Free
Genetics  |   October 2014
Dependable and Efficient Clinical Utility of Target Capture-Based Deep Sequencing in Molecular Diagnosis of Retinitis Pigmentosa
Author Affiliations & Notes
  • Jing Wang
    Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, Texas, United States
  • Victor W. Zhang
    Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, Texas, United States
  • Yanming Feng
    Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, Texas, United States
  • Xia Tian
    Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, Texas, United States
  • Fang-Yuan Li
    Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, Texas, United States
  • Cavatina Truong
    Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, Texas, United States
  • Guoli Wang
    Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, Texas, United States
  • Pei-Wen Chiang
    Casey Eye Institute, Oregon Health and Science University, Portland, Oregon, United States
  • Richard A. Lewis
    Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, Texas, United States
    Department of Ophthalmology, Baylor College of Medicine, Houston, Texas, United States
  • Lee-Jun C. Wong
    Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, Texas, United States
Investigative Ophthalmology & Visual Science October 2014, Vol.55, 6213-6223. doi:10.1167/iovs.14-14936
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Jing Wang, Victor W. Zhang, Yanming Feng, Xia Tian, Fang-Yuan Li, Cavatina Truong, Guoli Wang, Pei-Wen Chiang, Richard A. Lewis, Lee-Jun C. Wong; Dependable and Efficient Clinical Utility of Target Capture-Based Deep Sequencing in Molecular Diagnosis of Retinitis Pigmentosa. Invest. Ophthalmol. Vis. Sci. 2014;55(10):6213-6223. doi: 10.1167/iovs.14-14936.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: The purpose of this study was to establish a fully validated, high-throughput next-generation sequencing (NGS) approach for comprehensive, cost-effective, clinical molecular diagnosis of retinitis pigmentosa (RP).

Methods.: Target sequences of a panel of 66 genes known to cause all nonsyndromic and a few syndromic forms of RP were enriched by using custom-designed probe hybridization. A total of 939 coding exons and 20 bp of their flanking intron regions with a total of 202,800 bp of target sequences were captured, followed by massively parallel sequencing (MPS) on the Illumina HiSeq2000 device.

Results.: Twelve samples with known mutations were used for test validation. We achieved an average sequence depth of ∼1000× per base. Exons with <20× insufficient coverage were completed by PCR/Sanger sequencing to ensure 100% coverage. We analyzed DNA from 65 unrelated RP patients and detected deleterious mutations in 53 patients with a diagnostic yield of ∼82%.

Conclusions.: Clinical validation and consistently deep coverage of individual exons allow for the accurate identification of all types of mutations including point mutations, exonic deletions, and large insertions. Our comprehensive MPS approach greatly improves diagnostic acumen for RP in a cost- and time-efficient manner.

Introduction
Retinitis pigmentosa (RP, Mendelian Inheritance in Man [MIM] 268000) is a diverse group of inherited disorders that result in progressive retinal degeneration and affect millions of people worldwide. 14 These disorders are caused by inexorable loss of rod and cone photoreceptor cells and/or the retinal pigment epithelium (RPE). More than 190 genes have been identified as causing one or more forms of inherited retinal disorders. 4 The clinical presentation of RP can be nonsyndromic or syndromic. 2,4 In syndromic forms, RP occurs in addition to abnormalities in nonocular tissues and organs. Retinitis pigmentosa is a highly complex phenotype due to both genetic and clinical heterogeneities. Different mutations in the same gene may cause different phenotypes. 59 Furthermore, clinical symptoms, appearance, and severity may differ among individuals with the same mutation(s) in the same gene, even in the same family. Since the number of genes involved is large, digenic events, such as the simultaneous presence of heterozygous mutations in two autosomal dominant (AD) genes, PRPH2 and ROM1, have been reported. 10,11 In addition, a patient from a consanguineous family has been reported to have a “double hit” with homozygous mutations in two autosomal recessive (AR) RP-causative genes: MYO7A and PDE6B. 12  
The clinical diagnosis of RP often relies on results of both electroretinography (ERG) and visual field testing. However, the similarity of the progressive pigmentary degeneration in the fundus from case to case and from family to family affords no clues as to the specific molecular defect in any one case or family. Thus, identification of disease-causing mutations in affected individuals is essential for genetic counseling, carrier testing, and evolving gene-specific therapies. High-throughput sequencing with deep coverage of the coding regions offers a cost-effective approach to those molecular diagnostic challenges. 
Materials and Methods
Patients
Patient samples were submitted to the Medical Genetics Laboratories at Baylor College of Medicine (BCM) for sequence analysis of gene(s) responsible for RP. The analyses were performed according to the BCM Institutional Review Board–approved protocols for human subjects and complied with the tenets of the Declaration of Helsinki. Deoxyribonucleic acid samples from 12 patients with known mutations in RP genes were included as positive controls for validation. Sixty-five samples from patients with the clinical diagnosis of nonsyndromic RP, judged by skilled ophthalmologists, were analyzed. 
Design of Capture Probe Library and Target Gene Enrichment
Target sequences of a panel of 66 genes known to cause all nonsyndromic and a few syndromic forms of RP (Table 1) were enriched by using custom-designed NimbleGen SeqCap probe hybridization (Roche NimbleGen, Inc., Madison, WI, USA). The captured target sequences include all coding exons and 20 bp of their flanking intronic regions. Deoxyribonucleic acid template libraries were prepared according to the manufacturer's recommendation. Equal molar ratios of 10 indexed samples were pooled to be loaded onto each lane of the flow cells for sequencing on a HiSeq2000 (Illumina, Inc., San Diego, CA, USA) with 100 cycle single-end reads. The GenBank accession numbers for each gene are listed in Supplementary Table S1
Table 1
 
Genes Included in the RP Panel and Summary of Capture and Sequencing
Table 1
 
Genes Included in the RP Panel and Summary of Capture and Sequencing
Next-Generation Sequencing Panel Retinitis Pigmentosa Panel—66 Genes
Genes included in this panel ABCA4, ABHD12, AIPL1, BEST1, C2orf71, C8orf37, CA4, CDHR1, CEP290, CERKL, CLRN1, CNGA1, CNGB1, CRB1, CRX, DHDDS, EYS, FAM161A, FLVCR1, FSCN2, GUCA1B, GUCY2D, IDH3B, IMPDH1, IMPG2, KLHL7, LCA5, LRAT, MAK, MERTK, MFRP, NR2E3, NRL, PDE6A, PDE6B, PDE6G, PRCD, PRKCG, PROM1, PRPF3, PRPF31, PRPF6, PRPF8, PRPH2, RBP3, RD3, RDH12, RGR, RHO, RLBP1, ROM1, RP1, RP2, RP9, RPE65, RPGR, RPGRIP1, SAG, SEMA4A, SNRNP200, SPATA7, TOPORS, TTC8, TULP1, USH2A, ZNF513
Number of CDS 939
Targeted size 202,800 bases (coding exon sequences ± 20 bp)
Average coverage Approximately 1000× per base
Enrichment In solution capture library (NimbleGen)
Sequencing information Illumina HiSeq2000, 100 cycle, single-end
Massively Parallel Sequencing (MPS) Data Analysis and Variant Calling
Raw data in base call files (.bcl format) were converted to qseq files before demultiplexing with CASAVA v1.7 software (Illumina, Inc.). Demultiplexed data were processed further by NextGENe software for alignment (SoftGenetics, State College, PA, USA). We used a proprietary intramural bioinformatics pipeline for the variant annotations and classifications (Fig. 1). All mutations and novel variants were confirmed by Sanger sequencing followed by careful classification according to American College of Medical Genetics and Genomics (ACMG) recommendations for standards of interpretation of sequence variation. 13  
Figure 1
 
Analytical pipeline for variant annotation and classification. Estimated time needed for each step and the approximate number of variants at each analysis step are shown.
Figure 1
 
