package jalview.ext.ensembl; import static org.testng.AssertJUnit.assertEquals; import static org.testng.AssertJUnit.assertFalse; import static org.testng.AssertJUnit.assertTrue; import jalview.datamodel.SequenceDummy; import jalview.datamodel.SequenceFeature; import jalview.datamodel.SequenceI; import jalview.io.gff.SequenceOntologyFactory; import jalview.io.gff.SequenceOntologyLite; import jalview.util.MapList; import java.util.List; import org.testng.Assert; import org.testng.annotations.AfterClass; import org.testng.annotations.BeforeClass; import org.testng.annotations.Test; public class EnsemblCdsTest { @BeforeClass public void setUp() { SequenceOntologyFactory.setInstance(new SequenceOntologyLite()); } @AfterClass public void tearDown() { SequenceOntologyFactory.setInstance(null); } /** * Test that the cdna part of genomic sequence is correctly identified by * 'CDS' features (or subtypes) with the desired transcript as parent */ @Test(groups = "Functional") public void testGetGenomicRangesFromFeatures() { EnsemblCds testee = new EnsemblCds(); SequenceI genomic = new SequenceDummy("chr7"); genomic.setStart(10000); genomic.setEnd(50000); String transcriptId = "ABC123"; // CDS at (start+10000) length 501 SequenceFeature sf = new SequenceFeature("CDS", "", 20000, 20500, 0f, null); sf.setValue("Parent", "transcript:" + transcriptId); sf.setStrand("+"); genomic.addSequenceFeature(sf); // CDS (sub-type) at (start + 10500) length 101 sf = new SequenceFeature("CDS_predicted", "", 10500, 10600, 0f, null); sf.setValue("Parent", "transcript:" + transcriptId); sf.setStrand("+"); genomic.addSequenceFeature(sf); // CDS belonging to a different transcript doesn't count sf = new SequenceFeature("CDS", "", 11500, 12600, 0f, null); sf.setValue("Parent", "transcript:anotherOne"); genomic.addSequenceFeature(sf); // exon feature doesn't count sf = new SequenceFeature("exon", "", 10000, 50000, 0f, null); genomic.addSequenceFeature(sf); // mRNA_region feature doesn't count (parent of CDS) sf = new SequenceFeature("mRNA_region", "", 10000, 50000, 0f, null); genomic.addSequenceFeature(sf); MapList ranges = testee.getGenomicRangesFromFeatures(genomic, transcriptId, 23); List fromRanges = ranges.getFromRanges(); assertEquals(2, fromRanges.size()); // from ranges should be sorted by start order assertEquals(10500, fromRanges.get(0)[0]); assertEquals(10600, fromRanges.get(0)[1]); assertEquals(20000, fromRanges.get(1)[0]); assertEquals(20500, fromRanges.get(1)[1]); // to range should start from given start numbering List toRanges = ranges.getToRanges(); assertEquals(1, toRanges.size()); assertEquals(23, toRanges.get(0)[0]); assertEquals(624, toRanges.get(0)[1]); } /** * Test the method that retains features except for 'CDS' (or subtypes), or * features with parent other than the given id */ @Test(groups = "Functional") public void testRetainFeature() { String accId = "ABC123"; EnsemblCds testee = new EnsemblCds(); SequenceFeature sf = new SequenceFeature("CDS", "", 20000, 20500, 0f, null); assertFalse(testee.retainFeature(sf, accId)); sf.setType("CDS_predicted"); assertFalse(testee.retainFeature(sf, accId)); // other feature with no parent is retained sf.setType("sequence_variant"); assertTrue(testee.retainFeature(sf, accId)); // other feature with desired parent is retained sf.setValue("Parent", "transcript:" + accId); assertTrue(testee.retainFeature(sf, accId)); // feature with wrong parent is not retained sf.setValue("Parent", "transcript:XYZ"); assertFalse(testee.retainFeature(sf, accId)); } /** * Test the method that picks out 'CDS' (or subtype) features with the * accession id as parent */ @Test(groups = "Functional") public void testIdentifiesSequence() { String accId = "ABC123"; EnsemblCds testee = new EnsemblCds(); // cds with no parent not valid SequenceFeature sf = new SequenceFeature("CDS", "", 1, 2, 0f, null); assertFalse(testee.identifiesSequence(sf, accId)); // cds with wrong parent not valid sf.setValue("Parent", "transcript:XYZ"); assertFalse(testee.identifiesSequence(sf, accId)); // cds with right parent is valid sf.setValue("Parent", "transcript:" + accId); assertTrue(testee.identifiesSequence(sf, accId)); // cds sub-type with right parent is valid sf.setType("CDS_predicted"); assertTrue(testee.identifiesSequence(sf, accId)); // transcript not valid: sf.setType("transcript"); assertFalse(testee.identifiesSequence(sf, accId)); // exon not valid: sf.setType("exon"); assertFalse(testee.identifiesSequence(sf, accId)); } @Test(groups = "Functional") public void testIsValidReference() throws Exception { EnsemblSequenceFetcher esq = new EnsemblCds(); Assert.assertTrue(esq.isValidReference("CCDS5863.1")); Assert.assertTrue(esq.isValidReference("ENST00000288602")); Assert.assertTrue(esq.isValidReference("ENSG00000288602")); Assert.assertTrue(esq.isValidReference("ENSP00000288602")); Assert.assertFalse(esq.isValidReference("ENST0000288602")); // non-human species have a 3 character identifier included: Assert.assertTrue(esq.isValidReference("ENSMUSG00000099398")); } }