+ long ntCount = 0; // nucleotide symbol count (does not include ntaCount)
+ long aaCount = 0; // non-nucleotide, non-gap symbol count (includes nCount
+ // and ntaCount)
+ long nCount = 0; // "Unknown" (N) symbol count
+ long xCount = 0; // Also used as "Unknown" (X) symbol count
+ long ntaCount = 0; // nucleotide ambiguity symbol count
+
+ int len = seq.getLength();
+ for (int i = 0; i < len; i++)
+ {
+ char c = seq.getCharAt(i);
+ if (isNucleotide(c))
+ {
+ ntCount++;
+ }
+ else if (!isGap(c))
+ {
+ aaCount++;
+ if (isN(c))
+ {
+ nCount++;
+ }
+ else
+ {
+ if (isX(c))
+ {
+ xCount++;
+ }
+ if (isNucleotideAmbiguity(c))
+ {
+ ntaCount++;
+ }
+ }
+ }
+ }
+ long allCount = ntCount + aaCount;
+
+ if (NUCLEOTIDE_AMBIGUITY_DETECTION)
+ {
+ Console.debug("Performing new nucleotide detection routine");
+ if (allCount > NUCLEOTIDE_COUNT_SHORT_SEQUENCE)
+ {
+ // a long sequence.
+ // check for at least 55% nucleotide, and nucleotide and ambiguity codes
+ // (including N) must make up 95%
+ return ntCount * 100 >= NUCLEOTIDE_COUNT_PERCENT * allCount
+ && 100 * (ntCount + nCount
+ + ntaCount) >= NUCLEOTIDE_COUNT_LONG_SEQUENCE_AMBIGUITY_PERCENT
+ * allCount;
+ }
+ else if (allCount > NUCLEOTIDE_COUNT_VERY_SHORT_SEQUENCE)
+ {
+ // a short sequence.
+ // check if a short sequence is at least 55% nucleotide and the rest of
+ // the symbols are all X or all N
+ if (ntCount * 100 >= NUCLEOTIDE_COUNT_PERCENT * allCount
+ && (nCount == aaCount || xCount == aaCount))
+ {
+ return true;
+ }
+
+ // a short sequence.
+ // check for at least x% nucleotide and all the rest nucleotide
+ // ambiguity codes (including N), where x slides from 75% for sequences
+ // of length 4 (i.e. only one non-nucleotide) to 55% for sequences of
+ // length 100
+ return myShortSequenceNucleotideProportionCount(ntCount, allCount)
+ && nCount + ntaCount == aaCount;
+ }
+ else
+ {
+ // a very short sequence. (<4)
+ // all bases must be nucleotide
+ return ntCount > 0 && ntCount == allCount;
+ }
+ }
+ else