for transferring features between mapped sequences
1562 @Test(groups = { "Functional" })
1563 public void testTransferFeatures_withSelect()
1565 SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
1566 SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
1568 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1569 new int[] { 1, 6 }, 1, 1);
1571 // [5, 11] maps to [2, 5]
1572 dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
1574 // [4, 12] maps to [1, 6]
1575 dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
1577 // [12, 12] maps to [6, 6]
1578 dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
1581 // "type5" is the 'select this type' argument
1582 AlignmentUtils.transferFeatures(dna, cds, map, "type5");
1583 SequenceFeature[] sfs = cds.getSequenceFeatures();
1584 assertEquals(1, sfs.length);
1586 SequenceFeature sf = sfs[0];
1587 assertEquals("type5", sf.getType());
1588 assertEquals(1, sf.getBegin());
1589 assertEquals(6, sf.getEnd());
1593 * Test the method that extracts the cds-only part of a dna alignment, for the
1594 * case where the cds should be aligned to match its nucleotide sequence.
1596 @Test(groups = { "Functional" })
1597 public void testMakeCdsAlignment_alternativeTranscripts()
1599 SequenceI dna1 = new Sequence("dna1", "aaaGGGCC-----CTTTaaaGGG");
1600 // alternative transcript of same dna skips CCC codon
1601 SequenceI dna2 = new Sequence("dna2", "aaaGGGCC-----cttTaaaGGG");
1602 // dna3 has no mapping (protein product) so should be ignored here
1603 SequenceI dna3 = new Sequence("dna3", "aaaGGGCCCCCGGGcttTaaaGGG");
1604 SequenceI pep1 = new Sequence("pep1", "GPFG");
1605 SequenceI pep2 = new Sequence("pep2", "GPG");
1606 dna1.createDatasetSequence();
1607 dna2.createDatasetSequence();
1608 dna3.createDatasetSequence();
1609 pep1.createDatasetSequence();
1610 pep2.createDatasetSequence();
1612 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
1613 dna.setDataset(null);
1615 MapList map = new MapList(new int[] { 4, 12, 16, 18 },
1616 new int[] { 1, 4 }, 3, 1);
1617 AlignedCodonFrame acf = new AlignedCodonFrame();
1618 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
1619 dna.addCodonFrame(acf);
1620 map = new MapList(new int[] { 4, 8, 12, 12, 16, 18 },
1621 new int[] { 1, 3 }, 3, 1);
1622 acf = new AlignedCodonFrame();
1623 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map);
1624 dna.addCodonFrame(acf);
1626 AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
1627 dna1, dna2, dna3 }, dna.getDataset(), null);
1628 List<SequenceI> cdsSeqs = cds.getSequences();
1629 assertEquals(2, cdsSeqs.size());
1630 assertEquals("GGGCCCTTTGGG", cdsSeqs.get(0).getSequenceAsString());
1631 assertEquals("GGGCCTGGG", cdsSeqs.get(1).getSequenceAsString());
1634 * verify shared, extended alignment dataset
1636 assertSame(dna.getDataset(), cds.getDataset());
1637 assertTrue(dna.getDataset().getSequences()
1638 .contains(cdsSeqs.get(0).getDatasetSequence()));
1639 assertTrue(dna.getDataset().getSequences()
1640 .contains(cdsSeqs.get(1).getDatasetSequence()));
1643 * Verify 6 mappings: dna1 to cds1, cds1 to pep1, dna1 to pep1
1644 * and the same for dna2/cds2/pep2
1646 List<AlignedCodonFrame> mappings = cds.getCodonFrames();
1647 assertEquals(6, mappings.size());
1650 * 2 mappings involve pep1
1652 List<AlignedCodonFrame> pep1Mappings = MappingUtils
1653 .findMappingsForSequence(pep1, mappings);
1654 assertEquals(2, pep1Mappings.size());
1657 * Get mapping of pep1 to cds1 and verify it
1658 * maps GPFG to 1-3,4-6,7-9,10-12
1660 List<AlignedCodonFrame> pep1CdsMappings = MappingUtils
1661 .findMappingsForSequence(cds.getSequenceAt(0), pep1Mappings);
1662 assertEquals(1, pep1CdsMappings.size());
1663 SearchResultsI sr = MappingUtils.buildSearchResults(pep1, 1,
1665 assertEquals(1, sr.getResults().size());
1666 SearchResultMatchI m = sr.getResults().get(0);
1667 assertEquals(cds.getSequenceAt(0).getDatasetSequence(), m.getSequence());
1668 assertEquals(1, m.getStart());
1669 assertEquals(3, m.getEnd());
1670 sr = MappingUtils.buildSearchResults(pep1, 2, pep1CdsMappings);
1671 m = sr.getResults().get(0);
1672 assertEquals(4, m.getStart());
1673 assertEquals(6, m.getEnd());
1674 sr = MappingUtils.buildSearchResults(pep1, 3, pep1CdsMappings);
1675 m = sr.getResults().get(0);
1676 assertEquals(7, m.getStart());
1677 assertEquals(9, m.getEnd());
1678 sr = MappingUtils.buildSearchResults(pep1, 4, pep1CdsMappings);
1679 m = sr.getResults().get(0);
1680 assertEquals(10, m.getStart());
1681 assertEquals(12, m.getEnd());
1684 * Get mapping of pep2 to cds2 and verify it
1685 * maps GPG in pep2 to 1-3,4-6,7-9 in second CDS sequence
1687 List<AlignedCodonFrame> pep2Mappings = MappingUtils
1688 .findMappingsForSequence(pep2, mappings);
1689 assertEquals(2, pep2Mappings.size());
1690 List<AlignedCodonFrame> pep2CdsMappings = MappingUtils
1691 .findMappingsForSequence(cds.getSequenceAt(1), pep2Mappings);
1692 assertEquals(1, pep2CdsMappings.size());
1693 sr = MappingUtils.buildSearchResults(pep2, 1, pep2CdsMappings);
1694 assertEquals(1, sr.getResults().size());
1695 m = sr.getResults().get(0);
1696 assertEquals(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
1697 assertEquals(1, m.getStart());
1698 assertEquals(3, m.getEnd());
1699 sr = MappingUtils.buildSearchResults(pep2, 2, pep2CdsMappings);
1700 m = sr.getResults().get(0);
1701 assertEquals(4, m.getStart());
1702 assertEquals(6, m.getEnd());
1703 sr = MappingUtils.buildSearchResults(pep2, 3, pep2CdsMappings);
1704 m = sr.getResults().get(0);
1705 assertEquals(7, m.getStart());
1706 assertEquals(9, m.getEnd());
1710 * Test the method that realigns protein to match mapped codon alignment.
