X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fext%2Fensembl%2FEnsemblCds.java;h=2086eba7ad667f1a2cb0557a7ddaaa0bab43e65d;hb=a7c519749ec05f55e11494c0d60df6e1af670a98;hp=e366569791ac6d13008d330f022ad038905d7956;hpb=ecc50d775a514f9840cdcc10a9d5b1e49c60582c;p=jalview.git

diff --git a/src/jalview/ext/ensembl/EnsemblCds.java b/src/jalview/ext/ensembl/EnsemblCds.java
index e366569..2086eba 100644
--- a/src/jalview/ext/ensembl/EnsemblCds.java
+++ b/src/jalview/ext/ensembl/EnsemblCds.java
@@ -1,25 +1,51 @@
 package jalview.ext.ensembl;
 
 import jalview.datamodel.SequenceFeature;
-import jalview.io.gff.SequenceOntology;
+import jalview.datamodel.SequenceI;
+import jalview.io.gff.SequenceOntologyFactory;
+import jalview.io.gff.SequenceOntologyI;
 
+import java.util.ArrayList;
+import java.util.List;
+
+/**
+ * A client for direct fetching of CDS sequences from Ensembl (i.e. that part of
+ * the genomic sequence that is translated to protein)
+ * 
+ * TODO: not currently used as CDS sequences are computed from CDS features on
+ * transcripts - delete this class?
+ * 
+ * @author gmcarstairs
+ *
+ */
 public class EnsemblCds extends EnsemblSeqProxy
 {
   /*
    * fetch cds features on genomic sequence (to identify the CDS regions)
-   * and variation features (to retain)
+   * and exon and variation features (to retain for display)
    */
   private static final EnsemblFeatureType[] FEATURES_TO_FETCH = {
-      EnsemblFeatureType.cds, EnsemblFeatureType.variation };
+      EnsemblFeatureType.cds, EnsemblFeatureType.exon,
+      EnsemblFeatureType.variation };
 
   /**
-   * Constructor
+   * Default constructor (to use rest.ensembl.org)
    */
   public EnsemblCds()
   {
     super();
   }
 
+  /**
+   * Constructor given the target domain to fetch data from
+   * 
+   * @param d
+   */
+  public EnsemblCds(String d)
+  {
+    super(d);
+  }
+
   @Override
   public String getDbName()
   {
@@ -45,9 +71,14 @@ public class EnsemblCds extends EnsemblSeqProxy
    * itself.
    */
   @Override
-  protected boolean retainFeature(String type)
+  protected boolean retainFeature(SequenceFeature sf, String accessionId)
   {
-    return !SequenceOntology.getInstance().isA(type, SequenceOntology.CDS);
+    if (SequenceOntologyFactory.getInstance().isA(sf.getType(),
+            SequenceOntologyI.CDS))
+    {
+      return false;
+    }
+    return featureMayBelong(sf, accessionId);
   }
 
   /**
@@ -58,8 +89,8 @@ public class EnsemblCds extends EnsemblSeqProxy
   @Override
   protected boolean identifiesSequence(SequenceFeature sf, String accId)
   {
-    if (SequenceOntology.getInstance().isA(sf.getType(),
-            SequenceOntology.CDS))
+    if (SequenceOntologyFactory.getInstance().isA(sf.getType(),
+            SequenceOntologyI.CDS))
     {
       String parentFeature = (String) sf.getValue(PARENT);
       if (("transcript:" + accId).equals(parentFeature))
@@ -70,4 +101,18 @@ public class EnsemblCds extends EnsemblSeqProxy
     return false;
   }
 
+  /**
+   * Overrides this method to trivially return a range which is the whole of the
+   * nucleotide sequence. This is both faster than scanning for CDS features,
+   * and also means we don't need to keep CDS features on CDS sequence (where
+   * they are redundant information).
+   */
+  protected List<int[]> getCdsRanges(SequenceI dnaSeq)
+  {
+    int len = dnaSeq.getLength();
+    List<int[]> ranges = new ArrayList<int[]>();
+    ranges.add(new int[] { 1, len });
+    return ranges;
+  }
+
 }