Analytical pipeline for variant annotation and classification. Estimated time needed for each step and the approximate number of variants at each analysis step are shown.
Sanger Sequencing of Insufficiently Covered Regions and Confirmation of MPS Findings
All exons containing any base with <20× coverage were completed by PCR/Sanger sequencing of the particular coding sequences (CDS). In addition, a known deep intronic splicing mutation, c.2991+1655A>G, in the CEP290 gene was also included by Sanger sequencing. All mutations and novel variants detected by MPS were confirmed independently by Sanger sequencing. Any ambiguous MPS variant calls were also verified by Sanger sequencing. 
Results
Characteristics of Target Gene Capture and Sequence Depth
More than 99.2% of the target sequences are enriched in an unbiased fashion with a mean coverage of ∼1000×. Seven exons are consistently insufficiently covered (<20×) due to high guanine and cytosine (GC) content. These exons and 10 additional exons with homologous sequences in the genome are completed by PCR/Sanger sequencing. Thus, this clinically fully validated panel has 100% coverage of every single base of target sequences. The sensitivity and specificity for all 202,800 base pairs per sample are 100% when compared with the results of Sanger sequencing of the same set of genes of the same sample. 
Verification of Known Mutations and Identification of Additional Mutations
We analyzed DNA samples from 12 patients with mutations identified previously by Sanger sequencing. All mutations and polymorphisms were identified correctly by our capture/MPS-based analysis (Supplementary Table S2). Since all these cases had been analyzed previously by Sanger sequencing for only one or a few candidate genes, the search stopped once disease-causing mutations were identified. In contrast, our capture/MPS approach analyzes 66 RP genes in parallel, allowing the identification of additional mutations and variants in genes that have not been analyzed. Mutations and unclassified variants detected by this panel as likely to be pathogenic are listed under each tested subject in Supplementary Table S2
Identification of Mutations in Patients With RP
Samples from 65 unrelated patients who met the clinical criteria for RP were evaluated with this capture-based MPS analysis. Approximately 150 to 200 sequence variations per individual were identified after passing through MPS quality control (QC) check (Fig. 1). These variations were then filtered through a pipeline according to ACMG guidelines for variant classification. 13 Variants with a minor allele frequency (MAF) > 0.01 in the dbSNP database, the 1000 Genomes database, or the ESP 4500 database are considered common polymorphisms and likely benign, while novel variants with MAF < 0.01 are categorized as unclassified. After being filtered through the variant classification pipeline, all deleterious mutations and unclassified variants deemed likely pathogenic were confirmed by Sanger sequencing (Table 2). Only mutations and unclassified variants are listed in Table 2. Additional single heterozygous variants of unknown clinical significance (VUS) in one or more AR genes are listed in Supplementary Table S3. Deoxyribonucleic acid samples from the parents or other family members, if available, were analyzed to confirm parental carrier status and the segregation of mutations with disease. 
Table 2
 
Mutations and Unclassified Variants Identified in Previously Undiagnosed Patients
Table 2
 
