The Journal of Molecular Diagnostics, Vol. 24, No. 5, May 2022 jmdjournal.org Systematic Collaborative Reanalysis of Genomic Data Improves Diagnostic Yield in Neurologic Rare Diseases Gemma Bullich,* Leslie Matalonga,* Montserrat Pujadas,y Anastasios Papakonstantinou,* Davide Piscia,* Raúl Tonda,* Rafael Artuch,zx Pia Gallano,x{ Glòria Garrabou,xk Juan R. González,**yyzz Daniel Grinberg,x,xx Míriam Guitart,{{ Steven Laurie,* Conxi Lázaro,kk Cristina Luengo,* Ramon Martí,x*** Montserrat Milà,xyyy David Ovelleiro,* Genís Parra,* Aurora Pujol,xzzz Eduardo Tizzano,xxx Alfons Macaya,{{{ Francesc Palau,xkkk Antònia Ribes,x**** Luis A. Pérez-Jurado,yxyyyy and Sergi Beltran,*yyzzzz Undiagnosed Rare Disease Program of Catalonia (URD-Cat) Consortium From the Centro Nacional Análisis Genómico (CNAG)eCentre for Genomic Regulation (CRG),* The Barcelona Institute of Science and Technology, Barcelona, Spain; the Genetics Unit,y University Pompeu Fabra, Institut Hospital del Mar d'Investigacions Mèdiques (IMIM), Barcelona, Spain; the Clinical Biochemistry Department,z Institut de Recerca Sant Joan de Déu (IRSJD), Barcelona, Spain; the Centro de Investigaciones Biomédicas en Red de Enfermedades Raras (CIBERER)x and the Centro de Investigaciones Biomédicas en Red de Epidemiología y Salud Pública (CIBERESP),zz Instituto de Salud Carlos III, Madrid, Spain; the Genetics Department,{ Institut d'Investigacions Biomèdiques (IIB) Sant Pau, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain; the Muscle Research and Mitochondrial Function Laboratory,k CELLEX-Institut d'Investigació Biomèdica August Pi i Sunyer (IDIBAPS), Internal Medicine Department, Hospital Clinic of Barcelona, Faculty of Medicine and Health Sciences, University of Barcelona, Barcelona, Spain; the Barcelona Institute for Global Health (ISGlobal),** Barcelona, Spain; the Universitat Pompeu Fabra,yy Barcelona, Spain; the Department of Genetics, Microbiology and Statistics,xx Faculty of Biology, University of Barcelona, Institute of Biomedicine of the University of Barcelona (IBUB), Institut de Recerca Sant Joan de Déu (IRSJD), Barcelona, Spain; the Genetics Laboratory,{{ Paediatric Unit, Parc Taulí Hospital Universitari, Institut d’Investigació i Innovació Parc Taulí I3PT, Universitat Autònoma de Barcelona, Sabadell, Spain; the Molecular Diagnostic Unit,kk Hereditary Cancer Program, Institut d'Investigació Biomèdica de Bellvitge (IDIBELL), Catalan Institute of Oncology, Centro de Investigación Biomédica en Red de Cáncer (CIBERONC), Barcelona, Spain; the Research Group on Neuromuscular and Mitochondrial Diseases,*** Vall d'Hebron Institut de Recerca (VHIR), Universitat Autònoma de Barcelona, Barcelona, Spain; the Biochemistry and Molecular Genetics Department,yyy Hospital Clínic de Barcelona, Institut d’Investigació Biomèdica August Pi I Sunyer (IDIBAPS), Barcelona, Spain; the Neurometabolic Diseases Laboratory,zzz Institut d'Investigació Biomèdica de Bellvitge (IDIBELL)eHospital Duran i Reynals, Institucio Catalana de Recerca i Estudis Avançats (ICREA), Barcelona, Spain; the Department of Clinical and Molecular Genetics,xxx Medicine Genetics Group Vall d'Hebron Institut de Recerca (VHIR), European Reference Network on Rare Congenital Malformations and Rare Intellectual Disability ERN-ITHACA, Universitat Autònoma de Barcelona, Hospital Vall dHebron, Barcelona, Spain; the Pediatric Neurology Research Group,{{{ Vall d’Hebron Institut de Recerca (VHIR), Universitat Autònoma de Barcelona, Barcelona, Spain; the Department of Genetic and Molecular Medicine,kkk Pediatric Institute of Rare Diseases (IPER), Hospital Sant Joan de Déu, Clinic Institute of Medicine and Dermatology, Hospital Clínic de Barcelona and Division of Pediatrics, Faculty of Medicine and Health Sciences, University of Barcelona, Barcelona, Spain; the Secció d’Errors Congènits del MetabolismeeInstitute of Clinical Biochemistry (IBC),**** Servei de Bioquímica i Genètìca Molecular, Hospital Clínic de Barcelona, Institut d'Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain; the Women’s and Children’s Hospital,yyyy South Australian Health and Medical Research Institute and The University of Adelaide, Adelaide, South Australia, Australia; and the Department of Genetics, Microbiology and Statistics,zzzz Facultat de Biologia, Universitat de Barcelona, Barcelona, Spain Accepted for publication February 3, 2022. Many patients experiencing a rare disease remain undiagnosed even after genomic testing. Reanalysis of existing genomic data has shown to increase diagnostic yield, although there are few systematic Address correspondence to Sergi Beltran, Ph.D., Centro and comprehensive reanalysis efforts that enable collaborative interpretation and future reinter- pretation. The Undiagnosed Rare Disease Program of Catalonia project collated previously Supported by Generalitat de Catalunya through Departament de Salut III grants PT13/0001/0044 and PT17/0009/0019 for CNAG-CRG and (SLT002/16/00174 to URD-Cat consortium) and Departament d’Empresa i grants PI16/01048, PI19/01310, PI18/00451, PI18/00498, and PI21/00935 Coneixement and the CERCA Program; FP7 and H2020 EU projects RD- for IDIBAPS (Instituto Nacional de Bioinformática); ELIXIR Imple- Connect, Solve-RD, and EJP-RD grants FP7 305444, H2020 779257, and mentation Studies (CNAG-CRG); Centro de Investigaciones Biomédicas H2020 825575 for CNAG-CRG; Spanish Ministry of Science and Inno- en Red de Enfermedades Raras; Centro de Excelencia Severo Ochoa grant vation to the EMBL partnership and through the Instituto de Salud Carlos SEV-2016-0571 (CNAG-CRG) and grant CEX 2018-000806-S Copyright ª 2022 Association for Molecular Pathology and American Society for Investigative Pathology. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0). https://doi.org/10.1016/j.jmoldx.2022.02.003 Bullich et al Nacional Análisis Genómico inconclusive good quality genomic data (panels, exomes, and genomes) and standardized phenotypic eCentre for Genomic Regula- profiles from 323 families (543 individuals) with a neurologic rare disease. The data were reanalyzed tion, Parc Científic de Barce- systematically to identify relatedness, runs of homozygosity, consanguinity, single-nucleotide vari- lonaeTorre I, Baldiri Reixac 4, ants, insertions and deletions, and copy number variants. Data were shared and collaboratively 08028 Barcelona, Spain. E- interpreted within the consortium through a customized Genome-Phenome Analysis Platform, which mail: sergi.beltran@cnag.crg. eu. also enables future data reinterpretation. Reanalysis of existing genomic data provided a diagnosis for 20.7% of the patients, including 1.8% diagnosed after the generation of additional genomic data to identify a second pathogenic heterozygous variant. Diagnostic rate was significantly higher for family-based exome/genome reanalysis compared with singleton panels. Most new diagnoses were attributable to recent gene-disease associations (50.8%), additional or improved bioinformatic analysis (19.7%), and standardized phenotyping data integrated within the Undiagnosed Rare Dis- ease Program of Catalonia Genome-Phenome Analysis Platform functionalities (18%). (J Mol Diagn 2022, 24: 529e542; https://doi.org/10.1016/j.jmoldx.2022.02.003) Rare diseases collectively affect 3.5% to 5.9% of the The Undiagnosed Rare Disease Program of Catalonia worldwide population, and around 72% of them are of (URD-Cat) aims to provide the Catalan Health System with genetic origin.1 Patients with rare diseases often undergo a personalized genomic medicine as a fully integrated service years-long diagnostic odyssey characterized by multiple for patients with rare diseases, initially as a pilot project for tests with little or no success. Health system costs ascribed rare diseases with neurologic involvement. This project in- to rare disease patients are an important public health issue, volves the main groups that work in rare diseases in Cata- highlighting the need for improved access to early diag- lonia: 15 consolidated research groups belonging to 7 nosis and care coordination.2,3 Reaching a molecular Health Research Institutes in Barcelona Hospital del Mar diagnosis in a timely manner shortens the diagnostic od- Research Institute (IMIM); Institut d’Investigacions Bio- yssey, and can guide therapeutic strategies, improve clin- mèdiques August