+
+ /*
+ * Verify there are mappings from each cds sequence to its protein product
+ * and also to its dna source
+ */
+ Iterator<AlignedCodonFrame> newMappingsIterator = mappings.iterator();
+
+ // mappings for dna1 - exon1 - pep1
+ AlignedCodonFrame cdsMapping = newMappingsIterator.next();
+ List<Mapping> dnaMappings = cdsMapping.getMappingsForSequence(dna1);
+ assertEquals(1, dnaMappings.size());
+ assertSame(cds.get(0).getDatasetSequence(), dnaMappings.get(0)
+ .getTo());
+ assertEquals("G(1) in CDS should map to G(4) in DNA", 4, dnaMappings
+ .get(0).getMap().getToPosition(1));
+ List<Mapping> peptideMappings = cdsMapping
+ .getMappingsForSequence(pep1);
+ assertEquals(1, peptideMappings.size());
+ assertSame(pep1.getDatasetSequence(), peptideMappings.get(0).getTo());
+
+ // mappings for dna1 - cds2 - pep2
+ cdsMapping = newMappingsIterator.next();
+ dnaMappings = cdsMapping.getMappingsForSequence(dna1);
+ assertEquals(1, dnaMappings.size());
+ assertSame(cds.get(1).getDatasetSequence(), dnaMappings.get(0)
+ .getTo());
+ assertEquals("c(4) in CDS should map to c(7) in DNA", 7, dnaMappings
+ .get(0).getMap().getToPosition(4));
+ peptideMappings = cdsMapping.getMappingsForSequence(pep2);
+ assertEquals(1, peptideMappings.size());
+ assertSame(pep2.getDatasetSequence(), peptideMappings.get(0).getTo());
+
+ // mappings for dna1 - cds3 - pep3
+ cdsMapping = newMappingsIterator.next();
+ dnaMappings = cdsMapping.getMappingsForSequence(dna1);
+ assertEquals(1, dnaMappings.size());
+ assertSame(cds.get(2).getDatasetSequence(), dnaMappings.get(0)
+ .getTo());
+ assertEquals("T(4) in CDS should map to T(10) in DNA", 10, dnaMappings
+ .get(0).getMap().getToPosition(4));
+ peptideMappings = cdsMapping.getMappingsForSequence(pep3);
+ assertEquals(1, peptideMappings.size());
+ assertSame(pep3.getDatasetSequence(), peptideMappings.get(0).getTo());
+ }
+
+ @Test(groups = { "Functional" })
+ public void testIsMappable()
+ {
+ SequenceI dna1 = new Sequence("dna1", "cgCAGtgGT");
+ SequenceI aa1 = new Sequence("aa1", "RSG");
+ AlignmentI al1 = new Alignment(new SequenceI[] { dna1 });
+ AlignmentI al2 = new Alignment(new SequenceI[] { aa1 });
+
+ assertFalse(AlignmentUtils.isMappable(null, null));
+ assertFalse(AlignmentUtils.isMappable(al1, null));
+ assertFalse(AlignmentUtils.isMappable(null, al1));
+ assertFalse(AlignmentUtils.isMappable(al1, al1));
+ assertFalse(AlignmentUtils.isMappable(al2, al2));
+
+ assertTrue(AlignmentUtils.isMappable(al1, al2));
+ assertTrue(AlignmentUtils.isMappable(al2, al1));
+ }
+
+ /**
+ * Test creating a mapping when the sequences involved do not start at residue
+ * 1
+ *
+ * @throws IOException
+ */
+ @Test(groups = { "Functional" })
+ public void testMapCdnaToProtein_forSubsequence()
+ throws IOException
+ {
+ SequenceI prot = new Sequence("UNIPROT|V12345", "E-I--Q", 10, 12);
+ prot.createDatasetSequence();
+
+ SequenceI dna = new Sequence("EMBL|A33333", "GAA--AT-C-CAG", 40, 48);
+ dna.createDatasetSequence();
+
+ MapList map = AlignmentUtils.mapCdnaToProtein(prot, dna);
