Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Whole exome sequencing as a diagnostic tool for patients with ciliopathy-like phenotypes

  • Sheila Castro-Sánchez,

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Software, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliations Departamento de Bioquímica, Genética e Inmunología, Facultad de Biología, Universidad de Vigo, Vigo, Spain, Research Group of Rare Diseases & Pediatric Medicine, Instituto de Investigación Sanitaria Galicia Sur (IISGS), SERGAS-UVIGO, Hospital Álvaro Cunqueiro, Vigo, Spain, Centro de Investigaciones Biomédicas (CINBIO) (Centro Singular de Investigación de Galicia), Universidad de Vigo, Vigo, Spain

  • María Álvarez-Satta,

    Roles Formal analysis, Funding acquisition, Investigation, Validation, Writing – review & editing

    Affiliations Departamento de Bioquímica, Genética e Inmunología, Facultad de Biología, Universidad de Vigo, Vigo, Spain, Research Group of Rare Diseases & Pediatric Medicine, Instituto de Investigación Sanitaria Galicia Sur (IISGS), SERGAS-UVIGO, Hospital Álvaro Cunqueiro, Vigo, Spain, Centro de Investigaciones Biomédicas (CINBIO) (Centro Singular de Investigación de Galicia), Universidad de Vigo, Vigo, Spain

  • Mohamed A. Tohamy,

    Roles Formal analysis, Investigation, Validation, Writing – review & editing

    Current address: Department of Biochemistry, Faculty of Pharmacy, Beni-Suef University, Beni-Suef, Egypt

    Affiliation Departamento de Bioquímica, Genética e Inmunología, Facultad de Biología, Universidad de Vigo, Vigo, Spain

  • Sergi Beltran,

    Roles Data curation, Formal analysis, Investigation, Methodology, Resources, Software, Validation, Writing – review & editing

    Affiliations CNAG-CRG, Centre for Genomic Regulation (CRG), Barcelona Institute of Science and Technology (BIST), Barcelona, Spain, Universitat Pompeu Fabra (UPF), Barcelona, Spain

  • Sophia Derdak,

    Roles Data curation, Formal analysis, Investigation, Methodology, Software, Validation, Writing – review & editing

    Affiliations CNAG-CRG, Centre for Genomic Regulation (CRG), Barcelona Institute of Science and Technology (BIST), Barcelona, Spain, Universitat Pompeu Fabra (UPF), Barcelona, Spain

  • Diana Valverde

    Roles Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing – original draft, Writing – review & editing

    dianaval@uvigo.es

    Affiliations Departamento de Bioquímica, Genética e Inmunología, Facultad de Biología, Universidad de Vigo, Vigo, Spain, Research Group of Rare Diseases & Pediatric Medicine, Instituto de Investigación Sanitaria Galicia Sur (IISGS), SERGAS-UVIGO, Hospital Álvaro Cunqueiro, Vigo, Spain, Centro de Investigaciones Biomédicas (CINBIO) (Centro Singular de Investigación de Galicia), Universidad de Vigo, Vigo, Spain

    ORCID http://orcid.org/0000-0002-7024-1657

Whole exome sequencing as a diagnostic tool for patients with ciliopathy-like phenotypes

  • Sheila Castro-Sánchez, 
  • María Álvarez-Satta, 
  • Mohamed A. Tohamy, 
  • Sergi Beltran, 
  • Sophia Derdak, 
  • Diana Valverde
PLOS
x

Abstract

Ciliopathies are a group of rare disorders characterized by a high genetic and phenotypic variability, which complicates their molecular diagnosis. Hence the need to use the latest powerful approaches to faster identify the genetic defect in these patients. We applied whole exome sequencing to six consanguineous families clinically diagnosed with ciliopathy-like disease, and for which mutations in predominant Bardet-Biedl syndrome (BBS) genes had previously been excluded. Our strategy, based on first applying several filters to ciliary variants and using many of the bioinformatics tools available, allowed us to identify causal mutations in BBS2, ALMS1 and CRB1 genes in four families, thus confirming the molecular diagnosis of ciliopathy. In the remaining two families, after first rejecting the presence of pathogenic variants in common cilia-related genes, we adopted a new filtering strategy combined with prioritisation tools to rank the final candidate genes for each case. Thus, we propose CORO2B, LMO7 and ZNF17 as novel candidate ciliary genes, but further functional studies will be needed to confirm their role. Our data show the usefulness of this strategy to diagnose patients with unclear phenotypes, and therefore the success of applying such technologies to achieve a rapid and reliable molecular diagnosis, improving genetic counselling for these patients. In addition, the described pipeline also highlights the common pitfalls associated to the large volume of data we have to face and the difficulty of assigning a functional role to these changes, hence the importance of designing the most appropriate strategy according to each case.

Introduction

Over the last 15 years, our knowledge about the group of human genetic diseases called ciliopathies has rapidly grown [1, 2]. These multisystemic disorders, characterized by ciliary dysfunction, are caused by mutations in highly conserved genes mainly involved in the correct assembly and maintenance of cilia [3, 4].

It is well-known that one ciliary gene can be involved in the development of two or more distinct disorders, such as MKKS, which may be related to McKusick-Kauffman syndrome (MKKS, MIM #236700) or Bardet-Biedl syndrome (BBS, MIM #209900). This genetic heterogeneity, together with the high phenotypic variability, overlapping phenotypes and the progressive onset of many features during childhood and adolescence, complicates both the clinical and molecular diagnosis [1, 5].

In this respect, genetic tools have evolved considerably over the years. These syndromes were initially confirmed by the time-consuming Sanger sequencing, often combined with linkage analysis and autozygosity mapping strategies, or by DNA microarrays, which are restricted to a previously defined set of polymorphisms and rarely allow for the detection of rare/novel mutations. Thus, the shortcomings associated with these methods have encouraged most researchers to take advantage of next generation sequencing technologies [6, 7]. These approaches have emerged as a powerful way to improve molecular diagnosis in ciliopathy-related families.

Today, the ever lower cost of these techniques has promoted the implementation of these tools in molecular diagnosis laboratories and has led to an increased identification of novel ciliary-related genes in the last few years [8, 9]. It is worthwhile using WES as diagnostic tool since it allows us to get the mutational screening of nearly all coding regions of the genome at once, especially in such heterogeneous diseases as most of ciliopathies [4, 7, 10, 11].

