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
140 * @return true if alignment references were in order, otherwise false.
142 public static boolean verifyAlignmentDatasetRefs(AlignmentI alignment,
143 boolean raiseAssert, String message)
149 if (alignment == null)
153 Assert.fail(message + "Alignment for verification was null.");
157 if (alignment.getDataset() != null)
159 AlignmentI dataset = alignment.getDataset();
160 // check all alignment sequences have their dataset within the dataset
161 for (SequenceI seq : alignment.getSequences())
163 SequenceI seqds = seq.getDatasetSequence();
164 if (seqds.getDatasetSequence() != null)
169 + " Alignment contained a sequence who's dataset sequence has a second dataset reference.");
173 if (dataset.findIndex(seqds) == -1)
178 + " Alignment contained a sequence who's dataset sequence was not in the dataset.");
183 return verifyAlignmentDatasetRefs(alignment.getDataset(),
184 raiseAssert, message);
189 // verify all dataset sequences
190 for (SequenceI seqds : alignment.getSequences())
193 if (seqds.getDatasetSequence() != null)
198 + " Dataset contained a sequence with non-null dataset reference (ie not a dataset sequence!)");
202 int foundp = alignment.findIndex(seqds);
208 + " Dataset sequence array contains a reference at "
209 + dsp + " to a sequence first seen at " + foundp + " ("
210 + seqds.toString() + ")");
214 if (seqds.getDBRefs() != null)
216 for (DBRefEntry dbr : seqds.getDBRefs())
218 if (dbr.getMap() != null)
220 SequenceI seqdbrmapto = dbr.getMap().getTo();
221 if (seqdbrmapto != null)
223 if (seqdbrmapto.getDatasetSequence() != null)
228 + " DBRefEntry for sequence in alignment had map to sequence which was not a dataset sequence");
233 if (alignment.findIndex(dbr.getMap().getTo()) == -1)
238 + " DBRefEntry for sequence in alignment had map to sequence not in dataset");
247 // finally, verify codonmappings involve only dataset sequences.
248 if (alignment.getCodonFrames() != null)
250 for (AlignedCodonFrame alc : alignment.getCodonFrames())
252 for (SequenceToSequenceMapping ssm : alc.getMappings())
254 if (ssm.getFromSeq().getDatasetSequence() != null)
259 + " CodonFrame-SSM-FromSeq is not a dataset sequence");
263 if (alignment.findIndex(ssm.getFromSeq()) == -1)
269 + " CodonFrame-SSM-FromSeq is not contained in dataset");
273 if (ssm.getMapping().getTo().getDatasetSequence() != null)
278 + " CodonFrame-SSM-Mapping-ToSeq is not a dataset sequence");
282 if (alignment.findIndex(ssm.getMapping().getTo()) == -1)
288 + " CodonFrame-SSM-Mapping-ToSeq is not contained in dataset");
296 return true; // all relationships verified!
