Targeted adaptive sampling for pharmacogenomics and genome-wide variant analysis

Biography

Dr Pamela Gan is the Bioinformatics Team Lead at NalaGenetics, whose current primary research focus is developing bioinformatic pipelines and applications for diagnostics and personalized medicine. She holds a PhD from the Australian National University and has prior experience at RIKEN and Genesis Healthcare in Japan. In this presentation, Pamela will present a novel workflow for reporting pharmacogenes using targeted long-read sequencing.

Abstract

More than 90% of people carry genetic variants affecting drug metabolism, increasing the risk of adverse drug reactions (ADRs). Pre-emptive pharmacogenomic (PGx) testing can reduce ADRs by up to 30%, improving patient safety and lowering healthcare costs. However, current methods, such as short-read sequencing and microarrays, face issues resolving complex structural variants in key pharmacogenes like CYP2D6. Long-read sequencing offers a promising alternative to address these limitations. We developed a PGx workflow using targeted adaptive sampling long-read sequencing (TAS-LRS), a cost-effective alternative to whole-genome sequencing. Our pipeline integrates existing tools with custom algorithms, including a new CYP2D6 caller, and utilizes the off-target sequences for genome-wide genotyping. Validation on reference cell lines and clinical samples was conducted to assess its performance. The TAS-LRS workflow showed improved accuracy in PGx diplotype detection, including an increase in concordance for CYP2D6 compared to other methods. For DPYD, SLCO1B1, CYP2B6, UGT1A1 and CYP2D6, the pipeline improved calls made by established methods (microarrays, short-read sequencing) demonstrating the advantage of long reads in phasing haplotypes. Additionally, genome-wide genotyping based on imputation using the off-target sequence outperformed microarray-based imputation. This study highlights the potential of TAS-LRS for pharmacogenomic testing and shows proof-of-concept for its clinical utility in pharmacogenomic reporting.