1712 @Test(groups = { "Functional" })
1713 public void testAlignProteinAsDna_incompleteStartCodon()
1715 // seq1: incomplete start codon (not mapped), then [3, 11]
1716 SequenceI dna1 = new Sequence("Seq1", "ccAAA-TTT-GGG-");
1717 // seq2 codons are [4, 5], [8, 11]
1718 SequenceI dna2 = new Sequence("Seq2", "ccaAA-ttT-GGG-");
1719 // seq3 incomplete start codon at 'tt'
1720 SequenceI dna3 = new Sequence("Seq3", "ccaaa-ttt-GGG-");
1721 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
1722 dna.setDataset(null);
1724 // prot1 has 'X' for incomplete start codon (not mapped)
1725 SequenceI prot1 = new Sequence("Seq1", "XKFG"); // X for incomplete start
1726 SequenceI prot2 = new Sequence("Seq2", "NG");
1727 SequenceI prot3 = new Sequence("Seq3", "XG"); // X for incomplete start
1728 AlignmentI protein = new Alignment(new SequenceI[] { prot1, prot2,
1730 protein.setDataset(null);
1732 // map dna1 [3, 11] to prot1 [2, 4] KFG
1733 MapList map = new MapList(new int[] { 3, 11 }, new int[] { 2, 4 }, 3, 1);
1734 AlignedCodonFrame acf = new AlignedCodonFrame();
1735 acf.addMap(dna1.getDatasetSequence(), prot1.getDatasetSequence(), map);
1737 // map dna2 [4, 5] [8, 11] to prot2 [1, 2] NG
1738 map = new MapList(new int[] { 4, 5, 8, 11 }, new int[] { 1, 2 }, 3, 1);
1739 acf.addMap(dna2.getDatasetSequence(), prot2.getDatasetSequence(), map);
1741 // map dna3 [9, 11] to prot3 [2, 2] G
1742 map = new MapList(new int[] { 9, 11 }, new int[] { 2, 2 }, 3, 1);
1743 acf.addMap(dna3.getDatasetSequence(), prot3.getDatasetSequence(), map);
1745 ArrayList<AlignedCodonFrame> acfs = new ArrayList<AlignedCodonFrame>();
1747 protein.setCodonFrames(acfs);
1750 * verify X is included in the aligned proteins, and placed just
1751 * before the first mapped residue
1752 * CCT is between CCC and TTT
1754 AlignmentUtils.alignProteinAsDna(protein, dna);
1755 assertEquals("XK-FG", prot1.getSequenceAsString());
1756 assertEquals("--N-G", prot2.getSequenceAsString());
1757 assertEquals("---XG", prot3.getSequenceAsString());
1761 * Tests for the method that maps the subset of a dna sequence that has CDS
1762 * (or subtype) feature - case where the start codon is incomplete.
1764 @Test(groups = "Functional")
1765 public void testFindCdsPositions_fivePrimeIncomplete()
1767 SequenceI dnaSeq = new Sequence("dna", "aaagGGCCCaaaTTTttt");
1768 dnaSeq.createDatasetSequence();
1769 SequenceI ds = dnaSeq.getDatasetSequence();
1771 // CDS for dna 5-6 (incomplete codon), 7-9
1772 SequenceFeature sf = new SequenceFeature("CDS", "", 5, 9, 0f, null);
1773 sf.setPhase("2"); // skip 2 bases to start of next codon
1774 ds.addSequenceFeature(sf);
1775 // CDS for dna 13-15
1776 sf = new SequenceFeature("CDS_predicted", "", 13, 15, 0f, null);
1777 ds.addSequenceFeature(sf);
1779 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
1782 * check the mapping starts with the first complete codon
1784 assertEquals(6, MappingUtils.getLength(ranges));
1785 assertEquals(2, ranges.size());
1786 assertEquals(7, ranges.get(0)[0]);
1787 assertEquals(9, ranges.get(0)[1]);
1788 assertEquals(13, ranges.get(1)[0]);
1789 assertEquals(15, ranges.get(1)[1]);
1793 * Tests for the method that maps the subset of a dna sequence that has CDS
1794 * (or subtype) feature.
