X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=test%2Fjalview%2Fanalysis%2FAlignmentUtilsTests.java;h=eccfe500bb5812953584212160cec04f21237c81;hb=ee198b3ca3687f18a2ee186f4e7c7330f4ea30f0;hp=71b1bcb9ed8c0665ced512fc11e4b9d0deae4dce;hpb=07bb0da70ad46444f5f9bc0b2fa80e9ee3805394;p=jalview.git diff --git a/test/jalview/analysis/AlignmentUtilsTests.java b/test/jalview/analysis/AlignmentUtilsTests.java index 71b1bcb..eccfe50 100644 --- a/test/jalview/analysis/AlignmentUtilsTests.java +++ b/test/jalview/analysis/AlignmentUtilsTests.java @@ -20,20 +20,13 @@ */ package jalview.analysis; -import static org.junit.Assert.assertEquals; -import static org.junit.Assert.assertFalse; -import static org.junit.Assert.assertSame; -import static org.junit.Assert.assertTrue; - -import java.io.IOException; -import java.util.ArrayList; -import java.util.Arrays; -import java.util.Collections; -import java.util.List; -import java.util.Map; - -import org.junit.Test; +import static org.testng.AssertJUnit.assertEquals; +import static org.testng.AssertJUnit.assertFalse; +import static org.testng.AssertJUnit.assertNull; +import static org.testng.AssertJUnit.assertSame; +import static org.testng.AssertJUnit.assertTrue; +import jalview.analysis.AlignmentUtils.DnaVariant; import jalview.datamodel.AlignedCodonFrame; import jalview.datamodel.Alignment; import jalview.datamodel.AlignmentAnnotation; @@ -41,85 +34,186 @@ import jalview.datamodel.AlignmentI; 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.SequenceFeature; import jalview.datamodel.SequenceI; import jalview.io.AppletFormatAdapter; +import jalview.io.DataSourceType; +import jalview.io.FileFormat; +import jalview.io.FileFormatI; import jalview.io.FormatAdapter; import jalview.util.MapList; +import jalview.util.MappingUtils; + +import java.io.IOException; +import java.util.ArrayList; +import java.util.Arrays; +import java.util.Iterator; +import java.util.LinkedHashMap; +import java.util.List; +import java.util.Map; +import java.util.TreeMap; + +import org.testng.annotations.Test; -public class AlignmentUtilsTests +public class AlignmentUtilsTests { - // @formatter:off - private static final String TEST_DATA = - "# STOCKHOLM 1.0\n" + - "#=GS D.melanogaster.1 AC AY119185.1/838-902\n" + - "#=GS D.melanogaster.2 AC AC092237.1/57223-57161\n" + - "#=GS D.melanogaster.3 AC AY060611.1/560-627\n" + - "D.melanogaster.1 G.AGCC.CU...AUGAUCGA\n" + - "#=GR D.melanogaster.1 SS ................((((\n" + - "D.melanogaster.2 C.AUUCAACU.UAUGAGGAU\n" + - "#=GR D.melanogaster.2 SS ................((((\n" + - "D.melanogaster.3 G.UGGCGCU..UAUGACGCA\n" + - "#=GR D.melanogaster.3 SS (.(((...(....(((((((\n" + - "//"; - - private static final String AA_SEQS_1 = - ">Seq1Name\n" + - "K-QY--L\n" + - ">Seq2Name\n" + - "-R-FP-W-\n"; - - private static final String CDNA_SEQS_1 = - ">Seq1Name\n" + - "AC-GG--CUC-CAA-CT\n" + - ">Seq2Name\n" + - "-CG-TTA--ACG---AAGT\n"; - - private static final String CDNA_SEQS_2 = - ">Seq1Name\n" + - "GCTCGUCGTACT\n" + - ">Seq2Name\n" + - "GGGTCAGGCAGT\n"; - // @formatter:on - - public static Sequence ts=new Sequence("short","ASDASDASDASDASDASDASDASDASDASDASDASDASD"); - - @Test - public void testExpandFlanks() + public static Sequence ts = new Sequence("short", + "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklm"); + + @Test(groups = { "Functional" }) + public void testExpandContext() { AlignmentI al = new Alignment(new Sequence[] {}); - for (int i=4;i<14;i+=3) + for (int i = 4; i < 14; i += 2) { - SequenceI s1=ts.deriveSequence().getSubSequence(i, i+7); + SequenceI s1 = ts.deriveSequence().getSubSequence(i, i + 7); al.addSequence(s1); } - System.out.println(new AppletFormatAdapter().formatSequences("Clustal", al, true)); - for (int flnk=-1;flnk<25; flnk++) + System.out.println(new AppletFormatAdapter().formatSequences( + FileFormat.Clustal, + al, true)); + for (int flnk = -1; flnk < 25; flnk++) { - AlignmentI exp; - System.out.println("\nFlank size: "+flnk); - System.out.println(new AppletFormatAdapter().formatSequences("Clustal", exp=AlignmentUtils.expandContext(al, flnk), true)); - if (flnk==-1) { - for (SequenceI sq:exp.getSequences()) + AlignmentI exp = AlignmentUtils.expandContext(al, flnk); + System.out.println("\nFlank size: " + flnk); + System.out.println(new AppletFormatAdapter().formatSequences( + FileFormat.Clustal, exp, true)); + if (flnk == -1) { + /* + * Full expansion to complete sequences + */ + for (SequenceI sq : exp.getSequences()) + { String ung = sq.getSequenceAsString().replaceAll("-+", ""); - assertTrue("Flanking sequence not the same as original dataset sequence.\n"+ung+"\n"+sq.getDatasetSequence().getSequenceAsString(),ung.equalsIgnoreCase(sq.getDatasetSequence().getSequenceAsString())); + final String errorMsg = "Flanking sequence not the same as original dataset sequence.\n" + + ung + + "\n" + + sq.getDatasetSequence().getSequenceAsString(); + assertTrue(errorMsg, ung.equalsIgnoreCase(sq.getDatasetSequence() + .getSequenceAsString())); + } } + else if (flnk == 24) + { + /* + * Last sequence is fully expanded, others have leading gaps to match + */ + assertTrue(exp.getSequenceAt(4).getSequenceAsString() + .startsWith("abc")); + assertTrue(exp.getSequenceAt(3).getSequenceAsString() + .startsWith("--abc")); + assertTrue(exp.getSequenceAt(2).getSequenceAsString() + .startsWith("----abc")); + assertTrue(exp.getSequenceAt(1).getSequenceAsString() + .startsWith("------abc")); + assertTrue(exp.getSequenceAt(0).getSequenceAsString() + .startsWith("--------abc")); } } - } + } + + /** + * Test that annotations are correctly adjusted by expandContext + */ + @Test(groups = { "Functional" }) + public void testExpandContext_annotation() + { + AlignmentI al = new Alignment(new Sequence[] {}); + SequenceI ds = new Sequence("Seq1", "ABCDEFGHI"); + // subsequence DEF: + SequenceI seq1 = ds.deriveSequence().getSubSequence(3, 6); + al.addSequence(seq1); + + /* + * Annotate DEF with 4/5/6 respectively + */ + Annotation[] anns = new Annotation[] { new Annotation(4), + new Annotation(5), new Annotation(6) }; + AlignmentAnnotation ann = new AlignmentAnnotation("SS", + "secondary structure", anns); + seq1.addAlignmentAnnotation(ann); + + /* + * The annotations array should match aligned positions + */ + assertEquals(3, ann.annotations.length); + assertEquals(4, ann.annotations[0].value, 0.001); + assertEquals(5, ann.annotations[1].value, 0.001); + assertEquals(6, ann.annotations[2].value, 0.001); + + /* + * Check annotation to sequence position mappings before expanding the + * sequence; these are set up in Sequence.addAlignmentAnnotation -> + * Annotation.setSequenceRef -> createSequenceMappings + */ + assertNull(ann.getAnnotationForPosition(1)); + assertNull(ann.getAnnotationForPosition(2)); + assertNull(ann.getAnnotationForPosition(3)); + assertEquals(4, ann.getAnnotationForPosition(4).value, 0.001); + assertEquals(5, ann.getAnnotationForPosition(5).value, 0.001); + assertEquals(6, ann.getAnnotationForPosition(6).value, 0.001); + assertNull(ann.getAnnotationForPosition(7)); + assertNull(ann.getAnnotationForPosition(8)); + assertNull(ann.getAnnotationForPosition(9)); + + /* + * Expand the subsequence to the full sequence abcDEFghi + */ + AlignmentI expanded = AlignmentUtils.expandContext(al, -1); + assertEquals("abcDEFghi", expanded.getSequenceAt(0) + .getSequenceAsString()); + + /* + * Confirm the alignment and sequence have the same SS annotation, + * referencing the expanded sequence + */ + ann = expanded.getSequenceAt(0).getAnnotation()[0]; + assertSame(ann, expanded.getAlignmentAnnotation()[0]); + assertSame(expanded.getSequenceAt(0), ann.sequenceRef); + + /* + * The annotations array should have null values except for annotated + * positions + */ + assertNull(ann.annotations[0]); + assertNull(ann.annotations[1]); + assertNull(ann.annotations[2]); + assertEquals(4, ann.annotations[3].value, 0.001); + assertEquals(5, ann.annotations[4].value, 0.001); + assertEquals(6, ann.annotations[5].value, 0.001); + assertNull(ann.annotations[6]); + assertNull(ann.annotations[7]); + assertNull(ann.annotations[8]); + + /* + * sequence position mappings should be unchanged + */ + assertNull(ann.getAnnotationForPosition(1)); + assertNull(ann.getAnnotationForPosition(2)); + assertNull(ann.getAnnotationForPosition(3)); + assertEquals(4, ann.getAnnotationForPosition(4).value, 0.001); + assertEquals(5, ann.getAnnotationForPosition(5).value, 0.001); + assertEquals(6, ann.getAnnotationForPosition(6).value, 0.001); + assertNull(ann.getAnnotationForPosition(7)); + assertNull(ann.getAnnotationForPosition(8)); + assertNull(ann.getAnnotationForPosition(9)); + } /** * Test method that returns a map of lists of sequences by sequence name. * * @throws IOException */ - @Test + @Test(groups = { "Functional" }) public void testGetSequencesByName() throws IOException { final String data = ">Seq1Name\nKQYL\n" + ">Seq2Name\nRFPW\n" + ">Seq1Name\nABCD\n"; - AlignmentI