Pi i Sunyer (IDIBAPS); Institut d’Inves- ical management, and provide genetic counseling for tigació Biomèdica de Bellvitge (IDIBELL); Hospital Sant patients and their families with respect to recurrence risk Joan de Déu (HSJD); Vall d’Hebron Institut de Recerca and prenatal options.3 (VHIR); Hospital de la Santa Creu i Sant Pau (HSP); Institut Reanalysis of existing genomic data has emerged as an Germans Trias i Pujol (IGTP); the National Center for effective approach to increasing the diagnostic yield of Genomic AnalysiseCenter for Genomic Regulation previously undiagnosed patients, not only because of the (CNAG-CRG); the Barcelona Institute for Global Health rapid path of discovery of novel gene-disease associations, (ISGlobal); the National Supercomputing Center of Barce- but also due to improvements in analytical workflows, lona (BSC); and the Spanish Rare Disease Patient Federa- reclassification of previously unrecognized variants, and/or tion (FEDER). A multidisciplinary team of >140 availability of new phenotypic data.4 However, systematic professionals, including clinicians, geneticists, bio- reanalysis coupled with reinterpretation of the results re- informaticians, biochemists, technicians, and software en- quires iterative communication between researchers, cli- gineers, participate in the project. A customized version of nicians, and families as diagnoses can be made years after the GPAP has been deployed for the URD-Cat project to the initial sequencing and analysis were performed.3 meet the specific requirements needed to be integrated with Systems like the RD-Connect Genome-Phenome Analysis a National Health System regarding data privacy, interop- Platform (GPAP; https://platform.rd-connect.eu, last erability, availability, sustainability, and scalability, among accessed December 6, 2021) facilitate such others. communication and collaboration between clinicians and To date, the project has systematically collated clinical researchers within a trustworthy environment. The RD- and phenotypic information from 928 undiagnosed index Connect GPAP is an International Rare Diseases cases and their relatives (total of 1569 individuals) for Research Consortium (IRDiRC)erecognized resource that which most of the available diagnostic tests had been brings together pseudonymized clinical/phenotypic and performed without yielding a positive result. For 323 of genomic data with tools and services to enable data those index cases (543 of those individuals, including sharing, analysis and interpretation for rare disease diag- relatives), previously existing good quality genomic data nosis, and gene discovery.5,6 (panel, exome, or genome) were reanalyzed. Furthermore, (ISGLOBAL); and cofinancing with funds from the European Regional Departament d’Empresa i Coneixement of the Generalitat de Catalunya Development Fund by the Spanish Ministry of Science and Innovation corresponding to the Programa Operatiu FEDER de Catalunya 2014 corresponding to the Programa Operativo FEDER Plurirregional de España to 2020. (POPE) 2014 to 2020 and by the Secretaria d’Universitats i Recerca, Disclosures: None declared. 530 jmdjournal.org - The Journal of Molecular Diagnostics Collaborative Reanalysis in Rare Disease additional sequencing data (exome, genome, or tran- study was approved by the local ethics committees from scriptome) were generated in nine index cases with a each URD-Cat partnering institution. single heterozygous pathogenic or likely pathogenic variant in an autosomal recessive gene identified through Data Processing and Variant Detection the reanalysis. This study describes the workflow applied and the diagnostic yield after reanalysis. The effect of the The URD-Cat GPAP, a customized version of the RD- phenotype annotation quality and the sequencing approach Connect GPAP, was implemented for the URD-Cat project. as well as the reasons why diagnosed patients were not Existing FASTQ files, all of them obtained with Illumina diagnosed in the original analysis are also explored. (San Diego, CA) sequencing platforms, and corresponding metadata (sequencing approach, capture kit, singleton or extended family analysis, and DNA source) were submitted for processing to the URD-Cat GPAP. Materials and Methods All of the samples were bioinformatically processed at the The URD-Cat Reanalysis Cohort CNAG-CRG using the RD-Connect pipeline,11 which is based on GATK best practices.12 Briefly, sequencing reads Pseudonymized clinical information from undiagnosed pa- were mapped to human genome build GRCh37d5 using tients and their family members was collected from medical BWA-MEM version 0.7.15.13 The resulting BAM files were records by clinicians or geneticists from participating hos- sorted, and duplicate reads were removed using Picard pitals using PhenoTips.7 A specific form was generated in version 1.110 (http://broadinstitute.github.io/picard, last collaboration with them to collect all relevant data: de- accessed September 6, 2021). Insertion and deletion mographic data, disease category, family history and pedi- realignment and base quality score recalibration were gree, personal history, clinical symptoms, biochemical performed using GATK version 3.6.14 analysis, and previous genomic and nongenomic tests per- Single-nucleotide variants and short insertions and de- formed. Clinical symptoms were collected using standard- letions were called using GATK version 3.6 Hap- ized Human Phenotype Ontology (HPO) terms.8 Orphanet lotypeCaller tool.12 Single-nucleotide variants and Rare Disease Ontology (https://www.orpha.net/consor/cgi- insertions and deletions with a minimum depth of coverage bin/index.php, last accessed December 6, 2021) and of 8 and a minimum genotype quality of 30 were released to Online Mendelian Inheritance in Man (https://www.omim. the URD-Cat GPAP. Copy number variants (CNVs) were org, last accessed December 6, 2021) codes were used to detected with ExomeDepth for exomes and gene panels,15 enter a clinical and molecular diagnosis, respectively.9 analyzing together all of the samples captured with the Quality of the phenotypic data was assessed using the same kit. CNVs on the sex chromosomes were evaluated by Monarch star rating system integrated within PhenoTips. comparison against samples from the same sex only. CNVs The Monarch star rating system is an annotation were called only for groups of samples in which there were sufficiency meter developed by the Monarch initiative that at least 10 samples captured with the same kit assesses the breadth and depth of the phenotype (Supplemental Table S1). The CNV results were crossed annotation profile for a given patient in the context of all with sets of common CNVs from Conrad et al16 and curated human and model organisms using a five-star Database of Genomic Variants Gold Standard data set rating system: bad (0 to 1.4), fair (1.5 to 2.4), good (2.5 (http://dgv.tcag.ca, last accessed May 21, 2021).17 All of the to 3.4), very good (3.5 to 4.4), and excellent (4.5 to 5).10 CNVs obtained were released to the URD-Cat platform. All phenotypic entries were reviewed and approved by two clinical experts from another participating center. The Additional Sequencing reviewers were allowed to ask for clarifications or additional information to be included in the records during Additional sequencing data were obtained for nine index the review process. cases with a single heterozygous pathogenic or likely Among all of the undiagnosed patients and relatives pathogenic variant in an autosomal recessive gene identified whose information was submitted to PhenoTips, 331 index through the reanalysis. Genome (n Z 5) or transcriptome cases (560 individuals including relatives) from seven (n Z 2) sequencing was performed if the existing data were hospitals were considered for reanalysis. The inclusion an exome, whereas exome sequencing (n Z 2) was per- criteria were as follows: i) available and previously formed if the existing data were a panel. analyzed genomic data (genome, exome, or panel), ii) clinical suspicion of rare disease with neurologic involve- Relatedness and Consanguinity Analysis ment, iii) clinical suspicion of a genetic etiology, iv) availability of clinical information and disease progression Kinship between all individuals with a genome or an exome (from the patients