1796 @Test(groups = "Functional")
1797 public void testFindCdsPositions()
1799 SequenceI dnaSeq = new Sequence("dna", "aaaGGGcccAAATTTttt");
1800 dnaSeq.createDatasetSequence();
1801 SequenceI ds = dnaSeq.getDatasetSequence();
1803 // CDS for dna 10-12
1804 SequenceFeature sf = new SequenceFeature("CDS_predicted", "", 10, 12,
1807 ds.addSequenceFeature(sf);
1809 sf = new SequenceFeature("CDS", "", 4, 6, 0f, null);
1811 ds.addSequenceFeature(sf);
1812 // exon feature should be ignored here
1813 sf = new SequenceFeature("exon", "", 7, 9, 0f, null);
1814 ds.addSequenceFeature(sf);
1816 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
1818 * verify ranges { [4-6], [12-10] }
1819 * note CDS ranges are ordered ascending even if the CDS
1822 assertEquals(6, MappingUtils.getLength(ranges));
1823 assertEquals(2, ranges.size());
1824 assertEquals(4, ranges.get(0)[0]);
1825 assertEquals(6, ranges.get(0)[1]);
1826 assertEquals(10, ranges.get(1)[0]);
1827 assertEquals(12, ranges.get(1)[1]);
1831 * Test the method that computes a map of codon variants for each protein
1832 * position from "sequence_variant" features on dna
1834 @Test(groups = "Functional")
1835 public void testBuildDnaVariantsMap()
1837 SequenceI dna = new Sequence("dna", "atgAAATTTGGGCCCtag");
1838 MapList map = new MapList(new int[] { 1, 18 }, new int[] { 1, 5 }, 3, 1);
1841 * first with no variants on dna
1843 LinkedHashMap<Integer, List<DnaVariant>[]> variantsMap = AlignmentUtils
1844 .buildDnaVariantsMap(dna, map);
1845 assertTrue(variantsMap.isEmpty());
1848 * single allele codon 1, on base 1
1850 SequenceFeature sf1 = new SequenceFeature("sequence_variant", "", 1, 1,
1852 sf1.setValue("alleles", "T");
1853 sf1.setValue("ID", "sequence_variant:rs758803211");
1854 dna.addSequenceFeature(sf1);
1857 * two alleles codon 2, on bases 2 and 3 (distinct variants)
1859 SequenceFeature sf2 = new SequenceFeature("sequence_variant", "", 5, 5,
1861 sf2.setValue("alleles", "T");
1862 sf2.setValue("ID", "sequence_variant:rs758803212");
1863 dna.addSequenceFeature(sf2);
1864 SequenceFeature sf3 = new SequenceFeature("sequence_variant", "", 6, 6,
1866 sf3.setValue("alleles", "G");
1867 sf3.setValue("ID", "sequence_variant:rs758803213");
1868 dna.addSequenceFeature(sf3);
1871 * two alleles codon 3, both on base 2 (one variant)
1873 SequenceFeature sf4 = new SequenceFeature("sequence_variant", "", 8, 8,
1875 sf4.setValue("alleles", "C, G");
1876 sf4.setValue("ID", "sequence_variant:rs758803214");
1877 dna.addSequenceFeature(sf4);
1879 // no alleles on codon 4
1882 * alleles on codon 5 on all 3 bases (distinct variants)
1884 SequenceFeature sf5 = new SequenceFeature("sequence_variant", "", 13,
1886 sf5.setValue("alleles", "C, G"); // (C duplicates given base value)
1887 sf5.setValue("ID", "sequence_variant:rs758803215");
1888 dna.addSequenceFeature(sf5);
1889 SequenceFeature sf6 = new SequenceFeature("sequence_variant", "", 14,
1891 sf6.setValue("alleles", "g, a"); // should force to upper-case
1892 sf6.setValue("ID", "sequence_variant:rs758803216");
1893 dna.addSequenceFeature(sf6);
1894 SequenceFeature sf7 = new SequenceFeature("sequence_variant", "", 15,
1896 sf7.setValue("alleles", "A, T");
1897 sf7.setValue("ID", "sequence_variant:rs758803217");
1898 dna.addSequenceFeature(sf7);
1901 * build map - expect variants on positions 1, 2, 3, 5
1903 variantsMap = AlignmentUtils.buildDnaVariantsMap(dna, map);
1904 assertEquals(4, variantsMap.size());
1907 * protein residue 1: variant on codon (ATG) base 1, not on 2 or 3
1909 List<DnaVariant>[] pep1Variants = variantsMap.get(1);
1910 assertEquals(3, pep1Variants.length);
1911 assertEquals(1, pep1Variants[0].size());
1912 assertEquals("A", pep1Variants[0].get(0).base); // codon[1] base
1913 assertSame(sf1, pep1Variants[0].get(0).variant); // codon[1] variant
1914 assertEquals(1, pep1Variants[1].size());
1915 assertEquals("T", pep1Variants[1].get(0).base); // codon[2] base
1916 assertNull(pep1Variants[1].get(0).variant); // no variant here
1917 assertEquals(1, pep1Variants[2].size());
1918 assertEquals("G", pep1Variants[2].get(0).base); // codon[3] base
1919 assertNull(pep1Variants[2].get(0).variant); // no variant here
1922 * protein residue 2: variants on codon (AAA) bases 2 and 3
1924 List<DnaVariant>[] pep2Variants = variantsMap.get(2);
1925 assertEquals(3, pep2Variants.length);
