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.datamodel.AlignedCodonFrame.SequenceToSequenceMapping;
31 import jalview.io.AppletFormatAdapter;
32 import jalview.io.FormatAdapter;
33 import jalview.util.MapList;
35 import java.io.IOException;
36 import java.util.ArrayList;
37 import java.util.Arrays;
38 import java.util.Iterator;
39 import java.util.List;
41 import org.testng.Assert;
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, String format)
100 AlignmentI a = new FormatAdapter().readFile(data,
101 AppletFormatAdapter.PASTE, format);
107 * assert wrapper: tests all references in the given alignment are consistent
111 public static void assertAlignmentDatasetRefs(AlignmentI alignment)
113 verifyAlignmentDatasetRefs(alignment, true, null);
117 * assert wrapper: tests all references in the given alignment are consistent
121 * - prefixed to any assert failed messages
123 public static void assertAlignmentDatasetRefs(AlignmentI alignment,
126 verifyAlignmentDatasetRefs(alignment, true, message);
130 * verify sequence and dataset references are properly contained within
134 * - the alignmentI object to verify (either alignment or dataset)
136 * - when set, testng assertions are raised.
138 * - null or a string message to prepend to the assert failed messages.
139 * @return true if alignment references were in order, otherwise false.
141 public static boolean verifyAlignmentDatasetRefs(AlignmentI alignment,
142 boolean raiseAssert, String message)
144 if (message==null) { message = ""; }
145 if (alignment == null)
149 Assert.fail(message+"Alignment for verification was null.");
153 if (alignment.getDataset() != null)
155 AlignmentI dataset = alignment.getDataset();
156 // check all alignment sequences have their dataset within the dataset
157 for (SequenceI seq : alignment.getSequences())
159 SequenceI seqds = seq.getDatasetSequence();
160 if (seqds.getDatasetSequence() != null)
164 Assert.fail(message+" Alignment contained a sequence who's dataset sequence has a second dataset reference.");
168 if (dataset.findIndex(seqds) == -1)
172 Assert.fail(message+" Alignment contained a sequence who's dataset sequence was not in the dataset.");
177 return verifyAlignmentDatasetRefs(alignment.getDataset(), raiseAssert, message);
182 // verify all dataset sequences
183 for (SequenceI seqds : alignment.getSequences())
186 if (seqds.getDatasetSequence() != null)
190 Assert.fail(message+" Dataset contained a sequence with non-null dataset reference (ie not a dataset sequence!)");
194 int foundp = alignment.findIndex(seqds);
200 + " Dataset sequence array contains a reference at "
201 + dsp + " to a sequence first seen at " + foundp + " ("
202 + seqds.toString() + ")");
206 if (seqds.getDBRefs() != null)
208 for (DBRefEntry dbr : seqds.getDBRefs())
210 if (dbr.getMap() != null)
212 SequenceI seqdbrmapto = dbr.getMap().getTo();
213 if (seqdbrmapto != null)
215 if (seqdbrmapto.getDatasetSequence() != null)
219 Assert.fail(message+" DBRefEntry for sequence in alignment had map to sequence which was not a dataset sequence");
224 if (alignment.findIndex(dbr.getMap().getTo()) == -1)
228 Assert.fail(message+" DBRefEntry for sequence in alignment had map to sequence not in dataset");
237 // finally, verify codonmappings involve only dataset sequences.
238 if (alignment.getCodonFrames() != null)
240 for (AlignedCodonFrame alc : alignment.getCodonFrames())
242 for (SequenceToSequenceMapping ssm : alc.getMappings())
244 if (ssm.getFromSeq().getDatasetSequence() != null)
248 Assert.fail(message+" CodonFrame-SSM-FromSeq is not a dataset sequence");
252 if (alignment.findIndex(ssm.getFromSeq()) == -1)
257 Assert.fail(message+" CodonFrame-SSM-FromSeq is not contained in dataset");
261 if (ssm.getMapping().getTo().getDatasetSequence() != null)
265 Assert.fail(message+" CodonFrame-SSM-Mapping-ToSeq is not a dataset sequence");
269 if (alignment.findIndex(ssm.getMapping().getTo()) == -1)
274 Assert.fail(message+" CodonFrame-SSM-Mapping-ToSeq is not contained in dataset");
282 return true; // all relationships verified!
