+ }
+
+ /**
+ * @see assertEqualSecondaryStructure - test if two secondary structure
+ * annotations are not equal
+ * @param message
+ * @param an_orig
+ * @param an_new
+ * @param allowNullEquivalence
+ */
+ public static void assertNotEqualSecondaryStructure(String message,
+ AlignmentAnnotation an_orig, AlignmentAnnotation an_new,
+ boolean allowNullEquivalence)
+ {
+ boolean thrown = false;
+ try
+ {
+ assertEqualSecondaryStructure("", an_orig, an_new,
+ allowNullEquivalence);
+ } catch (AssertionError af)
+ {
+ thrown = true;
+ }
+ if (!thrown)
+ {
+ fail("Expected difference for [" + an_orig + "] and [" + an_new + "]");
+ }
+ }
+ private AlignmentAnnotation makeAnnot(Annotation ae)
+ {
+ return new AlignmentAnnotation("label", "description", new Annotation[]
+ { ae });
+ }
+
+ @Test(groups={"Functional"})
+ public void testAnnotationEquivalence()
+ {
+ AlignmentAnnotation one = makeAnnot(new Annotation("", "", ' ', 1));
+ AlignmentAnnotation anotherOne = makeAnnot(new Annotation("", "", ' ',
+ 1));
+ AlignmentAnnotation sheet = makeAnnot(new Annotation("","",'E',0f));
+ AlignmentAnnotation anotherSheet = makeAnnot(new Annotation("","",'E',0f));
+ AlignmentAnnotation sheetWithLabel = makeAnnot(new Annotation("1", "",
+ 'E', 0f));
+ AlignmentAnnotation anotherSheetWithLabel = makeAnnot(new Annotation(
+ "1", "", 'E', 0f));
+ AlignmentAnnotation rnaNoDC = makeAnnot(new Annotation("","",'<',0f));
+ AlignmentAnnotation anotherRnaNoDC = makeAnnot(new Annotation("","",'<',0f));
+ AlignmentAnnotation rnaWithDC = makeAnnot(new Annotation("B", "", '<',
+ 0f));
+ AlignmentAnnotation anotherRnaWithDC = makeAnnot(new Annotation("B",
+ "", '<', 0f));
+
+ // check self equivalence
+ for (boolean allowNull : new boolean[] { true, false })
+ {
+ assertEqualSecondaryStructure("Should be equal", one, anotherOne,
+ allowNull);
+ assertEqualSecondaryStructure("Should be equal", sheet, anotherSheet,
+ allowNull);
+ assertEqualSecondaryStructure("Should be equal", sheetWithLabel,
+ anotherSheetWithLabel, allowNull);
+ assertEqualSecondaryStructure("Should be equal", rnaNoDC,
+ anotherRnaNoDC, allowNull);
+ assertEqualSecondaryStructure("Should be equal", rnaWithDC,
+ anotherRnaWithDC, allowNull);
+ // display character doesn't matter for RNA structure (for 2.10.2)
+ assertEqualSecondaryStructure("Should be equal", rnaWithDC, rnaNoDC,
+ allowNull);
+ assertEqualSecondaryStructure("Should be equal", rnaNoDC, rnaWithDC,
+ allowNull);
+ }
+
+ // verify others are different
+ List<AlignmentAnnotation> aaSet = Arrays.asList(one, sheet,
+ sheetWithLabel, rnaWithDC);
+ for (int p = 0; p < aaSet.size(); p++)
+ {
+ for (int q = 0; q < aaSet.size(); q++)
+ {
+ if (p != q)
+ {
+ assertNotEqualSecondaryStructure("Should be different",
+ aaSet.get(p), aaSet.get(q), false);
+ }
+ else
+ {
+ assertEqualSecondaryStructure("Should be same", aaSet.get(p),
+ aaSet.get(q), false);
+ assertEqualSecondaryStructure("Should be same", aaSet.get(p),
+ aaSet.get(q), true);
+ assertNotEqualSecondaryStructure(
+ "Should be different to empty anot", aaSet.get(p),