300 * call verifyAlignmentDatasetRefs with and without assertion raising enabled,
301 * to check expected pass/fail actually occurs in both conditions
307 private void assertVerifyAlignment(AlignmentI al, boolean expected,
315 Assert.assertTrue(verifyAlignmentDatasetRefs(al, true, null),
316 "Valid test alignment failed when raiseAsserts enabled:"
318 } catch (AssertionError ae)
320 ae.printStackTrace();
322 "Valid test alignment raised assertion errors when raiseAsserts enabled: "
325 // also check validation passes with asserts disabled
326 Assert.assertTrue(verifyAlignmentDatasetRefs(al, false, null),
327 "Valid test alignment tested false when raiseAsserts disabled:"
332 boolean assertRaised = false;
335 verifyAlignmentDatasetRefs(al, true, null);
336 } catch (AssertionError ae)
338 // expected behaviour
343 Assert.fail("Invalid test alignment passed when raiseAsserts enabled:"
346 // also check validation passes with asserts disabled
347 Assert.assertFalse(verifyAlignmentDatasetRefs(al, false, null),
348 "Invalid test alignment tested true when raiseAsserts disabled:"
353 @Test(groups = { "Functional" })
354 public void testVerifyAlignmentDatasetRefs()
356 SequenceI sq1 = new Sequence("sq1", "ASFDD"), sq2 = new Sequence("sq2",
359 // construct simple valid alignment dataset
360 Alignment al = new Alignment(new SequenceI[] { sq1, sq2 });
361 // expect this to pass
362 assertVerifyAlignment(al, true, "Simple valid alignment didn't verify");
364 // check test for sequence->datasetSequence validity
365 sq1.setDatasetSequence(sq2);
366 assertVerifyAlignment(al, false,
367 "didn't detect dataset sequence with a dataset sequence reference.");
369 sq1.setDatasetSequence(null);
370 assertVerifyAlignment(
373 "didn't reinstate validity after nulling dataset sequence dataset reference");
375 // now create dataset and check again
376 al.createDatasetAlignment();
377 assertNotNull(al.getDataset());
379 assertVerifyAlignment(al, true,
380 "verify failed after createDatasetAlignment");
382 // create a dbref on sq1 with a sequence ref to sq2
383 DBRefEntry dbrs1tos2 = new DBRefEntry("UNIPROT", "1", "Q111111");
384 dbrs1tos2.setMap(new Mapping(sq2.getDatasetSequence(),
385 new int[] { 1, 5 }, new int[] { 2, 6 }, 1, 1));
386 sq1.getDatasetSequence().addDBRef(dbrs1tos2);
387 assertVerifyAlignment(al, true,
388 "verify failed after addition of valid DBRefEntry/map");
389 // now create a dbref on a new sequence which maps to another sequence
390 // outside of the dataset
391 SequenceI sqout = new Sequence("sqout", "ututututucagcagcag"), sqnew = new Sequence(
393 DBRefEntry sqnewsqout = new DBRefEntry("ENAFOO", "1", "R000001");
394 sqnewsqout.setMap(new Mapping(sqout, new int[] { 1, 6 }, new int[] { 1,
396 al.getDataset().addSequence(sqnew);
398 assertVerifyAlignment(al, true,
399 "verify failed after addition of new sequence to dataset");
400 // now start checking exception conditions
401 sqnew.addDBRef(sqnewsqout);
402 assertVerifyAlignment(
405 "verify passed when a dbref with map to sequence outside of dataset was added");
406 // make the verify pass by adding the outsider back in
407 al.getDataset().addSequence(sqout);
408 assertVerifyAlignment(al, true,
409 "verify should have passed after adding dbref->to sequence in to dataset");
410 // and now the same for a codon mapping...
411 SequenceI sqanotherout = new Sequence("sqanotherout",
412 "aggtutaggcagcagcag");
414 AlignedCodonFrame alc = new AlignedCodonFrame();
415 alc.addMap(sqanotherout, sqnew, new MapList(new int[] { 1, 6 },
416 new int[] { 1, 18 }, 3, 1));
418 al.addCodonFrame(alc);
419 Assert.assertEquals(al.getDataset().getCodonFrames().size(), 1);
421 assertVerifyAlignment(
424 "verify passed when alCodonFrame mapping to sequence outside of dataset was added");
425 // make the verify pass by adding the outsider back in
426 al.getDataset().addSequence(sqanotherout);
427 assertVerifyAlignment(
430 "verify should have passed once all sequences involved in alCodonFrame were added to dataset");
431 al.getDataset().addSequence(sqanotherout);
432 assertVerifyAlignment(al, false,
433 "verify should have failed when a sequence was added twice to the dataset");
434 al.getDataset().deleteSequence(sqanotherout);
435 assertVerifyAlignment(al, true,
436 "verify should have passed after duplicate entry for sequence was removed");
440 * checks that the sequence data for an alignment's dataset is non-redundant.
441 * Fails if there are sequences with same id, sequence, start, and.
444 public static void assertDatasetIsNormalised(AlignmentI al)
446 assertDatasetIsNormalised(al, null);
450 * checks that the sequence data for an alignment's dataset is non-redundant.
451 * Fails if there are sequences with same id, sequence, start, and.
