Benchmarking and optimisation of bait-capture metagenomics for sequencing of respiratory viruses at scale

Abstract Background Sequencing respiratory virus genomes is essential for public health surveillance and research. Although shotgun metagenomics is pathogen agnostic, its sensitivity is limited by abundant off-target host nucleic acids. Hybridization bait capture overcomes this limitation by selectively enriching viral sequences prior to sequencing. Methods We evaluated three respiratory viral bait capture workflows (veSEQ, RVI-seq, and Illumina) to compare their performance and assess their suitability for scalable respiratory virus sequencing. Synthetic RNA controls and clinical samples containing SARS-CoV-2, influenza A, influenza B, human parainfluenza virus, and respiratory syncytial virus (RSV) were analysed. Results All three workflows demonstrated high efficiency, reproducibility and broadly comparable performance across respiratory viruses. Complete viral genomes were consistently recovered from samples containing 10,000 viral copies, while viral reads remained detectable at substantially lower viral loads, including approximately 100 copies. Workflow optimisation reduced reagent costs and enabled laboratory automation without compromising sequencing sensitivity. Conclusion Hybridization bait capture provides an effective and scalable approach for respiratory virus genome sequencing. Cost reductions and automation can be implemented without compromising performance, supporting its use in routine genomic surveillance and public health preparedness.