X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=test%2Fjalview%2Fdatamodel%2FSequenceTest.java;h=c7e53a9211e17c8e8a63797481dcbaca1e61fcd7;hb=e44876a5a8801c0118b67793ebd57d21079be7b0;hp=0e9ab28de03645e4a83b2aaa51b6cb070d7c6e0e;hpb=81962dc8c5cd2088ef2b3aff2cbdded0230dde53;p=jalview.git diff --git a/test/jalview/datamodel/SequenceTest.java b/test/jalview/datamodel/SequenceTest.java index 0e9ab28..c7e53a9 100644 --- a/test/jalview/datamodel/SequenceTest.java +++ b/test/jalview/datamodel/SequenceTest.java @@ -29,11 +29,15 @@ import static org.testng.AssertJUnit.assertTrue; import static org.testng.internal.junit.ArrayAsserts.assertArrayEquals; import jalview.datamodel.PDBEntry.Type; +import jalview.util.MapList; +import java.io.File; +import java.util.ArrayList; import java.util.Arrays; import java.util.List; import java.util.Vector; +import org.testng.Assert; import org.testng.annotations.BeforeMethod; import org.testng.annotations.Test; @@ -62,6 +66,30 @@ public class SequenceTest assertEquals("Gap interval 2 end wrong", 8, gapInt.get(1)[1]); } + @Test(groups = ("Functional")) + public void testIsProtein() + { + // test Protein + assertTrue(new Sequence("prot","ASDFASDFASDF").isProtein()); + // test DNA + assertFalse(new Sequence("prot","ACGTACGTACGT").isProtein()); + // test RNA + SequenceI sq = new Sequence("prot","ACGUACGUACGU"); + assertFalse(sq.isProtein()); + // change sequence, should trigger an update of cached result + sq.setSequence("ASDFASDFADSF"); + assertTrue(sq.isProtein()); + /* + * in situ change of sequence doesn't change hashcode :-O + * (sequence should not expose internal implementation) + */ + for (int i = 0; i < sq.getSequence().length; i++) + { + sq.getSequence()[i] = "acgtu".charAt(i % 5); + } + assertTrue(sq.isProtein()); // but it isn't + } + @Test(groups = { "Functional" }) public void testGetAnnotation() { @@ -83,8 +111,7 @@ public class SequenceTest { AlignmentAnnotation ann1 = addAnnotation("label1", "desc1", "calcId1", 1f); - AlignmentAnnotation ann2 = addAnnotation("label2", "desc2", "calcId2", - 1f); + addAnnotation("label2", "desc2", "calcId2", 1f); AlignmentAnnotation ann3 = addAnnotation("label1", "desc3", "calcId3", 1f); AlignmentAnnotation[] anns = seq.getAnnotation("label1"); @@ -106,16 +133,15 @@ public class SequenceTest @Test(groups = { "Functional" }) public void testGetAlignmentAnnotations_forCalcIdAndLabel() { - AlignmentAnnotation ann1 = addAnnotation("label1", "desc1", "calcId1", - 1f); + addAnnotation("label1", "desc1", "calcId1", 1f); AlignmentAnnotation ann2 = addAnnotation("label2", "desc2", "calcId2", 1f); - AlignmentAnnotation ann3 = addAnnotation("label2", "desc3", "calcId3", - 1f); + addAnnotation("label2", "desc3", "calcId3", 1f); AlignmentAnnotation ann4 = addAnnotation("label2", "desc3", "calcId2", 1f); - AlignmentAnnotation ann5 = addAnnotation("label5", "desc3", null, 1f); - AlignmentAnnotation ann6 = addAnnotation(null, "desc3", "calcId3", 1f); + addAnnotation("label5", "desc3", null, 1f); + addAnnotation(null, "desc3", "calcId3", 1f); + List anns = seq.getAlignmentAnnotations("calcId2", "label2"); assertEquals(2, anns.size()); @@ -310,9 +336,13 @@ public