Mutations and Unclassified Variants Identified in Previously Undiagnosed Patients
Patients Sex Age, y Clinical Diagnosis Genes Protein Allele 1 Allele 2
Autosomal recessive
1 F 19 Pigmentary retinal dystrophy ABCA4 (AR) ATP-binding cassette transporter—retina c.4469G>A (p.C1490Y) c.4469G>A (p.C1490Y)
2* F 27 Decreased vision ABCA4 (AR) ATP-binding cassette transporter—retina c.1365+5_+6insC c.1927G>A (p.V643M) c.3602T>G (p.L1201R)
USH2A (AR) Usherin c.10858A>G (p.I3620V) c.15377T>C (p.I5126T)
3* M 11 Bull's-eye maculopathy, cone dystrophy ABCA4 (AR) ATP-binding cassette transporter—retina c.3385C>T (p.R1129C) c.5461-10T>C
USH2A (AR) Usherin c.10073G>A (p.C3358Y) c.11026A>G (p.T3676A) c.14753C>T (p.T4918M)
CA4 (AD) Carbonic anhydrase IV c.700G>A (V234I) Neg
4* F 14 Rod-cone dystrophy ABCA4 (AR) ATP-binding cassette transporter—retinaChromosome 8 open reading frame 37 c.2588G>C (p.G863A) c.2828G>A (p.R943Q)
C8orf37 (AR) c.243+3A>C c.243+3A>C
5 M 39 Cone dystrophy ABCA4 (AR) ATP-binding cassette transporter—retina c.5196+1G>A c.5882G>A (p.G196E) c.2828G>A (p.R943Q)
6 F 1 mo RP ABCA4 (AR) ATP-binding cassette transporter—retina c.2828G>A (p.R943Q) c.2588G>C (p.G863A)
7* F 25 RP CEP290 (AR) Centrosomal protein 290 c.1078C>T (p.R360*) c.6851_6855delCTGAT (p.T2284Nfs*10)
AIPL1 (AR) Arylhydrocarbon-interacting receptor protein-like 1 c.140C>G (p.T47R) c.937G>T (p.A313S)
8 F 1 Senior-Loken syndrome CEP290 (AR) Centrosomal protein 290 c.2587-2A>G c.2587-2A>G
9 F 35 RP, muscle weakness CEP290 (AR) Centrosomal protein 290 c.297+1G>T c.6401T>C (p.I2134T)
10 M 26 Decreased vision, field loss, night vision blindness CERKL (AR) Ceramide kinase-like protein c.812T>C (p.I271T) c.812T>C (p.I271T)
11 M 20 Progressive central and peripheral vision loss, abnormal constructed visual field CERKL (AR) Ceramide kinase-like protein c.598A>T (p.K200*) c.769C>T (p.R257*)
12* M 35 RP, loss of peripheral vision CERKL (AR) Ceramide kinase-like protein c.1462G>A (p.E488*) c.1462G>A (p.E488*)
NR2E3 (AD, AR) Nuclear receptor subfamily 2 group E3 c.932G>A (p.R311Q) Neg
RP2 (XL) RP2 protein c.844C>T (p.R282W) NA
EYS (AR) Eyes shut/spacemaker (Drosophila) homolog c.1765A>G (p.R589G) Neg
13 F 48 Decreased peripheral vision, photophobia, difficult light-to-dark adaptation, night blindness EYS (AR) Eyes shut/spacemaker (Drosophila) homolog exon 14 del exon 14 del
14* M 18 Bone spicule pigmentary clumping in fundus, VA = 20/20–25, peripheral field constriction EYS (AR) Eyes shut/spacemaker (Drosophila) homolog c.5600C>T (p.S1867F) c.8422G>A (p.A2808T)
BEST (AD) Bestrophin 1 c.714+13C>T Neg
15 M 24 Decreased peripheral vision, night blindness EYS (AR) Eyes shut/spacemaker (Drosophila) homolog c.5681T>C (p.L1894P) c.8779T>C (p.C2927R)
16 M 31 RP FAM161A (AR) Family with sequence similarity 161 member A c.1309A>T (p.R437*) c.1501delT(p.C501Vfs*4)
17 F 77 RP MAK (AR) Male germ cell-associated kinase c.1297_1298 insAlu c.1297_1298 insAlu
18 M 76 RP MAK (AR) Male germ cell-associated kinase c.1297_1298 insAlu c.1297_1298 insAlu
19 M 68 RP MAK (AR) Male germ cell-associated kinase c.1297_1298 insAlu c.1297_1298 insAlu
20 M 18 Pigmentary retinal dystrophy MERTK (AR) c-mer protooncogene receptor tyrosine kinase c.62-1G>A c.1296+1G>A
21 F 20 Hereditary retinal dystrophy, rod-cone dystrophy MERTK (AR) c-mer protooncogene receptor tyrosine kinase c.1951C>T (p.R651*) c.345C>G (p.C115W)
22* F 45 RP MFRP (AR) Membrane-type frizzled-related protein c.772+2T>G c.1124+1G>T
CERKL (AR) Ceramide kinase-like protein c.769C>T (p.R257*) Neg
23 M 35 RP PDE6B (AR, AD) Rod cGMP phosphodiesterase beta subunit c.1833-1G>C c.1833-1G>C
24 F 21 Nyctalopia, reduced peripheral vision PDE6B (AR, AD) Human homolog of yeast pre-mRNA splicing factor 6 phosphodiesterase beta subunit c.869G>A (p.W290*) c.1280G>A (p.W427*)
25 M 35 NA PDE6G (AR) Phosphodiesterase 6G cGMP-specific rod gamma c.109C>T (p.Q37*) c.109C>T (p.Q37*)
26 M  9 Usher syndrome type II USH2A (AR) Usherin c.2299delG (p.E767Sfs*21) c.4714C>T (p.L1572F) c.7595-3C>G
27 F 39 Nyctalopia, decreased peripheral vision, history of plaquenil USH2A (AR) Usherin c.3395G>A (p.G1132D) c.5624A>G (p.N1875S)
28 M 17 Usher syndrome USH2A (AR) Usherin c.5776+1G>A c.14131C>T (p.Q4711*)
29* M 36 Usher syndrome USH2A (AR) Usherin c.1036A>C (p.N346H) c.2299delG (p.E767Sfs*21) c.4714C>T (p.L1572F)
NR2E3 (AD, AR) Nuclear receptor subfamily 2 group E3 c.767C>A (p.A256E) Wild type
RP1 (AD, AR) RP1 protein c.1118C>T (p.T373I) Wild type
RP2 (XL) RP2 protein c.844C>T (p.R282W) NA
30* M 37 RP USH2A (AR) Usherin c.2299delG (p.E767Sfs*21) c.2276G>T (p.C759F) c.4714C>T (p.L1572F)
RP IMPDH1 (AD) Inosine monophosphate dehydrogenase 1 c.1057G>A (p.V353I) Wild type
31* F 32 RP MERTK (AR) c-mer protooncogene receptor tyrosine kinase c.2164C>T (p.R722*) c.2219C>T (p.A740V)
CA4 (AD) Carbonic anhydrase IV c.198_199delACinsG (p.L67Wfs*24) Wild type
32 F 51 RP ABCA4 (AR) ATP-binding cassette transporter—retina c.766G>T (p.V256L) c.5755G>T (p.D1919Y)
33 F 8 Progressive pigmentary retinopathy CEP290 (AR) Centrosomal protein 290 c.4834_4835delAC (p.T1612Sfs*13) c.2980G>A (p.E994K)
34 M 11 Peripheral dystrophy EYS (AR) Eyes shut/spacemaker (Drosophila) homolog c.3443+1G>T c.2412G>C (p.Q804H) c.3250A>C (p.T1084P) c.4402G>C (p.D1468H)
Autosomal dominant
35 F 64 Progressive pigmentary retinopathy FSCN2 (AD) Retinal fascin homolog 2, actin bundling protein c.467C>T (p.P156L) Wild type
36* M 29 RP FSCN2 (AD) Retinal fascin homolog 2, actin bundling protein c.72delG (p.T25Qfs*120) Wild type
IMPDH1 (AD) Inosine monophosphate dehydrogenase 1 c.569G>T (p.R190L) Wild type
37 F 54 RP PDE6B (AR, AD) Human homolog of yeast pre-mRNA splicing factor 6phosphodiesterase beta subunit c.973G>C (p.E325Q) Wild type
38 M  7 Rod dystrophy PRPF8 (AD) Human homolog of yeast pre-mRNA splicing factor C8 c.6961C>T (p.Q2321*) Wild type
39 M 55 RP PRPH2 (AD) Peripherin 2 c.422A>G (p.Y141C) Wild type
40 F 70 RP RHO (AD) Rhodopsin c.936+1G>T Wild type
41* F  7 RP, bone spicules, optic disc pallor, nyctalopia, RHO (AD) Rhodopsin c.697-11G>A Wild type
RP, bone spicules, optic disc pallor, nyctalopia PRPF6 (AD) Human homolog of yeast pre-mRNA splicing factor 6 c.867-7C>G c.2431+9delG
42 M 19 RP SNRNP200 (AD) Small nuclear ribonucleoprotein 200 kDa c.2359G>A (p.A787T) Wild type
43 M 11 RP, likely AD inheritance TOPORS (AD) Topoisomerase I binding arginine/serine rich protein c.74C>G(p.S25W) Wild type
44 F 22 Nyctalopia, loss of peripheral vision, progressive pigmentary retinopathy NR2E3 (AD, AR) Nuclear receptor subfamily 2 group E3 c.119-2A>C c.119-2A>C
45 F 14 Retinal dystrophy PDE6B (AR, AD) Human homolog of yeast pre-mRNA splicing factor 6phosphodiesterase beta subunit c.2193+1G>A c.1624C>T (p.R542W)
46* M 45 Night blindness, visual field peripheral restriction, decreased vision, ERG consistent with RP RP1 (AD, AR) RP1 protein c.4196delG (p.C1399Lfs*5)c.6353G>A (p.S2118N) c.4196delG (p.C1399Lfs*5)c.6353G>A (p.S2118N)
47 M  6 Juvenile RP/LCA, nonrecordable electoretinogram, retinopathy RP1 (AD, AR) RP1 protein c.796delinsTA(p.H266*) c.1625C>G (p.S542*)
48* M 41 NA RPE65 (AD, AR) Retinal pigment epithelium-specific 65-kDa protein c.1597T>A (p.S533T) c.89T>C (p.V30A)
AIPL1 (AR) Arylhydrocarbon-interacting receptor protein-like 1 c.905G>T (p.R302L) Neg
CNGA1 (AR) Rod cGMP-gated channel alpha subunit c.521T>A (p.L174*) Neg
49 F 54 RP, macular subatrophy, pseudohole RHO (AD) Rhodopsin c.562G>A (p.G188R) Neg
X-linked
50 M  5 RP RP2 (XL) RP2 protein c.358C>T (p.R120*) NA
51 M 15 RP RPGR (XL) Retinitis pigmentosa GTPase regulator c.1202_1206del5 (p.V401Afs*50) NA
52 M 18 RP RPGR (XL) Retinitis pigmentosa GTPase regulator c.92G>A (p.W31*) NA
53 M 44 Rod-cone dystrophy RPGR ORF15 (XL) Retinitis pigmentosa GTPase regulator c.3178_3179delAG (p.E1060Rfs*18) NA
We confirmed molecular diagnoses in 53 unrelated patients (Table 2), achieving an overall positive diagnostic rate of 82% (53/65). Defects in AR, AD, and X-linked (XL) genes were detected in 34 (52%), 15 (23%), and 4 (6%) patients, respectively. Notably, mutations involving two (double hit) or more (multiple hit) genes were detected in 13 of 53 patients (Table 2). Interestingly, two patients (patients 12 and 29) had mutations in genes that have been reported to cause RP with different inheritance patterns, including AR, AD, and XL. For example, mutations in CERKL (AR), NR2E3 (AD and AR), and RP2 (XL) genes were detected in patient 12; mutations in the USH2A (AR), NR2E3 (AD and AR), RP1 (AD and AR), and RP2 (XL) genes were detected in patient 29. Twelve patients were heterozygous carriers for unclassified variants in one or more AR genes. Diagnoses in these patients cannot be confirmed. Below are notable highlights in this cohort. 
The Detection of Large Deletions and Insertions by Capture-Based MPS.
The strategy of capture-based target enrichment followed by deep sequencing (∼1000×) allows us to identify large exonic deletions and insertions in target genes. 
Case 13 is a 48-year-old female with decreased peripheral vision, photodysphoria, difficult light-to-dark adaptation, and night blindness (Table 2). While no deleterious point mutations or small indels were detected, the MPS data showed no coverage for exon 14 of the EYS gene. The presence of technical artifact was excluded, since this exon was adequately covered in all other specimens analyzed. These results suggested a homozygous deletion of exon 14. Subsequent exon-targeted array comparative genomic hybridization (CGH) analysis confirmed a homozygous deletion of approximately 28 kb involving exon 14 of the EYS gene (Fig. 2). 
Figure 2
 
EYS exonic deletion detected by MPS and confirmed by array CGH. (A) MPS exonic coverage profile for the EYS gene. Exon 14 had no coverage, suggesting a homozygous deletion. (B) Targeted array CGH detected a loss of copy number in exon 14. No copy number variations were detected in other exons of the EYS gene. (C) Enlarged view of exon 14 from array CGH. The log2 ratios for the probes in the deleted region are consistent with a homozygous deletion.
Figure 2
 
EYS exonic deletion detected by MPS and confirmed by array CGH. (A) MPS exonic coverage profile for the EYS gene. Exon 14 had no coverage, suggesting a homozygous deletion. (B) Targeted array CGH detected a loss of copy number in exon 14. No copy number variations were detected in other exons of the EYS gene. (C) Enlarged view of exon 14 from array CGH. The log2 ratios for the probes in the deleted region are consistent with a homozygous deletion.
Cases 17, 18, and 19 are three unrelated individuals of Jewish ancestry, each with a clinical diagnosis of RP; the age range is late 60s to mid-70s. An apparently homozygous insertion mutation, c.1297_1298insAlu in exon 10 of the MAK gene, was detected in each patient (Fig. 3). The inserted Alu sequence is approximately 355 bp, including a poly A tail of ∼65 bp. The insertion interrupts the normal reading frame and results in skipping of exon 10. This insertion mutation has been reported in multiple patients with AR RP and is a known common mutation in the Jewish population. 14 The heterozygous c.1297_1298insAlu in the asymptomatic daughter of case 18 was also detected by MPS (Fig. 3B). Both homozygous and heterozygous Alu insertions were confirmed by PCR and Sanger sequence analysis (Fig. 3C). 
Figure 3
 
