From c14d9e8c9ff59de1857d4834ee70e80abf623415 Mon Sep 17 00:00:00 2001 From: gmungoc Date: Wed, 6 Jul 2016 10:10:53 +0100 Subject: [PATCH] JAL-2110 fixes and tests for adding dbrefs to EMBL sequence and products --- src/jalview/datamodel/xdb/embl/EmblEntry.java | 72 +++++++--- test/jalview/datamodel/xdb/embl/EmblEntryTest.java | 152 ++++++++++++++------ 2 files changed, 160 insertions(+), 64 deletions(-) diff --git a/src/jalview/datamodel/xdb/embl/EmblEntry.java b/src/jalview/datamodel/xdb/embl/EmblEntry.java index 67d7059..56b1325 100644 --- a/src/jalview/datamodel/xdb/embl/EmblEntry.java +++ b/src/jalview/datamodel/xdb/embl/EmblEntry.java @@ -187,12 +187,12 @@ public class EmblEntry */ public SequenceI getSequence(String sourceDb, List peptides) { - SequenceI dna = new Sequence(sourceDb + "|" + accession, - sequence.getSequence()); + SequenceI dna = makeSequence(sourceDb); dna.setDescription(description); DBRefEntry retrievedref = new DBRefEntry(sourceDb, getSequenceVersion(), accession); dna.addDBRef(retrievedref); + dna.setSourceDBRef(retrievedref); // add map to indicate the sequence is a valid coordinate frame for the // dbref retrievedref.setMap(new Mapping(null, new int[] { 1, dna.getLength() }, @@ -235,6 +235,17 @@ public class EmblEntry } /** + * @param sourceDb + * @return + */ + SequenceI makeSequence(String sourceDb) + { + SequenceI dna = new Sequence(sourceDb + "|" + accession, + sequence.getSequence()); + return dna; + } + + /** * Extracts coding region and product from a CDS feature and properly decorate * it with annotations. * @@ -322,13 +333,15 @@ public class EmblEntry Mapping dnaToProteinMapping = null; if (translation != null && proteinName != null && proteinId != null) { + int translationLength = translation.length(); + /* * look for product in peptides list, if not found, add it */ product = matcher.findIdMatch(proteinId); if (product == null) { - product = new Sequence(proteinId, translation, 1, translation.length()); + product = new Sequence(proteinId, translation, 1, translationLength); product.setDescription(((proteinName.length() == 0) ? "Protein Product from " + sourceDb : proteinName)); @@ -340,30 +353,32 @@ public class EmblEntry // sequence if (exons == null || exons.length == 0) { + /* + * workaround until we handle dna location for CDS sequence + * e.g. location="X53828.1:60..1058" correctly + */ System.err .println("Implementation Notice: EMBLCDS records not properly supported yet - Making up the CDNA region of this sequence... may be incorrect (" + sourceDb + ":" + getAccession() + ")"); - if (translation.length() * 3 == (1 - codonStart + dna.getSequence().length)) + if (translationLength * 3 == (1 - codonStart + dna.getSequence().length)) { System.err .println("Not allowing for additional stop codon at end of cDNA fragment... !"); - // this might occur for CDS sequences where no features are - // marked. + // this might occur for CDS sequences where no features are marked exons = new int[] { dna.getStart() + (codonStart - 1), dna.getEnd() }; dnaToProteinMapping = new Mapping(product, exons, new int[] { 1, - translation.length() }, - 3, 1); + translationLength }, 3, 1); } - if ((translation.length() + 1) * 3 == (1 - codonStart + dna.getSequence().length)) + if ((translationLength + 1) * 3 == (1 - codonStart + dna + .getSequence().length)) { System.err .println("Allowing for