and the family members), and v) written was computed on alignment files (BAM) with Somalier informed consent of the patient or parents/guardians, version 2.618 to identify putative duplicates and inconsistent enabling the use of the data in the URD-Cat project. This family relationships. Runs of homozygosity (RoHs) were The Journal of Molecular Diagnostics - jmdjournal.org 531 Bullich et al Table 1 Characteristics of the 323 Probands Classified by the Main Disease Category Sex Consanguinity (PhenoTips) Disease category Probands, N Male Female Yes No Unknown Progressive neurodegenerative diseases 108 70 (64.8) 38 (35.2) 12 (11.1) 74 (68.5) 22 (20.4) Neuromuscular diseases 56 31 (55.4) 25 (44.6) 6 (10.7) 41 (73.2) 9 (16.1) Epilepsy/nonepileptic paroxysmal disorders 52 27 (51.9) 25 (48.1) 5 (9.6) 28 (53.8) 19 (36.5) Inherited metabolic disorders 39 14 (35.9) 25 (64.1) 6 (15.4) 27 (69.2) 6 (15.4) Intellectual disabilities/autism spectrum disorders 36 25 (69.4) 11 (30.6) 0 (0.0) 26 (72.2) 10 (27.8) Movement disorders 23 12 (52.2) 11 (47.8) 2 (8.7) 13 (56.5) 8 (34.8) Central nervous system malformations 5 4 (80.0) 1 (20.0) 1 (20.0) 4 (80.0) 0 (0.0) Other diseases 4 3 (75.0) 1 (25.0) 1 (25.0) 3 (75.0) 0 (0.0) Total 323 186 (57.6) 137 (42.4) 33 (10.2) 216 (66.9) 74 (22.9) (table continues) Data are given as number (percentage). Probands were classified into each disease category by the clinicians or researchers from the referring hospital. Other diseases included syndromic diseases that did not fit in any other group. computed on the genetic variants from all genomic experi- prediction22; and iv) candidate pathogenic CNVs based on ments using the PLINK software version 1.9019 following Online Mendelian Inheritance in Man database (https://www. previously suggested parameters.20 The total length of the omim.org, last accessed November 22, 2021) and Database RoH from each individual was used to estimate if it was an of Genomic Variants (http://dgv.tcag.ca, last accessed May offspring from a consanguineous couple, according to pre- 21, 2021).17 The filtering settings can be saved and applied viously described thresholds.21 recurrently to further analyses to speed up the filtering process. Variants were prioritized by clinical researchers Variant Filtering and Interpretation with the URD-Cat leveraging their expertise and the functionalities in the platform, including scored prioritization of variants GPAP according to patient’s HPO terms with Exomiser.23 In cases Genomic data were analyzed by geneticists from partici- with suspected consanguinity, the RoH filter was used to pating centers using the URD-Cat GPAP, which has many narrow down the list of candidate variants to only those functionalities, including standard filters and annotations within RoH of at least 500 Kb. Specific chromosomal (population databases and variant pathogenicity prediction positions (such as known pathogenic variants) could be tools), filters by clinical data (genes of interest, genes asso- selected individually or through the upload of a BED file. ciated with patient’s HPO terms entered in PhenoTips, On- Variant interpretation was done by geneticists and clinical line Mendelian Inheritance in Man codes, and in silico experts based on variant classification following the panels), links to multiple external resources, and data sharing American College of Medical Genetics and Genomics between authorized users (internal matchmaking by querying guidelines,24,25 the clinical fit, and the familial segregation. all of the data). Variant filtering and prioritization followed Validation and segregation of the candidate pathogenic the guidelines established by geneticists in the URD-Cat variants were performed by Sanger sequencing or project. Briefly, users selected the singleton, pair, trio, or comparative genomic hybridization arrays. Segregation with quartet to analyze and applied all possible inheritance pat- the disease was assessed for all patients unless otherwise terns to filter out the variants accordingly. Cutoffs for the indicated. Functional studies were performed when the population databases filters (GnomAD, 1000 Genomes, and variant and/or the gene had not been previously associated internal database) were set on the basis of the inheritance: with the disease. Reporting of findings to diagnosed minor allele frequency <0.02 for autosomal recessive or X- patients, or their families, was done according to the linked in men; and minor allele frequency <0.01 for auto- procedures of the corresponding managing hospital, which somal dominant or X-linked in women. Then, different typically includes a genetic report and counseling. filtering criteria were applied to look for the following: i) previously known variants: previously tagged by another user Statistical Analysis or reported in ClinVar database (https://www.ncbi.nlm.nih. gov/clinvar/, last accessed December 2, 2021); ii) variants The Fisher exact test was used to evaluate whether pheno- in genes associated with patient’s phenotype: filter by type annotation quality and sequencing approaches were genes associated with at least one patient’s HPO term, significantly different between diagnosed and undiagnosed predefined or custom gene lists, or Online Mendelian groups of patients. Nonparametric Kruskal-Wallis test was Inheritance in Man clinical features; iii) variants predicted done to compare the median number of HPO terms between to be highly likely pathogenic based on SnpEff diagnosed and undiagnosed patients. To remove annotation 532 jmdjournal.org - The Journal of Molecular Diagnostics Collaborative Reanalysis in Rare Disease Table 1 (continued) Sequencing strategy Family analysis Panel Exome Genome Singletons Pairs Trios Quads 2 (1.9) 103 (95.4) 3 (2.8) 73 (67.6) 18 (16.7) 16 (14.8) 1 (0.9) 28 (50.0) 28 (50.0) 0 (0.0) 49 (87.5) 4 (7.1) 2 (3.6) 1 (1.8) 2 (3.8) 50 (96.2) 0 (0.0) 17 (32.7) 3 (5.8) 30 (57.7) 2 (3.8) 9 (23.1) 30 (76.9) 0 (0.0) 21 (53.8) 3 (7.7) 13 (33.3) 2 (5.1) 6 (16.7) 30 (83.3) 0 (0.0) 17 (47.2) 2 (5.6) 12 (33.3) 5 (13.9) 13 (56.5) 10 (43.5) 0 (0.0) 18 (78.3) 1 (4.3) 4 (17.4) 0 (0.0) 2 (40.0) 3 (60.0) 0 (0.0) 5 (100.0) 0 (0.0) 0 (0.0) 0 (0.0) 1 (25.0) 3 (75.0) 0 (0.0) 3 (75.0) 0 (0.0) 1 (25.0) 0 (0.0) 63 (19.5) 257 (79.6) 3 (0.9) 203 (62.8) 31 (9.6) 78 (24.1) 11 (3.4) redundancy, only the most specific HPO terms were the same individual (n Z 2). These inconsistencies were considered by counting only terms from leaf nodes or nodes corrected in the URD-Cat platform. After complete without selected parent or child nodes. P < 0.05 was evaluation of the cohort, a total of 17 individuals were considered statistically significant for all analyses. excluded because they were duplicates (n Z 3), no pro- band was available (n Z 4), the proband was affected by a nonneurologic disease (Munchausen syndrome; n Z 3), Results or they were solved by another ongoing project (n Z 7). Systematic Evaluation of the Reanalysis Cohort Therefore, reanalysis of genomic data and variant inter- pretation was finally performed in a total of 323 probands Available genomics data from a total of 331 patients with (543 individuals including relatives) between 1 and 8 neurologic diseases (560 individuals including relatives) years (median, 4 years) after the data were originally were uploaded to the URD-Cat GPAP. Data were pro- generated. cessed as indicated in Materials and Methods. The mean The 323 probands included 186 males (57.6%) and 137 depth of coverage was 78.9 for panels, 74.1 for females (42.4%) (sex ratio, 1.4). They were classified into exomes, and 31.2 for genomes (Supplemental Table eight disease categories, according to the main clinical S2). Genomic relatedness was computed between 491 features: progressive neurodegenerative diseases (33.4%), individuals for which there was a genome or an exome neuromuscular diseases (17.3%), epilepsy/nonepileptic with average coverage >10. Six exomes with average paroxysmal disorders (16.1%), inherited metabolic disorders coverage 10 and 63 panels were not included in the (12.1%), intellectual disabilities/autism spectrum disorders relatedness analysis. Predicted kinship was compared with (11.2%), movement disorders (7.1%), central nervous sys- the reported family relationships. Inconsistencies were tem malformations (1.6%), and other diseases (1.2%) found in seven families, including duplicated individuals (Table 1). (n Z 3), true siblings reported as individuals belonging to Genetic variant-derived RoH analysis to predict the different families (n Z 2), and true siblings reported as absence or presence of consanguinity (consanguinity status) Table 2 Consanguinity Status Comparison between the Medical Records and the Runs of Homozygosity Analysis for the 257 Probands with Exome Sequencing Data Runs of homozygosity analysis Consanguineous/likely Variable consanguineous (n Z 24) Nonconsanguineous (n Z 183) Uncertain (n Z 50) Medical records (PhenoTips) Consanguineous (n Z 23) 15 3 5 Nonconsanguineous (n Z 164) 4 126 34 Unknown (n Z 70) 5 54 11 Four consanguinity ranges were established on the basis of the accumulated length of the runs of homozygosity (RoHs), as described in Matalonga et al21: consanguineous (total RoH size > 123 Mb), likely consanguineous (79 Mb < total RoH size < 123 Mb), uncertain (22 Mb < total RoH size < 79 Mb), and nonconsanguineous (total RoH size < 22 Mb). The Journal of Molecular Diagnostics - jmdjournal.org 533 Bullich et al Figure 1 Molecular results. A: Results overview from the reanalysis of 323 index cases. B: Results of the reanalysis by disease group. ID/ASD, intellectual disabilities/autism spectrum disorders; IMD, inherited metabolic disorders; NEPD, nonepileptic paroxysmal disorders. was performed in the 257 probands with exome data included: 3 targeted panels and 12 whole-exome kits available. A total of 24 probands were predicted to be part (Supplemental Table S1). DNA was mostly obtained from of a consanguineous (n Z 18) or likely consanguineous blood (535 individuals), although in some samples, the (n Z 6) family. In the medical records, the consanguinity DNA was extracted from muscle (5 individuals), fibroblasts status reported by the patient/family was available for 249 (3 individuals), or liver (1 individual). probands, 33 of which were tagged as consanguineous (Table 1). The consanguineous status predicted through the Molecular Diagnosis RoH analysis was compared with that reported in the medical records for the 257 probands with exome data Reanalysis of existing genomic data provided a conclusive available (Table 2). Consanguinity was confirmed in 15 genetic diagnosis for 67 probands (20.7%), including the probands. Discrepancies between the predicted and the re- identification of six novel gene-disease associations26e29 ported consanguinity were found in seven probands: four (V. Salpietro et al, unpublished data; and A. Pujol, unpub- cases predicted to be consanguineous had been reported as lished data). Sixty-one of the diagnoses (18.9%) were ach- nonconsanguineous, and three cases predicted to be non- ieved with just the reanalysis of existing data. Six of the consanguineous had been reported as consanguineous. The diagnoses (1.8% of the total probands with existing data) predicted consanguinity of 59 probands with no available were possible after identifying a single heterozygous information in the medical records indicated 5 would be candidate pathogenic variant in an autosomal recessive gene consanguineous (n Z 3) or likely consanguineous (n Z 2) through the reanalysis and the second causative variant and 54 would be nonconsanguineous. through additional sequencing (Figure 1A). Different sequencing strategies (panels, exomes, and The outcomes of the reanalysis for the different disease genomes) and different family analyses (singleton cases, groups are depicted in Figure 1B. Intellectual disabilities/ pairs, trios, or quartets) were included in the reanalysis autism spectrum disorders had the highest diagnostic rate (Table 1). Fifteen different sequence capture kits were (22.2%, 8 of 36), followed by progressive 534 jmdjournal.org - The Journal of Molecular Diagnostics Collaborative Reanalysis in Rare Disease Figure 2 Diagnostic yield by type of family analyses and sequencing strategy. The diagnostic yield of the different family analysis (singleton and nonsingletons) and sequencing approaches (panels, exomes, and genomes) was compared for the 61 probands fully diagnosed through reanalysis and the 256 undiagnosed probands. Patients diagnosed after additional sequencing were not considered for this analysis. Nonsingleton panel category is not shown because only one undiagnosed patient fell into this category. Asterisk denotes a statistically significant difference. *P < 0.05. WES, whole exome sequencing; WGS, whole genome sequencing. neurodegenerative diseases (21.3%, 23 of 108), epilepsy/ in the 67 diagnosed patients, consisting of 59 single- nonepileptic paroxysmal disorders (21.2%, 11 of 52), nucleotide variants, 16 insertions and deletions, and 6 inherited metabolic disorders (20.5%, 8 of 39), movement CNVs. The variants identified by additional sequencing disorders (17.4%, 4 of 23), and neuromuscular diseases were deep intronic variants (n Z 4) or variants located in (12.5%, 7 of 56). None of the patients classified within the poorly covered regions in the reanalyzed panel or exome central nervous system malformations group (n Z 5) or the (n Z 2). other diseases group (n Z 4) were diagnosed. Additional sequencing increased the diagnostic rate of inherited meta- bolic disorders, progressive neurodegenerative diseases, and neuromuscular diseases by around 5% (two patients), 3% (three patients), and 2% (one patient), respectively. Effect of the Sequencing Strategy on the Diagnostic Supplemental Table S326e62 provides further details on Rate the 67 probands fully diagnosed through reanalysis (n Z 61) and diagnosed through reanalysis plus additional The most prevalent sequencing approaches in the cohort sequencing (n Z 6). Autosomal dominant conditions were exome sequencing of singletons (42.7%, 138 of 323 accounted for 22 of the 67 (32.8%) diagnosed patients, 14 probands) and nonsingletons (36.9%, 119 of 323 probands). (63.6%) with a de novo variant, 4 (18.2%) inherited, 1 Panels and genomes accounted for 19.5% (63 of 323) and (4.6%) inherited from a mosaic-unaffected parent, and 3 0.9% (3 of 323) of the probands, respectively, and all cases (13.6%) undetermined because of unavailability of one or but one (a panel) were sequenced in singletons. both parental samples. Autosomal recessive conditions Analysis of the diagnostic rate for the different were found in 42 of the 67 (62.7%) diagnosed patients: 20 sequencing approaches revealed a significantly higher (47.6%) compound heterozygous cases and 22 (52.4%) diagnostic rate for exome or genome reanalysis compared homozygous cases. X-linked inheritance was reported in 3 with panels (22.0% and 8.1%, respectively; P Z 0.0116). cases of the 67 (4.5%) diagnosed patients, being dominant Reanalysis of more than one family member (pair, trio, or in 1 of them and recessive in the remaining 2 cases. CNVs quartet) also resulted in a higher diagnostic rate compared accounted for 8.9% of all of the diagnosed patients (6 of 67 with singleton analysis (23.7% and 16.6%, respectively; cases) and included four deletions (two homozygous, one P Z 0.1406), although no statistical significance was ach- X-linked hemizygous, and one heterozygous) and two ieved. Taken together, the highest diagnostic rate was ach- heterozygous duplications, ranging from 114 bp to almost ieved for exome or genome reanalysis with more than one 2 Mb. Altogether, a total of 81 different pathogenic vari- family member (23.9%) and the lowest for panel reanalysis ants in 61 different genes were detected through reanalysis in single cases (8.2%) (P Z 0.0137) (Figure 2). The Journal of Molecular Diagnostics - jmdjournal.org 535 Bullich et al Figure 3 Phenotypic quality. A: Monarch score distribution according to the number of Human Phenotype Ontology (HPO) terms for the 323 probands in the cohort. The Monarch score categories are as follows: bad (0 to 1.4), fair (1.5 to 2.4), good (2.5 to 3.4), very good (3.5 to 4.4), and excellent (4.5 to 5) phenotypic quality. B: Monarch category distribution for diagnosed and undiagnosed probands. The Monarch score categories for the 61 patients fully diagnosed through the reanalysis are excellent (n Z 2), very good (n Z 45), good (n Z 12), and fair (n Z 2). The Monarch score categories for the 256 undiagnosed patients are excellent (n Z 7), very good (n Z 187), good (n