Oxford Nanopore Biopharma Day, South San Francisco

Sequencing antibody libraries requires single-nucleotide resolution due to the importance of variants affecting biinding and functional characteristics. Existing sequencing technologies either didn't provide enough length, depth, or accuracy relating to these libraries. Using recently published methods, ONT sequencing can be used to generate high throughput ultra-accurate reads of antibody variants via RCA and consensus calling.

PRINT is a powerful platform for site-specific transgene integration. We aimed to characterize the specificity of PRINT in clinically relevant Primary Human Hepatocytes, free of bias introduced through amplification jackpotting and dropout. Furthermore, the retrotransposon used in PRINT, R2, inserts into the rDNA. This long and repetitive region, with small duplications throughout the genome and complicated regulation by DNA methylation, makes short read sequencing an ineffective option to accurately map reads to their genomic site. Here, we use long-read sequencing technologies, including Oxford Nanopore Technologies’ MinION and PromethION, to investigate insertion specificity, rDNA methylation, and transgene integrity. We also developed highly sensitive alignment strategies to accurately characterize PRINT mediated transgene integration.

Generating genetically engineered mouse models requires robust validation strategies capable of resolving edits ranging from single nucleotide variants to large knock-in constructs. Creation of these models begins with pronuclear injection, the direct microinjection of nuclease components and donor DNA into the pronucleus of a fertilized oocyte, producing founder animals that must be rigorously screened prior to colony expansion. Confirming edit integrity, zygosity, and transgene architecture of these larger donors require methods that go beyond what conventional short-read and Sanger-based approaches can reliably deliver. Limited read length, poor resolution of structural complexity, and indirect zygosity inference presented recurring bottlenecks as the scale and complexity of our engineering work grew. To address these limitations, we developed a long-read nanopore sequencing framework for in-house model characterization and validation. Droplet digital PCR serves as the initial screening step across all samples, providing copy number estimates that guide downstream sequencing decisions. Across all edit sizes, successful donor incorporation relies on homology-directed repair to precisely insert donor sequence at the target locus. For edits under a certain size, targeted amplicon sequencing via native barcoding library preparation provides rapid and cost-effective founder screening. For larger and more complex knock-in models, droplet digital PCR results inform the selection of samples taken forward for whole genome sequencing. Workflow development focused on high molecular weight DNA extraction, library preparation optimization, sequencing depth strategies, and analysis approaches capable of resolving full transgene structure, detecting concatemerization, and identifying integration site context. This framework has meaningfully improved founder selection confidence, reduced validation timelines, and enabled more informed decisions prior to colony expansion.