286 * call verifyAlignmentDatasetRefs with and without assertion raising enabled,
287 * to check expected pass/fail actually occurs in both conditions
293 private void assertVerifyAlignment(AlignmentI al, boolean expected,
301 Assert.assertTrue(verifyAlignmentDatasetRefs(al, true, null),
302 "Valid test alignment failed when raiseAsserts enabled:"
304 } catch (AssertionError ae)
306 ae.printStackTrace();
308 "Valid test alignment raised assertion errors when raiseAsserts enabled: "
311 // also check validation passes with asserts disabled
312 Assert.assertTrue(verifyAlignmentDatasetRefs(al, false, null),
313 "Valid test alignment tested false when raiseAsserts disabled:"
318 boolean assertRaised = false;
321 verifyAlignmentDatasetRefs(al, true, null);
322 } catch (AssertionError ae)
324 // expected behaviour
329 Assert.fail("Invalid test alignment passed when raiseAsserts enabled:"
332 // also check validation passes with asserts disabled
333 Assert.assertFalse(verifyAlignmentDatasetRefs(al, false, null),
334 "Invalid test alignment tested true when raiseAsserts disabled:"
338 @Test(groups = { "Functional" })
339 public void testVerifyAlignmentDatasetRefs()
341 SequenceI sq1 = new Sequence("sq1", "ASFDD"), sq2 = new Sequence("sq2",
344 // construct simple valid alignment dataset
345 Alignment al = new Alignment(new SequenceI[] {
347 // expect this to pass
348 assertVerifyAlignment(al, true, "Simple valid alignment didn't verify");
350 // check test for sequence->datasetSequence validity
351 sq1.setDatasetSequence(sq2);
352 assertVerifyAlignment(
355 "didn't detect dataset sequence with a dataset sequence reference.");
357 sq1.setDatasetSequence(null);
358 assertVerifyAlignment(
361 "didn't reinstate validity after nulling dataset sequence dataset reference");
363 // now create dataset and check again
364 al.createDatasetAlignment();
365 assertNotNull(al.getDataset());
367 assertVerifyAlignment(al, true,
368 "verify failed after createDatasetAlignment");
370 // create a dbref on sq1 with a sequence ref to sq2
371 DBRefEntry dbrs1tos2 = new DBRefEntry("UNIPROT", "1", "Q111111");
372 dbrs1tos2.setMap(new Mapping(sq2.getDatasetSequence(),
373 new int[] { 1, 5 }, new int[] { 2, 6 }, 1, 1));
374 sq1.getDatasetSequence().addDBRef(dbrs1tos2);
375 assertVerifyAlignment(al, true,
376 "verify failed after addition of valid DBRefEntry/map");
377 // now create a dbref on a new sequence which maps to another sequence
378 // outside of the dataset
379 SequenceI sqout = new Sequence("sqout", "ututututucagcagcag"), sqnew = new Sequence(
381 DBRefEntry sqnewsqout = new DBRefEntry("ENAFOO", "1", "R000001");
382 sqnewsqout.setMap(new Mapping(sqout, new int[] { 1, 6 }, new int[] { 1,
384 al.getDataset().addSequence(sqnew);
386 assertVerifyAlignment(al, true,
387 "verify failed after addition of new sequence to dataset");
388 // now start checking exception conditions
389 sqnew.addDBRef(sqnewsqout);
390 assertVerifyAlignment(
393 "verify passed when a dbref with map to sequence outside of dataset was added");
394 // make the verify pass by adding the outsider back in
395 al.getDataset().addSequence(sqout);
396 assertVerifyAlignment(al, true,
397 "verify should have passed after adding dbref->to sequence in to dataset");
398 // and now the same for a codon mapping...