Z 44), fair (n Z 11), and bad (n Z 7). Patients diagnosed after additional sequencing were not considered for this analysis. n Z 61 diagnosed probands (B); n Z 256 undiagnosed probands (B). Quality of Standardized Phenotyping (Figure 3B). However, these results are not statistically significant (P Z 0.3863). Evaluation of the standardized HPO-coded phenotypic data rated 93.8% of the probands (303/323) as having good Reasons for Increasing the Diagnostic Rate with Data (17.6%, 57/323), very good (73.4%, 237/323), or excellent Reanalysis (2.8%, 9/323) quality, according to the Monarch score. The median Monarch score was 3.9 (very good category), The geneticists from the participating hospitals were asked ranging from 0 to 4.92. The Monarch score gradually to indicate why new diagnoses were achieved in the 61 increased with the number of HPO terms per patient until it probands fully diagnosed through reanalysis (Supplemental reached a plateau at approximately 25 HPO terms Table S3). Novel gene-disease associations explained (Figure 3A). The median number of nonredundant HPO 50.8% of the diagnoses (n Z 31): 41% (n Z 25) being in terms per patient was 24, ranging from 0 to 120. genes published in the literature after the original analysis Phenotype quality based on Monarch score categories and 9.8% (n Z 6) in genes discovered within this study. was compared between probands fully diagnosed through Additional or improved bioinformatics analyses were reanalysis (n Z 61) and undiagnosed probands (n Z 256). responsible for 19.7% of the diagnoses (n Z 12): 8.2% The median Monarch score was 3.9 for both diagnosed and (n Z 5) due to CNV detection, 8.2% (n Z 5) thanks to undiagnosed patients, and the median number of nonre- improved variant calling, and 3.3% (n Z 2) due to con- dundant HPO terms per patient was 25 for both groups. A sanguinity assessment, as the RoH filter could be used in slightly better phenotype quality was observed in diagnosed these two consanguineous cases to narrow down the list of patients, with 96.7% of them classified as having excellent, candidate pathogenic variants. The integration of standard- very good, or good phenotype quality compared with 93.0% ized HPO-coded phenotypic information with genomic data of undiagnosed patients falling into those categories was claimed to be crucial for 18% of the diagnoses 536 jmdjournal.org - The Journal of Molecular Diagnostics Collaborative Reanalysis in Rare Disease Figure 4 Reasons for increasing diagnostic rate with data reanalysis. The chart indicates reasons for which cases were diagnosed with the reanalysis according to the clinicians or researchers who submitted the data. The figure only includes the 61 probands who were diagnosed exclusively with reanalysis of previously existing data. CNV, copy number variant; GPAP, Genome-Phenome Analysis Platform; URD-Cat, Undiagnosed Rare Disease Program of Catalonia. (n Z 11). In the remaining 11.5% of diagnoses (n Z 7), the URD-Cat GPAP system for an efficient and comprehensive reason why the patient was not diagnosed in the original genomic analysis. Similarly, accurate and updated pheno- analysis was unknown (Figure 4). typic information was the main contributing factor to diagnosing patients in several studies.65e68 This study contemplates a broad spectrum of sequencing Discussion strategiesdsingleton and familial analysis in gene panel, exome, and genome sequencingdgiving us the opportunity The URD-Cat data reanalysis contributed to ending the to evaluate the diagnostic yield of the different approaches. diagnostic odyssey of 67 patients with a neurologic disease. The diagnostic rate progressively increased as a wider According to their referring clinicians, the genetic diagnosis approach was considered, from reanalysis of panels in single was important to improve patients’ medical management, to cases (8.2%), through exome/genome reanalysis in single- enable genetic counseling for parents, and to diagnose other tons (20.3%), to exome/genome reanalysis in more than one affected family members. In addition, families were able to family member (23.9%). Similarly, trio exome analysis has contact patients’ associations, which made them feel more been associated with an approximately 7% to 9% higher supported. molecular diagnostic rate compared with proband-only The diagnostic rate of 18.9% obtained in this reanalysis analysis.69,70 Besides the higher diagnostic rate, data inter- study (without including additional data) is similar to that pretation in trio exome sequencing is more straightforward reported in the literature, with 15% being the median as segregation can be used to prioritize variants.64,71 How- diagnostic rate for reanalysis studies.63 Most new diagnoses ever, the higher cost derived from sequencing more than one resulted from novel gene-disease associations (50.8%), family member still hampers its routine use in diagnostic either published after initial analysis or discovered with the settings.72 Thus, singleton exome reanalysis is one of the reanalysis26e29 (V. Salpietro et al, unpublished data; and A. most prevalent sequencing strategies reported in reanalysis Pujol, unpublished data). Because approximately 250 novel studies.64,66,71,73,74 The limited number of genome samples gene-disease associations are reported each year, this was in the cohort makes it difficult to draw any conclusions quite expected.64 New molecular diagnoses were also about the utility of genome sequencing compared with possible due to additional or improved bioinformatics ana- exome sequencing, but several studies have highlighted the lyses (19.7%), supporting the idea that reanalysis studies importance of exome reanalysis before performing genome should include consanguinity assessment, updated pipelines, sequencing.67,75,76 and analysis of single-nucleotide variants and CNVs. In The reanalysis approach presented herein is innovative fact, CNVs accounted for 8.9% of the diagnosed patients. for at least three reasons. First, the systematic evaluation of The standardized high-quality phenotyping promoted within the data proved to be effective to predict and check familial the URD-Cat project, and its integration with the genomic relationships and consanguineous status. The large-scale data, was the reason for the new diagnoses in 18% of the and highly collaborative nature of the URD-Cat project, diagnosed patients, highlighting the valuable role of the with samples coming from the main groups that attend and The Journal of Molecular Diagnostics - jmdjournal.org 537 Bullich et al investigate patients with undiagnosed diseases in Catalonia, the professionals implicated in a patient’s genetic diagnosis, implied an increased risk of duplicated samples or incorrect bringing together otherwise scattered expertise. Clinical familial relationships that could hamper the identification of information can be updated at any time, and reinterpretation the pathogenic variant. In fact, three duplicated patients of genomic variants can efficiently be done through all were excluded, and four inconsistent family relationships functionalities included in the system. Data sharing is were corrected thanks to the relatedness analysis. Consan- necessary to advance faster in rare disease research and guinity estimation through RoH analysis was used to predict diagnosis because of the low prevalence (<5/10,000 in the the consanguinity status in patients with no previous infor- European Union), high genetic heterogeneity, and pheno- mation (59 cases) and to detect discrepancies between the typic variability of these diseases.79 International data- consanguinity status reported in the medical records and the sharing platforms have been reported to be relevant for estimated one (7 cases). This information was useful to the diagnosis of reanalysis patients.63,72 Herein, genomic guide the filtering strategy and to focus the analysis in the data sharing among all centers involved in the URD-Cat homozygous regions for the consanguineous cases. How- project was crucial to diagnose some patients, such as Pa- ever, all of the inheritance patterns were assessed in all tient URD-Cat_02, who was diagnosed by a researcher from patients because de novo variants can also cause a disease in another center by internal matchmaking. The platform is consanguineous