Alu insertion in the MAK gene was identified by MPS and confirmed by PCR and Sanger sequence. (A, B) MPS data showing misaligned reads from patient 18 (A) and his unaffected daughter (B). The next-generation sequencing raw data themselves cannot tell what the sequence variation was or the zygosity. However, the misaligned reads suggested the possible sequence aberration. (C) Pedigree and schematic Alu insertion in exon 10. Expected PCR product length is 367 bp for the normal allele and is ∼722 bp for the allele with Alu insertion. A single ∼722-bp application band detected in patient 18 indicated the homozygous insertion; a normal band and an insertion band detected in daughter of patient 18 indicated that she is a heterozygous carrier for the Alu insertion. Sanger sequencing of the 722 bp PCR product confirmed an insertion of ∼355 bp Alu sequence (data not shown).
Figure 3
 
Alu insertion in the MAK gene was identified by MPS and confirmed by PCR and Sanger sequence. (A, B) MPS data showing misaligned reads from patient 18 (A) and his unaffected daughter (B). The next-generation sequencing raw data themselves cannot tell what the sequence variation was or the zygosity. However, the misaligned reads suggested the possible sequence aberration. (C) Pedigree and schematic Alu insertion in exon 10. Expected PCR product length is 367 bp for the normal allele and is ∼722 bp for the allele with Alu insertion. A single ∼722-bp application band detected in patient 18 indicated the homozygous insertion; a normal band and an insertion band detected in daughter of patient 18 indicated that she is a heterozygous carrier for the Alu insertion. Sanger sequencing of the 722 bp PCR product confirmed an insertion of ∼355 bp Alu sequence (data not shown).
Mutations in Multiple Genes Involving Different Inheritance Patterns in a Single Patient.
Case 29 is a 36-year-old European male with a history of profound hearing and associated speech impairment detected before age 2. He was diagnosed with RP by age 7 years. At age 36 years, his functional visual field was less than 5° in all meridians. His best corrected visual acuity was 20/50 OD and 20/40 OS. Vitreous syneresis was evident. Both optic nerves were diffusely pale, and the caliber of the major retinal vessels was moderately attenuated. No cystoid macular edema was noted; a broad circumferential belt of intraretinal “bone spicule” pigment migration filled the periphery of each eye. The impression of a progressive pigmentary retinopathy with neurosensory deafness yielded a clinical diagnosis of Usher syndrome type 1. 
The RP panel analysis identified six deleterious mutations in four different genes in this patient: three heterozygous mutations, c.1036A>C (p.N346H), c.2299delG (p.E767Sfs*21), and c.4714C>T (p.L1572F), in the USH2A gene; a heterozygous c.767C>A (p.A256E) mutation in the NR2E3 gene; a heterozygous c.1118C>T (p.T373I) mutation in the RP1 gene; and a hemizygous c.844C>T (p.R282W) mutation in the XL RP2 gene. The c.1036A>C (p.N346H) mutation in the USH2A gene has been reported in a compound heterozygous state in families with Usher syndrome type IIa. 15 The c.2299delG (p.E767Sfs*21) mutation is a common mutation found in Usher syndrome. 16,17 The c.4714C>T (p.L1572F) has often occurred in cis with the c.2299delG (p.E767Sfs*21) mutation. While mutations in the USH2A gene appeared to be most consistent with this patient's clinical presentation, the original clinical diagnosis of Usher syndrome type 1 cannot be explained by USH2A mutations alone. Mutations in the other RP genes may impact the eye involvement. 
Mutations in the RPGR ORF15 Gene.
The RPGR ORF15 is prominently expressed in the retina and is a mutational hot spot. 18 It contains a highly repetitive domain. All previously reported ORF15 mutations are null (nonsense or frame shift) mutations; no missense mutations have been reported in the Human Gene Mutation Database (HGMD, http://www.hgmd.cf.ac.uk/ac/index.php). Our capture/MPS approach was able to cover the majority of the ORF15 sequence. Only ∼300 bp in the highly repetitive region were covered insufficiently. Case 53 is a 44-year-old male with cone-rod dystrophy. A hemizygous c.3178_3179delAG (p.E1060Rfs*18) mutation in the RPGR ORF15 was detected. This mutation has been reported in a patient with XL RP. 19 Mutations in RPGR locus are implicated in RP, cone-rod, and cone dystrophies. 20 Compared to mutations in early-onset XL RP2 gene, all male patients with mutations in RPGR showed a later age of onset of disease. 19  
Discussion
High-Throughput Sequencing Technologies and High Diagnostic Yield
Despite the complexity of retinal disorders, substantial progress has recently been made in identifying new RP genes and developing high-throughput technologies to screen a panel of RP genes for disease-causing mutations. 2124 The overall diagnostic yields in these reports ranged from 16% to 70% (Table 3). We attribute our much higher diagnostic yield of 82% (Table 3) to full validation by Sanger, complete deep coverage, consistent average coverage of individual exons with low variation among different samples, high-quality data, and 100% sensitivity and specificity. Our clinically validated comprehensive approach also allows simultaneous detection of point mutations and exonic deletions and large insertions. 25  
Table 3
 
Diagnostic Yield Comparison in Different Platforms
Table 3
 
Diagnostic Yield Comparison in Different Platforms
Authors Platform Coverage Depth Targeted Genes/CDS No. of Patients Positive/Total Diagnostic Yield Disease
Song et al.21 Affymetrix resequencing chip NA  93/1470 19 patients with known mutation Inherited retinal dystrophy
Clark et al.22 Resequencing chip APEX NA 19/100 9/56 16% arRP
Shanks et al.28 NimbleGen 12-plex capture array, 454GS FLX 918–1239×  73/1413 9/36 25% RP
Coppieters at al.23 PCR capture, GAII 24–117×, avg. 58× 16/252 3/17 18% LCA
Wang et al.24 NimbleGen SeqCap, HiSeq2000 20–30×, avg. 26× 163/2560 72/179 40% RP + LCA
Neveling et al.27 NimbleGen 12-plex capture array, 454GS FLX 62× 111/2011 36/100 36% RP
Eisenberger et al.26 NimbleGen SeqCap, 454GS FLX and MiSeq 75× (GS FLX) 250× (MiSeq) 55/876 88/126 70% RP + LCA
This study NimbleGen SeqCap, HiSeq2000 1000× 66/939 53/65 82% RP
Since classic nonsyndromic RP exclusively affects eyes and has distinct, definable, and observable clinical features, whole exome sequencing (WES) is not a necessary first-tier test. In general, WES has approximately 3% to 6% uncharacterized insufficiently covered target sequences that have high variation among different batches due to interference of the enormous number of probes. In contrast, target capture sequence approaches, which focus on a panel of known candidate RP genes with deep coverage, minimize interference among probes, and provide consistent coverage of individual exons, are the most efficient way to yield a definitive molecular diagnosis. 
While augmenting the number of genes in the panel increases the chance to identify disease-causing mutations, it also increases the cost. A recent study included a panel of 55 genes with a diagnostic yield of 70%, 26 whereas some published RP panels contain more than 100 genes, with diagnostic yield ranging from 16% to 40%. 2224,27,28 These wide variations are due to quality of data, depth and consistency of coverage, and percentage of insufficiently covered sequences and uncharacterized regions. Our RP panel includes only 66 genes. Nevertheless, our diagnostic yield is at least 82%, which outperforms previously published panels at a yield of 16% to 70% (Table 3). Notably, the USH2A and ABCA4 genes are the two most commonly mutated genes in our patient cohort, consistent with previous reports. 26,29 These results also suggest that some Usher syndrome patients may present RP only at the beginning without hearing loss. 
Nonsyndromic RP clinically affects the eyes only, without the involvement of other organ systems. Syndromic RP usually occurs with other neurosensory or systemic developmental defects. The two most common syndromic associations are the Usher syndromes (USH) (RP with neurosensory hearing impairment) and the Bardet-Biedl syndromes (BBS) (RP with obesity, various renal disorders, postaxial polydactyly, and intellectual disabilities). In general, the specific diagnosis of the nonsyndromic RPs is more challenging than for syndromic RPs in terms of revealing genetic defects. We included all 56 genes known to cause nonsyndromic RP in this panel. 4 We have also included a few Leber's congenital amaurosis (LCA) and syndromic RP genes for differential diagnosis due to phenotypic and clinical overlap. For example, some mutations in Usher syndrome genes, CLRN1 and USH2A, have been reported to cause RP only, without hearing impairment. 30,31 Different mutations in the RPE65 gene can cause either RP or LCA. 8,9,32  
Deep Coverage of Target Sequences.
The average coverage of this RP panel is ∼1000×, which is at least 10 times higher than that of published RP or LCA panels (Table 3). 24,27,28 While most panels will have some consistently insufficiently covered targets, due either to high GC content or to repetitive sequences, the deep coverage reduces the number of insufficiently sequenced exons from a reported 5% to less than 1% in our panel. 24,27 Moreover, our data demonstrate that, by reducing coverage from 1000× to 100× in the RP panel described here, the number of insufficiently covered CDS (<20×) increased from 6.7 (0.7%) to 72 (7.7%) (Fig. 4). This indicates clearly that sequence depth is essential in reducing the number of insufficiently sequenced regions and increasing the accuracy of variant calls. 
Figure 4
 