1926 assertEquals(1, pep2Variants[0].size());
1927 // codon[1] base recorded while processing variant on codon[2]
1928 assertEquals("A", pep2Variants[0].get(0).base);
1929 assertNull(pep2Variants[0].get(0).variant); // no variant here
1930 // codon[2] base and variant:
1931 assertEquals(1, pep2Variants[1].size());
1932 assertEquals("A", pep2Variants[1].get(0).base);
1933 assertSame(sf2, pep2Variants[1].get(0).variant);
1934 // codon[3] base was recorded when processing codon[2] variant
1935 // and then the variant for codon[3] added to it
1936 assertEquals(1, pep2Variants[2].size());
1937 assertEquals("A", pep2Variants[2].get(0).base);
1938 assertSame(sf3, pep2Variants[2].get(0).variant);
1941 * protein residue 3: variants on codon (TTT) base 2 only
1943 List<DnaVariant>[] pep3Variants = variantsMap.get(3);
1944 assertEquals(3, pep3Variants.length);
1945 assertEquals(1, pep3Variants[0].size());
1946 assertEquals("T", pep3Variants[0].get(0).base); // codon[1] base
1947 assertNull(pep3Variants[0].get(0).variant); // no variant here
1948 assertEquals(1, pep3Variants[1].size());
1949 assertEquals("T", pep3Variants[1].get(0).base); // codon[2] base
1950 assertSame(sf4, pep3Variants[1].get(0).variant); // codon[2] variant
1951 assertEquals(1, pep3Variants[2].size());
1952 assertEquals("T", pep3Variants[2].get(0).base); // codon[3] base
1953 assertNull(pep3Variants[2].get(0).variant); // no variant here
1956 * three variants on protein position 5
1958 List<DnaVariant>[] pep5Variants = variantsMap.get(5);
1959 assertEquals(3, pep5Variants.length);
1960 assertEquals(1, pep5Variants[0].size());
1961 assertEquals("C", pep5Variants[0].get(0).base); // codon[1] base
1962 assertSame(sf5, pep5Variants[0].get(0).variant); // codon[1] variant
1963 assertEquals(1, pep5Variants[1].size());
1964 assertEquals("C", pep5Variants[1].get(0).base); // codon[2] base
1965 assertSame(sf6, pep5Variants[1].get(0).variant); // codon[2] variant
1966 assertEquals(1, pep5Variants[2].size());
1967 assertEquals("C", pep5Variants[2].get(0).base); // codon[3] base
1968 assertSame(sf7, pep5Variants[2].get(0).variant); // codon[3] variant
1972 * Tests for the method that computes all peptide variants given codon
1975 @Test(groups = "Functional")
1976 public void testComputePeptideVariants()
1979 * scenario: AAATTTCCC codes for KFP
1981 * GAA -> E source: Ensembl
1982 * CAA -> Q source: dbSNP
1983 * AAG synonymous source: COSMIC
1984 * AAT -> N source: Ensembl
1985 * ...TTC synonymous source: dbSNP
1986 * ......CAC,CGC -> H,R source: COSMIC
1987 * (one variant with two alleles)
1989 SequenceI peptide = new Sequence("pep/10-12", "KFP");
1992 * two distinct variants for codon 1 position 1
1993 * second one has clinical significance
1995 String ensembl = "Ensembl";
1996 String dbSnp = "dbSNP";
1997 String cosmic = "COSMIC";
1998 SequenceFeature sf1 = new SequenceFeature("sequence_variant", "", 1, 1,
2000 sf1.setValue("alleles", "A,G"); // GAA -> E
2001 sf1.setValue("ID", "var1.125A>G");
2002 SequenceFeature sf2 = new SequenceFeature("sequence_variant", "", 1, 1,
2004 sf2.setValue("alleles", "A,C"); // CAA -> Q
2005 sf2.setValue("ID", "var2");
2006 sf2.setValue("clinical_significance", "Dodgy");
2007 SequenceFeature sf3 = new SequenceFeature("sequence_variant", "", 3, 3,
2009 sf3.setValue("alleles", "A,G"); // synonymous
2010 sf3.setValue("ID", "var3");
2011 sf3.setValue("clinical_significance", "None");
2012 SequenceFeature sf4 = new SequenceFeature("sequence_variant", "", 3, 3,
2014 sf4.setValue("alleles", "A,T"); // AAT -> N
2015 sf4.setValue("ID", "sequence_variant:var4"); // prefix gets stripped off
2016 sf4.setValue("clinical_significance", "Benign");
2017 SequenceFeature sf5 = new SequenceFeature("sequence_variant", "", 6, 6,
2019 sf5.setValue("alleles", "T,C"); // synonymous
2020 sf5.setValue("ID", "var5");
2021 sf5.setValue("clinical_significance", "Bad");
2022 SequenceFeature sf6 = new SequenceFeature("sequence_variant", "", 8, 8,
2024 sf6.setValue("alleles", "C,A,G"); // CAC,CGC -> H,R
2025 sf6.setValue("ID", "var6");
2026 sf6.setValue("clinical_significance", "Good");
2028 List<DnaVariant> codon1Variants = new ArrayList<DnaVariant>();
2029 List<DnaVariant> codon2Variants = new ArrayList<DnaVariant>();
2030 List<DnaVariant> codon3Variants = new ArrayList<DnaVariant>();
