import static org.testng.AssertJUnit.assertEquals;
import static org.testng.AssertJUnit.assertFalse;
+import static org.testng.AssertJUnit.assertNotNull;
import static org.testng.AssertJUnit.assertNull;
import static org.testng.AssertJUnit.assertSame;
import static org.testng.AssertJUnit.assertTrue;
+import jalview.analysis.AlignmentUtils.DnaVariant;
import jalview.datamodel.AlignedCodonFrame;
import jalview.datamodel.Alignment;
import jalview.datamodel.AlignmentAnnotation;
import java.io.IOException;
import java.util.ArrayList;
import java.util.Arrays;
-import java.util.Iterator;
import java.util.LinkedHashMap;
import java.util.List;
import java.util.Map;
@Test(groups = { "Functional" })
public void testMakeCdsAlignment()
{
+ /*
+ * scenario:
+ * dna1 --> [4, 6] [10,12] --> pep1
+ * dna2 --> [1, 3] [7, 9] [13,15] --> pep1
+ */
SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
SequenceI dna2 = new Sequence("dna2", "GGGcccTTTaaaCCC");
SequenceI pep1 = new Sequence("pep1", "GF");
SequenceI pep2 = new Sequence("pep2", "GFP");
+ pep1.addDBRef(new DBRefEntry("UNIPROT", "0", "pep1"));
+ pep2.addDBRef(new DBRefEntry("UNIPROT", "0", "pep2"));
dna1.createDatasetSequence();
dna2.createDatasetSequence();
pep1.createDatasetSequence();
pep2.createDatasetSequence();
- dna1.addSequenceFeature(new SequenceFeature("CDS", "cds1", 4, 6, 0f,
- null));
- dna1.addSequenceFeature(new SequenceFeature("CDS", "cds2", 10, 12, 0f,
- null));
- dna2.addSequenceFeature(new SequenceFeature("CDS", "cds3", 1, 3, 0f,
- null));
- dna2.addSequenceFeature(new SequenceFeature("CDS", "cds4", 7, 9, 0f,
- null));
- dna2.addSequenceFeature(new SequenceFeature("CDS", "cds5", 13, 15, 0f,
- null));
AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 });
dna.setDataset(null);
- List<AlignedCodonFrame> mappings = new ArrayList<AlignedCodonFrame>();
+ /*
+ * need a sourceDbRef if we are to construct dbrefs to the CDS
+ * sequence
+ */
+ DBRefEntry dbref = new DBRefEntry("ENSEMBL", "0", "dna1");
+ dna1.getDatasetSequence().addDBRef(dbref);
+ org.testng.Assert.assertEquals(dbref, dna1.getPrimaryDBRefs().get(0));
+ dbref = new DBRefEntry("ENSEMBL", "0", "dna2");
+ dna2.getDatasetSequence().addDBRef(dbref);
+ org.testng.Assert.assertEquals(dbref, dna2.getPrimaryDBRefs().get(0));
+
+ /*
+ * CDS sequences are 'discovered' from dna-to-protein mappings on the alignment
+ * dataset (e.g. added from dbrefs by CrossRef.findXrefSequences)
+ */
MapList map = new MapList(new int[] { 4, 6, 10, 12 },
new int[] { 1, 2 }, 3, 1);
AlignedCodonFrame acf = new AlignedCodonFrame();
acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
- mappings.add(acf);
+ dna.addCodonFrame(acf);
map = new MapList(new int[] { 1, 3, 7, 9, 13, 15 }, new int[] { 1, 3 },
3, 1);
acf = new AlignedCodonFrame();
acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map);
- mappings.add(acf);
+ dna.addCodonFrame(acf);
/*
* execute method under test:
*/
AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
- dna1, dna2 }, mappings, dna);
+ dna1, dna2 }, dna.getDataset(), null);
+ /*
+ * verify cds sequences
+ */
assertEquals(2, cds.getSequences().size());
- assertEquals("GGGTTT", cds.getSequenceAt(0)
- .getSequenceAsString());
- assertEquals("GGGTTTCCC", cds.getSequenceAt(1)
- .getSequenceAsString());
+ 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()));
+ SequenceI cds1Dss = cds.getSequenceAt(0).getDatasetSequence();
+ SequenceI cds2Dss = cds.getSequenceAt(1).getDatasetSequence();
+ assertTrue(dna.getDataset().getSequences().contains(cds1Dss));
+ assertTrue(dna.getDataset().getSequences().contains(cds2Dss));
+
+ /*
+ * verify CDS has a dbref with mapping to peptide
+ */
+ assertNotNull(cds1Dss.getDBRefs());
+ assertEquals(1, cds1Dss.getDBRefs().length);
+ dbref = cds1Dss.getDBRefs()[0];
+ assertEquals("UNIPROT", dbref.getSource());
+ assertEquals("0", dbref.getVersion());
+ assertEquals("pep1", dbref.getAccessionId());
