Deciphering the genetic architecture of IRD by integrated analysis of 425 whole genomes

Abstract

Purpose :

To define the genetic architecture of retinal dystrophies (RD) by examining whole genome sequence (WGS) data from pedigrees with inherited retinal dystrophies (IRDs).

Methods : The WGS of 425 members from 105 pedigrees with IRD have been analyzed. Among these, 30 were previously analyzed by sequencing exomes (WES), while the remaining were analyzed using a broad range of targeted gene/mutation panels. All patients underwent standard ophthalmic evaluation while selected members underwent ERG, OCT, visual field, and color vision testing. WGS was performed using Illumina X10 (average coverage 30x). Sequence reads were mapped against hg19, SNVs and indels were called using GATK and structural variants (SVs) were called using GenomeSTRiP and LUMPY. The genome variants were annotated using SnpEff, PolyPhen2, and CADD score. Variants were filtered to detect rare, potentially deleterious changes segregating with IRD. Candidate variants were validated by dideoxy sequencing. Potential mutations in novel genes were further evaluated to study their impact on the gene.

Results : Clinical evaluation revealed the inheritance of IRD to be recessive in 76, dominant in 20 and X-linked in 9 pedigrees. Potential causative mutations were detected in 73 pedigrees (70%). We established the involvement of novel genes AGBL5, IFT88 and ARFGAP2 in IRD in 3 pedigrees and detected mutations in known RD genes in 70 (67%). Mutations in known genes comprised 50 previously reported and 58 novel variants. Among the novel variants, 9 were SVs, one de-novo variant and one was non-coding (in the promoter region of TMEM216). Eleven of the 70 pedigrees had patients with causal mutations in multiple genes, for instance one pedigree had causal mutations in both C2orf71 and CLN3. These patterns of inheritance suggest that complex genic interactions may play a role in some IRD pedigrees. Causative mutations are yet to be identified in the remaining 32 pedigrees (30%).

Conclusions : WGS analysis of 105 pedigrees that underwent varying levels of prior screening identified mutations in known RD genes in 67% and novel genes (AGBL5, IFT88 and ARFGAP2) in 3%. We determined that SVs and/or complex genotypes are likely contributing to IRD in 22 families. Potential involvement of novel genes and/or complex gene interactions may also be responsible for disease in the remaining 32 families.

This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.