import java.util.ArrayList;
import java.util.Arrays;
import java.util.Collections;
+import java.util.HashSet;
import java.util.List;
import java.util.Map;
+import java.util.Set;
import org.junit.Test;
import jalview.datamodel.Annotation;
import jalview.datamodel.DBRefEntry;
import jalview.datamodel.Mapping;
+import jalview.datamodel.SearchResults;
+import jalview.datamodel.SearchResults.Match;
import jalview.datamodel.Sequence;
import jalview.datamodel.SequenceI;
import jalview.io.AppletFormatAdapter;
import jalview.io.FormatAdapter;
import jalview.util.MapList;
+import jalview.util.MappingUtils;
public class AlignmentUtilsTests
{
assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
}
+
+ /**
+ * Test the method that extracts the exon-only part of a dna alignment.
+ */
+ @Test
+ public void testMakeExonAlignment()
+ {
+ SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
+ SequenceI dna2 = new Sequence("dna2", "GGGcccTTTaaaCCC");
+ SequenceI pep1 = new Sequence("pep1", "GF");
+ SequenceI pep2 = new Sequence("pep2", "GFP");
+ dna1.createDatasetSequence();
+ dna2.createDatasetSequence();
+ pep1.createDatasetSequence();
+ pep2.createDatasetSequence();
+
+ Set<AlignedCodonFrame> mappings = new HashSet<AlignedCodonFrame>();
+ 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);
+ 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);
+
+ AlignmentI exons = AlignmentUtils.makeExonAlignment(new SequenceI[]
+ { dna1, dna2 }, mappings);
+ assertEquals(2, exons.getSequences().size());
+ assertEquals("GGGTTT", exons.getSequenceAt(0).getSequenceAsString());
+ assertEquals("GGGTTTCCC", exons.getSequenceAt(1).getSequenceAsString());
+
+ /*
+ * Verify updated mappings
+ */
+ assertEquals(2, mappings.size());
+
+ /*
+ * Mapping from pep1 to GGGTTT in first new exon sequence
+ */
+ List<AlignedCodonFrame> pep1Mapping = MappingUtils
+ .findMappingsForSequence(pep1, mappings);
+ assertEquals(1, pep1Mapping.size());
+ // map G to GGG
+ SearchResults sr = MappingUtils.buildSearchResults(pep1, 1, mappings);
+ assertEquals(1, sr.getResults().size());
+ Match m = sr.getResults().get(0);
+ assertEquals(exons.getSequenceAt(0).getDatasetSequence(),
+ m.getSequence());
+ assertEquals(1, m.getStart());
+ assertEquals(3, m.getEnd());
+ // map F to TTT
+ sr = MappingUtils.buildSearchResults(pep1, 2, mappings);
+ m = sr.getResults().get(0);
+ assertEquals(exons.getSequenceAt(0).getDatasetSequence(),
+ m.getSequence());
+ assertEquals(4, m.getStart());
+ assertEquals(6, m.getEnd());
+
+ /*
+ * Mapping from pep2 to GGGTTTCCC in second new exon sequence
+ */
+ List<AlignedCodonFrame> pep2Mapping = MappingUtils
+ .findMappingsForSequence(pep2, mappings);
+ assertEquals(1, pep2Mapping.size());
+ // map G to GGG
+ sr = MappingUtils.buildSearchResults(pep2, 1, mappings);
+ assertEquals(1, sr.getResults().size());
+ m = sr.getResults().get(0);
+ assertEquals(exons.getSequenceAt(1).getDatasetSequence(),
+ m.getSequence());
+ assertEquals(1, m.getStart());
+ assertEquals(3, m.getEnd());
+ // map F to TTT
+ sr = MappingUtils.buildSearchResults(pep2, 2, mappings);
+ m = sr.getResults().get(0);
+ assertEquals(exons.getSequenceAt(1).getDatasetSequence(),
+ m.getSequence());
+ assertEquals(4, m.getStart());
+ assertEquals(6, m.getEnd());
+ // map P to CCC
+ sr = MappingUtils.buildSearchResults(pep2, 3, mappings);
+ m = sr.getResults().get(0);
+ assertEquals(exons.getSequenceAt(1).getDatasetSequence(),
+ m.getSequence());
+ assertEquals(7, m.getStart());
+ assertEquals(9, m.getEnd());
+ }
+
+ /**
+ * Test the method that makes an exon-only sequence from a DNA sequence and
+ * its product mapping. Test includes the expected case that the DNA sequence
+ * already has a protein product (Uniprot translation) which in turn has an
+ * x-ref to the EMBLCDS record.
+ */
+ @Test
+ public void testMakeExonSequence()
+ {
+ SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
+ SequenceI pep1 = new Sequence("pep1", "GF");
+ dna1.createDatasetSequence();
+ pep1.createDatasetSequence();
+ pep1.getDatasetSequence().addDBRef(
+ new DBRefEntry("EMBLCDS", "2", "A12345"));
+
+ /*
+ * Make the mapping from dna to protein. The protein sequence has a DBRef to
+ * EMBLCDS|A12345.
+ */
+ Set<AlignedCodonFrame> mappings = new HashSet<AlignedCodonFrame>();
+ 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);
+
+ AlignedCodonFrame newMapping = new AlignedCodonFrame();
+ SequenceI exon = AlignmentUtils.makeExonSequence(dna1, acf, newMapping);
+
+ assertEquals("GGGTTT", exon.getSequenceAsString());
+ assertEquals("dna1|A12345", exon.getName());
+ assertEquals(1, exon.getDBRef().length);
+ DBRefEntry cdsRef = exon.getDBRef()[0];
+ assertEquals("EMBLCDS", cdsRef.getSource());
+ assertEquals("2", cdsRef.getVersion());
+ assertEquals("A12345", cdsRef.getAccessionId());
+
+ }
}