+ assertEquals(10, map.getToLowest());
+ assertEquals(12, map.getToHighest());
+ assertEquals(40, map.getFromLowest());
+ assertEquals(48, map.getFromHighest());
+ }
+
+ /**
+ * Test for the alignSequenceAs method where we have protein mapped to protein
+ */
+ @Test(groups = { "Functional" })
+ public void testAlignSequenceAs_mappedProteinProtein()
+ {
+
+ SequenceI alignMe = new Sequence("Match", "MGAASEV");
+ alignMe.createDatasetSequence();
+ SequenceI alignFrom = new Sequence("Query", "LQTGYMGAASEVMFSPTRR");
+ alignFrom.createDatasetSequence();
+
+ AlignedCodonFrame acf = new AlignedCodonFrame();
+ // this is like a domain or motif match of part of a peptide sequence
+ MapList map = new MapList(new int[] { 6, 12 }, new int[] { 1, 7 }, 1, 1);
+ acf.addMap(alignFrom.getDatasetSequence(),
+ alignMe.getDatasetSequence(), map);
+
+ AlignmentUtils.alignSequenceAs(alignMe, alignFrom, acf, "-", '-', true,
+ true);
+ assertEquals("-----MGAASEV-------", alignMe.getSequenceAsString());
+ }
+
+ /**
+ * Test for the alignSequenceAs method where there are trailing unmapped
+ * residues in the model sequence
+ */
+ @Test(groups = { "Functional" })
+ public void testAlignSequenceAs_withTrailingPeptide()
+ {
+ // map first 3 codons to KPF; G is a trailing unmapped residue
+ MapList map = new MapList(new int[] { 1, 9 }, new int[] { 1, 3 }, 3, 1);
+
+ checkAlignSequenceAs("AAACCCTTT", "K-PFG", true, true, map,
+ "AAA---CCCTTT---");
+ }
+
+ /**
+ * Tests for transferring features between mapped sequences
+ */
+ @Test(groups = { "Functional" })
+ public void testTransferFeatures()
+ {
+ SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
+ SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
+
+ // no overlap
+ dna.addSequenceFeature(new SequenceFeature("type1", "desc1", 1, 2, 1f,
+ null));
+ // partial overlap - to [1, 1]
+ dna.addSequenceFeature(new SequenceFeature("type2", "desc2", 3, 4, 2f,
+ null));
+ // exact overlap - to [1, 3]
+ dna.addSequenceFeature(new SequenceFeature("type3", "desc3", 4, 6, 3f,
+ null));
+ // spanning overlap - to [2, 5]
+ dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
+ null));
+ // exactly overlaps whole mapped range [1, 6]
+ dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
+ null));
+ // no overlap (internal)
+ dna.addSequenceFeature(new SequenceFeature("type6", "desc6", 7, 9, 6f,
+ null));
+ // no overlap (3' end)
+ dna.addSequenceFeature(new SequenceFeature("type7", "desc7", 13, 15,
+ 7f, null));
+ // overlap (3' end) - to [6, 6]
+ dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
+ 8f, null));
+ // extended overlap - to [6, +]
+ dna.addSequenceFeature(new SequenceFeature("type9", "desc9", 12, 13,
+ 9f, null));
+
+ MapList map = new MapList(new int[] { 4, 6, 10, 12 },
+ new int[] { 1, 6 }, 1, 1);
+
+ /*
+ * transferFeatures() will build 'partial overlap' for regions
+ * that partially overlap 5' or 3' (start or end) of target sequence
+ */
+ AlignmentUtils.transferFeatures(dna, cds, map, null);
+ SequenceFeature[] sfs = cds.getSequenceFeatures();
+ assertEquals(6, sfs.length);
+