However, the main challenge is how to interpret the great volume of resulting data, since variants with clinical relevance are difficult to pinpoint considering the high number of polymorphisms and possible false positives [5, 10]. This arduous task is being addressed thanks to the range of bioinformatics tools which help to adopt different strategies to narrow down the list of candidate variants [12, 13], rejecting and prioritizing variations by applying different filters based on (i) mode of inheritance, (ii) absence in public databases, (iii) predicted pathogenicity, (iv) role of the affected gene in pathways of interest or (v) impact on protein-protein interactions, among others.

The described identification of genes strongly related to cilia and the need to obtain an accurate diagnosis prompted us to perform whole exome sequencing (WES) in six consanguineous families with a clinical diagnosis of ciliopathy-like disorders, primarily BBS. The main objective of this study is to analyse massive sequencing data from these affected families following a strategy based on different filters like those mentioned above, as well as the use of some prioritisation tools (as described in “Materials and methods” section) on a reduced list of candidate genes, enhancing the possibilities to identify the causative mutation in each case.

Materials and methods

Patients

Six patients from unrelated families with clinical diagnosis of cilia-related disease were included. Consanguineous relationships have been described for all these families, which are from Western Europe, Africa and Asia. This group of patients consisted of five women and one man. When pedigree information was acquired, peripheral blood from all participants and available family members was collected for DNA extraction using the Flexigene DNA kit 250 (Qiagen, Hilden, Germany), following the manufacturer’s protocol. After a prior analysis by direct sequencing to exclude mutations in predominant BBS genes, BBS1 and BBS10, and also BBS12 according to our cohort data [14], the selected patients were studied by WES. In case 5, MKKS/BBS6 gene was also previously sequenced considering the phenotype displayed. Genomic DNA fulfilled the quality criteria required for WES.

This study was approved by the Galician Ethical Committee for Clinical Research (Spain—no.2006/08) and adhered to the tenets of the Declaration of Helsinki. Written informed consent was obtained from all patients or their guardians.

Library preparation and sequencing

Exome sequencing and analysis was performed at the Centro Nacional de Análisis Genómico (CNAG-CRG, Barcelona, Spain). For exome enrichment the NimbleGen SeqCap EZ v3.0 system following manufacturer's protocol version 4.2 was used and pre-capture multiplexing was applied. Briefly, 1 μg of genomic DNA was fragmented with Covaris E210 and used for ligation of the adapters containing Illumina specific indexes with a KAPA Library Preparation kit (Kapa Biosystems). Adapter ligated DNA fragments were enriched by 7 cycles of pre-capture PCR using KAPA HiFi HotStart ReadyMix (2X) (Kapa Biosystems) and analysed on an Agilent 2100 Bioanalyzer with the DNA 1000 assay. Five libraries were pooled with a combined mass of 1250 ng for the baits hybridisation step (47°C; 68 h). After washing (47°C), multiplexed captured library was recovered with Capture beads and amplified with 14 cycles of post-capture PCR using KAPA HiFi HotStart ReadyMix (2X). Size, concentration and quality of the captured library were determined using an Agilent DNA 1000 chip. The success of the enrichment was measured by qPCR SYBR Green assay on a Roche LightCycler® 480 Instrument evaluating one genomic locus with pre- and post-captured material.

Each library pool was sequenced on an Illumina HiSeq 2000 instrument in a fraction of a sequencing lane following the manufacturer’s protocol, with a paired end run of 2x101bp. Image analysis, base calling and quality scoring of the run were processed using the manufacturer’s software Real Time Analysis (RTA 1.13.48) and followed by generation of FASTQ sequence files by CASAVA.

Data analysis

Sequencing reads were trimmed from the 3’ end up to the first base with a Phred quality >9 and were mapped to the Human Genome Reference v37 with decoy sequences (Broad Institute), using GEM [15]. BAM files containing only properly paired and uniquely mapped reads were processed with picard tools v1.110 to remove duplicates, and local realignment was performed with the Genome Analysis Tool Kit (GATK) v3.1 [16]. Samtools v0.1.19 [17] was used on the processed BAM files to call single nucleotide variants (SNVs) and small insertion deletions (INDELs). Functional annotations from Ensembl release 75 [18] were added to the resulting Variant Call Format (VCF) file using snpEff [19]. snpSift [20] was used to add information from dbSNP v137 [21], the 1000 Genomes Project (1000GP) [22], the NHLBI Exome Sequencing Project [Exome Variant Server, NHLBI GO Exome Sequencing Project (ESP), Seattle, WA] and a variety of conservation and deleteriousness predictions included in dbNSFP v2.5 [23].

Prioritisation of genetic variants

This process was carried out following a strategy of our own design to reach a small number of candidate variants (Fig 1). Homozygous variants in known cilia-related genes were analysed first, as well as compound heterozygous variants in these genes, and then, we proceeded in the same way with homozygous variants in other genes. Only positions with a coverage of at least 15X and a genotype quality >20 (indicating only confident positions) were considered. Subsequently, variants with a minor allele frequency >0.01 in the dbSNP, and also with an alternative allele frequency >0.01 in NHLBI ESP or 1000GP databases were excluded. Next, based on functional annotation, variants with a snpEff predicted high or moderate effect were kept, whereas synonymous coding and intronic variants not associated with splice site alterations were excluded. Finally, the variants selected in the previous step were evaluated with several in silico prediction algorithms to evaluate the predicted effect at protein level (PolyPhen-2 [24], SIFT [25], Mutation Taster [26] and the likelihood ratio test (LRT) [27]). Once potentially pathogenic variants were selected, COBALT [28] was used to analyse protein residue conservation across species. Additionally, Endeavour [29] and ToppGene Suite [30] prioritisation tools were used to rank the final list of variants, with a training set of 51 known ciliary genes (S1 Table). For those top genes, STRING [31] and GENEMANIA [32] tools were used to elucidate possible co-expression and/or interactions between the corresponding encoded proteins and other ciliary proteins. When appropriate, the potential effect on splice sites was assessed using several prediction tools: NNSplice [33], NetGene2 [34], Human Splicing Finder [35] and ASSEDA [36], all with default settings. Finally, WES data were used to check if candidate variants localize in runs of homozygosity (ROH) regions, what is expected in consanguineous cases. Thus, homozygosity mapping with exome data was carried out with PLINK v1.9 [37] using optimized settings for Exome data [38].

