Updated with latest from mchmmer branch
[jalview.git] / test / jalview / io / StockholmFileTest.java
index 4273e6c..1c9c79d 100644 (file)
@@ -22,17 +22,22 @@ package jalview.io;
 
 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;
@@ -230,8 +235,8 @@ public class StockholmFileTest
     // 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)
     {
@@ -654,4 +659,273 @@ public class StockholmFileTest
     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);
+  }
 }