Cristina Rodilla ^1,2 Gonzalo Núñez Moreno ^1,2,3 Yolanda Benitez ^1,3 Raquel Romero ^1,2,3 Lidia Fernández-Caballero ^1,2* Pablo Minguez ^1,2,3 Marta Corton ^1,2* Carmen Ayuso ^1,2*

• ¹ Department of Genetics & Genomics, Health Research Institute Foundation Jimenez Diaz (IIS-FJD), Madrid, Spain
• ² Center for Biomedical Network Research on Rare Diseases (CIBERER), Madrid, Spain
• ³ Bioinformatics Unit, Health Research Institute Foundation Jimenez Diaz (IIS-FJD), Madrid, Spain

Long-read sequencing (LRS) enables accurate structural variant detection and variant phasing. When a molecular diagnosis is suspected, employing target enrichment can reduce the cost and duration of sequencing. LRS was conducted in five inherited retinal dystrophies (IRD) patients harboring a monoallelic variant in RPE65 that remained uncharacterized after clinical exome sequencing (CES). CRISPR-Cas9 guide RNA probes were designed to target a 31 kb region including the entire RPE65 locus. DNA was sequenced on a MinION platform. For 5 patients, short-read 30x whole genome sequencing (WGS) was performed to validate nanopore results. The nanopore sequencing process yielded a median of 271 reads within the targeted region, with a 109 mean depth and a median read size of 8 kb. All identified variants by CES have been detected using this approach, no additional RPE65 gene causative variants were found. Nanopore variant detection demonstrated performance akin to short read WGS at similar coverage levels, although exhibiting increased false positive calls at lower coverage. In this study, we explore the advantages of using a targeted approach together with long-read sequencing for identifying variants associated with IRD. The results underscore the utility of targeted long reads for the characterization of patients affected by rare diseases when first-tier diagnostic test are non-conclusive.