+ SequenceFeature sf = sfs[0];
+ assertEquals("type2", sf.getType());
+ assertEquals("desc2", sf.getDescription());
+ assertEquals(2f, sf.getScore());
+ assertEquals(1, sf.getBegin());
+ assertEquals(1, sf.getEnd());
+
+ sf = sfs[1];
+ assertEquals("type3", sf.getType());
+ assertEquals("desc3", sf.getDescription());
+ assertEquals(3f, sf.getScore());
+ assertEquals(1, sf.getBegin());
+ assertEquals(3, sf.getEnd());
+
+ sf = sfs[2];
+ assertEquals("type4", sf.getType());
+ assertEquals(2, sf.getBegin());
+ assertEquals(5, sf.getEnd());
+
+ sf = sfs[3];
+ assertEquals("type5", sf.getType());
+ assertEquals(1, sf.getBegin());
+ assertEquals(6, sf.getEnd());
+
+ sf = sfs[4];
+ assertEquals("type8", sf.getType());
+ assertEquals(6, sf.getBegin());
+ assertEquals(6, sf.getEnd());
+
+ sf = sfs[5];
+ assertEquals("type9", sf.getType());
+ assertEquals(6, sf.getBegin());
+ assertEquals(6, sf.getEnd());
+ }
+
+ /**
+ * Tests for transferring features between mapped sequences
+ */
+ @Test(groups = { "Functional" })
+ public void testTransferFeatures_withOmit()
+ {
+ SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
+ SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
+
+ MapList map = new MapList(new int[] { 4, 6, 10, 12 },
+ new int[] { 1, 6 }, 1, 1);
+
+ // [5, 11] maps to [2, 5]
+ dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
+ null));
+ // [4, 12] maps to [1, 6]
+ dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
+ null));
+ // [12, 12] maps to [6, 6]
+ dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
+ 8f, null));
+
+ // desc4 and desc8 are the 'omit these' varargs
+ AlignmentUtils.transferFeatures(dna, cds, map, null, "type4", "type8");
+ SequenceFeature[] sfs = cds.getSequenceFeatures();
+ assertEquals(1, sfs.length);
+
+ SequenceFeature sf = sfs[0];
+ assertEquals("type5", sf.getType());
+ assertEquals(1, sf.getBegin());
+ assertEquals(6, sf.getEnd());
+ }
+
+ /**
+ * Tests for transferring features between mapped sequences
+ */
+ @Test(groups = { "Functional" })
+ public void testTransferFeatures_withSelect()
+ {
+ SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
+ SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
+
+ MapList map = new MapList(new int[] { 4, 6, 10, 12 },
+ new int[] { 1, 6 }, 1, 1);
+
+ // [5, 11] maps to [2, 5]
+ dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
+ null));
+ // [4, 12] maps to [1, 6]
+ dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
+ null));
+ // [12, 12] maps to [6, 6]
+ dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
+ 8f, null));
+
+ // "type5" is the 'select this type' argument
+ AlignmentUtils.transferFeatures(dna, cds, map, "type5");
+ SequenceFeature[] sfs = cds.getSequenceFeatures();
+ assertEquals(1, sfs.length);
+
+ SequenceFeature sf = sfs[0];
+ assertEquals("type5", sf.getType());
+ assertEquals(1, sf.getBegin());
+ assertEquals(6, sf.getEnd());
+ }
+
+ /**
+ * Test the method that extracts the cds-only part of a dna alignment, for the
+ * case where the cds should be aligned to match its nucleotide sequence.
