X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=test%2Fjalview%2Fanalysis%2FAlignmentUtilsTests.java;h=4aed7e71d2a62018ce35196175f4255612204d00;hb=refs%2Fheads%2Freleases%2FRelease_2_10_0_Branch;hp=a8562314f36318d739620df1dade7cafdab95854;hpb=6b5f772a654104d856da1651c3d411981be463df;p=jalview.git diff --git a/test/jalview/analysis/AlignmentUtilsTests.java b/test/jalview/analysis/AlignmentUtilsTests.java index a856231..4aed7e7 100644 --- a/test/jalview/analysis/AlignmentUtilsTests.java +++ b/test/jalview/analysis/AlignmentUtilsTests.java @@ -439,7 +439,8 @@ public class AlignmentUtilsTests SequenceI alignFrom = new Sequence("Seq2", alignModel); alignFrom.createDatasetSequence(); AlignedCodonFrame acf = new AlignedCodonFrame(); - acf.addMap(alignMe.getDatasetSequence(), alignFrom.getDatasetSequence(), map); + acf.addMap(alignMe.getDatasetSequence(), + alignFrom.getDatasetSequence(), map); AlignmentUtils.alignSequenceAs(alignMe, alignFrom, acf, "---", '-', preserveMappedGaps, preserveUnmappedGaps); @@ -977,7 +978,7 @@ public class AlignmentUtilsTests /* * scenario: * dna1 --> [4, 6] [10,12] --> pep1 - * dna2 --> [1, 3] [7, 9] [13,15] --> pep1 + * dna2 --> [1, 3] [7, 9] [13,15] --> pep2 */ SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa"); SequenceI dna2 = new Sequence("dna2", "GGGcccTTTaaaCCC"); @@ -993,30 +994,51 @@ public class AlignmentUtilsTests dna.setDataset(null); /* + * put a variant feature on dna2 base 8 + * - should transfer to cds2 base 5 + */ + dna2.addSequenceFeature(new SequenceFeature("variant", "hgmd", 8, 8, + 0f, null)); + + /* * need a sourceDbRef if we are to construct dbrefs to the CDS - * sequence + * sequence from the dna contig sequences */ DBRefEntry dbref = new DBRefEntry("ENSEMBL", "0", "dna1"); - dna1.getDatasetSequence().setSourceDBRef(dbref); + dna1.getDatasetSequence().addDBRef(dbref); + org.testng.Assert.assertEquals(dbref, dna1.getPrimaryDBRefs().get(0)); dbref = new DBRefEntry("ENSEMBL", "0", "dna2"); - dna2.getDatasetSequence().setSourceDBRef(dbref); + 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); + MapList mapfordna1 = 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); + acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), + mapfordna1); dna.addCodonFrame(acf); - map = new MapList(new int[] { 1, 3, 7, 9, 13, 15 }, new int[] { 1, 3 }, - 3, 1); + MapList mapfordna2 = 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); + acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), + mapfordna2); dna.addCodonFrame(acf); /* + * In this case, mappings originally came from matching Uniprot accessions - so need an xref on dna involving those regions. These are normally constructed from CDS annotation + */ + DBRefEntry dna1xref = new DBRefEntry("UNIPROT", "ENSEMBL", "pep1", + new Mapping(mapfordna1)); + dna1.getDatasetSequence().addDBRef(dna1xref); + DBRefEntry dna2xref = new DBRefEntry("UNIPROT", "ENSEMBL", "pep2", + new Mapping(mapfordna2)); + dna2.getDatasetSequence().addDBRef(dna2xref); + + /* * execute method under test: */ AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] { @@ -1042,11 +1064,12 @@ public class AlignmentUtilsTests * verify CDS has a dbref with mapping to peptide */ assertNotNull(cds1Dss.getDBRefs()); - assertEquals(1, cds1Dss.getDBRefs().length); + assertEquals(2, cds1Dss.getDBRefs().length); dbref = cds1Dss.getDBRefs()[0]; - assertEquals("UNIPROT", dbref.getSource()); - assertEquals("0", dbref.getVersion()); - assertEquals("pep1", dbref.getAccessionId()); + assertEquals(dna1xref.getSource(), dbref.getSource()); + // version is via ensembl's primary ref + assertEquals(dna1xref.getVersion(), dbref.getVersion()); + assertEquals(dna1xref.getAccessionId(), dbref.getAccessionId()); assertNotNull(dbref.getMap()); assertSame(pep1.getDatasetSequence(), dbref.getMap().getTo()); MapList