al = loadAlignment(data, "FASTA"); + AlignmentI al = loadAlignment(data, FileFormat.Fasta); Map> map = AlignmentUtils .getSequencesByName(al); assertEquals(2, map.keySet().size()); @@ -129,22 +223,25 @@ public class AlignmentUtilsTests assertEquals(1, map.get("Seq2Name").size()); assertEquals("RFPW", map.get("Seq2Name").get(0).getSequenceAsString()); } + /** * Helper method to load an alignment and ensure dataset sequences are set up. * * @param data - * @param format TODO + * @param format + * TODO * @return * @throws IOException */ - protected AlignmentI loadAlignment(final String data, String format) throws IOException + protected AlignmentI loadAlignment(final String data, FileFormatI format) + throws IOException { - Alignment a = new FormatAdapter().readFile(data, - AppletFormatAdapter.PASTE, format); + AlignmentI a = new FormatAdapter().readFile(data, + DataSourceType.PASTE, format); a.setDataset(null); return a; } - + /** * Test mapping of protein to cDNA, for the case where we have no sequence * cross-references, so mappings are made first-served 1-1 where sequences @@ -152,8 +249,8 @@ public class AlignmentUtilsTests * * @throws IOException */ - @Test - public void testMapProteinToCdna_noXrefs() throws IOException + @Test(groups = { "Functional" }) + public void testMapProteinAlignmentToCdna_noXrefs() throws IOException { List protseqs = new ArrayList(); protseqs.add(new Sequence("UNIPROT|V12345", "EIQ")); @@ -170,7 +267,7 @@ public class AlignmentUtilsTests AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[4])); cdna.setDataset(null); - assertTrue(AlignmentUtils.mapProteinToCdna(protein, cdna)); + assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna)); // 3 mappings made, each from 1 to 1 sequence assertEquals(3, protein.getCodonFrames().size()); @@ -179,8 +276,8 @@ public class AlignmentUtilsTests assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size()); // V12345 mapped to A22222 - AlignedCodonFrame acf = protein.getCodonFrame( - protein.getSequenceAt(0)).get(0); + AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0)) + .get(0); assertEquals(1, acf.getdnaSeqs().length); assertEquals(cdna.getSequenceAt(1).getDatasetSequence(), acf.getdnaSeqs()[0]); @@ -189,11 +286,11 @@ public class AlignmentUtilsTests MapList mapList = protMappings[0].getMap(); assertEquals(3, mapList.getFromRatio()); assertEquals(1, mapList.getToRatio()); - assertTrue(Arrays.equals(new int[] - { 1, 9 }, mapList.getFromRanges().get(0))); + assertTrue(Arrays.equals(new int[] { 1, 9 }, mapList.getFromRanges() + .get(0))); assertEquals(1, mapList.getFromRanges().size()); - assertTrue(Arrays.equals(new int[] - { 1, 3 }, mapList.getToRanges().get(0))); + assertTrue(Arrays.equals(new int[] { 1, 3 }, + mapList.getToRanges().get(0))); assertEquals(1, mapList.getToRanges().size()); // V12346 mapped to A33333 @@ -215,12 +312,10 @@ public class AlignmentUtilsTests /** * Test for the alignSequenceAs method that takes two sequences and a mapping. */ - @Test + @Test(groups = { "Functional" }) public void testAlignSequenceAs_withMapping_noIntrons() { - MapList map = new MapList(new int[] - { 1, 6 }, new int[] - { 1, 2 }, 3, 1); + MapList map = new MapList(new int[] { 1, 6 }, new int[] { 1, 2 }, 3, 1); /* * No existing gaps in dna: @@ -246,8 +341,8 @@ public class AlignmentUtilsTests * region). The leading gap, and the gaps between codons, are subsumed by * the protein alignment gap. */ - checkAlignSequenceAs("-G-GG--AA-A-", "-A-L-", true, true, map, - "---G-GG---AA-A-"); + checkAlignSequenceAs("-G-GG--AA-A---", "-A-L-", true, true, map, + "---G-GG---AA-A---"); /* * Include only unmapped gaps in dna when realigning (outside the exon @@ -255,21 +350,20 @@ public class AlignmentUtilsTests * the protein alignment gap. */ checkAlignSequenceAs("-G-GG--AA-A-", "-A-L-", false, true, map, - "---GGG---AAA-"); + "---GGG---AAA---"); } /** * Test for the alignSequenceAs method that takes two sequences and a mapping. */ - @Test + @Test(groups = { "Functional" }) public void testAlignSequenceAs_withMapping_withIntrons() { /* * Exons at codon 2 (AAA) and 4 (TTT) */ - MapList map = new MapList(new int[] - { 4, 6, 10, 12 }, new int[] - { 1, 2 }, 3, 1); + MapList map = new MapList(new int[] { 4, 6, 10, 12 }, + new int[] { 1, 2 }, 3, 1); /* * Simple case: no gaps in dna @@ -311,100 +405,71 @@ public class AlignmentUtilsTests /** * Test for the case where not all of the protein sequence is mapped to cDNA. */ - @Test + @Test(groups = { "Functional" }) public void testAlignSequenceAs_withMapping_withUnmappedProtein() { - /* * Exons at codon 2 (AAA) and 4 (TTT) mapped to A and P */ - final MapList map = new MapList(new int[] - { 4, 6, 10, 12 }, new int[] - { 1, 1, 3, 3 }, 3, 1); - + final MapList map = new MapList(new int[] { 4, 6, 10, 12 }, new int[] { + 1, 1, 3, 3 }, 3, 1); /* - * Expect alignment does nothing (aborts realignment). Change this test - * first if different behaviour wanted. + * -L- 'aligns' ccc------ */ - checkAlignSequenceAs("GGGAAACCCTTTGGG", "-A-L-P-", false, - false, map, "GGGAAACCCTTTGGG"); + checkAlignSequenceAs("gggAAAcccTTTggg", "-A-L-P-", false, false, map, + "gggAAAccc------TTTggg"); } /** * Helper method that performs and verifies the method under test. * - * @param dnaSeq - * @param proteinSeq + * @param alignee + * the sequence to be realigned + * @param alignModel + * the sequence whose alignment is to be copied * @param preserveMappedGaps * @param preserveUnmappedGaps * @param map * @param expected */ - protected void checkAlignSequenceAs(final String dnaSeq, - final String proteinSeq, final boolean preserveMappedGaps, + protected void checkAlignSequenceAs(final String alignee, + final String alignModel, final boolean preserveMappedGaps, final boolean preserveUnmappedGaps, MapList map, final String expected) { - SequenceI dna = new Sequence("Seq1", dnaSeq); - dna.createDatasetSequence(); - SequenceI protein = new Sequence("Seq1", proteinSeq); - protein.createDatasetSequence(); + SequenceI alignMe = new Sequence("Seq1", alignee); + alignMe.createDatasetSequence(); + SequenceI alignFrom = new Sequence("Seq2", alignModel); + alignFrom.createDatasetSequence(); AlignedCodonFrame acf = new AlignedCodonFrame(); - acf.addMap(dna.getDatasetSequence(), protein.getDatasetSequence(), map); + acf.addMap(alignMe.getDatasetSequence(), alignFrom.getDatasetSequence(), map); - AlignmentUtils.alignSequenceAs(dna, protein, acf, "---", '-', + AlignmentUtils.alignSequenceAs(alignMe, alignFrom, acf, "---", '-', preserveMappedGaps, preserveUnmappedGaps); - assertEquals(expected, dna.getSequenceAsString()); + assertEquals(expected, alignMe.getSequenceAsString()); } /** * Test for the alignSequenceAs method where we preserve gaps in introns only. */ - @Test + @Test(groups = { "Functional" }) public void testAlignSequenceAs_keepIntronGapsOnly() { /* * Intron GGGAAA followed by exon CCCTTT */ - MapList map = new MapList(new int[] - { 7, 12 }, new int[] - { 1, 2 }, 3, 1); - - checkAlignSequenceAs("GG-G-AA-A-C-CC-T-TT", "AL", - false, true, map, "GG-G-AA-ACCCTTT"); - } + MapList map = new MapList(new int[] { 7, 12 }, new int[] { 1, 2 }, 3, 1); - /** - * Test for the method that generates an aligned translated sequence from one - * mapping. - */ - @Test - public void testGetAlignedTranslation_dnaLikeProtein() - { - // dna alignment will be replaced - SequenceI dna = new Sequence("Seq1", "T-G-CC-A--T-TAC-CAG-"); - dna.createDatasetSequence(); - // protein alignment will be 'applied' to dna - SequenceI protein = new Sequence("Seq1", "-CH-Y--Q-"); - protein.createDatasetSequence(); - MapList map = new MapList(new int[] - { 1, 12 }, new int[] - { 1, 4 }, 3, 1); - AlignedCodonFrame acf = new AlignedCodonFrame(); - acf.addMap(dna.getDatasetSequence(), protein.getDatasetSequence(), map); - - final SequenceI aligned = AlignmentUtils - .getAlignedTranslation(protein, '-', acf); - assertEquals("---TGCCAT---TAC------CAG---", aligned.getSequenceAsString()); - assertSame(aligned.getDatasetSequence(), dna.getDatasetSequence()); + checkAlignSequenceAs("GG-G-AA-A-C-CC-T-TT", "AL", false, true, map, + "GG-G-AA-ACCCTTT"); } /** * Test the method that realigns protein to match mapped codon alignment. */ - @Test + @Test(groups = { "Functional" }) public void testAlignProteinAsDna() { // seq1 codons are [1,2,3] [4,5,6] [7,8,9] [10,11,12] @@ -413,26 +478,26 @@ public class AlignmentUtilsTests SequenceI dna2 = new Sequence("Seq2", "T-GCCATTACCAG"); // seq3 codons are [1,2,3] [4,5,7] [8,9,10] [11,12,13] SequenceI dna3 = new Sequence("Seq3", "TGCCA-TTACCAG"); - AlignmentI dna = new Alignment(new SequenceI[] - { dna1, dna2, dna3 }); + AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 }); dna.setDataset(null); // protein alignment will be realigned like dna SequenceI prot1 = new Sequence("Seq1", "CHYQ"); SequenceI prot2 = new Sequence("Seq2", "CHYQ"); SequenceI prot3 = new Sequence("Seq3", "CHYQ"); - AlignmentI protein = new