GradHC: highly reliable gradual hash-based clustering for DNA storage systems.

Area of Science:

• Biotechnology
• Bioinformatics
• Data Science

Background:

• Growing data storage demands necessitate novel solutions beyond traditional methods.
• Synthetic DNA offers exceptional density and durability for long-term data archiving.
• Current DNA data storage faces challenges with read errors and accurate data reconstruction.

Purpose of the Study:

• To address the critical task of clustering DNA reads in DNA storage systems.
• To introduce and evaluate a novel clustering algorithm for DNA data storage.

Main Methods:

• Reviewed existing methods for evaluating clustering algorithms.
• Developed and implemented Gradual Hash-based clustering (GradHC).
• Benchmarked GradHC against other clustering algorithms for DNA storage.

Main Results:

• GradHC demonstrates high accuracy in clustering diverse DNA designs, including varying strand lengths and cluster sizes.
• The algorithm effectively handles different error ranges inherent in DNA sequencing.
• Benchmark analysis shows GradHC is more stable and robust than prior algorithms.

Conclusions:

• GradHC provides a reliable and robust solution for clustering DNA reads in DNA storage systems.
• The algorithm's performance across various conditions makes it suitable for practical DNA data storage applications.