Confirmation of variants and segregation analysis

Selected candidate causative variants were verified by PCR amplification of the corresponding coding and adjacent intronic regions following standard protocols. Purified DNA fragments were directly sequenced using BigDye Terminator v.3.1 Cycle Sequencing Kit (Life Technologies, Foster City, CA, USA) and analysed on an ABI PRISM 3130 genetic analyser (Life Technologies, Foster City, CA, USA). Validated mutations were then analysed in the available family members to confirm co-segregation.

Results

Clinical findings

All the cases included in this study were clinically diagnosed with some type of ciliopathy, mainly BBS. Case 1 is a Caucasian female fulfilling diagnostic criteria for BBS, showing five out of the six primary features established for this disease: retinal dystrophy, obesity, polydactyly, urogenital anomalies and cognitive impairment [39]. Case 2, a female of Algerian origin, also displayed the same BBS primary features as case 1, in addition to several secondary features such as brachydactyly in four limbs, syndactyly in feet, craniofacial anomalies, psychomotor and speech delay, late menarche, hearing loss and hypothyroidism. Case 3 is an Indian male with clinically diagnosed BBS, showing mild retinal dystrophy, obesity, postaxial polydactyly in four limbs, pulmonary artery stenosis and some craniofacial defects. Case 4, an Iranian female, displayed retinal dystrophy, mild obesity, cognitive impairment and several facial anomalies (deep-set eyes, orbital hypertelorism and downward slanting palpebral fissures, among others), who did not strictly fulfilled diagnostic criteria but seemed to be closely related to BBS. Case 5, a Moroccan female child, was diagnosed with MKKS at birth since she showed postaxial polydactyly and hydrometrocolpos, together with vaginal agenesis. Case 6 belongs to a Caucasian family with an initial clinical diagnosis of BBS and consists of two affected siblings. At the time of submitting all samples to WES, only the sample of one of the siblings was available, a female who displayed retinal dystrophy and obesity. While WES data was being analysed, the phenotype reassessment of this family by clinicians only confirmed retinal dystrophy in the studied patient, whereas her brother was seen to fulfil diagnostic criteria of a more complex ciliopathy like BBS, showing retinal dystrophy, obesity, polydactyly and cognitive impairment. The main clinical data are summarized in Table 1.

thumbnail
Table 1. Summary of clinical data and molecular findings of the patients included in this study.

https://doi.org/10.1371/journal.pone.0183081.t001

Exome sequencing reveals causative mutations in cilia-related genes

Given the consanguineous nature of our families, we expected a homozygous mutation to be the cause of the disease in all cases, but the responsible gene not to be the same given the different phenotypes.

The filtering strategy based on first considering only homozygous variants in the most common ciliopathy genes, then selecting those with good sequencing quality and high effect on coding sequence, and excluding those with a high frequency in public databases, allowed us to identify three candidate variants in three different cases. Two homozygous nonsense variants were found in BBS2 gene (RefSeq NM_031885.3): c.565C>T; p.(Arg189*) (exon 5) and c.1932T>A; p.(Tyr644*) (exon 16), a novel mutation, in cases 2 and 3, respectively. Another homozygous nonsense variant, c.8005C>T; p.(Arg2669*) (exon 10), was detected in ALMS1 gene (RefSeq NM_015120.4) in case 4. Although we assumed homozygous nonsense variants to be pathogenic, we checked MutationTaster and LRT, obtaining a deleterious prediction for all three variants (Table 2). The two BBS2 variants are located in residues which are highly conserved across species. BBS2 encodes a 722-amino acid protein; therefore both nonsense BBS2 variants would lead to a shorter protein whose function could presumably be altered or even be subject to nonsense-mediated decay (NMD). The same occurs with the ALMS1 variant, in which the length of the normal 4169-amino acid protein would be reduced by almost half. After that, these variants were validated by Sanger sequencing and then, the segregation analysis, which was performed in available members of cases 2 and 3 (no family members available for case 4), confirmed the recessive inheritance of the mutations identified in both cases (Table 1). Anyway, compound heterozygosity in known ciliary genes was also checked, without finding convincing results.

Identification of novel candidate genes

In the remaining three families, no candidate causal variants were detected in the most commonly mutated cilia-related genes. Then, we extended the mutational search to all WES data, focusing again on a selection of homozygous variants fulfilling the same criteria proposed for the previous families as a starting point. On the resulting file containing a set of positions (~300) for each case, a series of exclusion steps were followed to obtain the final candidate variants.

Case 1.

In this case, no candidate variants with a snpEff predicted high effect were found after following the general filtering strategy, so different filters were applied: (a) only variants with a predicted moderate effect associated (modifier or low effects excluded), (b) very low population frequency (<0.005) in public databases, and (c) at least two (out of four) positive predictions with functionality prediction tools. Only seven variants in seven different genes fulfilled all of these criteria; the genes were ranked with Endeavour and Toppgene based on functional similarity to a training gene list including those human ciliary genes reported in the SYSCILIA gold standard v1 [43]. CORO2B (coronin 2B) was the top candidate in both algorithms, followed by SLC3A1 (solute carrier family 3 member 1) gene. To improve the robustness of these results, we also tried with subsets of training genes associated with BBS and MKS separately. For the first case, SLC3A1 was the top gene followed by CORO2B, whereas for the second case the opposite occurred. Finally, we decided to train Toppgene with a set of 51 common ciliary genes, also included in SYSCILIA (S1 Table). SLC3A1 was again the top gene in this setting, with CORO2B in the second position. Cildb database [44] was also consulted to gain more information about these genes not previously related to any ciliopathy. While coronin-2B protein has been identified in ciliary studies in C. elegans, D. melanogaster and M. musculus, the neutral and basic amino acid transport protein rBAT (SLC3A1) has no record related to the cilium to date. We also checked the existence of protein-protein interactions (PPIs) which could help to find relations between the proteins encoded by these candidate genes and other ciliary proteins. STRING and GENEMANIA tools revealed the co-expression of coronin-2B protein with BBS7 (Bardet-Biedl syndrome 7 protein) and FRZB/SFRP3 (secreted frizzled-related protein 3; belongs to the soluble frizzled-related proteins, sFRPS, that function as modulators of Wnt signalling) [31, 32]. FRZB/SFRP3 is co-expressed with BBS10 (Bardet-Biedl syndrome 10 protein) [45]. Furthermore, coronin-2B is known to interact with ACTA1 [46], an actin-binding protein, and also with coronin-2A [47], which in turn directly interacts with another actin-binding protein, ACTB [48]. SCL3A1 seems to have a direct interaction with another solute carrier member SLC4A1 (Band 3 anion transport protein), a Cl-/HCO3- exchanger expressed in sensory cilia of olfactory epithelium [49].