Alignment(new SequenceI[] - { prot1, prot2, prot3 }); + SequenceI prot4 = new Sequence("Seq4", "R-QSV"); // unmapped, unchanged + AlignmentI protein = new Alignment(new SequenceI[] { prot1, prot2, + prot3, prot4 }); protein.setDataset(null); - MapList map = new MapList(new int[] - { 1, 12 }, new int[] - { 1, 4 }, 3, 1); + MapList map = new MapList(new int[] { 1, 12 }, new int[] { 1, 4 }, 3, 1); AlignedCodonFrame acf = new AlignedCodonFrame(); acf.addMap(dna1.getDatasetSequence(), prot1.getDatasetSequence(), map); acf.addMap(dna2.getDatasetSequence(), prot2.getDatasetSequence(), map); acf.addMap(dna3.getDatasetSequence(), prot3.getDatasetSequence(), map); - protein.setCodonFrames(Collections.singleton(acf)); + ArrayList acfs = new ArrayList(); + acfs.add(acf); + protein.setCodonFrames(acfs); /* * Translated codon order is [1,2,3] [1,3,4] [4,5,6] [4,5,7] [5,6,7] [7,8,9] @@ -442,43 +507,80 @@ public class AlignmentUtilsTests assertEquals("C-H--Y-Q-", prot1.getSequenceAsString()); assertEquals("-C--H-Y-Q", prot2.getSequenceAsString()); assertEquals("C--H--Y-Q", prot3.getSequenceAsString()); + assertEquals("R-QSV", prot4.getSequenceAsString()); } /** * Test the method that tests whether a CDNA sequence translates to a protein * sequence */ - @Test + @Test(groups = { "Functional" }) public void testTranslatesAs() { + // null arguments check + assertFalse(AlignmentUtils.translatesAs(null, 0, null)); + assertFalse(AlignmentUtils.translatesAs(new char[] { 't' }, 0, null)); + assertFalse(AlignmentUtils.translatesAs(null, 0, new char[] { 'a' })); + + // straight translation assertTrue(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(), 0, "FPKG".toCharArray())); - // with start codon + // with extra start codon (not in protein) assertTrue(AlignmentUtils.translatesAs("atgtttcccaaaggg".toCharArray(), 3, "FPKG".toCharArray())); - // with stop codon1 + // with stop codon1 (not in protein) assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtaa".toCharArray(), 0, "FPKG".toCharArray())); - // with stop codon2 + // with stop codon1 (in protein as *) + assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtaa".toCharArray(), + 0, "FPKG*".toCharArray())); + // with stop codon2 (not in protein) assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtag".toCharArray(), 0, "FPKG".toCharArray())); - // with stop codon3 + // with stop codon3 (not in protein) assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtga".toCharArray(), 0, "FPKG".toCharArray())); // with start and stop codon1 assertTrue(AlignmentUtils.translatesAs( - "atgtttcccaaaggtaa".toCharArray(), 3, "FPKG".toCharArray())); + "atgtttcccaaagggtaa".toCharArray(), 3, "FPKG".toCharArray())); + // with start and stop codon1 (in protein as *) + assertTrue(AlignmentUtils.translatesAs( + "atgtttcccaaagggtaa".toCharArray(), 3, "FPKG*".toCharArray())); // with start and stop codon2 assertTrue(AlignmentUtils.translatesAs( - "atgtttcccaaaggtag".toCharArray(), 3, "FPKG".toCharArray())); + "atgtttcccaaagggtag".toCharArray(), 3, "FPKG".toCharArray())); // with start and stop codon3 assertTrue(AlignmentUtils.translatesAs( - "atgtttcccaaaggtga".toCharArray(), 3, "FPKG".toCharArray())); + "atgtttcccaaagggtga".toCharArray(), 3, "FPKG".toCharArray())); + + // with embedded stop codons + assertTrue(AlignmentUtils.translatesAs( + "atgtttTAGcccaaaTAAgggtga".toCharArray(), 3, + "F*PK*G".toCharArray())); // wrong protein assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(), - 0, - "FPMG".toCharArray())); + 0, "FPMG".toCharArray())); + + // truncated dna + assertFalse(AlignmentUtils.translatesAs("tttcccaaagg".toCharArray(), 0, + "FPKG".toCharArray())); + + // truncated protein + assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(), + 0, "FPK".toCharArray())); + + // overlong dna (doesn't end in stop codon) + assertFalse(AlignmentUtils.translatesAs( + "tttcccaaagggttt".toCharArray(), 0, "FPKG".toCharArray())); + + // dna + stop codon + more + assertFalse(AlignmentUtils.translatesAs( + "tttcccaaagggttaga".toCharArray(), 0, "FPKG".toCharArray())); + + // overlong protein + assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(), + 0, "FPKGQ".toCharArray())); } /** @@ -487,8 +589,8 @@ public class AlignmentUtilsTests * * @throws IOException */ - @Test - public void testMapProteinToCdna_withStartAndStopCodons() + @Test(groups = { "Functional" }) + public void testMapProteinAlignmentToCdna_withStartAndStopCodons() throws IOException { List protseqs = new ArrayList(); @@ -497,7 +599,7 @@ public class AlignmentUtilsTests protseqs.add(new Sequence("UNIPROT|V12347", "SAR")); AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3])); protein.setDataset(null); - + List dnaseqs = new ArrayList(); // start + SAR: dnaseqs.add(new Sequence("EMBL|A11111", "ATGTCAGCACGC")); @@ -508,18 +610,18 @@ public class AlignmentUtilsTests dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG")); AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[4])); cdna.setDataset(null); - - assertTrue(AlignmentUtils.mapProteinToCdna(protein, cdna)); + + assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna)); // 3 mappings made, each from 1 to 1 sequence assertEquals(3, protein.getCodonFrames().size()); assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size()); assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size()); assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size()); - + // V12345 mapped from A22222 - AlignedCodonFrame acf = protein.getCodonFrame( - protein.getSequenceAt(0)).get(0); + AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0)) + .get(0); assertEquals(1, acf.getdnaSeqs().length); assertEquals(cdna.getSequenceAt(1).getDatasetSequence(), acf.getdnaSeqs()[0]); @@ -528,11 +630,11 @@ public class AlignmentUtilsTests MapList mapList = protMappings[0].getMap(); assertEquals(3, mapList.getFromRatio()); assertEquals(1, mapList.getToRatio()); - assertTrue(Arrays.equals(new int[] - { 1, 9 }, mapList.getFromRanges().get(0))); + assertTrue(Arrays.equals(new int[] { 1, 9 }, mapList.getFromRanges() + .get(0))); assertEquals(1, mapList.getFromRanges().size()); - assertTrue(Arrays.equals(new int[] - { 1, 3 }, mapList.getToRanges().get(0))); + assertTrue(Arrays.equals(new int[] { 1, 3 }, + mapList.getToRanges().get(0))); assertEquals(1, mapList.getToRanges().size()); // V12346 mapped from A33333 starting position 4 @@ -545,13 +647,13 @@ public class AlignmentUtilsTests mapList = protMappings[0].getMap(); assertEquals(3, mapList.getFromRatio()); assertEquals(1, mapList.getToRatio()); - assertTrue(Arrays.equals(new int[] - { 4, 12 }, mapList.getFromRanges().get(0))); + assertTrue(Arrays.equals(new int[] { 4, 12 }, mapList.getFromRanges() + .get(0))); assertEquals(1, mapList.getFromRanges().size()); - assertTrue(Arrays.equals(new int[] - { 1, 3 }, mapList.getToRanges().get(0))); + assertTrue(Arrays.equals(new int[] { 1, 3 }, + mapList.getToRanges().get(0))); assertEquals(1, mapList.getToRanges().size()); - + // V12347 mapped to A11111 starting position 4 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0); assertEquals(1, acf.getdnaSeqs().length); @@ -562,13 +664,13 @@ public class AlignmentUtilsTests mapList = protMappings[0].getMap(); assertEquals(3, mapList.getFromRatio()); assertEquals(1, mapList.getToRatio()); - assertTrue(Arrays.equals(new int[] - { 4, 12 }, mapList.getFromRanges().get(0))); + assertTrue(Arrays.equals(new int[] { 4, 12 }, mapList.getFromRanges() + .get(0))); assertEquals(1, mapList.getFromRanges().size()); - assertTrue(Arrays.equals(new int[] - { 1, 3 }, mapList.getToRanges().get(0))); + assertTrue(Arrays.equals(new int[] { 1, 3 }, + mapList.getToRanges().get(0))); assertEquals(1, mapList.getToRanges().size()); - + // no mapping involving the 'extra' A44444 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(3)).isEmpty()); } @@ -580,8 +682,8 @@ public class AlignmentUtilsTests * * @throws IOException */ - @Test - public void testMapProteinToCdna_withXrefs() throws IOException + @Test(groups = { "Functional" }) + public void testMapProteinAlignmentToCdna_withXrefs() throws IOException { List protseqs = new ArrayList(); protseqs.add(new Sequence("UNIPROT|V12345", "EIQ")); @@ -589,7 +691,7 @@ public class AlignmentUtilsTests protseqs.add(new Sequence("UNIPROT|V12347", "SAR")); AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3])); protein.setDataset(null); - + List dnaseqs = new ArrayList(); dnaseqs.add(new Sequence("EMBL|A11111", "TCAGCACGC")); // = SAR dnaseqs.add(new Sequence("EMBL|A22222", "ATGGAGATACAA")); // = start + EIQ @@ -598,7 +700,7 @@ public class AlignmentUtilsTests dnaseqs.add(new Sequence("EMBL|A55555", "GAGATTCAG")); // = EIQ AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[5])); cdna.setDataset(null); - + // Xref A22222 to V12345 (should get mapped) dnaseqs.get(1).addDBRef(new DBRefEntry("UNIPROT", "1", "V12345")); // Xref V12345 to A44444 (should get mapped) @@ -610,7 +712,7 @@ public class