454 * - alignment to verify
456 * - null or message prepended to exception message.
458 public static void assertDatasetIsNormalised(AlignmentI al, String message)
460 if (al.getDataset() != null)
462 assertDatasetIsNormalised(al.getDataset(), message);
466 * look for pairs of sequences with same ID, start, end, and sequence
468 List<SequenceI> seqSet = al.getSequences();
469 for (int p = 0; p < seqSet.size(); p++)
471 SequenceI pSeq = seqSet.get(p);
472 for (int q = p + 1; q < seqSet.size(); q++)
474 SequenceI qSeq = seqSet.get(q);
475 if (pSeq.getStart() != qSeq.getStart())
479 if (pSeq.getEnd() != qSeq.getEnd())
483 if (!pSeq.getName().equals(qSeq.getName()))
487 if (!Arrays.equals(pSeq.getSequence(), qSeq.getSequence()))
491 Assert.fail((message == null ? "" : message + " :")
492 + "Found similar sequences at position " + p + " and " + q
493 + "\n" + pSeq.toString());
498 @Test(groups = { "Functional", "Asserts" })
499 public void testAssertDatasetIsNormalised()
501 Sequence sq1 = new Sequence("s1/1-4", "asdf");
502 Sequence sq1shift = new Sequence("s1/2-5", "asdf");
503 Sequence sq1seqd = new Sequence("s1/1-4", "asdt");
504 Sequence sq2 = new Sequence("s2/1-4", "asdf");
505 Sequence sq1dup = new Sequence("s1/1-4", "asdf");
507 Alignment al = new Alignment(new SequenceI[] { sq1 });
512 assertDatasetIsNormalised(al);
513 } catch (AssertionError ae)
515 Assert.fail("Single sequence should be valid normalised dataset.");
520 assertDatasetIsNormalised(al);
521 } catch (AssertionError ae)
523 Assert.fail("Two different sequences should be valid normalised dataset.");
526 * now change sq2's name in the alignment. should still be valid
528 al.findName(sq2.getName()).setName("sq1");
531 assertDatasetIsNormalised(al);
532 } catch (AssertionError ae)
534 Assert.fail("Two different sequences in dataset, but same name in alignment, should be valid normalised dataset.");
537 al.addSequence(sq1seqd);
540 assertDatasetIsNormalised(al);
541 } catch (AssertionError ae)
543 Assert.fail("sq1 and sq1 with different sequence should be distinct.");
546 al.addSequence(sq1shift);
549 assertDatasetIsNormalised(al);
550 } catch (AssertionError ae)
552 Assert.fail("sq1 and sq1 with different start/end should be distinct.");
555 * finally, the failure case
557 al.addSequence(sq1dup);
558 boolean ssertRaised = false;
561 assertDatasetIsNormalised(al);
563 } catch (AssertionError ae)
569 Assert.fail("Expected identical sequence to raise exception.");
574 * Read in Stockholm format test data including secondary structure
577 @BeforeMethod(alwaysRun = true)
578 public void setUp() throws IOException
580 al = loadAlignment(TEST_DATA, "STH");
582 for (AlignmentAnnotation ann : al.getAlignmentAnnotation())
584 ann.setCalcId("CalcIdFor" + al.getSequenceAt(i).getName());
590 * Test method that returns annotations that match on calcId.