The two homozygous variants identified, c.581T>A, p.(Leu194Gln), in CORO2B (RefSeq NM_006091.4) and c.1381T>C, p.(Tyr461His), in SLC3A1 (RefSeq NM_000341.3), were confirmed by Sanger sequencing, as well as the autosomal recessive pattern of inheritance after segregation analysis in the only member available (mother) (Table 1). The second copy of these variants is assumed to be inherited via the proband’s father. Both variants were predicted to be pathogenic by three out of four bioinformatics prediction tools (Table 2).

Although none of these two genes has been previously related to any ciliopathy, these evidences suggest that CORO2B could be the causative gene in this family. The variant identified in SLC3A1 has already been described in patients with cystinuria (#220100) [50, 51], but according to their phenotypes this variant cannot explain case 1 phenotype, that is why SLC3A1 was excluded as causative gene.

Case 5.

The same filters as in the previous case were applied, considering variants with both high and moderate snpEff predicted effect, narrowing the list down to 8 candidate genes. These were also prioritised by using different subsets of training genes, resulting in LMO7 (LIM domain 7) and ZNF17 (zinc finger protein 17) as top candidates. Cildb database revealed the detection of ZNF17 in previous ciliary studies in C. elegans, but no evidence was found for LMO7. Tools predicting PPIs revealed a link between ZNF17 and TTC8/BBS8 (Tetratricopeptide repeat protein 8) via OFIP/KIAA0753 (OFD1 and FOR20 interacting protein), and also the involvement of LMO7 in the same pathway as ACTN2 (Alpha-actinin-2), an actin-binding protein which physically interacts with TTC8. All these proteins have been associated to either ciliary disease or pathways [5254].

Both homozygous variants, c.890C>T; p.(Pro297Leu), in LMO7 (RefSeq NM_015842.2) and c.1903G>T, p.(Glu635*); in ZNF17 (RefSeq NM_006959.2), were confirmed by direct sequencing, but no family members were available to perform segregation analysis. All prediction tools classified these two variants as deleterious (Table 2).

Case 6.

In this case, a homozygous deletion of 7 nucleotides, c.613_619delATAGGAA; p.(Ile205Aspfs*13), in CRB1 (RefSeq NM_201253.2), a less common ciliary gene, was identified after examining those variants with a high impact prediction. CRB1 has been previously related to retinitis pigmentosa and Leber Congenital Amaurosis [55, 56]. This variant has been assumed as deleterious since the protein presumably would be truncated, and also affects a highly conserved residue. Furthermore, since this change is located at the end of exon 2, and considering that CRB1 gene contains 12 exons, we also analysed the putative effect on splice donor/acceptor sequences. Three out of four bioinformatics tools did predict an alteration of splicing efficiency, which reinforces the presumable pathogenicity of the variant (Table 2). Direct sequencing confirmed its presence in homozygous state in the studied patient (female). Segregation analysis in the parents supported the autosomal recessive inheritance of this variant (Table 1), and revealed its presence in heterozygous state in the patient’s sibling (male). This made us think in a more complex inheritance in the heterozygous carrier, but the negative results obtained in our previous genetic screening suggest that there should be an additional gene harbouring two mutated alleles, which could explain the phenotype displayed.

ROH analysis from WES data

The analysis of WES data also identified ROHs consistent with the presence of consanguinity (ROH >100,000 kb in length) in all patients. In addition, we confirmed that all candidate variants localize within these genomic ROH regions. These results can be found in S2 Table.

Discussion

Molecular diagnosis of rare diseases has evolved considerably over the last years, primarily due to the advent and subsequent improvements in next-generation sequencing technologies. Until recently, traditional Sanger sequencing was the most recurrent approach, involving long working hours, particularly when analysing large genes or multigenic diseases. This, together with the high genetic heterogeneity associated with these disorders and the frequent difficulty to obtain an accurate clinical assessment, has hampered the molecular diagnosis, and therefore the possibility for patients to receive an appropriate genetic counselling and treatment [8, 55, 57].

Here we report the molecular analysis using whole-exome sequencing of six consanguineous families with different genetic background and primary clinical diagnosis of a ciliopathy, mainly BBS. The initial revision of clinical features showed a variety of phenotypes, among which only three patients fulfilled the BBS diagnostic criteria [39]. Therefore, we followed the molecular algorithm previously proposed by our group [58] consisting in first testing for prevalent mutations in BBS, whether through the BBS/AS Asper Ophthalmics genotyping microarray (Asper Biotech; Tartu, Estonia) or direct sequencing of the main genes involved. In the particular case of BBS and Alström syndrome (ALMS, MIM #203800), the genotyping microarray has facilitated the molecular diagnosis by the rapid detection of known variations in the main genes related to them [59, 60], but however limits the identification of novel mutations or novel candidate genes. These tests turned up no results. Conversely, WES enabled us to identify potential causative mutations in all families, even after applying strict criteria for data filtering as proposed by other authors [61].

The suspected clinical diagnosis of BBS was confirmed in two patients (cases 2 and 3) by identifying two nonsense mutations in BBS2 gene. In addition, a mutation in ALMS1, the only gene associated with ALMS to date, was found in case 4, a patient suspected of having BBS. Since these positions are not included in the Asper Ophthalmics genotyping microarray and BBS2 is not a predominant gene according our cohort, these mutations could not be detected in our initial screening. It is well-known that both syndromes show a significant phenotypic overlap [7, 62, 63], primarily early-onset obesity, diabetes and retinal dystrophy. This fact is consistent with the idea that both entities could share similar molecular etiologies, so it is no wonder that patients can be clinically misdiagnosed, although recent efforts are being made to clarify the mechanisms underlying these syndromes [62, 63].