+ assertNotNull(dbref.getMap());
+ assertSame(pep1.getDatasetSequence(), dbref.getMap().getTo());
+ MapList cdsMapping = new MapList(new int[] { 1, 6 },
+ new int[] { 1, 2 }, 3, 1);
+ assertEquals(cdsMapping, dbref.getMap().getMap());
/*
- * Verify mappings from CDS to peptide and cDNA to CDS
+ * verify peptide has added a dbref with reverse mapping to CDS
+ */
+ assertNotNull(pep1.getDBRefs());
+ assertEquals(2, pep1.getDBRefs().length);
+ dbref = pep1.getDBRefs()[1];
+ assertEquals("ENSEMBL", dbref.getSource());
+ assertEquals("0", dbref.getVersion());
+ assertEquals("CDS|dna1", dbref.getAccessionId());
+ assertNotNull(dbref.getMap());
+ assertSame(cds1Dss, dbref.getMap().getTo());
+ assertEquals(cdsMapping.getInverse(), dbref.getMap().getMap());
+
+ /*
+ * Verify mappings from CDS to peptide, cDNA to CDS, and cDNA to peptide
* the mappings are on the shared alignment dataset
+ * 6 mappings, 2*(DNA->CDS), 2*(DNA->Pep), 2*(CDS->Pep)
*/
- assertSame(dna.getCodonFrames(), cds.getCodonFrames());
- List<AlignedCodonFrame> cdsMappings = cds.getCodonFrames();
- assertEquals(2, cdsMappings.size());
-
+ List<AlignedCodonFrame> cdsMappings = cds.getDataset().getCodonFrames();
+ 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 pep1 (dna to pep1, cds to pep1)
* Mapping from pep1 to GGGTTT in first new exon sequence
*/
- List<AlignedCodonFrame> pep1Mapping = MappingUtils
+ List<AlignedCodonFrame> pep1Mappings = MappingUtils
.findMappingsForSequence(pep1, cdsMappings);
- assertEquals(1, pep1Mapping.size());
+ assertEquals(2, pep1Mappings.size());
+ List<AlignedCodonFrame> mappings = MappingUtils
+ .findMappingsForSequence(cds.getSequenceAt(0), pep1Mappings);
+ assertEquals(1, mappings.size());
+
// map G to GGG
- SearchResults sr = MappingUtils
- .buildSearchResults(pep1, 1, cdsMappings);
+ SearchResults sr = MappingUtils.buildSearchResults(pep1, 1, mappings);
assertEquals(1, sr.getResults().size());
Match m = sr.getResults().get(0);
- assertSame(cds.getSequenceAt(0).getDatasetSequence(),
- m.getSequence());
+ assertSame(cds1Dss, m.getSequence());
assertEquals(1, m.getStart());
assertEquals(3, m.getEnd());
// map F to TTT
- sr = MappingUtils.buildSearchResults(pep1, 2, cdsMappings);
+ sr = MappingUtils.buildSearchResults(pep1, 2, mappings);
m = sr.getResults().get(0);
- assertSame(cds.getSequenceAt(0).getDatasetSequence(),
- m.getSequence());
+ assertSame(cds1Dss, m.getSequence());
assertEquals(4, m.getStart());
assertEquals(6, m.getEnd());
/*
- * Mapping from pep2 to GGGTTTCCC in second new exon sequence
+ * Two mappings involve pep2 (dna to pep2, cds to pep2)
+ * Verify mapping from pep2 to GGGTTTCCC in second new exon sequence
*/
- List<AlignedCodonFrame> pep2Mapping = MappingUtils
+ List<AlignedCodonFrame> pep2Mappings = MappingUtils
.findMappingsForSequence(pep2, cdsMappings);
- assertEquals(1, pep2Mapping.size());
+ assertEquals(2, pep2Mappings.size());
+ mappings = MappingUtils.findMappingsForSequence(cds.getSequenceAt(1),
+ pep2Mappings);
+ assertEquals(1, mappings.size());
// map G to GGG
- sr = MappingUtils.buildSearchResults(pep2, 1, cdsMappings);
+ sr = MappingUtils.buildSearchResults(pep2, 1, mappings);
assertEquals(1, sr.getResults().size());
m = sr.getResults().get(0);
- assertSame(cds.getSequenceAt(1).getDatasetSequence(),
- m.getSequence());
+ assertSame(cds2Dss, m.getSequence());
assertEquals(1, m.getStart());
assertEquals(3, m.getEnd());
// map F to TTT
- sr = MappingUtils.buildSearchResults(pep2, 2, cdsMappings);
+ sr = MappingUtils.buildSearchResults(pep2, 2, mappings);
m = sr.getResults().get(0);
- assertSame(cds.getSequenceAt(1).getDatasetSequence(),
- m.getSequence());
+ assertSame(cds2Dss, m.getSequence());
assertEquals(4, m.getStart());
assertEquals(6, m.getEnd());
// map P to CCC
- sr = MappingUtils.buildSearchResults(pep2, 3, cdsMappings);
+ sr = MappingUtils.buildSearchResults(pep2, 3, mappings);
m = sr.getResults().get(0);
