Abstract 2706: Ultra-Low Input High-Fidelity (ULI-HiFi) long-reads uncover variants in genomic dark matter from pre-cancer polyp and tumor samples

Abstract

Abstract Long-read sequencing technologies have been constrained by high error rates and large sample input requirements. The recent introduction of HiFi sequencing has enabled highly accurate detection of single-nucleotide variants (SNVs) from long reads (10-15 kb). However, a major limiting factor is the amount of sample input required, where current protocols require ≥15 micrograms of DNA. Here we report Ultra-Low Input High Fidelity Sequencing (ULI-HiFi), an amplification-based library preparation that reduces this input to 10 nanograms of DNA, a more than 10,000-fold reduction in sample input. When the ultra-low input method was tested using a benchmark genome, HG002, we observed high precision and improved recall of structural variants compared to a ground truth dataset from the Genome in a Bottle Consortium. We also benchmarked small variants and reported the accurate analysis of many variants in clinically important genes that are unmappable with short reads. We used this strategy to sequence tumor and polyp samples from a patient with familial adenomatous polyposis, a hereditary form of colon cancer, which revealed SNVs and structural variants not detected by short-read techniques. This strategy will enable the improved characterization of genetic variants from limited clinical samples. Citation Format: Hayan Lee, Graham Erwin, Aaron Horning, Raushun Kirtikar, Emmett Griffin-Baldwin, William Rowell, Philip Li, Sarah Kingan, Michael P. Snyder. Ultra-Low Input High-Fidelity (ULI-HiFi) long-reads uncover variants in genomic dark matter from pre-cancer polyp and tumor samples [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2706.