+ */
+ @Test(groups = { "Functional" })
+ public void testMakeCdsAlignment_alternativeTranscripts()
+ {
+ SequenceI dna1 = new Sequence("dna1", "aaaGGGCC-----CTTTaaaGGG");
+ // alternative transcript of same dna skips CCC codon
+ SequenceI dna2 = new Sequence("dna2", "aaaGGGCC-----cttTaaaGGG");
+ // dna3 has no mapping (protein product) so should be ignored here
+ SequenceI dna3 = new Sequence("dna3", "aaaGGGCCCCCGGGcttTaaaGGG");
+ SequenceI pep1 = new Sequence("pep1", "GPFG");
+ SequenceI pep2 = new Sequence("pep2", "GPG");
+ dna1.createDatasetSequence();
+ dna2.createDatasetSequence();
+ dna3.createDatasetSequence();
+ pep1.createDatasetSequence();
+ pep2.createDatasetSequence();
+ dna1.addSequenceFeature(new SequenceFeature("CDS", "cds1", 4, 8, 0f,
+ null));
+ dna1.addSequenceFeature(new SequenceFeature("CDS", "cds2", 9, 12, 0f,
+ null));
+ dna1.addSequenceFeature(new SequenceFeature("CDS", "cds3", 16, 18, 0f,
+ null));
+ dna2.addSequenceFeature(new SequenceFeature("CDS", "cds", 4, 8, 0f,
+ null));
+ dna2.addSequenceFeature(new SequenceFeature("CDS", "cds", 12, 12, 0f,
+ null));
+ dna2.addSequenceFeature(new SequenceFeature("CDS", "cds", 16, 18, 0f,
+ null));
+
+ List<AlignedCodonFrame> mappings = new ArrayList<AlignedCodonFrame>();
+ MapList map = new MapList(new int[] { 4, 12, 16, 18 },
+ new int[] { 1, 4 }, 3, 1);
+ AlignedCodonFrame acf = new AlignedCodonFrame();
+ acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
+ mappings.add(acf);
+ map = new MapList(new int[] { 4, 8, 12, 12, 16, 18 },
+ new int[] { 1, 3 },
+ 3, 1);
+ acf = new AlignedCodonFrame();
+ acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map);
+ mappings.add(acf);
+
+ AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
+ dna.setDataset(null);
+ AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
+ dna1, dna2, dna3 }, mappings, dna);
+ List<SequenceI> cdsSeqs = cds.getSequences();
+ assertEquals(2, cdsSeqs.size());
+ assertEquals("GGGCCCTTTGGG", cdsSeqs.get(0).getSequenceAsString());
+ assertEquals("GGGCC---TGGG", cdsSeqs.get(1).getSequenceAsString());
+
+ /*
+ * verify shared, extended alignment dataset
+ */
+ assertSame(dna.getDataset(), cds.getDataset());
+ assertTrue(dna.getDataset().getSequences()
+ .contains(cdsSeqs.get(0).getDatasetSequence()));
+ assertTrue(dna.getDataset().getSequences()
+ .contains(cdsSeqs.get(1).getDatasetSequence()));
+
+ /*
+ * Verify updated mappings
+ */
+ assertEquals(2, mappings.size());
+
+ /*
+ * Mapping from pep1 to GGGTTT in first new CDS sequence
+ */
+ List<AlignedCodonFrame> pep1Mapping = MappingUtils
+ .findMappingsForSequence(pep1, mappings);
+ assertEquals(1, pep1Mapping.size());
+ /*
+ * maps GPFG to 1-3,4-6,7-9,10-12
+ */
+ SearchResults sr = MappingUtils.buildSearchResults(pep1, 1, mappings);
+ assertEquals(1, sr.getResults().size());
+ Match m = sr.getResults().get(0);
+ assertEquals(cds.getSequenceAt(0).getDatasetSequence(),
+ m.getSequence());
+ assertEquals(1, m.getStart());
+ assertEquals(3, m.getEnd());
+ sr = MappingUtils.buildSearchResults(pep1, 2, mappings);
+ m = sr.getResults().get(0);
+ assertEquals(4, m.getStart());
+ assertEquals(6, m.getEnd());
+ sr = MappingUtils.buildSearchResults(pep1, 3, mappings);
+ m = sr.getResults().get(0);
+ assertEquals(7, m.getStart());
+ assertEquals(9, m.getEnd());
+ sr = MappingUtils.buildSearchResults(pep1, 4, mappings);
+ m = sr.getResults().get(0);
+ assertEquals(10, m.getStart());
+ assertEquals(12, m.getEnd());
+
+ /*
+ * GPG in pep2 map to 1-3,4-6,7-9 in second CDS sequence
+ */
+ List<AlignedCodonFrame> pep2Mapping = MappingUtils
+ .findMappingsForSequence(pep2, mappings);