- assertSame(cds.getSequenceAt(1).getDatasetSequence(),
- m.getSequence());
+ assertSame(cds2Dss, m.getSequence());
assertEquals(7, m.getStart());
assertEquals(9, m.getEnd());
}
pep1.createDatasetSequence();
pep2.createDatasetSequence();
pep3.createDatasetSequence();
- dna1.addSequenceFeature(new SequenceFeature("CDS", "cds1", 4, 6, 0f,
- null));
- dna1.addSequenceFeature(new SequenceFeature("CDS", "cds2", 10, 12, 0f,
- null));
- dna1.addSequenceFeature(new SequenceFeature("CDS", "cds3", 1, 3, 0f,
- null));
- dna1.addSequenceFeature(new SequenceFeature("CDS", "cds4", 7, 9, 0f,
- null));
- dna1.addSequenceFeature(new SequenceFeature("CDS", "cds5", 1, 3, 0f,
- null));
- dna1.addSequenceFeature(new SequenceFeature("CDS", "cds6", 10, 12, 0f,
- null));
pep1.getDatasetSequence().addDBRef(
new DBRefEntry("EMBLCDS", "2", "A12345"));
pep2.getDatasetSequence().addDBRef(
new DBRefEntry("EMBLCDS", "4", "A12347"));
/*
+ * Create the CDS alignment
+ */
+ AlignmentI dna = new Alignment(new SequenceI[] { dna1 });
+ dna.setDataset(null);
+
+ /*
* Make the mappings from dna to protein
*/
- List<AlignedCodonFrame> mappings = new ArrayList<AlignedCodonFrame>();
// map ...GGG...TTT to GF
MapList map = new MapList(new int[] { 4, 6, 10, 12 },
new int[] { 1, 2 }, 3, 1);
AlignedCodonFrame acf = new AlignedCodonFrame();
acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
- mappings.add(acf);
+ dna.addCodonFrame(acf);
// map aaa...ccc to KP
map = new MapList(new int[] { 1, 3, 7, 9 }, new int[] { 1, 2 }, 3, 1);
acf = new AlignedCodonFrame();
acf.addMap(dna1.getDatasetSequence(), pep2.getDatasetSequence(), map);
- mappings.add(acf);
+ dna.addCodonFrame(acf);
// map aaa......TTT to KF
map = new MapList(new int[] { 1, 3, 10, 12 }, new int[] { 1, 2 }, 3, 1);
acf = new AlignedCodonFrame();
acf.addMap(dna1.getDatasetSequence(), pep3.getDatasetSequence(), map);
- mappings.add(acf);
-
- /*
- * Create the CDS alignment; also augments the dna-to-protein mappings with
- * exon-to-protein and exon-to-dna mappings
- */
- AlignmentI dna = new Alignment(new SequenceI[] { dna1 });
- dna.setDataset(null);
+ dna.addCodonFrame(acf);
/*
* execute method under test
*/
AlignmentI cdsal = AlignmentUtils.makeCdsAlignment(
- new SequenceI[] { dna1 }, mappings, dna);
+ new SequenceI[] { dna1 }, dna.getDataset(), null);
/*
* Verify we have 3 cds sequences, mapped to pep1/2/3 respectively
SequenceI cdsSeq = cds.get(0);
assertEquals("GGGTTT", cdsSeq.getSequenceAsString());
// assertEquals("dna1|A12345", cdsSeq.getName());
- assertEquals("dna1|pep1", cdsSeq.getName());
+ assertEquals("CDS|dna1", cdsSeq.getName());
// assertEquals(1, cdsSeq.getDBRefs().length);
// DBRefEntry cdsRef = cdsSeq.getDBRefs()[0];
// assertEquals("EMBLCDS", cdsRef.getSource());
cdsSeq = cds.get(1);
assertEquals("aaaccc", cdsSeq.getSequenceAsString());
// assertEquals("dna1|A12346", cdsSeq.getName());
- assertEquals("dna1|pep2", cdsSeq.getName());
+ assertEquals("CDS|dna1", cdsSeq.getName());
// assertEquals(1, cdsSeq.getDBRefs().length);
// cdsRef = cdsSeq.getDBRefs()[0];
// assertEquals("EMBLCDS", cdsRef.getSource());
cdsSeq = cds.get(2);
assertEquals("aaaTTT", cdsSeq.getSequenceAsString());
// assertEquals("dna1|A12347", cdsSeq.getName());
- assertEquals("dna1|pep3", cdsSeq.getName());
+ assertEquals("CDS|dna1", cdsSeq.getName());
// assertEquals(1, cdsSeq.getDBRefs().length);
// cdsRef = cdsSeq.getDBRefs()[0];
// assertEquals("EMBLCDS", cdsRef.getSource());
* Verify there are mappings from each cds sequence to its protein product
* and also to its dna source
*/
- Iterator<AlignedCodonFrame> newMappingsIterator = cdsal
- .getCodonFrames().iterator();
+ List<AlignedCodonFrame> newMappings = cdsal.getCodonFrames();
- // mappings for dna1 - exon1 - pep1
- AlignedCodonFrame cdsMapping = newMappingsIterator.next();
- List<Mapping> dnaMappings = cdsMapping.getMappingsFromSequence(dna1);
- assertEquals(3, 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.getMappingsFromSequence(cds
