SPRING: a next-generation compressor for FASTQ data.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• High-throughput sequencing (HTS) generates massive genomic datasets.
• Standard compression tools struggle with the inherent redundancy in FASTQ files.
• Existing FASTQ compressors often lack crucial features like variable read length support or scalability.

Purpose of the Study:

• Introduce SPRING, a new reference-free compressor for FASTQ files.
• Address limitations of current FASTQ compression tools.
• Provide a scalable and feature-rich compression solution for genomic data.

Main Methods:

• Developed SPRING, a novel compression algorithm for FASTQ data.
• Implemented features including lossless compression, pairing preservation, and lossy quality value compression.
• Evaluated performance on large-scale whole-genome sequencing datasets.

Main Results:

• SPRING achieves significantly better compression ratios compared to state-of-the-art tools.
• Compressed 195 GB of human whole-genome data to under 7 GB (1.6x improvement).
• Maintained comparable computational resource usage while offering advanced features.

Conclusions:

• SPRING offers a substantial advancement in FASTQ data compression.
• The tool supports diverse compression needs, including lossy and lossless modes.
• SPRING provides an efficient and effective solution for managing large genomic datasets.