Abstract
Purpose :
To date, approximately 60% of inherited retinal degenerations patients still do not have a molecular diagnosis despite genetic testing. In this study, we aimed to design a bioinformatic pipeline to identify the second variant in patients in whom only a single variant had previously been identified in a presumed autosomal recessively inherited gene consistent with the clinical phenotype
Methods :
We included 27 patients with IRDs recruited to the 100,000 Genomes Project (GEL) from the Oxford Eye Hospital in whom whole-genome sequencing (WGS) was performed for the proband and 2 family members where possible.
A bioinformatics workflow was designed encompassing a wide range of publicly available bioinformatics tools such as PolyPhen-2, SIFT, CADD, and UTRannotator in addition to a custom-made tool designed by one of the authors called SVrare. The latter analyses and prioritises structural variants identified by CANVAS and MANTA
Results :
Of the 27 patients included, 9 were solved at different stages of the pipeline. The novel variants varied between structural variants, intronic variants that affect splicing, and single nucleotide variants in coding regions that skipped the conventional bioinformatics pipeline as they either had low pathogenicity scores or were considered too common to be pathogenic. In this study, we present the phenotype of the solved patients and their genotype as identified at each stage of the purpose-built pipeline
Conclusions :
We demonstrate a substantial increase in the rate of confirming a genetic diagnosis in patients with IRDs using a bioinformatic pipeline to perform a comprehensive analysis of candidate genes. This work also shows the added value of scrutiny of whole-genome sequencing data of undiagnosed patients by a research team working within a multidisciplinary framework
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.