cdsMapping = new MapList(new int[] { 1, 6 }, @@ -1057,6 +1080,7 @@ public class AlignmentUtilsTests * verify peptide has added a dbref with reverse mapping to CDS */ assertNotNull(pep1.getDBRefs()); + // FIXME pep1.getDBRefs() is 1 - is that the correct behaviour ? assertEquals(2, pep1.getDBRefs().length); dbref = pep1.getDBRefs()[1]; assertEquals("ENSEMBL", dbref.getSource()); @@ -1137,6 +1161,16 @@ public class AlignmentUtilsTests assertSame(cds2Dss, m.getSequence()); assertEquals(7, m.getStart()); assertEquals(9, m.getEnd()); + + /* + * check cds2 acquired a variant feature in position 5 + */ + SequenceFeature[] sfs = cds2Dss.getSequenceFeatures(); + assertNotNull(sfs); + assertEquals(1, sfs.length); + assertEquals("variant", sfs[0].type); + assertEquals(5, sfs[0].begin); + assertEquals(5, sfs[0].end); } /** @@ -1276,8 +1310,7 @@ public class AlignmentUtilsTests .findMappingsForSequence(cds.get(0), dnaMappings); Mapping mapping = dnaToCds1Mappings.get(0).getMappings().get(0) .getMapping(); - assertSame(cds.get(0).getDatasetSequence(), mapping - .getTo()); + assertSame(cds.get(0).getDatasetSequence(), mapping.getTo()); assertEquals("G(1) in CDS should map to G(4) in DNA", 4, mapping .getMap().getToPosition(1)); @@ -1345,8 +1378,7 @@ public class AlignmentUtilsTests * @throws IOException */ @Test(groups = { "Functional" }) - public void testMapCdnaToProtein_forSubsequence() - throws IOException + public void testMapCdnaToProtein_forSubsequence() throws IOException { SequenceI prot = new Sequence("UNIPROT|V12345", "E-I--Q", 10, 12); prot.createDatasetSequence(); @@ -1367,7 +1399,7 @@ public class AlignmentUtilsTests @Test(groups = { "Functional" }) public void testAlignSequenceAs_mappedProteinProtein() { - + SequenceI alignMe = new Sequence("Match", "MGAASEV"); alignMe.createDatasetSequence(); SequenceI alignFrom = new Sequence("Query", "LQTGYMGAASEVMFSPTRR"); @@ -1378,7 +1410,7 @@ public class AlignmentUtilsTests MapList map = new MapList(new int[] { 6, 12 }, new int[] { 1, 7 }, 1, 1); acf.addMap(alignFrom.getDatasetSequence(), alignMe.getDatasetSequence(), map); - + AlignmentUtils.alignSequenceAs(alignMe, alignFrom, acf, "-", '-', true, true); assertEquals("-----MGAASEV-------", alignMe.getSequenceAsString()); @@ -1393,7 +1425,7 @@ public class AlignmentUtilsTests { // map first 3 codons to KPF; G is a trailing unmapped residue MapList map = new MapList(new int[] { 1, 9 }, new int[] { 1, 3 }, 3, 1); - + checkAlignSequenceAs("AAACCCTTT", "K-PFG", true, true, map, "AAA---CCCTTT---"); } @@ -1492,7 +1524,7 @@ public class AlignmentUtilsTests MapList map = new MapList(new int[] { 4, 6, 10, 12 }, new int[] { 1, 6 }, 1, 1); - + // [5, 11] maps to [2, 5] dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f, null)); @@ -1502,12 +1534,12 @@ public class AlignmentUtilsTests // [12, 12] maps to [6, 6] dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12, 8f, null)); - + // desc4 and desc8 are the 'omit these' varargs AlignmentUtils.transferFeatures(dna, cds, map, null, "type4", "type8"); SequenceFeature[] sfs = cds.getSequenceFeatures(); assertEquals(1, sfs.length); - + SequenceFeature sf = sfs[0]; assertEquals("type5", sf.getType()); assertEquals(1, sf.getBegin()); @@ -1522,10 +1554,10 @@ public class AlignmentUtilsTests { SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt"); SequenceI cds = new Sequence("cds/10-15", "TAGGCC"); - + MapList map = new MapList(new int[] { 4, 6, 10, 12 }, new int[] { 1, 6 }, 1, 1); - + // [5, 11] maps to [2, 5] dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f, null)); @@ -1535,12 +1567,12 @@ public class AlignmentUtilsTests // [12, 