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.assertNull;
26 import static org.testng.AssertJUnit.assertSame;
27 import static org.testng.AssertJUnit.assertTrue;
29 import jalview.datamodel.AlignedCodonFrame;
30 import jalview.datamodel.Alignment;
31 import jalview.datamodel.AlignmentAnnotation;
32 import jalview.datamodel.AlignmentI;
33 import jalview.datamodel.Annotation;
34 import jalview.datamodel.DBRefEntry;
35 import jalview.datamodel.Mapping;
36 import jalview.datamodel.SearchResults;
37 import jalview.datamodel.SearchResults.Match;
38 import jalview.datamodel.Sequence;
39 import jalview.datamodel.SequenceI;
40 import jalview.io.AppletFormatAdapter;
41 import jalview.io.FormatAdapter;
42 import jalview.util.MapList;
43 import jalview.util.MappingUtils;
45 import java.io.IOException;
46 import java.util.ArrayList;
47 import java.util.Arrays;
48 import java.util.HashSet;
49 import java.util.Iterator;
50 import java.util.List;
54 import org.testng.annotations.Test;
56 public class AlignmentUtilsTests
59 private static final String TEST_DATA =
61 "#=GS D.melanogaster.1 AC AY119185.1/838-902\n" +
62 "#=GS D.melanogaster.2 AC AC092237.1/57223-57161\n" +
63 "#=GS D.melanogaster.3 AC AY060611.1/560-627\n" +
64 "D.melanogaster.1 G.AGCC.CU...AUGAUCGA\n" +
65 "#=GR D.melanogaster.1 SS ................((((\n" +
66 "D.melanogaster.2 C.AUUCAACU.UAUGAGGAU\n" +
67 "#=GR D.melanogaster.2 SS ................((((\n" +
68 "D.melanogaster.3 G.UGGCGCU..UAUGACGCA\n" +
69 "#=GR D.melanogaster.3 SS (.(((...(....(((((((\n" +
72 private static final String AA_SEQS_1 =
78 private static final String CDNA_SEQS_1 =
80 "AC-GG--CUC-CAA-CT\n" +
82 "-CG-TTA--ACG---AAGT\n";
84 private static final String CDNA_SEQS_2 =
91 // public static Sequence ts=new
92 // Sequence("short","ASDASDASDASDASDASDASDASDASDASDASDASDASD");
93 public static Sequence ts = new Sequence("short",
94 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklm");
96 @Test(groups = { "Functional" })
97 public void testExpandContext()
99 AlignmentI al = new Alignment(new Sequence[] {});
100 for (int i = 4; i < 14; i += 2)
102 SequenceI s1 = ts.deriveSequence().getSubSequence(i, i + 7);
105 System.out.println(new AppletFormatAdapter().formatSequences("Clustal",
107 for (int flnk = -1; flnk < 25; flnk++)
109 AlignmentI exp = AlignmentUtils.expandContext(al, flnk);
110 System.out.println("\nFlank size: " + flnk);
111 System.out.println(new AppletFormatAdapter().formatSequences(
112 "Clustal", exp, true));
116 * Full expansion to complete sequences
118 for (SequenceI sq : exp.getSequences())
120 String ung = sq.getSequenceAsString().replaceAll("-+", "");
121 final String errorMsg = "Flanking sequence not the same as original dataset sequence.\n"
124 + sq.getDatasetSequence().getSequenceAsString();
125 assertTrue(errorMsg, ung.equalsIgnoreCase(sq.getDatasetSequence()
126 .getSequenceAsString()));
132 * Last sequence is fully expanded, others have leading gaps to match
134 assertTrue(exp.getSequenceAt(4).getSequenceAsString()
136 assertTrue(exp.getSequenceAt(3).getSequenceAsString()
137 .startsWith("--abc"));
138 assertTrue(exp.getSequenceAt(2).getSequenceAsString()
139 .startsWith("----abc"));
140 assertTrue(exp.getSequenceAt(1).getSequenceAsString()
141 .startsWith("------abc"));
142 assertTrue(exp.getSequenceAt(0).getSequenceAsString()
143 .startsWith("--------abc"));
149 * Test that annotations are correctly adjusted by expandContext
151 @Test(groups = { "Functional" })
152 public void testExpandContext_annotation()
154 AlignmentI al = new Alignment(new Sequence[] {});
155 SequenceI ds = new Sequence("Seq1", "ABCDEFGHI");
157 SequenceI seq1 = ds.deriveSequence().getSubSequence(3, 6);
158 al.addSequence(seq1);
161 * Annotate DEF with 4/5/6 respectively
163 Annotation[] anns = new Annotation[] { new Annotation(4),
164 new Annotation(5), new Annotation(6) };
165 AlignmentAnnotation ann = new AlignmentAnnotation("SS",
166 "secondary structure", anns);
167 seq1.addAlignmentAnnotation(ann);
170 * The annotations array should match aligned positions
172 assertEquals(3, ann.annotations.length);
173 assertEquals(4, ann.annotations[0].value, 0.001);
174 assertEquals(5, ann.annotations[1].value, 0.001);
175 assertEquals(6, ann.annotations[2].value, 0.001);
178 * Check annotation to sequence position mappings before expanding the
179 * sequence; these are set up in Sequence.addAlignmentAnnotation ->
180 * Annotation.setSequenceRef -> createSequenceMappings
182 assertNull(ann.getAnnotationForPosition(1));
183 assertNull(ann.getAnnotationForPosition(2));
184 assertNull(ann.getAnnotationForPosition(3));
185 assertEquals(4, ann.getAnnotationForPosition(4).value, 0.001);
186 assertEquals(5, ann.getAnnotationForPosition(5).value, 0.001);
187 assertEquals(6, ann.getAnnotationForPosition(6).value, 0.001);
188 assertNull(ann.getAnnotationForPosition(7));
189 assertNull(ann.getAnnotationForPosition(8));
190 assertNull(ann.getAnnotationForPosition(9));
193 * Expand the subsequence to the full sequence abcDEFghi
195 AlignmentI expanded = AlignmentUtils.expandContext(al, -1);
196 assertEquals("abcDEFghi", expanded.getSequenceAt(0)
197 .getSequenceAsString());
200 * Confirm the alignment and sequence have the same SS annotation,
201 * referencing the expanded sequence
203 ann = expanded.getSequenceAt(0).getAnnotation()[0];
204 assertSame(ann, expanded.getAlignmentAnnotation()[0]);
205 assertSame(expanded.getSequenceAt(0), ann.sequenceRef);
208 * The annotations array should have null values except for annotated
211 assertNull(ann.annotations[0]);
212 assertNull(ann.annotations[1]);
213 assertNull(ann.annotations[2]);
214 assertEquals(4, ann.annotations[3].value, 0.001);
215 assertEquals(5, ann.annotations[4].value, 0.001);
216 assertEquals(6, ann.annotations[5].value, 0.001);
217 assertNull(ann.annotations[6]);
218 assertNull(ann.annotations[7]);
219 assertNull(ann.annotations[8]);
222 * sequence position mappings should be unchanged
224 assertNull(ann.getAnnotationForPosition(1));
225 assertNull(ann.getAnnotationForPosition(2));
226 assertNull(ann.getAnnotationForPosition(3));
227 assertEquals(4, ann.getAnnotationForPosition(4).value, 0.001);
228 assertEquals(5, ann.getAnnotationForPosition(5).value, 0.001);
229 assertEquals(6, ann.getAnnotationForPosition(6).value, 0.001);
230 assertNull(ann.getAnnotationForPosition(7));
231 assertNull(ann.getAnnotationForPosition(8));
232 assertNull(ann.getAnnotationForPosition(9));
236 * Test method that returns a map of lists of sequences by sequence name.