+ assertEquals(1, pep2Mapping.size());
+ sr = MappingUtils.buildSearchResults(pep2, 1, mappings);
+ assertEquals(1, sr.getResults().size());
+ m = sr.getResults().get(0);
+ assertEquals(cds.getSequenceAt(1).getDatasetSequence(),
+ m.getSequence());
+ assertEquals(1, m.getStart());
+ assertEquals(3, m.getEnd());
+ sr = MappingUtils.buildSearchResults(pep2, 2, mappings);
+ m = sr.getResults().get(0);
+ assertEquals(4, m.getStart());
+ assertEquals(6, m.getEnd());
+ sr = MappingUtils.buildSearchResults(pep2, 3, mappings);
+ m = sr.getResults().get(0);
+ assertEquals(7, m.getStart());
+ assertEquals(9, m.getEnd());
+ }
+
+ /**
+ * Tests for gapped column in sequences
+ */
+ @Test(groups = { "Functional" })
+ public void testIsGappedColumn()
+ {
+ SequenceI seq1 = new Sequence("Seq1", "a--c.tc-a-g");
+ SequenceI seq2 = new Sequence("Seq2", "aa---t--a-g");
+ SequenceI seq3 = new Sequence("Seq3", "ag-c t-g-");
+ List<SequenceI> seqs = Arrays
+ .asList(new SequenceI[] { seq1, seq2, seq3 });
+ // the column number is base 1
+ assertFalse(AlignmentUtils.isGappedColumn(seqs, 1));
+ assertFalse(AlignmentUtils.isGappedColumn(seqs, 2));
+ assertTrue(AlignmentUtils.isGappedColumn(seqs, 3));
+ assertFalse(AlignmentUtils.isGappedColumn(seqs, 4));
+ assertTrue(AlignmentUtils.isGappedColumn(seqs, 5));
+ assertFalse(AlignmentUtils.isGappedColumn(seqs, 6));
+ assertFalse(AlignmentUtils.isGappedColumn(seqs, 7));
+ assertFalse(AlignmentUtils.isGappedColumn(seqs, 8));
+ assertFalse(AlignmentUtils.isGappedColumn(seqs, 9));
+ assertTrue(AlignmentUtils.isGappedColumn(seqs, 10));
+ assertFalse(AlignmentUtils.isGappedColumn(seqs, 11));
+ // out of bounds:
+ assertTrue(AlignmentUtils.isGappedColumn(seqs, 0));
+ assertTrue(AlignmentUtils.isGappedColumn(seqs, 100));
+ assertTrue(AlignmentUtils.isGappedColumn(seqs, -100));
+ assertTrue(AlignmentUtils.isGappedColumn(null, 0));
+ }
+
+ @Test(groups = { "Functional" })
+ public void testFindCdsColumns()
+ {
+ // TODO target method belongs in a general-purpose alignment
+ // analysis method to find columns for feature
+
+ /*
+ * NB this method assumes CDS ranges are contiguous (no introns)
+ */
+ SequenceI gene = new Sequence("gene", "aaacccgggtttaaacccgggttt");
+ SequenceI seq1 = new Sequence("Seq1", "--ac-cgGG-GGaaACC--GGtt-");
+ SequenceI seq2 = new Sequence("Seq2", "AA--CCGG--g-AAA--cG-GTTt");
+ seq1.createDatasetSequence();
+ seq2.createDatasetSequence();
+ seq1.addSequenceFeature(new SequenceFeature("CDS", "cds", 5, 6, 0f,
+ null));
+ seq1.addSequenceFeature(new SequenceFeature("CDS", "cds", 7, 8, 0f,
+ null));
+ seq1.addSequenceFeature(new SequenceFeature("CDS", "cds", 11, 13, 0f,
+ null));
+ seq1.addSequenceFeature(new SequenceFeature("CDS", "cds", 14, 15, 0f,
+ null));
+ seq2.addSequenceFeature(new SequenceFeature("CDS", "cds", 1, 2, 0f,
+ null));
+ seq2.addSequenceFeature(new SequenceFeature("CDS", "cds", 3, 6, 0f,
+ null));
+ seq2.addSequenceFeature(new SequenceFeature("CDS", "cds", 8, 10, 0f,
+ null));
+ seq2.addSequenceFeature(new SequenceFeature("CDS", "cds", 12, 12, 0f,
+ null));
+ seq2.addSequenceFeature(new SequenceFeature("CDS", "cds", 13, 15, 0f,
+ null));
+
+ List<int[]> cdsColumns = AlignmentUtils.findCdsColumns(new SequenceI[] {
+ seq1, seq2 });
+ assertEquals(4, cdsColumns.size());
+ assertEquals("[1, 2]", Arrays.toString(cdsColumns.get(0)));
+ assertEquals("[5, 9]", Arrays.toString(cdsColumns.get(1)));
+ assertEquals("[11, 17]", Arrays.toString(cdsColumns.get(2)));
+ assertEquals("[19, 23]", Arrays.toString(cdsColumns.get(3)));
+ }
+
+ /**
+ * Test the method that realigns protein to match mapped codon alignment.
