X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=test%2Fjalview%2Fanalysis%2FAlignmentUtilsTests.java;h=35196faaccb4e58d7ea4386716c3dc4df68d75c4;hb=07194be147ec7a652d1029c9460fb6426db13b7e;hp=4439bb9368dd94f3bc0a640ee0858c6872269185;hpb=df2bf54c7050c42db156e96e73bbabce63e718d7;p=jalview.git diff --git a/test/jalview/analysis/AlignmentUtilsTests.java b/test/jalview/analysis/AlignmentUtilsTests.java index 4439bb9..35196fa 100644 --- a/test/jalview/analysis/AlignmentUtilsTests.java +++ b/test/jalview/analysis/AlignmentUtilsTests.java @@ -47,6 +47,7 @@ import jalview.io.DataSourceType; import jalview.io.FileFormat; import jalview.io.FileFormatI; import jalview.io.FormatAdapter; +import jalview.io.gff.SequenceOntologyI; import jalview.util.MapList; import jalview.util.MappingUtils; @@ -263,14 +264,14 @@ public class AlignmentUtilsTests @Test(groups = { "Functional" }) public void testMapProteinAlignmentToCdna_noXrefs() throws IOException { - List protseqs = new ArrayList(); + List protseqs = new ArrayList<>(); protseqs.add(new Sequence("UNIPROT|V12345", "EIQ")); protseqs.add(new Sequence("UNIPROT|V12346", "EIQ")); protseqs.add(new Sequence("UNIPROT|V12347", "SAR")); AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3])); protein.setDataset(null); - List dnaseqs = new ArrayList(); + List dnaseqs = new ArrayList<>(); dnaseqs.add(new Sequence("EMBL|A11111", "TCAGCACGC")); // = SAR dnaseqs.add(new Sequence("EMBL|A22222", "GAGATACAA")); // = EIQ dnaseqs.add(new Sequence("EMBL|A33333", "GAAATCCAG")); // = EIQ @@ -507,7 +508,7 @@ public class AlignmentUtilsTests acf.addMap(dna1.getDatasetSequence(), prot1.getDatasetSequence(), map); acf.addMap(dna2.getDatasetSequence(), prot2.getDatasetSequence(), map); acf.addMap(dna3.getDatasetSequence(), prot3.getDatasetSequence(), map); - ArrayList acfs = new ArrayList(); + ArrayList acfs = new ArrayList<>(); acfs.add(acf); protein.setCodonFrames(acfs); @@ -605,14 +606,14 @@ public class AlignmentUtilsTests public void testMapProteinAlignmentToCdna_withStartAndStopCodons() throws IOException { - List protseqs = new ArrayList(); + List protseqs = new ArrayList<>(); protseqs.add(new Sequence("UNIPROT|V12345", "EIQ")); protseqs.add(new Sequence("UNIPROT|V12346", "EIQ")); protseqs.add(new Sequence("UNIPROT|V12347", "SAR")); AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3])); protein.setDataset(null); - List dnaseqs = new ArrayList(); + List dnaseqs = new ArrayList<>(); // start + SAR: dnaseqs.add(new Sequence("EMBL|A11111", "ATGTCAGCACGC")); // = EIQ + stop @@ -697,14 +698,14 @@ public class AlignmentUtilsTests @Test(groups = { "Functional" }) public void testMapProteinAlignmentToCdna_withXrefs() throws IOException { - List protseqs = new ArrayList(); + List protseqs = new ArrayList<>(); protseqs.add(new Sequence("UNIPROT|V12345", "EIQ")); protseqs.add(new Sequence("UNIPROT|V12346", "EIQ")); protseqs.add(new Sequence("UNIPROT|V12347", "SAR")); AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3])); protein.setDataset(null); - List dnaseqs = new ArrayList(); + List dnaseqs = new ArrayList<>(); dnaseqs.add(new Sequence("EMBL|A11111", "TCAGCACGC")); // = SAR dnaseqs.add(new Sequence("EMBL|A22222", "ATGGAGATACAA")); // = start + EIQ dnaseqs.add(new Sequence("EMBL|A33333", "GAAATCCAG")); // = EIQ @@ -774,14 +775,14 @@ public class AlignmentUtilsTests public void testMapProteinAlignmentToCdna_prioritiseXrefs() throws IOException { - List protseqs = new ArrayList(); + List protseqs = new ArrayList<>(); protseqs.add(new Sequence("UNIPROT|V12345", "EIQ")); protseqs.add(new Sequence("UNIPROT|V12346", "EIQ")); AlignmentI protein = new Alignment( protseqs.toArray(new SequenceI[protseqs.size()])); protein.setDataset(null); - List dnaseqs = new ArrayList(); + List dnaseqs = new ArrayList<>(); dnaseqs.add(new Sequence("EMBL|A11111", "GAAATCCAG")); // = EIQ dnaseqs.add(new Sequence("EMBL|A22222", "GAAATTCAG")); // = EIQ AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[dnaseqs @@ -848,8 +849,8 @@ public class AlignmentUtilsTests al.addAnnotation(ann4); // Temp for seq1 al.addAnnotation(ann5); // Temp for seq2 al.addAnnotation(ann6); // Temp for no sequence - List types = new ArrayList(); - List scope = new ArrayList(); + List types = new ArrayList<>(); + List scope = new ArrayList<>(); /* * Set all sequence related Structure to hidden (ann1, ann2) @@ -1747,7 +1748,7 @@ public class AlignmentUtilsTests map = new MapList(new int[] { 9, 11 }, new int[] { 2, 2 }, 3, 1); acf.addMap(dna3.getDatasetSequence(), prot3.getDatasetSequence(), map); - ArrayList acfs = new ArrayList(); + ArrayList acfs = new ArrayList<>(); acfs.add(acf); protein.setCodonFrames(acfs); @@ -2030,9 +2031,9 @@ public class AlignmentUtilsTests sf6.setValue("ID", "var6"); sf6.setValue("clinical_significance", "Good"); - List codon1Variants = new ArrayList(); - List codon2Variants = new ArrayList(); - List codon3Variants = new ArrayList(); + List codon1Variants = new ArrayList<>(); + List codon2Variants = new ArrayList<>(); + List codon3Variants = new ArrayList<>(); List codonVariants[] = new ArrayList[3]; codonVariants[0] = codon1Variants; codonVariants[1] = codon2Variants; @@ -2272,7 +2273,7 @@ public class AlignmentUtilsTests seq1.createDatasetSequence(); Mapping mapping = new Mapping(seq1, new MapList( new int[] { 3, 6, 9, 10 }, new int[] { 1, 6 }, 1, 1)); - Map> map = new TreeMap>(); + Map> map = new TreeMap<>(); AlignmentUtils.addMappedPositions(seq1, from, mapping, map); /* @@ -2304,7 +2305,7 @@ public class AlignmentUtilsTests seq1.createDatasetSequence(); Mapping mapping = new Mapping(seq1, new MapList( new int[] { 3, 6, 9, 10 }, new int[] { 1, 6 }, 1, 1)); - Map> map = new TreeMap>(); + Map> map = new TreeMap<>(); AlignmentUtils.addMappedPositions(seq1, from, mapping, map); /* @@ -2533,4 +2534,243 @@ public class AlignmentUtilsTests assertEquals(s_as3, uas3.getSequenceAsString()); } + /** + * Tests for the method that maps nucleotide to protein based on CDS features + */ + @Test(groups = "Functional") + public void testMapCdsToProtein() + { + SequenceI peptide = new Sequence("pep", "KLQ"); + + /* + * Case 1: CDS 3 times length of peptide + * NB method only checks lengths match, not translation + */ + SequenceI dna = new Sequence("dna", "AACGacgtCTCCT"); + dna.createDatasetSequence(); + dna.addSequenceFeature(new SequenceFeature("CDS", "", 1, 4, null)); + dna.addSequenceFeature(new SequenceFeature("CDS", "", 9, 13, null)); + MapList ml = AlignmentUtils.mapCdsToProtein(dna, peptide); + assertEquals(3, ml.getFromRatio()); + assertEquals(1, ml.getToRatio()); + assertEquals("[[1, 3]]", + Arrays.deepToString(ml.getToRanges().toArray())); + assertEquals("[[1, 4], [9, 13]]", + Arrays.deepToString(ml.getFromRanges().toArray())); + + /* + * Case 2: CDS 3 times length of peptide + stop codon + * (note code does not currently check trailing codon is a stop codon) + */ + dna = new Sequence("dna", "AACGacgtCTCCTCCC"); + dna.createDatasetSequence(); + dna.addSequenceFeature(new SequenceFeature("CDS", "", 1, 4, null)); + dna.addSequenceFeature(new SequenceFeature("CDS", "", 9, 16, null)); + ml = AlignmentUtils.mapCdsToProtein(dna, peptide); + assertEquals(3, ml.getFromRatio()); + assertEquals(1, ml.getToRatio()); + assertEquals("[[1, 