12] maps to [6, 6] dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12, 8f, null)); - + // "type5" is the 'select this type' argument AlignmentUtils.transferFeatures(dna, cds, map, "type5"); SequenceFeature[] sfs = cds.getSequenceFeatures(); assertEquals(1, sfs.length); - + SequenceFeature sf = sfs[0]; assertEquals("type5", sf.getType()); assertEquals(1, sf.getBegin()); @@ -1569,26 +1601,25 @@ public class AlignmentUtilsTests AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 }); dna.setDataset(null); - + 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); dna.addCodonFrame(acf); map = new MapList(new int[] { 4, 8, 12, 12, 16, 18 }, - new int[] { 1, 3 }, - 3, 1); + new int[] { 1, 3 }, 3, 1); acf = new AlignedCodonFrame(); acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map); dna.addCodonFrame(acf); - + AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] { dna1, dna2, dna3 }, dna.getDataset(), null); List cdsSeqs = cds.getSequences(); assertEquals(2, cdsSeqs.size()); assertEquals("GGGCCCTTTGGG", cdsSeqs.get(0).getSequenceAsString()); assertEquals("GGGCCTGGG", cdsSeqs.get(1).getSequenceAsString()); - + /* * verify shared, extended alignment dataset */ @@ -1604,7 +1635,7 @@ public class AlignmentUtilsTests */ List mappings = cds.getCodonFrames(); assertEquals(6, mappings.size()); - + /* * 2 mappings involve pep1 */ @@ -1623,8 +1654,7 @@ public class AlignmentUtilsTests pep1CdsMappings); assertEquals(1, sr.getResults().size()); Match m = sr.getResults().get(0); - assertEquals(cds.getSequenceAt(0).getDatasetSequence(), - m.getSequence()); + assertEquals(cds.getSequenceAt(0).getDatasetSequence(), m.getSequence()); assertEquals(1, m.getStart()); assertEquals(3, m.getEnd()); sr = MappingUtils.buildSearchResults(pep1, 2, pep1CdsMappings); @@ -1639,7 +1669,7 @@ public class AlignmentUtilsTests m = sr.getResults().get(0); assertEquals(10, m.getStart()); assertEquals(12, m.getEnd()); - + /* * Get mapping of pep2 to cds2 and verify it * maps GPG in pep2 to 1-3,4-6,7-9 in second CDS sequence @@ -1653,8 +1683,7 @@ public class AlignmentUtilsTests sr = MappingUtils.buildSearchResults(pep2, 1, pep2CdsMappings); assertEquals(1, sr.getResults().size()); m = sr.getResults().get(0); - assertEquals(cds.getSequenceAt(1).getDatasetSequence(), - m.getSequence()); + assertEquals(cds.getSequenceAt(1).getDatasetSequence(), m.getSequence()); assertEquals(1, m.getStart()); assertEquals(3, m.getEnd()); sr = MappingUtils.buildSearchResults(pep2, 2, pep2CdsMappings); @@ -1681,7 +1710,7 @@ public class AlignmentUtilsTests SequenceI dna3 = new Sequence("Seq3", "ccaaa-ttt-GGG-"); AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 }); dna.setDataset(null); - + // prot1 has 'X' for incomplete start codon (not mapped) SequenceI prot1 = new Sequence("Seq1", "XKFG"); // X for incomplete start SequenceI prot2 = new Sequence("Seq2", "NG"); @@ -1689,7 +1718,7 @@ public class AlignmentUtilsTests AlignmentI protein = new Alignment(new SequenceI[] { prot1, prot2, prot3 }); protein.setDataset(null); - + // map dna1 [3, 11] to prot1 [2, 4] KFG MapList map = new MapList(new int[] { 3, 11 }, new int[] { 2, 4 }, 3, 1); AlignedCodonFrame acf = new AlignedCodonFrame(); @@ -1728,7 +1757,7 @@ public class AlignmentUtilsTests SequenceI dnaSeq = new Sequence("dna", "aaagGGCCCaaaTTTttt"); dnaSeq.createDatasetSequence(); SequenceI ds = dnaSeq.getDatasetSequence(); - + // CDS for dna 5-6 (incomplete codon), 7-9 SequenceFeature sf = new SequenceFeature("CDS", "", 5, 9, 0f, null); sf.setPhase("2"); // skip 2 bases to start of next codon @@ -1736,9 +1765,9 @@ public class AlignmentUtilsTests // CDS for dna 13-15 sf = new SequenceFeature("CDS_predicted", "", 13, 15, 0f, null); ds.addSequenceFeature(sf); - + List