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.AppletFormatAdapter;
31 import jalview.io.FormatAdapter;
32 import jalview.util.MapList;
34 import java.io.IOException;
35 import java.util.ArrayList;
36 import java.util.Arrays;
37 import java.util.Iterator;
38 import java.util.List;
40 import org.testng.annotations.BeforeMethod;
41 import org.testng.annotations.Test;
44 * Unit tests for Alignment datamodel.
49 public class AlignmentTest
52 private static final String TEST_DATA =
54 "#=GS D.melanogaster.1 AC AY119185.1/838-902\n" +
55 "#=GS D.melanogaster.2 AC AC092237.1/57223-57161\n" +
56 "#=GS D.melanogaster.3 AC AY060611.1/560-627\n" +
57 "D.melanogaster.1 G.AGCC.CU...AUGAUCGA\n" +
58 "#=GR D.melanogaster.1 SS ................((((\n" +
59 "D.melanogaster.2 C.AUUCAACU.UAUGAGGAU\n" +
60 "#=GR D.melanogaster.2 SS ................((((\n" +
61 "D.melanogaster.3 G.UGGCGCU..UAUGACGCA\n" +
62 "#=GR D.melanogaster.3 SS (.(((...(....(((((((\n" +
65 private static final String AA_SEQS_1 =
71 private static final String CDNA_SEQS_1 =
72 ">Seq1Name/100-111\n" +
73 "AC-GG--CUC-CAA-CT\n" +
74 ">Seq2Name/200-211\n" +
75 "-CG-TTA--ACG---AAGT\n";
77 private static final String CDNA_SEQS_2 =
84 private AlignmentI al;
87 * Helper method to load an alignment and ensure dataset sequences are set up.
95 protected AlignmentI loadAlignment(final String data, String format)
98 AlignmentI a = new FormatAdapter().readFile(data,
99 AppletFormatAdapter.PASTE, format);
105 * Read in Stockholm format test data including secondary structure
108 @BeforeMethod(alwaysRun = true)
109 public void setUp() throws IOException
111 al = loadAlignment(TEST_DATA, "STH");
113 for (AlignmentAnnotation ann : al.getAlignmentAnnotation())
115 ann.setCalcId("CalcIdFor" + al.getSequenceAt(i).getName());
121 * Test method that returns annotations that match on calcId.
123 @Test(groups = { "Functional" })
124 public void testFindAnnotation_byCalcId()
126 Iterable<AlignmentAnnotation> anns = al
127 .findAnnotation("CalcIdForD.melanogaster.2");
128 Iterator<AlignmentAnnotation> iter = anns.iterator();
129 assertTrue(iter.hasNext());
130 AlignmentAnnotation ann = iter.next();
131 assertEquals("D.melanogaster.2", ann.sequenceRef.getName());
132 assertFalse(iter.hasNext());
135 @Test(groups = { "Functional" })
136 public void testDeleteAllAnnotations_includingAutocalculated()
138 AlignmentAnnotation aa = new AlignmentAnnotation("Consensus",
140 aa.autoCalculated = true;
141 al.addAnnotation(aa);
142 AlignmentAnnotation[] anns = al.getAlignmentAnnotation();
143 assertEquals("Wrong number of annotations before deleting", 4,
145 al.deleteAllAnnotations(true);
146 assertEquals("Not all deleted", 0, al.getAlignmentAnnotation().length);
149 @Test(groups = { "Functional" })
150 public void testDeleteAllAnnotations_excludingAutocalculated()
152 AlignmentAnnotation aa = new AlignmentAnnotation("Consensus",
154 aa.autoCalculated = true;
155 al.addAnnotation(aa);
156 AlignmentAnnotation[] anns = al.getAlignmentAnnotation();
157 assertEquals("Wrong number of annotations before deleting", 4,
159 al.deleteAllAnnotations(false);
160 assertEquals("Not just one annotation left", 1,
161 al.getAlignmentAnnotation().length);
165 * Tests for realigning as per a supplied alignment: Dna as Dna.
167 * Note: AlignedCodonFrame's state variables are named for protein-to-cDNA
168 * mapping, but can be exploited for a general 'sequence-to-sequence' mapping
171 * @throws IOException
173 @Test(groups = { "Functional" })
174 public void testAlignAs_dnaAsDna() throws IOException
177 AlignmentI al1 = loadAlignment(CDNA_SEQS_1, "FASTA");
179 AlignmentI al2 = loadAlignment(CDNA_SEQS_2, "FASTA");
182 * Make mappings between sequences. The 'aligned cDNA' is playing the role
183 * of what would normally be protein here.