After exploring the involvement of known ciliary genes, three cases remained without a clear molecular diagnosis. Although case 1 was clinically diagnosed with BBS since the patient showed five primary features commonly related to this syndrome, no mutation was found either in any BBS gene or in other ciliopathy-related genes. This led us to believe in the possibility of a novel candidate gene responsible for the phenotype displayed by this patient. Thus, an alternative filtering strategy revealed CORO2B as the main candidate causative gene in this family. CORO2B encodes Coronin-2B protein, which seems to play a role in the reorganization of neuronal actin structure (GeneCards database, http://www.genecards.org/, identifier GC15P068559). Although it is predominantly expressed in brain, it can also be found in other human tissues such as retina, heart, kidney, adipocytes and reproductive organs (GeneCards), which are commonly affected in many ciliopathies. In addition, this protein contains several WD-repeat domains, frequently found in intraflagellar transport proteins (IFT80, IFT122, IFT172), dynein-related proteins (DNAI1, DNAI2, DYNC1I1) and other proteins associated with ciliogenesis (POC1A, POC1B) [64, 65]. Proteins containing these WD-repeat domains are thought to play important roles in the assembly of multiprotein complexes involved in the majority of pathways in eukaryotic cells, this being the reason why this kind of proteins are commonly associated with numerous genetic diseases [66]. Interestingly, the p.(Leu194Gln) mutation found in this gene localizes in one of these WD40-repeat domains. The fact that PPIs tools have shown the co-expression and physical interaction of Coronin-2B with well-known ciliary proteins via a third protein, together with the identification of Coronin-2B in ciliary studies in other organisms, reinforces our hypothesis of CORO2B as the main candidate gene for this family.

A similar situation occurs with case 5, with a clinical diagnosis of MKKS. After ruling out the presence of mutations in MKKS, the only gene linked to the disease to date, and other ciliopathy genes, only two possible candidate genes remained: ZNF17 and LMO7. To date, there is little information about these two genes. ZNF17 encodes the zinc finger protein 17, which may be involved in transcriptional regulation (GeneCards, identifier GC19P057411) and is evolutionary conserved across species. This protein seems to be co-expressed with KIAA0753/OFIP [53], whose gene has been recently identified as causative in a patient with Oral-facial-digital VI syndrome and forms a ternary complex with OFD1 and FOR20 proteins, which are necessary for basal body anchoring and, consequently, for a non-defective primary cilia ciliogenesis. This, together with its association with centrosome and pericentriolar satellites in human cells, made us think in a possible involvement of ZNF17 in ciliogenesis. The second gene, LMO7, encodes the LIM-only 7 protein, which is known to be involved in PPIs(GeneCards, identifier GC13P075620). It seems to take part in the same pathway as ACTN2, which in turn physically interacts with TTC8, often related to BBS and retinitis pigmentosa (Genemania). Additionally, LMO7 seems to participate in regulation of actin cytoskeleton as part of the adherens junction pathway (KEGG Pathway database, http://www.genome.jp/kegg/pathway.html), thus the involvement of LMO7 in ciliary disease would not be unlikely. Given the above-mentioned information, and considering that both proteins are expressed in multiple human ciliated tissues such as brain, retina, heart, adipocytes and reproductive organs (GeneCards), it would not be rare to find these proteins playing an important role in ciliogenesis and/or ciliary function, but further studies will be needed to confirm their involvement in ciliary pathogenesis.

The two cases described above clearly show the genetic and phenotypic complexity of ciliary diseases, leading to misdiagnosis in some instances. Thus, case 5 seemed to be a phenotypic overlap between two syndromes. This child was clinically diagnosed with MKKS because of the symptoms displayed at the first assessment but no mutation was found in MKKS gene, so we decided to review our data in search of mutations in other BBS genes, since it is well-known that many patients diagnosed with MKKS during neonatal period or early childhood may turn out to be a case of BBS years later, developing retinal dystrophy and obesity [67, 68]. However, it was not possible to find causative mutations either in BBS or in other ciliopathy-related genes, which led us to seek and propose novel genes as candidates. Case 6 also turned out to be a case of misdiagnosis. While the studied patient was initially referred to as a case of BBS with retinal dystrophy and obesity, the recent reassessment of the phenotype only confirmed the presence of retinal dystrophy, which can be explained by defective CRB1 expression, and thus indicating that the previously diagnosed obesity was nonsyndromic. By contrast, her heterozygous sibling evolved from polydactyly into a more complex phenotype, suggesting that although CRB1 might be the causative gene in the studied patient it would be necessary to perform WES in the sibling with the more severe phenotype. This highlights the importance of a full clinical evaluation and, above all, a patient follow-up for providing an accurate molecular diagnosis, prognosis and efficient genetic counselling.

Anyway, we cannot discard the presence of pathogenic variants in known ciliopathy-related genes that are not likely to be detected in WES, such as variants in non-coding parts of the transcripts, promoters, or long distance regulatory elements. In addition, some types of variants, such as copy number variations (CNVs), structural variants (SVs), middle-sized indels and deep intronic variants may be missed [10, 69]. While some of these issues could be addressed better with whole genome sequencing (WGS), this approach entails higher costs and the large volume of data to analyse requires more advanced pipelines to filter and detect disease-causing variants, which made it less suitable for clinical application [70]. Given the consanguinity of the families included in this study, the combination of homozygosity mapping and WES would have been a good strategy to follow, since it allows us to identify loss of heterozygosity (LOH) regions where homozygous disease-causing variants are frequently located [71, 72]. Instead, we obtained ROHs from WES data and confirmed that all candidate variants are located in ROH regions. This has been described as a useful approach to identify deleterious mutations in consanguineous families, particularly those with low frequency in general population [73].

In summary, we show WES as a useful and cost-effective tool for molecular discovery and diagnosis in families with suspected ciliopathy phenotype, allowing us to identify three homozygous mutations, one of them novel, which confirm two cases of BBS and another of ALMS. In addition, the strategy we adopted has also enabled us to propose novel candidate genes in two families, which seem to be potentially related to cilia. Future functional studies will be needed to understand in depth the functional impact of all the variants identified herein and also to verify the possible role of the novel candidate genes in the molecular pathogenesis of ciliopathies, thus giving new insights into cilia biology.

Supporting information

S1 Table. List of common ciliary genes used as training set for prioritisation with Endeavour and ToppGene Suite tools.

https://doi.org/10.1371/journal.pone.0183081.s001

(DOCX)

S2 Table. ROH regions identified from WES data.

Abbreviations: ROH, runs of homozygosity.

https://doi.org/10.1371/journal.pone.0183081.s002

(DOCX)

Acknowledgments

The authors would like to thank the patients and their family members for participating in this research, all the colleagues for the collection of patients and clinical data and the Centro Nacional de Análisis Genómico (CNAG-CRG), specially Leslie Matalonga and Anastasios Papakonstantinou, for their technical help. We also thank the Registro Español de los Síndromes de Wolfram, Bardet-Biedl y Alström (REWBA), the European Union Rare Diseases Registry for Wolfram syndrome, Alström syndrome, Bardet-Biedl syndrome and other rare diabetes syndromes (EURO-WABB) and the BIOCAPS Project (from European Commission under the 7th Framework Programme, FP-7-REGPOT 2012-2013-1).

