X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Futil%2FComparison.java;h=581d9e83dd4f0cde37bb4ff2ddc1ef4cdc3a9383;hb=808f36e6bb443068330ec5c19b89e2dc1c9b0bd0;hp=835a1b4c58c20712fbc58eddb22c08e6239a64d8;hpb=17fe82c9d7f9da1b17f3091e4b3992a35e3e411e;p=jalview.git

diff --git a/src/jalview/util/Comparison.java b/src/jalview/util/Comparison.java
index 835a1b4..581d9e8 100644
--- a/src/jalview/util/Comparison.java
+++ b/src/jalview/util/Comparison.java
@@ -20,11 +20,11 @@
  */
 package jalview.util;
 
+import jalview.datamodel.SequenceI;
+
 import java.util.ArrayList;
 import java.util.List;
 
-import jalview.datamodel.SequenceI;
-
 /**
  * Assorted methods for analysing or comparing sequences.
  */
@@ -40,8 +40,8 @@ public class Comparison
 
   private static final char GAP_DASH = '-';
 
-  public static final String GapChars = new String(new char[]
-  { GAP_SPACE, GAP_DOT, GAP_DASH });
+  public static final String GapChars = new String(new char[] { GAP_SPACE,
+      GAP_DOT, GAP_DASH });
 
   /**
    * DOCUMENT ME!
@@ -245,7 +245,7 @@ public class Comparison
    */
   public static final boolean isGap(char c)
   {
-    return (c == GAP_DASH || c == GAP_DOT || c == GAP_SPACE) ? true : false;
+    return (c == GAP_DASH || c == GAP_DOT || c == GAP_SPACE);
   }
 
   /**
@@ -262,9 +262,35 @@ public class Comparison
     {
       return false;
     }
+    char[][] letters = new char[seqs.length][];
+    for (int i = 0; i < seqs.length; i++)
+    {
+      if (seqs[i] != null)
+      {
+        char[] sequence = seqs[i].getSequence();
+        if (sequence != null)
+        {
+          letters[i] = sequence;
+        }
+      }
+    }
+
+    return areNucleotide(letters);
+  }
+
+  /**
+   * Answers true if more than 85% of the sequence residues (ignoring gaps) are
+   * A, G, C, T or U, else false. This is just a heuristic guess and may give a
+   * wrong answer (as AGCT are also amino acid codes).
+   * 
+   * @param letters
+   * @return
+   */
+  public static final boolean areNucleotide(char[][] letters)
+  {
     int ntCount = 0;
     int aaCount = 0;
-    for (SequenceI seq : seqs)
+    for (char[] seq : letters)
     {
       if (seq == null)
       {
@@ -272,7 +298,7 @@ public class Comparison
       }
       // TODO could possibly make an informed guess just from the first sequence
       // to save a lengthy calculation
-      for (char c : seq.getSequence())
+      for (char c : seq)
       {
         if ('a' <= c && c <= 'z')
         {
@@ -320,11 +346,13 @@ public class Comparison
     List<SequenceI> flattened = new ArrayList<SequenceI>();
     for (SequenceI[] ss : seqs)
     {
-      for (SequenceI s : ss) {
-      flattened.add(s);
+      for (SequenceI s : ss)
+      {
+        flattened.add(s);
       }
     }
-    final SequenceI[] oneDArray = flattened.toArray(new SequenceI[flattened.size()]);
+    final SequenceI[] oneDArray = flattened.toArray(new SequenceI[flattened
+            .size()]);
     return isNucleotide(oneDArray);
   }
 }