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.datamodel;
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.io.DataSourceType;
31 import jalview.io.FileFormat;
32 import jalview.io.FileFormatI;
33 import jalview.io.FormatAdapter;
34 import jalview.util.MapList;
36 import java.io.IOException;
37 import java.util.ArrayList;
38 import java.util.Arrays;
39 import java.util.Iterator;
40 import java.util.List;
42 import org.testng.annotations.BeforeMethod;
43 import org.testng.annotations.Test;
46 * Unit tests for Alignment datamodel.
51 public class AlignmentTest
54 private static final String TEST_DATA =
56 "#=GS D.melanogaster.1 AC AY119185.1/838-902\n" +
57 "#=GS D.melanogaster.2 AC AC092237.1/57223-57161\n" +
58 "#=GS D.melanogaster.3 AC AY060611.1/560-627\n" +
59 "D.melanogaster.1 G.AGCC.CU...AUGAUCGA\n" +
60 "#=GR D.melanogaster.1 SS ................((((\n" +
61 "D.melanogaster.2 C.AUUCAACU.UAUGAGGAU\n" +
62 "#=GR D.melanogaster.2 SS ................((((\n" +
63 "D.melanogaster.3 G.UGGCGCU..UAUGACGCA\n" +
64 "#=GR D.melanogaster.3 SS (.(((...(....(((((((\n" +
67 private static final String AA_SEQS_1 =
73 private static final String CDNA_SEQS_1 =
74 ">Seq1Name/100-111\n" +
75 "AC-GG--CUC-CAA-CT\n" +
76 ">Seq2Name/200-211\n" +
77 "-CG-TTA--ACG---AAGT\n";
79 private static final String CDNA_SEQS_2 =
86 private AlignmentI al;
89 * Helper method to load an alignment and ensure dataset sequences are set up.
97 protected AlignmentI loadAlignment(final String data, FileFormatI format)
100 AlignmentI a = new FormatAdapter().readFile(data, DataSourceType.PASTE,
107 * Read in Stockholm format test data including secondary structure
110 @BeforeMethod(alwaysRun = true)
111 public void setUp() throws IOException
113 al = loadAlignment(TEST_DATA, FileFormat.Stockholm);
115 for (AlignmentAnnotation ann : al.getAlignmentAnnotation())
117 ann.setCalcId("CalcIdFor" + al.getSequenceAt(i).getName());
123 * Test method that returns annotations that match on calcId.
125 @Test(groups = { "Functional" })
126 public void testFindAnnotation_byCalcId()
128 Iterable<AlignmentAnnotation> anns = al
129 .findAnnotation("CalcIdForD.melanogaster.2");
130 Iterator<AlignmentAnnotation> iter = anns.iterator();
131 assertTrue(iter.hasNext());
132 AlignmentAnnotation ann = iter.next();
133 assertEquals("D.melanogaster.2", ann.sequenceRef.getName());
134 assertFalse(iter.hasNext());
137 @Test(groups = { "Functional" })
138 public void testDeleteAllAnnotations_includingAutocalculated()
140 AlignmentAnnotation aa = new AlignmentAnnotation("Consensus",
142 aa.autoCalculated = true;
143 al.addAnnotation(aa);
144 AlignmentAnnotation[] anns = al.getAlignmentAnnotation();
145 assertEquals("Wrong number of annotations before deleting", 4,
147 al.deleteAllAnnotations(true);
148 assertEquals("Not all deleted", 0, al.getAlignmentAnnotation().length);
151 @Test(groups = { "Functional" })
152 public void testDeleteAllAnnotations_excludingAutocalculated()
154 AlignmentAnnotation aa = new AlignmentAnnotation("Consensus",
156 aa.autoCalculated = true;
157 al.addAnnotation(aa);
158 AlignmentAnnotation[] anns = al.getAlignmentAnnotation();
159 assertEquals("Wrong number of annotations before deleting", 4,
161 al.deleteAllAnnotations(false);
162 assertEquals("Not just one annotation left", 1,
163 al.getAlignmentAnnotation().length);
167 * Tests for realigning as per a supplied alignment: Dna as Dna.