238 * @throws IOException
240 @Test(groups = { "Functional" })
241 public void testGetSequencesByName() throws IOException
243 final String data = ">Seq1Name\nKQYL\n" + ">Seq2Name\nRFPW\n"
244 + ">Seq1Name\nABCD\n";
245 AlignmentI al = loadAlignment(data, "FASTA");
246 Map<String, List<SequenceI>> map = AlignmentUtils
247 .getSequencesByName(al);
248 assertEquals(2, map.keySet().size());
249 assertEquals(2, map.get("Seq1Name").size());
250 assertEquals("KQYL", map.get("Seq1Name").get(0).getSequenceAsString());
251 assertEquals("ABCD", map.get("Seq1Name").get(1).getSequenceAsString());
252 assertEquals(1, map.get("Seq2Name").size());
253 assertEquals("RFPW", map.get("Seq2Name").get(0).getSequenceAsString());
257 * Helper method to load an alignment and ensure dataset sequences are set up.
263 * @throws IOException
265 protected AlignmentI loadAlignment(final String data, String format)
268 AlignmentI a = new FormatAdapter().readFile(data,
269 AppletFormatAdapter.PASTE, format);
275 * Test mapping of protein to cDNA, for the case where we have no sequence
276 * cross-references, so mappings are made first-served 1-1 where sequences
279 * @throws IOException
281 @Test(groups = { "Functional" })
282 public void testMapProteinAlignmentToCdna_noXrefs() throws IOException
284 List<SequenceI> protseqs = new ArrayList<SequenceI>();
285 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
286 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
287 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
288 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
289 protein.setDataset(null);
291 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
292 dnaseqs.add(new Sequence("EMBL|A11111", "TCAGCACGC")); // = SAR
293 dnaseqs.add(new Sequence("EMBL|A22222", "GAGATACAA")); // = EIQ
294 dnaseqs.add(new Sequence("EMBL|A33333", "GAAATCCAG")); // = EIQ
295 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG")); // = EIQ
296 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[4]));
297 cdna.setDataset(null);
299 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
301 // 3 mappings made, each from 1 to 1 sequence
302 assertEquals(3, protein.getCodonFrames().size());
303 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
304 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
305 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
307 // V12345 mapped to A22222
308 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
310 assertEquals(1, acf.getdnaSeqs().length);
311 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
312 acf.getdnaSeqs()[0]);
313 Mapping[] protMappings = acf.getProtMappings();
314 assertEquals(1, protMappings.length);
315 MapList mapList = protMappings[0].getMap();
316 assertEquals(3, mapList.getFromRatio());
317 assertEquals(1, mapList.getToRatio());
318 assertTrue(Arrays.equals(new int[] { 1, 9 }, mapList.getFromRanges()
320 assertEquals(1, mapList.getFromRanges().size());
321 assertTrue(Arrays.equals(new int[] { 1, 3 },
322 mapList.getToRanges().get(0)));
323 assertEquals(1, mapList.getToRanges().size());
325 // V12346 mapped to A33333
326 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
327 assertEquals(1, acf.getdnaSeqs().length);
328 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
329 acf.getdnaSeqs()[0]);
331 // V12347 mapped to A11111
332 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
333 assertEquals(1, acf.getdnaSeqs().length);
334 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
335 acf.getdnaSeqs()[0]);
337 // no mapping involving the 'extra' A44444
338 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(3)).isEmpty());
342 * Test for the alignSequenceAs method that takes two sequences and a mapping.
344 @Test(groups = { "Functional" })
345 public void testAlignSequenceAs_withMapping_noIntrons()
347 MapList map = new MapList(new int[] { 1, 6 }, new int[] { 1, 2 }, 3, 1);
350 * No existing gaps in dna:
352 checkAlignSequenceAs("GGGAAA", "-A-L-", false, false, map,
356 * Now introduce gaps in dna but ignore them when realigning.
358 checkAlignSequenceAs("-G-G-G-A-A-A-", "-A-L-", false, false, map,
362 * Now include gaps in dna when realigning. First retaining 'mapped' gaps
363 * only, i.e. those within the exon region.
365 checkAlignSequenceAs("-G-G--G-A--A-A-", "-A-L-", true, false, map,
366 "---G-G--G---A--A-A");
369 * Include all gaps in dna when realigning (within and without the exon
370 * region). The leading gap, and the gaps between codons, are subsumed by
371 * the protein alignment gap.
373 checkAlignSequenceAs("-G-GG--AA-A-", "-A-L-", true, true, map,
377 * Include only unmapped gaps in dna when realigning (outside the exon
378 * region). The leading gap, and the gaps between codons, are subsumed by
379 * the protein alignment gap.
381 checkAlignSequenceAs("-G-GG--AA-A-", "-A-L-", false, true, map,
386 * Test for the alignSequenceAs method that takes two sequences and a mapping.