References

  1. 1. Hildebrandt F, Benzing T, Katsanis N. Ciliopathies. N Engl J Med. 2011;364: 1533–1543. pmid:21506742
  2. 2. van Reeuwijk J, Arts HH, Roepman R. Scrutinizing ciliopathies by unravelling ciliary interaction networks. Hum Mol Genet. 2011;20: R149–R157. pmid:21862450
  3. 3. Adly N, Alhashem A, Ammari A, Alkuraya FS. Ciliary genes TBC1D32/C6orf170 and SCLT1 are mutated in patients with OFD Type IX. Hum Mut. 2014;35: 36–40. pmid:24285566
  4. 4. Girisha KM, Shukla A, Trujillano D, Bhavani GS, Hebbar M, Kadavigere R, et al. A homozygous nonsense variant in IFT52 is associated with a human skeletal ciliopathy. Clin Genet. 2016;90: 536–539. pmid:26880018
  5. 5. Bachmann-Gagescu R. Genetic complexity of ciliopathies and novel genes identification. Med Sci (Paris). 2014;30: 1011–1023.
  6. 6. Alkuraya FS. The application of next-generation sequencing in the autozygosity mapping of human recessive diseases. Hum Genet. 2013;132: 1197–1211. pmid:23907654
  7. 7. Kim MK, Kwak SH, Kang S, Jung HS, Cho YM, Kim SY, et al. Identification of Two Cases of Ciliopathy-Associated Diabetes and Their Mutation Analysis Using Whole Exome Sequencing. Diabetes Metab J. 2015;39: 439–443. pmid:26566502
  8. 8. Abu-Safieh L, Alrashed M, Anazi S, Alkuraya H, Khan AO, Al-Owain M, et al. Autozygome-guided exome sequencing in retinal dystrophy patients reveals pathogenic mutations and novel candidate disease genes. Genome Res. 2013;23: 236–247. pmid:23105016
  9. 9. Shaheen R, Faqeih E, Alshammari MJ, Swaid A, Al-Gazali L, Mardawi E, et al. Genomic analysis of Meckel-Gruber syndrome in Arabs reveals marked genetic heterogeneity and novel candidate genes. Eur J Hum Genet. 2013;21: 762–768. pmid:23169490
  10. 10. Bamshad MJ, Ng SB, Bigham AW, Tabor HK, Emond MJ, Nickerson DA, et al. Exome sequencing as a tool for Mendelian disease gene discovery. Nat Rev Genet. 2011;12: 745–755. pmid:21946919
  11. 11. Roosing S, Romani M, Isrie M, Rosti RO, Micalizzi A, Musaev D, et al. Mutations in CEP120 cause Joubert syndrome as well as complex cilipathy phenotypes. J Med Genet. 2016;53: 608–615. pmid:27208211
  12. 12. Erlich Y, Edvardson S, Hodges E, Zenvirt S, Thekkat P, Shaag A, et al. Exome sequencing and disease-network analysis of a single family implicate a mutation in KIF1A in hereditary spastic paraparesis. Genome Res. 2011;21: 658–664. pmid:21487076
  13. 13. Li MX, Gui HS, Kwan JS, Bao SY, Sham PC. A comprehensive framework for prioritizing variants in exome sequencing studies of Mendelian diseases. Nucleic Acids Res. 2012;40: e53. http://nar.oxfordjournals.org/content/40/7/e53.long pmid:22241780
  14. 14. Álvarez-Satta M, Castro-Sánchez S, Pereiro I, Piñeiro-Gallego T, Baiget M, Ayuso C, et al. Overview of Bardet-Biedl syndrome in Spain: identification of novel mutations in BBS1, BBS10 and BBS12 genes. Clin Genet. 2014;86: 601–602. pmid:24611592
  15. 15. Marco-Sola S, Sammeth M, Guigó R, Ribeca P. The GEM mapper: fast, accurate and versatile alignment by filtration. Nature methods. 2012;9: 1185–1188. pmid:23103880
  16. 16. McKenna A, Hanna M, Banks E, Sivachenko A, Cibulskis K, Kernytsky A, et al. The Genome Analysis Toolkit: a MapReduce framework for analysing next-generation DNA sequencing data. Genome Res. 2010;20: 1297–1303. pmid:20644199
  17. 17. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics. 2009;25: 2078–2079. pmid:19505943
  18. 18. Cunningham F, Amode MR, Barrell D, Beal K, Billis K, Brent S, et al. Ensembl 2015. Nucleic Acids Res. 2015;43: D662–D669. pmid:25352552
  19. 19. Cingolani P, Platts A, Wang Ie L, Coon M, Nguyen T, Wang L, et al. A program for annotating and predicting the effects of single nucleotide polymorphisms, SnpEff: SNPs in the genome of Drosophila melanogaster strain w1118; iso-2; iso-3. Fly (Austin). 2012;6: 80–92.
  20. 20. Cingolani P, Patel VM, Coon M, Nguyen T, Land SJ, Ruden DM, et al. Using Drosophila melanogaster as a model for genotoxic chemical mutational studies with a new program, SnpSift. Front Genet. 2012;3: 35. pmid:22435069
  21. 21. Sherry ST, Ward MH, Kholodov M, Baker J, Phan L, Smigielski EM, et al. dbSNP: the NCBI database of genetic variation. Nucleic Acids Res. 2001;29: 308–311. pmid:11125122
  22. 22. 1000 Genomes Project Consortium, Abecasis GR, Altshuler D, Auton A, Brooks LD, Durbin RM, et al. A map of human genome variation from population-scale sequencing. Nature. 2010;467: 1061–1073. pmid:20981092
  23. 23. Liu X, Jian X, Boerwinkle E. dbNSFP v2.0: a database of human non-synonymous SNVs and their functional predictions and annotations. Hum Mutat. 2013;34: E2393–E2402. pmid:23843252
  24. 24. Adzhubei IA, Schmidt S, Peshkin L, Ramensky VE, Gerasimova A, Bork P, et al. A method and server for predicting damaging missense mutations. Nat Methods. 2010;7: 248–249. pmid:20354512
  25. 25. Kumar P, Henikoff S, Ng PC. Predicting the effects of coding non-synonymous variants on protein function using the SIFT algorithm. Nat Protoc. 2009;4: 1073–1081. pmid:19561590
  26. 26. Schwarz JM, Cooper DN, Schuelke M, Seelow D. MutationTaster2: mutation prediction for the deep-sequencing age. Nat Methods. 2014;11: 361–362. pmid:24681721