169 * Note: AlignedCodonFrame's state variables are named for protein-to-cDNA
170 * mapping, but can be exploited for a general 'sequence-to-sequence' mapping
173 * @throws IOException
175 @Test(groups = { "Functional" })
176 public void testAlignAs_dnaAsDna() throws IOException
179 AlignmentI al1 = loadAlignment(CDNA_SEQS_1, FileFormat.Fasta);
181 AlignmentI al2 = loadAlignment(CDNA_SEQS_2, FileFormat.Fasta);
184 * Make mappings between sequences. The 'aligned cDNA' is playing the role
185 * of what would normally be protein here.
187 makeMappings(al1, al2);
189 ((Alignment) al2).alignAs(al1, false, true);
190 assertEquals("GC-TC--GUC-GTACT", al2.getSequenceAt(0)
191 .getSequenceAsString());
192 assertEquals("-GG-GTC--AGG--CAGT", al2.getSequenceAt(1)
193 .getSequenceAsString());
197 * Aligning protein from cDNA.
199 * @throws IOException
201 @Test(groups = { "Functional" })
202 public void testAlignAs_proteinAsCdna() throws IOException
204 // see also AlignmentUtilsTests
205 AlignmentI al1 = loadAlignment(CDNA_SEQS_1, FileFormat.Fasta);
206 AlignmentI al2 = loadAlignment(AA_SEQS_1, FileFormat.Fasta);
207 makeMappings(al1, al2);
209 // Fudge - alignProteinAsCdna expects mappings to be on protein
210 al2.getCodonFrames().addAll(al1.getCodonFrames());
212 ((Alignment) al2).alignAs(al1, false, true);
213 assertEquals("K-Q-Y-L-", al2.getSequenceAt(0).getSequenceAsString());
214 assertEquals("-R-F-P-W", al2.getSequenceAt(1).getSequenceAsString());
218 * Test aligning cdna as per protein alignment.
220 * @throws IOException
222 @Test(groups = { "Functional" }, enabled = true)
223 // TODO review / update this test after redesign of alignAs method
224 public void testAlignAs_cdnaAsProtein() throws IOException
227 * Load alignments and add mappings for cDNA to protein
229 AlignmentI al1 = loadAlignment(CDNA_SEQS_1, FileFormat.Fasta);
230 AlignmentI al2 = loadAlignment(AA_SEQS_1, FileFormat.Fasta);
231 makeMappings(al1, al2);
234 * Realign DNA; currently keeping existing gaps in introns only
236 ((Alignment) al1).alignAs(al2, false, true);
237 assertEquals("ACG---GCUCCA------ACT---", al1.getSequenceAt(0)
238 .getSequenceAsString());
239 assertEquals("---CGT---TAACGA---AGT---", al1.getSequenceAt(1)
240 .getSequenceAsString());
244 * Test aligning cdna as per protein - single sequences
246 * @throws IOException
248 @Test(groups = { "Functional" }, enabled = true)
249 // TODO review / update this test after redesign of alignAs method
250 public void testAlignAs_cdnaAsProtein_singleSequence() throws IOException
253 * simple case insert one gap
255 verifyAlignAs(">dna\nCAAaaa\n", ">protein\nQ-K\n", "CAA---aaa");
258 * simple case but with sequence offsets
260 verifyAlignAs(">dna/5-10\nCAAaaa\n", ">protein/20-21\nQ-K\n",
264 * insert gaps as per protein, drop gaps within codons
266 verifyAlignAs(">dna/10-18\nCA-Aa-aa--AGA\n", ">aa/6-8\n-Q-K--R\n",
267 "---CAA---aaa------AGA");
271 * Helper method that makes mappings and then aligns the first alignment as
277 * @throws IOException
279 public void verifyAlignAs(String fromSeqs, String toSeqs, String expected)
283 * Load alignments and add mappings from nucleotide to protein (or from
284 * first to second if both the same type)
286 AlignmentI al1 = loadAlignment(fromSeqs, FileFormat.Fasta);
287 AlignmentI al2 = loadAlignment(toSeqs, FileFormat.Fasta);
288 makeMappings(al1, al2);
291 * Realign DNA; currently keeping existing gaps in introns only
293 ((Alignment) al1).alignAs(al2, false, true);
294 assertEquals(expected, al1.getSequenceAt(0).getSequenceAsString());
298 * Helper method to make mappings between sequences, and add the mappings to
299 * the 'mapped from' alignment
304 public void makeMappings(AlignmentI alFrom, AlignmentI alTo)
306 int ratio = (alFrom.isNucleotide() == alTo.isNucleotide() ? 1 : 3);
308 AlignedCodonFrame acf = new AlignedCodonFrame();
310 for (int i = 0; i < alFrom.getHeight(); i++)
312 SequenceI seqFrom = alFrom.getSequenceAt(i);
313 SequenceI seqTo = alTo.getSequenceAt(i);
314 MapList ml = new MapList(new int[] { seqFrom.getStart(),
316 new int[] { seqTo.getStart(), seqTo.getEnd() }, ratio, 1);
317 acf.addMap(seqFrom, seqTo, ml);
321 * not sure whether mappings 'belong' or protein or nucleotide
322 * alignment, so adding to both ;~)
324 alFrom.addCodonFrame(acf);
325 alTo.addCodonFrame(acf);
329 * Test aligning dna as per protein alignment, for the case where there are
330 * introns (i.e. some dna sites have no mapping from a peptide).
