+ 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<>();
+ 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 testFindCdsPositions_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 testFindCdsPositions()
+ {
+ SequenceI dnaSeq = new Sequence("dna", "aaaGGGcccAAATTTttt");
+ dnaSeq.createDatasetSequence();
+ SequenceI ds = dnaSeq.getDatasetSequence();
+
+ // CDS for dna 10-12
+ SequenceFeature sf = new SequenceFeature("CDS_predicted", "", 10, 12,
+ 0f, null);
+ sf.setStrand("+");
+ ds.addSequenceFeature(sf);
+ // CDS for dna 4-6
+ 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);
+
+ List<int[]> ranges = AlignmentUtils.findCdsPositions(dnaSeq);
+ /*
+ * verify ranges { [4-6], [12-10] }
+ * note CDS ranges are ordered ascending even if the CDS
+ * features are not
+ */
+ 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, List<DnaVariant>[]> variantsMap = AlignmentUtils
+ .buildDnaVariantsMap(dna, map);
+ assertTrue(variantsMap.isEmpty());
+
+ /*
+ * single allele codon 1, on base 1
+ */
+ SequenceFeature sf1 = new SequenceFeature("sequence_variant", "", 1, 1,
+ 0f, null);
+ sf1.setValue("alleles", "T");
+ sf1.setValue("ID", "sequence_variant:rs758803211");
+ dna.addSequenceFeature(sf1);
+
+ /*
+ * two alleles codon 2, on bases 2 and 3 (distinct variants)
+ */
+ SequenceFeature sf2 = new SequenceFeature("sequence_variant", "", 5, 5,
+ 0f, null);
+ sf2.setValue("alleles", "T");
+ sf2.setValue("ID", "sequence_variant:rs758803212");
+ dna.addSequenceFeature(sf2);
+ SequenceFeature sf3 = new SequenceFeature("sequence_variant", "", 6, 6,
+ 0f, null);
+ sf3.setValue("alleles", "G");
+ sf3.setValue("ID", "sequence_variant:rs758803213");
+ dna.addSequenceFeature(sf3);
+
+ /*
+ * two alleles codon 3, both on base 2 (one variant)
+ */
+ SequenceFeature sf4 = new SequenceFeature("sequence_variant", "", 8, 8,
+ 0f, null);
+ sf4.setValue("alleles", "C, G");
+ sf4.setValue("ID", "sequence_variant:rs758803214");
+ dna.addSequenceFeature(sf4);
+
+ // no alleles on codon 4
+
+ /*
+ * alleles on codon 5 on all 3 bases (distinct variants)
+ */
+ SequenceFeature sf5 = new SequenceFeature("sequence_variant", "", 13,
+ 13, 0f, null);
+ sf5.setValue("alleles", "C, G"); // (C duplicates given base value)
+ sf5.setValue("ID", "sequence_variant:rs758803215");
+ dna.addSequenceFeature(sf5);
+ SequenceFeature sf6 = new SequenceFeature("sequence_variant", "", 14,
+ 14, 0f, null);
+ sf6.setValue("alleles", "g, a"); // should force to upper-case
+ sf6.setValue("ID", "sequence_variant:rs758803216");
+ dna.addSequenceFeature(sf6);
+
+ SequenceFeature sf7 = new SequenceFeature("sequence_variant", "", 15,
+ 15, 0f, null);
+ sf7.setValue("alleles", "A, T");
+ sf7.setValue("ID", "sequence_variant:rs758803217");
+ dna.addSequenceFeature(sf7);
+
+ /*
+ * build map - expect variants on positions 1, 2, 3, 5
+ */
+ variantsMap = AlignmentUtils.buildDnaVariantsMap(dna, map);
+ assertEquals(4, variantsMap.size());
+
+ /*
+ * protein residue 1: variant on codon (ATG) base 1, not on 2 or 3
+ */
+ List<DnaVariant>[] pep1Variants = variantsMap.get(1);
+ assertEquals(3, pep1Variants.length);
+ assertEquals(1, pep1Variants[0].size());
+ assertEquals("A", pep1Variants[0].get(0).base); // codon[1] base
+ assertSame(sf1, pep1Variants[0].get(0).variant); // codon[1] variant
+ assertEquals(1, pep1Variants[1].size());
+ assertEquals("T", pep1Variants[1].get(0).base); // codon[2] base
+ assertNull(pep1Variants[1].get(0).variant); // no variant here
+ assertEquals(1, pep1Variants[2].size());
+ assertEquals("G", pep1Variants[2].get(0).base); // codon[3] base
+ assertNull(pep1Variants[2].get(0).variant); // no variant here
+
+ /*
+ * protein residue 2: variants on codon (AAA) bases 2 and 3
+ */
+ List<DnaVariant>[] pep2Variants = variantsMap.get(2);
