HGSA 2025

Background: Diagnosing hereditary conditions like Lynch syndrome remains challenging despite advances in clinical genomics. A major hurdle is distinguishing pathogenic variants in the PMS2 from its highly homologous pseudogene, PMS2CL, which complicates accurate diagnosis. Aim: This study aims to leverage long read nanopore sequencing to overcome these diagnostic barriers and improve the detection of complex genomic variants associated with Lynch syndrome. Methods: Using Oxford Nanopore Technologies’ PromethION P2 platform with adaptive sampling, we analysed target genes in 40 genomic DNA clinical samples. The approach enabled comprehensive detection of single nucleotide variants, structural variants, copy number changes, and epigenetic modifications. Specific targets included PMS2/PMS2CL, EPCAM deletions, and MLH1 promoter methylation. Results: Pathogenic PMS2 variants distinct from PMS2CL were validated in 25 cases, resolving variants that would previously have been classified as uncertain. Notable findings include structural variants in exon 11, exon 12, exon 14, and large deletions spanning exon 11-15 of PMS2 and adjacent genes. Screening for MLH1 promoter methylation ruled out constitutional methylation in non-familial cases. Additionally, intergenic deletions between EPCAM and MSH2 were identified, highlighting their impact on MSH2 promoter methylation and the utility of long read sequencing in these complex scenarios. Conclusion: By resolving pseudogene interference and analysing repetitive regions, long read sequencing represents a significant advancement in Lynch syndrome diagnostics. Its adoption in clinical workflows could reduce diagnostic uncertainty, enhance treatment pathways, and improve patient outcomes, offering a novel and transformative approach to hereditary cancer diagnostics.

Background: Short read sequencing is the dominant technology in clinical genomics and continues to improve in detecting complex variant types such as structural variants and repeat expansions. However, it provides limited information about a variant’s genomic context and may not be clinically accredited across all variants. Orthogonal testing is frequently required to confirm or fully resolve variants and achieve diagnoses. These often-bespoke assays can be expensive and labour intensive. Long read sequencing offers a promising alternative. Aim: To explore the utility of nanopore long read sequencing with adaptive sampling to resolve complex diagnostic cases. Methods: We applied long read sequencing with adaptive sampling to 30 diagnostic cases with variants including structural variants, repeat expansions, high homology regions, transposon insertions, imprinting disorders, and phasing of de novo and inherited variants in biallelic disease genes. Results: Long read sequencing confirmed or clarified the variants in over 90% of cases, resulting in high confidence clinical diagnosis. In one case, both short and long-read data were required to show a de novo variant in TTN was in trans with an inherited pathogenic variant 150kb downstream. Variant phasing was reported within 8 days, resulting in a time and cost-efficient solution. Conclusions: Long read sequencing with adaptive sampling is a fast and cost-effective approach to orthogonally confirm a range of complex clinically relevant variants. It has the potential to replace many traditional custom assays while resolving previously unaddressed challenges such as long-distance phasing of de novo and inherited variants in recessive genes and variant-specific imprinting analysis.

The hereditary spastic-ataxia spectrum disorders are a group of disabling neurological diseases. The current standard clinical genetic testing pathway is often prolonged, multistep and many cases remain unsolved. With the aim of streamlining the diagnostic pathway, and improving diagnostic rate, we developed a strategy using Oxford Nanopore Technologies targeted long-read sequencing, with the capacity to characterise genetic variation of all variant types within 469 spastic-ataxia-associated genes. We applied this approach to a cohort of 34 individuals with unsolved spastic-ataxia, as well as five positive controls.

We identified pathogenic variants, that would be sufficient for genetic diagnosis, in 14 unsolved individuals (41%). SCA27B, caused by a short tandem repeat expansion in FGF14 was the most common diagnosis, present in 7 individuals (21%). Two individuals (6%) had biallelic pathogenic expansions in RFC1, and one individual had a monoallelic pathogenic expansion in ATXN8OS/ATXN8. Causative pathogenic sequence variants were found in four individuals (involving VCP, STUB1, ANO10 and SPG7). Furthermore, all positive controls were successfully identified.

Our results demonstrate the utility of ONT targeted long-read sequencing in the genetic evaluation of patients with spastic-ataxia spectrum disorders and highlight the capacity to increase overall diagnostic yield, and streamline the testing pathway by permitting evaluation of all relevant genes and gene variant types in a single assay.