  27. 27. Chun S, Fay JC. Identification of deleterious mutations within three human genomes. Genome Res. 2009;19: 1553–1561. pmid:19602639
  28. 28. Papadopoulos JS, Agarwala R. COBALT: constraint-based alignment tool for multiple protein sequences. Bioinformatics. 2007;23: 1073–1079. pmid:17332019
  29. 29. Aerts S, Lambrechts D, Maity S, Van Loo P, Coessens B, De Smet F, et al. Gene prioritization through genomic data fusion. Nat Biotechnol. 2006;24: 537–544. pmid:16680138
  30. 30. Chen J, Bardes EE, Aronow BJ, Jegga AG. ToppGene Suite for gene list enrichment analysis and candidate gene prioritization. Nucleic Acids Res. 2009;37: W305–W311. pmid:19465376
  31. 31. Szklarczyk D, Franceschini A, Wyder S, Forslund K, Heller D, Huerta-Cepas J, et al. STRING v10: protein-protein interaction networks, integrated over the tree of life. Nucleic Acids Res. 2015;43: D447–D452. pmid:25352553
  32. 32. Warde-Farley D, Donaldson SL, Comes O, Zuberi K, Badrawi R, Chao P, et al. The GeneMANIA prediction server: biological network integration for gene prioritization and predicting gene function. Nucleic Acids Res. 2010;38: W214–W220. http://nar.oxfordjournals.org/content/38/suppl_2/W214.long pmid:20576703
  33. 33. Reese MG, Eeckman FH, Kulp D, Haussler D. Improved Splice Site Detection in Genie. J Comp Biol. 1997;4: 311–323.
  34. 34. Hebsgaard SM, Korning PG, Tolstrup N, Engelbrecht J, Rouze P, Brunak S. Splice site prediction in Arabidopsis thaliana DNA by combining local and global sequence information. Nucleic Acids Res. 1996;24: 3439–3452.
  35. 35. Desmet FO, Hamroun D, Lalande M, Collod-Béroud G, Claustres M, Béroud C. Human Splice Finder: an online bioinformatics tool to predict splicing signals. Nucleic Acids Res. 2009;37: e67. http://nar.oxfordjournals.org/content/37/9/e67.long pmid:19339519
  36. 36. Mucaki EJ, Shirley BC, Rogan PK. Prediction of mutant mRNA splice isoforms by information theory-based exon definition. Informatics. 2013;34: 557–565.
  37. 37. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007;81: 559–575. pmid:17701901
  38. 38. Kancheva D, Atkinson D, De Rijk P, Zimon M, Chamova T, Mitev V, et al. Novel mutations in genes causing hereditary spastic paraplegia and Charcot-Marie-Tooth neuropathy identified by an optimized protocol for homozygosity mapping based on whole-exome sequencing. Genet Med. 2016;18: 600–607. Erratum in: Genet Med. 2016;18: 108. Parma, Yesim [corrected to Parman, Yesim]. pmid:26492578
  39. 39. Beales PL, Elcioglu N, Woolf AS, Parker D, Flinter FA. New criteria for improved diagnosis of Bardet-Biedl syndrome: results of a population survey. J Med Genet. 1999;36: 437–446. pmid:10874630
  40. 40. Endsley JK, Phillips JA 3rd, Hruska KA, Denneberg T, Carlson J, George AL Jr. Genomic organization of a human cystine transporter gene (SLC3A1) and identification of novel mutations causing cystinuria. Kidney Int. 1997;51: 1893–1899. pmid:9186880
  41. 41. Smaoui N, Chaabouni M, Sergeev YV, Kallel H, Li S, Mahfoudh N, et al. Screening of the eight BBS genes in Tunisian families: no evidence of triallelism. Invest Ophthalmol Vis Sci. 2006;47: 3487–3495. pmid:16877420
  42. 42. Bond J, Flintoff K, Higgins J, Scott S, Bennet C, Parsons J, et al. The importance of seeking ALMS1 mutations in infants with dilated cardiomyopathy. J Med Genet. 2005;42: e10. pmid:15689433
  43. 43. van Dam TJ, Wheway G, Slaats GG, SYSCILIA Study Group, Huynen MA, Giles RH. The SYSCILIA Gold Standard (SCGSv.1) of known ciliary components and its applications within a systems biology consortium. Cilia. 2013;2: 7. pmid:23725226
  44. 44. Arnaiz O, Malinowska A, Klotz C, Sperling L, Dadlez M, Cohen J. Cildb: a knowledgebase for centrosomes and cilia. Database (Oxford). 2009;2009: bap022.
  45. 45. Vinayagam A, Stelzl U, Foulle R, Plassmann S, Zenkner M, Timm J, et al. A directed protein interaction network for investigating intracellular signal transduction. Sci Signal. 2011:4: rs8. pmid:21900206
  46. 46. Nakamura T, Takeuchi K, Muraoka S, Takezoe H, Takahashi N, Mori N. A neurally enriched coronin-like protein, ClipinC, is a novel candidate for an actin cytoskeleton-cortical membrane-linking protein. J Biol Chem. 1999;274: 13322–13327. pmid:10224093
  47. 47. Huttlin EL, Bruckner RJ, Paulo JA, Cannon JR, Ting L, Baltier K, et al. Architecture of the human interactome defines protein communities and disease networks. Nature. 2017;545: 505–509. pmid:28514442
  48. 48. Hein MY, Hubner NC, Poser I, Cox J, Nagaraj N, Toyoda Y, et al. A human interactome in three quantitative dimensions organized by stoichiometries and abundances. Cell. 2015;163: 712–723. pmid:26496610
  49. 49. Hengl T, Kaneko H, Dauner K, Vocke K, Frings S, Möhrlen F. Molecular components of signal amplification in olfactory sensory cilia. Proc Natl Acad Sci U S A. 2010;107: 6052–6057. pmid:20231443
  50. 50. Endsley JK, Phillips JA 3rd, Hruska KA, Denneberg T, Carlson J, George AL Jr. Genomic organization of a human cysteine transporter gene (SLC3A1) and identification of novel mutations causing cystinuria. Kidney Int. 1997;51: 1893–1899. pmid:9186880
