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 * checks that the sequence data for an alignment's dataset is non-redundant.
429 * Fails if there are sequences with same id, sequence, start, and.
432 public static void assertDatasetIsNormalised(AlignmentI al)
434 assertDatasetIsNormalised(al, null);
438 * checks that the sequence data for an alignment's dataset is non-redundant.
439 * Fails if there are sequences with same id, sequence, start, and.
442 * - alignment to verify
444 * - null or message prepended to exception message.
446 public static void assertDatasetIsNormalised(AlignmentI al, String message)
448 if (al.getDataset()!=null)
450 assertDatasetIsNormalised(al.getDataset(), message);
454 * look for pairs of sequences with same ID, start, end, and sequence
456 List<SequenceI> seqSet = al.getSequences();
457 for (int p=0;p<seqSet.size(); p++)
459 SequenceI pSeq = seqSet.get(p);
460 for (int q=p+1; q<seqSet.size(); q++)
462 SequenceI qSeq = seqSet.get(q);
463 if (pSeq.getStart()!=qSeq.getStart())
467 if (pSeq.getEnd()!=qSeq.getEnd())
471 if (!pSeq.getName().equals(qSeq.getName()))
475 if (!Arrays.equals(pSeq.getSequence(), qSeq.getSequence()))
479 Assert.fail((message == null ? "" : message + " :")
480 + "Found similar sequences at position " + p + " and " + q
481 + "\n" + pSeq.toString());
486 @Test(groups = { "Functional", "Asserts" })
487 public void testAssertDatasetIsNormalised()
489 Sequence sq1 = new Sequence("s1/1-4", "asdf");
490 Sequence sq1shift = new Sequence("s1/2-5", "asdf");
491 Sequence sq1seqd = new Sequence("s1/1-4", "asdt");
492 Sequence sq2 = new Sequence("s2/1-4", "asdf");
493 Sequence sq1dup = new Sequence("s1/1-4", "asdf");
495 Alignment al = new Alignment(new SequenceI[] { sq1 });
500 assertDatasetIsNormalised(al);
501 } catch (AssertionError ae)
503 Assert.fail("Single sequence should be valid normalised dataset.");
508 assertDatasetIsNormalised(al);
509 } catch (AssertionError ae)
511 Assert.fail("Two different sequences should be valid normalised dataset.");
514 * now change sq2's name in the alignment. should still be valid
516 al.findName(sq2.getName()).setName("sq1");
519 assertDatasetIsNormalised(al);
520 } catch (AssertionError ae)
522 Assert.fail("Two different sequences in dataset, but same name in alignment, should be valid normalised dataset.");
525 al.addSequence(sq1seqd);
528 assertDatasetIsNormalised(al);
529 } catch (AssertionError ae)
531 Assert.fail("sq1 and sq1 with different sequence should be distinct.");
534 al.addSequence(sq1shift);
537 assertDatasetIsNormalised(al);
538 } catch (AssertionError ae)
540 Assert.fail("sq1 and sq1 with different start/end should be distinct.");
543 * finally, the failure case
545 al.addSequence(sq1dup);
546 boolean ssertRaised = false;
549 assertDatasetIsNormalised(al);
551 } catch (AssertionError ae)
557 Assert.fail("Expected identical sequence to raise exception.");
561 * Read in Stockholm format test data including secondary structure
564 @BeforeMethod(alwaysRun = true)
565 public void setUp() throws IOException
567 al = loadAlignment(TEST_DATA, "STH");
569 for (AlignmentAnnotation ann : al.getAlignmentAnnotation())
571 ann.setCalcId("CalcIdFor" + al.getSequenceAt(i).getName());
577 * Test method that returns annotations that match on calcId.