- .get(0).getDatasetSequence());
- assertEquals(1, peptideMappings.size());
- assertSame(pep1.getDatasetSequence(), peptideMappings.get(0).getTo());
-
- // mappings for dna1 - cds2 - pep2
- assertSame(cds.get(1).getDatasetSequence(), dnaMappings.get(1)
- .getTo());
- assertEquals("c(4) in CDS should map to c(7) in DNA", 7, dnaMappings
- .get(1).getMap().getToPosition(4));
- peptideMappings = cdsMapping.getMappingsFromSequence(cds.get(1)
- .getDatasetSequence());
- assertEquals(1, peptideMappings.size());
- assertSame(pep2.getDatasetSequence(), peptideMappings.get(0).getTo());
-
- // mappings for dna1 - cds3 - pep3
- assertSame(cds.get(2).getDatasetSequence(), dnaMappings.get(2)
+ /*
+ * 6 mappings involve dna1 (to pep1/2/3, cds1/2/3)
+ */
+ List<AlignedCodonFrame> dnaMappings = MappingUtils
+ .findMappingsForSequence(dna1, newMappings);
+ assertEquals(6, dnaMappings.size());
+
+ /*
+ * dna1 to pep1
+ */
+ List<AlignedCodonFrame> mappings = MappingUtils
+ .findMappingsForSequence(pep1, dnaMappings);
+ assertEquals(1, mappings.size());
+ assertEquals(1, mappings.get(0).getMappings().size());
+ assertSame(pep1.getDatasetSequence(), mappings.get(0).getMappings()
+ .get(0).getMapping().getTo());
+
+ /*
+ * dna1 to cds1
+ */
+ List<AlignedCodonFrame> dnaToCds1Mappings = MappingUtils
+ .findMappingsForSequence(cds.get(0), dnaMappings);
+ Mapping mapping = dnaToCds1Mappings.get(0).getMappings().get(0)
+ .getMapping();
+ assertSame(cds.get(0).getDatasetSequence(), mapping
.getTo());
- assertEquals("T(4) in CDS should map to T(10) in DNA", 10, dnaMappings
- .get(2).getMap().getToPosition(4));
- peptideMappings = cdsMapping.getMappingsFromSequence(cds.get(2)
- .getDatasetSequence());
- assertEquals(1, peptideMappings.size());
- assertSame(pep3.getDatasetSequence(), peptideMappings.get(0).getTo());
+ assertEquals("G(1) in CDS should map to G(4) in DNA", 4, mapping
+ .getMap().getToPosition(1));
+
+ /*
+ * dna1 to pep2
+ */
+ mappings = MappingUtils.findMappingsForSequence(pep2, dnaMappings);
+ assertEquals(1, mappings.size());
+ assertEquals(1, mappings.get(0).getMappings().size());
+ assertSame(pep2.getDatasetSequence(), mappings.get(0).getMappings()
+ .get(0).getMapping().getTo());
+
+ /*
+ * dna1 to cds2
+ */
+ List<AlignedCodonFrame> dnaToCds2Mappings = MappingUtils
+ .findMappingsForSequence(cds.get(1), dnaMappings);
+ mapping = dnaToCds2Mappings.get(0).getMappings().get(0).getMapping();
+ assertSame(cds.get(1).getDatasetSequence(), mapping.getTo());
+ assertEquals("c(4) in CDS should map to c(7) in DNA", 7, mapping
+ .getMap().getToPosition(4));
+
+ /*
+ * dna1 to pep3
+ */
+ mappings = MappingUtils.findMappingsForSequence(pep3, dnaMappings);
+ assertEquals(1, mappings.size());
+ assertEquals(1, mappings.get(0).getMappings().size());
+ assertSame(pep3.getDatasetSequence(), mappings.get(0).getMappings()
+ .get(0).getMapping().getTo());
+
+ /*
+ * dna1 to cds3
+ */
+ List<AlignedCodonFrame> dnaToCds3Mappings = MappingUtils
+ .findMappingsForSequence(cds.get(2), dnaMappings);
+ mapping = dnaToCds3Mappings.get(0).getMappings().get(0).getMapping();
+ assertSame(cds.get(2).getDatasetSequence(), mapping.getTo());
+ assertEquals("T(4) in CDS should map to T(10) in DNA", 10, mapping
+ .getMap().getToPosition(4));
}
@Test(groups = { "Functional" })
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));
+
+ AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
+ dna.setDataset(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);
+ dna.addCodonFrame(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);
+ dna.addCodonFrame(acf);
- AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 });
- dna.setDataset(null);
AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] {
- dna1, dna2, dna3 }, mappings, dna);
+ dna1, dna2, dna3 }, dna.getDataset(), null);
List<SequenceI> cdsSeqs = cds.getSequences();
assertEquals(2, cdsSeqs.size());