+ assertEquals(3, pep2Variants.length);
+ assertEquals(1, pep2Variants[0].size());
+ // codon[1] base recorded while processing variant on codon[2]
+ assertEquals("A", pep2Variants[0].get(0).base);
+ assertNull(pep2Variants[0].get(0).variant); // no variant here
+ // codon[2] base and variant:
+ assertEquals(1, pep2Variants[1].size());
+ assertEquals("A", pep2Variants[1].get(0).base);
+ assertSame(sf2, pep2Variants[1].get(0).variant);
+ // codon[3] base was recorded when processing codon[2] variant
+ // and then the variant for codon[3] added to it
+ assertEquals(1, pep2Variants[2].size());
+ assertEquals("A", pep2Variants[2].get(0).base);
+ assertSame(sf3, pep2Variants[2].get(0).variant);
+
+ /*
+ * protein residue 3: variants on codon (TTT) base 2 only
+ */
+ List<DnaVariant>[] pep3Variants = variantsMap.get(3);
+ assertEquals(3, pep3Variants.length);
+ assertEquals(1, pep3Variants[0].size());
+ assertEquals("T", pep3Variants[0].get(0).base); // codon[1] base
+ assertNull(pep3Variants[0].get(0).variant); // no variant here
+ assertEquals(1, pep3Variants[1].size());
+ assertEquals("T", pep3Variants[1].get(0).base); // codon[2] base
+ assertSame(sf4, pep3Variants[1].get(0).variant); // codon[2] variant
+ assertEquals(1, pep3Variants[2].size());
+ assertEquals("T", pep3Variants[2].get(0).base); // codon[3] base
+ assertNull(pep3Variants[2].get(0).variant); // no variant here
+
+ /*
+ * three variants on protein position 5
+ */
+ List<DnaVariant>[] pep5Variants = variantsMap.get(5);
+ assertEquals(3, pep5Variants.length);
+ assertEquals(1, pep5Variants[0].size());
+ assertEquals("C", pep5Variants[0].get(0).base); // codon[1] base
+ assertSame(sf5, pep5Variants[0].get(0).variant); // codon[1] variant
+ assertEquals(1, pep5Variants[1].size());
+ assertEquals("C", pep5Variants[1].get(0).base); // codon[2] base
+ assertSame(sf6, pep5Variants[1].get(0).variant); // codon[2] variant
+ assertEquals(1, pep5Variants[2].size());
+ assertEquals("C", pep5Variants[2].get(0).base); // codon[3] base
+ assertSame(sf7, pep5Variants[2].get(0).variant); // codon[3] variant
+ }
+
+ /**
+ * Tests for the method that computes all peptide variants given codon
+ * variants
+ */
+ @Test(groups = "Functional")
+ public void testComputePeptideVariants()
+ {
+ /*
+ * scenario: AAATTTCCC codes for KFP
+ * variants:
+ * GAA -> E source: Ensembl
+ * CAA -> Q source: dbSNP
+ * TAA -> STOP source: dnSNP
+ * AAG synonymous source: COSMIC
+ * AAT -> N source: Ensembl
+ * ...TTC synonymous source: dbSNP
+ * ......CAC,CGC -> H,R source: COSMIC
+ * (one variant with two alleles)
+ */
+ SequenceI peptide = new Sequence("pep/10-12", "KFP");
+
+ /*
+ * two distinct variants for codon 1 position 1
+ * second one has clinical significance
+ */
+ String ensembl = "Ensembl";
+ String dbSnp = "dbSNP";
+ String cosmic = "COSMIC";
+
+ /*
+ * NB setting "id" (as returned by Ensembl for features in JSON format);
+ * previously "ID" (as returned for GFF3 format)
+ */
+ SequenceFeature sf1 = new SequenceFeature("sequence_variant", "", 1, 1,
+ 0f, ensembl);
+ sf1.setValue("alleles", "A,G"); // AAA -> GAA -> K/E
+ sf1.setValue("id", "var1.125A>G");
+
+ SequenceFeature sf2 = new SequenceFeature("sequence_variant", "", 1, 1,
+ 0f, dbSnp);
+ sf2.setValue("alleles", "A,C"); // AAA -> CAA -> K/Q
+ sf2.setValue("id", "var2");
+ sf2.setValue("clinical_significance", "Dodgy");
+
+ SequenceFeature sf3 = new SequenceFeature("sequence_variant", "", 1, 1,
+ 0f, dbSnp);
+ sf3.setValue("alleles", "A,T"); // AAA -> TAA -> stop codon
+ sf3.setValue("id", "var3");
+ sf3.setValue("clinical_significance", "Bad");
+
+ SequenceFeature sf4 = new SequenceFeature("sequence_variant", "", 3, 3,
+ 0f, cosmic);
+ sf4.setValue("alleles", "A,G"); // AAA -> AAG synonymous
+ sf4.setValue("id", "var4");
+ sf4.setValue("clinical_significance", "None");