579 @Test(groups = { "Functional" })
580 public void testFindAnnotation_byCalcId()
582 Iterable<AlignmentAnnotation> anns = al
583 .findAnnotation("CalcIdForD.melanogaster.2");
584 Iterator<AlignmentAnnotation> iter = anns.iterator();
585 assertTrue(iter.hasNext());
586 AlignmentAnnotation ann = iter.next();
587 assertEquals("D.melanogaster.2", ann.sequenceRef.getName());
588 assertFalse(iter.hasNext());
591 @Test(groups = { "Functional" })
592 public void testDeleteAllAnnotations_includingAutocalculated()
594 AlignmentAnnotation aa = new AlignmentAnnotation("Consensus",
596 aa.autoCalculated = true;
597 al.addAnnotation(aa);
598 AlignmentAnnotation[] anns = al.getAlignmentAnnotation();
599 assertEquals("Wrong number of annotations before deleting", 4,
601 al.deleteAllAnnotations(true);
602 assertEquals("Not all deleted", 0, al.getAlignmentAnnotation().length);
605 @Test(groups = { "Functional" })
606 public void testDeleteAllAnnotations_excludingAutocalculated()
608 AlignmentAnnotation aa = new AlignmentAnnotation("Consensus",
610 aa.autoCalculated = true;
611 al.addAnnotation(aa);
612 AlignmentAnnotation[] anns = al.getAlignmentAnnotation();
613 assertEquals("Wrong number of annotations before deleting", 4,
615 al.deleteAllAnnotations(false);
616 assertEquals("Not just one annotation left", 1,
617 al.getAlignmentAnnotation().length);
621 * Tests for realigning as per a supplied alignment: Dna as Dna.
623 * Note: AlignedCodonFrame's state variables are named for protein-to-cDNA
624 * mapping, but can be exploited for a general 'sequence-to-sequence' mapping
627 * @throws IOException
629 @Test(groups = { "Functional" })
630 public void testAlignAs_dnaAsDna() throws IOException
633 AlignmentI al1 = loadAlignment(CDNA_SEQS_1, "FASTA");
635 AlignmentI al2 = loadAlignment(CDNA_SEQS_2, "FASTA");
638 * Make mappings between sequences. The 'aligned cDNA' is playing the role
639 * of what would normally be protein here.
641 makeMappings(al1, al2);
643 ((Alignment) al2).alignAs(al1, false, true);
644 assertEquals("GC-TC--GUC-GTACT", al2.getSequenceAt(0)
645 .getSequenceAsString());
646 assertEquals("-GG-GTC--AGG--CAGT", al2.getSequenceAt(1)
647 .getSequenceAsString());
651 * Aligning protein from cDNA.
653 * @throws IOException
655 @Test(groups = { "Functional" })
656 public void testAlignAs_proteinAsCdna() throws IOException
658 // see also AlignmentUtilsTests
659 AlignmentI al1 = loadAlignment(CDNA_SEQS_1, "FASTA");
660 AlignmentI al2 = loadAlignment(AA_SEQS_1, "FASTA");
661 makeMappings(al1, al2);
663 // Fudge - alignProteinAsCdna expects mappings to be on protein
664 al2.getCodonFrames().addAll(al1.getCodonFrames());
666 ((Alignment) al2).alignAs(al1, false, true);
667 assertEquals("K-Q-Y-L-", al2.getSequenceAt(0).getSequenceAsString());
668 assertEquals("-R-F-P-W", al2.getSequenceAt(1).getSequenceAsString());
672 * Test aligning cdna as per protein alignment.
674 * @throws IOException
676 @Test(groups = { "Functional" }, enabled = true)
677 // TODO review / update this test after redesign of alignAs method
678 public void testAlignAs_cdnaAsProtein() throws IOException
681 * Load alignments and add mappings for cDNA to protein
683 AlignmentI al1 = loadAlignment(CDNA_SEQS_1, "FASTA");
684 AlignmentI al2 = loadAlignment(AA_SEQS_1, "FASTA");
685 makeMappings(al1, al2);
688 * Realign DNA; currently keeping existing gaps in introns only
690 ((Alignment) al1).alignAs(al2, false, true);
691 assertEquals("ACG---GCUCCA------ACT---", al1.getSequenceAt(0)
692 .getSequenceAsString());
693 assertEquals("---CGT---TAACGA---AGT---", al1.getSequenceAt(1)
694 .getSequenceAsString());
698 * Test aligning cdna as per protein - single sequences
700 * @throws IOException
702 @Test(groups = { "Functional" }, enabled = true)
703 // TODO review / update this test after redesign of alignAs method
704 public void testAlignAs_cdnaAsProtein_singleSequence() throws IOException
707 * simple case insert one gap
709 verifyAlignAs(">dna\nCAAaaa\n", ">protein\nQ-K\n", "CAA---aaa");
712 * simple case but with sequence offsets
714 verifyAlignAs(">dna/5-10\nCAAaaa\n", ">protein/20-21\nQ-K\n",
718 * insert gaps as per protein, drop gaps within codons
720 verifyAlignAs(">dna/10-18\nCA-Aa-aa--AGA\n", ">aa/6-8\n-Q-K--R\n",
721 "---CAA---aaa------AGA");
725 * Helper method that makes mappings and then aligns the first alignment as
731 * @throws IOException
