OReO: optimizing read order for practical compression.

Area of Science:

• Genomics
• Bioinformatics
• Data Science

Background:

• High-throughput and third-generation sequencing generate massive datasets, posing storage and management challenges.
• Existing specialized compression tools often require dedicated software, hindering widespread adoption.
• Generic compressors (e.g., gzip) are widely used for convenience but offer suboptimal compression ratios for sequencing data.

Purpose of the Study:

• To introduce Optimizing Read Order (OReO), a novel read-reordering framework for efficient sequencing data compression.
• To achieve high compression performance comparable to state-of-the-art tools without specialized decompression software.
• To provide a user-friendly and sustainable solution for managing large sequencing datasets.

Main Methods:

• Developed OReO, a framework that groups overlapping reads before applying generic compression algorithms.
• Evaluated OReO on diverse Oxford Nanopore Technologies (ONT) and HiFi genomic and metagenomic datasets.
• Compared OReO's compression performance, resource usage, and decompression speed against existing methods.

Main Results:

• OReO achieved substantial compression gains across various datasets.
• Compression ratios were comparable to specialized state-of-the-art tools.
• OReO demonstrated faster decompression speeds than dedicated methods with similar resource usage.

Conclusions:

• Read reordering is a viable strategy to enhance the effectiveness of generic compression tools for sequencing data.
• OReO offers an efficient, sustainable, and user-friendly solution to sequencing data storage challenges.
• The open-source OReO framework removes barriers to adoption for researchers and bioinformaticians.