+ */
+ @Test(groups = { "Functional" })
+ public void testAlignProteinAsDna_incompleteStartCodon()
+ {
+ // seq1: incomplete start codon (not mapped), then [3, 11]
+ SequenceI dna1 = new Sequence("Seq1", "ccAAA-TTT-GGG-");
+ // seq2 codons are [4, 5], [8, 11]
+ SequenceI dna2 = new Sequence("Seq2", "ccaAA-ttT-GGG-");
+ // seq3 incomplete start codon at 'tt'
+ SequenceI dna3 = new Sequence("Seq3", "ccaaa-ttt-GGG-");
+ AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
+ dna.setDataset(null);
+
+ // prot1 has 'X' for incomplete start codon (not mapped)
+ SequenceI prot1 = new Sequence("Seq1", "XKFG"); // X for incomplete start
+ SequenceI prot2 = new Sequence("Seq2", "NG");
+ SequenceI prot3 = new Sequence("Seq3", "XG"); // X for incomplete start
+ AlignmentI protein = new Alignment(new SequenceI[] { prot1, prot2,
+ prot3 });
+ protein.setDataset(null);
+
+ // map dna1 [3, 11] to prot1 [2, 4] KFG
+ MapList map = new MapList(new int[] { 3, 11 }, new int[] { 2, 4 }, 3, 1);
+ AlignedCodonFrame acf = new AlignedCodonFrame();
+ acf.addMap(dna1.getDatasetSequence(), prot1.getDatasetSequence(), map);
+
+ // map dna2 [4, 5] [8, 11] to prot2 [1, 2] NG
+ map = new MapList(new int[] { 4, 5, 8, 11 }, new int[] { 1, 2 }, 3, 1);
+ acf.addMap(dna2.getDatasetSequence(), prot2.getDatasetSequence(), map);
+
+ // map dna3 [9, 11] to prot3 [2, 2] G
+ map = new MapList(new int[] { 9, 11 }, new int[] { 2, 2 }, 3, 1);
+ acf.addMap(dna3.getDatasetSequence(), prot3.getDatasetSequence(), map);
+
+ ArrayList<AlignedCodonFrame> acfs = new ArrayList<AlignedCodonFrame>();
+ acfs.add(acf);
+ protein.setCodonFrames(acfs);
+
+ /*
+ * verify X is included in the aligned proteins, and placed just
+ * before the first mapped residue
+ * CCT is between CCC and TTT
+ */
+ AlignmentUtils.alignProteinAsDna(protein, dna);
+ assertEquals("XK-FG", prot1.getSequenceAsString());
+ assertEquals("--N-G", prot2.getSequenceAsString());
+ assertEquals("---XG", prot3.getSequenceAsString());
+ }
+
+ /**
+ * Tests for the method that maps the subset of a dna sequence that has CDS
+ * (or subtype) feature - case where the start codon is incomplete.