733 public void verifyAlignAs(String fromSeqs, String toSeqs, String expected)
737 * Load alignments and add mappings from nucleotide to protein (or from
738 * first to second if both the same type)
740 AlignmentI al1 = loadAlignment(fromSeqs, "FASTA");
741 AlignmentI al2 = loadAlignment(toSeqs, "FASTA");
742 makeMappings(al1, al2);
745 * Realign DNA; currently keeping existing gaps in introns only
747 ((Alignment) al1).alignAs(al2, false, true);
748 assertEquals(expected, al1.getSequenceAt(0).getSequenceAsString());
752 * Helper method to make mappings between sequences, and add the mappings to
753 * the 'mapped from' alignment
758 public void makeMappings(AlignmentI alFrom, AlignmentI alTo)
760 int ratio = (alFrom.isNucleotide() == alTo.isNucleotide() ? 1 : 3);
762 AlignedCodonFrame acf = new AlignedCodonFrame();
764 for (int i = 0; i < alFrom.getHeight(); i++)
766 SequenceI seqFrom = alFrom.getSequenceAt(i);
767 SequenceI seqTo = alTo.getSequenceAt(i);
768 MapList ml = new MapList(new int[] { seqFrom.getStart(),
770 new int[] { seqTo.getStart(), seqTo.getEnd() }, ratio, 1);
771 acf.addMap(seqFrom, seqTo, ml);
775 * not sure whether mappings 'belong' or protein or nucleotide
776 * alignment, so adding to both ;~)
778 alFrom.addCodonFrame(acf);
779 alTo.addCodonFrame(acf);
783 * Test aligning dna as per protein alignment, for the case where there are
784 * introns (i.e. some dna sites have no mapping from a peptide).
786 * @throws IOException
788 @Test(groups = { "Functional" }, enabled = false)
789 // TODO review / update this test after redesign of alignAs method
790 public void testAlignAs_dnaAsProtein_withIntrons() throws IOException
793 * Load alignments and add mappings for cDNA to protein
795 String dna1 = "A-Aa-gG-GCC-cT-TT";
796 String dna2 = "c--CCGgg-TT--T-AA-A";
797 AlignmentI al1 = loadAlignment(">Dna1/6-17\n" + dna1
798 + "\n>Dna2/20-31\n" + dna2 + "\n", "FASTA");
799 AlignmentI al2 = loadAlignment(
800 ">Pep1/7-9\n-P--YK\n>Pep2/11-13\nG-T--F\n", "FASTA");
801 AlignedCodonFrame acf = new AlignedCodonFrame();
802 // Seq1 has intron at dna positions 3,4,9 so splice is AAG GCC TTT
803 // Seq2 has intron at dna positions 1,5,6 so splice is CCG TTT AAA
804 MapList ml1 = new MapList(new int[] { 6, 7, 10, 13, 15, 17 }, new int[]
806 acf.addMap(al1.getSequenceAt(0), al2.getSequenceAt(0), ml1);
807 MapList ml2 = new MapList(new int[] { 21, 23, 26, 31 }, new int[] { 11,
809 acf.addMap(al1.getSequenceAt(1), al2.getSequenceAt(1), ml2);
810 al2.addCodonFrame(acf);
813 * Align ignoring gaps in dna introns and exons
815 ((Alignment) al1).alignAs(al2, false, false);
816 assertEquals("---AAagG------GCCcTTT", al1.getSequenceAt(0)
817 .getSequenceAsString());
818 // note 1 gap in protein corresponds to 'gg-' in DNA (3 positions)
819 assertEquals("cCCGgg-TTT------AAA", al1.getSequenceAt(1)
820 .getSequenceAsString());
823 * Reset and realign, preserving gaps in dna introns and exons
825 al1.getSequenceAt(0).setSequence(dna1);
826 al1.getSequenceAt(1).setSequence(dna2);
827 ((Alignment) al1).alignAs(al2, true, true);
828 // String dna1 = "A-Aa-gG-GCC-cT-TT";
829 // String dna2 = "c--CCGgg-TT--T-AA-A";
830 // assumption: we include 'the greater of' protein/dna gap lengths, not both
831 assertEquals("---A-Aa-gG------GCC-cT-TT", al1.getSequenceAt(0)
832 .getSequenceAsString());
833 assertEquals("c--CCGgg-TT--T------AA-A", al1.getSequenceAt(1)
834 .getSequenceAsString());
837 @Test(groups = "Functional")
838 public void testCopyConstructor() throws IOException
840 AlignmentI protein = loadAlignment(AA_SEQS_1, FormatAdapter.PASTE);
841 // create sequence and alignment datasets
842 protein.setDataset(null);
843 AlignedCodonFrame acf = new AlignedCodonFrame();
844 List<AlignedCodonFrame> acfList = Arrays.asList(new AlignedCodonFrame[]
846 protein.getDataset().setCodonFrames(acfList);
847 AlignmentI copy = new Alignment(protein);
850 * copy has different aligned sequences but the same dataset sequences
852 assertFalse(copy.getSequenceAt(0) == protein.getSequenceAt(0));
853 assertFalse(copy.getSequenceAt(1) == protein.getSequenceAt(1));
854 assertSame(copy.getSequenceAt(0).getDatasetSequence(), protein
855 .getSequenceAt(0).getDatasetSequence());
856 assertSame(copy.getSequenceAt(1).getDatasetSequence(), protein
857 .getSequenceAt(1).getDatasetSequence());
859 // TODO should the copy constructor copy the dataset?