AlignmentUtilsTests // A11111 should be mapped to V12347 // A55555 is spare and has no xref so is not mapped - assertTrue(AlignmentUtils.mapProteinToCdna(protein, cdna)); + assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna)); // 4 protein mappings made for 3 proteins, 2 to V12345, 1 each to V12346/7 assertEquals(3, protein.getCodonFrames().size()); @@ -623,28 +725,28 @@ public class AlignmentUtilsTests assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(1)).size()); assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(2)).size()); assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(3)).size()); - + // V12345 mapped to A22222 and A44444 - AlignedCodonFrame acf = protein.getCodonFrame( - protein.getSequenceAt(0)).get(0); + AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0)) + .get(0); assertEquals(2, acf.getdnaSeqs().length); assertEquals(cdna.getSequenceAt(1).getDatasetSequence(), acf.getdnaSeqs()[0]); assertEquals(cdna.getSequenceAt(3).getDatasetSequence(), acf.getdnaSeqs()[1]); - + // V12346 mapped to A33333 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0); assertEquals(1, acf.getdnaSeqs().length); assertEquals(cdna.getSequenceAt(2).getDatasetSequence(), acf.getdnaSeqs()[0]); - + // V12347 mapped to A11111 acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0); assertEquals(1, acf.getdnaSeqs().length); assertEquals(cdna.getSequenceAt(0).getDatasetSequence(), acf.getdnaSeqs()[0]); - + // no mapping involving the 'extra' A55555 assertTrue(protein.getCodonFrame(cdna.getSequenceAt(4)).isEmpty()); } @@ -656,8 +758,9 @@ public class AlignmentUtilsTests * * @throws IOException */ - @Test - public void testMapProteinToCdna_prioritiseXrefs() throws IOException + @Test(groups = { "Functional" }) + public void testMapProteinAlignmentToCdna_prioritiseXrefs() + throws IOException { List protseqs = new ArrayList(); protseqs.add(new Sequence("UNIPROT|V12345", "EIQ")); @@ -665,36 +768,36 @@ public class AlignmentUtilsTests AlignmentI protein = new Alignment( protseqs.toArray(new SequenceI[protseqs.size()])); protein.setDataset(null); - + List dnaseqs = new ArrayList(); dnaseqs.add(new Sequence("EMBL|A11111", "GAAATCCAG")); // = EIQ dnaseqs.add(new Sequence("EMBL|A22222", "GAAATTCAG")); // = EIQ AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[dnaseqs .size()])); cdna.setDataset(null); - + // Xref A22222 to V12345 (should get mapped) // A11111 should then be mapped to the unmapped V12346 dnaseqs.get(1).addDBRef(new DBRefEntry("UNIPROT", "1", "V12345")); - - assertTrue(AlignmentUtils.mapProteinToCdna(protein, cdna)); - + + assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna)); + // 2 protein mappings made assertEquals(2, protein.getCodonFrames().size()); assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size()); assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size()); - + // one mapping for each of the cDNA sequences assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(0)).size()); assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(1)).size()); - + // V12345 mapped to A22222 AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0)) .get(0); assertEquals(1, acf.getdnaSeqs().length); assertEquals(cdna.getSequenceAt(1).getDatasetSequence(), acf.getdnaSeqs()[0]); - + // V12346 mapped to A11111 acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0); assertEquals(1, acf.getdnaSeqs().length); @@ -706,14 +809,13 @@ public class AlignmentUtilsTests * Test the method that shows or hides sequence annotations by type(s) and * selection group. */ - @Test + @Test(groups = { "Functional" }) public void testShowOrHideSequenceAnnotations() { SequenceI seq1 = new Sequence("Seq1", "AAA"); SequenceI seq2 = new Sequence("Seq2", "BBB"); SequenceI seq3 = new Sequence("Seq3", "CCC"); - Annotation[] anns = new Annotation[] - { new Annotation(2f) }; + Annotation[] anns = new Annotation[] { new Annotation(2f) }; AlignmentAnnotation ann1 = new AlignmentAnnotation("Structure", "ann1", anns); ann1.setSequenceRef(seq1); @@ -727,7 +829,7 @@ public class AlignmentUtilsTests AlignmentAnnotation ann5 = new AlignmentAnnotation("Temp", "ann5", anns); ann5.setSequenceRef(seq2); AlignmentAnnotation ann6 = new AlignmentAnnotation("Temp", "ann6", anns); - AlignmentI al = new Alignment(new SequenceI[] {seq1, seq2, seq3}); + AlignmentI al = new Alignment(new SequenceI[] { seq1, seq2, seq3 }); al.addAnnotation(ann1); // Structure for Seq1 al.addAnnotation(ann2); // Structure for Seq2 al.addAnnotation(ann3); // Structure for no sequence @@ -814,7 +916,7 @@ public class AlignmentUtilsTests /** * Tests for the method that checks if one sequence cross-references another */ - @Test + @Test(groups = { "Functional" }) public void testHasCrossRef() { assertFalse(AlignmentUtils.hasCrossRef(null, null)); @@ -823,15 +925,15 @@ public class AlignmentUtilsTests assertFalse(AlignmentUtils.hasCrossRef(null, seq1)); SequenceI seq2 = new Sequence("UNIPROT|V20192", "ABCDEF"); assertFalse(AlignmentUtils.hasCrossRef(seq1, seq2)); - + // different ref seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "v20193")); assertFalse(AlignmentUtils.hasCrossRef(seq1, seq2)); - + // case-insensitive; version number is ignored seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "v20192")); assertTrue(AlignmentUtils.hasCrossRef(seq1, seq2)); - + // right case! seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192")); assertTrue(AlignmentUtils.hasCrossRef(seq1, seq2)); @@ -843,7 +945,7 @@ public class AlignmentUtilsTests * Tests for the method that checks if either sequence cross-references the * other */ - @Test + @Test(groups = { "Functional" }) public void testHaveCrossRef() { assertFalse(AlignmentUtils.hasCrossRef(null, null)); @@ -852,21 +954,1247 @@ public class AlignmentUtilsTests assertFalse(AlignmentUtils.haveCrossRef(null, seq1)); SequenceI seq2 = new Sequence("UNIPROT|V20192", "ABCDEF"); assertFalse(AlignmentUtils.haveCrossRef(seq1, seq2)); - + seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192")); assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2)); // next is true for haveCrossRef, false for hasCrossRef assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1)); - + // now the other way round - seq1.setDBRef(null); + seq1.setDBRefs(null); seq2.addDBRef(new DBRefEntry("EMBL", "1", "A12345")); assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2)); assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1)); - + // now both ways seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192")); assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2)); assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1)); } + + /** + * Test the method that extracts the cds-only part of a dna alignment. + */ + @Test(groups = { "Functional" }) + public void testMakeCdsAlignment() + { + 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(); + dna1.addSequenceFeature(new SequenceFeature("CDS", "cds1", 4, 6, 0f, + null)); + dna1.addSequenceFeature(new SequenceFeature("CDS", "cds2", 10, 12, 0f, + null)); + dna2.addSequenceFeature(new SequenceFeature("CDS", "cds3", 1, 3, 0f, + null)); + dna2.addSequenceFeature(new SequenceFeature("CDS", "cds4", 7, 9, 0f, + null)); + dna2.addSequenceFeature(new SequenceFeature("CDS", "cds5", 13, 15, 0f, + null)); + AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 }); + dna.setDataset(null); + + List mappings = new ArrayList(); + 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); + + /* + * execute method under test: + */ + AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] { + dna1, dna2 }, mappings, dna); + + assertEquals(2, cds.getSequences().size()); + assertEquals("GGGTTT", cds.getSequenceAt(0) + .getSequenceAsString()); + assertEquals("GGGTTTCCC", cds.getSequenceAt(1) + .getSequenceAsString()); + + /* + * verify shared, extended alignment dataset + */ + assertSame(dna.getDataset(), cds.getDataset()); + assertTrue(dna.getDataset().getSequences() + .contains(cds.getSequenceAt(0).getDatasetSequence())); + assertTrue(dna.getDataset().getSequences() + .contains(cds.getSequenceAt(1).getDatasetSequence())); + + /* + * Verify mappings from CDS to peptide and cDNA to CDS + * the mappings are on the shared alignment dataset + */ + assertSame(dna.getCodonFrames(), cds.getCodonFrames()); + List cdsMappings = cds.getCodonFrames(); + assertEquals(2, cdsMappings.size()); + + /* + * Mapping from pep1 to GGGTTT in first new exon sequence + */ + List pep1Mapping = MappingUtils + .findMappingsForSequence(pep1, cdsMappings); + assertEquals(1, pep1Mapping.size()); + // map G to GGG + SearchResults sr = MappingUtils + .buildSearchResults(pep1, 1, cdsMappings); + assertEquals(1, sr.getResults().size()); + Match m = sr.getResults().get(0); + assertSame(cds.getSequenceAt(0).getDatasetSequence(), + m.getSequence()); + assertEquals(1, m.getStart()); + assertEquals(3, m.getEnd()); + // map F to TTT + sr = MappingUtils.buildSearchResults(pep1, 