import static org.testng.AssertJUnit.assertEquals;
import static org.testng.AssertJUnit.assertNotNull;
+import static org.testng.AssertJUnit.assertNull;
import static org.testng.AssertJUnit.assertTrue;
import static org.testng.AssertJUnit.fail;
+import jalview.datamodel.Alignment;
import jalview.datamodel.AlignmentAnnotation;
import jalview.datamodel.AlignmentI;
import jalview.datamodel.Annotation;
+import jalview.datamodel.DBRefEntry;
+import jalview.datamodel.Sequence;
import jalview.datamodel.SequenceFeature;
import jalview.datamodel.SequenceI;
import jalview.gui.JvOptionPane;
import java.io.File;
+import java.io.IOException;
+import java.util.Arrays;
import java.util.BitSet;
import java.util.HashMap;
+import java.util.List;
import java.util.Map;
+import org.testng.Assert;
import org.testng.annotations.BeforeClass;
import org.testng.annotations.Test;
@Test(groups = { "Functional" })
public void pfamFileIO() throws Exception
{
- testFileIOwithFormat(new File(PfamFile), "STH", -1, 0);
+ testFileIOwithFormat(new File(PfamFile), FileFormat.Stockholm, -1, 0,
+ false, false, false);
}
@Test(groups = { "Functional" })
public void pfamFileDataExtraction() throws Exception
{
AppletFormatAdapter af = new AppletFormatAdapter();
- AlignmentI al = af.readFile(PfamFile, af.FILE,
- new IdentifyFile().identify(PfamFile, af.FILE));
+ AlignmentI al = af.readFile(PfamFile, DataSourceType.FILE,
+ new IdentifyFile().identify(PfamFile, DataSourceType.FILE));
int numpdb = 0;
for (SequenceI sq : al.getSequences())
{
@Test(groups = { "Functional" })
public void rfamFileIO() throws Exception
{
- testFileIOwithFormat(new File(RfamFile), "STH", 2, 1);
+ testFileIOwithFormat(new File(RfamFile), FileFormat.Stockholm, 2, 1,
+ false, false, false);
}
/**
* @param ioformat
* - label for IO class used to write and read back in the data from
* f
+ * @param ignoreFeatures
+ * @param ignoreRowVisibility
+ * @param allowNullAnnotations
*/
- public static void testFileIOwithFormat(File f, String ioformat,
- int naliannot, int nminseqann)
+ public static void testFileIOwithFormat(File f, FileFormatI ioformat,
+ int naliannot, int nminseqann, boolean ignoreFeatures,
+ boolean ignoreRowVisibility, boolean allowNullAnnotations)
{
System.out.println("Reading file: " + f);
String ff = f.getPath();
{
AppletFormatAdapter rf = new AppletFormatAdapter();
- AlignmentI al = rf.readFile(ff, AppletFormatAdapter.FILE,
- new IdentifyFile().identify(ff, AppletFormatAdapter.FILE));
+ AlignmentI al = rf.readFile(ff, DataSourceType.FILE,
+ new IdentifyFile().identify(ff, DataSourceType.FILE));
assertNotNull("Couldn't read supplied alignment data.", al);
+ outputfile + "\n<<EOF\n");
// test for consistency in io
AlignmentI al_input = new AppletFormatAdapter().readFile(outputfile,
- AppletFormatAdapter.PASTE, ioformat);
+ DataSourceType.PASTE, ioformat);
assertNotNull("Couldn't parse reimported alignment data.", al_input);
- String identifyoutput = new IdentifyFile().identify(outputfile,
- AppletFormatAdapter.PASTE);
+ FileFormatI identifyoutput = new IdentifyFile().identify(outputfile,
+ DataSourceType.PASTE);
assertNotNull("Identify routine failed for outputformat " + ioformat,
identifyoutput);
assertTrue(
"Identify routine could not recognise output generated by '"
+ ioformat + "' writer",
ioformat.equals(identifyoutput));
- testAlignmentEquivalence(al, al_input, false);
+ testAlignmentEquivalence(al, al_input, ignoreFeatures,
+ ignoreRowVisibility, allowNullAnnotations);
int numaliannot = 0, numsqswithali = 0;
for (AlignmentAnnotation ala : al_input.getAlignmentAnnotation())
{
public static void testAlignmentEquivalence(AlignmentI al,
AlignmentI al_input, boolean ignoreFeatures)
{
+ testAlignmentEquivalence(al, al_input, ignoreFeatures, false, false);
+ }
+
+ /**
+ * assert alignment equivalence - uses special comparators for RNA structure
+ * annotation rows.