592 @Test(groups = { "Functional" })
593 public void testFindAnnotation_byCalcId()
595 Iterable<AlignmentAnnotation> anns = al
596 .findAnnotation("CalcIdForD.melanogaster.2");
597 Iterator<AlignmentAnnotation> iter = anns.iterator();
598 assertTrue(iter.hasNext());
599 AlignmentAnnotation ann = iter.next();
600 assertEquals("D.melanogaster.2", ann.sequenceRef.getName());
601 assertFalse(iter.hasNext());
604 @Test(groups = { "Functional" })
605 public void testDeleteAllAnnotations_includingAutocalculated()
607 AlignmentAnnotation aa = new AlignmentAnnotation("Consensus",
609 aa.autoCalculated = true;
610 al.addAnnotation(aa);
611 AlignmentAnnotation[] anns = al.getAlignmentAnnotation();
612 assertEquals("Wrong number of annotations before deleting", 4,
614 al.deleteAllAnnotations(true);
615 assertEquals("Not all deleted", 0, al.getAlignmentAnnotation().length);
618 @Test(groups = { "Functional" })
619 public void testDeleteAllAnnotations_excludingAutocalculated()
621 AlignmentAnnotation aa = new AlignmentAnnotation("Consensus",
623 aa.autoCalculated = true;
624 al.addAnnotation(aa);
625 AlignmentAnnotation[] anns = al.getAlignmentAnnotation();
626 assertEquals("Wrong number of annotations before deleting", 4,
628 al.deleteAllAnnotations(false);
629 assertEquals("Not just one annotation left", 1,
630 al.getAlignmentAnnotation().length);
634 * Tests for realigning as per a supplied alignment: Dna as Dna.
636 * Note: AlignedCodonFrame's state variables are named for protein-to-cDNA
637 * mapping, but can be exploited for a general 'sequence-to-sequence' mapping
640 * @throws IOException
642 @Test(groups = { "Functional" })
643 public void testAlignAs_dnaAsDna() throws IOException
646 AlignmentI al1 = loadAlignment(CDNA_SEQS_1, "FASTA");
648 AlignmentI al2 = loadAlignment(CDNA_SEQS_2, "FASTA");
651 * Make mappings between sequences. The 'aligned cDNA' is playing the role
652 * of what would normally be protein here.
654 makeMappings(al1, al2);
656 ((Alignment) al2).alignAs(al1, false, true);
657 assertEquals("GC-TC--GUC-GTACT", al2.getSequenceAt(0)
658 .getSequenceAsString());
659 assertEquals("-GG-GTC--AGG--CAGT", al2.getSequenceAt(1)
660 .getSequenceAsString());
664 * Aligning protein from cDNA.
666 * @throws IOException
668 @Test(groups = { "Functional" })
669 public void testAlignAs_proteinAsCdna() throws IOException
671 // see also AlignmentUtilsTests
672 AlignmentI al1 = loadAlignment(CDNA_SEQS_1, "FASTA");
673 AlignmentI al2 = loadAlignment(AA_SEQS_1, "FASTA");
674 makeMappings(al1, al2);
676 // Fudge - alignProteinAsCdna expects mappings to be on protein
677 al2.getCodonFrames().addAll(al1.getCodonFrames());
679 ((Alignment) al2).alignAs(al1, false, true);
680 assertEquals("K-Q-Y-L-", al2.getSequenceAt(0).getSequenceAsString());
681 assertEquals("-R-F-P-W", al2.getSequenceAt(1).getSequenceAsString());
685 * Test aligning cdna as per protein alignment.
687 * @throws IOException
689 @Test(groups = { "Functional" }, enabled = true)
690 // TODO review / update this test after redesign of alignAs method
691 public void testAlignAs_cdnaAsProtein() throws IOException
694 * Load alignments and add mappings for cDNA to protein
696 AlignmentI al1 = loadAlignment(CDNA_SEQS_1, "FASTA");
697 AlignmentI al2 = loadAlignment(AA_SEQS_1, "FASTA");
698 makeMappings(al1, al2);
701 * Realign DNA; currently keeping existing gaps in introns only
703 ((Alignment) al1).alignAs(al2, false, true);
704 assertEquals("ACG---GCUCCA------ACT---", al1.getSequenceAt(0)
705 .getSequenceAsString());
706 assertEquals("---CGT---TAACGA---AGT---", al1.getSequenceAt(1)
707 .getSequenceAsString());
