X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fanalysis%2FAAFrequency.java;h=b1505d6827ce8401af83960a821cc6d4b3a1a5ea;hb=da11f432f67f80f1848d75f862495e5e9dd9076b;hp=e4f2dfa01a9e41cc2b945dbe5c6e573e85ed0818;hpb=be762d8d9c71a7aa3121e845c45911c7192b7827;p=jalview.git

diff --git a/src/jalview/analysis/AAFrequency.java b/src/jalview/analysis/AAFrequency.java
index e4f2dfa..b1505d6 100755
--- a/src/jalview/analysis/AAFrequency.java
+++ b/src/jalview/analysis/AAFrequency.java
@@ -398,7 +398,7 @@ public class AAFrequency
    * contains
    * 
    * <pre>
-   *    [profileType, numberOfValues, nonGapCount, charValue1, percentage1, charValue2, percentage2, ...]
+   *    [profileType, numberOfValues, totalPercent, charValue1, percentage1, charValue2, percentage2, ...]
    * in descending order of percentage value
    * </pre>
    * 
@@ -411,7 +411,6 @@ public class AAFrequency
    */
   public static int[] extractProfile(ProfileI profile, boolean ignoreGaps)
   {
-    int[] rtnval = new int[64];
     ResidueCount counts = profile.getCounts();
     if (counts == null)
     {
@@ -422,7 +421,6 @@ public class AAFrequency
     char[] symbols = symbolCounts.symbols;
     int[] values = symbolCounts.values;
     QuickSort.sort(values, symbols);
-    int nextArrayPos = 2;
     int totalPercentage = 0;
     final int divisor = ignoreGaps ? profile.getNonGapped()
             : profile.getHeight();
@@ -430,21 +428,44 @@ public class AAFrequency
     /*
      * traverse the arrays in reverse order (highest counts first)
      */
+    int[] result = new int[3 + 2 * symbols.length];
+    int nextArrayPos = 3;
+    int nonZeroCount = 0;
+
     for (int i = symbols.length - 1; i >= 0; i--)
     {
       int theChar = symbols[i];
       int charCount = values[i];
-
-      rtnval[nextArrayPos++] = theChar;
       final int percentage = (charCount * 100) / divisor;
-      rtnval[nextArrayPos++] = percentage;
+      if (percentage == 0)
+      {
+        /*
+         * this count (and any remaining) round down to 0% - discard
+         */
+        break;
+      }
+      nonZeroCount++;
+      result[nextArrayPos++] = theChar;
+      result[nextArrayPos++] = percentage;
       totalPercentage += percentage;
     }
-    rtnval[0] = symbols.length;
-    rtnval[1] = totalPercentage;
-    int[] result = new int[rtnval.length + 1];
+
+    /*
+     * truncate array if any zero values were discarded
+     */
+    if (nonZeroCount < symbols.length)
+    {
+      int[] tmp = new int[3 + 2 * nonZeroCount];
+      System.arraycopy(result, 0, tmp, 0, tmp.length);
+      result = tmp;
+    }
+
+    /*
+     * fill in 'header' values
+     */
     result[0] = AlignmentAnnotation.SEQUENCE_PROFILE;
-    System.arraycopy(rtnval, 0, result, 1, rtnval.length);
+    result[1] = nonZeroCount;
+    result[2] = totalPercentage;
 
     return result;
   }
@@ -454,15 +475,15 @@ public class AAFrequency
    * contains
    * 
    * <pre>
-   *    [profileType, numberOfValues, totalCount, charValue1, percentage1, charValue2, percentage2, ...]
+   *    [profileType, numberOfValues, totalPercentage, charValue1, percentage1, charValue2, percentage2, ...]
    * in descending order of percentage value, where the character values encode codon triplets
    * </pre>
    * 
    * @param hashtable
    * @return
    */
-  public static int[] extractCdnaProfile(Hashtable hashtable,
-          boolean ignoreGaps)
+  public static int[] extractCdnaProfile(
+          Hashtable<String, Object> hashtable, boolean ignoreGaps)
   {
     // this holds #seqs, #ungapped, and then codon count, indexed by encoded
     // codon triplet
@@ -492,9 +513,16 @@ public class AAFrequency
       {
         break; // nothing else of interest here
       }
+      final int percentage = codonCount * 100 / divisor;
+      if (percentage == 0)
+      {
+        /*
+         * this (and any remaining) values rounded down to 0 - discard
+         */
+        break;
+      }
       distinctValuesCount++;
       result[j++] = codons[i];
-      final int percentage = codonCount * 100 / divisor;
       result[j++] = percentage;
       totalPercentage += percentage;
     }
@@ -518,7 +546,7 @@ public class AAFrequency
    *          the consensus data stores to be populated (one per column)
    */
   public static void calculateCdna(AlignmentI alignment,
-          Hashtable[] hconsensus)
+          Hashtable<String, Object>[] hconsensus)
   {
     final char gapCharacter = alignment.getGapCharacter();
     List<AlignedCodonFrame> mappings = alignment.getCodonFrames();
@@ -531,7 +559,7 @@ public class AAFrequency
     for (int col = 0; col < cols; col++)
     {
       // todo would prefer a Java bean for consensus data
-      Hashtable<String, int[]> columnHash = new Hashtable<String, int[]>();
+      Hashtable<String, Object> columnHash = new Hashtable<>();
       // #seqs, #ungapped seqs, counts indexed by (codon encoded + 1)
       int[] codonCounts = new int[66];
       codonCounts[0] = alignment.getSequences().size();
@@ -551,6 +579,7 @@ public class AAFrequency
           {
             codonCounts[codonEncoded + 2]++;
             ungappedCount++;
+            break;
           }
         }
       }
@@ -576,7 +605,8 @@ public class AAFrequency
    */
   public static void completeCdnaConsensus(
           AlignmentAnnotation consensusAnnotation,
-          Hashtable[] consensusData, boolean showProfileLogo, int nseqs)
+          Hashtable<String, Object>[] consensusData,
+          boolean showProfileLogo, int nseqs)
   {
     if (consensusAnnotation == null
             || consensusAnnotation.annotations == null
@@ -591,7 +621,7 @@ public class AAFrequency
     consensusAnnotation.scaleColLabel = true;
     for (int col = 0; col < consensusData.length; col++)
     {
-      Hashtable hci = consensusData[col];
+      Hashtable<String, Object> hci = consensusData[col];
       if (hci == null)
       {
         // gapped protein column?