+ */
+ @Test(groups = "Functional")
+ public void testGetCdsRanges_fivePrimeIncomplete()
+ {
+ SequenceI dnaSeq = new Sequence("dna", "aaagGGCCCaaaTTTttt");
+ dnaSeq.createDatasetSequence();
+ SequenceI ds = dnaSeq.getDatasetSequence();
+
+ // CDS for dna 5-6 (incomplete codon), 7-9
+ SequenceFeature sf = new SequenceFeature("CDS", "", 5, 9, 0f, null);
+ sf.setPhase("2"); // skip 2 bases to start of next codon
+ ds.addSequenceFeature(sf);
+ // CDS for dna 13-15
+ sf = new SequenceFeature("CDS_predicted", "", 13, 15, 0f, null);
+ ds.addSequenceFeature(sf);
+
+ List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
+
+ /*
+ * check the mapping starts with the first complete codon
+ */
+ assertEquals(6, MappingUtils.getLength(ranges));
+ assertEquals(2, ranges.size());
+ assertEquals(7, ranges.get(0)[0]);
+ assertEquals(9, ranges.get(0)[1]);
+ assertEquals(13, ranges.get(1)[0]);
+ assertEquals(15, ranges.get(1)[1]);
+ }
+
+ /**
+ * Tests for the method that maps the subset of a dna sequence that has CDS
+ * (or subtype) feature.
+ */
+ @Test(groups = "Functional")
+ public void testGetCdsRanges()
+ {
+ SequenceI dnaSeq = new Sequence("dna", "aaaGGGcccAAATTTttt");
+ dnaSeq.createDatasetSequence();
+ SequenceI ds = dnaSeq.getDatasetSequence();
+
+ // CDS for dna 3-6
+ SequenceFeature sf = new SequenceFeature("CDS", "", 4, 6, 0f, null);
+ 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);
+ ds.addSequenceFeature(sf);
+
+ List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
+ assertEquals(6, MappingUtils.getLength(ranges));
+ assertEquals(2, ranges.size());
+ assertEquals(4, ranges.get(0)[0]);
+ assertEquals(6, ranges.get(0)[1]);
+ assertEquals(10, ranges.get(1)[0]);
+ assertEquals(12, ranges.get(1)[1]);
+ }
+
+ /**
+ * Test the method that computes a map of codon variants for each protein
+ * position from "sequence_variant" features on dna
+ */
+ @Test(groups = "Functional")
+ public void testBuildDnaVariantsMap()
+ {
+ SequenceI dna = new Sequence("dna", "atgAAATTTGGGCCCtag");
+ MapList map = new MapList(new int[] { 1, 18 }, new int[] { 1, 5 }, 3, 1);
+
+ /*
+ * first with no variants on dna
+ */
+ LinkedHashMap<Integer, String[][]> variantsMap = AlignmentUtils
+ .buildDnaVariantsMap(dna, map);
+ assertTrue(variantsMap.isEmpty());
+
+ // single allele codon 1, on base 1
+ SequenceFeature sf = new SequenceFeature("sequence_variant", "", 1, 1,
+ 0f, null);
+ sf.setValue("alleles", "T");
+ dna.addSequenceFeature(sf);
+
+ // two alleles codon 2, on bases 2 and 3
+ sf = new SequenceFeature("sequence_variant", "", 5, 5, 0f, null);
+ sf.setValue("alleles", "T");
+ dna.addSequenceFeature(sf);
+ sf = new SequenceFeature("sequence_variant", "", 6, 6, 0f, null);
+ sf.setValue("alleles", "G");
+ dna.addSequenceFeature(sf);
+
+ // two alleles codon 3, both on base 2
+ sf = new SequenceFeature("sequence_variant", "", 8, 8, 0f, null);
+ sf.setValue("alleles", "C, G");
+ dna.addSequenceFeature(sf);
+
+ // no alleles on codon 4
+ // alleles on codon 5 on all 3 bases
+ sf = new SequenceFeature("sequence_variant", "", 13, 13, 0f, null);
+ sf.setValue("alleles", "C, G"); // (C duplicates given base value)
+ dna.addSequenceFeature(sf);
+ sf = new SequenceFeature("sequence_variant", "", 14, 14, 0f, null);