388 @Test(groups = { "Functional" })
389 public void testAlignSequenceAs_withMapping_withIntrons()
392 * Exons at codon 2 (AAA) and 4 (TTT)
394 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
395 new int[] { 1, 2 }, 3, 1);
398 * Simple case: no gaps in dna
400 checkAlignSequenceAs("GGGAAACCCTTTGGG", "--A-L-", false, false, map,
401 "GGG---AAACCCTTTGGG");
404 * Add gaps to dna - but ignore when realigning.
406 checkAlignSequenceAs("-G-G-G--A--A---AC-CC-T-TT-GG-G-", "--A-L-",
407 false, false, map, "GGG---AAACCCTTTGGG");
410 * Add gaps to dna - include within exons only when realigning.
412 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
413 true, false, map, "GGG---A--A---ACCCT-TTGGG");
416 * Include gaps outside exons only when realigning.
418 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
419 false, true, map, "-G-G-GAAAC-CCTTT-GG-G-");
422 * Include gaps following first intron if we are 'preserving mapped gaps'
424 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
425 true, true, map, "-G-G-G--A--A---A-C-CC-T-TT-GG-G-");
428 * Include all gaps in dna when realigning.
430 checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
431 true, true, map, "-G-G-G--A--A---A-C-CC-T-TT-GG-G-");
435 * Test for the case where not all of the protein sequence is mapped to cDNA.
437 @Test(groups = { "Functional" })
438 public void testAlignSequenceAs_withMapping_withUnmappedProtein()
441 * Exons at codon 2 (AAA) and 4 (TTT) mapped to A and P
443 final MapList map = new MapList(new int[] { 4, 6, 10, 12 }, new int[] {
447 * Expect alignment does nothing (aborts realignment). Change this test
448 * first if different behaviour wanted.
450 checkAlignSequenceAs("gggAAAcccTTTggg", "-A-L-P-", false, false, map,
451 "gggAAAccc---TTTggg");
455 * Helper method that performs and verifies the method under test.
458 * the sequence to be realigned
460 * the sequence whose alignment is to be copied
461 * @param preserveMappedGaps
462 * @param preserveUnmappedGaps
466 protected void checkAlignSequenceAs(final String alignee,
467 final String alignModel, final boolean preserveMappedGaps,
468 final boolean preserveUnmappedGaps, MapList map,
469 final String expected)
471 SequenceI alignMe = new Sequence("Seq1", alignee);
472 alignMe.createDatasetSequence();
473 SequenceI alignFrom = new Sequence("Seq2", alignModel);
474 alignFrom.createDatasetSequence();
475 AlignedCodonFrame acf = new AlignedCodonFrame();
476 acf.addMap(alignMe.getDatasetSequence(), alignFrom.getDatasetSequence(), map);
478 AlignmentUtils.alignSequenceAs(alignMe, alignFrom, acf, "---", '-',
479 preserveMappedGaps, preserveUnmappedGaps);
480 assertEquals(expected, alignMe.getSequenceAsString());
484 * Test for the alignSequenceAs method where we preserve gaps in introns only.
486 @Test(groups = { "Functional" })
487 public void testAlignSequenceAs_keepIntronGapsOnly()
491 * Intron GGGAAA followed by exon CCCTTT
493 MapList map = new MapList(new int[] { 7, 12 }, new int[] { 1, 2 }, 3, 1);
495 checkAlignSequenceAs("GG-G-AA-A-C-CC-T-TT", "AL", false, true, map,
500 * Test for the method that generates an aligned translated sequence from one
503 @Test(groups = { "Functional" })
504 public void testGetAlignedTranslation_dnaLikeProtein()
506 // dna alignment will be replaced
507 SequenceI dna = new Sequence("Seq1", "T-G-CC-A--T-TAC-CAG-");
508 dna.createDatasetSequence();
509 // protein alignment will be 'applied' to dna
510 SequenceI protein = new Sequence("Seq1", "-CH-Y--Q-");
511 protein.createDatasetSequence();
512 MapList map = new MapList(new int[] { 1, 12 }, new int[] { 1, 4 }, 3, 1);
513 AlignedCodonFrame acf = new AlignedCodonFrame();
514 acf.addMap(dna.getDatasetSequence(), protein.getDatasetSequence(), map);
516 final SequenceI aligned = AlignmentUtils.getAlignedTranslation(protein,
518 assertEquals("---TGCCAT---TAC------CAG---",
519 aligned.getSequenceAsString());
520 assertSame(aligned.getDatasetSequence(), dna.getDatasetSequence());
524 * Test the method that realigns protein to match mapped codon alignment.
526 @Test(groups = { "Functional" })
527 public void testAlignProteinAsDna()
529 // seq1 codons are [1,2,3] [4,5,6] [7,8,9] [10,11,12]
530 SequenceI dna1 = new Sequence("Seq1", "TGCCATTACCAG-");
531 // seq2 codons are [1,3,4] [5,6,7] [8,9,10] [11,12,13]
532 SequenceI dna2 = new Sequence("Seq2", "T-GCCATTACCAG");
533 // seq3 codons are [1,2,3] [4,5,7] [8,9,10] [11,12,13]
534 SequenceI dna3 = new Sequence("Seq3", "TGCCA-TTACCAG");
535 AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
536 dna.setDataset(null);
538 // protein alignment will be realigned like dna
539 SequenceI prot1 = new Sequence("Seq1", "CHYQ");
540 SequenceI prot2 = new Sequence("Seq2", "CHYQ");
541 SequenceI prot3 = new Sequence("Seq3", "CHYQ");
542 SequenceI prot4 = new Sequence("Seq4", "R-QSV"); // unmapped, unchanged
543 AlignmentI protein = new Alignment(new SequenceI[] { prot1, prot2,
545 protein.setDataset(null);
547 MapList map = new MapList(new int[] { 1, 12 }, new int[] { 1, 4 }, 3, 1);
548 AlignedCodonFrame acf = new AlignedCodonFrame();
549 acf.addMap(dna1.getDatasetSequence(), prot1.getDatasetSequence(), map);
550 acf.addMap(dna2.getDatasetSequence(), prot2.getDatasetSequence(), map);
551 acf.addMap(dna3.getDatasetSequence(), prot3.getDatasetSequence(), map);
552 protein.setCodonFrames(new ArrayList<AlignedCodonFrame>());
555 * Translated codon order is [1,2,3] [1,3,4] [4,5,6] [4,5,7] [5,6,7] [7,8,9]
556 * [8,9,10] [10,11,12] [11,12,13]
558 AlignmentUtils.alignProteinAsDna(protein, dna);
559 assertEquals("C-H--Y-Q-", prot1.getSequenceAsString());
560 assertEquals("-C--H-Y-Q", prot2.getSequenceAsString());
561 assertEquals("C--H--Y-Q", prot3.getSequenceAsString());