2, cdsMappings); + m = sr.getResults().get(0); + assertSame(cds.getSequenceAt(0).getDatasetSequence(), + m.getSequence()); + assertEquals(4, m.getStart()); + assertEquals(6, m.getEnd()); + + /* + * Mapping from pep2 to GGGTTTCCC in second new exon sequence + */ + List pep2Mapping = MappingUtils + .findMappingsForSequence(pep2, cdsMappings); + assertEquals(1, pep2Mapping.size()); + // map G to GGG + sr = MappingUtils.buildSearchResults(pep2, 1, cdsMappings); + assertEquals(1, sr.getResults().size()); + m = sr.getResults().get(0); + assertSame(cds.getSequenceAt(1).getDatasetSequence(), + m.getSequence()); + assertEquals(1, m.getStart()); + assertEquals(3, m.getEnd()); + // map F to TTT + sr = MappingUtils.buildSearchResults(pep2, 2, cdsMappings); + m = sr.getResults().get(0); + assertSame(cds.getSequenceAt(1).getDatasetSequence(), + m.getSequence()); + assertEquals(4, m.getStart()); + assertEquals(6, m.getEnd()); + // map P to CCC + sr = MappingUtils.buildSearchResults(pep2, 3, cdsMappings); + m = sr.getResults().get(0); + assertSame(cds.getSequenceAt(1).getDatasetSequence(), + m.getSequence()); + assertEquals(7, m.getStart()); + assertEquals(9, m.getEnd()); + } + + /** + * Test the method that makes a cds-only alignment from a DNA sequence and its + * product mappings, for the case where there are multiple exon mappings to + * different protein products. + */ + @Test(groups = { "Functional" }) + public void testMakeCdsAlignment_multipleProteins() + { + SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa"); + SequenceI pep1 = new Sequence("pep1", "GF"); // GGGTTT + SequenceI pep2 = new Sequence("pep2", "KP"); // aaaccc + SequenceI pep3 = new Sequence("pep3", "KF"); // aaaTTT + dna1.createDatasetSequence(); + pep1.createDatasetSequence(); + pep2.createDatasetSequence(); + pep3.createDatasetSequence(); + dna1.addSequenceFeature(new SequenceFeature("CDS", "cds1", 4, 6, 0f, + null)); + dna1.addSequenceFeature(new SequenceFeature("CDS", "cds2", 10, 12, 0f, + null)); + dna1.addSequenceFeature(new SequenceFeature("CDS", "cds3", 1, 3, 0f, + null)); + dna1.addSequenceFeature(new SequenceFeature("CDS", "cds4", 7, 9, 0f, + null)); + dna1.addSequenceFeature(new SequenceFeature("CDS", "cds5", 1, 3, 0f, + null)); + dna1.addSequenceFeature(new SequenceFeature("CDS", "cds6", 10, 12, 0f, + null)); + pep1.getDatasetSequence().addDBRef( + new DBRefEntry("EMBLCDS", "2", "A12345")); + pep2.getDatasetSequence().addDBRef( + new DBRefEntry("EMBLCDS", "3", "A12346")); + pep3.getDatasetSequence().addDBRef( + new DBRefEntry("EMBLCDS", "4", "A12347")); + + /* + * Make the mappings from dna to protein + */ + List mappings = new ArrayList(); + // map ...GGG...TTT to GF + 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 aaa...ccc to KP + map = new MapList(new int[] { 1, 3, 7, 9 }, new int[] { 1, 2 }, 3, 1); + acf = new AlignedCodonFrame(); + acf.addMap(dna1.getDatasetSequence(), pep2.getDatasetSequence(), map); + mappings.add(acf); + + // map aaa......TTT to KF + map = new MapList(new int[] { 1, 3, 10, 12 }, new int[] { 1, 2 }, 3, 1); + acf = new AlignedCodonFrame(); + acf.addMap(dna1.getDatasetSequence(), pep3.getDatasetSequence(), map); + mappings.add(acf); + + /* + * Create the CDS alignment; also augments the dna-to-protein mappings with + * exon-to-protein and exon-to-dna mappings + */ + AlignmentI dna = new Alignment(new SequenceI[] { dna1 }); + dna.setDataset(null); + + /* + * execute method under test + */ + AlignmentI cdsal = AlignmentUtils.makeCdsAlignment( + new SequenceI[] { dna1 }, mappings, dna); + + /* + * Verify we have 3 cds sequences, mapped to pep1/2/3 respectively + */ + List cds = cdsal.getSequences(); + assertEquals(3, cds.size()); + + /* + * verify shared, extended alignment dataset + */ + assertSame(cdsal.getDataset(), dna.getDataset()); + assertTrue(dna.getDataset().getSequences() + .contains(cds.get(0).getDatasetSequence())); + assertTrue(dna.getDataset().getSequences() + .contains(cds.get(1).getDatasetSequence())); + assertTrue(dna.getDataset().getSequences() + .contains(cds.get(2).getDatasetSequence())); + + /* + * verify aligned cds sequences and their xrefs + */ + SequenceI cdsSeq = cds.get(0); + assertEquals("GGGTTT", cdsSeq.getSequenceAsString()); + // assertEquals("dna1|A12345", cdsSeq.getName()); + assertEquals("dna1|pep1", cdsSeq.getName()); + // assertEquals(1, cdsSeq.getDBRefs().length); + // DBRefEntry cdsRef = cdsSeq.getDBRefs()[0]; + // assertEquals("EMBLCDS", cdsRef.getSource()); + // assertEquals("2", cdsRef.getVersion()); + // assertEquals("A12345", cdsRef.getAccessionId()); + + cdsSeq = cds.get(1); + assertEquals("aaaccc", cdsSeq.getSequenceAsString()); + // assertEquals("dna1|A12346", cdsSeq.getName()); + assertEquals("dna1|pep2", cdsSeq.getName()); + // assertEquals(1, cdsSeq.getDBRefs().length); + // cdsRef = cdsSeq.getDBRefs()[0]; + // assertEquals("EMBLCDS", cdsRef.getSource()); + // assertEquals("3", cdsRef.getVersion()); + // assertEquals("A12346", cdsRef.getAccessionId()); + + cdsSeq = cds.get(2); + assertEquals("aaaTTT", cdsSeq.getSequenceAsString()); + // assertEquals("dna1|A12347", cdsSeq.getName()); + assertEquals("dna1|pep3", cdsSeq.getName()); + // assertEquals(1, cdsSeq.getDBRefs().length); + // cdsRef = cdsSeq.getDBRefs()[0]; + // assertEquals("EMBLCDS", cdsRef.getSource()); + // assertEquals("4", cdsRef.getVersion()); + // assertEquals("A12347", cdsRef.getAccessionId()); + + /* + * Verify there are mappings from each cds sequence to its protein product + * and also to its dna source + */ + Iterator newMappingsIterator = cdsal + .getCodonFrames().iterator(); + + // mappings for dna1 - exon1 - pep1 + AlignedCodonFrame cdsMapping = newMappingsIterator.next(); + List dnaMappings = cdsMapping.getMappingsFromSequence(dna1); + assertEquals(3, dnaMappings.size()); + assertSame(cds.get(0).getDatasetSequence(), dnaMappings.get(0) + .getTo()); + assertEquals("G(1) in CDS should map to G(4) in DNA", 4, dnaMappings + .get(0).getMap().getToPosition(1)); + List peptideMappings = cdsMapping.getMappingsFromSequence(cds + .get(0).getDatasetSequence()); + assertEquals(1, peptideMappings.size()); + assertSame(pep1.getDatasetSequence(), peptideMappings.get(0).getTo()); + + // mappings for dna1 - cds2 - pep2 + assertSame(cds.get(1).getDatasetSequence(), dnaMappings.get(1) + .getTo()); + assertEquals("c(4) in CDS should map to c(7) in DNA", 7, dnaMappings + .get(1).getMap().getToPosition(4)); + peptideMappings = cdsMapping.getMappingsFromSequence(cds.get(1) + .getDatasetSequence()); + assertEquals(1, peptideMappings.size()); + assertSame(pep2.getDatasetSequence(), peptideMappings.get(0).getTo()); + + // mappings for dna1 - cds3 - pep3 + assertSame(cds.get(2).getDatasetSequence(), dnaMappings.get(2) + .getTo()); + assertEquals("T(4) in CDS should map to T(10) in DNA", 10, dnaMappings + .get(2).getMap().getToPosition(4)); + peptideMappings = cdsMapping.getMappingsFromSequence(cds.get(2) + .getDatasetSequence()); + assertEquals(1, peptideMappings.size()); + assertSame(pep3.getDatasetSequence(), peptideMappings.get(0).getTo()); + } + + @Test(groups = { "Functional" }) + public void testIsMappable() + { + SequenceI dna1 = new Sequence("dna1", "cgCAGtgGT"); + SequenceI aa1 = new Sequence("aa1", "RSG"); + AlignmentI al1 = new Alignment(new SequenceI[] { dna1 }); + AlignmentI al2 = new Alignment(new SequenceI[] { aa1 }); + + assertFalse(AlignmentUtils.isMappable(null, null)); + assertFalse(AlignmentUtils.isMappable(al1, null)); + assertFalse(AlignmentUtils.isMappable(null, al1)); + assertFalse(AlignmentUtils.isMappable(al1, al1)); + assertFalse(AlignmentUtils.isMappable(al2, al2)); + + assertTrue(AlignmentUtils.isMappable(al1, al2)); + assertTrue(AlignmentUtils.isMappable(al2, al1)); + } + + /** + * Test creating a mapping when the sequences involved do not start at residue + * 1 + * + * @throws IOException + */ + @Test(groups = { "Functional" }) + public void testMapCdnaToProtein_forSubsequence() + throws IOException + { + SequenceI prot = new Sequence("UNIPROT|V12345", "E-I--Q", 10, 12); + prot.createDatasetSequence(); + + SequenceI dna = new Sequence("EMBL|A33333", "GAA--AT-C-CAG", 40, 48); + dna.createDatasetSequence(); + + MapList map = AlignmentUtils.mapCdnaToProtein(prot, dna); + assertEquals(10, map.getToLowest()); + assertEquals(12, map.getToHighest()); + assertEquals(40, map.getFromLowest()); + assertEquals(48, map.getFromHighest()); + } + + /** + * Test for the alignSequenceAs method where we have protein mapped to protein + */ + @Test(groups = { "Functional" }) + public void testAlignSequenceAs_mappedProteinProtein() + { + + SequenceI alignMe = new Sequence("Match", "MGAASEV"); + alignMe.createDatasetSequence(); + SequenceI alignFrom = new Sequence("Query", "LQTGYMGAASEVMFSPTRR"); + alignFrom.createDatasetSequence(); + + AlignedCodonFrame acf = new AlignedCodonFrame(); + // this is like a domain or motif match of part of a peptide sequence + 