+ sf.setValue("alleles", "g, a"); // should force to upper-case
+ dna.addSequenceFeature(sf);
+ sf = new SequenceFeature("sequence_variant", "", 15, 15, 0f, null);
+ sf.setValue("alleles", "A, T");
+ dna.addSequenceFeature(sf);
+
+ variantsMap = AlignmentUtils.buildDnaVariantsMap(dna, map);
+ assertEquals(4, variantsMap.size());
+ assertTrue(Arrays.deepEquals(new String[][] { { "A", "T" }, { "T" },
+ { "G" } }, variantsMap.get(1)));
+ assertTrue(Arrays.deepEquals(new String[][] { { "A" }, { "A", "T" },
+ { "A", "G" } }, variantsMap.get(2)));
+ assertTrue(Arrays.deepEquals(new String[][] { { "T" },
+ { "T", "C", "G" }, { "T" } }, variantsMap.get(3)));
+ // duplicated bases are not removed here, handled in computePeptideVariants
+ assertTrue(Arrays.deepEquals(new String[][] { { "C", "C", "G" },
+ { "C", "G", "A" }, { "C", "A", "T" } }, variantsMap.get(5)));
+ }
+
+ /**
+ * Tests for the method that computes all peptide variants given codon
+ * variants
+ */
+ @Test(groups = "Functional")
+ public void testComputePeptideVariants()
+ {
+ String[][] codonVariants = new String[][] { { "A" }, { "G" }, { "T" } };
+
+ /*
+ * AGT codes for S - this is not included in the variants returned
+ */
+ List<String> variants = AlignmentUtils.computePeptideVariants(codonVariants, "S");
+ assertEquals("[]", variants.toString());
+
+ // S is reported if it differs from the current value (A):
+ variants = AlignmentUtils.computePeptideVariants(codonVariants, "A");
+ assertEquals("[S]", variants.toString());
+
+ /*
+ * synonymous variant is not reported
+ */
+ codonVariants = new String[][] { { "A" }, { "G" }, { "C", "T" } };
+ // AGC and AGT both code for S
+ variants = AlignmentUtils.computePeptideVariants(codonVariants, "s");
+ assertEquals("[]", variants.toString());
+
+ /*
+ * equivalent variants are only reported once
+ */
+ codonVariants = new String[][] { { "C" }, { "T" },
+ { "A", "C", "G", "T" } };
+ // CTA CTC CTG CTT all code for L
+ variants = AlignmentUtils.computePeptideVariants(codonVariants, "S");
+ assertEquals("[L]", variants.toString());
+
+ /*
+ * vary codons 1 and 2; variant products are sorted and non-redundant
+ */
+ codonVariants = new String[][] { { "a", "C" }, { "g", "T" }, { "A" } };
+ // aga ata cga cta code for R, I, R, L
+ variants = AlignmentUtils.computePeptideVariants(codonVariants, "S");
+ assertEquals("[I, L, R]", variants.toString());
+
+ /*
+ * vary codons 2 and 3
+ */
+ codonVariants = new String[][] { { "a" }, { "g", "T" }, { "A", "c" } };
+ // aga agc ata atc code for R, S, I, I
+ variants = AlignmentUtils.computePeptideVariants(codonVariants, "S");
+ assertEquals("[I, R]", variants.toString());
+
+ /*
+ * vary codons 1 and 3
+ */
+ codonVariants = new String[][] { { "a", "t" }, { "a" }, { "t", "g" } };
+ // aat aag tat tag code for N, K, Y, STOP - STOP sorted to end
+ variants = AlignmentUtils.computePeptideVariants(codonVariants, "S");
+ assertEquals("[K, N, Y, STOP]", variants.toString());
+
+ /*
+ * vary codons 1, 2 and 3
+ */
+ codonVariants = new String[][] { { "a", "t" }, { "G", "C" },
+ { "t", "g" } };
+ // agt agg act acg tgt tgg tct tcg code for S, R, T, T, C, W, S, S
+ variants = AlignmentUtils.computePeptideVariants(codonVariants, "S");
+ assertEquals("[C, R, T, W]", variants.toString());