2031 List<DnaVariant> codonVariants[] = new ArrayList[3];
2032 codonVariants[0] = codon1Variants;
2033 codonVariants[1] = codon2Variants;
2034 codonVariants[2] = codon3Variants;
2037 * compute variants for protein position 1
2039 codon1Variants.add(new DnaVariant("A", sf1));
2040 codon1Variants.add(new DnaVariant("A", sf2));
2041 codon2Variants.add(new DnaVariant("A"));
2042 codon2Variants.add(new DnaVariant("A"));
2043 codon3Variants.add(new DnaVariant("A", sf3));
2044 codon3Variants.add(new DnaVariant("A", sf4));
2045 AlignmentUtils.computePeptideVariants(peptide, 1, codonVariants);
2048 * compute variants for protein position 2
2050 codon1Variants.clear();
2051 codon2Variants.clear();
2052 codon3Variants.clear();
2053 codon1Variants.add(new DnaVariant("T"));
2054 codon2Variants.add(new DnaVariant("T"));
2055 codon3Variants.add(new DnaVariant("T", sf5));
2056 AlignmentUtils.computePeptideVariants(peptide, 2, codonVariants);
2059 * compute variants for protein position 3
2061 codon1Variants.clear();
2062 codon2Variants.clear();
2063 codon3Variants.clear();
2064 codon1Variants.add(new DnaVariant("C"));
2065 codon2Variants.add(new DnaVariant("C", sf6));
2066 codon3Variants.add(new DnaVariant("C"));
2067 AlignmentUtils.computePeptideVariants(peptide, 3, codonVariants);
2070 * verify added sequence features for
2071 * var1 K -> E Ensembl
2073 * var4 K -> N Ensembl
2074 * var6 P -> H COSMIC
2075 * var6 P -> R COSMIC
2077 SequenceFeature[] sfs = peptide.getSequenceFeatures();
2078 assertEquals(5, sfs.length);
2080 SequenceFeature sf = sfs[0];
2081 assertEquals(1, sf.getBegin());
2082 assertEquals(1, sf.getEnd());
2083 assertEquals("p.Lys1Glu", sf.getDescription());
2084 assertEquals("var1.125A>G", sf.getValue("ID"));
2085 assertNull(sf.getValue("clinical_significance"));
2086 assertEquals("ID=var1.125A>G", sf.getAttributes());
2087 assertEquals(1, sf.links.size());
2088 // link to variation is urlencoded
2090 "p.Lys1Glu var1.125A>G|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var1.125A%3EG",
2092 assertEquals(ensembl, sf.getFeatureGroup());
2095 assertEquals(1, sf.getBegin());
2096 assertEquals(1, sf.getEnd());
2097 assertEquals("p.Lys1Gln", sf.getDescription());
2098 assertEquals("var2", sf.getValue("ID"));
2099 assertEquals("Dodgy", sf.getValue("clinical_significance"));
2100 assertEquals("ID=var2;clinical_significance=Dodgy", sf.getAttributes());
2101 assertEquals(1, sf.links.size());
2103 "p.Lys1Gln var2|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var2",
2105 assertEquals(dbSnp, sf.getFeatureGroup());
2108 assertEquals(1, sf.getBegin());
2109 assertEquals(1, sf.getEnd());
2110 assertEquals("p.Lys1Asn", sf.getDescription());
2111 assertEquals("var4", sf.getValue("ID"));
2112 assertEquals("Benign", sf.getValue("clinical_significance"));
2113 assertEquals("ID=var4;clinical_significance=Benign", sf.getAttributes());
2114 assertEquals(1, sf.links.size());
2116 "p.Lys1Asn var4|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var4",
2118 assertEquals(ensembl, sf.getFeatureGroup());
2120 // var5 generates two distinct protein variant features
2122 assertEquals(3, sf.getBegin());
2123 assertEquals(3, sf.getEnd());
2124 assertEquals("p.Pro3His", sf.getDescription());
2125 assertEquals("var6", sf.getValue("ID"));
2126 assertEquals("Good", sf.getValue("clinical_significance"));
2127 assertEquals("ID=var6;clinical_significance=Good", sf.getAttributes());
2128 assertEquals(1, sf.links.size());
2130 "p.Pro3His var6|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var6",
2132 assertEquals(cosmic, sf.getFeatureGroup());
2135 assertEquals(3, sf.getBegin());
2136 assertEquals(3, sf.getEnd());
2137 assertEquals("p.Pro3Arg", sf.getDescription());
2138 assertEquals("var6", sf.getValue("ID"));
2139 assertEquals("Good", sf.getValue("clinical_significance"));
2140 assertEquals("ID=var6;clinical_significance=Good", sf.getAttributes());
2141 assertEquals(1, sf.links.size());
2143 "p.Pro3Arg var6|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var6",
2145 assertEquals(cosmic, sf.getFeatureGroup());
2149 * Tests for the method that maps the subset of a dna sequence that has CDS
2150 * (or subtype) feature, with CDS strand = '-' (reverse)
2152 // test turned off as currently findCdsPositions is not strand-dependent
2153 // left in case it comes around again...