assertEquals("GGGCCCTTTGGG", cdsSeqs.get(0).getSequenceAsString());
.contains(cdsSeqs.get(1).getDatasetSequence()));
/*
- * Verify updated mappings
+ * Verify 6 mappings: dna1 to cds1, cds1 to pep1, dna1 to pep1
+ * and the same for dna2/cds2/pep2
*/
- List<AlignedCodonFrame> cdsMappings = cds.getCodonFrames();
- assertEquals(2, cdsMappings.size());
+ List<AlignedCodonFrame> mappings = cds.getCodonFrames();
+ assertEquals(6, mappings.size());
/*
- * Mapping from pep1 to GGGTTT in first new CDS sequence
+ * 2 mappings involve pep1
*/
- List<AlignedCodonFrame> pep1Mapping = MappingUtils
- .findMappingsForSequence(pep1, cdsMappings);
- assertEquals(1, pep1Mapping.size());
+ List<AlignedCodonFrame> pep1Mappings = MappingUtils
+ .findMappingsForSequence(pep1, mappings);
+ assertEquals(2, pep1Mappings.size());
+
/*
+ * Get mapping of pep1 to cds1 and verify it
* maps GPFG to 1-3,4-6,7-9,10-12
*/
- SearchResults sr = MappingUtils
- .buildSearchResults(pep1, 1, cdsMappings);
+ List<AlignedCodonFrame> pep1CdsMappings = MappingUtils
+ .findMappingsForSequence(cds.getSequenceAt(0), pep1Mappings);
+ assertEquals(1, pep1CdsMappings.size());
+ SearchResults sr = MappingUtils.buildSearchResults(pep1, 1,
+ pep1CdsMappings);
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, cdsMappings);
+ sr = MappingUtils.buildSearchResults(pep1, 2, pep1CdsMappings);
m = sr.getResults().get(0);
assertEquals(4, m.getStart());
assertEquals(6, m.getEnd());
- sr = MappingUtils.buildSearchResults(pep1, 3, cdsMappings);
+ sr = MappingUtils.buildSearchResults(pep1, 3, pep1CdsMappings);
m = sr.getResults().get(0);
assertEquals(7, m.getStart());
assertEquals(9, m.getEnd());
- sr = MappingUtils.buildSearchResults(pep1, 4, cdsMappings);
+ sr = MappingUtils.buildSearchResults(pep1, 4, pep1CdsMappings);
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
+ * Get mapping of pep2 to cds2 and verify it
+ * maps GPG in pep2 to 1-3,4-6,7-9 in second CDS sequence
*/
- List<AlignedCodonFrame> pep2Mapping = MappingUtils
- .findMappingsForSequence(pep2, cdsMappings);
- assertEquals(1, pep2Mapping.size());
- sr = MappingUtils.buildSearchResults(pep2, 1, cdsMappings);
+ List<AlignedCodonFrame> pep2Mappings = MappingUtils
+ .findMappingsForSequence(pep2, mappings);
+ assertEquals(2, pep2Mappings.size());
+ List<AlignedCodonFrame> pep2CdsMappings = MappingUtils
+ .findMappingsForSequence(cds.getSequenceAt(1), pep2Mappings);
+ assertEquals(1, pep2CdsMappings.size());
+ sr = MappingUtils.buildSearchResults(pep2, 1, pep2CdsMappings);
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, cdsMappings);
+ sr = MappingUtils.buildSearchResults(pep2, 2, pep2CdsMappings);
m = sr.getResults().get(0);
assertEquals(4, m.getStart());
assertEquals(6, m.getEnd());
- sr = MappingUtils.buildSearchResults(pep2, 3, cdsMappings);
+ sr = MappingUtils.buildSearchResults(pep2, 3, pep2CdsMappings);
m = sr.getResults().get(0);
assertEquals(7, m.getStart());
assertEquals(9, m.getEnd());
/*
* first with no variants on dna
*/
- LinkedHashMap<Integer, List<String[][]>> variantsMap = AlignmentUtils
+ LinkedHashMap<Integer, List<DnaVariant>[]> variantsMap = AlignmentUtils
.buildDnaVariantsMap(dna, map);
assertTrue(variantsMap.isEmpty());
/*
* single allele codon 1, on base 1
*/
- SequenceFeature sf = new SequenceFeature("sequence_variant", "", 1, 1,
+ SequenceFeature sf1 = new SequenceFeature("sequence_variant", "", 1, 1,
0f, null);
- sf.setValue("alleles", "T");
- sf.setValue("ID", "sequence_variant:rs758803211");
- dna.addSequenceFeature(sf);
+ sf1.setValue("alleles", "T");
+ sf1.setValue("ID", "sequence_variant:rs758803211");
+ dna.addSequenceFeature(sf1);
/*
* two alleles codon 2, on bases 2 and 3 (distinct variants)
*/
- sf = new SequenceFeature("sequence_variant", "", 5, 5, 0f, null);
- sf.setValue("alleles", "T");
- sf.setValue("ID", "sequence_variant:rs758803212");
- dna.addSequenceFeature(sf);