399 SequenceI sqanotherout = new Sequence("sqanotherout",
400 "aggtutaggcagcagcag");
402 AlignedCodonFrame alc = new AlignedCodonFrame();
403 alc.addMap(sqanotherout, sqnew, new MapList(new int[] { 1, 6 },
404 new int[] { 1, 18 }, 3, 1));
406 al.addCodonFrame(alc);
407 Assert.assertEquals(al.getDataset().getCodonFrames().size(), 1);
409 assertVerifyAlignment(
412 "verify passed when alCodonFrame mapping to sequence outside of dataset was added");
413 // make the verify pass by adding the outsider back in
414 al.getDataset().addSequence(sqanotherout);
415 assertVerifyAlignment(
418 "verify should have passed once all sequences involved in alCodonFrame were added to dataset");
419 al.getDataset().addSequence(sqanotherout);
420 assertVerifyAlignment(al, false,
421 "verify should have failed when a sequence was added twice to the dataset");
422 al.getDataset().deleteSequence(sqanotherout);
423 assertVerifyAlignment(al, true,
424 "verify should have passed after duplicate entry for sequence was removed");
428 * Read in Stockholm format test data including secondary structure
431 @BeforeMethod(alwaysRun = true)
432 public void setUp() throws IOException
434 al = loadAlignment(TEST_DATA, "STH");
436 for (AlignmentAnnotation ann : al.getAlignmentAnnotation())
438 ann.setCalcId("CalcIdFor" + al.getSequenceAt(i).getName());
444 * Test method that returns annotations that match on calcId.
446 @Test(groups = { "Functional" })
447 public void testFindAnnotation_byCalcId()
449 Iterable<AlignmentAnnotation> anns = al
450 .findAnnotation("CalcIdForD.melanogaster.2");
451 Iterator<AlignmentAnnotation> iter = anns.iterator();
452 assertTrue(iter.hasNext());
453 AlignmentAnnotation ann = iter.next();
454 assertEquals("D.melanogaster.2", ann.sequenceRef.getName());
455 assertFalse(iter.hasNext());
458 @Test(groups = { "Functional" })
459 public void testDeleteAllAnnotations_includingAutocalculated()
461 AlignmentAnnotation aa = new AlignmentAnnotation("Consensus",
463 aa.autoCalculated = true;
464 al.addAnnotation(aa);
465 AlignmentAnnotation[] anns = al.getAlignmentAnnotation();
466 assertEquals("Wrong number of annotations before deleting", 4,
468 al.deleteAllAnnotations(true);
469 assertEquals("Not all deleted", 0, al.getAlignmentAnnotation().length);
472 @Test(groups = { "Functional" })
473 public void testDeleteAllAnnotations_excludingAutocalculated()
475 AlignmentAnnotation aa = new AlignmentAnnotation("Consensus",
477 aa.autoCalculated = true;
478 al.addAnnotation(aa);
479 AlignmentAnnotation[] anns = al.getAlignmentAnnotation();
480 assertEquals("Wrong number of annotations before deleting", 4,
482 al.deleteAllAnnotations(false);
483 assertEquals("Not just one annotation left", 1,
484 al.getAlignmentAnnotation().length);
488 * Tests for realigning as per a supplied alignment: Dna as Dna.
490 * Note: AlignedCodonFrame's state variables are named for protein-to-cDNA
491 * mapping, but can be exploited for a general 'sequence-to-sequence' mapping
494 * @throws IOException
496 @Test(groups = { "Functional" })
497 public void testAlignAs_dnaAsDna() throws IOException
500 AlignmentI al1 = loadAlignment(CDNA_SEQS_1, "FASTA");
502 AlignmentI al2 = loadAlignment(CDNA_SEQS_2, "FASTA");
505 * Make mappings between sequences. The 'aligned cDNA' is playing the role
506 * of what would normally be protein here.
