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;
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;
* 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", "AACGacgtCTCCTTGA");
+ dna = new Sequence("dna", "AACGacgtCTCCTCCC");
dna.createDatasetSequence();
dna.addSequenceFeature(new SequenceFeature("CDS", "", 1, 4, null));
dna.addSequenceFeature(new SequenceFeature("CDS", "", 9, 16, null));
Arrays.deepToString(ml.getFromRanges().toArray()));
/*
- * Case 3: CDS not 3 times length of peptide - no mapping is made
+ * 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);
- assertNull(ml);
+ 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 4: incomplete start codon corresponding to X in peptide
+ * Case 6: incomplete start codon corresponding to X in peptide
*/
dna = new Sequence("dna", "ACGacgtCTCCTTGG");
dna.createDatasetSequence();
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<AlignedCodonFrame> 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<AlignedCodonFrame> 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<AlignedCodonFrame> 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<AlignedCodonFrame> 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());
+ }
}
git://source.jalview.org / jalview.git / blobdiff