185 makeMappings(al1, al2);
187 ((Alignment) al2).alignAs(al1, false, true);
188 assertEquals("GC-TC--GUC-GTACT", al2.getSequenceAt(0)
189 .getSequenceAsString());
190 assertEquals("-GG-GTC--AGG--CAGT", al2.getSequenceAt(1)
191 .getSequenceAsString());
195 * Aligning protein from cDNA.
197 * @throws IOException
199 @Test(groups = { "Functional" })
200 public void testAlignAs_proteinAsCdna() throws IOException
202 // see also AlignmentUtilsTests
203 AlignmentI al1 = loadAlignment(CDNA_SEQS_1, "FASTA");
204 AlignmentI al2 = loadAlignment(AA_SEQS_1, "FASTA");
205 makeMappings(al1, al2);
207 // Fudge - alignProteinAsCdna expects mappings to be on protein
208 al2.getCodonFrames().addAll(al1.getCodonFrames());
210 ((Alignment) al2).alignAs(al1, false, true);
211 assertEquals("K-Q-Y-L-", al2.getSequenceAt(0).getSequenceAsString());
212 assertEquals("-R-F-P-W", al2.getSequenceAt(1).getSequenceAsString());
216 * Test aligning cdna as per protein alignment.
218 * @throws IOException
220 @Test(groups = { "Functional" }, enabled = true)
221 // TODO review / update this test after redesign of alignAs method
222 public void testAlignAs_cdnaAsProtein() throws IOException
225 * Load alignments and add mappings for cDNA to protein
227 AlignmentI al1 = loadAlignment(CDNA_SEQS_1, "FASTA");
228 AlignmentI al2 = loadAlignment(AA_SEQS_1, "FASTA");
229 makeMappings(al1, al2);
232 * Realign DNA; currently keeping existing gaps in introns only
234 ((Alignment) al1).alignAs(al2, false, true);
235 assertEquals("ACG---GCUCCA------ACT---", al1.getSequenceAt(0)
236 .getSequenceAsString());
237 assertEquals("---CGT---TAACGA---AGT---", al1.getSequenceAt(1)
238 .getSequenceAsString());
242 * Test aligning cdna as per protein - single sequences
244 * @throws IOException
246 @Test(groups = { "Functional" }, enabled = true)
247 // TODO review / update this test after redesign of alignAs method
248 public void testAlignAs_cdnaAsProtein_singleSequence() throws IOException
251 * simple case insert one gap
253 verifyAlignAs(">dna\nCAAaaa\n", ">protein\nQ-K\n", "CAA---aaa");
256 * simple case but with sequence offsets
258 verifyAlignAs(">dna/5-10\nCAAaaa\n", ">protein/20-21\nQ-K\n",
262 * insert gaps as per protein, drop gaps within codons
264 verifyAlignAs(">dna/10-18\nCA-Aa-aa--AGA\n", ">aa/6-8\n-Q-K--R\n",
265 "---CAA---aaa------AGA");
269 * Helper method that makes mappings and then aligns the first alignment as
275 * @throws IOException
277 public void verifyAlignAs(String fromSeqs, String toSeqs, String expected)
281 * Load alignments and add mappings from nucleotide to protein (or from
282 * first to second if both the same type)
284 AlignmentI al1 = loadAlignment(fromSeqs, "FASTA");
285 AlignmentI al2 = loadAlignment(toSeqs, "FASTA");
286 makeMappings(al1, al2);
289 * Realign DNA; currently keeping existing gaps in introns only
291 ((Alignment) al1).alignAs(al2, false, true);
292 assertEquals(expected, al1.getSequenceAt(0).getSequenceAsString());
296 * Helper method to make mappings between sequences, and add the mappings to
297 * the 'mapped from' alignment
302 public void makeMappings(AlignmentI alFrom, AlignmentI alTo)
304 int ratio = (alFrom.isNucleotide() == alTo.isNucleotide() ? 1 : 3);
306 AlignedCodonFrame acf = new AlignedCodonFrame();
308 for (int i = 0; i < alFrom.getHeight(); i++)
310 SequenceI seqFrom = alFrom.getSequenceAt(i);
311 SequenceI seqTo = alTo.getSequenceAt(i);
312 MapList ml = new MapList(new int[] { seqFrom.getStart(),
314 new int[] { seqTo.getStart(), seqTo.getEnd() }, ratio, 1);
315 acf.addMap(seqFrom, seqTo, ml);
319 * not sure whether mappings 'belong' or protein or nucleotide
320 * alignment, so adding to both ;~)
322 alFrom.addCodonFrame(acf);
323 alTo.addCodonFrame(acf);
327 * Test aligning dna as per protein alignment, for the case where there are
328 * introns (i.e. some dna sites have no mapping from a peptide).