562 assertEquals("R-QSV", prot4.getSequenceAsString());
566 * Test the method that tests whether a CDNA sequence translates to a protein
569 @Test(groups = { "Functional" })
570 public void testTranslatesAs()
572 assertTrue(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(), 0,
573 "FPKG".toCharArray()));
574 // with start codon (not in protein)
575 assertTrue(AlignmentUtils.translatesAs("atgtttcccaaaggg".toCharArray(),
576 3, "FPKG".toCharArray()));
577 // with stop codon1 (not in protein)
578 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtaa".toCharArray(),
579 0, "FPKG".toCharArray()));
580 // with stop codon1 (in protein as *)
581 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtaa".toCharArray(),
582 0, "FPKG*".toCharArray()));
583 // with stop codon2 (not in protein)
584 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtag".toCharArray(),
585 0, "FPKG".toCharArray()));
586 // with stop codon3 (not in protein)
587 assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtga".toCharArray(),
588 0, "FPKG".toCharArray()));
589 // with start and stop codon1
590 assertTrue(AlignmentUtils.translatesAs(
591 "atgtttcccaaagggtaa".toCharArray(), 3, "FPKG".toCharArray()));
592 // with start and stop codon1 (in protein as *)
593 assertTrue(AlignmentUtils.translatesAs(
594 "atgtttcccaaagggtaa".toCharArray(), 3, "FPKG*".toCharArray()));
595 // with start and stop codon2
596 assertTrue(AlignmentUtils.translatesAs(
597 "atgtttcccaaagggtag".toCharArray(), 3, "FPKG".toCharArray()));
598 // with start and stop codon3
599 assertTrue(AlignmentUtils.translatesAs(
600 "atgtttcccaaagggtga".toCharArray(), 3, "FPKG".toCharArray()));
602 // with embedded stop codon
603 assertTrue(AlignmentUtils.translatesAs(
604 "atgtttTAGcccaaaTAAgggtga".toCharArray(), 3,
605 "F*PK*G".toCharArray()));
608 assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(),
609 0, "FPMG".toCharArray()));
613 * Test mapping of protein to cDNA, for cases where the cDNA has start and/or
614 * stop codons in addition to the protein coding sequence.
616 * @throws IOException
618 @Test(groups = { "Functional" })
619 public void testMapProteinAlignmentToCdna_withStartAndStopCodons()
622 List<SequenceI> protseqs = new ArrayList<SequenceI>();
623 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
624 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
625 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
626 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
627 protein.setDataset(null);
629 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
631 dnaseqs.add(new Sequence("EMBL|A11111", "ATGTCAGCACGC"));
633 dnaseqs.add(new Sequence("EMBL|A22222", "GAGATACAATAA"));
634 // = start +EIQ + stop
635 dnaseqs.add(new Sequence("EMBL|A33333", "ATGGAAATCCAGTAG"));
636 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG"));
637 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[4]));
638 cdna.setDataset(null);
640 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
642 // 3 mappings made, each from 1 to 1 sequence
643 assertEquals(3, protein.getCodonFrames().size());
644 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
645 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
646 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
648 // V12345 mapped from A22222
649 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
651 assertEquals(1, acf.getdnaSeqs().length);
652 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
653 acf.getdnaSeqs()[0]);
654 Mapping[] protMappings = acf.getProtMappings();
655 assertEquals(1, protMappings.length);
656 MapList mapList = protMappings[0].getMap();
657 assertEquals(3, mapList.getFromRatio());
658 assertEquals(1, mapList.getToRatio());
659 assertTrue(Arrays.equals(new int[] { 1, 9 }, mapList.getFromRanges()
661 assertEquals(1, mapList.getFromRanges().size());
662 assertTrue(Arrays.equals(new int[] { 1, 3 },
663 mapList.getToRanges().get(0)));
664 assertEquals(1, mapList.getToRanges().size());
666 // V12346 mapped from A33333 starting position 4
667 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
668 assertEquals(1, acf.getdnaSeqs().length);
669 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
670 acf.getdnaSeqs()[0]);
671 protMappings = acf.getProtMappings();
672 assertEquals(1, protMappings.length);
673 mapList = protMappings[0].getMap();
674 assertEquals(3, mapList.getFromRatio());
675 assertEquals(1, mapList.getToRatio());
676 assertTrue(Arrays.equals(new int[] { 4, 12 }, mapList.getFromRanges()
678 assertEquals(1, mapList.getFromRanges().size());
679 assertTrue(Arrays.equals(new int[] { 1, 3 },
680 mapList.getToRanges().get(0)));
681 assertEquals(1, mapList.getToRanges().size());
683 // V12347 mapped to A11111 starting position 4
684 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
685 assertEquals(1, acf.getdnaSeqs().length);
686 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
687 acf.getdnaSeqs()[0]);
688 protMappings = acf.getProtMappings();
689 assertEquals(1, protMappings.length);
690 mapList = protMappings[0].getMap();
691 assertEquals(3, mapList.getFromRatio());
692 assertEquals(1, mapList.getToRatio());
693 assertTrue(Arrays.equals(new int[] { 4, 12 }, mapList.getFromRanges()
695 assertEquals(1, mapList.getFromRanges().size());
696 assertTrue(Arrays.equals(new int[] { 1, 3 },
697 mapList.getToRanges().get(0)));
698 assertEquals(1, mapList.getToRanges().size());
700 // no mapping involving the 'extra' A44444
701 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(3)).isEmpty());
705 * Test mapping of protein to cDNA, for the case where we have some sequence
706 * cross-references. Verify that 1-to-many mappings are made where
707 * cross-references exist and sequences are mappable.