508 makeMappings(al1, al2);
510 ((Alignment) al2).alignAs(al1, false, true);
511 assertEquals("GC-TC--GUC-GTACT", al2.getSequenceAt(0)
512 .getSequenceAsString());
513 assertEquals("-GG-GTC--AGG--CAGT", al2.getSequenceAt(1)
514 .getSequenceAsString());
518 * Aligning protein from cDNA.
520 * @throws IOException
522 @Test(groups = { "Functional" })
523 public void testAlignAs_proteinAsCdna() throws IOException
525 // see also AlignmentUtilsTests
526 AlignmentI al1 = loadAlignment(CDNA_SEQS_1, "FASTA");
527 AlignmentI al2 = loadAlignment(AA_SEQS_1, "FASTA");
528 makeMappings(al1, al2);
530 // Fudge - alignProteinAsCdna expects mappings to be on protein
531 al2.getCodonFrames().addAll(al1.getCodonFrames());
533 ((Alignment) al2).alignAs(al1, false, true);
534 assertEquals("K-Q-Y-L-", al2.getSequenceAt(0).getSequenceAsString());
535 assertEquals("-R-F-P-W", al2.getSequenceAt(1).getSequenceAsString());
539 * Test aligning cdna as per protein alignment.
541 * @throws IOException
543 @Test(groups = { "Functional" }, enabled = true)
544 // TODO review / update this test after redesign of alignAs method
545 public void testAlignAs_cdnaAsProtein() throws IOException
548 * Load alignments and add mappings for cDNA to protein
550 AlignmentI al1 = loadAlignment(CDNA_SEQS_1, "FASTA");
551 AlignmentI al2 = loadAlignment(AA_SEQS_1, "FASTA");
552 makeMappings(al1, al2);
555 * Realign DNA; currently keeping existing gaps in introns only
557 ((Alignment) al1).alignAs(al2, false, true);
558 assertEquals("ACG---GCUCCA------ACT---", al1.getSequenceAt(0)
559 .getSequenceAsString());
560 assertEquals("---CGT---TAACGA---AGT---", al1.getSequenceAt(1)
561 .getSequenceAsString());
565 * Test aligning cdna as per protein - single sequences
567 * @throws IOException
569 @Test(groups = { "Functional" }, enabled = true)
570 // TODO review / update this test after redesign of alignAs method
571 public void testAlignAs_cdnaAsProtein_singleSequence() throws IOException
574 * simple case insert one gap
576 verifyAlignAs(">dna\nCAAaaa\n", ">protein\nQ-K\n", "CAA---aaa");
579 * simple case but with sequence offsets
581 verifyAlignAs(">dna/5-10\nCAAaaa\n", ">protein/20-21\nQ-K\n",
585 * insert gaps as per protein, drop gaps within codons
587 verifyAlignAs(">dna/10-18\nCA-Aa-aa--AGA\n", ">aa/6-8\n-Q-K--R\n",
588 "---CAA---aaa------AGA");
592 * Helper method that makes mappings and then aligns the first alignment as
598 * @throws IOException
600 public void verifyAlignAs(String fromSeqs, String toSeqs, String expected)
604 * Load alignments and add mappings from nucleotide to protein (or from
605 * first to second if both the same type)
607 AlignmentI al1 = loadAlignment(fromSeqs, "FASTA");
608 AlignmentI al2 = loadAlignment(toSeqs, "FASTA");
609 makeMappings(al1, al2);
612 * Realign DNA; currently keeping existing gaps in introns only
614 ((Alignment) al1).alignAs(al2, false, true);
615 assertEquals(expected, al1.getSequenceAt(0).getSequenceAsString());
619 * Helper method to make mappings between sequences, and add the mappings to
620 * the 'mapped from' alignment
625 public void makeMappings(AlignmentI alFrom, AlignmentI alTo)
627 int ratio = (alFrom.isNucleotide() == alTo.isNucleotide() ? 1 : 3);
629 AlignedCodonFrame acf = new AlignedCodonFrame();
631 for (int i = 0; i < alFrom.getHeight(); i++)
633 SequenceI seqFrom = alFrom.getSequenceAt(i);
634 SequenceI seqTo = alTo.getSequenceAt(i);
635 MapList ml = new MapList(new int[] { seqFrom.getStart(),
637 new int[] { seqTo.getStart(), seqTo.getEnd() }, ratio, 1);
638 acf.addMap(seqFrom, seqTo, ml);
642 * not sure whether mappings 'belong' or protein or nucleotide
643 * alignment, so adding to both ;~)
645 alFrom.addCodonFrame(acf);
646 alTo.addCodonFrame(acf);
650 * Test aligning dna as per protein alignment, for the case where there are
651 * introns (i.e. some dna sites have no mapping from a peptide).