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()); + } + + /** + * Test for the alignSequenceAs method where there are trailing unmapped + * residues in the model sequence + */ + @Test(groups = { "Functional" }) + public void testAlignSequenceAs_withTrailingPeptide() + { + // 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---"); + } + + /** + * Tests for transferring features between mapped sequences + */ + @Test(groups = { "Functional" }) + public void testTransferFeatures() + { + SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt"); + SequenceI cds = new Sequence("cds/10-15", "TAGGCC"); + + // no overlap + dna.addSequenceFeature(new SequenceFeature("type1", "desc1", 1, 2, 1f, + null)); + // partial overlap - to [1, 1] + dna.addSequenceFeature(new SequenceFeature("type2", "desc2", 3, 4, 2f, + null)); + // exact overlap - to [1, 3] + dna.addSequenceFeature(new SequenceFeature("type3", "desc3", 4, 6, 3f, + null)); + // spanning overlap - to [2, 5] + dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f, + null)); + // exactly overlaps whole mapped range [1, 6] + dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f, + null)); + // no overlap (internal) + dna.addSequenceFeature(new SequenceFeature("type6", "desc6", 7, 9, 6f, + null)); + // no overlap (3' end) + dna.addSequenceFeature(new SequenceFeature("type7", "desc7", 13, 15, + 7f, null)); + // overlap (3' end) - to [6, 6] + dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12, + 8f, null)); + // extended overlap - to [6, +] + dna.addSequenceFeature(new SequenceFeature("type9", "desc9", 12, 13, + 9f, null)); + + MapList map = new MapList(new int[] { 4, 6, 10, 12 }, + new int[] { 1, 6 }, 1, 1); + + /* + * transferFeatures() will build 'partial overlap' for regions + * that partially overlap 5' or 3' (start or end) of target sequence + */ + AlignmentUtils.transferFeatures(dna, cds, map, null); + SequenceFeature[] sfs = cds.getSequenceFeatures(); + assertEquals(6, sfs.length); + + SequenceFeature sf = sfs[0]; + assertEquals("type2", sf.getType()); + assertEquals("desc2", sf.getDescription()); + assertEquals(2f, sf.getScore()); + assertEquals(1, sf.getBegin()); + assertEquals(1, sf.getEnd()); + + sf = sfs[1]; + assertEquals("type3", sf.getType()); + assertEquals("desc3", sf.getDescription()); + assertEquals(3f, sf.getScore()); + assertEquals(1, sf.getBegin()); + assertEquals(3, sf.getEnd()); + + sf = sfs[2]; + assertEquals("type4", sf.getType()); + assertEquals(2, sf.getBegin()); + assertEquals(5, sf.getEnd()); + + sf = sfs[3]; + assertEquals("type5", sf.getType()); + assertEquals(1, sf.getBegin()); + assertEquals(6, sf.getEnd()); + + sf = sfs[4]; + assertEquals("type8", sf.getType()); + assertEquals(6, sf.getBegin()); + assertEquals(6, sf.getEnd()); + + sf = sfs[5]; + assertEquals("type9", sf.getType()); + assertEquals(6, sf.getBegin()); + assertEquals(6, sf.getEnd()); + } + + /** + * Tests for transferring features between mapped sequences + */ + @Test(groups = { "Functional" }) + public void testTransferFeatures_withOmit() + { + 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)); + // [4, 12] maps to [1, 6] + dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f, + null)); + // [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()); + assertEquals(6, sf.getEnd()); + } + + /** + * Tests for transferring features between mapped sequences + */ + @Test(groups = { "Functional" }) + public void testTransferFeatures_withSelect() + { + 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)); + // [4, 12] maps to [1, 6] + dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f, + null)); + // [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()); + assertEquals(6, sf.getEnd()); + } + + /** + * Test the method that extracts the cds-only part of a dna alignment, for the + * case where the cds should be aligned to match its nucleotide sequence. + */ + @Test(groups = { "Functional" }) + public void testMakeCdsAlignment_alternativeTranscripts() + { + SequenceI dna1 = new Sequence("dna1", "aaaGGGCC-----CTTTaaaGGG"); + // alternative transcript of same dna skips CCC codon + SequenceI dna2 = new Sequence("dna2", "aaaGGGCC-----cttTaaaGGG"); + // dna3 has no mapping (protein product) so should be ignored here + SequenceI dna3 = new Sequence("dna3", "aaaGGGCCCCCGGGcttTaaaGGG"); + SequenceI pep1 = new Sequence("pep1", "GPFG"); + SequenceI pep2 = new Sequence("pep2", "GPG"); + dna1.createDatasetSequence(); + dna2.createDatasetSequence(); + dna3.createDatasetSequence(); + pep1.createDatasetSequence(); + pep2.createDatasetSequence(); + dna1.addSequenceFeature(new SequenceFeature("CDS", "cds1", 4, 8, 0f, + null)); + dna1.addSequenceFeature(new SequenceFeature("CDS", "cds2", 9, 12, 0f, + null)); + dna1.addSequenceFeature(new SequenceFeature("CDS", "cds3", 16, 18, 0f, + null)); + dna2.addSequenceFeature(new SequenceFeature("CDS", "cds", 4, 8, 0f, + null)); + dna2.addSequenceFeature(new SequenceFeature("CDS", "cds", 12, 12, 0f, + null)); + dna2.addSequenceFeature(new SequenceFeature("CDS", "cds", 16, 18, 0f, + null)); + + List mappings = new ArrayList(); + 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); + mappings.add(acf); + map = new MapList(new int[] { 4, 8, 12, 12, 16, 18 }, + new int[] { 1, 3 }, + 3, 1); + acf = new AlignedCodonFrame(); + acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map); + mappings.add(acf); + + AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2, dna3 }); + dna.setDataset(null); + AlignmentI cds = AlignmentUtils.makeCdsAlignment(new SequenceI[] { + dna1, dna2, dna3 }, mappings, dna); + 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 + */ + assertSame(dna.getDataset(), cds.getDataset()); + assertTrue(dna.getDataset().getSequences() + .contains(cdsSeqs.get(0).getDatasetSequence())); + assertTrue(dna.getDataset().getSequences() + .contains(cdsSeqs.get(1).getDatasetSequence())); + + /* + * Verify updated mappings + */ + List cdsMappings = cds.getCodonFrames(); + assertEquals(2, cdsMappings.size()); + + /* + * Mapping from pep1 to GGGTTT in first new CDS sequence + */ + List pep1Mapping = MappingUtils + .findMappingsForSequence(pep1, cdsMappings); + assertEquals(1, pep1Mapping.size()); + /* + * maps GPFG to 1-3,4-6,7-9,10-12 + */ + SearchResults sr = MappingUtils + .buildSearchResults(pep1, 1, cdsMappings); + assertEquals(1, sr.getResults().size()); + Match m = sr.getResults().get(0); + assertEquals(cds.getSequenceAt(0).getDatasetSequence(), + m.getSequence()); + assertEquals(1, m.getStart()); + assertEquals(3, m.getEnd()); + sr = MappingUtils.buildSearchResults(pep1, 2, cdsMappings); + m = sr.getResults().get(0); + assertEquals(4, m.getStart()); + assertEquals(6, m.getEnd()); + sr = MappingUtils.buildSearchResults(pep1, 3, cdsMappings); + m = sr.getResults().get(0); + assertEquals(7, m.getStart()); + assertEquals(9, m.getEnd()); + sr = MappingUtils.buildSearchResults(pep1, 4, cdsMappings); + m = sr.getResults().get(0); + assertEquals(10, m.getStart()); + assertEquals(12, m.getEnd()); + + /* + * GPG in pep2 map to 1-3,4-6,7-9 in second CDS sequence + */ + List pep2Mapping = MappingUtils + .findMappingsForSequence(pep2, cdsMappings); + assertEquals(1, pep2Mapping.size()); + sr = MappingUtils.buildSearchResults(pep2, 1, cdsMappings); + assertEquals(1, sr.getResults().size()); + m = sr.getResults().get(0); + assertEquals(cds.getSequenceAt(1).getDatasetSequence(), + m.getSequence()); + assertEquals(1, m.getStart()); + assertEquals(3, m.getEnd()); + sr = MappingUtils.buildSearchResults(pep2, 2, cdsMappings); + m = sr.getResults().get(0); + assertEquals(4, m.getStart()); + assertEquals(6, m.getEnd()); + sr = MappingUtils.buildSearchResults(pep2, 3, cdsMappings); + m = sr.getResults().get(0); + assertEquals(7, m.getStart()); + assertEquals(9, m.getEnd()); + } + + /** + * Test the method that realigns protein to match mapped codon alignment. + */ + @Test(groups = { "Functional" }) + public void testAlignProteinAsDna_incompleteStartCodon() + { + // seq1: incomplete start codon (not mapped), then [3, 11] + SequenceI dna1 = new Sequence("Seq1", "ccAAA-TTT-GGG-"); + // seq2 codons are [4, 5], [8, 11] + SequenceI dna2 = new Sequence("Seq2", "ccaAA-ttT-GGG-"); + // seq3 incomplete start codon at 'tt' + 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"); + SequenceI prot3 = new Sequence("Seq3", "XG"); // X for incomplete start + 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(); + acf.addMap(dna1.getDatasetSequence(), prot1.getDatasetSequence(), map); + + // map dna2 [4, 5] [8, 11] to prot2 [1, 2] NG + map = new MapList(new int[] { 4, 5, 8, 11 }, new int[] { 1, 2 }, 3, 1); + acf.addMap(dna2.getDatasetSequence(), prot2.getDatasetSequence(), map); + + // map dna3 [9, 11] to prot3 [2, 2] G + map = new MapList(new int[] { 9, 11 }, new int[] { 2, 2 }, 3, 1); + acf.addMap(dna3.getDatasetSequence(), prot3.getDatasetSequence(), map); + + ArrayList acfs = new