ranges = AlignmentUtils.findCdsPositions(dnaSeq); - + /* * check the mapping starts with the first complete codon */ @@ -1760,7 +1789,7 @@ public class AlignmentUtilsTests SequenceI dnaSeq = new Sequence("dna", "aaaGGGcccAAATTTttt"); dnaSeq.createDatasetSequence(); SequenceI ds = dnaSeq.getDatasetSequence(); - + // CDS for dna 10-12 SequenceFeature sf = new SequenceFeature("CDS_predicted", "", 10, 12, 0f, null); @@ -1773,7 +1802,7 @@ public class AlignmentUtilsTests // exon feature should be ignored here sf = new SequenceFeature("exon", "", 7, 9, 0f, null); ds.addSequenceFeature(sf); - + List ranges = AlignmentUtils.findCdsPositions(dnaSeq); /* * verify ranges { [4-6], [12-10] } @@ -2118,7 +2147,7 @@ public class AlignmentUtilsTests SequenceI dnaSeq = new Sequence("dna", "aaaGGGcccAAATTTttt"); dnaSeq.createDatasetSequence(); SequenceI ds = dnaSeq.getDatasetSequence(); - + // CDS for dna 4-6 SequenceFeature sf = new SequenceFeature("CDS", "", 4, 6, 0f, null); sf.setStrand("-"); @@ -2130,7 +2159,7 @@ public class AlignmentUtilsTests sf = new SequenceFeature("CDS_predicted", "", 10, 12, 0f, null); sf.setStrand("-"); ds.addSequenceFeature(sf); - + List ranges = AlignmentUtils.findCdsPositions(dnaSeq); /* * verify ranges { [12-10], [6-4] } @@ -2156,7 +2185,7 @@ public class AlignmentUtilsTests SequenceI dnaSeq = new Sequence("dna", "aaagGGCCCaaaTTTttt"); dnaSeq.createDatasetSequence(); SequenceI ds = dnaSeq.getDatasetSequence(); - + // CDS for dna 5-9 SequenceFeature sf = new SequenceFeature("CDS", "", 5, 9, 0f, null); sf.setStrand("-"); @@ -2166,9 +2195,9 @@ public class AlignmentUtilsTests sf.setStrand("-"); sf.setPhase("2"); // skip 2 bases to start of next codon ds.addSequenceFeature(sf); - + List ranges = AlignmentUtils.findCdsPositions(dnaSeq); - + /* * check the mapping starts with the first complete codon * expect ranges [13, 13], [9, 5] @@ -2221,8 +2250,7 @@ public class AlignmentUtilsTests from.createDatasetSequence(); seq1.createDatasetSequence(); Mapping mapping = new Mapping(seq1, new MapList( - new int[] { 3, 6, 9, 10 }, - new int[] { 1, 6 }, 1, 1)); + new int[] { 3, 6, 9, 10 }, new int[] { 1, 6 }, 1, 1)); Map> map = new TreeMap>(); AlignmentUtils.addMappedPositions(seq1, from, mapping, map); @@ -2254,11 +2282,10 @@ public class AlignmentUtilsTests from.createDatasetSequence(); seq1.createDatasetSequence(); Mapping mapping = new Mapping(seq1, new MapList( - new int[] { 3, 6, 9, 10 }, - new int[] { 1, 6 }, 1, 1)); + new int[] { 3, 6, 9, 10 }, new int[] { 1, 6 }, 1, 1)); Map> map = new TreeMap>(); AlignmentUtils.addMappedPositions(seq1, from, mapping, map); - + /* * verify map has seq1 residues in columns 3,4,6,7,11,12 */ @@ -2296,7 +2323,7 @@ public class AlignmentUtilsTests 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); @@ -2310,17 +2337,17 @@ public class AlignmentUtilsTests 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 */ @@ -2329,7 +2356,7 @@ public class AlignmentUtilsTests .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 @@ -2339,7 +2366,7 @@ public class AlignmentUtilsTests * 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] @@ -2348,7 +2375,7 @@ public class AlignmentUtilsTests // 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 @@ -2359,7 +2386,7 @@ public class AlignmentUtilsTests List 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()); @@ -2373,7 +2400,7 @@ public class AlignmentUtilsTests 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 @@ -2427,7 +2454,7 @@ public class AlignmentUtilsTests 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()); @@ -2484,5 +2511,5 @@ public class AlignmentUtilsTests assertEquals(s_as2, uas2.getSequenceAsString()); assertEquals(s_as3, uas3.getSequenceAsString()); } - + }