- sf = new SequenceFeature("sequence_variant", "", 6, 6, 0f, null);
- sf.setValue("alleles", "G");
- sf.setValue("ID", "sequence_variant:rs758803213");
- dna.addSequenceFeature(sf);
+ 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)
*/
- sf = new SequenceFeature("sequence_variant", "", 8, 8, 0f, null);
- sf.setValue("alleles", "C, G");
- sf.setValue("ID", "sequence_variant:rs758803214");
- dna.addSequenceFeature(sf);
+ 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)
*/
- sf = new SequenceFeature("sequence_variant", "", 13, 13, 0f, null);
- sf.setValue("alleles", "C, G"); // (C duplicates given base value)
- sf.setValue("ID", "sequence_variant:rs758803215");
- dna.addSequenceFeature(sf);
- sf = new SequenceFeature("sequence_variant", "", 14, 14, 0f, null);
- sf.setValue("alleles", "g, a"); // should force to upper-case
- sf.setValue("ID", "sequence_variant:rs758803216");
- dna.addSequenceFeature(sf);
- sf = new SequenceFeature("sequence_variant", "", 15, 15, 0f, null);
- sf.setValue("alleles", "A, T");
- sf.setValue("ID", "sequence_variant:rs758803217");
- dna.addSequenceFeature(sf);
+ 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
assertEquals(4, variantsMap.size());
/*
- * one variant on protein position 1
- */
- assertEquals(1, variantsMap.get(1).size());
- assertTrue(Arrays.deepEquals(new String[][] { { "A", "T" }, { "T" },
- { "G" } }, variantsMap.get(1).get(0)));
-
- /*
- * two variants on protein position 2
- */
- assertEquals(2, variantsMap.get(2).size());
- assertTrue(Arrays.deepEquals(new String[][] { { "A" }, { "A", "T" },
- { "A" } }, variantsMap.get(2).get(0)));
- assertTrue(Arrays.deepEquals(new String[][] { { "A" }, { "A" },
- { "A", "G" } }, variantsMap.get(2).get(1)));
-
- /*
- * one variant on protein position 3
- */
- assertEquals(1, variantsMap.get(3).size());
- assertTrue(Arrays.deepEquals(new String[][] { { "T" },
- { "T", "C", "G" }, { "T" } }, variantsMap.get(3).get(0)));
+ * 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
- * duplicated bases are not removed here, handled in computePeptideVariants
- */
- assertEquals(3, variantsMap.get(5).size());
- assertTrue(Arrays.deepEquals(new String[][] { { "C", "C", "G" },
- { "C" }, { "C" } }, variantsMap.get(5).get(0)));
- assertTrue(Arrays.deepEquals(new String[][] { { "C" },
- { "C", "G", "A" }, { "C" } }, variantsMap.get(5).get(1)));
- assertTrue(Arrays.deepEquals(new String[][] { { "C" }, { "C" },
- { "C", "A", "T" } }, variantsMap.get(5).get(2)));
+ */
+ 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
}
/**
@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
+ * scenario: AAATTTCCC codes for KFP, with variants
+ * GAA -> E
+ * CAA -> Q
+ * AAG synonymous
+ * AAT -> N
+ * TTC synonymous
+ * CAC,CGC -> H,R (as one variant)
*/
- codonVariants = new String[][] { { "A" }, { "G" }, { "C", "T" } };
- // AGC and AGT both code for S
- variants = AlignmentUtils.computePeptideVariants(codonVariants, "s");
- assertEquals("[]", variants.toString());
-
+ SequenceI peptide = new Sequence("pep/10-12", "KFP");
+
/*
- * equivalent variants are only reported once
+ * two distinct variants for codon 1 position 1
+ * second one has clinical significance
*/
- 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());
-
+ SequenceFeature sf1 = new SequenceFeature("sequence_variant", "", 1, 1,
+ 0f, null);
+ sf1.setValue("alleles", "A,G"); // GAA -> E
+ sf1.setValue("ID", "var1.125A>G");
+ SequenceFeature sf2 = new SequenceFeature("sequence_variant", "", 1, 1,
+ 0f, null);
+ sf2.setValue("alleles", "A,C"); // CAA -> Q
+ sf2.setValue("ID", "var2");
+ sf2.setValue("clinical_significance", "Dodgy");
+ SequenceFeature sf3 = new SequenceFeature("sequence_variant", "", 3, 3,
+ 0f, null);
+ sf3.setValue("alleles", "A,G"); // synonymous
+ sf3.setValue("ID", "var3");
+ sf3.setValue("clinical_significance", "None");
+ SequenceFeature sf4 = new SequenceFeature("sequence_variant", "", 3, 3,
+ 0f, null);
