Abstract
Purpose :
Inherited retinal degenerations (IRDs) are an important cause of blindness affecting over two million people worldwide. Even though IRDs are mostly monogenic, they are genetically heterogeneous with mutations in over 200 genes leading to disease. Despite substantial progress in sequencing and new disease gene discovery, current strategies can genetically solve only about 60% of IRD cases. The high number of unsolved cases can be attributed to the yet-unidentified genes, large insertion/deletions also called copy number variations (CNVs), and deep intronic mutations, which are not easily detected by targeted next-generation sequencing (NGS) approaches. The purpose of this study was to test the use of the read-depth data from NGS to predict CNVs in the genetically unsolved patients.
Methods :
The patients included in the study were recruited and clinically examined at the Massachusetts Eye and Ear Infirmary. Patients underwent a full ophthalmic examination and their DNA samples were studied by Genetic Eye Disease (GEDi) diagnostic test, SNP genotyping array, quantitative real-time PCR (qRT-PCR), PCR and Sanger sequencing. The read-depth NGS analysis was performed with two methods (Exome-Depth and an in-house algorithm). The study protocol adhered to the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board. All subjects signed informed consent.
Results :
350 probands were sequenced with the same GEDi design (v5), of which 158 (45%) remained unsolved after the initial analysis of rare single nucleotide variants (SNVs) and small insertion/deletions. NGS read-depth analysis of the unsolved patients predicted likely heterozygous deletions of the coding regions, matching rare SNVs in the same gene, in 37 probands (10%). Homozygous coding deletions were seen in 2 patients (0.6%) and heterozygous likely causal deletions in autosomal dominant genes were seen in 3 patients (0.9%). Deletions in USH2A were the most common (22%). An additional 27 patients (8%) were predicted to carry duplications of the coding regions or deletions in the intronic regions. These were considered as low confidence CNVs and warrant further validation.
Conclusions :
Analysis of a large IRD cohort indicate that CNVs are important contributors to the etiology of IRDs, being responsible for 10-20% of genetic cases. Analysis of the NGS sequence depth proved to be a robust tool to predict CNVs.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.