653 * @throws IOException
655 @Test(groups = { "Functional" }, enabled = false)
656 // TODO review / update this test after redesign of alignAs method
657 public void testAlignAs_dnaAsProtein_withIntrons() throws IOException
660 * Load alignments and add mappings for cDNA to protein
662 String dna1 = "A-Aa-gG-GCC-cT-TT";
663 String dna2 = "c--CCGgg-TT--T-AA-A";
664 AlignmentI al1 = loadAlignment(">Dna1/6-17\n" + dna1
665 + "\n>Dna2/20-31\n" + dna2 + "\n", "FASTA");
666 AlignmentI al2 = loadAlignment(
667 ">Pep1/7-9\n-P--YK\n>Pep2/11-13\nG-T--F\n", "FASTA");
668 AlignedCodonFrame acf = new AlignedCodonFrame();
669 // Seq1 has intron at dna positions 3,4,9 so splice is AAG GCC TTT
670 // Seq2 has intron at dna positions 1,5,6 so splice is CCG TTT AAA
671 MapList ml1 = new MapList(new int[] { 6, 7, 10, 13, 15, 17 }, new int[]
673 acf.addMap(al1.getSequenceAt(0), al2.getSequenceAt(0), ml1);
674 MapList ml2 = new MapList(new int[] { 21, 23, 26, 31 }, new int[] { 11,
676 acf.addMap(al1.getSequenceAt(1), al2.getSequenceAt(1), ml2);
677 al2.addCodonFrame(acf);
680 * Align ignoring gaps in dna introns and exons
682 ((Alignment) al1).alignAs(al2, false, false);
683 assertEquals("---AAagG------GCCcTTT", al1.getSequenceAt(0)
684 .getSequenceAsString());
685 // note 1 gap in protein corresponds to 'gg-' in DNA (3 positions)
686 assertEquals("cCCGgg-TTT------AAA", al1.getSequenceAt(1)
687 .getSequenceAsString());
690 * Reset and realign, preserving gaps in dna introns and exons
692 al1.getSequenceAt(0).setSequence(dna1);
693 al1.getSequenceAt(1).setSequence(dna2);
694 ((Alignment) al1).alignAs(al2, true, true);
695 // String dna1 = "A-Aa-gG-GCC-cT-TT";
696 // String dna2 = "c--CCGgg-TT--T-AA-A";
697 // assumption: we include 'the greater of' protein/dna gap lengths, not both
698 assertEquals("---A-Aa-gG------GCC-cT-TT", al1.getSequenceAt(0)
699 .getSequenceAsString());
700 assertEquals("c--CCGgg-TT--T------AA-A", al1.getSequenceAt(1)
701 .getSequenceAsString());
704 @Test(groups = "Functional")
705 public void testCopyConstructor() throws IOException
707 AlignmentI protein = loadAlignment(AA_SEQS_1, FormatAdapter.PASTE);
708 // create sequence and alignment datasets
709 protein.setDataset(null);
710 AlignedCodonFrame acf = new AlignedCodonFrame();
711 List<AlignedCodonFrame> acfList = Arrays.asList(new AlignedCodonFrame[]
713 protein.getDataset().setCodonFrames(acfList);
714 AlignmentI copy = new Alignment(protein);
717 * copy has different aligned sequences but the same dataset sequences
719 assertFalse(copy.getSequenceAt(0) == protein.getSequenceAt(0));
720 assertFalse(copy.getSequenceAt(1) == protein.getSequenceAt(1));
721 assertSame(copy.getSequenceAt(0).getDatasetSequence(), protein
722 .getSequenceAt(0).getDatasetSequence());
723 assertSame(copy.getSequenceAt(1).getDatasetSequence(), protein
724 .getSequenceAt(1).getDatasetSequence());
726 // TODO should the copy constructor copy the dataset?
