Hybrid capture is a critical technology for selective enrichment of genomic regions of interest in genomic analysis. Despite its widespread adoption, the core methodology has remained largely unchanged for over 15 years, with traditional workflows involving time-consuming bead-based capture steps, multiple temperature-controlled washes, and post-hybridization PCR. These steps introduce workflow complexity, increase turnaround time, and can negatively impact library complexity and variant calling accuracy. We present a simplified hybrid capture workflow that eliminates these complexities by directly loading the hybridization product onto the sequencing flow cell. The approach is enabled by the development of a streptavidin flow cell surface, a method to circularize and amplify captured targets on the flow cell, and a fast hybridization protocol. Our workflow reduces the time from the start of library preparation to the start of sequencing by over 50% while maintaining or improving capture specificity and library complexity. We demonstrate improved variant calling performance with indel false positive and false negative reductions of 89% and 67%, respectively. We also show how the approach can be used to create an entirely PCR-free targeted sequencing workflow. We present a targeted sequencing workflow that eliminates bead-based capture, multiple washes, and post-hybridization PCR, while improving various aspects of data quality. The performance of the approach was evaluated by sequencing hundreds of samples and demonstrating high on-target rates, reduced duplicates, and improved indel accuracy. By combining the approach with a PCR-free library preparation, we enable an entirely PCR-free targeted sequencing assay which further improves indel calling, and shows the ability to call an HTT expansion, associated with Huntington's disease. This streamlined approach addresses key operational challenges in targeted sequencing, offering potential benefits for applications requiring rapid turnaround times or increased capability in variant detection.
Mah et al. (Tue,) studied this question.