KCOSS: an ultra-fast k-mer counter for assembled genome analysis

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• K-mer frequency analysis is crucial for understanding genomic complexity, comparative genomics, metagenomics, and phylogeny.
• Existing k-mer counting tools often suffer from slow performance and high resource requirements (memory, disk space).

Purpose of the Study:

• To introduce KCOSS, a novel and highly efficient algorithm for k-mer counting.
• To address the limitations of current k-mer counting tools in terms of speed and resource consumption.

Main Methods:

• KCOSS utilizes a segmented Bloom filter, lock-free queue, lock-free thread pool, and cuckoo hash table.
• Optimization strategies include memory block recycling, merging k-mers into C-reads, and asynchronous disk writing.

Main Results:

• KCOSS demonstrated superior performance compared to Jellyfish2, CHTKC, and KMC3 across seven assembled genomes and three sequencing datasets.
• The algorithm achieved shorter running times with reduced memory and disk space requirements for assembled genomes.

Conclusions:

• KCOSS offers a significant advancement in k-mer counting efficiency for both assembled genomes and raw sequencing data.
• The software is freely available, facilitating broader adoption in genomic research.