additional stop codon at end of cDNA fragment... will probably cause an error in VAMSAs!"); exons = new int[] { dna.getStart() + (codonStart - 1), dna.getEnd() - 3 }; dnaToProteinMapping = new Mapping(product, exons, new int[] { 1, - translation.length() }, - 3, 1); + translationLength }, 3, 1); } } else @@ -382,28 +397,37 @@ public class EmblEntry else { // final product length truncation check - int[] cdsRanges = adjustForProteinLength(translation.length(), exons); - dnaToProteinMapping = new Mapping(product, cdsRanges, new int[] { 1, - translation.length() }, 3, 1); + int[] cdsRanges = adjustForProteinLength(translationLength, exons); + dnaToProteinMapping = new Mapping(product, cdsRanges, new int[] { + 1, translationLength }, 3, 1); if (product != null) { /* + * make xref with mapping from protein to EMBL dna + */ + DBRefEntry proteinToEmblRef = new DBRefEntry(DBRefSource.EMBL, + getSequenceVersion(), proteinId, new Mapping( + dnaToProteinMapping.getMap().getInverse())); + product.addDBRef(proteinToEmblRef); + + /* * make xref from protein to EMBLCDS; we assume here that the * CDS sequence version is same as dna sequence (?!) */ MapList proteinToCdsMapList = new MapList(new int[] { 1, - translation.length() }, new int[] { 1 + (codonStart - 1), - (codonStart - 1) + 3 * translation.length() }, 1, 3); + translationLength }, new int[] { 1 + (codonStart - 1), + (codonStart - 1) + 3 * translationLength }, 1, 3); DBRefEntry proteinToEmblCdsRef = new DBRefEntry( DBRefSource.EMBLCDS, getSequenceVersion(), proteinId, new Mapping(proteinToCdsMapList)); product.addDBRef(proteinToEmblCdsRef); /* - * make xref from protein to EMBLCDSPROTEIN + * make 'direct' xref from protein to EMBLCDSPROTEIN */ proteinToEmblProteinRef = new DBRefEntry(proteinToEmblCdsRef); proteinToEmblProteinRef.setSource(DBRefSource.EMBLCDSProduct); + proteinToEmblProteinRef.setMap(null); product.addDBRef(proteinToEmblProteinRef); } } @@ -414,8 +438,8 @@ public class EmblEntry */ for (int xint = 0; exons != null && xint < exons.length; xint += 2) { - SequenceFeature sf = makeCdsFeature(exons, xint, proteinName, proteinId, vals, - codonStart); + SequenceFeature sf = makeCdsFeature(exons, xint, proteinName, + proteinId, vals, codonStart); sf.setType(feature.getName()); // "CDS" sf.setEnaLocation(feature.getLocation()); sf.setFeatureGroup(sourceDb); @@ -437,7 +461,7 @@ public class EmblEntry */ String source = DBRefUtils.getCanonicalName(ref.getSource()); ref.setSource(source); - DBRefEntry proteinToDnaRef = new DBRefEntry(ref.getSource(), ref.getVersion(), ref + DBRefEntry proteinDbRef = new DBRefEntry(ref.getSource(), ref.getVersion(), ref .getAccessionId()); if (source.equals(DBRefSource.UNIPROT)) { @@ -469,7 +493,9 @@ public class EmblEntry peptides.add(proteinSeq); } dnaToProteinMapping.setTo(proteinSeq); - proteinSeq.addDBRef(proteinToDnaRef); + dnaToProteinMapping.setMappedFromId(proteinId); + proteinSeq.addDBRef(proteinDbRef); + proteinSeq.setSourceDBRef(proteinDbRef); ref.setMap(dnaToProteinMapping); } hasUniprotDbref = true; @@ -479,7 +505,7 @@ public class EmblEntry /* * copy feature dbref to our protein product */ - DBRefEntry pref = proteinToDnaRef; + DBRefEntry pref = proteinDbRef; pref.setMap(null); // reference is direct product.addDBRef(pref); // Add converse mapping reference @@ -514,6 +540,7 @@ public class EmblEntry