3]]", + Arrays.deepToString(ml.getToRanges().toArray())); + assertEquals("[[1, 4], [9, 13]]", + Arrays.deepToString(ml.getFromRanges().toArray())); + + /* + * Case 3: CDS longer than 3 * peptide + stop codon - no mapping is made + */ + dna = new Sequence("dna", "AACGacgtCTCCTTGATCA"); + dna.createDatasetSequence(); + dna.addSequenceFeature(new SequenceFeature("CDS", "", 1, 4, null)); + dna.addSequenceFeature(new SequenceFeature("CDS", "", 9, 19, null)); + ml = AlignmentUtils.mapCdsToProtein(dna, peptide); + assertNull(ml); + + /* + * Case 4: CDS shorter than 3 * peptide - no mapping is made + */ + dna = new Sequence("dna", "AACGacgtCTCC"); + dna.createDatasetSequence(); + dna.addSequenceFeature(new SequenceFeature("CDS", "", 1, 4, null)); + dna.addSequenceFeature(new SequenceFeature("CDS", "", 9, 12, null)); + ml = AlignmentUtils.mapCdsToProtein(dna, peptide); + assertNull(ml); + + /* + * Case 5: CDS 3 times length of peptide + part codon - mapping is truncated + */ + dna = new Sequence("dna", "AACGacgtCTCCTTG"); + dna.createDatasetSequence(); + dna.addSequenceFeature(new SequenceFeature("CDS", "", 1, 4, null)); + dna.addSequenceFeature(new SequenceFeature("CDS", "", 9, 15, null)); + ml = AlignmentUtils.mapCdsToProtein(dna, peptide); + assertEquals(3, ml.getFromRatio()); + assertEquals(1, ml.getToRatio()); + assertEquals("[[1, 3]]", + Arrays.deepToString(ml.getToRanges().toArray())); + assertEquals("[[1, 4], [9, 13]]", + Arrays.deepToString(ml.getFromRanges().toArray())); + + /* + * Case 6: incomplete start codon corresponding to X in peptide + */ + dna = new Sequence("dna", "ACGacgtCTCCTTGG"); + dna.createDatasetSequence(); + SequenceFeature sf = new SequenceFeature("CDS", "", 1, 3, null); + sf.setPhase("2"); // skip 2 positions (AC) to start of next codon (GCT) + dna.addSequenceFeature(sf); + dna.addSequenceFeature(new SequenceFeature("CDS", "", 8, 15, null)); + peptide = new Sequence("pep", "XLQ"); + ml = AlignmentUtils.mapCdsToProtein(dna, peptide); + assertEquals("[[2, 3]]", + Arrays.deepToString(ml.getToRanges().toArray())); + assertEquals("[[3, 3], [8, 12]]", + Arrays.deepToString(ml.getFromRanges().toArray())); + } + + /** + * Tests for the method that locates the CDS sequence that has a mapping to + * the given protein. That is, given a transcript-to-peptide mapping, find the + * cds-to-peptide mapping that relates to both, and return the CDS sequence. + */ + @Test + public void testFindCdsForProtein() + { + List mappings = new ArrayList<>(); + AlignedCodonFrame acf1 = new AlignedCodonFrame(); + mappings.add(acf1); + + SequenceI dna1 = new Sequence("dna1", "cgatATcgGCTATCTATGacg"); + dna1.createDatasetSequence(); + + // NB we currently exclude STOP codon from CDS sequences + // the test would need to change if this changes in future + SequenceI cds1 = new Sequence("cds1", "ATGCTATCT"); + cds1.createDatasetSequence(); + + SequenceI pep1 = new Sequence("pep1", "MLS"); + pep1.createDatasetSequence(); + List seqMappings = new ArrayList<>(); + MapList mapList = new MapList( + new int[] + { 5, 6, 9, 15 }, new int[] { 1, 3 }, 3, 1); + Mapping dnaToPeptide = new Mapping(pep1.getDatasetSequence(), mapList); + + // add dna to peptide mapping + seqMappings.add(acf1); + acf1.