Comprehensive Rare Variant Analysis via Whole-Genome Sequencing to Determine the Molecular Pathology of Inherited Retinal Disease

Abstract

Inherited retinal disease is a common cause of visual impairment and represents a highly heterogeneous group of conditions. Here, we present findings from a cohort of 722 individuals with inherited retinal disease, who have had whole-genome sequencing (n = 605), whole-exome sequencing (n = 72), or both (n = 45) performed, as part of the NIHR-BioResource Rare Diseases research study. We identified pathogenic variants (single-nucleotide variants, indels, or structural variants) for 404/722 (56%) individuals. Whole-genome sequencing gives unprecedented power to detect three categories of pathogenic variants in particular: structural variants, variants in GC-rich regions, which have significantly improved coverage compared to whole-exome sequencing, and variants in non-coding regulatory regions. In addition to previously reported pathogenic regulatory variants, we have identified a previously unreported pathogenic intronic variant in CHM in two males with choroideremia. We have also identified 19 genes not previously known to be associated with inherited retinal disease, which harbor biallelic predicted protein-truncating variants in unsolved cases. Whole-genome sequencing is an increasingly important comprehensive method with which to investigate the genetic causes of inherited retinal disease.

Keywords: copy-number variants; rare sequence variant; retinal dystrophy; whole-genome sequence.

Figure 1

Case Study 1: WGS, Compared to WES, Increases Power to Detect SVs (A) One individual (W000325) with retinitis pigmentosa has a pathogenic heterozygous deletion within EYS (6:656,02,819–65,658,187). IGV plots showing the deleted region in both WGS and WES data in the individual. The deletion was identified by WGS due to the drop in coverage and increased distance between mate-paired reads. However, it was not identified from WES data. (B) The individual also has a pathogenic missense variant ENST00000503581.1; c.6473T>C (p.Leu2158Pro).

Figure 2

Case Study 2: Identification by WGS of Pathogenic Tandem Duplication and Likely UPD One individual (W000170) with atypical, early-onset retinal dystrophy, has a homozygous pathogenic tandem duplication within KCNV2, likely due to partial maternal uniparental isodisomy of chromosome 9. (A) IGV plot showing the 184 bp tandem duplication of 9: 2,717,844–2,718,028 in WGS data. (B) The tandem duplication was confirmed by PCR and Sanger sequencing. The arrows represent the positions of the primers. (C) There is a ∼25 Mb region of homozygosity in chromosome 9 in this individual, which encompasses KCNV2. Approximate coordinates of this region of homozygosity are 9:2,100,000–27,400,000. (D) Sanger sequencing also confirms a homozygous combined in-frame insertion/deletion in KCNV2, within the tandem duplication: ENST00000382082.3; c.222_232delGGACCAGCAGGinsGGTCACCACCACCTTGG (ENSP00000371514.3; p.Asp75_Gln77delinsValThrThrThrLeu). The mother of the affected individual is heterozygous for this variant, but the father is homozygous for the reference allele (not shown).

Figure 3

Case Study 3: Identification and Characterization of Compound Heterozygous Deletions in EYS by WGS (A) One individual (W000164) with retinitis pigmentosa has compound heterozygous deletions in EYS, shown by the green dashed lines. (B) Sanger sequencing confirmed both deletions. Sequencing across the breakpoint (vertical dashed line) of deletion 2 is shown.

Figure 4

Regions of IRD-Associated Genes that Are Low or High in GC Content Have Significantly Higher Coverage in WGS Data than in the ExAC WES Project Coverage was calculated across protein-coding regions of autosomal IRD-associated genes (Ensembl canonical transcript), split into 50 bp bins. Coverage shown is relative to the average coverage of each dataset. Error bars show standard deviation. ^∗p < 1 × 10⁻¹⁵.

Figure 5

Case Study 4: Identification of Pathogenic Variants by WGS in GC-Rich Regions Not Covered by WES One individual (G004991) with Leber congenital amaurosis has pathogenic compound heterozygous variants ENST00000254854.4; c.238_252delGCCGCCGCCCGCCTG (p.Ala80_Leu84del) and ENST00000254854.4; c.307G>A (p.Glu103Lys) in exon 1 of GUCY2D, which is 76% GC-rich. This exon is not covered by WES in our cohort, as demonstrated by the WES data of a control sample shown here. It is also not well covered in ExAC.

Figure 6

The Deep Intronic CHM Variant c.315-1536A>G Results in the Inclusion of a Cryptic Exon (A) Sequence of the cryptic exon included by the deep intronic CHM variant chrX:85,220,593T>C (ENST00000357749.2; c.315-1536A>G). (B) RT-PCR showing increased size of fragment due to inclusion of the cryptic exon.