2154 @Test(groups = "Functional", enabled = false)
2155 public void testFindCdsPositions_reverseStrand()
2157 SequenceI dnaSeq = new Sequence("dna", "aaaGGGcccAAATTTttt");
2158 dnaSeq.createDatasetSequence();
2159 SequenceI ds = dnaSeq.getDatasetSequence();
2162 SequenceFeature sf = new SequenceFeature("CDS", "", 4, 6, 0f, null);
2164 ds.addSequenceFeature(sf);
2165 // exon feature should be ignored here
2166 sf = new SequenceFeature("exon", "", 7, 9, 0f, null);
2167 ds.addSequenceFeature(sf);
2168 // CDS for dna 10-12
2169 sf = new SequenceFeature("CDS_predicted", "", 10, 12, 0f, null);
2171 ds.addSequenceFeature(sf);
2173 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
2175 * verify ranges { [12-10], [6-4] }
2177 assertEquals(6, MappingUtils.getLength(ranges));
2178 assertEquals(2, ranges.size());
2179 assertEquals(12, ranges.get(0)[0]);
2180 assertEquals(10, ranges.get(0)[1]);
2181 assertEquals(6, ranges.get(1)[0]);
2182 assertEquals(4, ranges.get(1)[1]);
2186 * Tests for the method that maps the subset of a dna sequence that has CDS
2187 * (or subtype) feature - reverse strand case where the start codon is
2190 @Test(groups = "Functional", enabled = false)
2191 // test turned off as currently findCdsPositions is not strand-dependent
2192 // left in case it comes around again...
2193 public void testFindCdsPositions_reverseStrandThreePrimeIncomplete()
2195 SequenceI dnaSeq = new Sequence("dna", "aaagGGCCCaaaTTTttt");
2196 dnaSeq.createDatasetSequence();
2197 SequenceI ds = dnaSeq.getDatasetSequence();
2200 SequenceFeature sf = new SequenceFeature("CDS", "", 5, 9, 0f, null);
2202 ds.addSequenceFeature(sf);
2203 // CDS for dna 13-15
2204 sf = new SequenceFeature("CDS_predicted", "", 13, 15, 0f, null);
2206 sf.setPhase("2"); // skip 2 bases to start of next codon
2207 ds.addSequenceFeature(sf);
2209 List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
2212 * check the mapping starts with the first complete codon
2213 * expect ranges [13, 13], [9, 5]
2215 assertEquals(6, MappingUtils.getLength(ranges));
2216 assertEquals(2, ranges.size());
2217 assertEquals(13, ranges.get(0)[0]);
2218 assertEquals(13, ranges.get(0)[1]);
2219 assertEquals(9, ranges.get(1)[0]);
2220 assertEquals(5, ranges.get(1)[1]);
2223 @Test(groups = "Functional")
2224 public void testAlignAs_alternateTranscriptsUngapped()
2226 SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
2227 SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
2228 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
2229 ((Alignment) dna).createDatasetAlignment();
2230 SequenceI cds1 = new Sequence("cds1", "GGGTTT");
2231 SequenceI cds2 = new Sequence("cds2", "CCCAAA");
2232 AlignmentI cds = new Alignment(new SequenceI[] { cds1, cds2 });
2233 ((Alignment) cds).createDatasetAlignment();
2235 AlignedCodonFrame acf = new AlignedCodonFrame();
2236 MapList map = new MapList(new int[] { 4, 9 }, new int[] { 1, 6 }, 1, 1);
2237 acf.addMap(dna1.getDatasetSequence(), cds1.getDatasetSequence(), map);
2238 map = new MapList(new int[] { 1, 3, 10, 12 }, new int[] { 1, 6 }, 1, 1);
2239 acf.addMap(dna2.getDatasetSequence(), cds2.getDatasetSequence(), map);
2242 * verify CDS alignment is as:
2243 * cccGGGTTTaaa (cdna)
2244 * CCCgggtttAAA (cdna)
2246 * ---GGGTTT--- (cds)
2247 * CCC------AAA (cds)
2249 dna.addCodonFrame(acf);
2250 AlignmentUtils.alignAs(cds, dna);
2251 assertEquals("---GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
2252 assertEquals("CCC------AAA", cds.getSequenceAt(1).getSequenceAsString());
2255 @Test(groups = { "Functional" })
2256 public void testAddMappedPositions()
2258 SequenceI from = new Sequence("dna", "ggAA-ATcc-TT-g");
2259 SequenceI seq1 = new Sequence("cds", "AAATTT");
2260 from.createDatasetSequence();
2261 seq1.createDatasetSequence();
2262 Mapping mapping = new Mapping(seq1, new MapList(
2263 new int[] { 3, 6, 9, 10 }, new int[] { 1, 6 }, 1, 1));
2264 Map<Integer, Map<SequenceI, Character>> map = new TreeMap<Integer, Map<SequenceI, Character>>();
2265 AlignmentUtils.addMappedPositions(seq1, from, mapping, map);
2268 * verify map has seq1 residues in columns 3,4,6,7,11,12
2270 assertEquals(6, map.size());
2271 assertEquals('A', map.get(3).get(seq1).charValue());
2272 assertEquals('A', map.get(4).get(seq1).charValue());
2273 assertEquals('A', map.get(6).get(seq1).charValue());
2274 assertEquals('T', map.get(7).get(seq1).charValue());
2275 assertEquals('T', map.get(11).get(seq1).charValue());
2276 assertEquals('T', map.get(12).get(seq1).charValue());
2284 * Test case where the mapping 'from' range includes a stop codon which is
2285 * absent in the 'to' range
2287 @Test(groups = { "Functional" })
2288 public void testAddMappedPositions_withStopCodon()
2290 SequenceI from = new Sequence("dna", "ggAA-ATcc-TT-g");
2291 SequenceI seq1 = new Sequence("cds", "AAATTT");
2292 from.createDatasetSequence();
2293 seq1.createDatasetSequence();
2294 Mapping mapping = new Mapping(seq1, new MapList(
2295 new int[] { 3, 6, 9, 10 }, new int[] { 1, 6 }, 1, 1));
2296 Map<Integer, Map<SequenceI, Character>> map = new TreeMap<Integer, Map<SequenceI, Character>>();
2297 AlignmentUtils.addMappedPositions(seq1, from, mapping, map);
2300 * verify map has seq1 residues in columns 3,4,6,7,11,12
2302 assertEquals(6, map.size());
2303 assertEquals('A', map.get(3).get(seq1).charValue());
2304 assertEquals('A', map.get(4).get(seq1).charValue());
2305 assertEquals('A', map.get(6).get(seq1).charValue());
2306 assertEquals('T', map.get(7).get(seq1).charValue());
2307 assertEquals('T', map.get(11).get(seq1).charValue());
2308 assertEquals('T', map.get(12).get(seq1).charValue());
2312 * Test for the case where the products for which we want CDS are specified.