709 * @throws IOException
711 @Test(groups = { "Functional" })
712 public void testMapProteinAlignmentToCdna_withXrefs() throws IOException
714 List<SequenceI> protseqs = new ArrayList<SequenceI>();
715 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
716 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
717 protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
718 AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
719 protein.setDataset(null);
721 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
722 dnaseqs.add(new Sequence("EMBL|A11111", "TCAGCACGC")); // = SAR
723 dnaseqs.add(new Sequence("EMBL|A22222", "ATGGAGATACAA")); // = start + EIQ
724 dnaseqs.add(new Sequence("EMBL|A33333", "GAAATCCAG")); // = EIQ
725 dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG")); // = EIQ
726 dnaseqs.add(new Sequence("EMBL|A55555", "GAGATTCAG")); // = EIQ
727 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[5]));
728 cdna.setDataset(null);
730 // Xref A22222 to V12345 (should get mapped)
731 dnaseqs.get(1).addDBRef(new DBRefEntry("UNIPROT", "1", "V12345"));
732 // Xref V12345 to A44444 (should get mapped)
733 protseqs.get(0).addDBRef(new DBRefEntry("EMBL", "1", "A44444"));
734 // Xref A33333 to V12347 (sequence mismatch - should not get mapped)
735 dnaseqs.get(2).addDBRef(new DBRefEntry("UNIPROT", "1", "V12347"));
736 // as V12345 is mapped to A22222 and A44444, this leaves V12346 unmapped.
737 // it should get paired up with the unmapped A33333
738 // A11111 should be mapped to V12347
739 // A55555 is spare and has no xref so is not mapped
741 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
743 // 4 protein mappings made for 3 proteins, 2 to V12345, 1 each to V12346/7
744 assertEquals(3, protein.getCodonFrames().size());
745 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
746 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
747 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
749 // one mapping for each of the first 4 cDNA sequences
750 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(0)).size());
751 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(1)).size());
752 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(2)).size());
753 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(3)).size());
755 // V12345 mapped to A22222 and A44444
756 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
758 assertEquals(2, acf.getdnaSeqs().length);
759 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
760 acf.getdnaSeqs()[0]);
761 assertEquals(cdna.getSequenceAt(3).getDatasetSequence(),
762 acf.getdnaSeqs()[1]);
764 // V12346 mapped to A33333
765 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
766 assertEquals(1, acf.getdnaSeqs().length);
767 assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
768 acf.getdnaSeqs()[0]);
770 // V12347 mapped to A11111
771 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
772 assertEquals(1, acf.getdnaSeqs().length);
773 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
774 acf.getdnaSeqs()[0]);
776 // no mapping involving the 'extra' A55555
777 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(4)).isEmpty());
781 * Test mapping of protein to cDNA, for the case where we have some sequence
782 * cross-references. Verify that once we have made an xref mapping we don't
783 * also map un-xrefd sequeces.
785 * @throws IOException
787 @Test(groups = { "Functional" })
788 public void testMapProteinAlignmentToCdna_prioritiseXrefs()
791 List<SequenceI> protseqs = new ArrayList<SequenceI>();
792 protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
793 protseqs.add(new Sequence("UNIPROT|V12346", "EIQ"));
794 AlignmentI protein = new Alignment(
795 protseqs.toArray(new SequenceI[protseqs.size()]));
796 protein.setDataset(null);
798 List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
799 dnaseqs.add(new Sequence("EMBL|A11111", "GAAATCCAG")); // = EIQ
800 dnaseqs.add(new Sequence("EMBL|A22222", "GAAATTCAG")); // = EIQ
801 AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[dnaseqs
803 cdna.setDataset(null);
805 // Xref A22222 to V12345 (should get mapped)
806 // A11111 should then be mapped to the unmapped V12346
807 dnaseqs.get(1).addDBRef(new DBRefEntry("UNIPROT", "1", "V12345"));
809 assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
811 // 2 protein mappings made
812 assertEquals(2, protein.getCodonFrames().size());
813 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
814 assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
816 // one mapping for each of the cDNA sequences
817 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(0)).size());
818 assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(1)).size());
820 // V12345 mapped to A22222
821 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
823 assertEquals(1, acf.getdnaSeqs().length);
824 assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
825 acf.getdnaSeqs()[0]);
827 // V12346 mapped to A11111
828 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
829 assertEquals(1, acf.getdnaSeqs().length);
830 assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
831 acf.getdnaSeqs()[0]);
835 * Test the method that shows or hides sequence annotations by type(s) and
838 @Test(groups = { "Functional" })
839 public void testShowOrHideSequenceAnnotations()
841 SequenceI seq1 = new Sequence("Seq1", "AAA");
842 SequenceI seq2 = new Sequence("Seq2", "BBB");
843 SequenceI seq3 = new Sequence("Seq3", "CCC");
844 Annotation[] anns = new Annotation[] { new Annotation(2f) };
845 AlignmentAnnotation ann1 = new AlignmentAnnotation("Structure", "ann1",
847 ann1.setSequenceRef(seq1);
848 AlignmentAnnotation ann2 = new AlignmentAnnotation("Structure", "ann2",
850 ann2.setSequenceRef(seq2);
851 AlignmentAnnotation ann3 = new AlignmentAnnotation("Structure", "ann3",
853 AlignmentAnnotation ann4 = new AlignmentAnnotation("Temp", "ann4", anns);
854 ann4.setSequenceRef(seq1);
855 AlignmentAnnotation ann5 = new AlignmentAnnotation("Temp", "ann5", anns);
856 ann5.setSequenceRef(seq2);
857 AlignmentAnnotation ann6 = new AlignmentAnnotation("Temp", "ann6", anns);
858 AlignmentI al = new Alignment(new SequenceI[] { seq1, seq2, seq3 });
859 al.addAnnotation(ann1); // Structure for Seq1
860 al.addAnnotation(ann2); // Structure for Seq2
861 al.addAnnotation(ann3); // Structure for no sequence
862 al.addAnnotation(ann4); // Temp for seq1
863 al.addAnnotation(ann5); // Temp for seq2
864 al.addAnnotation(ann6); // Temp for no sequence
865 List<String> types = new ArrayList<String>();
866 List<SequenceI> scope = new ArrayList<SequenceI>();
869 * Set all sequence related Structure to hidden (ann1, ann2)
871 types.add("Structure");
872 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, false,
874 assertFalse(ann1.visible);
875 assertFalse(ann2.visible);
876 assertTrue(ann3.visible); // not sequence-related, not affected
877 assertTrue(ann4.visible); // not Structure, not affected
878 assertTrue(ann5.visible); // "
879 assertTrue(ann6.visible); // not sequence-related, not affected
882 * Set Temp in {seq1, seq3} to hidden