Relationship between average coverage and number of CDS with coverage lower than 20× in nine randomly selected samples. Reducing coverage from 1000× to 100× in the RP panel described here, the number of <20× covered CDS (<20×) increased from 6.7 (0.7%) to 72 (7.7%).
Figure 4
 
Relationship between average coverage and number of CDS with coverage lower than 20× in nine randomly selected samples. Reducing coverage from 1000× to 100× in the RP panel described here, the number of <20× covered CDS (<20×) increased from 6.7 (0.7%) to 72 (7.7%).
Most importantly, we have used Sanger sequencing to complete all exons containing any base with <20× coverage. In our cohort, we perform PCR/Sanger sequencing routinely on 17 exons/amplicons in nine different genes that have homologous sequences or pseudogenes, or those consistently poorly sequenced due to the presence of repetitive sequences or high GC content. Therefore, the coverage for this clinically validated test is 100% for every single nucleotide in the target regions. 
Appropriate and Accurate Clinical Evaluation Is Essential for High Diagnostic Yield.
Retinitis pigmentosa disorders are intrinsically complicated. The clinical diagnosis encompasses different causes and diverse biological pathways but with overlapping symptoms and signs, and in such progressive diseases, variations with time/age at presentation. Therefore, careful and deliberate assessment of the patient's clinical and family history is essential to distinguish syndromic RP from nonsyndromic RP and juvenile RP from LCA, to establish mode of inheritance, and to conclude which gene(s) are most likely consistent with the patient's clinical presentation. In our patient cohort, 12 patients (18%) were heterozygous carriers for unclassified variants in one or more AR retinal genes (Table 4). Therefore, the molecular diagnoses for these patients remain unclear. It could be that the clinical diagnosis is inaccurate, or that the clinical indications or assessments are “incomplete.” A good example is patient 65 (Supplementary Table S3), who visited the ophthalmologist because of isolated RP. His RP panel testing was negative. The follow-up expanded eye gene panel test and whole exome test revealed a homozygous c.1169T>G (p.M390R) mutation in the BBS1 gene. Both parents have been confirmed to be heterozygous carriers for this mutation. The c.1169T>G (p.M390R) mutation has been reported as a common disease-causing mutation in the BBS1 gene. 33,34 Defects in the BBS1 gene are the cause of Bardet-Biedl syndrome 1, characterized by RP, obesity, polydactyly, hypogenitalism, renal malformation, and intellectual disability. Patient 65 did not have obesity and was not intellectually delayed. However, he did have excision of an extra digit, but did not tell the ophthalmologist until results for a BBS1 mutation were revealed. Collectively, the diagnosis for patient 65 is Bardet-Biedl syndrome 1. 
Table 4
 