711 * Test aligning cdna as per protein - single sequences
713 * @throws IOException
715 @Test(groups = { "Functional" }, enabled = true)
716 // TODO review / update this test after redesign of alignAs method
717 public void testAlignAs_cdnaAsProtein_singleSequence() throws IOException
720 * simple case insert one gap
722 verifyAlignAs(">dna\nCAAaaa\n", ">protein\nQ-K\n", "CAA---aaa");
725 * simple case but with sequence offsets
727 verifyAlignAs(">dna/5-10\nCAAaaa\n", ">protein/20-21\nQ-K\n",
731 * insert gaps as per protein, drop gaps within codons
733 verifyAlignAs(">dna/10-18\nCA-Aa-aa--AGA\n", ">aa/6-8\n-Q-K--R\n",
734 "---CAA---aaa------AGA");
738 * Helper method that makes mappings and then aligns the first alignment as
744 * @throws IOException
746 public void verifyAlignAs(String fromSeqs, String toSeqs, String expected)
750 * Load alignments and add mappings from nucleotide to protein (or from
751 * first to second if both the same type)
753 AlignmentI al1 = loadAlignment(fromSeqs, "FASTA");
754 AlignmentI al2 = loadAlignment(toSeqs, "FASTA");
755 makeMappings(al1, al2);
758 * Realign DNA; currently keeping existing gaps in introns only
760 ((Alignment) al1).alignAs(al2, false, true);
761 assertEquals(expected, al1.getSequenceAt(0).getSequenceAsString());
765 * Helper method to make mappings between sequences, and add the mappings to
766 * the 'mapped from' alignment
771 public void makeMappings(AlignmentI alFrom, AlignmentI alTo)
773 int ratio = (alFrom.isNucleotide() == alTo.isNucleotide() ? 1 : 3);
775 AlignedCodonFrame acf = new AlignedCodonFrame();
777 for (int i = 0; i < alFrom.getHeight(); i++)
779 SequenceI seqFrom = alFrom.getSequenceAt(i);
780 SequenceI seqTo = alTo.getSequenceAt(i);
781 MapList ml = new MapList(new int[] { seqFrom.getStart(),
783 new int[] { seqTo.getStart(), seqTo.getEnd() }, ratio, 1);
784 acf.addMap(seqFrom, seqTo, ml);
788 * not sure whether mappings 'belong' or protein or nucleotide
789 * alignment, so adding to both ;~)
791 alFrom.addCodonFrame(acf);
792 alTo.addCodonFrame(acf);
796 * Test aligning dna as per protein alignment, for the case where there are
797 * introns (i.e. some dna sites have no mapping from a peptide).
799 * @throws IOException
801 @Test(groups = { "Functional" }, enabled = false)
802 // TODO review / update this test after redesign of alignAs method
803 public void testAlignAs_dnaAsProtein_withIntrons() throws IOException
806 * Load alignments and add mappings for cDNA to protein
808 String dna1 = "A-Aa-gG-GCC-cT-TT";
809 String dna2 = "c--CCGgg-TT--T-AA-A";
810 AlignmentI al1 = loadAlignment(">Dna1/6-17\n" + dna1
811 + "\n>Dna2/20-31\n" + dna2 + "\n", "FASTA");
812 AlignmentI al2 = loadAlignment(
813 ">Pep1/7-9\n-P--YK\n>Pep2/11-13\nG-T--F\n", "FASTA");
814 AlignedCodonFrame acf = new AlignedCodonFrame();
815 // Seq1 has intron at dna positions 3,4,9 so splice is AAG GCC TTT
816 // Seq2 has intron at dna positions 1,5,6 so splice is CCG TTT AAA
817 MapList ml1 = new MapList(new int[] { 6, 7, 10, 13, 15, 17 }, new int[]
819 acf.addMap(al1.getSequenceAt(0), al2.getSequenceAt(0), ml1);
820 MapList ml2 = new MapList(new int[] { 21, 23, 26, 31 }, new int[] { 11,
822 acf.addMap(al1.getSequenceAt(1), al2.getSequenceAt(1), ml2);
823 al2.addCodonFrame(acf);
826 * Align ignoring gaps in dna introns and exons
828 ((Alignment) al1).alignAs(al2, false, false);
829 assertEquals("---AAagG------GCCcTTT", al1.getSequenceAt(0)
830 .getSequenceAsString());
831 // note 1 gap in protein corresponds to 'gg-' in DNA (3 positions)
832 assertEquals("cCCGgg-TTT------AAA", al1.getSequenceAt(1)
833 .getSequenceAsString());
836 * Reset and realign, preserving gaps in dna introns and exons