families, as illustrated by cases URD- periodically updated with disease and population databases, Cat_06 (likely de novo variant in the SPAST gene) and and new functionalities are included on the basis of genet- URD-Cat_35 (de novo duplication in the HNRNPH1 gene). icists’ needs in a nice example of close collaboration among Two cases in the cohort were diagnosed with a homozygous clinicians, geneticists, bioinformaticians, and software en- pathogenic variant despite being reported in the medical gineers. A model based on continuous interaction between record as nonconsanguineous (Patient URD-Cat_21) or with all of the parties implicated in a patient’s genetic diagnosis unknown consanguinity status (Patient URD-Cat_41). In has been proposed for automated iterative reanalysis.77 light of this, conducting a systematic evaluation of the data In conclusion, the systematic collaborative reanalysis of through relatedness and consanguinity analysis to ensure a genomic data within the URD-Cat project diagnosed 20.7% correct characterization of the cohort is highly of probands, thanks to standardized high-quality pheno- recommended. typing, reprocessing of genomic data, reinterpretation of the Second, standardized high-quality phenotyping was ach- results, and data sharing. The URD-Cat GPAP has been key ieved in the cohort through collection of clinical symptoms for systematic collation, analysis, and interpretation of the in the form of HPO terms. HPO terms have been used to data in a collaborative manner. collect phenotypic data in some reanalysis studies.4,64,71,76 However, the phenotype quality achieved was not evalu- ated. Herein, 93.8% of patients had good or higher pheno- Acknowledgments type quality based on Monarch score. The specificity of the phenotype gradually increased with the number of HPO We thank the Undiagnosed Rare Disease Program of Cat- terms per patient until it reached a plateau at approximately alonia (URD-Cat) consortium: Agatha Schlüter, Agustí 25 HPO terms. Diagnosed patients had a higher proportion Rodriguez-Palmero, Alejandro Cáceres, Andrés Nasci- of good, very good, or excellent Monarch scores (96.7%) mento, Àngels García-Cazorla, Anna Cueto-González, compared with undiagnosed (93%) ones, although no sta- Anna Marcé-Grau, Anna Ruiz Nel.lo, Antonio Martínez- tistical significance was obtained, likely because of the high Monseny, Aurora Sànchez, Belén García, Belén Pérez- phenotype quality achieved for the whole cohort. Similarly, Dueñas, Bernat Gel, Berta Fusté, Carles Hernández-Ferrer, Wright et al4 did not find phenotypic differences between Carlos Casasnovas, Carlos Ortez, César Arjona, Cristina diagnosed and undiagnosed in terms of number of HPO Hernando-Davalillo, Daniel Natera de Benito, Daniel Picó terms per patient, similarity by most informative term, or Amador, David Gómez-Andrés, Dèlia Yubero, Dolors similarity by Jaccard index. Pelegrí-Sisó, Edgard Verdura, Elena García-Arumí, Eli- Third, the URD-Cat GPAP allows efficient and contin- sabeth Castellanos, Elisabeth Gabau, Ester Tobías, Fermina uous reanalysis of genomic data and is extremely useful for López-Grondona, Francesc Cardellach, Francesc Josep collaborative efforts as it can be used by clinicians and Garcia-Garcia, Francina Munell, Frederic Tort, Gemma technicians thanks to its user-friendly interface. Periodic Aznar, Gemma Olivé-Cirera, Gemma Tell, Gerard Muñoz- reanalysis and reinterpretation of sequencing data coupled Pujol, Ida Paramonov, Ignacio Blanco, Irene Madrigal, Irene with updates in patients’ phenotype is recommended by Valenzuela, Ivo Gut, Ivon Cusco, Jean-Rémi Trotta, Jordi several studies, including the American College of Medical Cruz, Jordi Díaz-Manera, José César Milisenda, Josep Ma Genetics and Genomics.4,24,64,77,78 However, bioinformatics Grau, Judit Garcia-Villoria, Judith Armstrong, Judith Cantó, infrastructures must be able to cope with the organizational Júlia Sala-Coromina, Laia Rodríguez-Revenga, Laura Alias, and communicative challenge that iterative reanalysis Laura Gort, Lídia González-Quereda, Mar Costa, Marcos require.4,71,77 The URD-Cat platform is a key tool for Fernández-Callejo, Marcos López-Sánchez, Maria Isabel continuous communication and data sharing among all of Álvarez-Mora, Marta Gut, Mercedes Serrano, Miquel 538 jmdjournal.org - The Journal of Molecular Diagnostics Collaborative Reanalysis in Rare Disease Raspall-Chaure, Mireia del Toro, Mònica Bayés, Neus EURenOmics: collaborative European initiative for rare diseases. Eur Baena Díez, Nino Spataro, Núria Capdevila, Olatz Ugarte- J Hum Genet 2018, 26:778e785 7. Girdea M, Dumitriu S, Fiume M, Bowdin S, Boycott KM, Chénier S, buru, Patricia Muñoz-Cabello, Penélope Romero Duque, Chitayat D, Faghfoury H, Meyn MS, Ray PN, So J, Stavropoulos DJ, Raquel Rabionet, Ricard Rojas-García, Rosa Calvo, Roser Brudno M: PhenoTips: patient phenotyping software for clinical and Urreizti, Sara Bernal, Susana Boronat, Susanna Balcells, research use. Hum Mutat 2013, 34:1057e1065 Teresa Vendrell. 8. Robinson PN, Köhler S, Bauer S, Seelow D, Horn D, Mundlos S: The human phenotype ontology: a tool for annotating and analyzing human hereditary disease. Am J Hum Genet 2008, 83:610e615 Author Contributions 9. Rath A, Olry A, Dhombres F, Brandt MM, Urbero B, Ayme S: Representation of rare diseases in health information systems: the Orphanet approach to serve a wide range of end users. Hum Mutat G.B., L.M., G.P., and D.O. processed and analyzed data; 2012, 33:803e808 M.P. coordinated the Undiagnosed Rare Disease Program of 10. Mungall CJ, McMurry JA, Köhler S, Balhoff JP, Borromeo C, Catalonia (URD-Cat) project; D.P., A.Pa., and C.Lu. Brush M, Carbon S, Conlin T, Dunn N, Engelstad M, Foster E, developed and released data to the URD-Cat platform; S.L., Gourdine JP, Jacobsen JOB, Keith D, Laraway B, Lewis SE, R.T., and D.O. implemented the analysis pipelines; R.A., NguyenXuan J, Shefchek K, Vasilevsky N, Yuan Z, Washington N, Hochheiser H, Groza T, Smedley D, Robinson PN, Haendel MA: The P.G., G.G., J.R.G., D.G., M.G., C.La., R.M., M.M., A.Pu., Monarch Initiative: an integrative data and analytic platform con- E.T., A.M., F.P., A.R., and L.A.P.-J. recruited patient necting phenotypes to genotypes across species. Nucleic Acids Res clinical and genomic data and coordinated molecular diag- 2017, 45:D712eD722 nosis feedback; S.B. and G.B. designed the study and wrote 11. Laurie S, Fernandez-Callejo M, Marco-Sola S, Trotta J-R, Camps J, the manuscript; all authors revised the article and read and Chacón A, Espinosa A, Gut M, Gut I, Heath S, Beltran S: From wet- lab to variations: concordance and speed of bioinformatics pipelines approved the final manuscript. S.B. is the guarantor of this for whole genome and whole exome sequencing. Hum Mutat 2016, work and, as such, had full access to all of the data in the 37:1263e1271 study and takes responsibility for the integrity of the data 12. DePristo MA, Banks E, Poplin R, Garimella KV, Maguire JR, and the accuracy of the data analysis. Hartl C, Philippakis AA, del Angel G, Rivas MA, Hanna M, McKenna A, Fennell TJ, Kernytsky AM, Sivachenko AY, Cibulskis K, Gabriel SB, Altshuler D, Daly MJ: A framework for Supplemental Data variation discovery and genotyping using next-generation DNA sequencing data. Nat Genet 2011, 43:491e498 13. Li H: Aligning sequence reads, clone sequences and assembly contigs Supplemental material for this article can be found at with BWA-MEM. arXiv 2013. [Preprint] doi:10.48550/arXiv.1303. http://doi.org/10.1016/j.jmoldx.2022.02.003. 