Relative Proportion of Patients by Mode of Inheritance
Table 4
 
Relative Proportion of Patients by Mode of Inheritance
Inheritance Patient Number This Study, % Gene Review, %
AR 34 52.3 15–25
AD 15 23.1 5–20
XL 4 6.2 5–15
Unknown 12 18.4 40–50
Total 65 100
The Advantage of Capture-Based Enrichment
Capture-based enrichment followed by deep sequencing allows accurate identification not only of all point mutations and small indels, but also of large exonic deletions (case 13, Fig. 2) and large insertions (cases 17–19, Fig. 3). 35 As for the heterozygous exonic deletions, the coverage depth of the deleted exons is reduced to approximately half of the coverage of undeleted exons. 25 In addition, unbiased capture allows the determination of the genomic rearrangements should the junction points reside in the targeted regions. The major advantage of target gene capture followed by the deep MPS approach is the ability to detect exonic deletions, whose breakpoints can be confirmed by Sanger sequencing of the breakpoints or microarray-based Comparative Genomic Hybridization (aCGH). Compared to the multiplexed PCR-based target gene enrichment method, the capture method can avoid any allele dropout due to rare single nucleotide polymorphisms in the primer binding sites resulting in missed mutations or false homozygosity. 36 The Alu insertions that were found in cases 17 through 19 would not have been detected by PCR-based enrichment methods. In addition, heterozygous exonic deletions cannot be revealed by PCR-based enrichment due to the amplification of the undeleted allele. The PCR enrichment step is the primary source of base-composition bias in fragment libraries. 37 Furthermore, approximately 5% of CDS failed to be amplified due to high GC content or secondary structure. 23 Nevertheless, capture-based enrichment has its limitations. It usually has low coverage in exons with high GC content or regions with repetitive sequences, or with highly homologous sequences in other part of the genome. Therefore, PCR and Sanger are required in order to achieve 100% coverage of these regions. 
Detecting Mutations in More Than One Gene
High-throughput parallel sequencing analysis of 66 RP genes greatly increases the diagnostic yield. At the same time, it inevitably increases the probability of identifying mutations or novel variants in one or more genes. As shown in Supplementary Table S2, analyses of 10 out of 12 cases revealed mutations and likely pathogenic variants in genes not previously detected. In our cohort, 13 patients (20%) harbor mutations in two or more genes (Table 2). Notably, patients 12 and 29 were found to have mutations in genes involving AR, AD and XL RP (Table 2). The identification of mutations in multiple genes implicated in different patterns of inheritance makes the interpretation even more complicated. Patient 29 had six mutations identified in four different genes, USH2A, NR2E3, RP1, and RP2. While the detection of two mutations in the USH2A is consistent with this patient's clinical presentation, it is unclear whether there is a synergistic effect of mutations in the NR2E3, RP1, and RP2 genes that might contribute to the clinical severity of the patient. Therefore, careful clinical examination and diligent correlation of molecular information may elucidate novel phenotypes or, indeed, whether one has a primary effect on the presenting disorder or whether multiple genes interacting influence a complex phenotype. Testing at-risk family members for the familial mutations will be helpful to elucidate the clinical significance of each mutation in this family. 
The use of high-throughput MPS to analyze simultaneously a group of genes responsible for RP increases the possibility to identify novel mutations/variants in multiple genes in one patient, especially for the genes not being extensively studied in multiple patient cohorts. The panel test may recognize the mutations with cumulative effect and genes with digenic or polygenic effect. In this study, 34 (52%), 15 (23%), and 4 (6%) individuals among the 65 patients were due to mutations in AR, AD, and 4 XL genes, respectively (Table 4). Compared to the proportion of patients by mode of inheritance in RP GeneReviews (http://www.ncbi.nlm.nih.gov/books/NBK1417/ [in the public domain]), we have a higher fraction of patients with AR RP, a similar proportion of patients with either AD or XL RP, and a lower fraction of unsolved cases (Table 4). 
In summary, the ability to analyze a panel of carefully selected RP genes greatly improves the diagnosis in a cost- and time-efficient manner. This panel serves as an efficient screening tool for most known genes causing nonsyndromic RP. It will not, nor it is intended to, discover novel genes for RP. For those patients whose DNA analysis does not yield a molecular diagnosis by this panel, careful reassessment of clinical and family history, including possible examination of extended family members, followed with array CGH and then WES, may be considered. 
Supplementary Materials
Acknowledgments
We would like to thank the patients for their participation in the study. We are also grateful to the physicians for providing patient information and samples. We thank Janice Smith, PhD, for critical review of this manuscript. 
The authors have no conflict of interest. 
Disclosure: J. Wang, None; V.W. Zhang, None; Y. Feng, None; X. Tian, None; F.-Y. Li, None; C. Truong, None; G. Wang, None; P.-W. Chiang, None; R.A. Lewis, None; L.-J.C. Wong, None 
References
Marmor MF Aguirre G Arden GB Retinitis pigmentosa. A symposium on terminology and methods of examination. Ophthalmology . 1983; 90: 126–131. [CrossRef] [PubMed]
Daiger SP Bowne SJ Sullivan LS. Perspective on genes and mutations causing retinitis pigmentosa. Arch Ophthalmol . 2007; 125: 151–158. [CrossRef] [PubMed]
Haim M. Epidemiology of retinitis pigmentosa in Denmark. Acta Ophthalmol Scand Suppl . 2002; 1–34.
Daiger SP Sullivan LS Bowne SJ. Genes and mutations causing retinitis pigmentosa. Clin Genet . 2013; 84: 132–141. [CrossRef] [PubMed]
Dryja TP McGee TL Hahn LB Mutations within the rhodopsin gene in patients with autosomal dominant retinitis pigmentosa. N Engl J Med . 1990; 323: 1302–1307. [CrossRef] [PubMed]
Kartasasmita A Fujiki K Iskandar E Sovani I Fujimaki T Murakami A. A novel nonsense mutation in rhodopsin gene in two Indonesian families with autosomal recessive retinitis pigmentosa. Ophthalmic Genet . 2011; 32: 57–63. [CrossRef] [PubMed]
Sieving PA Richards JE Naarendorp F Bingham EL Scott K Alpern M. Dark-light: model for nightblindness from the human rhodopsin Gly-90-->Asp mutation. Proc Natl Acad Sci U S A . 1995; 92: 880–884. [CrossRef] [PubMed]
Bowne SJ Humphries MM Sullivan LS A dominant mutation in RPE65 identified by whole-exome sequencing causes retinitis pigmentosa with choroidal involvement. Eur J Hum Genet . 2011; 19: 1074–1081. [CrossRef] [PubMed]
Morimura H Fishman GA Grover SA Fulton AB Berson EL Dryja TP. Mutations in the RPE65 gene in patients with autosomal recessive retinitis pigmentosa or leber congenital amaurosis. Proc Natl Acad Sci U S A . 1998; 95: 3088–3093. [CrossRef] [PubMed]
Dryja TP Hahn LB Kajiwara K Berson EL. Dominant and digenic mutations in the peripherin/RDS and ROM1 genes in retinitis pigmentosa. Invest Ophthalmol Vis Sci . 1997; 38: 1972–1982. [PubMed]
Daiger SP Sullivan LS Gire AI Birch DG Heckenlively JR Bowne SJ. Mutations in known genes account for 58% of autosomal dominant retinitis pigmentosa (adRP). Adv Exp Med Biol . 2008; 613: 203–209. [PubMed]
Goldenberg-Cohen N Banin E Zalzstein Y Genetic heterogeneity and consanguinity lead to a “double hit”: homozygous mutations of MYO7A and PDE6B in a patient with retinitis pigmentosa. Mol Vis . 2013; 19: 1565–1571. [PubMed]
Richards CS Bale S Bellissimo DB ACMG recommendations for standards for interpretation and reporting of sequence variations: revisions 2007. Genet Med . 2008; 10: 294–300. [CrossRef] [PubMed]
Tucker BA Scheetz TE Mullins RF Exome sequencing and analysis of induced pluripotent stem cells identify the cilia-related gene male germ cell-associated kinase (MAK) as a cause of retinitis pigmentosa. Proc Natl Acad Sci U S A . 2011; 108: E569–E576. [CrossRef] [PubMed]
Weston MD Eudy JD Fujita S Genomic structure and identification of novel mutations in usherin, the gene responsible for Usher syndrome type IIa. Am J Hum Genet . 2000; 66: 1199–1210. [CrossRef] [PubMed]
Aller E Larrieu L Jaijo T The USH2A c.2299delG mutation: dating its common origin in a Southern European population. Eur J Hum Genet . 2010; 18: 788–793. [CrossRef] [PubMed]
Eudy JD Weston MD Yao S Mutation of a gene encoding a protein with extracellular matrix motifs in Usher syndrome type IIa. Science . 1998; 280: 1753–1757. [CrossRef] [PubMed]
Bader I Brandau O Achatz H X-linked retinitis pigmentosa: RPGR mutations in most families with definite X linkage and clustering of mutations in a short sequence stretch of exon ORF15. Invest Ophthalmol Vis Sci . 2003; 44: 1458–1463. [CrossRef] [PubMed]
Garcia-Hoyos M Garcia-Sandoval B Cantalapiedra D Mutational screening of the RP2 and RPGR genes in Spanish families with X-linked retinitis pigmentosa. Invest Ophthalmol Vis Sci . 2006; 47: 3777–3782. [CrossRef] [PubMed]
Ferreira PA. Insights into X-linked retinitis pigmentosa type 3, allied diseases and underlying pathomechanisms. Hum Mol Genet . 2005; 14( spec no. 2): R259–R267. [CrossRef] [PubMed]
Song J Smaoui N Ayyagari R High-throughput retina-array for screening 93 genes involved in inherited retinal dystrophy. Invest Ophthalmol Vis Sci . 2011; 52: 9053–9060. [CrossRef] [PubMed]
Clark GR Crowe P Muszynska D Development of a diagnostic genetic test for simplex and autosomal recessive retinitis pigmentosa. Ophthalmology . 2010; 117: 2169–2177, e2163. [CrossRef] [PubMed]
Coppieters F De Wilde B Lefever S Massively parallel sequencing for early molecular diagnosis in Leber congenital amaurosis. Genet Med . 2012; 14: 576–585. [CrossRef] [PubMed]
Wang X Wang H Sun V Comprehensive molecular diagnosis of 179 Leber congenital amaurosis and juvenile retinitis pigmentosa patients by targeted next generation sequencing. J Med Genet . 2013; 50: 674–688. [CrossRef] [PubMed]
Feng Y Chen D Wang G-L Zhang VW Wong L-J. Improved molecular diagnosis by the detection of exonic deletions with target gene capture and deep sequencing. Genet Med . 2014. In press.
Eisenberger T Neuhaus C Khan AO Increasing the yield in targeted next-generation sequencing by implicating CNV analysis, non-coding exons and the overall variant load: the example of retinal dystrophies. PLoS ONE . 2013; 8: e78496. [CrossRef] [PubMed]
Neveling K Collin RW Gilissen C Next-generation genetic testing for retinitis pigmentosa. Hum Mutat . 2012; 33: 963–972. [CrossRef] [PubMed]
Shanks ME Downes SM Copley RR Next-generation sequencing (NGS) as a diagnostic tool for retinal degeneration reveals a much higher detection rate in early-onset disease. Eur J Hum Genet . 2013; 21: 274–280. [CrossRef] [PubMed]
Fahim AT Daiger SP Weleber RG. Retinitis pigmentosa overview. In: Pagon RA MP Adam Bird TD , eds. GeneReviews . Seattle, WA: University of Washington; 1993.
Khan MI Kersten FF Azam M CLRN1 mutations cause nonsyndromic retinitis pigmentosa. Ophthalmology . 2011; 118: 1444–1448. [CrossRef] [PubMed]
Rivolta C Sweklo EA Berson EL Dryja TP. Missense mutation in the USH2A gene: association with recessive retinitis pigmentosa without hearing loss. Am J Hum Genet . 2000; 66: 1975–1978. [CrossRef] [PubMed]
Marlhens F Bareil C Griffoin JM Mutations in RPE65 cause Leber's congenital amaurosis. Nat Genet . 1997; 17: 139–141. [CrossRef] [PubMed]
Mykytyn K Nishimura DY Searby CC Identification of the gene (BBS1) most commonly involved in Bardet-Biedl syndrome, a complex human obesity syndrome. Nat Genet . 2002; 31: 435–438. [PubMed]
O'Sullivan J Mullaney BG Bhaskar SS A paradigm shift in the delivery of services for diagnosis of inherited retinal disease. J Med Genet . 2012; 49: 322–326. [CrossRef] [PubMed]
Wang J Cui H Lee NC Clinical application of massively parallel sequencing in the molecular diagnosis of glycogen storage diseases of genetically heterogeneous origin. Genet Med . 2013; 15: 106–114. [CrossRef] [PubMed]
Landsverk ML Douglas GV Tang S Diagnostic approaches to apparent homozygosity. Genet Med . 2012; 14: 877–882. [CrossRef] [PubMed]
Aird D Ross MG Chen WS Analyzing and minimizing PCR amplification bias in Illumina sequencing libraries. Genome Biol . 2011; 12: R18. [CrossRef] [PubMed]
Figure 1
 
Analytical pipeline for variant annotation and classification. Estimated time needed for each step and the approximate number of variants at each analysis step are shown.
Figure 1
 
Analytical pipeline for variant annotation and classification. Estimated time needed for each step and the approximate number of variants at each analysis step are shown.
Figure 2
 
EYS exonic deletion detected by MPS and confirmed by array CGH. (A) MPS exonic coverage profile for the EYS gene. Exon 14 had no coverage, suggesting a homozygous deletion. (B) Targeted array CGH detected a loss of copy number in exon 14. No copy number variations were detected in other exons of the EYS gene. (C) Enlarged view of exon 14 from array CGH. The log2 ratios for the probes in the deleted region are consistent with a homozygous deletion.
Figure 2
 
