A benchmark study of k-mer counting methods for high-throughput sequencing

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• High-throughput sequencing generates massive datasets (hundreds of gigabytes).
• Bioinformatics tools often require counting k-mers (substrings of length k) in DNA/RNA sequencing reads.
• Applications include genome assembly, error correction, and repeat detection.

Purpose of the Study:

• To assess and compare the performance of various k-mer counting programs.
• To evaluate trade-offs between time and memory usage for k-mer counting.
• To provide recommendations for current state-of-the-art k-mer counting tool setup.

Main Methods:

• Evaluated multiple k-mer counting programs.
• Measured performance based on runtime and memory usage.
• Assessed additional parameters: disk usage, accuracy, parallelism, compressed input, large k-value performance, and scalability.

Main Results:

• Identified performance trade-offs among different k-mer counting techniques.
• Detailed analysis of runtime, memory, and other key performance indicators.
• Comparative evaluation of tools for handling large sequencing datasets.

Conclusions:

• Specific recommendations for setting up optimal k-mer counting tools.
• Suggestions for future development directions in k-mer counting software.
• Guidance for researchers selecting k-mer counting methods for large-scale sequencing projects.