+ makeAnnot(Annotation.EMPTY_ANNOTATION), false);
+ assertNotEqualSecondaryStructure(
+ "Should be different to empty annot",
+ makeAnnot(Annotation.EMPTY_ANNOTATION), aaSet.get(q),
+ true);
+ assertNotEqualSecondaryStructure("Should be different to null",
+ aaSet.get(p), makeAnnot(null), false);
+ assertNotEqualSecondaryStructure("Should be different to null",
+ makeAnnot(null), aaSet.get(q), true);
+ }
+ }
+ }
+
+ // test null
+
+ }
+
+ String aliFile = ">Dm\nAAACCCUUUUACACACGGGAAAGGG";
+ String annFile = "JALVIEW_ANNOTATION\n# Created: Thu May 04 11:16:52 BST 2017\n\n"
+ + "SEQUENCE_REF\tDm\nNO_GRAPH\tsecondary structure\tsecondary structure\t"
+ + "(|(|(|(|, .|, .|, .|, .|)|)|)|)|\t0.0\nROWPROPERTIES\t"
+ + "secondary structure\tscaletofit=true\tshowalllabs=true\tcentrelabs=false";
+
+ String annFileCurlyWuss = "JALVIEW_ANNOTATION\n# Created: Thu May 04 11:16:52 BST 2017\n\n"
+ + "SEQUENCE_REF\tDm\nNO_GRAPH\tsecondary structure\tsecondary structure\t"
+ + "(|(|(|(||{|{||{|{||)|)|)|)||}|}|}|}|\t0.0\nROWPROPERTIES\t"
+ + "secondary structure\tscaletofit=true\tshowalllabs=true\tcentrelabs=false";
+ String annFileFullWuss = "JALVIEW_ANNOTATION\n# Created: Thu May 04 11:16:52 BST 2017\n\n"
+ + "SEQUENCE_REF\tDm\nNO_GRAPH\tsecondary structure\tsecondary structure\t"
+ + "(|(|(|(||{|{||[|[||)|)|)|)||}|}|]|]|\t0.0\nROWPROPERTIES\t"
+ + "secondary structure\tscaletofit=true\tshowalllabs=true\tcentrelabs=false";
+
+ @Test(groups = { "Functional" })
+ public void secondaryStructureForRNASequence() throws Exception
+ {
+ roundTripSSForRNA(aliFile, annFile);
+ }
+
+ @Test(groups = { "Functional" })
+ public void curlyWUSSsecondaryStructureForRNASequence() throws Exception
+ {
+ roundTripSSForRNA(aliFile, annFileCurlyWuss);
+ }
+
+ @Test(groups = { "Functional" })
+ public void fullWUSSsecondaryStructureForRNASequence() throws Exception
+ {
+ roundTripSSForRNA(aliFile, annFileFullWuss);
+ }
+
+ @Test(groups = { "Functional" })
+ public void detectWussBrackets()
+ {
+ for (char ch : new char[] { '{', '}', '[', ']', '(', ')', '<', '>' })
+ {
+ Assert.assertTrue(StockholmFile.DETECT_BRACKETS.matchAt("" + ch, 0),
+ "Didn't recognise " + ch + " as a WUSS bracket");
+ }
+ for (char ch : new char[] { '@', '!', 'V', 'Q', '*', ' ', '-', '.' })
+ {
+ Assert.assertFalse(StockholmFile.DETECT_BRACKETS.matchAt("" + ch, 0),
+ "Shouldn't recognise " + ch + " as a WUSS bracket");
+ }
+ }
+ private static void roundTripSSForRNA(String aliFile, String annFile)
+ throws Exception
+ {
+ AlignmentI al = new AppletFormatAdapter().readFile(aliFile,
+ DataSourceType.PASTE, jalview.io.FileFormat.Fasta);
+ AnnotationFile aaf = new AnnotationFile();
+ aaf.readAnnotationFile(al, annFile, DataSourceType.PASTE);
+ al.getAlignmentAnnotation()[0].visible = true;
+
+ // TODO: create a better 'save as <format>' pattern
+ StockholmFile sf = new StockholmFile(al);
+
+ String stockholmFile = sf.print(al.getSequencesArray(), true);
+
+ AlignmentI newAl = new AppletFormatAdapter().readFile(stockholmFile,
+ DataSourceType.PASTE, jalview.io.FileFormat.Stockholm);
+ // AlignmentUtils.showOrHideSequenceAnnotations(newAl.getViewport()
