+ * Tests for the method that maps the subset of a dna sequence that has CDS
+ * (or subtype) feature, with CDS strand = '-' (reverse)
+ */
+ // test turned off as currently findCdsPositions is not strand-dependent
+ // left in case it comes around again...
+ @Test(groups = "Functional", enabled = false)
+ public void testFindCdsPositions_reverseStrand()
+ {
+ SequenceI dnaSeq = new Sequence("dna", "aaaGGGcccAAATTTttt");
+ dnaSeq.createDatasetSequence();
+ SequenceI ds = dnaSeq.getDatasetSequence();
+
+ // CDS for dna 4-6
+ SequenceFeature sf = new SequenceFeature("CDS", "", 4, 6, 0f, null);
+ sf.setStrand("-");
+ ds.addSequenceFeature(sf);
+ // exon feature should be ignored here
+ sf = new SequenceFeature("exon", "", 7, 9, 0f, null);
+ ds.addSequenceFeature(sf);
+ // CDS for dna 10-12
+ sf = new SequenceFeature("CDS_predicted", "", 10, 12, 0f, null);
+ sf.setStrand("-");
+ ds.addSequenceFeature(sf);
+
+ List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
+ /*
+ * verify ranges { [12-10], [6-4] }
+ */
+ assertEquals(6, MappingUtils.getLength(ranges));
+ assertEquals(2, ranges.size());
+ assertEquals(12, ranges.get(0)[0]);
+ assertEquals(10, ranges.get(0)[1]);
+ assertEquals(6, ranges.get(1)[0]);
+ assertEquals(4, ranges.get(1)[1]);
+ }
+
+ /**
+ * Tests for the method that maps the subset of a dna sequence that has CDS
+ * (or subtype) feature - reverse strand case where the start codon is
+ * incomplete.
+ */
+ @Test(groups = "Functional", enabled = false)
+ // test turned off as currently findCdsPositions is not strand-dependent
+ // left in case it comes around again...
+ public void testFindCdsPositions_reverseStrandThreePrimeIncomplete()
+ {
+ SequenceI dnaSeq = new Sequence("dna", "aaagGGCCCaaaTTTttt");
+ dnaSeq.createDatasetSequence();
+ SequenceI ds = dnaSeq.getDatasetSequence();
+
+ // CDS for dna 5-9
+ SequenceFeature sf = new SequenceFeature("CDS", "", 5, 9, 0f, null);
+ sf.setStrand("-");
+ ds.addSequenceFeature(sf);
+ // CDS for dna 13-15
+ sf = new SequenceFeature("CDS_predicted", "", 13, 15, 0f, null);
+ sf.setStrand("-");
+ sf.setPhase("2"); // skip 2 bases to start of next codon
+ ds.addSequenceFeature(sf);
+
+ List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
+
+ /*
+ * check the mapping starts with the first complete codon
+ * expect ranges [13, 13], [9, 5]
+ */
+ assertEquals(6, MappingUtils.getLength(ranges));
+ assertEquals(2, ranges.size());
+ assertEquals(13, ranges.get(0)[0]);
+ assertEquals(13, ranges.get(0)[1]);
+ assertEquals(9, ranges.get(1)[0]);
+ assertEquals(5, ranges.get(1)[1]);
+ }
+
+ @Test(groups = "Functional")
+ public void testAlignAs_alternateTranscriptsUngapped()
+ {
+ SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
+ SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
+ AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
+ ((Alignment) dna).createDatasetAlignment();
+ SequenceI cds1 = new Sequence("cds1", "GGGTTT");
+ SequenceI cds2 = new Sequence("cds2", "CCCAAA");
+ AlignmentI cds = new Alignment(new SequenceI[] { cds1, cds2 });
+ ((Alignment) cds).createDatasetAlignment();
+
+ AlignedCodonFrame acf = new AlignedCodonFrame();
+ MapList map = new MapList(new int[] { 4, 9 }, new int[] { 1, 6 }, 1, 1);
+ acf.addMap(dna1.getDatasetSequence(), cds1.getDatasetSequence(), map);
+ map = new MapList(new int[] { 1, 3, 10, 12 }, new int[] { 1, 6 }, 1, 1);
+ acf.addMap(dna2.getDatasetSequence(), cds2.getDatasetSequence(), map);
+
+ /*
+ * verify CDS alignment is as:
+ * cccGGGTTTaaa (cdna)
+ * CCCgggtttAAA (cdna)
+ *
+ * ---GGGTTT--- (cds)
+ * CCC------AAA (cds)
+ */
+ dna.addCodonFrame(acf);
+ AlignmentUtils.alignAs(cds, dna);
+ assertEquals("---GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