class SequenceTest assertEquals(1, sfs.length); assertSame(sf, sfs[0]); + /* * SequenceFeature on sequence and dataset sequence; returns that on * sequence + * + * Note JAL-2046: spurious: we have no use case for this at the moment. + * This test also buggy - as sf2.equals(sf), no new feature is added */ SequenceFeature sf2 = new SequenceFeature(); sq.getDatasetSequence().addSequenceFeature(sf2); @@ -322,17 +352,20 @@ public class SequenceTest /* * SequenceFeature on dataset sequence only + * Note JAL-2046: spurious: we have no use case for setting a non-dataset sequence's feature array to null at the moment. */ sq.setSequenceFeatures(null); - sfs = sq.getSequenceFeatures(); - assertEquals(1, sfs.length); - assertSame(sf2, sfs[0]); + assertNull(sq.getDatasetSequence().getSequenceFeatures()); /* * Corrupt case - no SequenceFeature, dataset's dataset is the original * sequence. Test shows no infinite loop results. */ sq.getDatasetSequence().setSequenceFeatures(null); + /** + * is there a usecase for this ? setDatasetSequence should throw an error if + * this actually occurs. + */ sq.getDatasetSequence().setDatasetSequence(sq); // loop! assertNull(sq.getSequenceFeatures()); } @@ -378,6 +411,20 @@ public class SequenceTest } /** + * test createDatasetSequence behaves to doc + */ + @Test(groups = { "Functional" }) + public void testCreateDatasetSequence() + { + SequenceI sq = new Sequence("my","ASDASD"); + assertNull(sq.getDatasetSequence()); + SequenceI rds = sq.createDatasetSequence(); + assertNotNull(rds); + assertNull(rds.getDatasetSequence()); + assertEquals(sq.getDatasetSequence(), rds); + } + + /** * Test for deriveSequence applied to a sequence with a dataset */ @Test(groups = { "Functional" }) @@ -390,7 +437,91 @@ public class SequenceTest sq.setStart(3); sq.setEnd(4); + sq.setDescription("Test sequence description.."); + sq.setVamsasId("TestVamsasId"); + sq.addDBRef(new DBRefEntry("PDB", "version0", "1TST")); + + sq.addDBRef(new DBRefEntry("PDB", "version1", "1PDB")); + sq.addDBRef(new DBRefEntry("PDB", "version2", "2PDB")); + sq.addDBRef(new DBRefEntry("PDB", "version3", "3PDB")); + sq.addDBRef(new DBRefEntry("PDB", "version4", "4PDB")); + + sq.addPDBId(new PDBEntry("1PDB", "A", Type.PDB, "filePath/test1")); + sq.addPDBId(new PDBEntry("1PDB", "B", Type.PDB, "filePath/test1")); + sq.addPDBId(new PDBEntry("2PDB", "A", Type.MMCIF, "filePath/test2")); + sq.addPDBId(new PDBEntry("2PDB", "B", Type.MMCIF, "filePath/test2")); + + DBRefEntry pdb1pdb = new DBRefEntry("PDB", "version1", "1PDB"); + DBRefEntry pdb2pdb = new DBRefEntry("PDB", "version1", "2PDB"); + + + List primRefs = Arrays.asList(new DBRefEntry[] { pdb1pdb, + pdb2pdb }); + + sq.getDatasetSequence().addDBRef(pdb1pdb); + sq.getDatasetSequence().addDBRef(pdb2pdb); + sq.getDatasetSequence().addDBRef( + new DBRefEntry("PDB", "version3", "3PDB")); + sq.getDatasetSequence().addDBRef( + new DBRefEntry("PDB", "version4", "4PDB")); + + PDBEntry pdbe1a=new PDBEntry("1PDB", "A", Type.PDB, "filePath/test1"); + PDBEntry pdbe1b = new PDBEntry("1PDB", "B", Type.PDB, "filePath/test1"); + PDBEntry pdbe2a=new PDBEntry("2PDB", "A", Type.MMCIF, "filePath/test2"); + PDBEntry pdbe2b = new PDBEntry("2PDB", "B", Type.MMCIF, "filePath/test2"); + sq.getDatasetSequence().addPDBId( + pdbe1a); + sq.getDatasetSequence().addPDBId( + pdbe1b); + sq.getDatasetSequence().addPDBId(pdbe2a); + sq.getDatasetSequence().addPDBId(pdbe2b); + + /* + * test we added pdb entries to the dataset sequence + */ + Assert.assertEquals(sq.getDatasetSequence().getAllPDBEntries(), Arrays + .asList(new PDBEntry[] { pdbe1a, pdbe1b, pdbe2a, pdbe2b }), + "PDB Entries were not found on dataset sequence."); + + /* + * we should recover a pdb entry that is on the dataset sequence via PDBEntry + */ + Assert.assertEquals(pdbe1a, + sq.getDatasetSequence().getPDBEntry("1PDB"), + "PDB Entry '1PDB' not found on dataset sequence via getPDBEntry."); + ArrayList annotsList = new ArrayList(); + System.out.println(">>>>>> " + sq.getSequenceAsString().length()); + annotsList.add(new Annotation("A", "A", 'X', 0.1f)); + annotsList.add(new Annotation("A", "A", 'X', 0.1f)); + Annotation[] annots = annotsList.toArray(new Annotation[0]); + sq.addAlignmentAnnotation(new AlignmentAnnotation("Test annot", + "Test annot description", annots)); + sq.getDatasetSequence().addAlignmentAnnotation( + new AlignmentAnnotation("Test annot", "Test annot description", + annots)); + Assert.assertEquals(sq.getDescription(), "Test sequence description.."); + Assert.assertEquals(sq.getDBRefs().length, 5); + Assert.assertEquals(sq.getAllPDBEntries().size(), 4); + Assert.assertNotNull(sq.getAnnotation()); + Assert.assertEquals(sq.getAnnotation()[0].annotations.length, 2); + Assert.assertEquals(sq.getDatasetSequence().getDBRefs().length, 4); + Assert.assertEquals(sq.getDatasetSequence().getAllPDBEntries().size(), + 4); + Assert.assertNotNull(sq.getDatasetSequence().getAnnotation()); + Sequence derived = (Sequence) sq.deriveSequence(); + + Assert.assertEquals(derived.getDescription(), + "Test sequence description.."); + Assert.assertEquals(derived.getDBRefs().length, 4); // come from dataset + Assert.assertEquals(derived.getAllPDBEntries().size(), 4); + Assert.assertNotNull(derived.getAnnotation()); + Assert.assertEquals(derived.getAnnotation()[0].annotations.length, 2); + Assert.assertEquals(derived.getDatasetSequence().getDBRefs().length, 4); + Assert.assertEquals(derived.getDatasetSequence().getAllPDBEntries() + .size(), 4); + Assert.assertNotNull(derived.getDatasetSequence().getAnnotation()); + assertEquals("CD", derived.getSequenceAsString()); assertSame(sq.getDatasetSequence(), derived.getDatasetSequence()); @@ -400,6 +531,17 @@ public class SequenceTest assertNotNull(sq.getSequenceFeatures()); assertArrayEquals(sq.getSequenceFeatures(), derived.getSequenceFeatures()); + + /* + * verify we have primary db refs *just* for PDB IDs with associated + * PDBEntry objects + */ + + assertEquals(primRefs, sq.getPrimaryDBRefs()); + assertEquals(primRefs, sq.getDatasetSequence().getPrimaryDBRefs()); + + assertEquals(sq.getPrimaryDBRefs(), derived.getPrimaryDBRefs()); + } /** @@ -543,4 +685,203 @@ public class SequenceTest assertEquals(' ', sq.getCharAt(5)); assertEquals(' ', sq.getCharAt(-1)); } + + /** + * Tests for adding (or updating) dbrefs + * + * @see DBRefEntry#updateFrom(DBRefEntry) + */ + @Test(groups = { "Functional" }) + public void testAddDBRef() + { + SequenceI sq = new Sequence("", "abcde"); + assertNull(sq.getDBRefs()); + DBRefEntry dbref = new DBRefEntry("Uniprot", "1", "P00340"); + sq.addDBRef(dbref); + assertEquals(1, sq.getDBRefs().length); + assertSame(dbref, sq.getDBRefs()[0]); + + /* + * change of version - new entry + */ + DBRefEntry dbref2 = new DBRefEntry("Uniprot", "2", "P00340"); + sq.addDBRef(dbref2); + assertEquals(2, sq.getDBRefs().length); + assertSame(dbref, sq.getDBRefs()[0]); + assertSame(dbref2, sq.getDBRefs()[1]); + + /* + * matches existing entry - not added + */ + sq.addDBRef(new DBRefEntry("UNIPROT", "1", "p00340")); + assertEquals(2, sq.getDBRefs().length); + + /* + * different source = new entry + */ + DBRefEntry dbref3 = new DBRefEntry("UniRef", "1", "p00340"); + sq.addDBRef(dbref3); + assertEquals(3, sq.getDBRefs().length); + assertSame(dbref3, sq.getDBRefs()[2]); + + /* + * different ref = new entry + */ + DBRefEntry dbref4 = new DBRefEntry("UniRef", "1", "p00341"); + sq.addDBRef(dbref4); + assertEquals(4, sq.getDBRefs().length); + assertSame(dbref4, sq.getDBRefs()[3]); + + /* + * matching ref with a mapping - map updated + */ + DBRefEntry dbref5 = new DBRefEntry("UniRef", "1", "p00341"); + Mapping map = new Mapping(new MapList(new int[] { 1, 3 }, new int[] { + 1, 1 }, 3, 1)); + dbref5.setMap(map); + sq.addDBRef(dbref5); + assertEquals(4, sq.getDBRefs().length); + assertSame(dbref4, sq.getDBRefs()[3]); + assertSame(map, dbref4.getMap()); + + /* + * 'real' version replaces "0" version + */ + dbref2.setVersion("0"); + DBRefEntry dbref6 = new DBRefEntry(dbref2.getSource(), "3", + dbref2.getAccessionId()); + sq.addDBRef(dbref6); + assertEquals(4, sq.getDBRefs().length); + assertSame(dbref2, sq.getDBRefs()[1]); + assertEquals("3", dbref2.getVersion()); + + /* + * 'real' version replaces "source:0" version + */ + dbref3.setVersion("Uniprot:0"); + DBRefEntry dbref7 = new DBRefEntry(dbref3.getSource(), "3", + dbref3.getAccessionId()); + sq.addDBRef(dbref7); + assertEquals(4, sq.getDBRefs().length); + assertSame(dbref3, sq.getDBRefs()[2]); + assertEquals("3", dbref2.getVersion()); + } + + @Test(groups = { "Functional" }) + public void testGetPrimaryDBRefs_peptide() + { + SequenceI sq = new Sequence("aseq", "ASDFKYLMQPRST", 10, 22); + + // no dbrefs + List primaryDBRefs = sq.getPrimaryDBRefs(); + assertTrue(primaryDBRefs.isEmpty()); + + // empty dbrefs + sq.setDBRefs(new DBRefEntry[] {}); + primaryDBRefs = sq.getPrimaryDBRefs(); + assertTrue(primaryDBRefs.isEmpty()); + + // primary - uniprot + DBRefEntry upentry1 = new DBRefEntry("UNIPROT", "0", "Q04760"); + sq.addDBRef(upentry1); + + // primary - uniprot with congruent map + DBRefEntry upentry2 = new DBRefEntry("UNIPROT", "0", "Q04762"); + upentry2.setMap(new Mapping(null, new MapList(new int[] { 10, 22 }, + new int[] { 10, 22 }, 1, 1))); + sq.addDBRef(upentry2); + + // primary - uniprot with map of enclosing sequence + DBRefEntry upentry3 = new DBRefEntry("UNIPROT", "0", "Q04763"); + upentry3.setMap(new Mapping(null, new MapList(new int[] { 8, 24 }, + new int[] { 8, 24 }, 1, 1))); + sq.addDBRef(upentry3); + + // not