FastqCA: an effective FASTQ compressor through 2D spatial redundancy reduction

High-throughput sequencing generates vast amounts of FASTQ data, posing significant storage and transmission challenges. We propose FastqCA, a reference-free, cellular automaton-based compressor designed to exploit two-dimensional spatial redundancy across reads and positions. FastqCA applies a cellular automaton-based predictive-modeling algorithm to the nucleotide sequence stream in both modes, and to the Q4-quantised quality score stream in lossy mode. Evaluation against representative FASTQ compressors, including FQZcomp, DSRC2, fastqz, Scalce, Quip, NAF and SPRING, shows that FastqCA achieves competitive or improved compression ratios on the benchmarked datasets. Genome assembly analysis using QUAST confirms that data compressed by FastqCA maintains assembly quality comparable to the original data across key metrics such as N50 and total length. By effectively utilizing spatial context, FastqCA provides an efficient solution for large-scale sequencing data archival and transmission. FastqCA can be downloaded from https://github.com/XXhaos/FastqCA and freely available for non-commercial usage.