860 // or make a new one referring to the same dataset sequences??
861 assertNull(copy.getDataset());
862 // TODO test metadata is copied when AlignmentI is a dataset
864 // assertArrayEquals(copy.getDataset().getSequencesArray(), protein
865 // .getDataset().getSequencesArray());
869 * Test behaviour of createDataset
871 * @throws IOException
873 @Test(groups = "Functional")
874 public void testCreateDatasetAlignment() throws IOException
876 AlignmentI protein = new FormatAdapter().readFile(AA_SEQS_1,
877 AppletFormatAdapter.PASTE, "FASTA");
879 * create a dataset sequence on first sequence
880 * leave the second without one
882 protein.getSequenceAt(0).createDatasetSequence();
883 assertNotNull(protein.getSequenceAt(0).getDatasetSequence());
884 assertNull(protein.getSequenceAt(1).getDatasetSequence());
887 * add a mapping to the alignment
889 AlignedCodonFrame acf = new AlignedCodonFrame();
890 protein.addCodonFrame(acf);
891 assertNull(protein.getDataset());
892 assertTrue(protein.getCodonFrames().contains(acf));
895 * create the alignment dataset
896 * note this creates sequence datasets where missing
897 * as a side-effect (in this case, on seq2
899 // TODO promote this method to AlignmentI
900 ((Alignment) protein).createDatasetAlignment();
902 AlignmentI ds = protein.getDataset();
904 // side-effect: dataset created on second sequence
905 assertNotNull(protein.getSequenceAt(1).getDatasetSequence());
906 // dataset alignment has references to dataset sequences
907 assertEquals(ds.getSequenceAt(0), protein.getSequenceAt(0)
908 .getDatasetSequence());
909 assertEquals(ds.getSequenceAt(1), protein.getSequenceAt(1)
910 .getDatasetSequence());
912 // codon frames should have been moved to the dataset
913 // getCodonFrames() should delegate to the dataset:
914 assertTrue(protein.getCodonFrames().contains(acf));
915 // prove the codon frames are indeed on the dataset:
916 assertTrue(ds.getCodonFrames().contains(acf));
920 * tests the addition of *all* sequences referred to by a sequence being added
923 @Test(groups = "Functional")
924 public void testCreateDatasetAlignmentWithMappedToSeqs()
926 // Alignment with two sequences, gapped.
927 SequenceI sq1 = new Sequence("sq1", "A--SDF");
928 SequenceI sq2 = new Sequence("sq2", "G--TRQ");
930 // cross-references to two more sequences.
931 DBRefEntry dbr = new DBRefEntry("SQ1", "", "sq3");
932 SequenceI sq3 = new Sequence("sq3", "VWANG");
933 dbr.setMap(new Mapping(sq3, new MapList(new int[] { 1, 4 }, new int[] {
937 SequenceI sq4 = new Sequence("sq4", "ERKWI");
938 DBRefEntry dbr2 = new DBRefEntry("SQ2", "", "sq4");
939 dbr2.setMap(new Mapping(sq4, new MapList(new int[] { 1, 4 }, new int[] {
942 // and a 1:1 codonframe mapping between them.