838 al1.getSequenceAt(0).setSequence(dna1);
839 al1.getSequenceAt(1).setSequence(dna2);
840 ((Alignment) al1).alignAs(al2, true, true);
841 // String dna1 = "A-Aa-gG-GCC-cT-TT";
842 // String dna2 = "c--CCGgg-TT--T-AA-A";
843 // assumption: we include 'the greater of' protein/dna gap lengths, not both
844 assertEquals("---A-Aa-gG------GCC-cT-TT", al1.getSequenceAt(0)
845 .getSequenceAsString());
846 assertEquals("c--CCGgg-TT--T------AA-A", al1.getSequenceAt(1)
847 .getSequenceAsString());
850 @Test(groups = "Functional")
851 public void testCopyConstructor() throws IOException
853 AlignmentI protein = loadAlignment(AA_SEQS_1, FormatAdapter.PASTE);
854 // create sequence and alignment datasets
855 protein.setDataset(null);
856 AlignedCodonFrame acf = new AlignedCodonFrame();
857 List<AlignedCodonFrame> acfList = Arrays.asList(new AlignedCodonFrame[]
859 protein.getDataset().setCodonFrames(acfList);
860 AlignmentI copy = new Alignment(protein);
863 * copy has different aligned sequences but the same dataset sequences
865 assertFalse(copy.getSequenceAt(0) == protein.getSequenceAt(0));
866 assertFalse(copy.getSequenceAt(1) == protein.getSequenceAt(1));
867 assertSame(copy.getSequenceAt(0).getDatasetSequence(), protein
868 .getSequenceAt(0).getDatasetSequence());
869 assertSame(copy.getSequenceAt(1).getDatasetSequence(), protein
870 .getSequenceAt(1).getDatasetSequence());
872 // TODO should the copy constructor copy the dataset?
873 // or make a new one referring to the same dataset sequences??
874 assertNull(copy.getDataset());
875 // TODO test metadata is copied when AlignmentI is a dataset
877 // assertArrayEquals(copy.getDataset().getSequencesArray(), protein
878 // .getDataset().getSequencesArray());
882 * Test behaviour of createDataset
884 * @throws IOException
886 @Test(groups = "Functional")
887 public void testCreateDatasetAlignment() throws IOException
889 AlignmentI protein = new FormatAdapter().readFile(AA_SEQS_1,
890 AppletFormatAdapter.PASTE, "FASTA");
892 * create a dataset sequence on first sequence
893 * leave the second without one
895 protein.getSequenceAt(0).createDatasetSequence();
896 assertNotNull(protein.getSequenceAt(0).getDatasetSequence());
897 assertNull(protein.getSequenceAt(1).getDatasetSequence());
900 * add a mapping to the alignment
902 AlignedCodonFrame acf = new AlignedCodonFrame();
903 protein.addCodonFrame(acf);
904 assertNull(protein.getDataset());
905 assertTrue(protein.getCodonFrames().contains(acf));
908 * create the alignment dataset
909 * note this creates sequence datasets where missing
910 * as a side-effect (in this case, on seq2
912 // TODO promote this method to AlignmentI
913 ((Alignment) protein).createDatasetAlignment();
915 AlignmentI ds = protein.getDataset();
917 // side-effect: dataset created on second sequence
918 assertNotNull(protein.getSequenceAt(1).getDatasetSequence());
919 // dataset alignment has references to dataset sequences
920 assertEquals(ds.getSequenceAt(0), protein.getSequenceAt(0)
921 .getDatasetSequence());
922 assertEquals(ds.getSequenceAt(1), protein.getSequenceAt(1)
923 .getDatasetSequence());
925 // codon frames should have been moved to the dataset
926 // getCodonFrames() should delegate to the dataset:
927 assertTrue(protein.getCodonFrames().contains(acf));
928 // prove the codon frames are indeed on the dataset:
929 assertTrue(ds.getCodonFrames().contains(acf));
933 * tests the addition of *all* sequences referred to by a sequence being added
936 @Test(groups = "Functional")
937 public void testCreateDatasetAlignmentWithMappedToSeqs()
939 // Alignment with two sequences, gapped.