+ // .getAlignment(), Arrays.asList("Secondary Structure"), newAl
+ // .getViewport().getAlignment().getSequences(), true, true);
+ testAlignmentEquivalence(al, newAl, true, true, true);
+
+ }
+
+ @Test(groups = { "Functional" })
+ public void testTypeToDescription()
+ {
+ assertEquals("Secondary Structure",
+ StockholmFile.typeToDescription("SS"));
+ assertEquals("Surface Accessibility",
+ StockholmFile.typeToDescription("SA"));
+ assertEquals("transmembrane", StockholmFile.typeToDescription("TM"));
+ assertEquals("Posterior Probability",
+ StockholmFile.typeToDescription("PP"));
+ assertEquals("ligand binding", StockholmFile.typeToDescription("LI"));
+ assertEquals("active site", StockholmFile.typeToDescription("AS"));
+ assertEquals("intron", StockholmFile.typeToDescription("IN"));
+ assertEquals("interacting residue",
+ StockholmFile.typeToDescription("IR"));
+ assertEquals("accession", StockholmFile.typeToDescription("AC"));
+ assertEquals("organism", StockholmFile.typeToDescription("OS"));
+ assertEquals("class", StockholmFile.typeToDescription("CL"));
+ assertEquals("description", StockholmFile.typeToDescription("DE"));
+ assertEquals("reference", StockholmFile.typeToDescription("DR"));
+ assertEquals("look", StockholmFile.typeToDescription("LO"));
+ assertEquals("Reference Positions",
+ StockholmFile.typeToDescription("RF"));
+
+ // case-sensitive:
+ assertEquals("Rf", StockholmFile.typeToDescription("Rf"));
+ assertEquals("junk", StockholmFile.typeToDescription("junk"));
+ assertEquals("", StockholmFile.typeToDescription(""));
+ assertNull(StockholmFile.typeToDescription(null));
+ }
+
+ @Test(groups = { "Functional" })
+ public void testDescriptionToType()
+ {
+ assertEquals("SS",
+ StockholmFile.descriptionToType("Secondary Structure"));
+ assertEquals("SA",
+ StockholmFile.descriptionToType("Surface Accessibility"));
+ assertEquals("TM", StockholmFile.descriptionToType("transmembrane"));
+
+ // test is not case-sensitive:
+ assertEquals("SS",
+ StockholmFile.descriptionToType("secondary structure"));
+
+ // test is white-space sensitive:
+ assertNull(StockholmFile.descriptionToType("secondary structure "));
+
+ assertNull(StockholmFile.descriptionToType("any old junk"));
+ assertNull(StockholmFile.descriptionToType(""));
+ assertNull(StockholmFile.descriptionToType(null));
+ }
+
+ @Test(groups = { "Functional" })
+ public void testPrint()
+ {
+ SequenceI seq1 = new Sequence("seq1", "LKMF-RS-Q");
+ SequenceI seq2 = new Sequence("seq2/10-15", "RRS-LIP-");
+ SequenceI[] seqs = new SequenceI[] { seq1, seq2 };
+ AlignmentI al = new Alignment(seqs);
+
+ StockholmFile testee = new StockholmFile(al);
+
+ /*
+ * basic output (sequences only):
+ * sequence ids are padded with 9 spaces more than the widest id
+ */
+ String output = testee.print(seqs, true);
+ String expected = "# STOCKHOLM 1.0\n" + "seq1/1-7 LKMF-RS-Q\n"
+ + "seq2/10-15 RRS-LIP-\n//\n";
+ assertEquals(expected, output);
+
+ /*
+ * add some dbrefs
+ */
+ seq1.addDBRef(new DBRefEntry("PFAM", "1", "PF00111"));
+ seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "P83527"));