330 * @throws IOException
332 @Test(groups = { "Functional" }, enabled = false)
333 // TODO review / update this test after redesign of alignAs method
334 public void testAlignAs_dnaAsProtein_withIntrons() throws IOException
337 * Load alignments and add mappings for cDNA to protein
339 String dna1 = "A-Aa-gG-GCC-cT-TT";
340 String dna2 = "c--CCGgg-TT--T-AA-A";
341 AlignmentI al1 = loadAlignment(">Dna1/6-17\n" + dna1
342 + "\n>Dna2/20-31\n" + dna2 + "\n", "FASTA");
343 AlignmentI al2 = loadAlignment(
344 ">Pep1/7-9\n-P--YK\n>Pep2/11-13\nG-T--F\n", "FASTA");
345 AlignedCodonFrame acf = new AlignedCodonFrame();
346 // Seq1 has intron at dna positions 3,4,9 so splice is AAG GCC TTT
347 // Seq2 has intron at dna positions 1,5,6 so splice is CCG TTT AAA
348 MapList ml1 = new MapList(new int[] { 6, 7, 10, 13, 15, 17 }, new int[]
350 acf.addMap(al1.getSequenceAt(0), al2.getSequenceAt(0), ml1);
351 MapList ml2 = new MapList(new int[] { 21, 23, 26, 31 }, new int[] { 11,
353 acf.addMap(al1.getSequenceAt(1), al2.getSequenceAt(1), ml2);
354 al2.addCodonFrame(acf);
357 * Align ignoring gaps in dna introns and exons
359 ((Alignment) al1).alignAs(al2, false, false);
360 assertEquals("---AAagG------GCCcTTT", al1.getSequenceAt(0)
361 .getSequenceAsString());
362 // note 1 gap in protein corresponds to 'gg-' in DNA (3 positions)
363 assertEquals("cCCGgg-TTT------AAA", al1.getSequenceAt(1)
364 .getSequenceAsString());
367 * Reset and realign, preserving gaps in dna introns and exons
369 al1.getSequenceAt(0).setSequence(dna1);
370 al1.getSequenceAt(1).setSequence(dna2);
371 ((Alignment) al1).alignAs(al2, true, true);
372 // String dna1 = "A-Aa-gG-GCC-cT-TT";
373 // String dna2 = "c--CCGgg-TT--T-AA-A";
374 // assumption: we include 'the greater of' protein/dna gap lengths, not both
375 assertEquals("---A-Aa-gG------GCC-cT-TT", al1.getSequenceAt(0)
376 .getSequenceAsString());
377 assertEquals("c--CCGgg-TT--T------AA-A", al1.getSequenceAt(1)
378 .getSequenceAsString());
381 @Test(groups = "Functional")
382 public void testCopyConstructor() throws IOException
384 AlignmentI protein = loadAlignment(AA_SEQS_1, FormatAdapter.PASTE);
385 // create sequence and alignment datasets
386 protein.setDataset(null);
387 AlignedCodonFrame acf = new AlignedCodonFrame();
388 List<AlignedCodonFrame> acfList = Arrays.asList(new AlignedCodonFrame[]
390 protein.getDataset().setCodonFrames(acfList);
391 AlignmentI copy = new Alignment(protein);
394 * copy has different aligned sequences but the same dataset sequences
396 assertFalse(copy.getSequenceAt(0) == protein.getSequenceAt(0));
397 assertFalse(copy.getSequenceAt(1) == protein.getSequenceAt(1));
398 assertSame(copy.getSequenceAt(0).getDatasetSequence(), protein
399 .getSequenceAt(0).getDatasetSequence());
400 assertSame(copy.getSequenceAt(1).getDatasetSequence(), protein
401 .getSequenceAt(1).getDatasetSequence());
403 // TODO should the copy constructor copy the dataset?