332 * @throws IOException
334 @Test(groups = { "Functional" }, enabled = false)
335 // TODO review / update this test after redesign of alignAs method
336 public void testAlignAs_dnaAsProtein_withIntrons() throws IOException
339 * Load alignments and add mappings for cDNA to protein
341 String dna1 = "A-Aa-gG-GCC-cT-TT";
342 String dna2 = "c--CCGgg-TT--T-AA-A";
343 AlignmentI al1 = loadAlignment(">Dna1/6-17\n" + dna1
344 + "\n>Dna2/20-31\n" + dna2 + "\n", FileFormat.Fasta);
345 AlignmentI al2 = loadAlignment(
346 ">Pep1/7-9\n-P--YK\n>Pep2/11-13\nG-T--F\n", FileFormat.Fasta);
347 AlignedCodonFrame acf = new AlignedCodonFrame();
348 // Seq1 has intron at dna positions 3,4,9 so splice is AAG GCC TTT
349 // Seq2 has intron at dna positions 1,5,6 so splice is CCG TTT AAA
350 MapList ml1 = new MapList(new int[] { 6, 7, 10, 13, 15, 17 }, new int[]
352 acf.addMap(al1.getSequenceAt(0), al2.getSequenceAt(0), ml1);
353 MapList ml2 = new MapList(new int[] { 21, 23, 26, 31 }, new int[] { 11,
355 acf.addMap(al1.getSequenceAt(1), al2.getSequenceAt(1), ml2);
356 al2.addCodonFrame(acf);
359 * Align ignoring gaps in dna introns and exons
361 ((Alignment) al1).alignAs(al2, false, false);
362 assertEquals("---AAagG------GCCcTTT", al1.getSequenceAt(0)
363 .getSequenceAsString());
364 // note 1 gap in protein corresponds to 'gg-' in DNA (3 positions)
365 assertEquals("cCCGgg-TTT------AAA", al1.getSequenceAt(1)
366 .getSequenceAsString());
369 * Reset and realign, preserving gaps in dna introns and exons
371 al1.getSequenceAt(0).setSequence(dna1);
372 al1.getSequenceAt(1).setSequence(dna2);
373 ((Alignment) al1).alignAs(al2, true, true);
374 // String dna1 = "A-Aa-gG-GCC-cT-TT";
375 // String dna2 = "c--CCGgg-TT--T-AA-A";
376 // assumption: we include 'the greater of' protein/dna gap lengths, not both
377 assertEquals("---A-Aa-gG------GCC-cT-TT", al1.getSequenceAt(0)
378 .getSequenceAsString());
379 assertEquals("c--CCGgg-TT--T------AA-A", al1.getSequenceAt(1)
380 .getSequenceAsString());
383 @Test(groups = "Functional")
384 public void testCopyConstructor() throws IOException
386 AlignmentI protein = loadAlignment(AA_SEQS_1, FileFormat.Fasta);
387 // create sequence and alignment datasets
388 protein.setDataset(null);
389 AlignedCodonFrame acf = new AlignedCodonFrame();
390 List<AlignedCodonFrame> acfList = Arrays.asList(new AlignedCodonFrame[]
392 protein.getDataset().setCodonFrames(acfList);
393 AlignmentI copy = new Alignment(protein);
396 * copy has different aligned sequences but the same dataset sequences
398 assertFalse(copy.getSequenceAt(0) == protein.getSequenceAt(0));
399 assertFalse(copy.getSequenceAt(1) == protein.getSequenceAt(1));
400 assertSame(copy.getSequenceAt(0).getDatasetSequence(), protein
401 .getSequenceAt(0).getDatasetSequence());
402 assertSame(copy.getSequenceAt(1).getDatasetSequence(), protein
403 .getSequenceAt(1).getDatasetSequence());
405 // TODO should the copy constructor copy the dataset?