+ sf4.setValue("alleles", "A,T"); // AAT -> N
+ sf4.setValue("ID", "sequence_variant:var4"); // prefix gets stripped off
+ sf4.setValue("clinical_significance", "Benign");
+ SequenceFeature sf5 = new SequenceFeature("sequence_variant", "", 6, 6,
+ 0f, null);
+ sf5.setValue("alleles", "T,C"); // synonymous
+ sf5.setValue("ID", "var5");
+ sf5.setValue("clinical_significance", "Bad");
+ SequenceFeature sf6 = new SequenceFeature("sequence_variant", "", 8, 8,
+ 0f, null);
+ sf6.setValue("alleles", "C,A,G"); // CAC,CGC -> H,R
+ sf6.setValue("ID", "var6");
+ sf6.setValue("clinical_significance", "Good");
+
+ List<DnaVariant> codon1Variants = new ArrayList<DnaVariant>();
+ List<DnaVariant> codon2Variants = new ArrayList<DnaVariant>();
+ List<DnaVariant> codon3Variants = new ArrayList<DnaVariant>();
+ List<DnaVariant> codonVariants[] = new ArrayList[3];
+ codonVariants[0] = codon1Variants;
+ codonVariants[1] = codon2Variants;
+ codonVariants[2] = codon3Variants;
+
/*
- * vary codons 1 and 2; variant products are sorted and non-redundant
+ * compute variants for protein position 1
*/
- 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());
-
+ codon1Variants.add(new DnaVariant("A", sf1));
+ codon1Variants.add(new DnaVariant("A", sf2));
+ codon2Variants.add(new DnaVariant("A"));
+ codon2Variants.add(new DnaVariant("A"));
+ codon3Variants.add(new DnaVariant("A", sf3));
+ codon3Variants.add(new DnaVariant("A", sf4));
+ AlignmentUtils.computePeptideVariants(peptide, 1, codonVariants);
+
/*
- * vary codons 2 and 3
+ * compute variants for protein position 2
*/
- 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());
-
+ codon1Variants.clear();
+ codon2Variants.clear();
+ codon3Variants.clear();
+ codon1Variants.add(new DnaVariant("T"));
+ codon2Variants.add(new DnaVariant("T"));
+ codon3Variants.add(new DnaVariant("T", sf5));
+ AlignmentUtils.computePeptideVariants(peptide, 2, codonVariants);
+
/*
- * vary codons 1 and 3
+ * compute variants for protein position 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());
-
+ codon1Variants.clear();
+ codon2Variants.clear();
+ codon3Variants.clear();
+ codon1Variants.add(new DnaVariant("C"));
+ codon2Variants.add(new DnaVariant("C", sf6));
+ codon3Variants.add(new DnaVariant("C"));
+ AlignmentUtils.computePeptideVariants(peptide, 3, codonVariants);
+
/*
- * vary codons 1, 2 and 3
+ * verify added sequence features for
+ * var1 K -> E
+ * var2 K -> Q
+ * var4 K -> N
+ * var6 P -> H
+ * var6 P -> R
*/
- 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());
+ SequenceFeature[] sfs = peptide.getSequenceFeatures();
+ assertEquals(5, sfs.length);
+ SequenceFeature sf = sfs[0];
+ assertEquals(1, sf.getBegin());
+ assertEquals(1, sf.getEnd());
+ 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("Jalview", sf.getFeatureGroup());
+ sf = sfs[1];
+ assertEquals(1, sf.getBegin());
+ assertEquals(1, sf.getEnd());
+ 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("Jalview", sf.getFeatureGroup());
+ sf = sfs[2];
+ assertEquals(1, sf.getBegin());
+ assertEquals(1, sf.getEnd());
+ assertEquals("p.Lys1Asn", sf.getDescription());
+ assertEquals("var4", sf.getValue("ID"));
+ assertEquals("Benign", sf.getValue("clinical_significance"));
+ assertEquals("ID=var4;clinical_significance=Benign", sf.getAttributes());
+ assertEquals(1, sf.links.size());
+ assertEquals(
+ "p.Lys1Asn var4|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var4",
+ sf.links.get(0));
+ assertEquals("Jalview", sf.getFeatureGroup());
+ sf = sfs[3];
+ assertEquals(3, sf.getBegin());
+ assertEquals(3, sf.getEnd());
+ assertEquals("p.Pro3His", sf.getDescription());
+ assertEquals("var6", sf.getValue("ID"));
+ assertEquals("Good", sf.getValue("clinical_significance"));
+ assertEquals("ID=var6;clinical_significance=Good", sf.getAttributes());
+ assertEquals(1, sf.links.size());
+ assertEquals(
+ "p.Pro3His var6|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var6",