+ seq2.addDBRef(new DBRefEntry("RFAM", "1", "AY119185.1"));
+ seq2.addDBRef(new DBRefEntry("EMBL", "1", "AF125575"));
+ output = testee.print(seqs, true);
+ // PFAM and RFAM dbrefs should be output as AC, others as DR
+ expected = "# STOCKHOLM 1.0\n" + "#=GS seq1/1-7 AC PF00111\n"
+ + "#=GS seq1/1-7 DR UNIPROT ; P83527\n"
+ + "#=GS seq2/10-15 AC AY119185.1\n"
+ + "#=GS seq2/10-15 DR EMBL ; AF125575\n"
+ + "seq1/1-7 LKMF-RS-Q\n"
+ + "seq2/10-15 RRS-LIP-\n//\n";
+ assertEquals(expected, output);
+
+ /*
+ * add some sequence and alignment annotation
+ */
+ Annotation[] anns = new Annotation[5];
+ for (int i = 0; i < anns.length; i++)
+ {
+ anns[i] = new Annotation(String.valueOf((char) ('B' + i)),
+ "Desc " + i,
+ (char) ('C' + i), i + 3);
+ }
+
+ // expect "secondary structure" to be output as #=GR seqid SS
+ // using the secondary structure character (CDEFG) not display char (BCDEF)
+ AlignmentAnnotation aa1 = new AlignmentAnnotation("secondary structure",
+ "ssdesc", anns);
+ aa1.sequenceRef = seq1;
+ seq1.addAlignmentAnnotation(aa1);
+ al.addAnnotation(aa1);
+
+ // "sec structure" should not be output as no corresponding feature id
+ AlignmentAnnotation aa2 = new AlignmentAnnotation("sec structure",
+ "ssdesc", anns);
+ aa2.sequenceRef = seq2;
+ seq2.addAlignmentAnnotation(aa2);
+ al.addAnnotation(aa2);
+
+ // alignment annotation for Reference Positions: output as #=GC RF
+ AlignmentAnnotation aa3 = new AlignmentAnnotation("reference positions",
+ "refpos", anns);
+ al.addAnnotation(aa3);
+
+ // 'seq' annotation: output as seq_cons
+ AlignmentAnnotation aa4 = new AlignmentAnnotation("seq", "seqdesc",
+ anns);
+ al.addAnnotation(aa4);
+
+ // 'intron' annotation: output as IN_cons
+ AlignmentAnnotation aa5 = new AlignmentAnnotation("intron",
+ "introndesc", anns);
+ al.addAnnotation(aa5);
+
+ // 'binding site' annotation: output as binding_site
+ AlignmentAnnotation aa6 = new AlignmentAnnotation("binding site",
+ "bindingdesc", anns);
+ al.addAnnotation(aa6);
+
+ // 'autocalc' annotation should not be output
+ AlignmentAnnotation aa7 = new AlignmentAnnotation("Consensus",
+ "consensusdesc", anns);
+ aa7.autoCalculated = true;
+ al.addAnnotation(aa7);
+
+ // hidden annotation should not be output
+ AlignmentAnnotation aa8 = new AlignmentAnnotation("domains",
+ "domaindesc", anns);
+ aa8.visible = false;
+ al.addAnnotation(aa8);
+
+ output = testee.print(seqs, true);
+ //@formatter:off
+ expected =
+ "# STOCKHOLM 1.0\n"
+ + "#=GS seq1/1-7 AC PF00111\n"
+ + "#=GS seq1/1-7 DR UNIPROT ; P83527\n"
+ + "#=GS seq2/10-15 AC AY119185.1\n"
+ + "#=GS seq2/10-15 DR EMBL ; AF125575\n"
+ + "#=GR seq1/1-7 SS CDEFG\n"
+ + "seq1/1-7 LKMF-RS-Q\n"
+ + "seq2/10-15 RRS-LIP-\n"
+ + "#=GC RF BCDEF\n" + "#=GC seq_cons BCDEF\n"
+ + "#=GC IN_cons BCDEF\n" + "#=GC binding_site BCDEF\n"
+ + "//\n";
+ //@formatter:on
+ assertEquals(expected, output);
+ }
+
+ @Test(groups = "Functional")
+ public void testGetAnnotationCharacter()
+ {
+ SequenceI seq = new Sequence("seq", "abc--def-");
+
+ Annotation[] ann = new Annotation[8];