+
+ SequenceFeature sf5 = new SequenceFeature("sequence_variant", "", 3, 3,
+ 0f, ensembl);
+ sf5.setValue("alleles", "A,T"); // AAA -> AAT -> K/N
+ sf5.setValue("id", "sequence_variant:var5"); // prefix gets stripped off
+ sf5.setValue("clinical_significance", "Benign");
+
+ SequenceFeature sf6 = new SequenceFeature("sequence_variant", "", 6, 6,
+ 0f, dbSnp);
+ sf6.setValue("alleles", "T,C"); // TTT -> TTC synonymous
+ sf6.setValue("id", "var6");
+
+ SequenceFeature sf7 = new SequenceFeature("sequence_variant", "", 8, 8,
+ 0f, cosmic);
+ sf7.setValue("alleles", "C,A,G"); // CCC -> CAC,CGC -> P/H/R
+ sf7.setValue("id", "var7");
+ sf7.setValue("clinical_significance", "Good");
+
+ List<DnaVariant> codon1Variants = new ArrayList<>();
+ List<DnaVariant> codon2Variants = new ArrayList<>();
+ List<DnaVariant> codon3Variants = new ArrayList<>();
+
+ List<DnaVariant> codonVariants[] = new ArrayList[3];
+ codonVariants[0] = codon1Variants;
+ codonVariants[1] = codon2Variants;
+ codonVariants[2] = codon3Variants;
+
+ /*
+ * compute variants for protein position 1
+ */
+ codon1Variants.add(new DnaVariant("A", sf1));
+ codon1Variants.add(new DnaVariant("A", sf2));
+ codon1Variants.add(new DnaVariant("A", sf3));
+ codon2Variants.add(new DnaVariant("A"));
+ // codon2Variants.add(new DnaVariant("A"));
+ codon3Variants.add(new DnaVariant("A", sf4));
+ codon3Variants.add(new DnaVariant("A", sf5));
+ AlignmentUtils.computePeptideVariants(peptide, 1, codonVariants);
+
+ /*
+ * compute variants for protein position 2
+ */
+ codon1Variants.clear();
+ codon2Variants.clear();
+ codon3Variants.clear();
+ codon1Variants.add(new DnaVariant("T"));
+ codon2Variants.add(new DnaVariant("T"));
+ codon3Variants.add(new DnaVariant("T", sf6));
+ AlignmentUtils.computePeptideVariants(peptide, 2, codonVariants);
+
+ /*
+ * compute variants for protein position 3
+ */
+ codon1Variants.clear();
+ codon2Variants.clear();
+ codon3Variants.clear();
+ codon1Variants.add(new DnaVariant("C"));
+ codon2Variants.add(new DnaVariant("C", sf7));
+ codon3Variants.add(new DnaVariant("C"));
+ AlignmentUtils.computePeptideVariants(peptide, 3, codonVariants);
+
+ /*
+ * verify added sequence features for
+ * var1 K -> E Ensembl
+ * var2 K -> Q dbSNP
+ * var3 K -> stop
+ * var4 synonymous
+ * var5 K -> N Ensembl
+ * var6 synonymous
+ * var7 P -> H COSMIC
+ * var8 P -> R COSMIC
+ */
+ List<SequenceFeature> sfs = peptide.getSequenceFeatures();
+ SequenceFeatures.sortFeatures(sfs, true);
+ assertEquals(8, sfs.size());
+
+ /*
+ * features are sorted by start position ascending, but in no
+ * particular order where start positions match; asserts here
+ * simply match the data returned (the order is not important)
+ */
+ // AAA -> AAT -> K/N
+ SequenceFeature sf = sfs.get(0);
+ assertEquals(1, sf.getBegin());
+ assertEquals(1, sf.getEnd());
+ assertEquals("nonsynonymous_variant", sf.getType());
+ assertEquals("p.Lys1Asn", sf.getDescription());
+ assertEquals("var5", sf.getValue("id"));
+ assertEquals("Benign", sf.getValue("clinical_significance"));
+ assertEquals("id=var5;clinical_significance=Benign",
+ sf.getAttributes());
+ assertEquals(1, sf.links.size());
+ assertEquals(
+ "p.Lys1Asn var5|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var5",
+ sf.links.get(0));
+ assertEquals(ensembl, sf.getFeatureGroup());
+
+ // AAA -> CAA -> K/Q
+ sf = sfs.get(1);
+ assertEquals(1, sf.getBegin());
+ assertEquals(1, sf.getEnd());
+ assertEquals("nonsynonymous_variant", sf.getType());
+ assertEquals("p.Lys1Gln", sf.getDescription());
+ assertEquals("var2", sf.getValue("id"));
+ assertEquals("Dodgy", sf.getValue("clinical_significance"));
+ assertEquals("id=var2;clinical_significance=Dodgy", sf.getAttributes());
+ assertEquals(1, sf.links.size());
+ assertEquals(