  51. 51. Harnevick L, Fjellstedt E, Molbaek A, Tiselius HG, Denneberg T, Söderkvist P. Identification of 12 novel mutations in the SLC3A1 gene in Swedish cystinuria patients. Hum Mut. 2001;18: 516–525. pmid:11748844
  52. 52. Ansley SJ, Badano JL, Blacque OE, Hill J, Hoskins BE, Leitch CC, et al. Basal body dysfunction is a likely cause of pleiotropic Bardet-Biedl syndrome. Nature. 2003;425: 628–633. pmid:14520415
  53. 53. Chevrier V, Bruel AL, Van Dam TJ, Franco B, Lo Scalzo M, Lembo F, et al. OFIP/KIAA0753 forms a complex with OFD1 and FOR20 at pericentriolar satellites and centrosomes and is mutated in one individual with oral-facial-digital syndrome. Hum Mol Genet. 2016;25: 497–513. pmid:26643951
  54. 54. Li Q, Montalbetti N, Shen PY, Dai XQ, Cheeseman CI, Karpinski E, et al. Alpha-actinin associates with polycystin-2 and regulates its channel activity. Hum Mol Genet. 2005;14: 1587–1603. pmid:15843396
  55. 55. den Hollander AI, ten Brink JB, de Kok YJ, van Soest S, van den Born LI, van Driel MA, et al. Mutations in a human homologue of Drosophila crumbs cause retinitis pigmentosa (RP12). Nat Genet. 1999;23: 217–221. pmid:10508521
  56. 56. Lotery AJ, Jacobson SG, Fishman GA, Weleber RG, Fulton AB, Namperumalsamy P, et al. Mutations in the CRB1 gene cause Leber congenital amaurosis. Arch Ophthalmol. 2001;119: 415–420. pmid:11231775
  57. 57. Berger W, Kloeckner-Gruissem B, Neidhardt J. The molecular basis of human retinal and vitreoretinal diseases. Prog Retin Eye Res. 2010;29: 335–375. pmid:20362068
  58. 58. Castro-Sánchez S, Álvarez-Satta M, Pereiro I, Piñeiro-Gallego MT, Valverde D. Algorithm for the molecular analysis of Bardet-Biedl syndrome in Spain. Med Clin (Barc). 2015;145: 147–152.
  59. 59. Pereiro I, Valverde D, Piñeiro-Gallego T, Baiget M, Borrego S, Ayuso C, et al. New mutations in BBS genes in small consanguineous families with Bardet-Biedl syndrome: detection of candidate regions by homozygosity mapping. Mol Vis. 2010;16: 137–143. pmid:20142850
  60. 60. Pereiro I, Hoskins BE, Marshall JD, Collin GB, Naggert JK, Piñeiro-Gallego T, et al. Arrayed primer extension technology simplifies mutation detection in Bardet-Biedl and Alström syndrome. Eur J Hum Genet. 2011;19: 485–488. pmid:21157496
  61. 61. M’hamdi O, Ouertani I, Chaabouni-Bouhamed H. Update on the genetics of Bardet-Biedl syndrome. Mol Syndromol. 2014;5: 51–56. pmid:24715851
  62. 62. Hostelley TL, Lodh S, Zaghloul NA. Whole organism transcriptome analysis of zebrafish models of Bardet-Biedl syndrome and Alström syndrome provides mechanistic insight into shared and divergent phenotypes. BMC Genomics. 2016;17: 318. pmid:27142762
  63. 63. Lodh S, Hostelley TL, Leitch CC, O’Hare EA, Zaghloul NA. Differential effects on β-cell mass by disruption of Bardet-Biedl syndrome or Alström syndrome genes. Hum Mol Genet. 2016;25: 57–68. pmid:26494903
  64. 64. Beck BB, Phillips JB, Bartram MP, Wegner J, Thoenes M, Pannes A, et al. Mutation of POC1B in a severe syndromic retinal ciliopathy. Hum Mutat. 2014;35: 1153–1162. pmid:25044745
  65. 65. Pearson GC, Osborn DP, Giddings TH Jr, Beales PL, Winey M. Basal body stability and ciliogenesis requires the conserved component Poc1. J Cell Biol. 2009;187: 905–920. pmid:20008567
  66. 66. Smith TF. Diversity of WD-repeat proteins. Subcell Biochem. 2008;48: 20–30. pmid:18925368
  67. 67. David A, Bitoun P, Lacombe D, Lamert JC, Nivelon A, Vigneron J, et al. Hydrometrocolpos and polydactyly: a common neonatal presentation of Bardet-Biedl and McKusick-Kaufman syndromes. J Med Genet. 1999;36: 599–603. pmid:10465109
  68. 68. Schaefer E, Durand M, Stoetzel C, Doray B, Viville B, Hellé S, et al. Molecular diagnosis reveals genetic heterogeneity for the overlapping MKKS and BBS phenotypes. Eur J Med Genet. 2011;54: 157–160. pmid:21044901
  69. 69. Pérez-Carro R, Corton M, Sánchez-Navarro I, Zurita O, Sanchez-Bolivar N, Sánchez-Alcudia R, et al. Panel-based NGS reveals novel pathogenic mutations in autosomal recessive Retinitis Pigmentosa. Sci Rep. 2016;6: 19531. pmid:26806561
  70. 70. Rabbani B, Mahdieh N, Hosomichi K, Nakaoka H, Inoue I. Next-generation sequencing: impact of exome sequencing in characterizing Mendelian disorders. J Hum Genet. 2012;57: 621–632. pmid:22832387
  71. 71. Knopp C, Rudnik-Schöneborn S, Eggermann T, Bergmann C, Begemann M, Choner K, et al. Syndromic ciliopathies: From single gene to multi gene analysis by SNP arrays and next generation sequencing. Mol Cell Probes. 2015;29: 299–307. pmid:26003401
  72. 72. Zaki MS, Heller R, Thoenes M, Nürnberg G, Stern-Schneider G, Nürberg P, et al. PEX6 is expressed in photoreceptor cilia and mutated in deaf blindness with Enamel Dysplasia and microcephaly. Hum Mut. 2015;37: 170–174. pmid:26593283
  73. 73. Mezzavilla M, Vozzi D, Badii R, Alkowari MK, Abdulhadi K, Girotto G, et al. Increased rate of deleterious variants in long runs of homozgyosity of an inbred population from Qatar. Hum Hered. 2015;79: 14–19. pmid:25720536