+ assertEquals("CCC------AAA",
+ cds.getSequenceAt(1).getSequenceAsString());
+ }
+
+ @Test(groups = { "Functional" })
+ public void testAddMappedPositions()
+ {
+ SequenceI from = new Sequence("dna", "ggAA-ATcc-TT-g");
+ SequenceI seq1 = new Sequence("cds", "AAATTT");
+ from.createDatasetSequence();
+ seq1.createDatasetSequence();
+ Mapping mapping = new Mapping(seq1,
+ new MapList(new int[]
+ { 3, 6, 9, 10 }, new int[] { 1, 6 }, 1, 1));
+ Map<Integer, Map<SequenceI, Character>> map = new TreeMap<>();
+ AlignmentUtils.addMappedPositions(seq1, from, mapping, map);
+
+ /*
+ * verify map has seq1 residues in columns 3,4,6,7,11,12
+ */
+ assertEquals(6, map.size());
+ assertEquals('A', map.get(3).get(seq1).charValue());
+ assertEquals('A', map.get(4).get(seq1).charValue());
+ assertEquals('A', map.get(6).get(seq1).charValue());
+ assertEquals('T', map.get(7).get(seq1).charValue());
+ assertEquals('T', map.get(11).get(seq1).charValue());
+ assertEquals('T', map.get(12).get(seq1).charValue());
+
+ /*
+ *
+ */
+ }
+
+ /**
+ * Test case where the mapping 'from' range includes a stop codon which is
+ * absent in the 'to' range
+ */
+ @Test(groups = { "Functional" })
+ public void testAddMappedPositions_withStopCodon()
+ {
+ SequenceI from = new Sequence("dna", "ggAA-ATcc-TT-g");
+ SequenceI seq1 = new Sequence("cds", "AAATTT");
+ from.createDatasetSequence();
+ seq1.createDatasetSequence();
+ Mapping mapping = new Mapping(seq1,
+ new MapList(new int[]
+ { 3, 6, 9, 10 }, new int[] { 1, 6 }, 1, 1));
+ Map<Integer, Map<SequenceI, Character>> map = new TreeMap<>();
+ AlignmentUtils.addMappedPositions(seq1, from, mapping, map);
+
+ /*
+ * verify map has seq1 residues in columns 3,4,6,7,11,12
+ */
+ assertEquals(6, map.size());
+ assertEquals('A', map.get(3).get(seq1).charValue());
+ assertEquals('A', map.get(4).get(seq1).charValue());
+ assertEquals('A', map.get(6).get(seq1).charValue());
+ assertEquals('T', map.get(7).get(seq1).charValue());
+ assertEquals('T', map.get(11).get(seq1).charValue());
+ assertEquals('T', map.get(12).get(seq1).charValue());
+ }
+
+ /**
+ * Test for the case where the products for which we want CDS are specified.
+ * This is to represent the case where EMBL has CDS mappings to both Uniprot
+ * and EMBLCDSPROTEIN. makeCdsAlignment() should only return the mappings for
+ * the protein sequences specified.
+ */
+ @Test(groups = { "Functional" })
+ public void testMakeCdsAlignment_filterProducts()
+ {
+ SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
+ SequenceI dna2 = new Sequence("dna2", "GGGcccTTTaaaCCC");
+ SequenceI pep1 = new Sequence("Uniprot|pep1", "GF");
+ SequenceI pep2 = new Sequence("Uniprot|pep2", "GFP");
+ SequenceI pep3 = new Sequence("EMBL|pep3", "GF");
+ SequenceI pep4 = new Sequence("EMBL|pep4", "GFP");
+ dna1.createDatasetSequence();
+ dna2.createDatasetSequence();
+ pep1.createDatasetSequence();
+ pep2.createDatasetSequence();
+ pep3.createDatasetSequence();
+ pep4.createDatasetSequence();
+ AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
+ dna.setDataset(null);
+ AlignmentI emblPeptides = new Alignment(new SequenceI[] { pep3, pep4 });
+ emblPeptides.setDataset(null);
+
+ AlignedCodonFrame acf = new AlignedCodonFrame();
+ MapList map = new MapList(new int[] { 4, 6, 10, 12 },
+ new int[]
+ { 1, 2 }, 3, 1);
+ acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
+ acf.addMap(dna1.getDatasetSequence(), pep3.getDatasetSequence(), map);
+ dna.addCodonFrame(acf);
+
+ acf = new AlignedCodonFrame();
+ map = new MapList(new int[] { 1, 3, 7, 9, 13, 15 }, new int[] { 1, 3 },
+ 3, 1);
+ acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map);
+ acf.addMap(dna2.getDatasetSequence(), pep4.getDatasetSequence(), map);