404 // or make a new one referring to the same dataset sequences??
405 assertNull(copy.getDataset());
406 // TODO test metadata is copied when AlignmentI is a dataset
408 // assertArrayEquals(copy.getDataset().getSequencesArray(), protein
409 // .getDataset().getSequencesArray());
413 * Test behaviour of createDataset
415 * @throws IOException
417 @Test(groups = "Functional")
418 public void testCreateDatasetAlignment() throws IOException
420 AlignmentI protein = new FormatAdapter().readFile(AA_SEQS_1,
421 AppletFormatAdapter.PASTE, "FASTA");
423 * create a dataset sequence on first sequence
424 * leave the second without one
426 protein.getSequenceAt(0).createDatasetSequence();
427 assertNotNull(protein.getSequenceAt(0).getDatasetSequence());
428 assertNull(protein.getSequenceAt(1).getDatasetSequence());
431 * add a mapping to the alignment
433 AlignedCodonFrame acf = new AlignedCodonFrame();
434 protein.addCodonFrame(acf);
435 assertNull(protein.getDataset());
436 assertTrue(protein.getCodonFrames().contains(acf));
439 * create the alignment dataset
440 * note this creates sequence datasets where missing
441 * as a side-effect (in this case, on seq2
443 // TODO promote this method to AlignmentI
444 ((Alignment) protein).createDatasetAlignment();
446 AlignmentI ds = protein.getDataset();
448 // side-effect: dataset created on second sequence
449 assertNotNull(protein.getSequenceAt(1).getDatasetSequence());
450 // dataset alignment has references to dataset sequences
451 assertEquals(ds.getSequenceAt(0), protein.getSequenceAt(0)
452 .getDatasetSequence());
453 assertEquals(ds.getSequenceAt(1), protein.getSequenceAt(1)
454 .getDatasetSequence());
456 // codon frames should have been moved to the dataset
457 // getCodonFrames() should delegate to the dataset:
458 assertTrue(protein.getCodonFrames().contains(acf));
459 // prove the codon frames are indeed on the dataset:
460 assertTrue(ds.getCodonFrames().contains(acf));
463 @Test(groups = "Functional")
464 public void testAddCodonFrame()
466 AlignmentI align = new Alignment(new SequenceI[] {});
467 AlignedCodonFrame acf = new AlignedCodonFrame();
468 align.addCodonFrame(acf);
469 assertEquals(1, align.getCodonFrames().size());
470 assertTrue(align.getCodonFrames().contains(acf));
471 // can't add the same object twice:
472 align.addCodonFrame(acf);
473 assertEquals(1, align.getCodonFrames().size());
475 // create dataset alignment - mappings move to dataset
476 ((Alignment) align).createDatasetAlignment();
477 assertSame(align.getCodonFrames(), align.getDataset().getCodonFrames());
478 assertEquals(1, align.getCodonFrames().size());
480 AlignedCodonFrame acf2 = new AlignedCodonFrame();
481 align.addCodonFrame(acf2);
482 assertTrue(align.getDataset().getCodonFrames().contains(acf));
485 @Test(groups = "Functional")
486 public void getVisibleStartAndEndIndexTest()
488 Sequence seq = new Sequence("testSeq", "ABCDEFGHIJKLMNOPQRSTUVWXYZ");
489 AlignmentI align = new Alignment(new SequenceI[] { seq });
490 ArrayList<int[]> hiddenCols = new ArrayList<int[]>();
492 int[] startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
493 assertEquals(0, startEnd[0]);
494 assertEquals(25, startEnd[1]);
496 hiddenCols.add(new int[] { 0, 0 });
497 startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
498 assertEquals(1, startEnd[0]);
499 assertEquals(25, startEnd[1]);
501 hiddenCols.add(new int[] { 6, 9 });
502 hiddenCols.add(new int[] { 11, 12 });
503 startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
504 assertEquals(1, startEnd[0]);
505 assertEquals(25, startEnd[1]);
507 hiddenCols.add(new int[] { 24, 25 });
508 startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
509 System.out.println(startEnd[0] + " : " + startEnd[1]);
510 assertEquals(1, startEnd[0]);
511 assertEquals(23, startEnd[1]);