ArrayList(); + acfs.add(acf); + protein.setCodonFrames(acfs); + + /* + * verify X is included in the aligned proteins, and placed just + * before the first mapped residue + * CCT is between CCC and TTT + */ + AlignmentUtils.alignProteinAsDna(protein, dna); + assertEquals("XK-FG", prot1.getSequenceAsString()); + assertEquals("--N-G", prot2.getSequenceAsString()); + assertEquals("---XG", prot3.getSequenceAsString()); + } + + /** + * Tests for the method that maps the subset of a dna sequence that has CDS + * (or subtype) feature - case where the start codon is incomplete. + */ + @Test(groups = "Functional") + public void testFindCdsPositions_fivePrimeIncomplete() + { + 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 + ds.addSequenceFeature(sf); + // 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 + */ + assertEquals(6, MappingUtils.getLength(ranges)); + assertEquals(2, ranges.size()); + assertEquals(7, ranges.get(0)[0]); + assertEquals(9, ranges.get(0)[1]); + assertEquals(13, ranges.get(1)[0]); + assertEquals(15, ranges.get(1)[1]); + } + + /** + * Tests for the method that maps the subset of a dna sequence that has CDS + * (or subtype) feature. + */ + @Test(groups = "Functional") + public void testFindCdsPositions() + { + 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); + sf.setStrand("+"); + ds.addSequenceFeature(sf); + // CDS for dna 4-6 + sf = new SequenceFeature("CDS", "", 4, 6, 0f, null); + sf.setStrand("+"); + ds.addSequenceFeature(sf); + // 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] } + * note CDS ranges are ordered ascending even if the CDS + * features are not + */ + assertEquals(6, MappingUtils.getLength(ranges)); + assertEquals(2, ranges.size()); + assertEquals(4, ranges.get(0)[0]); + assertEquals(6, ranges.get(0)[1]); + assertEquals(10, ranges.get(1)[0]); + assertEquals(12, ranges.get(1)[1]); + } + + /** + * Test the method that computes a map of codon variants for each protein + * position from "sequence_variant" features on dna + */ + @Test(groups = "Functional") + public void testBuildDnaVariantsMap() + { + SequenceI dna = new Sequence("dna", "atgAAATTTGGGCCCtag"); + MapList map = new MapList(new int[] { 1, 18 }, new int[] { 1, 5 }, 3, 1); + + /* + * first with no variants on dna + */ + LinkedHashMap[]> variantsMap = AlignmentUtils + .buildDnaVariantsMap(dna, map); + assertTrue(variantsMap.isEmpty()); + + /* + * single allele codon 1, on base 1 + */ + SequenceFeature sf1 = new SequenceFeature("sequence_variant", "", 1, 1, + 0f, null); + sf1.setValue("alleles", "T"); + sf1.setValue("ID", "sequence_variant:rs758803211"); + dna.addSequenceFeature(sf1); + + /* + * two alleles codon 2, on bases 2 and 3 (distinct variants) + */ + SequenceFeature sf2 = new SequenceFeature("sequence_variant", "", 5, 5, + 0f, null); + sf2.setValue("alleles", "T"); + sf2.setValue("ID", "sequence_variant:rs758803212"); + dna.addSequenceFeature(sf2); + SequenceFeature sf3 = new SequenceFeature("sequence_variant", "", 6, 6, + 0f, null); + sf3.setValue("alleles", "G"); + sf3.setValue("ID", "sequence_variant:rs758803213"); + dna.addSequenceFeature(sf3); + + /* + * two alleles codon 3, both on base 2 (one variant) + */ + SequenceFeature sf4 = new SequenceFeature("sequence_variant", "", 8, 8, + 0f, null); + sf4.setValue("alleles", "C, G"); + sf4.setValue("ID", "sequence_variant:rs758803214"); + dna.addSequenceFeature(sf4); + + // no alleles on codon 4 + + /* + * alleles on codon 5 on all 3 bases (distinct variants) + */ + SequenceFeature sf5 = new SequenceFeature("sequence_variant", "", 13, + 13, 0f, null); + sf5.setValue("alleles", "C, G"); // (C duplicates given base value) + sf5.setValue("ID", "sequence_variant:rs758803215"); + dna.addSequenceFeature(sf5); + SequenceFeature sf6 = new SequenceFeature("sequence_variant", "", 14, + 14, 0f, null); + sf6.setValue("alleles", "g, a"); // should force to upper-case + sf6.setValue("ID", "sequence_variant:rs758803216"); + dna.addSequenceFeature(sf6); + SequenceFeature sf7 = new SequenceFeature("sequence_variant", "", 15, + 15, 0f, null); + sf7.setValue("alleles", "A, T"); + sf7.setValue("ID", "sequence_variant:rs758803217"); + dna.addSequenceFeature(sf7); + + /* + * build map - expect variants on positions 1, 2, 3, 5 + */ + variantsMap = AlignmentUtils.buildDnaVariantsMap(dna, map); + assertEquals(4, variantsMap.size()); + + /* + * protein residue 1: variant on codon (ATG) base 1, not on 2 or 3 + */ + List[] pep1Variants = variantsMap.get(1); + assertEquals(3, pep1Variants.length); + assertEquals(1, pep1Variants[0].size()); + assertEquals("A", pep1Variants[0].get(0).base); // codon[1] base + assertSame(sf1, pep1Variants[0].get(0).variant); // codon[1] variant + assertEquals(1, pep1Variants[1].size()); + assertEquals("T", pep1Variants[1].get(0).base); // codon[2] base + assertNull(pep1Variants[1].get(0).variant); // no variant here + assertEquals(1, pep1Variants[2].size()); + assertEquals("G", pep1Variants[2].get(0).base); // codon[3] base + assertNull(pep1Variants[2].get(0).variant); // no variant here + + /* + * protein residue 2: variants on codon (AAA) bases 2 and 3 + */ + List[] pep2Variants = variantsMap.get(2); + assertEquals(3, pep2Variants.length); + assertEquals(1, pep2Variants[0].size()); + // codon[1] base recorded while processing variant on codon[2] + assertEquals("A", pep2Variants[0].get(0).base); + assertNull(pep2Variants[0].get(0).variant); // no variant here + // codon[2] base and variant: + assertEquals(1, pep2Variants[1].size()); + assertEquals("A", pep2Variants[1].get(0).base); + assertSame(sf2, pep2Variants[1].get(0).variant); + // codon[3] base was recorded when processing codon[2] variant + // and then the variant for codon[3] added to it + assertEquals(1, pep2Variants[2].size()); + assertEquals("A", pep2Variants[2].get(0).base); + assertSame(sf3, pep2Variants[2].get(0).variant); + + /* + * protein residue 3: variants on codon (TTT) base 2 only + */ + List[] pep3Variants = variantsMap.get(3); + assertEquals(3, pep3Variants.length); + assertEquals(1, pep3Variants[0].size()); + assertEquals("T", pep3Variants[0].get(0).base); // codon[1] base + assertNull(pep3Variants[0].get(0).variant); // no variant here + assertEquals(1, pep3Variants[1].size()); + assertEquals("T", pep3Variants[1].get(0).base); // codon[2] base + assertSame(sf4, pep3Variants[1].get(0).variant); // codon[2] variant + assertEquals(1, pep3Variants[2].size()); + assertEquals("T", pep3Variants[2].get(0).base); // codon[3] base + assertNull(pep3Variants[2].get(0).variant); // no variant here + + /* + * three variants on protein position 5 + */ + List[] pep5Variants = variantsMap.get(5); + assertEquals(3, pep5Variants.length); + assertEquals(1, pep5Variants[0].size()); + assertEquals("C", pep5Variants[0].get(0).base); // codon[1] base + assertSame(sf5, pep5Variants[0].get(0).variant); // codon[1] variant + assertEquals(1, pep5Variants[1].size()); + assertEquals("C", pep5Variants[1].get(0).base); // codon[2] base + assertSame(sf6, pep5Variants[1].get(0).variant); // codon[2] variant + assertEquals(1, pep5Variants[2].size()); + assertEquals("C", pep5Variants[2].get(0).base); // codon[3] base + assertSame(sf7, pep5Variants[2].get(0).variant); // codon[3] variant + } + + /** + * Tests for the method that computes all peptide variants given codon + * variants + */ + @Test(groups = "Functional") + public void testComputePeptideVariants() + { + /* + * scenario: AAATTTCCC codes for KFP, with variants + * GAA -> E + * CAA -> Q + * AAG synonymous + * AAT -> N + * TTC synonymous + * CAC,CGC -> H,R (as one variant) + */ + SequenceI peptide = new Sequence("pep/10-12", "KFP"); + + /* + * two distinct variants for codon 1 position 1 + * second one has clinical significance + */ + SequenceFeature sf1 = new SequenceFeature("sequence_variant", "", 1, 1, + 0f, null); + sf1.setValue("alleles", "A,G"); // GAA -> E + sf1.setValue("ID", "var1.125A>G"); + SequenceFeature sf2 = new SequenceFeature("sequence_variant", "", 1, 1, + 0f, null); + sf2.setValue("alleles", "A,C"); // CAA -> Q + sf2.setValue("ID", "var2"); + sf2.setValue("clinical_significance", "Dodgy"); + SequenceFeature sf3 = new SequenceFeature("sequence_variant", "", 3, 3, + 0f, null); + sf3.setValue("alleles", "A,G"); // synonymous + sf3.setValue("ID", "var3"); + sf3.setValue("clinical_significance", "None"); + SequenceFeature sf4 = new SequenceFeature("sequence_variant", "", 3, 3, + 0f, null); + sf4.setValue("alleles", "A,T"); // AAT -> N + sf4.setValue("ID", "sequence_variant:var4"); // prefix gets stripped off + sf4.setValue("clinical_significance", "Benign"); + SequenceFeature sf5 = new SequenceFeature("sequence_variant", "", 6, 6, + 0f, null); + sf5.setValue("alleles", "T,C"); // synonymous + sf5.setValue("ID", "var5"); + sf5.setValue("clinical_significance", "Bad"); + SequenceFeature sf6 = new SequenceFeature("sequence_variant", "", 8, 8, + 0f, null); + sf6.setValue("alleles", "C,A,G"); // CAC,CGC -> H,R + sf6.setValue("ID", "var6"); + sf6.setValue("clinical_significance", "Good"); + + List codon1Variants = new ArrayList(); + List codon2Variants = new ArrayList(); + List codon3Variants = new ArrayList(); + List codonVariants[] = new ArrayList[3]; + codonVariants[0] = codon1Variants; + codonVariants[1] = codon2Variants; + codonVariants[2] = codon3Variants; + + /* + * compute variants for protein position 1 + */ + codon1Variants.add(new DnaVariant("A", sf1)); + codon1Variants.add(new DnaVariant("A", sf2)); + codon2Variants.add(new DnaVariant("A")); + codon2Variants.add(new DnaVariant("A")); + codon3Variants.add(new DnaVariant("A", sf3)); + codon3Variants.add(new DnaVariant("A", sf4)); + AlignmentUtils.computePeptideVariants(peptide, 1, codonVariants); + + /* + * compute variants for protein position 2 + */ + codon1Variants.clear(); + codon2Variants.clear(); + codon3Variants.clear(); + codon1Variants.add(new DnaVariant("T")); + codon2Variants.add(new DnaVariant("T")); + codon3Variants.add(new DnaVariant("T", sf5)); + AlignmentUtils.computePeptideVariants(peptide, 2, codonVariants); + + /* + * compute variants for protein position 3 + */ + codon1Variants.clear(); + codon2Variants.clear(); + codon3Variants.clear(); + codon1Variants.add(new DnaVariant("C")); + codon2Variants.add(new DnaVariant("C", sf6)); + codon3Variants.add(new DnaVariant("C")); + AlignmentUtils.computePeptideVariants(peptide, 3, codonVariants); + + /* + * verify added sequence features for + * var1 K -> E + * var2 K -> Q + * var4 K -> N + * var6 P -> H + * var6 P -> R + */ + SequenceFeature[] sfs = peptide.getSequenceFeatures(); + assertEquals(5, sfs.length); + SequenceFeature sf = sfs[0]; + assertEquals(1, sf.getBegin()); + assertEquals(1, sf.getEnd()); + assertEquals("p.Lys1Glu", sf.getDescription()); + assertEquals("var1.125A>G", sf.getValue("ID")); + assertNull(sf.getValue("clinical_significance")); + assertEquals("ID=var1.125A>G", sf.getAttributes()); + assertEquals(1, sf.links.size()); + // link to variation is urlencoded + assertEquals( + "p.Lys1Glu var1.125A>G|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var1.125A%3EG", + sf.links.get(0)); + assertEquals("Jalview", sf.getFeatureGroup()); + sf = sfs[1]; + assertEquals(1, sf.getBegin()); + assertEquals(1, sf.getEnd()); + assertEquals("p.Lys1Gln", sf.getDescription()); + assertEquals("var2", sf.getValue("ID")); + assertEquals("Dodgy", sf.getValue("clinical_significance")); + assertEquals("ID=var2;clinical_significance=Dodgy", sf.getAttributes()); + assertEquals(1, sf.links.size()); + assertEquals( + "p.Lys1Gln var2|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var2", + sf.links.get(0)); + assertEquals("Jalview", sf.getFeatureGroup()); + sf = sfs[2]; + assertEquals(1, sf.getBegin()); + assertEquals(1, sf.getEnd()); + assertEquals("p.Lys1Asn", sf.getDescription()); + assertEquals("var4", sf.getValue("ID")); + assertEquals("Benign", sf.getValue("clinical_significance")); + assertEquals("ID=var4;clinical_significance=Benign", sf.getAttributes()); + assertEquals(1, sf.links.size()); + assertEquals( + "p.Lys1Asn var4|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var4", + sf.links.get(0)); + assertEquals("Jalview", sf.getFeatureGroup()); + sf = sfs[3]; + assertEquals(3, sf.getBegin()); + assertEquals(3, sf.getEnd()); + assertEquals("p.Pro3His", sf.getDescription()); + assertEquals("var6", sf.getValue("ID")); + assertEquals("Good", sf.getValue("clinical_significance")); + assertEquals("ID=var6;clinical_significance=Good", sf.getAttributes()); + assertEquals(1, sf.links.size()); + assertEquals( + "p.Pro3His var6|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var6", + sf.links.get(0)); + // var5 generates two distinct protein variant features + assertEquals("Jalview", sf.getFeatureGroup()); + sf = sfs[4]; + assertEquals(3, sf.getBegin()); + assertEquals(3, sf.getEnd()); + assertEquals("p.Pro3Arg", sf.getDescription()); + assertEquals("var6", sf.getValue("ID")); + assertEquals("Good", sf.getValue("clinical_significance")); + assertEquals("ID=var6;clinical_significance=Good", sf.getAttributes()); + assertEquals(1, sf.links.size()); + assertEquals( + "p.Pro3Arg var6|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var6", + sf.links.get(0)); + assertEquals("Jalview", sf.getFeatureGroup()); + } + + /** + * Tests for the method that maps the subset of a dna sequence that has CDS + * (or subtype) feature, with CDS strand = '-' (reverse) + */ + // test turned off as currently findCdsPositions is not strand-dependent + // left in case it comes around again... + @Test(groups = "Functional", enabled = false) + public void testFindCdsPositions_reverseStrand() + { + 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("-"); + ds.addSequenceFeature(sf); + // exon feature should be ignored here + sf = new SequenceFeature("exon", "", 7, 9, 0f, null); + ds.addSequenceFeature(sf); + // CDS for dna 10-12 + 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] } + */ + assertEquals(6, MappingUtils.getLength(ranges)); + assertEquals(2, ranges.size()); + assertEquals(12, ranges.get(0)[0]); + assertEquals(10, ranges.get(0)[1]); + assertEquals(6, ranges.get(1)[0]); + assertEquals(4, ranges.get(1)[1]); + } + + /** + * Tests for the method that maps the subset of a dna sequence that has CDS + * (or subtype) feature - reverse strand case where the start codon is + * incomplete. + */ + @Test(groups = "Functional", enabled = false) + // test turned off as currently findCdsPositions is not strand-dependent + // left in case it comes around again... + public void testFindCdsPositions_reverseStrandThreePrimeIncomplete() + { + 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("-"); + ds.addSequenceFeature(sf); + // CDS for dna 13-15 + sf = new SequenceFeature("CDS_predicted", "", 13, 15, 0f, null); + 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] + */ + assertEquals(6, MappingUtils.getLength(ranges)); + assertEquals(2, ranges.size()); + assertEquals(13, ranges.get(0)[0]); + assertEquals(13, ranges.get(0)[1]); + assertEquals(9, ranges.get(1)[0]); + assertEquals(5, ranges.get(1)[1]); + } + + @Test(groups = "Functional") + public void testAlignAs_alternateTranscriptsUngapped() + { + SequenceI dna1 = new Sequence("dna1", "cccGGGTTTaaa"); + SequenceI dna2 = new Sequence("dna2", "CCCgggtttAAA"); + AlignmentI dna = new Alignment(new SequenceI[] { dna1, dna2 }); + ((Alignment) dna).createDatasetAlignment(); + SequenceI cds1 = new Sequence("cds1", "GGGTTT"); + SequenceI cds2 = new Sequence("cds2", "CCCAAA"); + AlignmentI cds = new Alignment(new SequenceI[] { cds1, cds2 }); + ((Alignment) cds).createDatasetAlignment(); + + AlignedCodonFrame acf = new AlignedCodonFrame(); + MapList map = new MapList(new int[] { 4, 9 }, new int[] { 1, 6 }, 1, 1); + acf.addMap(dna1.getDatasetSequence(), cds1.getDatasetSequence(), map); + map = new MapList(new int[] { 1, 3, 10, 12 }, new int[] { 1, 6 }, 1, 1); + acf.addMap(dna2.getDatasetSequence(), cds2.getDatasetSequence(), map); + + /* + * verify CDS alignment is as: + * cccGGGTTTaaa (cdna) + * CCCgggtttAAA (cdna) + * + * ---GGGTTT--- (cds) + * CCC------AAA (cds) + */ + dna.addCodonFrame(acf); + AlignmentUtils.alignAs(cds, dna); + assertEquals("---GGGTTT", cds.getSequenceAt(0).getSequenceAsString()); + assertEquals("CCC------AAA", cds.getSequenceAt(1).getSequenceAsString()); + } + + @Test(groups = { "Functional" }) + public void testAddMappedPositions() + { + SequenceI from = new Sequence("dna", "ggAA-ATcc-TT-g"); + SequenceI seq1 = new Sequence("cds", "AAATTT"); + from.createDatasetSequence(); + seq1.createDatasetSequence(); + Mapping mapping = new Mapping(seq1, new MapList( + 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 + */ + assertEquals(6, map.size()); + assertEquals('A', map.get(3).get(seq1).charValue()); + assertEquals('A', map.get(4).get(seq1).charValue()); + assertEquals('A', map.get(6).get(seq1).charValue()); + assertEquals('T', map.get(7).get(seq1).charValue()); + assertEquals('T', map.get(11).get(seq1).charValue()); + assertEquals('T', map.get(12).get(seq1).charValue()); + + /* + * + */ + } + + /** + * Test case where the mapping 'from' range includes a stop codon which is + * absent in the 'to' range + */ + @Test(groups = { "Functional" }) + public void testAddMappedPositions_withStopCodon() + { + SequenceI from = new Sequence("dna", "ggAA-ATcc-TT-g"); + SequenceI seq1 = new Sequence("cds", "AAATTT"); + from.createDatasetSequence(); + seq1.createDatasetSequence(); + Mapping mapping = new Mapping(seq1, new MapList( + 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 + */ + assertEquals(6, map.size()); + assertEquals('A', map.get(3).get(seq1).charValue()); + assertEquals('A', map.get(4).get(seq1).charValue()); + assertEquals('A', map.get(6).get(seq1).charValue()); + assertEquals('T', map.get(7).get(seq1).charValue()); + assertEquals('T', map.get(11).get(seq1).charValue()); + assertEquals('T', map.get(12).get(seq1).charValue()); + } }