+ dna.addCodonFrame(acf);
+
+ /*
+ * execute method under test to find CDS for EMBL peptides only
+ */
+ AlignmentI cds = AlignmentUtils
+ .makeCdsAlignment(new SequenceI[]
+ { dna1, dna2 }, dna.getDataset(),
+ emblPeptides.getSequencesArray());
+
+ assertEquals(2, cds.getSequences().size());
+ assertEquals("GGGTTT", cds.getSequenceAt(0).getSequenceAsString());
+ assertEquals("GGGTTTCCC", cds.getSequenceAt(1).getSequenceAsString());
+
+ /*
+ * verify shared, extended alignment dataset
+ */
+ assertSame(dna.getDataset(), cds.getDataset());
+ assertTrue(dna.getDataset().getSequences()
+ .contains(cds.getSequenceAt(0).getDatasetSequence()));
+ assertTrue(dna.getDataset().getSequences()
+ .contains(cds.getSequenceAt(1).getDatasetSequence()));
+
+ /*
+ * Verify mappings from CDS to peptide, cDNA to CDS, and cDNA to peptide
+ * the mappings are on the shared alignment dataset
+ */
+ List<AlignedCodonFrame> cdsMappings = cds.getDataset().getCodonFrames();
+ /*
+ * 6 mappings, 2*(DNA->CDS), 2*(DNA->Pep), 2*(CDS->Pep)
+ */
+ assertEquals(6, cdsMappings.size());
+
+ /*
+ * verify that mapping sets for dna and cds alignments are different
+ * [not current behaviour - all mappings are on the alignment dataset]
+ */
+ // select -> subselect type to test.
+ // Assert.assertNotSame(dna.getCodonFrames(), cds.getCodonFrames());
+ // assertEquals(4, dna.getCodonFrames().size());
+ // assertEquals(4, cds.getCodonFrames().size());
+
+ /*
+ * Two mappings involve pep3 (dna to pep3, cds to pep3)
+ * Mapping from pep3 to GGGTTT in first new exon sequence
+ */
+ List<AlignedCodonFrame> pep3Mappings = MappingUtils
+ .findMappingsForSequence(pep3, cdsMappings);
+ assertEquals(2, pep3Mappings.size());
+ List<AlignedCodonFrame> mappings = MappingUtils
+ .findMappingsForSequence(cds.getSequenceAt(0), pep3Mappings);
+ assertEquals(1, mappings.size());
+
+ // map G to GGG
+ SearchResultsI sr = MappingUtils.buildSearchResults(pep3, 1, mappings);
+ assertEquals(1, sr.getResults().size());
+ SearchResultMatchI m = sr.getResults().get(0);
+ assertSame(cds.getSequenceAt(0).getDatasetSequence(), m.getSequence());
+ assertEquals(1, m.getStart());
+ assertEquals(3, m.getEnd());
+ // map F to TTT
+ sr = MappingUtils.buildSearchResults(pep3, 2, mappings);
+ m = sr.getResults().get(0);
+ assertSame(cds.getSequenceAt(0).getDatasetSequence(), m.getSequence());
+ assertEquals(4, m.getStart());
+ assertEquals(6, m.getEnd());
+
+ /*
+ * Two mappings involve pep4 (dna to pep4, cds to pep4)
+ * Verify mapping from pep4 to GGGTTTCCC in second new exon sequence
+ */
+ List<AlignedCodonFrame> pep4Mappings = MappingUtils
+ .findMappingsForSequence(pep4, cdsMappings);
+ assertEquals(2, pep4Mappings.size());
+ mappings = MappingUtils.findMappingsForSequence(cds.getSequenceAt(1),
+ pep4Mappings);
+ assertEquals(1, mappings.size());
+ // map G to GGG
+ sr = MappingUtils.buildSearchResults(pep4, 1, mappings);
+ assertEquals(1, sr.getResults().size());
+ m = sr.getResults().get(0);
+ assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
+ assertEquals(1, m.getStart());
+ assertEquals(3, m.getEnd());
+ // map F to TTT
+ sr = MappingUtils.buildSearchResults(pep4, 2, mappings);
+ m = sr.getResults().get(0);
+ assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
+ assertEquals(4, m.getStart());
+ assertEquals(6, m.getEnd());
+ // map P to CCC
+ sr = MappingUtils.buildSearchResults(pep4, 3, mappings);
+ m = sr.getResults().get(0);
+ assertSame(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence());
+ assertEquals(7, m.getStart());
+ assertEquals(9, m.getEnd());
+ }
+
+ /**
+ * Test the method that just copies aligned sequences, provided all sequences
+ * to be aligned share the aligned sequence's dataset