RPGR, a common source of missed variants in exome sequencing experiments

Abstract

Purpose: The ORF15 of the RPGR gene is well known to harbor a large fraction of the variants that cause X-Linked Retinitis Pigmentosa (XLRP). XLRP can be difficult to diagnose clinically, because it has a wide spectrum of severity, can manifest early or late in life, it can selectively affect rods or cones in some patients, it can appear autosomal dominant in some families while presenting as a simplex case in others. As RP genetic testing becomes increasingly dependent on whole or targeted exome sequencing, accurately calling variants in this highly repetitive region is becoming increasingly important. In fact, most current commercial exome sequencing will fail to detect some disease-causing variants in ORF15 due to alignment errors. Therefore, there is a need for both optimized bioinformatic techniques and improved methods to sequence this highly repetitive region.

Methods: Exomes from 256 individuals suspected of having RP and related retinal degenerative diseases were aligned to the genome using BWA and variants were called using GATK. Sanger sequencing of ORF15 was also performed using an optimized TA cloning-based technique on individuals from 45 families.

Results: Exome sequencing revealed 3/256 patients with plausible disease-causing variations in RPGR ORF15. In contrast, 23 of the 45 families sequenced using the TA cloning approach harbored disease-causing variations in ORF15. Of these, nine fell in the most repetitive portion of ORF15 and would not have been detectable using exome sequencing alone. One exome was performed on a sample known to harbor a frame-shift mutation at the edge of this interval. This variant was not detected using either our standard bioinformatic analysis techniques or a survey of common alignment tools and variant callers.

Conclusions: Due to the shortcomings of exome sequencing of ORF15 and the prevalence of disease-causing variants in this region, exome-based RP genetic testing strategies should incorporate screening of the RPGR ORF15. We have developed a simple, efficient method to perform this screen. Without this additional tier of testing, RPGR represents a major source of false negatives in clinical exome sequencing experiments.

Keywords: 467 clinical laboratory testing • 702 retinitis • 537 gene screening