943 AlignedCodonFrame alc = new AlignedCodonFrame();
944 alc.addMap(sq1, sq2, new MapList(new int[] { 1, 4 },
945 new int[] { 1, 4 }, 1, 1));
947 AlignmentI protein = new Alignment(new SequenceI[] { sq1, sq2 });
950 * create the alignment dataset
951 * note this creates sequence datasets where missing
952 * as a side-effect (in this case, on seq2
955 // TODO promote this method to AlignmentI
956 ((Alignment) protein).createDatasetAlignment();
958 AlignmentI ds = protein.getDataset();
960 // should be 4 sequences in dataset - two materialised, and two propagated
962 assertEquals(4, ds.getHeight());
963 assertTrue(ds.getSequences().contains(sq1.getDatasetSequence()));
964 assertTrue(ds.getSequences().contains(sq2.getDatasetSequence()));
965 assertTrue(ds.getSequences().contains(sq3));
966 assertTrue(ds.getSequences().contains(sq4));
967 // Should have one codon frame mapping between sq1 and sq2 via dataset
969 assertEquals(ds.getCodonFrame(sq1.getDatasetSequence()),
970 ds.getCodonFrame(sq2.getDatasetSequence()));
973 @Test(groups = "Functional")
974 public void testAddCodonFrame()
976 AlignmentI align = new Alignment(new SequenceI[] {});
977 AlignedCodonFrame acf = new AlignedCodonFrame();
978 align.addCodonFrame(acf);
979 assertEquals(1, align.getCodonFrames().size());
980 assertTrue(align.getCodonFrames().contains(acf));
981 // can't add the same object twice:
982 align.addCodonFrame(acf);
983 assertEquals(1, align.getCodonFrames().size());
985 // create dataset alignment - mappings move to dataset
986 ((Alignment) align).createDatasetAlignment();
987 assertSame(align.getCodonFrames(), align.getDataset().getCodonFrames());
988 assertEquals(1, align.getCodonFrames().size());
990 AlignedCodonFrame acf2 = new AlignedCodonFrame();
991 align.addCodonFrame(acf2);
992 assertTrue(align.getDataset().getCodonFrames().contains(acf));
995 @Test(groups = "Functional")
996 public void testAddSequencePreserveDatasetIntegrity()
998 Sequence seq = new Sequence("testSeq", "ABCDEFGHIJKLMNOPQRSTUVWXYZ");
999 Alignment align = new Alignment(new SequenceI[] { seq });
1000 align.createDatasetAlignment();
1001 AlignmentI ds = align.getDataset();
1002 SequenceI copy = new Sequence(seq);
1003 copy.insertCharAt(3, 5, '-');
1004 align.addSequence(copy);
1005 Assert.assertEquals(align.getDataset().getHeight(), 1,
1006 "Dataset shouldn't have more than one sequence.");
1008 Sequence seq2 = new Sequence("newtestSeq", "ABCDEFGHIJKLMNOPQRSTUVWXYZ");
1009 align.addSequence(seq2);
1010 Assert.assertEquals(align.getDataset().getHeight(), 2,
1011 "Dataset should now have two sequences.");
1013 assertAlignmentDatasetRefs(align,
1014 "addSequence broke dataset reference integrity");
1016 @Test(groups = "Functional")
1017 public void getVisibleStartAndEndIndexTest()
1019 Sequence seq = new Sequence("testSeq", "ABCDEFGHIJKLMNOPQRSTUVWXYZ");
1020 AlignmentI align = new Alignment(new SequenceI[] { seq });
1021 ArrayList<int[]> hiddenCols = new ArrayList<int[]>();
1023 int[] startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
1024 assertEquals(0, startEnd[0]);
1025 assertEquals(25, startEnd[1]);
1027 hiddenCols.add(new int[] { 0, 0 });
1028 startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
1029 assertEquals(1, startEnd[0]);
1030 assertEquals(25, startEnd[1]);
1032 hiddenCols.add(new int[] { 6, 9 });
1033 hiddenCols.add(new int[] { 11, 12 });
1034 startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
1035 assertEquals(1, startEnd[0]);
1036 assertEquals(25, startEnd[1]);
1038 hiddenCols.add(new int[] { 24, 25 });
1039 startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
1040 System.out.println(startEnd[0] + " : " + startEnd[1]);
1041 assertEquals(1, startEnd[0]);
1042 assertEquals(23, startEnd[1]);