primary - uniprot with map of sub-sequence (5') + DBRefEntry upentry4 = new DBRefEntry("UNIPROT", "0", "Q04764"); + upentry4.setMap(new Mapping(null, new MapList(new int[] { 10, 18 }, + new int[] { 10, 18 }, 1, 1))); + sq.addDBRef(upentry4); + + // not primary - uniprot with map that overlaps 3' + DBRefEntry upentry5 = new DBRefEntry("UNIPROT", "0", "Q04765"); + upentry5.setMap(new Mapping(null, new MapList(new int[] { 12, 22 }, + new int[] { 12, 22 }, 1, 1))); + sq.addDBRef(upentry5); + + // not primary - uniprot with map to different coordinates frame + DBRefEntry upentry6 = new DBRefEntry("UNIPROT", "0", "Q04766"); + upentry6.setMap(new Mapping(null, new MapList(new int[] { 12, 18 }, + new int[] { 112, 118 }, 1, 1))); + sq.addDBRef(upentry6); + + // not primary - dbref to 'non-core' database + DBRefEntry upentry7 = new DBRefEntry("Pfam", "0", "PF00903"); + sq.addDBRef(upentry7); + + // primary - type is PDB + DBRefEntry pdbentry = new DBRefEntry("PDB", "0", "1qip"); + sq.addDBRef(pdbentry); + + // not primary - PDBEntry has no file + sq.addDBRef(new DBRefEntry("PDB", "0", "1AAA")); + + // not primary - no PDBEntry + sq.addDBRef(new DBRefEntry("PDB", "0", "1DDD")); + + // add corroborating PDB entry for primary DBref - + // needs to have a file as well as matching ID + // note PDB ID is not treated as case sensitive + sq.addPDBId(new PDBEntry("1QIP", null, Type.PDB, new File("/blah") + .toString())); + + // not valid DBRef - no file.. + sq.addPDBId(new PDBEntry("1AAA", null, null, null)); + + primaryDBRefs = sq.getPrimaryDBRefs(); + assertEquals(4, primaryDBRefs.size()); + assertTrue("Couldn't find simple primary reference (UNIPROT)", + primaryDBRefs.contains(upentry1)); + assertTrue("Couldn't find mapped primary reference (UNIPROT)", + primaryDBRefs.contains(upentry2)); + assertTrue("Couldn't find mapped context reference (UNIPROT)", + primaryDBRefs.contains(upentry3)); + assertTrue("Couldn't find expected PDB primary reference", + primaryDBRefs.contains(pdbentry)); + } + + @Test(groups = { "Functional" }) + public void testGetPrimaryDBRefs_nucleotide() + { + SequenceI sq = new Sequence("aseq", "TGATCACTCGACTAGCATCAGCATA", 10, 34); + + // primary - Ensembl + DBRefEntry dbr1 = new DBRefEntry("ENSEMBL", "0", "ENSG1234"); + sq.addDBRef(dbr1); + + // not primary - Ensembl 'transcript' mapping of sub-sequence + DBRefEntry dbr2 = new DBRefEntry("ENSEMBL", "0", "ENST1234"); + dbr2.setMap(new Mapping(null, new MapList(new int[] { 15, 25 }, + new int[] { 1, 11 }, 1, 1))); + sq.addDBRef(dbr2); + + // primary - EMBL with congruent map + DBRefEntry dbr3 = new DBRefEntry("EMBL", "0", "J1234"); + dbr3.setMap(new Mapping(null, new MapList(new int[] { 10, 34 }, + new int[] { 10, 34 }, 1, 1))); + sq.addDBRef(dbr3); + + // not primary - to non-core database + DBRefEntry dbr4 = new DBRefEntry("CCDS", "0", "J1234"); + sq.addDBRef(dbr4); + + // not primary - to protein + DBRefEntry dbr5 = new DBRefEntry("UNIPROT", "0", "Q87654"); + sq.addDBRef(dbr5); + + List primaryDBRefs = sq.getPrimaryDBRefs(); + assertEquals(2, primaryDBRefs.size()); + assertTrue(primaryDBRefs.contains(dbr1)); + assertTrue(primaryDBRefs.contains(dbr3)); + } }