940 SequenceI sq1 = new Sequence("sq1", "A--SDF");
941 SequenceI sq2 = new Sequence("sq2", "G--TRQ");
943 // cross-references to two more sequences.
944 DBRefEntry dbr = new DBRefEntry("SQ1", "", "sq3");
945 SequenceI sq3 = new Sequence("sq3", "VWANG");
946 dbr.setMap(new Mapping(sq3, new MapList(new int[] { 1, 4 }, new int[] {
950 SequenceI sq4 = new Sequence("sq4", "ERKWI");
951 DBRefEntry dbr2 = new DBRefEntry("SQ2", "", "sq4");
952 dbr2.setMap(new Mapping(sq4, new MapList(new int[] { 1, 4 }, new int[] {
955 // and a 1:1 codonframe mapping between them.
956 AlignedCodonFrame alc = new AlignedCodonFrame();
957 alc.addMap(sq1, sq2, new MapList(new int[] { 1, 4 },
958 new int[] { 1, 4 }, 1, 1));
960 AlignmentI protein = new Alignment(new SequenceI[] { sq1, sq2 });
963 * create the alignment dataset
964 * note this creates sequence datasets where missing
965 * as a side-effect (in this case, on seq2
968 // TODO promote this method to AlignmentI
969 ((Alignment) protein).createDatasetAlignment();
971 AlignmentI ds = protein.getDataset();
973 // should be 4 sequences in dataset - two materialised, and two propagated
975 assertEquals(4, ds.getHeight());
976 assertTrue(ds.getSequences().contains(sq1.getDatasetSequence()));
977 assertTrue(ds.getSequences().contains(sq2.getDatasetSequence()));
978 assertTrue(ds.getSequences().contains(sq3));
979 assertTrue(ds.getSequences().contains(sq4));
980 // Should have one codon frame mapping between sq1 and sq2 via dataset
982 assertEquals(ds.getCodonFrame(sq1.getDatasetSequence()),
983 ds.getCodonFrame(sq2.getDatasetSequence()));
986 @Test(groups = "Functional")
987 public void testAddCodonFrame()
989 AlignmentI align = new Alignment(new SequenceI[] {});
990 AlignedCodonFrame acf = new AlignedCodonFrame();
991 align.addCodonFrame(acf);
992 assertEquals(1, align.getCodonFrames().size());
993 assertTrue(align.getCodonFrames().contains(acf));
994 // can't add the same object twice:
995 align.addCodonFrame(acf);
996 assertEquals(1, align.getCodonFrames().size());
998 // create dataset alignment - mappings move to dataset
999 ((Alignment) align).createDatasetAlignment();
1000 assertSame(align.getCodonFrames(), align.getDataset().getCodonFrames());
1001 assertEquals(1, align.getCodonFrames().size());
1003 AlignedCodonFrame acf2 = new AlignedCodonFrame();
1004 align.addCodonFrame(acf2);
1005 assertTrue(align.getDataset().getCodonFrames().contains(acf));
1008 @Test(groups = "Functional")
1009 public void testAddSequencePreserveDatasetIntegrity()
1011 Sequence seq = new Sequence("testSeq", "ABCDEFGHIJKLMNOPQRSTUVWXYZ");
1012 Alignment align = new Alignment(new SequenceI[] { seq });
1013 align.createDatasetAlignment();
1014 AlignmentI ds = align.getDataset();
1015 SequenceI copy = new Sequence(seq);
1016 copy.insertCharAt(3, 5, '-');
1017 align.addSequence(copy);
1018 Assert.assertEquals(align.getDataset().getHeight(), 1,
1019 "Dataset shouldn't have more than one sequence.");
1021 Sequence seq2 = new Sequence("newtestSeq", "ABCDEFGHIJKLMNOPQRSTUVWXYZ");
1022 align.addSequence(seq2);
1023 Assert.assertEquals(align.getDataset().getHeight(), 2,
1024 "Dataset should now have two sequences.");