888 AlignmentUtils.showOrHideSequenceAnnotations(al, types, scope, false,
890 assertFalse(ann1.visible); // unchanged
891 assertFalse(ann2.visible); // unchanged
892 assertTrue(ann3.visible); // not sequence-related, not affected
893 assertFalse(ann4.visible); // Temp for seq1 hidden
894 assertTrue(ann5.visible); // not in scope, not affected
895 assertTrue(ann6.visible); // not sequence-related, not affected
898 * Set Temp in all sequences to hidden
904 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, false,
906 assertFalse(ann1.visible); // unchanged
907 assertFalse(ann2.visible); // unchanged
908 assertTrue(ann3.visible); // not sequence-related, not affected
909 assertFalse(ann4.visible); // Temp for seq1 hidden
910 assertFalse(ann5.visible); // Temp for seq2 hidden
911 assertTrue(ann6.visible); // not sequence-related, not affected
914 * Set all types in {seq1, seq3} to visible
920 AlignmentUtils.showOrHideSequenceAnnotations(al, types, scope, true,
922 assertTrue(ann1.visible); // Structure for seq1 set visible
923 assertFalse(ann2.visible); // not in scope, unchanged
924 assertTrue(ann3.visible); // not sequence-related, not affected
925 assertTrue(ann4.visible); // Temp for seq1 set visible
926 assertFalse(ann5.visible); // not in scope, unchanged
927 assertTrue(ann6.visible); // not sequence-related, not affected
930 * Set all types in all scope to hidden
932 AlignmentUtils.showOrHideSequenceAnnotations(al, types, null, true,
934 assertFalse(ann1.visible);
935 assertFalse(ann2.visible);
936 assertTrue(ann3.visible); // not sequence-related, not affected
937 assertFalse(ann4.visible);
938 assertFalse(ann5.visible);
939 assertTrue(ann6.visible); // not sequence-related, not affected
943 * Tests for the method that checks if one sequence cross-references another
945 @Test(groups = { "Functional" })
946 public void testHasCrossRef()
948 assertFalse(AlignmentUtils.hasCrossRef(null, null));
949 SequenceI seq1 = new Sequence("EMBL|A12345", "ABCDEF");
950 assertFalse(AlignmentUtils.hasCrossRef(seq1, null));
951 assertFalse(AlignmentUtils.hasCrossRef(null, seq1));
952 SequenceI seq2 = new Sequence("UNIPROT|V20192", "ABCDEF");
953 assertFalse(AlignmentUtils.hasCrossRef(seq1, seq2));
956 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "v20193"));
957 assertFalse(AlignmentUtils.hasCrossRef(seq1, seq2));
959 // case-insensitive; version number is ignored
960 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "v20192"));
961 assertTrue(AlignmentUtils.hasCrossRef(seq1, seq2));
964 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
965 assertTrue(AlignmentUtils.hasCrossRef(seq1, seq2));
966 // test is one-way only
967 assertFalse(AlignmentUtils.hasCrossRef(seq2, seq1));
971 * Tests for the method that checks if either sequence cross-references the
974 @Test(groups = { "Functional" })
975 public void testHaveCrossRef()
977 assertFalse(AlignmentUtils.hasCrossRef(null, null));
978 SequenceI seq1 = new Sequence("EMBL|A12345", "ABCDEF");
979 assertFalse(AlignmentUtils.haveCrossRef(seq1, null));
980 assertFalse(AlignmentUtils.haveCrossRef(null, seq1));
981 SequenceI seq2 = new Sequence("UNIPROT|V20192", "ABCDEF");
982 assertFalse(AlignmentUtils.haveCrossRef(seq1, seq2));
984 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
985 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
986 // next is true for haveCrossRef, false for hasCrossRef
987 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
989 // now the other way round
991 seq2.addDBRef(new DBRefEntry("EMBL", "1", "A12345"));
992 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
993 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
996 seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
997 assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
998 assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
1002 * Test the method that extracts the exon-only part of a dna alignment.
1004 @Test(groups = { "Functional" })
1005 public void testMakeExonAlignment()
1007 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
1008 SequenceI dna2 = new Sequence("dna2", "GGGcccTTTaaaCCC");
1009 SequenceI pep1 = new Sequence("pep1", "GF");
1010 SequenceI pep2 = new Sequence("pep2", "GFP");
1011 dna1.createDatasetSequence();
1012 dna2.createDatasetSequence();
1013 pep1.createDatasetSequence();
1014 pep2.createDatasetSequence();
1016 List<AlignedCodonFrame> mappings = new ArrayList<AlignedCodonFrame>();
1017 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1018 new int[] { 1, 2 }, 3, 1);
1019 AlignedCodonFrame acf = new AlignedCodonFrame();
1020 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
1022 map = new MapList(new int[] { 1, 3, 7, 9, 13, 15 }, new int[] { 1, 3 },
1024 acf = new AlignedCodonFrame();
1025 acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map);
1028 AlignmentI exons = AlignmentUtils.makeExonAlignment(new SequenceI[] {
1029 dna1, dna2 }, mappings);
1030 assertEquals(2, exons.getSequences().size());
1031 assertEquals("GGGTTT", exons.getSequenceAt(0).getSequenceAsString());
1032 assertEquals("GGGTTTCCC", exons.getSequenceAt(1).getSequenceAsString());
1035 * Verify updated mappings
1037 assertEquals(2, mappings.size());
1040 * Mapping from pep1 to GGGTTT in first new exon sequence
1042 List<AlignedCodonFrame> pep1Mapping = MappingUtils
1043 .findMappingsForSequence(pep1, mappings);
1044 assertEquals(1, pep1Mapping.size());
1046 SearchResults sr = MappingUtils.buildSearchResults(pep1, 1, mappings);
1047 assertEquals(1, sr.getResults().size());
1048 Match m = sr.getResults().get(0);
1049 assertEquals(exons.getSequenceAt(0).getDatasetSequence(),
1051 assertEquals(1, m.getStart());
1052 assertEquals(3, m.getEnd());
1054 sr = MappingUtils.buildSearchResults(pep1, 2, mappings);
1055 m = sr.getResults().get(0);
1056 assertEquals(exons.getSequenceAt(0).getDatasetSequence(),
1058 assertEquals(4, m.getStart());
1059 assertEquals(6, m.getEnd());
1062 * Mapping from pep2 to GGGTTTCCC in second new exon sequence
1064 List<AlignedCodonFrame> pep2Mapping = MappingUtils
1065 .findMappingsForSequence(pep2, mappings);
1066 assertEquals(1, pep2Mapping.size());
1068 sr = MappingUtils.buildSearchResults(pep2, 1, mappings);
1069 assertEquals(1, sr.getResults().size());
1070 m = sr.getResults().get(0);
1071 assertEquals(exons.getSequenceAt(1).getDatasetSequence(),
1073 assertEquals(1, m.getStart());
1074 assertEquals(3, m.getEnd());
1076 sr = MappingUtils.buildSearchResults(pep2, 2, mappings);
1077 m = sr.getResults().get(0);
1078 assertEquals(exons.getSequenceAt(1).getDatasetSequence(),
1080 assertEquals(4, m.getStart());
1081 assertEquals(6, m.getEnd());
1083 sr = MappingUtils.buildSearchResults(pep2, 3, mappings);
1084 m = sr.getResults().get(0);
1085 assertEquals(exons.getSequenceAt(1).getDatasetSequence(),
1087 assertEquals(7, m.getStart());
1088 assertEquals(9, m.getEnd());
1092 * Test the method that makes an exon-only sequence from a DNA sequence and
1093 * its product mapping. Test includes the expected case that the DNA sequence
1094 * already has a protein product (Uniprot translation) which in turn has an
1095 * x-ref to the EMBLCDS record.
