+ break; // nothing else of interest here
+ }
+ distinctValuesCount++;
+ result[j++] = codons[i];
+ final int percentage = codonCount * 100 / divisor;
+ result[j++] = percentage;
+ totalPercentage += percentage;
+ }
+ result[2] = totalPercentage;
+
+ /*
+ * Just return the non-zero values
+ */
+ // todo next value is redundant if we limit the array to non-zero counts
+ result[1] = distinctValuesCount;
+ return Arrays.copyOfRange(result, 0, j);
+ }
+
+ /**
+ * Compute a consensus for the cDNA coding for a protein alignment.
+ *
+ * @param alignment
+ * the protein alignment (which should hold mappings to cDNA
+ * sequences)
+ * @param hconsensus
+ * the consensus data stores to be populated (one per column)
+ */
+ public static void calculateCdna(AlignmentI alignment,
+ Hashtable[] hconsensus)
+ {
+ final char gapCharacter = alignment.getGapCharacter();
+ List<AlignedCodonFrame> mappings = alignment.getCodonFrames();
+ if (mappings == null || mappings.isEmpty())
+ {
+ return;
+ }
+
+ int cols = alignment.getWidth();
+ for (int col = 0; col < cols; col++)
+ {
+ // todo would prefer a Java bean for consensus data
+ Hashtable<String, int[]> columnHash = new Hashtable<String, int[]>();
+ // #seqs, #ungapped seqs, counts indexed by (codon encoded + 1)
+ int[] codonCounts = new int[66];
+ codonCounts[0] = alignment.getSequences().size();
+ int ungappedCount = 0;
+ for (SequenceI seq : alignment.getSequences())
+ {
+ if (seq.getCharAt(col) == gapCharacter)
+ {
+ continue;
+ }
+ List<char[]> codons = MappingUtils
+ .findCodonsFor(seq, col, mappings);
+ for (char[] codon : codons)
+ {
+ int codonEncoded = CodingUtils.encodeCodon(codon);
+ if (codonEncoded >= 0)
+ {
+ codonCounts[codonEncoded + 2]++;
+ ungappedCount++;
+ }
+ }
+ }
+ codonCounts[1] = ungappedCount;
+ // todo: sort values here, save counts and codons?
+ columnHash.put(PROFILE, codonCounts);
+ hconsensus[col] = columnHash;
+ }
+ }
+
+ /**
+ * Derive displayable cDNA consensus annotation from computed consensus data.
+ *
+ * @param consensusAnnotation
+ * the annotation row to be populated for display
+ * @param consensusData
+ * the computed consensus data
+ * @param showProfileLogo
+ * if true show all symbols present at each position, else only the
+ * modal value
+ * @param nseqs
+ * the number of sequences in the alignment
+ */
+ public static void completeCdnaConsensus(
+ AlignmentAnnotation consensusAnnotation,
+ Hashtable[] consensusData, boolean showProfileLogo, int nseqs)
+ {
+ if (consensusAnnotation == null
+ || consensusAnnotation.annotations == null
+ || consensusAnnotation.annotations.length < consensusData.length)
+ {
+ // called with a bad alignment annotation row - wait for it to be
+ // initialised properly
+ return;
+ }
+
+ // ensure codon triplet scales with font size
+ consensusAnnotation.scaleColLabel = true;
+ for (int col = 0; col < consensusData.length; col++)
+ {
+ Hashtable hci = consensusData[col];
+ if (hci == null)
+ {
+ // gapped protein column?