EYS exonic deletion detected by MPS and confirmed by array CGH. (A) MPS exonic coverage profile for the EYS gene. Exon 14 had no coverage, suggesting a homozygous deletion. (B) Targeted array CGH detected a loss of copy number in exon 14. No copy number variations were detected in other exons of the EYS gene. (C) Enlarged view of exon 14 from array CGH. The log2 ratios for the probes in the deleted region are consistent with a homozygous deletion.
Figure 3
 
Alu insertion in the MAK gene was identified by MPS and confirmed by PCR and Sanger sequence. (A, B) MPS data showing misaligned reads from patient 18 (A) and his unaffected daughter (B). The next-generation sequencing raw data themselves cannot tell what the sequence variation was or the zygosity. However, the misaligned reads suggested the possible sequence aberration. (C) Pedigree and schematic Alu insertion in exon 10. Expected PCR product length is 367 bp for the normal allele and is ∼722 bp for the allele with Alu insertion. A single ∼722-bp application band detected in patient 18 indicated the homozygous insertion; a normal band and an insertion band detected in daughter of patient 18 indicated that she is a heterozygous carrier for the Alu insertion. Sanger sequencing of the 722 bp PCR product confirmed an insertion of ∼355 bp Alu sequence (data not shown).
Figure 3
 
Alu insertion in the MAK gene was identified by MPS and confirmed by PCR and Sanger sequence. (A, B) MPS data showing misaligned reads from patient 18 (A) and his unaffected daughter (B). The next-generation sequencing raw data themselves cannot tell what the sequence variation was or the zygosity. However, the misaligned reads suggested the possible sequence aberration. (C) Pedigree and schematic Alu insertion in exon 10. Expected PCR product length is 367 bp for the normal allele and is ∼722 bp for the allele with Alu insertion. A single ∼722-bp application band detected in patient 18 indicated the homozygous insertion; a normal band and an insertion band detected in daughter of patient 18 indicated that she is a heterozygous carrier for the Alu insertion. Sanger sequencing of the 722 bp PCR product confirmed an insertion of ∼355 bp Alu sequence (data not shown).
Figure 4
 
Relationship between average coverage and number of CDS with coverage lower than 20× in nine randomly selected samples. Reducing coverage from 1000× to 100× in the RP panel described here, the number of <20× covered CDS (<20×) increased from 6.7 (0.7%) to 72 (7.7%).
Figure 4
 
Relationship between average coverage and number of CDS with coverage lower than 20× in nine randomly selected samples. Reducing coverage from 1000× to 100× in the RP panel described here, the number of <20× covered CDS (<20×) increased from 6.7 (0.7%) to 72 (7.7%).
Table 1
 
Genes Included in the RP Panel and Summary of Capture and Sequencing
Table 1
 
Genes Included in the RP Panel and Summary of Capture and Sequencing
Next-Generation Sequencing Panel Retinitis Pigmentosa Panel—66 Genes
Genes included in this panel ABCA4, ABHD12, AIPL1, BEST1, C2orf71, C8orf37, CA4, CDHR1, CEP290, CERKL, CLRN1, CNGA1, CNGB1, CRB1, CRX, DHDDS, EYS, FAM161A, FLVCR1, FSCN2, GUCA1B, GUCY2D, IDH3B, IMPDH1, IMPG2, KLHL7, LCA5, LRAT, MAK, MERTK, MFRP, NR2E3, NRL, PDE6A, PDE6B, PDE6G, PRCD, PRKCG, PROM1, PRPF3, PRPF31, PRPF6, PRPF8, PRPH2, RBP3, RD3, RDH12, RGR, RHO, RLBP1, ROM1, RP1, RP2, RP9, RPE65, RPGR, RPGRIP1, SAG, SEMA4A, SNRNP200, SPATA7, TOPORS, TTC8, TULP1, USH2A, ZNF513
Number of CDS 939
Targeted size 202,800 bases (coding exon sequences ± 20 bp)
Average coverage Approximately 1000× per base
Enrichment In solution capture library (NimbleGen)
Sequencing information Illumina HiSeq2000, 100 cycle, single-end
Table 2
 
Mutations and Unclassified Variants Identified in Previously Undiagnosed Patients
Table 2
 
