Revisiting WGS data reveals previously overlooked structural variants disrupting IRD-associated genes

Abstract

Purpose : Structural variants (SVs) play an important role in the development of inherited retinal diseases (IRDs). Although the identification of SVs has significantly improved with the arrival of whole genome sequencing (WGS), we hypothesize that the involvement of SVs in IRDs is larger than currently anticipated. In this study, we revisited short-read WGS (srWGS) data from an IRD cohort to identify gene-disruptive SVs. Prioritization criteria were re-established to allow optimal detection and interpretation of SVs.

Methods : Optical genome mapping (OGM) was performed to improve SV detection in srWGS-negative cases. OGM identified a pericentric 173 Mb inversion in a monoallelic USH2A case. One breakpoint of this inversion is located within USH2A and disrupts the coding region of the gene. Retrospectively, the variant could be observed in srWGS data but was previously deemed false positive, mainly because of the unexpected large size of the variant. The identification of this extremely large inversion urged us to revisit our srWGS data and reestablish SV quality and filtering criteria. In total srWGS data of 425 IRD cases were screened for disruptive SVs affecting IRD-associated genes.

Results : Reanalysis of srWGS data revealed large inversion events that were overlooked during initial analyses, including a 5.8 Mb USH2Aand 4.1 Mb EYS inversion. The EYS inversion was identified in a case that was previously considered genetically solved, as this individual harbors compound heterozygous EYS variants of which one variant of unknown significance (Pt-12, Fadaie et al., PMID:34795310). The identification of the SV prompted us to revise this genetic diagnosis and emphasizes the need for complete genome analysis including SV detection. Additionally several small homozygous or heterozygous deletions were identified affecting e.g. EYS and PRPF31. The corresponding samples are now considered genetically solved.

Conclusions : Our data confirm a significant contribution of small and large SVs to the mutational landscape of IRDs and thus possibly also of other inherited disorders. The optimized prioritization protocol significantly improved SV identification from srWGS data, as it yielded several pathogenic SVs that were missed during previous analyses. This strongly suggests that more attention should be paid to SV identification and interpretation, and that the current contribution of SVs is still underestimated.

This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.