A bio-sequence k-mer frequency counter (kFC)

The high-throughput sequencing data from next generation sequencing technologies demands the need for over presented k-mers for being de novo sequenced and assembled. Even though generating k-mer frequency distribution of a sequence seems to be a simple task, memory usage and time are two important concerns especially for higher order mers. This paper proposes a method to find count of over represented k-mers in bio-sequences. The approach uses a hash table with open address scheme for estimating the frequency count of k-mers. The algorithm support both overlapping and non-overlapping pattern for nucleotide sequences, amino acid sequences and Next generation sequencing read sequences. Moreover, it even accept nucleotide sequences from the extended alphabet set in contrast to the traditional k-mer tool which accepts only the standard alphabet.