1026 assertAlignmentDatasetRefs(align,
1027 "addSequence broke dataset reference integrity");
1030 @Test(groups = "Functional")
1031 public void getVisibleStartAndEndIndexTest()
1033 Sequence seq = new Sequence("testSeq", "ABCDEFGHIJKLMNOPQRSTUVWXYZ");
1034 AlignmentI align = new Alignment(new SequenceI[] { seq });
1035 ArrayList<int[]> hiddenCols = new ArrayList<int[]>();
1037 int[] startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
1038 assertEquals(0, startEnd[0]);
1039 assertEquals(25, startEnd[1]);
1041 hiddenCols.add(new int[] { 0, 0 });
1042 startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
1043 assertEquals(1, startEnd[0]);
1044 assertEquals(25, startEnd[1]);
1046 hiddenCols.add(new int[] { 6, 9 });
1047 hiddenCols.add(new int[] { 11, 12 });
1048 startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
1049 assertEquals(1, startEnd[0]);
1050 assertEquals(25, startEnd[1]);
1052 hiddenCols.add(new int[] { 24, 25 });
1053 startEnd = align.getVisibleStartAndEndIndex(hiddenCols);
1054 System.out.println(startEnd[0] + " : " + startEnd[1]);
1055 assertEquals(1, startEnd[0]);
1056 assertEquals(23, startEnd[1]);
1060 * Tests that dbrefs with mappings to sequence get updated if the sequence
1061 * acquires a dataset sequence
1063 @Test(groups = "Functional")
1064 public void testCreateDataset_updateDbrefMappings()
1066 SequenceI pep = new Sequence("pep", "ASD");
1067 SequenceI dna = new Sequence("dna", "aaaGCCTCGGATggg");
1068 SequenceI cds = new Sequence("cds", "GCCTCGGAT");
1070 // add dbref from dna to peptide
1071 DBRefEntry dbr = new DBRefEntry("UNIPROT", "", "pep");
1072 dbr.setMap(new Mapping(pep, new MapList(new int[] { 4, 15 }, new int[] {
1076 // add dbref from dna to peptide
1077 DBRefEntry dbr2 = new DBRefEntry("UNIPROT", "", "pep");
1078 dbr2.setMap(new Mapping(pep, new MapList(new int[] { 1, 12 }, new int[]
1082 // add dbref from peptide to dna
1083 DBRefEntry dbr3 = new DBRefEntry("EMBL", "", "dna");
1084 dbr3.setMap(new Mapping(dna, new MapList(new int[] { 1, 4 }, new int[] {
1088 // add dbref from peptide to cds
1089 DBRefEntry dbr4 = new DBRefEntry("EMBLCDS", "", "cds");
1090 dbr4.setMap(new Mapping(cds, new MapList(new int[] { 1, 4 }, new int[] {
1094 AlignmentI protein = new Alignment(new SequenceI[] { pep });
1097 * create the alignment dataset
1099 ((Alignment) protein).createDatasetAlignment();
1101 AlignmentI ds = protein.getDataset();
1103 // should be 3 sequences in dataset
1104 assertEquals(3, ds.getHeight());
1105 assertTrue(ds.getSequences().contains(pep.getDatasetSequence()));
1106 assertTrue(ds.getSequences().contains(dna));
1107 assertTrue(ds.getSequences().contains(cds));
1110 * verify peptide.cdsdbref.peptidedbref is now mapped to peptide dataset
1112 DBRefEntry[] dbRefs = pep.getDBRefs();
1113 assertEquals(2, dbRefs.length);
1114 assertSame(dna, dbRefs[0].map.to);
1115 assertSame(cds, dbRefs[1].map.to);
1116 assertEquals(1, dna.getDBRefs().length);
1117 assertSame(pep.getDatasetSequence(), dna.getDBRefs()[0].map.to);
1118 assertEquals(1, cds.getDBRefs().length);
1119 assertSame(pep.getDatasetSequence(), cds.getDBRefs()[0].map.to);