DBRefSource.EMBLCDSProduct, getSequenceVersion(), proteinId); } product.addDBRef(proteinToEmblProteinRef); + product.setSourceDBRef(proteinToEmblProteinRef); if (dnaToProteinMapping != null && dnaToProteinMapping.getTo() != null) @@ -521,6 +548,7 @@ public class EmblEntry DBRefEntry dnaToEmblProteinRef = new DBRefEntry( DBRefSource.EMBLCDSProduct, getSequenceVersion(), proteinId); dnaToEmblProteinRef.setMap(dnaToProteinMapping); + dnaToProteinMapping.setMappedFromId(proteinId); dna.addDBRef(dnaToEmblProteinRef); } } diff --git a/test/jalview/datamodel/xdb/embl/EmblEntryTest.java b/test/jalview/datamodel/xdb/embl/EmblEntryTest.java index d987e53..4b71417 100644 --- a/test/jalview/datamodel/xdb/embl/EmblEntryTest.java +++ b/test/jalview/datamodel/xdb/embl/EmblEntryTest.java @@ -1,12 +1,14 @@ package jalview.datamodel.xdb.embl; import static org.testng.AssertJUnit.assertEquals; +import static org.testng.AssertJUnit.assertNull; import static org.testng.AssertJUnit.assertSame; import jalview.analysis.SequenceIdMatcher; import jalview.datamodel.DBRefEntry; -import jalview.datamodel.Sequence; +import jalview.datamodel.DBRefSource; import jalview.datamodel.SequenceI; +import jalview.util.MapList; import java.util.ArrayList; import java.util.Arrays; @@ -22,7 +24,7 @@ public class EmblEntryTest EmblEntry testee = new EmblEntry(); /* - * Make a (CDS) Feature with 4 locations + * Make a (CDS) Feature with 5 locations */ EmblFeature cds = new EmblFeature(); cds.setLocation("join(10..20,complement(30..40),50..60,70..80,complement(110..120))"); @@ -36,20 +38,22 @@ public class EmblEntryTest public void testParseCodingFeature() { // not the whole sequence but enough for this test... - SequenceI dna = new Sequence("J03321", "GGATCCGTAAGTTAGACGAAATT"); List peptides = new ArrayList(); SequenceIdMatcher matcher = new SequenceIdMatcher(peptides); EmblFile ef = EmblTestHelper.getEmblFile(); + assertEquals(1, ef.getEntries().size()); + EmblEntry testee = ef.getEntries().get(0); + String sourceDb = "EMBL"; + SequenceI dna = testee.makeSequence(sourceDb); /* * parse three CDS features, with two/one/no Uniprot cross-refs */ - EmblEntry testee = new EmblEntry(); for (EmblFeature feature : ef.getEntries().get(0).getFeatures()) { if ("CDS".equals(feature.getName())) { - testee.parseCodingFeature(feature, "EMBL", dna, peptides, matcher); + testee.parseCodingFeature(feature, sourceDb, dna, peptides, matcher); } } @@ -57,7 +61,7 @@ public class EmblEntryTest * peptides should now have five entries: * EMBL product and two Uniprot accessions for the first CDS / translation * EMBL product and one Uniprot accession for the second CDS / " - * EMBL product and synthesized EMBLCDSPROTEINaccession for the third + * EMBL product only for the third */ assertEquals(6, peptides.size()); assertEquals("CAA30420.1", peptides.get(0).getName()); @@ -74,63 +78,128 @@ public class EmblEntryTest assertEquals("MSS", peptides.get(5).getSequenceAsString()); /* - * verify dna sequence has dbrefs with mappings to the peptide 'products' + * verify dna sequence has dbrefs with CDS mappings to the peptide 'products' */ + MapList cds1Map = new MapList(new int[] { 57, 46 }, new int[] { 1, 4 }, + 3, 1); + MapList cds2Map = new MapList(new int[] { 4, 15 }, new int[] { 1, 4 }, + 3, 1); + MapList cds3Map = new MapList(new int[] { 4, 6, 10, 15 }, new int[] { + 1, 3 }, 3, 1); DBRefEntry[] dbrefs = dna.getDBRefs(); assertEquals(4, dbrefs.length); DBRefEntry dbRefEntry = dbrefs[0]; assertEquals("UNIPROT", dbRefEntry.getSource()); assertEquals("B0BCM4", dbRefEntry.getAccessionId()); assertSame(peptides.get(1), dbRefEntry.getMap().getTo()); - List fromRanges = dbRefEntry.getMap().getMap().getFromRanges(); - assertEquals(1, fromRanges.size()); - assertEquals(57, fromRanges.get(0)[0]); - assertEquals(46, fromRanges.get(0)[1]); - List toRanges = dbRefEntry.getMap().getMap().getToRanges(); - assertEquals(1, toRanges.size()); - assertEquals(1, toRanges.get(0)[0]); - assertEquals(4, toRanges.get(0)[1]); + assertEquals(cds1Map, dbRefEntry.getMap().getMap()); dbRefEntry = dbrefs[1]; assertEquals("UNIPROT", dbRefEntry.getSource()); assertEquals("P0CE20", dbRefEntry.getAccessionId()); assertSame(peptides.get(2), dbRefEntry.getMap().getTo()); - fromRanges = dbRefEntry.getMap().getMap().getFromRanges(); - assertEquals(1, fromRanges.size()); - assertEquals(57, fromRanges.get(0)[0]); - assertEquals(46, fromRanges.get(0)[1]); - toRanges = dbRefEntry.getMap().getMap().getToRanges(); - assertEquals(1, toRanges.size()); - assertEquals(1, toRanges.get(0)[0]); - assertEquals(4, toRanges.get(0)[1]); + assertEquals(cds1Map, dbRefEntry.getMap().getMap()); dbRefEntry = dbrefs[2]; assertEquals("UNIPROT", dbRefEntry.getSource()); assertEquals("B0BCM3", dbRefEntry.getAccessionId()); assertSame(peptides.get(4), dbRefEntry.getMap().getTo()); - fromRanges = dbRefEntry.getMap().getMap().getFromRanges(); - assertEquals(1, fromRanges.size()); - assertEquals(4, fromRanges.get(0)[0]); - assertEquals(15, fromRanges.get(0)[1]); - toRanges = dbRefEntry.getMap().getMap().getToRanges(); - assertEquals(1, toRanges.size()); - assertEquals(1, toRanges.get(0)[0]); - assertEquals(4, toRanges.get(0)[1]); + assertEquals(cds2Map, dbRefEntry.getMap().getMap()); dbRefEntry = dbrefs[3]; assertEquals("EMBLCDSPROTEIN", dbRefEntry.getSource()); assertEquals("CAA12345.6", dbRefEntry.getAccessionId()); assertSame(peptides.get(5), dbRefEntry.getMap().getTo()); - fromRanges = dbRefEntry.getMap().getMap().getFromRanges(); - assertEquals(2, fromRanges.size()); - assertEquals(4, fromRanges.get(0)[0]); - assertEquals(6, fromRanges.get(0)[1]); - assertEquals(10, fromRanges.get(1)[0]); - assertEquals(15, fromRanges.get(1)[1]); - toRanges = dbRefEntry.getMap().getMap().getToRanges(); - assertEquals(1, toRanges.size()); - assertEquals(1, toRanges.get(0)[0]); - assertEquals(3, toRanges.get(0)[1]); + assertEquals(cds3Map, dbRefEntry.getMap().getMap()); + + /* + * verify peptides have dbrefs + * - to EMBL sequence (with inverse 1:3 cds mapping) + * - to EMBLCDS (with 1:3 mapping) + * - direct (no mapping) to other protein accessions + */ + MapList proteinToCdsMap1 = new MapList(new int[] { 1, 4 }, new int[] { + 1, 12 }, 1, 3); + MapList proteinToCdsMap2 = new MapList(new int[] { 1, 3 }, new int[] { + 1, 9 }, 1, 3); + + // dbrefs for first CDS EMBL product CAA30420.1 + dbrefs = peptides.get(0).getDBRefs(); + assertEquals(5, dbrefs.length); + assertEquals(DBRefSource.EMBL, dbrefs[0].getSource()); + assertEquals("CAA30420.1", dbrefs[0].getAccessionId()); + assertEquals(cds1Map.getInverse(), dbrefs[0].getMap().getMap()); + assertEquals(DBRefSource.EMBLCDS, dbrefs[1].getSource()); + assertEquals("CAA30420.1", dbrefs[1].getAccessionId()); + assertEquals(proteinToCdsMap1, dbrefs[1].getMap().getMap()); + assertEquals(DBRefSource.EMBLCDSProduct, dbrefs[2].getSource()); + assertEquals("CAA30420.1", dbrefs[2].getAccessionId()); + assertNull(dbrefs[2].getMap()); + assertEquals(new