+ *
+ * @param al
+ * 'original'
+ * @param al_input
+ * 'secondary' or generated alignment from some datapreserving
+ * transformation
+ * @param ignoreFeatures
+ * when true, differences in sequence feature annotation are ignored
+ *
+ * @param ignoreRowVisibility
+ * when true, do not fail if there are differences in the visibility
+ * of annotation rows
+ * @param allowNullAnnotation
+ * when true, positions in alignment annotation that are null will be
+ * considered equal to positions containing annotation where
+ * Annotation.isWhitespace() returns true.
+ *
+ */
+ public static void testAlignmentEquivalence(AlignmentI al,
+ AlignmentI al_input, boolean ignoreFeatures,
+ boolean ignoreRowVisibility, boolean allowNullAnnotation)
+ {
assertNotNull("Original alignment was null", al);
assertNotNull("Generated alignment was null", al_input);
// we might want to revise this in future
int aa_new_size = (aa_new == null ? 0 : aa_new.length);
int aa_original_size = (aa_original == null ? 0 : aa_original.length);
- Map<Integer, BitSet> orig_groups = new HashMap<Integer, BitSet>();
- Map<Integer, BitSet> new_groups = new HashMap<Integer, BitSet>();
+ Map<Integer, BitSet> orig_groups = new HashMap<>();
+ Map<Integer, BitSet> new_groups = new HashMap<>();
if (aa_new != null && aa_original != null)
{
{
if (aa_new.length > i)
{
- assertTrue("Different alignment annotation at position " + i,
- equalss(aa_original[i], aa_new[i]));
+ assertEqualSecondaryStructure(
+ "Different alignment annotation at position " + i,
+ aa_original[i], aa_new[i], allowNullAnnotation);
// compare graphGroup or graph properties - needed to verify JAL-1299
assertEquals("Graph type not identical.", aa_original[i].graph,
aa_new[i].graph);
- assertEquals("Visibility not identical.", aa_original[i].visible,
+ if (!ignoreRowVisibility)
+ {
+ assertEquals("Visibility not identical.",
+ aa_original[i].visible,
aa_new[i].visible);
+ }
assertEquals("Threshold line not identical.",
aa_original[i].threshold, aa_new[i].threshold);
// graphGroup may differ, but pattern should be the same
seq_original = al.getSequencesArray();
SequenceI[] seq_new = new SequenceI[al_input.getSequencesArray().length];
seq_new = al_input.getSequencesArray();
- SequenceFeature[] sequenceFeatures_original, sequenceFeatures_new;
+ List<SequenceFeature> sequenceFeatures_original;
+ List<SequenceFeature> sequenceFeatures_new;
AlignmentAnnotation annot_original, annot_new;
//
for (int i = 0; i < al.getSequencesArray().length; i++)
&& seq_new[in].getSequenceFeatures() != null)
{
System.out.println("There are feature!!!");
- sequenceFeatures_original = new SequenceFeature[seq_original[i]
- .getSequenceFeatures().length];
sequenceFeatures_original = seq_original[i]
.getSequenceFeatures();
- sequenceFeatures_new = new SequenceFeature[seq_new[in]
- .getSequenceFeatures().length];
sequenceFeatures_new = seq_new[in].getSequenceFeatures();
- assertEquals("different number of features",
- seq_original[i].getSequenceFeatures().length,
- seq_new[in].getSequenceFeatures().length);
+ assertEquals("different number of features", seq_original[i]
+ .getSequenceFeatures().size(), seq_new[in]
+ .getSequenceFeatures().size());
- for (int feat = 0; feat < seq_original[i].getSequenceFeatures().length; feat++)
+ for (int feat = 0; feat < seq_original[i].getSequenceFeatures()
+ .size(); feat++)
{
assertEquals("Different features",
- sequenceFeatures_original[feat],
- sequenceFeatures_new[feat]);
+ sequenceFeatures_original.get(feat),
+ sequenceFeatures_new.get(feat));
}
}
// compare alignment annotation
{
annot_original = al.getSequenceAt(i).getAnnotation()[j];
annot_new = al_input.getSequenceAt(in).getAnnotation()[j];
- assertTrue("Different annotation elements",
- equalss(annot_original, annot_new));
+ assertEqualSecondaryStructure(
+ "Different annotation elements", annot_original,
+ annot_new, allowNullAnnotation);
}
}
}
}
}
- /*
- * compare annotations
+ /**
+ * compare two annotation rows, with special support for secondary structure
+ * comparison. With RNA, only the value and the secondaryStructure symbols are
+ * compared, displayCharacter and description are ignored. Annotations where
+ * Annotation.isWhitespace() is true are always considered equal.