+ "p.Lys1Gln var2|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var2",
+ sf.links.get(0));
+ assertEquals(dbSnp, sf.getFeatureGroup());
+
+ // AAA -> GAA -> K/E
+ sf = sfs.get(2);
+ assertEquals(1, sf.getBegin());
+ assertEquals(1, sf.getEnd());
+ assertEquals("nonsynonymous_variant", sf.getType());
+ assertEquals("p.Lys1Glu", sf.getDescription());
+ assertEquals("var1.125A>G", sf.getValue("id"));
+ assertNull(sf.getValue("clinical_significance"));
+ assertEquals("id=var1.125A>G", sf.getAttributes());
+ assertEquals(1, sf.links.size());
+ // link to variation is urlencoded
+ assertEquals(
+ "p.Lys1Glu var1.125A>G|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var1.125A%3EG",
+ sf.links.get(0));
+ assertEquals(ensembl, sf.getFeatureGroup());
+
+ // AAA -> TAA -> stop codon
+ sf = sfs.get(3);
+ assertEquals(1, sf.getBegin());
+ assertEquals(1, sf.getEnd());
+ assertEquals("stop_gained", sf.getType());
+ assertEquals("Aaa/Taa", sf.getDescription());
+ assertEquals("var3", sf.getValue("id"));
+ assertEquals("Bad", sf.getValue("clinical_significance"));
+ assertEquals("id=var3;clinical_significance=Bad", sf.getAttributes());
+ assertEquals(1, sf.links.size());
+ assertEquals(
+ "Aaa/Taa var3|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var3",
+ sf.links.get(0));
+ assertEquals(dbSnp, sf.getFeatureGroup());
+
+ // AAA -> AAG synonymous
+ sf = sfs.get(4);
+ assertEquals(1, sf.getBegin());
+ assertEquals(1, sf.getEnd());
+ assertEquals("synonymous_variant", sf.getType());
+ assertEquals("aaA/aaG", sf.getDescription());
+ assertEquals("var4", sf.getValue("id"));
+ assertEquals("None", sf.getValue("clinical_significance"));
+ assertEquals("id=var4;clinical_significance=None", sf.getAttributes());
+ assertEquals(1, sf.links.size());
+ assertEquals(
+ "aaA/aaG var4|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var4",
+ sf.links.get(0));
+ assertEquals(cosmic, sf.getFeatureGroup());
+
+ // TTT -> TTC synonymous
+ sf = sfs.get(5);
+ assertEquals(2, sf.getBegin());
+ assertEquals(2, sf.getEnd());
+ assertEquals("synonymous_variant", sf.getType());
+ assertEquals("ttT/ttC", sf.getDescription());
+ assertEquals("var6", sf.getValue("id"));
+ assertNull(sf.getValue("clinical_significance"));
+ assertEquals("id=var6", sf.getAttributes());
+ assertEquals(1, sf.links.size());
+ assertEquals(
+ "ttT/ttC var6|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var6",
+ sf.links.get(0));
+ assertEquals(dbSnp, sf.getFeatureGroup());
+
+ // var7 generates two distinct protein variant features (two alleles)
+ // CCC -> CGC -> P/R
+ sf = sfs.get(6);
+ assertEquals(3, sf.getBegin());
+ assertEquals(3, sf.getEnd());
+ assertEquals("nonsynonymous_variant", sf.getType());
+ assertEquals("p.Pro3Arg", sf.getDescription());
+ assertEquals("var7", sf.getValue("id"));
+ assertEquals("Good", sf.getValue("clinical_significance"));
+ assertEquals("id=var7;clinical_significance=Good", sf.getAttributes());
+ assertEquals(1, sf.links.size());
+ assertEquals(
+ "p.Pro3Arg var7|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var7",
+ sf.links.get(0));
+ assertEquals(cosmic, sf.getFeatureGroup());
+
+ // CCC -> CAC -> P/H
+ sf = sfs.get(7);
+ assertEquals(3, sf.getBegin());
+ assertEquals(3, sf.getEnd());
+ assertEquals("nonsynonymous_variant", sf.getType());
+ assertEquals("p.Pro3His", sf.getDescription());
+ assertEquals("var7", sf.getValue("id"));
+ assertEquals("Good", sf.getValue("clinical_significance"));
+ assertEquals("id=var7;clinical_significance=Good", sf.getAttributes());
+ assertEquals(1, sf.links.size());
+ assertEquals(
+ "p.Pro3His var7|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var7",
+ sf.links.get(0));
+ assertEquals(cosmic, sf.getFeatureGroup());
+ }
+
+ /**
+ * 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()