import java.util.LinkedHashMap;
import java.util.List;
import java.util.Map;
-import java.util.StringTokenizer;
import java.util.Vector;
import com.stevesoft.pat.Regex;
private static final Regex CLOSE_PAREN = new Regex("(>|\\])", ")");
- private static final Regex DETECT_BRACKETS = new Regex(
- "(<|>|\\[|\\]|\\(|\\))");
+ public static final Regex DETECT_BRACKETS = new Regex(
+ "(<|>|\\[|\\]|\\(|\\)|\\{|\\})");
StringBuffer out; // output buffer
// add alignment annotation for this feature
String key = type2id(type);
+
+ /*
+ * have we added annotation rows for this type ?
+ */
+ boolean annotsAdded = false;
if (key != null)
{
if (accAnnotations != null
Vector vv = (Vector) accAnnotations.get(key);
for (int ii = 0; ii < vv.size(); ii++)
{
+ annotsAdded = true;
AlignmentAnnotation an = (AlignmentAnnotation) vv
.elementAt(ii);
seqO.addAlignmentAnnotation(an);
while (j.hasMoreElements())
{
String desc = j.nextElement().toString();
+ if ("annotations".equals(desc) && annotsAdded)
+ {
+ // don't add features if we already added an annotation row
+ continue;
+ }
String ns = content.get(desc).toString();
char[] byChar = ns.toCharArray();
for (int k = 0; k < byChar.length; k++)
{
String acc = s.stringMatched(1);
String type = s.stringMatched(2);
- String seq = new String(s.stringMatched(3));
- String description = null;
- // Check for additional information about the current annotation
- // We use a simple string tokenizer here for speed
- StringTokenizer sep = new StringTokenizer(seq, " \t");
- description = sep.nextToken();
- if (sep.hasMoreTokens())
- {
- seq = sep.nextToken();
- }
- else
- {
- seq = description;
- description = new String();
- }
- // sequence id with from-to fields
+ String oseq = s.stringMatched(3);
+ /*
+ * copy of annotation field that may be processed into whitespace chunks
+ */
+ String seq = new String(oseq);
Hashtable ann;
// Get an object with all the annotations for this sequence
ann = new Hashtable();
seqAnn.put(acc, ann);
}
+
+ // // start of block for appending annotation lines for wrapped
+ // stokchholm file
// TODO test structure, call parseAnnotationRow with vector from
// hashtable for specific sequence
+
Hashtable features;
// Get an object with all the content for an annotation
if (ann.containsKey("features"))
content = new Hashtable();
features.put(this.id2type(type), content);
}
- String ns = (String) content.get(description);
+ String ns = (String) content.get("annotation");
+
if (ns == null)
{
ns = "";
}
+ // finally, append the annotation line
ns += seq;
- content.put(description, ns);
+ content.put("annotation", ns);
+ // // end of wrapped annotation block.
+ // // Now a new row is created with the current set of data
- // if(type.equals("SS")){
Hashtable strucAnn;
if (seqAnn.containsKey(acc))
{
{
alan.visible = false;
}
- // annotations.addAll(newStruc);
+ // new annotation overwrites any existing annotation...
+
strucAnn.put(type, newStruc);
seqAnn.put(acc, strucAnn);
}
if (DETECT_BRACKETS.search(pos))
{
ann.secondaryStructure = Rna.getRNASecStrucState(pos).charAt(0);
+ ann.displayCharacter = "" + pos.charAt(0);
}
else
{
ann.secondaryStructure = ResidueProperties.getDssp3state(pos)
.charAt(0);
- }
if (ann.secondaryStructure == pos.charAt(0))
{
{
ann.displayCharacter = " " + ann.displayCharacter;
}
+ }
}
}
// output annotations
while (i < s.length && s[i] != null)
{
- if (s[i].getDatasetSequence() != null)
+ AlignmentAnnotation[] alAnot = s[i].getAnnotation();
+ if (alAnot != null)
{
- SequenceI ds = s[i].getDatasetSequence();
- AlignmentAnnotation[] alAnot;
Annotation[] ann;
- Annotation annot;
- alAnot = s[i].getAnnotation();
- String feature = "";
- if (alAnot != null)
+ for (int j = 0; j < alAnot.length; j++)
{
- for (int j = 0; j < alAnot.length; j++)
- {
- String key = type2id(alAnot[j].label);
- boolean isrna = alAnot[j].isValidStruc();
+ String key = type2id(alAnot[j].label);
+ boolean isrna = alAnot[j].isValidStruc();
- if (isrna)
- {
- // hardwire to secondary structure if there is RNA secondary
- // structure on the annotation
- key = "SS";
- }
- if (key == null)
- {
+ if (isrna)
+ {
+ // hardwire to secondary structure if there is RNA secondary
+ // structure on the annotation
+ key = "SS";
+ }
+ if (key == null)
+ {
- continue;
- }
+ continue;
+ }
- // out.append("#=GR ");
- out.append(new Format("%-" + maxid + "s").form("#=GR "
- + printId(s[i], jvSuffix) + " " + key + " "));
- ann = alAnot[j].annotations;
- String seq = "";
- for (int k = 0; k < ann.length; k++)
- {
- seq += outputCharacter(key, k, isrna, ann, s[i]);
- }
- out.append(seq);
- out.append(newline);
+ // out.append("#=GR ");
+ out.append(new Format("%-" + maxid + "s").form("#=GR "
+ + printId(s[i], jvSuffix) + " " + key + " "));
+ ann = alAnot[j].annotations;
+ String seq = "";
+ for (int k = 0; k < ann.length; k++)
+ {
+ seq += outputCharacter(key, k, isrna, ann, s[i]);
}
+ out.append(seq);
+ out.append(newline);
}
}
{
if (annot == null)
{
- // sensible gap character if one is available or make one up
- return sequenceI == null ? '-' : sequenceI.getCharAt(k);
+ // sensible gap character
+ return ' ';
}
else
{
import jalview.gui.JvOptionPane;
import java.io.File;
+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;
public static void testAlignmentEquivalence(AlignmentI al,
AlignmentI al_input, boolean ignoreFeatures)
{
+ testAlignmentEquivalence(al, al_input, ignoreFeatures, false, false);
}
/**
- * assert alignment equivalence
+ * assert alignment equivalence - uses special comparators for RNA structure
+ * annotation rows.
*
* @param al
* 'original'
* @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 ignoreRowVisibility, boolean allowNullAnnotation)
{
assertNotNull("Original alignment was null", al);
assertNotNull("Generated alignment was null", al_input);
{
assertEqualSecondaryStructure(
"Different alignment annotation at position " + i,
- aa_original[i], aa_new[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);
annot_new = al_input.getSequenceAt(in).getAnnotation()[j];
assertEqualSecondaryStructure(
"Different annotation elements", annot_original,
- annot_new);
+ annot_new, allowNullAnnotation);
}
}
}
}
}
+ /**
+ * 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 void assertEqualSecondaryStructure(String message,
- AlignmentAnnotation annot_or,
- AlignmentAnnotation annot_new)
+ 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 (isRna)
{
if (an_or.secondaryStructure != an_new.secondaryStructure
- || an_or.value != an_new.value)
+ || ((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()
else
{
// not RNA secondary structure, so expect all elements to match...
- if (!an_or.displayCharacter.trim().equals(
+ 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())
.trim().length() == 0)
|| (an_new.description == null && an_or.description
.trim().length() == 0) || an_or.description
- .trim().equals(an_new.description.trim()))))
+ .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()
}
else
{
- fail("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()));
+ 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()));
+ }
+ }
+ }
+
+ /**
+ * @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);
+
}
}