2313 * This is to represent the case where EMBL has CDS mappings to both Uniprot
2314 * and EMBLCDSPROTEIN. makeCdsAlignment() should only return the mappings for
2315 * the protein sequences specified.
2317 @Test(groups = { "Functional" })
2318 public void testMakeCdsAlignment_filterProducts()
2320 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
2321 SequenceI dna2 = new Sequence("dna2", "GGGcccTTTaaaCCC");
2322 SequenceI pep1 = new Sequence("Uniprot|pep1", "GF");
2323 SequenceI pep2 = new Sequence("Uniprot|pep2", "GFP");
2324 SequenceI pep3 = new Sequence("EMBL|pep3", "GF");
2325 SequenceI pep4 = new Sequence("EMBL|pep4", "GFP");
2326 dna1.createDatasetSequence();
2327 dna2.createDatasetSequence();
2328 pep1.createDatasetSequence();
2329 pep2.createDatasetSequence();
2330 pep3.createDatasetSequence();
2331 pep4.createDatasetSequence();
2332 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
2333 dna.setDataset(null);
2334 AlignmentI emblPeptides = new Alignment(new SequenceI[] { pep3, pep4 });
2335 emblPeptides.setDataset(null);
2337 AlignedCodonFrame acf = new AlignedCodonFrame();
2338 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
2339 new int[] { 1, 2 }, 3, 1);
2340 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
2341 acf.addMap(dna1.getDatasetSequence(), pep3.getDatasetSequence(), map);
2342 dna.addCodonFrame(acf);
2344 acf = new AlignedCodonFrame();
2345 map = new MapList(new int[] { 1, 3, 7, 9, 13, 15 }, new int[] { 1, 3 },
2347 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map);
2348 acf.addMap(dna2.getDatasetSequence(), pep4.getDatasetSequence(), map);
2349 dna.addCodonFrame(acf);
2352 * execute method under test to find CDS for EMBL peptides only
2354 AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
2355 dna1, dna2 }, dna.getDataset(), emblPeptides.getSequencesArray());
2357 assertEquals(2, cds.getSequences().size());
2358 assertEquals("GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
2359 assertEquals("GGGTTTCCC", cds.getSequenceAt(1).getSequenceAsString());
2362 * verify shared, extended alignment dataset
2364 assertSame(dna.getDataset(), cds.getDataset());
2365 assertTrue(dna.getDataset().getSequences()
2366 .contains(cds.getSequenceAt(0).getDatasetSequence()));
2367 assertTrue(dna.getDataset().getSequences()
2368 .contains(cds.getSequenceAt(1).getDatasetSequence()));
2371 * Verify mappings from CDS to peptide, cDNA to CDS, and cDNA to peptide
2372 * the mappings are on the shared alignment dataset
2374 List<AlignedCodonFrame> cdsMappings = cds.getDataset().getCodonFrames();
2376 * 6 mappings, 2*(DNA->CDS), 2*(DNA->Pep), 2*(CDS->Pep)
2378 assertEquals(6, cdsMappings.size());
2381 * verify that mapping sets for dna and cds alignments are different
2382 * [not current behaviour - all mappings are on the alignment dataset]
2384 // select -> subselect type to test.