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), + mapList); + + /* + * first case - no dna-to-CDS mapping exists - search fails + */ + SequenceI seq = AlignmentUtils.findCdsForProtein(mappings, dna1, + seqMappings, dnaToPeptide); + assertNull(seq); + + /* + * second case - CDS-to-peptide mapping exists but no dna-to-CDS + * - search fails + */ + // todo this test fails if the mapping is added to acf1, not acf2 + // need to tidy up use of lists of mappings in AlignedCodonFrame + AlignedCodonFrame acf2 = new AlignedCodonFrame(); + mappings.add(acf2); + MapList cdsToPeptideMapping = new MapList(new int[] + { 1, 9 }, new int[] { 1, 3 }, 3, 1); + acf2.addMap(cds1.getDatasetSequence(), pep1.getDatasetSequence(), + cdsToPeptideMapping); + assertNull(AlignmentUtils.findCdsForProtein(mappings, dna1, seqMappings, + dnaToPeptide)); + + /* + * third case - add dna-to-CDS mapping - CDS is now found! + */ + MapList dnaToCdsMapping = new MapList(new int[] { 5, 6, 9, 15 }, + new int[] + { 1, 9 }, 1, 1); + acf1.addMap(dna1.getDatasetSequence(), cds1.getDatasetSequence(), + dnaToCdsMapping); + seq = AlignmentUtils.findCdsForProtein(mappings, dna1, seqMappings, + dnaToPeptide); + assertSame(seq, cds1.getDatasetSequence()); + } + + /** + * Tests for the method that locates the CDS sequence that has a mapping to + * the given protein. That is, given a transcript-to-peptide mapping, find the + * cds-to-peptide mapping that relates to both, and return the CDS sequence. + * This test is for the case where transcript and CDS are the same length. + */ + @Test + public void testFindCdsForProtein_noUTR() + { + List mappings = new ArrayList<>(); + AlignedCodonFrame acf1 = new AlignedCodonFrame(); + mappings.add(acf1); + + SequenceI dna1 = new Sequence("dna1", "ATGCTATCTTAA"); + dna1.createDatasetSequence(); + + // NB we currently exclude STOP codon from CDS sequences + // the test would need to change if this changes in future + SequenceI cds1 = new Sequence("cds1", "ATGCTATCT"); + cds1.createDatasetSequence(); + + SequenceI pep1 = new Sequence("pep1", "MLS"); + pep1.createDatasetSequence(); + List seqMappings = new ArrayList<>(); + MapList mapList = new MapList( + new int[] + { 1, 9 }, new int[] { 1, 3 }, 3, 1); + Mapping dnaToPeptide = new Mapping(pep1.getDatasetSequence(), mapList); + + // add dna to peptide mapping + seqMappings.add(acf1); + acf1.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), + mapList); + + /* + * first case - transcript lacks CDS features - it appears to be + * the CDS sequence and is returned + */ + SequenceI seq = AlignmentUtils.findCdsForProtein(mappings, dna1, + seqMappings, dnaToPeptide); + assertSame(seq, dna1.getDatasetSequence()); + + /* + * second case - transcript has CDS feature - this means it is + * not returned as a match for CDS (CDS sequences don't have CDS features) + */ + dna1.addSequenceFeature( + new SequenceFeature(SequenceOntologyI.CDS, "cds", 1, 12, null)); + seq = AlignmentUtils.findCdsForProtein(mappings, dna1, seqMappings, + dnaToPeptide); + assertNull(seq); + + /* + * third case - CDS-to-peptide mapping exists but no dna-to-CDS + * - search fails + */ + // todo this test fails if the mapping is added to acf1, not acf2 + // need to tidy up use of lists of mappings in AlignedCodonFrame + AlignedCodonFrame acf2 = new AlignedCodonFrame(); + mappings.add(acf2); + MapList cdsToPeptideMapping = new MapList(new int[] + { 1, 9 }, new int[] { 1, 3 }, 3, 1); + acf2.addMap(cds1.getDatasetSequence(), pep1.getDatasetSequence(), + cdsToPeptideMapping); + assertNull(AlignmentUtils.findCdsForProtein(mappings, dna1, seqMappings, + dnaToPeptide)); + + /* + * fourth case - add dna-to-CDS mapping - CDS is now found! + */ + MapList dnaToCdsMapping = new MapList(new int[] { 1, 9 }, + new int[] + { 1, 9 }, 1, 1); + acf1.addMap(dna1.getDatasetSequence(), cds1.getDatasetSequence(), + dnaToCdsMapping); + seq = AlignmentUtils.findCdsForProtein(mappings, dna1, seqMappings, + dnaToPeptide); + assertSame(seq, cds1.getDatasetSequence()); + } }