+ ann[1] = new Annotation("Z", "desc", 'E', 1f);
+ ann[2] = new Annotation("Q", "desc", ' ', 1f);
+ ann[4] = new Annotation("", "desc", 'E', 1f);
+ ann[6] = new Annotation("ZH", "desc", 'E', 1f);
+
+ /*
+ * null annotation in column (not Secondary Structure annotation)
+ * should answer sequence character, or '-' if null sequence
+ */
+ assertEquals('-',
+ StockholmFile.getAnnotationCharacter("RF", 0, ann[0], null));
+ assertEquals('d',
+ StockholmFile.getAnnotationCharacter("RF", 5, ann[5], seq));
+ assertEquals('-',
+ StockholmFile.getAnnotationCharacter("RF", 8, null, seq));
+
+ /*
+ * null annotation in column (SS annotation) should answer underscore
+ */
+ assertEquals('_',
+ StockholmFile.getAnnotationCharacter("SS", 0, ann[0], seq));
+
+ /*
+ * SS secondary structure symbol
+ */
+ assertEquals('E',
+ StockholmFile.getAnnotationCharacter("SS", 1, ann[1], seq));
+
+ /*
+ * no SS symbol, use label instead
+ */
+ assertEquals('Q',
+ StockholmFile.getAnnotationCharacter("SS", 2, ann[2], seq));
+
+ /*
+ * SS with 2 character label - second character overrides SS symbol
+ */
+ assertEquals('H',
+ StockholmFile.getAnnotationCharacter("SS", 6, ann[6], seq));
+
+ /*
+ * empty display character, not SS - answers '.'
+ */
+ assertEquals('.',
+ StockholmFile.getAnnotationCharacter("RF", 4, ann[4], seq));
+ }
+
+ /**
+ * Test to verify that gaps are input/output as underscore in STO annotation
+ *
+ * @throws IOException
+ */
+ @Test(groups = "Functional")
+ public void testRoundtripWithGaps() throws IOException
+ {
+ /*
+ * small extract from RF00031_folded.stk
+ */
+ // @formatter:off
+ String stoData =
+ "# STOCKHOLM 1.0\n" +
+ "#=GR B.taurus.4 SS .._((.))_\n" +
+ "B.taurus.4 AC.UGCGU.\n" +
+ "#=GR B.taurus.5 SS ..((_._))\n" +
+ "B.taurus.5 ACUU.G.CG\n" +
+ "//\n";
+ // @formatter:on
+ StockholmFile parser = new StockholmFile(stoData, DataSourceType.PASTE);
+ SequenceI[] seqs = parser.getSeqsAsArray();
+ assertEquals(2, seqs.length);
+
+ /*
+ * B.taurus.4 has a trailing gap
+ * rendered as underscore in Stockholm annotation
+ */
+ assertEquals("AC.UGCGU.", seqs[0].getSequenceAsString());
+ AlignmentAnnotation[] anns = seqs[0].getAnnotation();
+ assertEquals(1, anns.length);
+ AlignmentAnnotation taurus4SS = anns[0];
+ assertEquals(9, taurus4SS.annotations.length);
+ assertEquals(" .", taurus4SS.annotations[0].displayCharacter);
+ assertNull(taurus4SS.annotations[2]); // gapped position
+ assertNull(taurus4SS.annotations[8]); // gapped position
+ assertEquals('(', taurus4SS.annotations[3].secondaryStructure);
+ assertEquals("(", taurus4SS.annotations[3].displayCharacter);
+ assertEquals(')', taurus4SS.annotations[7].secondaryStructure);
+
+ /*
+ * output as Stockholm and verify it matches the original input
+ * (gaps output as underscore in annotation lines)
+ * note: roundtrip test works with the input lines ordered as above;
+ * can also parse in other orders, but then input doesn't match output
+ */
+ AlignmentFileWriterI afile = FileFormat.Stockholm
+ .getWriter(new Alignment(seqs));
+ String output = afile.print(seqs, false);
+ assertEquals(stoData, output);