2385 // Assert.assertNotSame(dna.getCodonFrames(), cds.getCodonFrames());
2386 // assertEquals(4, dna.getCodonFrames().size());
2387 // assertEquals(4, cds.getCodonFrames().size());
2390 * Two mappings involve pep3 (dna to pep3, cds to pep3)
2391 * Mapping from pep3 to GGGTTT in first new exon sequence
2393 List<AlignedCodonFrame> pep3Mappings = MappingUtils
2394 .findMappingsForSequence(pep3, cdsMappings);
2395 assertEquals(2, pep3Mappings.size());
2396 List<AlignedCodonFrame> mappings = MappingUtils
2397 .findMappingsForSequence(cds.getSequenceAt(0), pep3Mappings);
2398 assertEquals(1, mappings.size());
2401 SearchResultsI sr = MappingUtils.buildSearchResults(pep3, 1, mappings);
2402 assertEquals(1, sr.getResults().size());
2403 SearchResultMatchI m = sr.getResults().get(0);
2404 assertSame(cds.getSequenceAt(0).getDatasetSequence(), m.getSequence());
2405 assertEquals(1, m.getStart());
2406 assertEquals(3, m.getEnd());
2408 sr = MappingUtils.buildSearchResults(pep3, 2, mappings);
2409 m = sr.getResults().get(0);
2410 assertSame(cds.getSequenceAt(0).getDatasetSequence(), m.getSequence());
2411 assertEquals(4, m.getStart());
2412 assertEquals(6, m.getEnd());
2415 * Two mappings involve pep4 (dna to pep4, cds to pep4)
2416 * Verify mapping from pep4 to GGGTTTCCC in second new exon sequence
2418 List<AlignedCodonFrame> pep4Mappings = MappingUtils
2419 .findMappingsForSequence(pep4, cdsMappings);
2420 assertEquals(2, pep4Mappings.size());
2421 mappings = MappingUtils.findMappingsForSequence(cds.getSequenceAt(1),
2423 assertEquals(1, mappings.size());
2425 sr = MappingUtils.buildSearchResults(pep4, 1, mappings);
2426 assertEquals(1, sr.getResults().size());
2427 m = sr.getResults().get(0);
2428 assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
2429 assertEquals(1, m.getStart());
2430 assertEquals(3, m.getEnd());
2432 sr = MappingUtils.buildSearchResults(pep4, 2, mappings);
2433 m = sr.getResults().get(0);
2434 assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
2435 assertEquals(4, m.getStart());
2436 assertEquals(6, m.getEnd());
2438 sr = MappingUtils.buildSearchResults(pep4, 3, mappings);
2439 m = sr.getResults().get(0);
2440 assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
2441 assertEquals(7, m.getStart());
2442 assertEquals(9, m.getEnd());
2446 * Test the method that just copies aligned sequences, provided all sequences
2447 * to be aligned share the aligned sequence's dataset
2449 @Test(groups = "Functional")
2450 public void testAlignAsSameSequences()
2452 SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
2453 SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
2454 AlignmentI al1 = new Alignment(new SequenceI[] { dna1, dna2 });
2455 ((Alignment) al1).createDatasetAlignment();
2457 SequenceI dna3 = new Sequence(dna1);
2458 SequenceI dna4 = new Sequence(dna2);
2459 assertSame(dna3.getDatasetSequence(), dna1.getDatasetSequence());
2460 assertSame(dna4.getDatasetSequence(), dna2.getDatasetSequence());
2461 String seq1 = "-cc-GG-GT-TT--aaa";
2462 dna3.setSequence(seq1);
2463 String seq2 = "C--C-Cgg--gtt-tAA-A-";
2464 dna4.setSequence(seq2);
2465 AlignmentI al2 = new Alignment(new SequenceI[] { dna3, dna4 });
2466 ((Alignment) al2).createDatasetAlignment();
2468 assertTrue(AlignmentUtils.alignAsSameSequences(al1, al2));
2469 assertEquals(seq1, al1.getSequenceAt(0).getSequenceAsString());
2470 assertEquals(seq2, al1.getSequenceAt(1).getSequenceAsString());
2473 * add another sequence to 'aligned' - should still succeed, since
2474 * unaligned sequences still share a dataset with aligned sequences
2476 SequenceI dna5 = new Sequence("dna5", "CCCgggtttAAA");
2477 dna5.createDatasetSequence();
2478 al2.addSequence(dna5);
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 'unaligned' - should fail, since now not
2485 * all unaligned sequences share a dataset with aligned sequences
2487 SequenceI dna6 = new Sequence("dna6", "CCCgggtttAAA");
2488 dna6.createDatasetSequence();
2489 al1.addSequence(dna6);
2490 // JAL-2110 JBP Comment: what's the use case for this behaviour ?
2491 assertFalse(AlignmentUtils.alignAsSameSequences(al1, al2));
2494 @Test(groups = "Functional")
2495 public void testAlignAsSameSequencesMultipleSubSeq()
2497 SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
2498 SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
2499 SequenceI as1 = dna1.deriveSequence();
2500 SequenceI as2 = dna1.deriveSequence().getSubSequence(3, 7);
2501 SequenceI as3 = dna2.deriveSequence();
2502 as1.insertCharAt(6, 5, '-');
2503 String s_as1 = as1.getSequenceAsString();
2504 as2.insertCharAt(6, 5, '-');
2505 String s_as2 = as2.getSequenceAsString();
2506 as3.insertCharAt(6, 5, '-');
2507 String s_as3 = as3.getSequenceAsString();
2508 AlignmentI aligned = new Alignment(new SequenceI[] { as1, as2, as3 });
2510 // why do we need to cast this still ?
2511 ((Alignment) aligned).createDatasetAlignment();
2512 SequenceI uas1 = dna1.deriveSequence();
2513 SequenceI uas2 = dna1.deriveSequence().getSubSequence(3, 7);
2514 SequenceI uas3 = dna2.deriveSequence();
2515 AlignmentI tobealigned = new Alignment(new SequenceI[] { uas1, uas2,
2517 ((Alignment) tobealigned).createDatasetAlignment();
2519 assertTrue(AlignmentUtils.alignAsSameSequences(tobealigned, aligned));
2520 assertEquals(s_as1, uas1.getSequenceAsString());
2521 assertEquals(s_as2, uas2.getSequenceAsString());
2522 assertEquals(s_as3, uas3.getSequenceAsString());