Diagnostic discovery of structural variants causing foveal hypoplasia using SVRare and long-read nanopore sequencing

Abstract

Advances in DNA sequencing technology are increasing the rate of molecular diagnosis for patients and families with inherited Mendelian diseases. However, some patients remain unsolved following standard-of-care testing (typically either exome or genome sequencing), with structural variants (SVs) likely to account for a significant proportion of these missed diagnoses. We re-analysed ten research cases with the visual disorder foveal hypoplasia (FH) that were previously unsolved through whole exome sequencing (WES). We used SVRare, a tool that integrates outputs from Canvas and Manta and combines these with allele frequencies and information regarding their genomic context to highlight plausible pathogenic SVs. This analysis solved 2/10 cases. Expanding this strategy, we then used SVRare in a reverse genetics approach to analyse an FH-relevant gene panel in the 100,000 Genomes Project (100KGP) rare disease cohort. This identified potentially pathogenic SVs as the cause of disease in a further 11 previously unsolved cases with phenotypes overlapping FH. In total, 11 SVs in five genes were identified as the likely cause of FH in 13 patients: SLC38A8 (1 patient), PAX6 (3), OCA2 (3), GPR143 (5) and CACNA1F (1). This analysis also identified multiple carriers of an apparently novel deletion in OCA2. However, further analysis suggested that this is, in fact, the relatively common founder variant responsible for oculocutaneous albinism (OCA) in African populations, first identified over 30 years ago. These findings show that SVRare effectively identifies deleterious SVs and illustrates the challenges in reporting SVs, including the importance of accurate and consistent reporting of such variants.