406 // or make a new one referring to the same dataset sequences??
407 assertNull(copy.getDataset());
408 // TODO test metadata is copied when AlignmentI is a dataset
410 // assertArrayEquals(copy.getDataset().getSequencesArray(), protein
411 // .getDataset().getSequencesArray());
415 * Test behaviour of createDataset
417 * @throws IOException
419 @Test(groups = "Functional")
420 public void testCreateDatasetAlignment() throws IOException
422 AlignmentI protein = new FormatAdapter().readFile(AA_SEQS_1,
423 DataSourceType.PASTE, FileFormat.Fasta);
425 * create a dataset sequence on first sequence
426 * leave the second without one
428 protein.getSequenceAt(0).createDatasetSequence();
429 assertNotNull(protein.getSequenceAt(0).getDatasetSequence());
430 assertNull(protein.getSequenceAt(1).getDatasetSequence());
433 * add a mapping to the alignment
435 AlignedCodonFrame acf = new AlignedCodonFrame();
436 protein.addCodonFrame(acf);
437 assertNull(protein.getDataset());
438 assertTrue(protein.getCodonFrames().contains(acf));
441 * create the alignment dataset
442 * note this creates sequence datasets where missing
443 * as a side-effect (in this case, on seq2
445 // TODO promote this method to AlignmentI
446 ((Alignment) protein).createDatasetAlignment();
448 AlignmentI ds = protein.getDataset();
450 // side-effect: dataset created on second sequence
451 assertNotNull(protein.getSequenceAt(1).getDatasetSequence());
452 // dataset alignment has references to dataset sequences
453 assertEquals(ds.getSequenceAt(0), protein.getSequenceAt(0)
454 .getDatasetSequence());
455 assertEquals(ds.getSequenceAt(1), protein.getSequenceAt(1)
456 .getDatasetSequence());
458 // codon frames should have been moved to the dataset
459 // getCodonFrames() should delegate to the dataset:
460 assertTrue(protein.getCodonFrames().contains(acf));
461 // prove the codon frames are indeed on the dataset:
462 assertTrue(ds.getCodonFrames().contains(acf));
465 @Test(groups = "Functional")
466 public void testAddCodonFrame()
468 AlignmentI align = new Alignment(new SequenceI[] {});
469 AlignedCodonFrame acf = new AlignedCodonFrame();
470 align.addCodonFrame(acf);
471 assertEquals(1, align.getCodonFrames().size());
472 assertTrue(align.getCodonFrames().contains(acf));
473 // can't add the same object twice:
474 align.addCodonFrame(acf);
475 assertEquals(1, align.getCodonFrames().size());
477 // create dataset alignment - mappings move to dataset
478 ((Alignment) align).createDatasetAlignment();
479 assertSame(align.getCodonFrames(), align.getDataset().getCodonFrames());
480 assertEquals(1, align.getCodonFrames().size());
482 AlignedCodonFrame acf2 = new AlignedCodonFrame();
483 align.addCodonFrame(acf2);
484 assertTrue(align.getDataset().getCodonFrames().contains(acf));
487 @Test(groups = "Functional")
488 public void getVisibleStartAndEndIndexTest()
490 Sequence seq = new Sequence("testSeq", "ABCDEFGHIJKLMNOPQRSTUVWXYZ");
491 AlignmentI align = new Alignment(new SequenceI[] { seq });
492 ArrayList<int[]> hiddenCols = new ArrayList<int[]>();
494 int[] startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
495 assertEquals(0, startEnd[0]);
496 assertEquals(25, startEnd[1]);
498 hiddenCols.add(new int[] { 0, 0 });
499 startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
500 assertEquals(1, startEnd[0]);
501 assertEquals(25, startEnd[1]);
503 hiddenCols.add(new int[] { 6, 9 });
504 hiddenCols.add(new int[] { 11, 12 });
505 startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
506 assertEquals(1, startEnd[0]);
507 assertEquals(25, startEnd[1]);
509 hiddenCols.add(new int[] { 24, 25 });
510 startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
511 System.out.println(startEnd[0] + " : " + startEnd[1]);
512 assertEquals(1, startEnd[0]);
513 assertEquals(23, startEnd[1]);