1097 @Test(groups = { "Functional" })
1098 public void testMakeExonSequences()
1100 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
1101 SequenceI pep1 = new Sequence("pep1", "GF");
1102 dna1.createDatasetSequence();
1103 pep1.createDatasetSequence();
1104 pep1.getDatasetSequence().addDBRef(
1105 new DBRefEntry("EMBLCDS", "2", "A12345"));
1108 * Make the mapping from dna to protein. The protein sequence has a DBRef to
1111 Set<AlignedCodonFrame> mappings = new HashSet<AlignedCodonFrame>();
1112 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1113 new int[] { 1, 2 }, 3, 1);
1114 AlignedCodonFrame acf = new AlignedCodonFrame();
1115 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
1118 AlignedCodonFrame newMapping = new AlignedCodonFrame();
1119 List<SequenceI> exons = AlignmentUtils.makeExonSequences(dna1, acf,
1121 assertEquals(1, exons.size());
1122 SequenceI exon = exons.get(0);
1124 assertEquals("GGGTTT", exon.getSequenceAsString());
1125 assertEquals("dna1|A12345", exon.getName());
1126 assertEquals(1, exon.getDBRef().length);
1127 DBRefEntry cdsRef = exon.getDBRef()[0];
1128 assertEquals("EMBLCDS", cdsRef.getSource());
1129 assertEquals("2", cdsRef.getVersion());
1130 assertEquals("A12345", cdsRef.getAccessionId());
1134 * Test the method that makes an exon-only alignment from a DNA sequence and
1135 * its product mappings, for the case where there are multiple exon mappings
1136 * to different protein products.
1138 @Test(groups = { "Functional" })
1139 public void testMakeExonAlignment_multipleProteins()
1141 SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
1142 SequenceI pep1 = new Sequence("pep1", "GF"); // GGGTTT
1143 SequenceI pep2 = new Sequence("pep2", "KP"); // aaaccc
1144 SequenceI pep3 = new Sequence("pep3", "KF"); // aaaTTT
1145 dna1.createDatasetSequence();
1146 pep1.createDatasetSequence();
1147 pep2.createDatasetSequence();
1148 pep3.createDatasetSequence();
1149 pep1.getDatasetSequence().addDBRef(
1150 new DBRefEntry("EMBLCDS", "2", "A12345"));
1151 pep2.getDatasetSequence().addDBRef(
1152 new DBRefEntry("EMBLCDS", "3", "A12346"));
1153 pep3.getDatasetSequence().addDBRef(
1154 new DBRefEntry("EMBLCDS", "4", "A12347"));
1157 * Make the mappings from dna to protein. Using LinkedHashset is a
1158 * convenience so results are in the input order. There is no assertion that
1159 * the generated exon sequences are in any particular order.
1161 List<AlignedCodonFrame> mappings = new ArrayList<AlignedCodonFrame>();
1162 // map ...GGG...TTT to GF
1163 MapList map = new MapList(new int[] { 4, 6, 10, 12 },
1164 new int[] { 1, 2 }, 3, 1);
1165 AlignedCodonFrame acf = new AlignedCodonFrame();
1166 acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
1169 // map aaa...ccc to KP
1170 map = new MapList(new int[] { 1, 3, 7, 9 }, new int[] { 1, 2 }, 3, 1);
1171 acf = new AlignedCodonFrame();
1172 acf.addMap(dna1.getDatasetSequence(), pep2.getDatasetSequence(), map);
1175 // map aaa......TTT to KF
1176 map = new MapList(new int[] { 1, 3, 10, 12 }, new int[] { 1, 2 }, 3, 1);
1177 acf = new AlignedCodonFrame();
1178 acf.addMap(dna1.getDatasetSequence(), pep3.getDatasetSequence(), map);
1182 * Create the Exon alignment; also replaces the dna-to-protein mappings with
1183 * exon-to-protein and exon-to-dna mappings
1185 AlignmentI exal = AlignmentUtils.makeExonAlignment(
1186 new SequenceI[] { dna1 }, mappings);
1189 * Verify we have 3 exon sequences, mapped to pep1/2/3 respectively
1191 List<SequenceI> exons = exal.getSequences();
1192 assertEquals(3, exons.size());
1194 SequenceI exon = exons.get(0);
1195 assertEquals("GGGTTT", exon.getSequenceAsString());
1196 assertEquals("dna1|A12345", exon.getName());
1197 assertEquals(1, exon.getDBRef().length);
1198 DBRefEntry cdsRef = exon.getDBRef()[0];
1199 assertEquals("EMBLCDS", cdsRef.getSource());
1200 assertEquals("2", cdsRef.getVersion());
1201 assertEquals("A12345", cdsRef.getAccessionId());
1203 exon = exons.get(1);
1204 assertEquals("aaaccc", exon.getSequenceAsString());
1205 assertEquals("dna1|A12346", exon.getName());
1206 assertEquals(1, exon.getDBRef().length);
1207 cdsRef = exon.getDBRef()[0];
1208 assertEquals("EMBLCDS", cdsRef.getSource());
1209 assertEquals("3", cdsRef.getVersion());
1210 assertEquals("A12346", cdsRef.getAccessionId());
1212 exon = exons.get(2);
1213 assertEquals("aaaTTT", exon.getSequenceAsString());
1214 assertEquals("dna1|A12347", exon.getName());
1215 assertEquals(1, exon.getDBRef().length);
1216 cdsRef = exon.getDBRef()[0];
1217 assertEquals("EMBLCDS", cdsRef.getSource());
1218 assertEquals("4", cdsRef.getVersion());