+ *
+ * @param message
+ * - not actually used yet..
+ * @param annot_or
+ * - the original annotation
+ * @param annot_new
+ * - the one compared to the original annotation
+ * @param allowNullEquivalence
+ * when true, positions in alignment annotation that are null will be
+ * considered equal to non-null positions for which
+ * Annotation.isWhitespace() is true.
*/
- private static boolean equalss(AlignmentAnnotation annot_or,
- AlignmentAnnotation annot_new)
+ private static void assertEqualSecondaryStructure(String message,
+ AlignmentAnnotation annot_or, AlignmentAnnotation annot_new,
+ boolean allowNullEqivalence)
{
+ // TODO: test to cover this assert behaves correctly for all allowed
+ // variations of secondary structure annotation row equivalence
if (annot_or.annotations.length != annot_new.annotations.length)
{
- System.err.println("Different lengths for annotation row elements: "
+ fail("Different lengths for annotation row elements: "
+ annot_or.annotations.length + "!="
+ annot_new.annotations.length);
- return false;
}
+ boolean isRna = annot_or.isRNA();
+ assertTrue("Expected " + (isRna ? " valid RNA " : " no RNA ")
+ + " secondary structure in the row.",
+ isRna == annot_new.isRNA());
for (int i = 0; i < annot_or.annotations.length; i++)
{
Annotation an_or = annot_or.annotations[i], an_new = annot_new.annotations[i];
if (an_or != null && an_new != null)
{
- if (!an_or.displayCharacter.trim().equals(
- an_new.displayCharacter.trim())
- || !("" + an_or.secondaryStructure).trim().equals(
- ("" + an_new.secondaryStructure).trim())
- || (an_or.description != an_new.description && !((an_or.description == null && an_new.description
- .trim().length() == 0)
- || (an_new.description == null && an_or.description
- .trim().length() == 0) || an_or.description
- .trim().equals(an_new.description.trim()))))
+
+ if (isRna)
+ {
+ if (an_or.secondaryStructure != an_new.secondaryStructure
+ || ((Float.isNaN(an_or.value) != Float
+ .isNaN(an_new.value)) || an_or.value != an_new.value))
+ {
+ fail("Different RNA secondary structure at column " + i
+ + " expected: [" + annot_or.annotations[i].toString()
+ + "] but got: [" + annot_new.annotations[i].toString()
+ + "]");
+ }
+ }
+ else
{
- System.err.println("Annotation Element Mismatch\nElement " + i
- + " in original: " + annot_or.annotations[i].toString()
- + "\nElement " + i + " in new: "
- + annot_new.annotations[i].toString());
- return false;
+ // not RNA secondary structure, so expect all elements to match...
+ if ((an_or.isWhitespace() != an_new.isWhitespace())
+ || !an_or.displayCharacter.trim().equals(
+ an_new.displayCharacter.trim())
+ || !("" + an_or.secondaryStructure).trim().equals(
+ ("" + an_new.secondaryStructure).trim())
+ || (an_or.description != an_new.description && !((an_or.description == null && an_new.description
+ .trim().length() == 0)
+ || (an_new.description == null && an_or.description
+ .trim().length() == 0) || an_or.description
+ .trim().equals(an_new.description.trim())))
+ || !((Float.isNaN(an_or.value) && Float
+ .isNaN(an_new.value)) || an_or.value == an_new.value))
+ {
+ fail("Annotation Element Mismatch\nElement " + i
+ + " in original: " + annot_or.annotations[i].toString()
+ + "\nElement " + i + " in new: "
+ + annot_new.annotations[i].toString());
+ }
}
}
else if (annot_or.annotations[i] == null
}
else
{
- System.err.println("Annotation Element Mismatch\nElement "
- + i
- + " in original: "
- + (annot_or.annotations[i] == null ? "is null"
- : annot_or.annotations[i].toString())
- + "\nElement "
- + i
- + " in new: "
- + (annot_new.annotations[i] == null ? "is null"
- : annot_new.annotations[i].toString()));
- return false;
+ if (allowNullEqivalence)
+ {
+ if (an_or != null && an_or.isWhitespace())
+
+ {
+ continue;
+ }
+ if (an_new != null && an_new.isWhitespace())
+ {
+ continue;
+ }
+ }
+ // need also to test for null in one, non-SS annotation in other...
+ fail("Annotation Element Mismatch\nElement " + i + " in original: "
+ + (an_or == null ? "is null" : an_or.toString())
+ + "\nElement " + i + " in new: "
+ + (an_new == null ? "is null" : an_new.toString()));
}
}
- return true;
+ }
+
+ /**
+ * @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);
}
}