727 // or make a new one referring to the same dataset sequences??
728 assertNull(copy.getDataset());
729 // TODO test metadata is copied when AlignmentI is a dataset
731 // assertArrayEquals(copy.getDataset().getSequencesArray(), protein
732 // .getDataset().getSequencesArray());
736 * Test behaviour of createDataset
738 * @throws IOException
740 @Test(groups = "Functional")
741 public void testCreateDatasetAlignment() throws IOException
743 AlignmentI protein = new FormatAdapter().readFile(AA_SEQS_1,
744 AppletFormatAdapter.PASTE, "FASTA");
746 * create a dataset sequence on first sequence
747 * leave the second without one
749 protein.getSequenceAt(0).createDatasetSequence();
750 assertNotNull(protein.getSequenceAt(0).getDatasetSequence());
751 assertNull(protein.getSequenceAt(1).getDatasetSequence());
754 * add a mapping to the alignment
756 AlignedCodonFrame acf = new AlignedCodonFrame();
757 protein.addCodonFrame(acf);
758 assertNull(protein.getDataset());
759 assertTrue(protein.getCodonFrames().contains(acf));
762 * create the alignment dataset
763 * note this creates sequence datasets where missing
764 * as a side-effect (in this case, on seq2
766 // TODO promote this method to AlignmentI
767 ((Alignment) protein).createDatasetAlignment();
769 AlignmentI ds = protein.getDataset();
771 // side-effect: dataset created on second sequence
772 assertNotNull(protein.getSequenceAt(1).getDatasetSequence());
773 // dataset alignment has references to dataset sequences
774 assertEquals(ds.getSequenceAt(0), protein.getSequenceAt(0)
775 .getDatasetSequence());
776 assertEquals(ds.getSequenceAt(1), protein.getSequenceAt(1)
777 .getDatasetSequence());
779 // codon frames should have been moved to the dataset
780 // getCodonFrames() should delegate to the dataset:
781 assertTrue(protein.getCodonFrames().contains(acf));
782 // prove the codon frames are indeed on the dataset:
783 assertTrue(ds.getCodonFrames().contains(acf));
787 * tests the addition of *all* sequences referred to by a sequence being added
790 @Test(groups = "Functional")
791 public void testCreateDatasetAlignmentWithMappedToSeqs()
793 // Alignment with two sequences, gapped.
794 SequenceI sq1 = new Sequence("sq1", "A--SDF");
795 SequenceI sq2 = new Sequence("sq2", "G--TRQ");
797 // cross-references to two more sequences.
798 DBRefEntry dbr = new DBRefEntry("SQ1", "", "sq3");
799 SequenceI sq3 = new Sequence("sq3", "VWANG");
800 dbr.setMap(new Mapping(sq3, new MapList(new int[] { 1, 4 }, new int[] {
804 SequenceI sq4 = new Sequence("sq4", "ERKWI");
805 DBRefEntry dbr2 = new DBRefEntry("SQ2", "", "sq4");
806 dbr2.setMap(new Mapping(sq4, new MapList(new int[] { 1, 4 }, new int[] {
809 // and a 1:1 codonframe mapping between them.