1219 assertEquals("A12347", cdsRef.getAccessionId());
1222 * Verify there are mappings from each exon sequence to its protein product
1223 * and also to its dna source
1225 Iterator<AlignedCodonFrame> newMappingsIterator = mappings.iterator();
1227 // mappings for dna1 - exon1 - pep1
1228 AlignedCodonFrame exonMapping = newMappingsIterator.next();
1229 List<Mapping> dnaMappings = exonMapping.getMappingsForSequence(dna1);
1230 assertEquals(1, dnaMappings.size());
1231 assertSame(exons.get(0).getDatasetSequence(), dnaMappings.get(0)
1233 assertEquals("G(1) in CDS should map to G(4) in DNA", 4, dnaMappings
1234 .get(0).getMap().getToPosition(1));
1235 List<Mapping> peptideMappings = exonMapping
1236 .getMappingsForSequence(pep1);
1237 assertEquals(1, peptideMappings.size());
1238 assertSame(pep1.getDatasetSequence(), peptideMappings.get(0).getTo());
1240 // mappings for dna1 - exon2 - pep2
1241 exonMapping = newMappingsIterator.next();
1242 dnaMappings = exonMapping.getMappingsForSequence(dna1);
1243 assertEquals(1, dnaMappings.size());
1244 assertSame(exons.get(1).getDatasetSequence(), dnaMappings.get(0)
1246 assertEquals("c(4) in CDS should map to c(7) in DNA", 7, dnaMappings
1247 .get(0).getMap().getToPosition(4));
1248 peptideMappings = exonMapping.getMappingsForSequence(pep2);
1249 assertEquals(1, peptideMappings.size());
1250 assertSame(pep2.getDatasetSequence(), peptideMappings.get(0).getTo());
1252 // mappings for dna1 - exon3 - pep3
1253 exonMapping = newMappingsIterator.next();
1254 dnaMappings = exonMapping.getMappingsForSequence(dna1);
1255 assertEquals(1, dnaMappings.size());
1256 assertSame(exons.get(2).getDatasetSequence(), dnaMappings.get(0)
1258 assertEquals("T(4) in CDS should map to T(10) in DNA", 10, dnaMappings
1259 .get(0).getMap().getToPosition(4));
1260 peptideMappings = exonMapping.getMappingsForSequence(pep3);
1261 assertEquals(1, peptideMappings.size());
1262 assertSame(pep3.getDatasetSequence(), peptideMappings.get(0).getTo());
1265 @Test(groups = { "Functional" })
1266 public void testIsMappable()
1268 SequenceI dna1 = new Sequence("dna1", "cgCAGtgGT");
1269 SequenceI aa1 = new Sequence("aa1", "RSG");
1270 AlignmentI al1 = new Alignment(new SequenceI[] { dna1 });
1271 AlignmentI al2 = new Alignment(new SequenceI[] { aa1 });
1273 assertFalse(AlignmentUtils.isMappable(null, null));
1274 assertFalse(AlignmentUtils.isMappable(al1, null));
1275 assertFalse(AlignmentUtils.isMappable(null, al1));
1276 assertFalse(AlignmentUtils.isMappable(al1, al1));
1277 assertFalse(AlignmentUtils.isMappable(al2, al2));
1279 assertTrue(AlignmentUtils.isMappable(al1, al2));
1280 assertTrue(AlignmentUtils.isMappable(al2, al1));
1284 * Test creating a mapping when the sequences involved do not start at residue
1287 * @throws IOException
1289 @Test(groups = { "Functional" })
1290 public void testMapProteinSequenceToCdna_forSubsequence()
1293 SequenceI prot = new Sequence("UNIPROT|V12345", "E-I--Q", 10, 12);
1294 prot.createDatasetSequence();
1296 SequenceI dna = new Sequence("EMBL|A33333", "GAA--AT-C-CAG", 40, 48);
1297 dna.createDatasetSequence();
1299 MapList map = AlignmentUtils.mapProteinSequenceToCdna(prot, dna);
1300 assertEquals(10, map.getToLowest());
1301 assertEquals(12, map.getToHighest());
1302 assertEquals(40, map.getFromLowest());
1303 assertEquals(48, map.getFromHighest());
1307 * Test for the alignSequenceAs method where we have protein mapped to protein
1309 @Test(groups = { "Functional" })
1310 public void testAlignSequenceAs_mappedProteinProtein()
1313 SequenceI alignMe = new Sequence("Match", "MGAASEV");
1314 alignMe.createDatasetSequence();
1315 SequenceI alignFrom = new Sequence("Query", "LQTGYMGAASEVMFSPTRR");
1316 alignFrom.createDatasetSequence();
1318 AlignedCodonFrame acf = new AlignedCodonFrame();
1319 // this is like a domain or motif match of part of a peptide sequence
1320 MapList map = new MapList(new int[] { 6, 12 }, new int[] { 1, 7 }, 1, 1);
1321 acf.addMap(alignFrom.getDatasetSequence(),
1322 alignMe.getDatasetSequence(), map);
1324 AlignmentUtils.alignSequenceAs(alignMe, alignFrom, acf, "-", '-', true,
1326 assertEquals("-----MGAASEV-------", alignMe.getSequenceAsString());
1330 * Test for the alignSequenceAs method where there are trailing unmapped
1331 * residues in the model sequence
1333 @Test(groups = { "Functional" })
1334 public void testAlignSequenceAs_withTrailingPeptide()
1336 // map first 3 codons to KPF; G is a trailing unmapped residue
1337 MapList map = new MapList(new int[] { 1, 9 }, new int[] { 1, 3 }, 3, 1);
1339 checkAlignSequenceAs("AAACCCTTT", "K-PFG", true, true, map,