DBRefEntry(DBRefSource.UNIPROT, "2.1", "B0BCM4"), + dbrefs[3]); + assertNull(dbrefs[3].getMap()); + assertEquals(new DBRefEntry(DBRefSource.UNIPROT, "0", "P0CE20"), + dbrefs[4]); + assertNull(dbrefs[4].getMap()); + + // dbrefs for first CDS first Uniprot xref + dbrefs = peptides.get(1).getDBRefs(); + assertEquals(2, dbrefs.length); + assertEquals(new DBRefEntry(DBRefSource.UNIPROT, "2.1", "B0BCM4"), + dbrefs[0]); + assertNull(dbrefs[0].getMap()); + assertEquals(DBRefSource.EMBL, dbrefs[1].getSource()); + assertEquals("X07547", dbrefs[1].getAccessionId()); + assertEquals(cds1Map.getInverse(), dbrefs[1].getMap().getMap()); + + // dbrefs for first CDS second Uniprot xref + dbrefs = peptides.get(2).getDBRefs(); + assertEquals(2, dbrefs.length); + assertEquals(new DBRefEntry(DBRefSource.UNIPROT, "0", "P0CE20"), + dbrefs[0]); + assertNull(dbrefs[0].getMap()); + assertEquals(DBRefSource.EMBL, dbrefs[1].getSource()); + assertEquals("X07547", dbrefs[1].getAccessionId()); + assertEquals(cds1Map.getInverse(), dbrefs[1].getMap().getMap()); + + // dbrefs for second CDS EMBL product CAA30421.1 + dbrefs = peptides.get(3).getDBRefs(); + assertEquals(4, dbrefs.length); + assertEquals(DBRefSource.EMBL, dbrefs[0].getSource()); + assertEquals("CAA30421.1", dbrefs[0].getAccessionId()); + assertEquals(cds2Map.getInverse(), dbrefs[0].getMap().getMap()); + assertEquals(DBRefSource.EMBLCDS, dbrefs[1].getSource()); + assertEquals("CAA30421.1", dbrefs[1].getAccessionId()); + assertEquals(proteinToCdsMap1, dbrefs[1].getMap().getMap()); + assertEquals(DBRefSource.EMBLCDSProduct, dbrefs[2].getSource()); + assertEquals("CAA30421.1", dbrefs[2].getAccessionId()); + assertNull(dbrefs[2].getMap()); + assertEquals(new DBRefEntry(DBRefSource.UNIPROT, "0", "B0BCM3"), + dbrefs[3]); + assertNull(dbrefs[3].getMap()); + + // dbrefs for second CDS second Uniprot xref + dbrefs = peptides.get(4).getDBRefs(); + assertEquals(2, dbrefs.length); + assertEquals(new DBRefEntry(DBRefSource.UNIPROT, "0", "B0BCM3"), + dbrefs[0]); + assertNull(dbrefs[0].getMap()); + assertEquals(DBRefSource.EMBL, dbrefs[1].getSource()); + assertEquals("X07547", dbrefs[1].getAccessionId()); + assertEquals(cds2Map.getInverse(), dbrefs[1].getMap().getMap()); + + // dbrefs for third CDS inferred EMBL product CAA12345.6 + dbrefs = peptides.get(5).getDBRefs(); + assertEquals(3, dbrefs.length); + assertEquals(DBRefSource.EMBL, dbrefs[0].getSource()); + assertEquals("CAA12345.6", dbrefs[0].getAccessionId()); + assertEquals(cds3Map.getInverse(), dbrefs[0].getMap().getMap()); + assertEquals(DBRefSource.EMBLCDS, dbrefs[1].getSource()); + assertEquals("CAA12345.6", dbrefs[1].getAccessionId()); + assertEquals(proteinToCdsMap2, dbrefs[1].getMap().getMap()); + assertEquals(DBRefSource.EMBLCDSProduct, dbrefs[2].getSource()); + assertEquals("CAA12345.6", dbrefs[2].getAccessionId()); + assertNull(dbrefs[2].getMap()); } @Test(groups = "Functional") @@ -161,6 +230,5 @@ public class EmblEntryTest // what if exons are too short for protein? truncated = EmblEntry.adjustForProteinLength(7, exons); assertSame(exons, truncated); - // assertEquals("[11, 15, 21, 27]", Arrays.toString(truncated)); } } -- 1.7.10.2