+ sf.links.get(0));
+ // var5 generates two distinct protein variant features
+ assertEquals("Jalview", sf.getFeatureGroup());
+ sf = sfs[4];
+ assertEquals(3, sf.getBegin());
+ assertEquals(3, sf.getEnd());
+ assertEquals("p.Pro3Arg", sf.getDescription());
+ assertEquals("var6", sf.getValue("ID"));
+ assertEquals("Good", sf.getValue("clinical_significance"));
+ assertEquals("ID=var6;clinical_significance=Good", sf.getAttributes());
+ assertEquals(1, sf.links.size());
+ assertEquals(
+ "p.Pro3Arg var6|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var6",
+ sf.links.get(0));
+ assertEquals("Jalview", sf.getFeatureGroup());
}
/**
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
+ SearchResults sr = MappingUtils.buildSearchResults(pep3, 1, mappings);
+ assertEquals(1, sr.getResults().size());
+ Match 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
+ */
+ @Test(groups = "Functional")
+ public void testAlignAsSameSequences()
+ {
+ SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
+ SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
+ AlignmentI al1 = new Alignment(new SequenceI[] { dna1, dna2 });
+ ((Alignment) al1).createDatasetAlignment();
+
+ SequenceI dna3 = new Sequence(dna1);
+ SequenceI dna4 = new Sequence(dna2);
+ assertSame(dna3.getDatasetSequence(), dna1.getDatasetSequence());
+ assertSame(dna4.getDatasetSequence(), dna2.getDatasetSequence());
+ String seq1 = "-cc-GG-GT-TT--aaa";
+ dna3.setSequence(seq1);
+ String seq2 = "C--C-Cgg--gtt-tAA-A-";
+ dna4.setSequence(seq2);
+ AlignmentI al2 = new Alignment(new SequenceI[] { dna3, dna4 });
+ ((Alignment) al2).createDatasetAlignment();
+
+ assertTrue(AlignmentUtils.alignAsSameSequences(al1, al2));
+ assertEquals(seq1, al1.getSequenceAt(0).getSequenceAsString());
+ assertEquals(seq2, al1.getSequenceAt(1).getSequenceAsString());
+
+ /*
+ * add another sequence to 'aligned' - should still succeed, since
+ * unaligned sequences still share a dataset with aligned sequences
+ */
+ SequenceI dna5 = new Sequence("dna5", "CCCgggtttAAA");
+ dna5.createDatasetSequence();
+ al2.addSequence(dna5);
+ assertTrue(AlignmentUtils.alignAsSameSequences(al1, al2));
+ assertEquals(seq1, al1.getSequenceAt(0).getSequenceAsString());
+ assertEquals(seq2, al1.getSequenceAt(1).getSequenceAsString());
+
+ /*
+ * add another sequence to 'unaligned' - should fail, since now not
+ * all unaligned sequences share a dataset with aligned sequences
+ */
+ SequenceI dna6 = new Sequence("dna6", "CCCgggtttAAA");
+ dna6.createDatasetSequence();
+ al1.addSequence(dna6);
+ // JAL-2110 JBP Comment: what's the use case for this behaviour ?
+ assertFalse(AlignmentUtils.alignAsSameSequences(al1, al2));
+ }
+
+ @Test(groups = "Functional")
+ public void testAlignAsSameSequencesMultipleSubSeq()
+ {
+ SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa");
+ SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA");
+ SequenceI as1 = dna1.deriveSequence(), as2 = dna1.deriveSequence()
+ .getSubSequence(3, 7), as3 = dna2.deriveSequence();
+ as1.insertCharAt(6, 5, '-');
+ String s_as1 = as1.getSequenceAsString();
+ as2.insertCharAt(6, 5, '-');
+ String s_as2 = as2.getSequenceAsString();
+ as3.insertCharAt(6, 5, '-');
+ String s_as3 = as3.getSequenceAsString();
+ AlignmentI aligned = new Alignment(new SequenceI[] { as1, as2, as3 });
+
+ // why do we need to cast this still ?
+ ((Alignment) aligned).createDatasetAlignment();
+ SequenceI uas1 = dna1.deriveSequence(), uas2 = dna1.deriveSequence()
+ .getSubSequence(3, 7), uas3 = dna2.deriveSequence();
+ AlignmentI tobealigned = new Alignment(new SequenceI[] { uas1, uas2,
+ uas3 });
+ ((Alignment) tobealigned).createDatasetAlignment();
+
+ assertTrue(AlignmentUtils.alignAsSameSequences(tobealigned, aligned));
+ assertEquals(s_as1, uas1.getSequenceAsString());
+ assertEquals(s_as2, uas2.getSequenceAsString());
+ assertEquals(s_as3, uas3.getSequenceAsString());
+ }
+
}