Mutations and Unclassified Variants Identified in Previously Undiagnosed Patients
Patients Sex Age, y Clinical Diagnosis Genes Protein Allele 1 Allele 2
Autosomal recessive
1 F 19 Pigmentary retinal dystrophy ABCA4 (AR) ATP-binding cassette transporter—retina c.4469G>A (p.C1490Y) c.4469G>A (p.C1490Y)
2* F 27 Decreased vision ABCA4 (AR) ATP-binding cassette transporter—retina c.1365+5_+6insC c.1927G>A (p.V643M) c.3602T>G (p.L1201R)
USH2A (AR) Usherin c.10858A>G (p.I3620V) c.15377T>C (p.I5126T)
3* M 11 Bull's-eye maculopathy, cone dystrophy ABCA4 (AR) ATP-binding cassette transporter—retina c.3385C>T (p.R1129C) c.5461-10T>C
USH2A (AR) Usherin c.10073G>A (p.C3358Y) c.11026A>G (p.T3676A) c.14753C>T (p.T4918M)
CA4 (AD) Carbonic anhydrase IV c.700G>A (V234I) Neg
4* F 14 Rod-cone dystrophy ABCA4 (AR) ATP-binding cassette transporter—retinaChromosome 8 open reading frame 37 c.2588G>C (p.G863A) c.2828G>A (p.R943Q)
C8orf37 (AR) c.243+3A>C c.243+3A>C
5 M 39 Cone dystrophy ABCA4 (AR) ATP-binding cassette transporter—retina c.5196+1G>A c.5882G>A (p.G196E) c.2828G>A (p.R943Q)
6 F 1 mo RP ABCA4 (AR) ATP-binding cassette transporter—retina c.2828G>A (p.R943Q) c.2588G>C (p.G863A)
7* F 25 RP CEP290 (AR) Centrosomal protein 290 c.1078C>T (p.R360*) c.6851_6855delCTGAT (p.T2284Nfs*10)
AIPL1 (AR) Arylhydrocarbon-interacting receptor protein-like 1 c.140C>G (p.T47R) c.937G>T (p.A313S)
8 F 1 Senior-Loken syndrome CEP290 (AR) Centrosomal protein 290 c.2587-2A>G c.2587-2A>G
9 F 35 RP, muscle weakness CEP290 (AR) Centrosomal protein 290 c.297+1G>T c.6401T>C (p.I2134T)
10 M 26 Decreased vision, field loss, night vision blindness CERKL (AR) Ceramide kinase-like protein c.812T>C (p.I271T) c.812T>C (p.I271T)
11 M 20 Progressive central and peripheral vision loss, abnormal constructed visual field CERKL (AR) Ceramide kinase-like protein c.598A>T (p.K200*) c.769C>T (p.R257*)
12* M 35 RP, loss of peripheral vision CERKL (AR) Ceramide kinase-like protein c.1462G>A (p.E488*) c.1462G>A (p.E488*)
NR2E3 (AD, AR) Nuclear receptor subfamily 2 group E3 c.932G>A (p.R311Q) Neg
RP2 (XL) RP2 protein c.844C>T (p.R282W) NA
EYS (AR) Eyes shut/spacemaker (Drosophila) homolog c.1765A>G (p.R589G) Neg
13 F 48 Decreased peripheral vision, photophobia, difficult light-to-dark adaptation, night blindness EYS (AR) Eyes shut/spacemaker (Drosophila) homolog exon 14 del exon 14 del
14* M 18 Bone spicule pigmentary clumping in fundus, VA = 20/20–25, peripheral field constriction EYS (AR) Eyes shut/spacemaker (Drosophila) homolog c.5600C>T (p.S1867F) c.8422G>A (p.A2808T)
BEST (AD) Bestrophin 1 c.714+13C>T Neg
15 M 24 Decreased peripheral vision, night blindness EYS (AR) Eyes shut/spacemaker (Drosophila) homolog c.5681T>C (p.L1894P) c.8779T>C (p.C2927R)
16 M 31 RP FAM161A (AR) Family with sequence similarity 161 member A c.1309A>T (p.R437*) c.1501delT(p.C501Vfs*4)
17 F 77 RP MAK (AR) Male germ cell-associated kinase c.1297_1298 insAlu c.1297_1298 insAlu
18 M 76 RP MAK (AR) Male germ cell-associated kinase c.1297_1298 insAlu c.1297_1298 insAlu
19 M 68 RP MAK (AR) Male germ cell-associated kinase c.1297_1298 insAlu c.1297_1298 insAlu
20 M 18 Pigmentary retinal dystrophy MERTK (AR) c-mer protooncogene receptor tyrosine kinase c.62-1G>A c.1296+1G>A
21 F 20 Hereditary retinal dystrophy, rod-cone dystrophy MERTK (AR) c-mer protooncogene receptor tyrosine kinase c.1951C>T (p.R651*) c.345C>G (p.C115W)
22* F 45 RP MFRP (AR) Membrane-type frizzled-related protein c.772+2T>G c.1124+1G>T
CERKL (AR) Ceramide kinase-like protein c.769C>T (p.R257*) Neg
23 M 35 RP PDE6B (AR, AD) Rod cGMP phosphodiesterase beta subunit c.1833-1G>C c.1833-1G>C
24 F 21 Nyctalopia, reduced peripheral vision PDE6B (AR, AD) Human homolog of yeast pre-mRNA splicing factor 6 phosphodiesterase beta subunit c.869G>A (p.W290*) c.1280G>A (p.W427*)
25 M 35 NA PDE6G (AR) Phosphodiesterase 6G cGMP-specific rod gamma c.109C>T (p.Q37*) c.109C>T (p.Q37*)
26 M  9 Usher syndrome type II USH2A (AR) Usherin c.2299delG (p.E767Sfs*21) c.4714C>T (p.L1572F) c.7595-3C>G
27 F 39 Nyctalopia, decreased peripheral vision, history of plaquenil USH2A (AR) Usherin c.3395G>A (p.G1132D) c.5624A>G (p.N1875S)
28 M 17 Usher syndrome USH2A (AR) Usherin c.5776+1G>A c.14131C>T (p.Q4711*)
29* M 36 Usher syndrome USH2A (AR) Usherin c.1036A>C (p.N346H) c.2299delG (p.E767Sfs*21) c.4714C>T (p.L1572F)
NR2E3 (AD, AR) Nuclear receptor subfamily 2 group E3 c.767C>A (p.A256E) Wild type
RP1 (AD, AR) RP1 protein c.1118C>T (p.T373I) Wild type
RP2 (XL) RP2 protein c.844C>T (p.R282W) NA
30* M 37 RP USH2A (AR) Usherin c.2299delG (p.E767Sfs*21) c.2276G>T (p.C759F) c.4714C>T (p.L1572F)
RP IMPDH1 (AD) Inosine monophosphate dehydrogenase 1 c.1057G>A (p.V353I) Wild type
31* F 32 RP MERTK (AR) c-mer protooncogene receptor tyrosine kinase c.2164C>T (p.R722*) c.2219C>T (p.A740V)
CA4 (AD) Carbonic anhydrase IV c.198_199delACinsG (p.L67Wfs*24) Wild type
32 F 51 RP ABCA4 (AR) ATP-binding cassette transporter—retina c.766G>T (p.V256L) c.5755G>T (p.D1919Y)
33 F 8 Progressive pigmentary retinopathy CEP290 (AR) Centrosomal protein 290 c.4834_4835delAC (p.T1612Sfs*13) c.2980G>A (p.E994K)
34 M 11 Peripheral dystrophy EYS (AR) Eyes shut/spacemaker (Drosophila) homolog c.3443+1G>T c.2412G>C (p.Q804H) c.3250A>C (p.T1084P) c.4402G>C (p.D1468H)
Autosomal dominant
35 F 64 Progressive pigmentary retinopathy FSCN2 (AD) Retinal fascin homolog 2, actin bundling protein c.467C>T (p.P156L) Wild type
36* M 29 RP FSCN2 (AD) Retinal fascin homolog 2, actin bundling protein c.72delG (p.T25Qfs*120) Wild type
IMPDH1 (AD) Inosine monophosphate dehydrogenase 1 c.569G>T (p.R190L) Wild type
37 F 54 RP PDE6B (AR, AD) Human homolog of yeast pre-mRNA splicing factor 6phosphodiesterase beta subunit c.973G>C (p.E325Q) Wild type
38 M  7 Rod dystrophy PRPF8 (AD) Human homolog of yeast pre-mRNA splicing factor C8 c.6961C>T (p.Q2321*) Wild type
39 M 55 RP PRPH2 (AD) Peripherin 2 c.422A>G (p.Y141C) Wild type
40 F 70 RP RHO (AD) Rhodopsin c.936+1G>T Wild type
41* F  7 RP, bone spicules, optic disc pallor, nyctalopia, RHO (AD) Rhodopsin c.697-11G>A Wild type
RP, bone spicules, optic disc pallor, nyctalopia PRPF6 (AD) Human homolog of yeast pre-mRNA splicing factor 6 c.867-7C>G c.2431+9delG
42 M 19 RP SNRNP200 (AD) Small nuclear ribonucleoprotein 200 kDa c.2359G>A (p.A787T) Wild type
43 M 11 RP, likely AD inheritance TOPORS (AD) Topoisomerase I binding arginine/serine rich protein c.74C>G(p.S25W) Wild type
44 F 22 Nyctalopia, loss of peripheral vision, progressive pigmentary retinopathy NR2E3 (AD, AR) Nuclear receptor subfamily 2 group E3 c.119-2A>C c.119-2A>C
45 F 14 Retinal dystrophy PDE6B (AR, AD) Human homolog of yeast pre-mRNA splicing factor 6phosphodiesterase beta subunit c.2193+1G>A c.1624C>T (p.R542W)
46* M 45 Night blindness, visual field peripheral restriction, decreased vision, ERG consistent with RP RP1 (AD, AR) RP1 protein c.4196delG (p.C1399Lfs*5)c.6353G>A (p.S2118N) c.4196delG (p.C1399Lfs*5)c.6353G>A (p.S2118N)
47 M  6 Juvenile RP/LCA, nonrecordable electoretinogram, retinopathy RP1 (AD, AR) RP1 protein c.796delinsTA(p.H266*) c.1625C>G (p.S542*)
48* M 41 NA RPE65 (AD, AR) Retinal pigment epithelium-specific 65-kDa protein c.1597T>A (p.S533T) c.89T>C (p.V30A)
AIPL1 (AR) Arylhydrocarbon-interacting receptor protein-like 1 c.905G>T (p.R302L) Neg
CNGA1 (AR) Rod cGMP-gated channel alpha subunit c.521T>A (p.L174*) Neg
49 F 54 RP, macular subatrophy, pseudohole RHO (AD) Rhodopsin c.562G>A (p.G188R) Neg
X-linked
50 M  5 RP RP2 (XL) RP2 protein c.358C>T (p.R120*) NA
51 M 15 RP RPGR (XL) Retinitis pigmentosa GTPase regulator c.1202_1206del5 (p.V401Afs*50) NA
52 M 18 RP RPGR (XL) Retinitis pigmentosa GTPase regulator c.92G>A (p.W31*) NA
53 M 44 Rod-cone dystrophy RPGR ORF15 (XL) Retinitis pigmentosa GTPase regulator c.3178_3179delAG (p.E1060Rfs*18) NA
Table 3
 
Diagnostic Yield Comparison in Different Platforms
Table 3
 
Diagnostic Yield Comparison in Different Platforms
Authors Platform Coverage Depth Targeted Genes/CDS No. of Patients Positive/Total Diagnostic Yield Disease
Song et al.21 Affymetrix resequencing chip NA  93/1470 19 patients with known mutation Inherited retinal dystrophy
Clark et al.22 Resequencing chip APEX NA 19/100 9/56 16% arRP
Shanks et al.28 NimbleGen 12-plex capture array, 454GS FLX 918–1239×  73/1413 9/36 25% RP
Coppieters at al.23 PCR capture, GAII 24–117×, avg. 58× 16/252 3/17 18% LCA
Wang et al.24 NimbleGen SeqCap, HiSeq2000 20–30×, avg. 26× 163/2560 72/179 40% RP + LCA
Neveling et al.27 NimbleGen 12-plex capture array, 454GS FLX 62× 111/2011 36/100 36% RP
Eisenberger et al.26 NimbleGen SeqCap, 454GS FLX and MiSeq 75× (GS FLX) 250× (MiSeq) 55/876 88/126 70% RP + LCA
This study NimbleGen SeqCap, HiSeq2000 1000× 66/939 53/65 82% RP
Table 4
 
Relative Proportion of Patients by Mode of Inheritance
Table 4
 
Relative Proportion of Patients by Mode of Inheritance
Inheritance Patient Number This Study, % Gene Review, %
AR 34 52.3 15–25
AD 15 23.1 5–20
XL 4 6.2 5–15
Unknown 12 18.4 40–50
Total 65 100
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×