810 AlignedCodonFrame alc = new AlignedCodonFrame();
811 alc.addMap(sq1, sq2, new MapList(new int[] { 1, 4 },
812 new int[] { 1, 4 }, 1, 1));
814 AlignmentI protein = new Alignment(new SequenceI[] { sq1, sq2 });
817 * create the alignment dataset
818 * note this creates sequence datasets where missing
819 * as a side-effect (in this case, on seq2
822 // TODO promote this method to AlignmentI
823 ((Alignment) protein).createDatasetAlignment();
825 AlignmentI ds = protein.getDataset();
827 // should be 4 sequences in dataset - two materialised, and two propagated
829 assertEquals(4, ds.getHeight());
830 assertTrue(ds.getSequences().contains(sq1.getDatasetSequence()));
831 assertTrue(ds.getSequences().contains(sq2.getDatasetSequence()));
832 assertTrue(ds.getSequences().contains(sq3));
833 assertTrue(ds.getSequences().contains(sq4));
834 // Should have one codon frame mapping between sq1 and sq2 via dataset
836 assertEquals(ds.getCodonFrame(sq1.getDatasetSequence()),
837 ds.getCodonFrame(sq2.getDatasetSequence()));
840 @Test(groups = "Functional")
841 public void testAddCodonFrame()
843 AlignmentI align = new Alignment(new SequenceI[] {});
844 AlignedCodonFrame acf = new AlignedCodonFrame();
845 align.addCodonFrame(acf);
846 assertEquals(1, align.getCodonFrames().size());
847 assertTrue(align.getCodonFrames().contains(acf));
848 // can't add the same object twice:
849 align.addCodonFrame(acf);
850 assertEquals(1, align.getCodonFrames().size());
852 // create dataset alignment - mappings move to dataset
853 ((Alignment) align).createDatasetAlignment();
854 assertSame(align.getCodonFrames(), align.getDataset().getCodonFrames());
855 assertEquals(1, align.getCodonFrames().size());
857 AlignedCodonFrame acf2 = new AlignedCodonFrame();
858 align.addCodonFrame(acf2);
859 assertTrue(align.getDataset().getCodonFrames().contains(acf));
862 @Test(groups = "Functional")
863 public void testAddSequencePreserveDatasetIntegrity()
865 Sequence seq = new Sequence("testSeq", "ABCDEFGHIJKLMNOPQRSTUVWXYZ");
866 Alignment align = new Alignment(new SequenceI[] { seq });
867 align.createDatasetAlignment();
868 AlignmentI ds = align.getDataset();
869 SequenceI copy = new Sequence(seq);
870 copy.insertCharAt(3, 5, '-');
871 align.addSequence(copy);
872 Assert.assertEquals(align.getDataset().getHeight(), 1,
873 "Dataset shouldn't have more than one sequence.");
875 Sequence seq2 = new Sequence("newtestSeq", "ABCDEFGHIJKLMNOPQRSTUVWXYZ");
876 align.addSequence(seq2);
877 Assert.assertEquals(align.getDataset().getHeight(), 2,
878 "Dataset should now have two sequences.");
880 assertAlignmentDatasetRefs(align,
881 "addSequence broke dataset reference integrity");
883 @Test(groups = "Functional")
884 public void getVisibleStartAndEndIndexTest()
886 Sequence seq = new Sequence("testSeq", "ABCDEFGHIJKLMNOPQRSTUVWXYZ");
887 AlignmentI align = new Alignment(new SequenceI[] { seq });
888 ArrayList<int[]> hiddenCols = new ArrayList<int[]>();
890 int[] startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
891 assertEquals(0, startEnd[0]);
892 assertEquals(25, startEnd[1]);
894 hiddenCols.add(new int[] { 0, 0 });
895 startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
896 assertEquals(1, startEnd[0]);
897 assertEquals(25, startEnd[1]);
899 hiddenCols.add(new int[] { 6, 9 });
900 hiddenCols.add(new int[] { 11, 12 });
901 startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
902 assertEquals(1, startEnd[0]);
903 assertEquals(25, startEnd[1]);
905 hiddenCols.add(new int[] { 24, 25 });
906 startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
907 System.out.println(startEnd[0] + " : " + startEnd[1]);
908 assertEquals(1, startEnd[0]);
909 assertEquals(23, startEnd[1]);