3997 14. McKenna A, Hanna M, Banks E, Sivachenko A, Cibulskis K, Kernytsky A, Garimella K, Altshuler D, Gabriel S, Daly M, References DePristo MA: The Genome Analysis Toolkit: a MapReduce frame- work for analyzing next-generation DNA sequencing data. Genome 1. Nguengang Wakap S, Lambert DM, Olry A, Rodwell C, Gueydan C, Res 2010, 20:1297e1303 Lanneau V, Murphy D, Le Cam Y, Rath A: Estimating cumulative 15. Plagnol V, Curtis J, Epstein M, Mok KY, Stebbings E, point prevalence of rare diseases: analysis of the Orphanet database. Grigoriadou S, Wood NW, Hambleton S, Burns SO, Thrasher AJ, Eur J Hum Genet 2020, 28:165e173 Kumararatne D, Doffinger R, Nejentsev S: A robust model for 2. Walker CE, Mahede T, Davis G, Miller LJ, Girschik J, Brameld K, read count data in exome sequencing experiments and implica- Sun W, Rath A, Aymé S, Zubrick SR, Baynam GS, Molster C, tions for copy number variant calling. Bioinformatics 2012, 28: Dawkins HJS, Weeramanthri TS: The collective impact of rare dis- 2747e2754 eases in Western Australia: an estimate using a population-based 16. Conrad DF, Pinto D, Redon R, Feuk L, Gokcumen O, Zhang Y, cohort. Genet Med 2017, 19:546e552 Aerts J, Andrews TD, Barnes C, Campbell P, Fitzgerald T, Hu M, 3. Neu MB, Bowling KM, Cooper GM: Clinical utility of genomic Ihm CH, Kristiansson K, Macarthur DG, Macdonald JR, Onyiah I, sequencing. Curr Opin Pediatr 2019, 31:732e738 Pang AWC, Robson S, Stirrups K, Valsesia A, Walter K, Wei J, 4. Wright CF, McRae JF, Clayton S, Gallone G, Aitken S, Wellcome Trust Case Control Consortium, Tyler-Smith C, Carter NP, FitzGerald TW, Jones P, Prigmore E, Rajan D, Lord J, Sifrim A, Lee C, Scherer SW, Hurles ME: Origins and functional impact of copy Kelsell R, Parker MJ, Barrett JC, Hurles ME, FitzPatrick DR, Firth HV, number variation in the human genome. Nature 2010, 464:704e712 DDD Study: Making new genetic diagnoses with old data: iterative 17. MacDonald JR, Ziman R, Yuen RKC, Feuk L, Scherer SW: The reanalysis and reporting from genome-wide data in 1,133 families with Database of Genomic Variants: a curated collection of structural developmental disorders. Genet Med 2018, 20:1216e1223 variation in the human genome. Nucleic Acids Res 2014, 42: 5. Thompson R, Johnston L, Taruscio D, Monaco L, Béroud C, Gut IG, D986eD992 Hansson MG, ’t Hoen P-BA, Patrinos GP, Dawkins H, Ensini M, 18. Pedersen BS, Bhetariya PJ, Brown J, Marth G, Jensen RL, Zatloukal K, Koubi D, Heslop E, Paschall JE, Posada M, Bronner MP, Underhill HR, Quinlan AR: Somalier: rapid relatedness Robinson PN, Bushby K, Lochmüller H: RD-Connect: an integrated estimation for cancer and germline studies using efficient genome platform connecting databases, registries, biobanks and clinical bio- sketches. Genome Med 2020, 12:62 informatics for rare disease research. J Gen Intern Med 2014, 19. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MAR, 29(Suppl 3):S780eS787 Bender D, Maller J, Sklar P, de Bakker PIW, Daly MJ, Sham PC: 6. Lochmüller H, Badowska DM, Thompson R, Knoers NV, Aartsma- PLINK: a tool set for whole-genome association and population- Rus A, Gut I, Wood L, Harmuth T, Durudas A, Graessner H, based linkage analyses. Am J Hum Genet 2007, 81:559e575 Schaefer F, Riess O, RD-Connect consortium, NeurOmics con- 20. Kancheva D, Atkinson D, De Rijk P, Zimon M, Chamova T, sortium, EURenOmics consortium: RD-Connect, NeurOmics and Mitev V, Yaramis A, Maria Fabrizi G, Topaloglu H, Tournev I, The Journal of Molecular Diagnostics - jmdjournal.org 539 Bullich et al Parman Y, Parma Y, Battaloglu E, Estrada-Cuzcano A, Jordanova A: novel chondrodysplasia with hypomyelinating leukodystrophy. Brain Novel mutations in genes causing hereditary spastic paraplegia and 2022 [Epub ahead of print] doi:10.1093/brain/awac110 Charcot-Marie-Tooth neuropathy identified by an optimized protocol 30. Bozarth X, Dines JN, Cong Q, Mirzaa GM, Foss K, Lawrence for homozygosity mapping based on whole-exome sequencing. Genet Merritt J 2nd, Thies J, Mefford HC, Novotny E: Expanding clinical Med 2016, 18:600e607 phenotype in CACNA1C related disorders: from neonatal onset se- 21. Matalonga L, Laurie S, Papakonstantinou A, Piscia D, Mereu E, vere epileptic encephalopathy to late-onset epilepsy. Am J Med Genet Bullich G, Thompson R, Horvath R, Pérez-Jurado L, Riess O, Gut I, A 2018, 176:2733e2739 van Ommen G-J, Lochmüller H, Beltran S, RDeConnect Genome- 31. Beck DB, Cho MT, Millan F, Yates C, Hannibal M, O’Connor B, Phenome Analysis Platform and URD-Cat Data Contributors: Shinawi M, Connolly AM, Waggoner D, Halbach S, Angle B, Improved diagnosis of rare disease patients through systematic Sanders V, Shen Y, Retterer K, Begtrup A, Bai R, Chung WK: A detection of runs of homozygosity. J Mol Diagn 2020, 22: recurrent de novo CTBP1 mutation is associated with developmental 1205e1215 delay, hypotonia, ataxia, and tooth enamel defects. Neurogenetics 22. Cingolani P, Platts A, Wang LL, Coon M, Nguyen T, Wang L, 2016, 17:173e178 Land SJ, Lu X, Ruden DM: A program for annotating and predicting 32. Gu S, Chen C-A, Rosenfeld JA, Cope H, Launay N, Flanigan KM, the effects of single nucleotide polymorphisms, SnpEff: SNPs in the Waldrop MA, Schrader R, Juusola J, Goker-Alpan O, Milunsky A, genome of Drosophila melanogaster strain w1118; iso-2; iso-3. Fly Schlüter A, Troncoso M, Pujol A, Tan QK-G, Schaaf CP, Meng L: 2012, 6:80e92 Truncating variants in UBAP1 associated with childhood-onset 23. Smedley D, Jacobsen JOB, Jäger M, Köhler S, Holtgrewe M, nonsyndromic hereditary spastic paraplegia. Hum Mutat 2020, 41: Schubach M, Siragusa E, Zemojtel T, Buske OJ, Washington NL, 632e640 Bone WP, Haendel MA, Robinson PN: Next-generation diagnostics 33. Marcogliese PC, Shashi V, Spillmann RC, Stong N, Rosenfeld JA, and disease-gene discovery with the Exomiser. Nat Protoc 2015, 10: Koenig MK, et al: IRF2BPL is associated with neurological pheno- 2004e2015 types. Am J Hum Genet 2018, 103:245e260 24. Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, 34. Fonknechten N, Mavel D, Byrne P, Davoine CS, Cruaud C, Grody WW, Hegde M, Lyon E, Spector E, Voelkerding K, Rehm HL, Bönsch D, Samson D, Coutinho P, Hutchinson M, McMonagle P, ACMG Laboratory Quality Assurance Committee: Standards and Burgunder JM, Tartaglione A, Heinzlef O, Feki I, Deufel T, guidelines for the interpretation of sequence variants: a joint Parfrey N, Brice A, Fontaine B, Prud’homme JF, Weissenbach J, consensus recommendation of the American College of Medical Dürr A, Hazan J: Spectrum of SPG4 mutations in autosomal domi- Genetics and Genomics and the Association for Molecular Pathology. nant spastic paraplegia. Hum Mol Genet 2000, 9:637e644 Genet Med 2015, 17:405e424 35. Arrudi-Moreno M, Fernández-Gómez A, Peña-Segura JL: A new 25. Riggs ER, Andersen EF, Cherry AM, Kantarci S, Kearney H, Patel A, mutation in the SEPSECS gene related to pontocerebellar hypoplasia Raca G, Ritter DI, South ST, Thorland EC, Pineda-Alvarez D, type 2D. Med Clin 2021, 156:94e95 Aradhya S, Martin CL: Technical standards for the interpretation and 36. Agamy O, Ben Zeev B, Lev D, Marcus B, Fine D, Su D, Narkis G, reporting of constitutional copy-number variants: a joint consensus Ofir R, Hoffmann C, Leshinsky-Silver E, Flusser H, Sivan S, Söll D, recommendation of the American College of Medical Genetics and Lerman-Sagie T, Birk OS: Mutations disrupting selenocysteine for- Genomics (ACMG) and the Clinical Genome Resource (ClinGen). mation cause progressive cerebello-cerebral atrophy. Am J Hum Genet Med 2020, 22:245e257 Genet 2010, 87:538e544 26. Reichert SC, Li R, A Turner S, van Jaarsveld RH, Massink MPG, van 37. La Piana R, Cayami FK, Tran LT, Guerrero K, van Spaendonk R, den Boogaard M-JH, Del Toro M, Rodríguez-Palmero A, Fourcade S, Õunap K, Pajusalu S, Haack T, Wassmer E, Timmann D, Schlüter A, Planas-Serra L, Pujol A, Iascone M, Maitz S, Loong L, Mierzewska H, Poll-Thé BT, Patel C, Cox H, Atik T, Onay H, Stewart H, De Franco E, Ellard S, Frank J, Lewandowski R: Ozkınay F, Vanderver A, van der Knaap MS, Wolf NI, Bernard G: HNRNPH1-related syndromic intellectual disability: seven additional Diffuse hypomyelination is not obligate for POLR3-related disorders. cases suggestive of a distinct syndromic neurodevelopmental syn- Neurology 2016, 86:1622e1626 drome. Clin Genet 2020, 98:91e98 38. Verdura E, Rodríguez-Palmero A, Vélez-Santamaria V, Planas- 27. García-Cazorla À, Verdura E, Juliá-Palacios N, Anderson EN, Serra L, de la Calle I, Raspall-Chaure M, Roubertie A, Benkirane M, Goicoechea L, Planas-Serra L, Tsogtbaatar E, Dsouza NR, Saettini F, Pavinato L, Mandrile G, O’Leary M, O’Heir E, Barredo E, Schlüter A, Urreizti R, Tarnowski JM, Gavrilova RH, SHMT2 Chacón A, Michaud V, Goizet C, Ruiz M, Schlüter A, Rouvet I, Sala- Working Group, Ruiz M, Rodríguez-Palmero A, Fourcade S, Coromina J, Fossati C, Iascone M, Canonico F, Marcé-Grau A, de Cogné B, Besnard T, Vincent M, Bézieau S, Folmes CD, Souza P, Adams DR, Casasnovas C, Rehm HL, Mefford HC, Gon- Zimmermann MT, Klee EW, Pandey UB, Artuch R, Cousin MA, zález Gutierrez-Solana L, Brusco A, Koenig M, Macaya A, Pujol A: Pujol A: Impairment of the mitochondrial one-carbon metabolism Biallelic PI4KA variants cause a novel neurodevelopmental syn- enzyme SHMT2 causes a novel brain and heart developmental syn- drome with hypomyelinating leukodystrophy. Brain 2021, 144: drome. Acta Neuropathol 2020, 140:971e975 2659e2669 28. Schlüter A, Rodríguez-Palmero A, Verdura E, Vélez-Santamaría V, 39. Pant DC, Dorboz I, Schluter A, Fourcade S, Launay N, Joya J, et al: Ruiz M, Fourcade S, Planas-Serra L, Martínez JJ, Guilera C, Girós M, Loss of the sphingolipid desaturase DEGS1 causes hypomyelinating Artuch R, Yoldi ME, O’Callaghan M, García-Cazorla A, leukodystrophy. J Clin Invest 2019, 129:1240e1256 Armstrong J, Redin C, Mandel JL, Conejo D, Sierra-Córcoles C, 40. Ramantani G, Kohlhase J, Hertzberg C, Innes AM, Engel K, Hunger S, Beltran S, Gut M, Vázquez E, del Toro M, Troncoso M, Pérez- Borozdin W, Mah JK, Ungerath K, Walkenhorst H, Richardt H-H, Jurado LA, Gutiérrez-Solana L, López de Munain A, Casanovas C, Buckard J, Bevot A, Siegel C, von Stülpnagel C, Ikonomidou C, Aguilera-Albesa S, Macaya A, Pujol A, GWMD working group: Thomas K, Proud V, Niemann F, Wieczorek D, Häusler M, Diagnosis of genetic white matter disorders by singleton whole- Niggemann P, Baltaci V, Conrad K, Lebon P, Lee-Kirsch MA: exome and genome sequencing using interactome-driven prioritiza- Expanding the phenotypic spectrum of lupus erythematosus in Aicardi- tion. Neurology 2022, 98:e912ee923 Goutières syndrome. Arthritis Rheum 2010, 62:1469e1477 29. Guasto A, Dubai J, Aguilera-Albesa S, Paganini C, Vanhulle C, 41. Vélez-Santamaría V, Verdura E, Macmurdo C, Planas-Serra L, Haouari W, Gorría-Redondo N, Aznal-Sainz E, Boddaert N, Planas- Schlüter A, Casas J, Martínez JJ, Casasnovas C, Si Y, Thompson SS, Serra L, Schlüter A, Verdura E, Bruneel A, Rossi A, Huber C, Maroofian R, Pujol A: Expanding the clinical and genetic spectrum of Pujol A, Cormier-Daire V: Biallelic variants in SLC35B2 cause a PCYT2-related disorders. Brain 2020, 143:e76 540 jmdjournal.org - The Journal of Molecular Diagnostics Collaborative Reanalysis in Rare Disease 42. Marti-Sanchez L, Baide-Mairena H, Marcé-Grau A, Pons R, Robb S, Munot P, Wei Liu W, Banka S, Fraser H, De Goede C, Skouma A, López-Laso E, Sigatullina M, Rizzo C, Semeraro M, Zanoteli E, Conti Reed U, Sage A, Gratacos M, Macaya A, Dusl M, Martinelli D, Carrozzo R, Dionisi-Vici C, González-Gutiérrez- Senderek J, Töpf A, Hofer M, Knight R, Ramdas S, Jayawant S, Solana L, Correa-Vela M, Ortigoza-Escobar JD, Sánchez- Lochmüller H, Palace J, Beeson D: The clinical spectrum of the Montañez Á, Vazquez É, Delgado I, Aguilera-Albesa S, Yoldi ME, congenital myasthenic syndrome resulting from COL13A1 mutations. Ribes A, Tort F, Pollini L, Galosi S, Leuzzi V, Tolve M, Pérez- Brain 2019, 142:1547e1560 Gay L, Aldamiz-Echevarría L, Del Toro M, Arranz A, Roelens F, 56. Bijlsma EK, Gijsbers ACJ, Schuurs-Hoeijmakers JHM, van Urreizti R, Artuch R, Macaya A, Pérez-Dueñas B: Delineating the Haeringen A, Fransen van de Putte DE, Anderlid B-M, Lundin J, neurological phenotype in children with defects in the ECHS1 or Lapunzina P, Pérez Jurado LA, Delle Chiaie B, Loeys B, Menten B, HIBCH gene. J Inherit Metab Dis 2021, 44:401e414 Oostra A, Verhelst H, Amor DJ, Bruno DL, van Essen AJ, Hordijk R, 43. Nakajima J, Okamoto N, Tohyama J, Kato M, Arai H, Funahashi O, Sikkema-Raddatz B, Verbruggen KT, Jongmans MCJ, Pfundt R, Tsurusaki Y, Nakashima M, Kawashima H, Saitsu H, Matsumoto N, Reeser HM, Breuning MH, Ruivenkamp CAL: Extending the Miyake N: De novo EEF1A2 mutations in patients with characteristic phenotype of recurrent rearrangements of 16p11.2: deletions in facial features, intellectual disability, autistic behaviors and epilepsy. mentally retarded patients without autism and in normal individuals. Clin Genet 2015, 87:356e361 Eur J Med Genet 2009, 52:77e87 44. Lazo PA, García JL, Gómez-Puertas P, Marcos-Alcalde Í, Arjona C, 57. Schollen E, Martens K, Geuzens E, Matthijs G: DHPLC analysis as a Villarroel A, González-Sarmiento R, Fons C: Novel dominant platform for molecular diagnosis of congenital disorders of glyco- KCNQ2 exon 7 partial in-frame duplication in a complex epileptic sylation (CDG). Eur J Hum Genet 2002, 10:643e648 and neurodevelopmental delay syndrome. Int J Mol Sci 2020, 21: 58. Salpietro V, Perez-Dueñas B, Nakashima K, San Antonio-Arce V, 4447 Manole A, Efthymiou S, Vandrovcova J, Bettencourt C, 45. Heron SE, Scheffer IE, Grinton BE, Eyre H, Oliver KL, Bain S, Mencacci NE, Klein C, Kelly MP, Davies CH, Kimura H, Macaya A, Berkovic SF, Mulley JC: Familial neonatal seizures with intellectual Houlden H: A homozygous loss-of-function mutation in PDE2A disability caused by a microduplication of chromosome 2q24.3. associated to early-onset hereditary chorea. Mov Disord 2018, 33: Epilepsia 2010, 51:1865e1869 482e488 46. Berkovic SF, Heron SE, Giordano L, Marini C, Guerrini R, 59. Verdura E, Schlüter A, Fernández-Eulate G, Ramos-Martín R, Kaplan RE, Gambardella A, Steinlein OK, Grinton BE, Dean JT, Zulaica M, Planas-Serra L, Ruiz M, Fourcade S, Casasnovas C, Bordo L, Hodgson BL, Yamamoto T, Mulley JC, Zara F, Scheffer IE: López de Munain A, Pujol A: A deep intronic splice variant advises Benign familial neonatal-infantile seizures: characterization of a new reexamination of presumably dominant SPG7 Cases. Ann Clin Transl sodium channelopathy. Ann Neurol 2004, 55:550e557 Neurol 2020, 7:105e111 47. Perea-Romero I, Blanco-Kelly F, Sanchez-Navarro I, Lorda- 60. Fernández-Marmiesse A, Carrascosa-Romero MC, Alfaro Ponce B, Sanchez I, Tahsin-Swafiri S, Avila-Fernandez A, Martin-Merida I, Nascimento A, Ortez C, Romero N, Palacios L, Jimenez-Mallebrera C, Trujillo-Tiebas MJ, Lopez-Rodriguez R, Rodriguez de Alba M, Jou C, Gouveia S, Couce ML: Homozygous truncating mutation in Iancu IF, Romero R, Quinodoz M, Hakonarson H, Garcia-Sandova B, prenatally expressed skeletal isoform of TTN gene results in arthrog- Minguez P, Corton M, Rivolta C, Ayuso C: NGS and phenotypic ryposis multiplex congenita and myopathy without cardiac involve- ontology-based approaches increase the diagnostic yield in syndromic ment. Neuromuscul Disord 2017, 27:188e192 retinal diseases. Hum Genet 2021, 140:1665e1678 61. Luzi P, Rafi MA, Rao HZ, Wenger DA: Sixteen novel mutations in 48. Reuber BE, Germain-Lee E, Collins CS, Morrell JC, Ameritunga R, the arylsulfatase A gene causing metachromatic leukodystrophy. Moser HW, Valle D, Gould SJ: Mutations in PEX1 are the most Gene 2013, 530:323e328 common cause of peroxisome biogenesis disorders. Nat Genet 1997, 62. Polten A, Fluharty AL, Fluharty CB, Kappler J, von Figura K, 17:445e448 Gieselmann V: Molecular basis of different forms of metachromatic 49. Hamdan FF, Myers CT, Cossette P, Lemay P, Spiegelman D, leukodystrophy. N Engl J Med 1991, 324:18e22 Laporte AD, et al: High rate of recurrent de novo mutations in 63. Tan NB, Stapleton R, Stark Z, Delatycki MB, Yeung A, Hunter MF, developmental and epileptic encephalopathies. Am J Hum Genet Amor DJ, Brown NJ, Stutterd CA, McGillivray G, Yap P, Regan M, 2017, 101:664e685 Chong B, Fanjul Fernandez M, Marum J, Phelan D, Pais LS, 50. Takenouchi T, Miura K, Uehara T, Mizuno S, Kosaki K: Establishing White SM, Lunke S, Tan TY: Evaluating systematic reanalysis of SON in 21q22.11 as a cause a new syndromic form of intellectual clinical genomic data in rare disease from single center experience disability: possible contribution to Braddock-Carey syndrome and literature review. Mol Genet Genomic Med 2020, 8:e1508 phenotype. Am J Med Genet A 2016, 170:2587e2590 64. Wenger AM, Guturu H, Bernstein JA, Bejerano G: Systematic 51. Gordo G, Tenorio J, Arias P, Santos-Simarro F, García-Miñaur S, reanalysis of clinical exome data yields additional diagnoses: impli- Moreno JC, Nevado J, Vallespin E, Rodriguez-Laguna L, de Mena R, cations for providers. Genet Med 2017, 19:209e214 Dapia I, Palomares-Bralo M, Pozo Á Del, Ibañez K, Silla JC, 65. Ewans LJ, Schofield D, Shrestha R, Zhu Y, Gayevskiy V, Ying K, Barroso E, Ruiz-Pérez VL, Martinez-Glez V, Lapunzina P: mTOR Walsh C, Lee E, Kirk EP, Colley A, Ellaway C, Turner A, Mowat D, mutations in Smith-Kingsmore syndrome: four additional patients and Worgan L, Freckmann M-L, Lipke M, Sachdev R, Miller D, Field M, a review. Clin Genet 2018, 93:762e775 Dinger ME, Buckley MF, Cowley MJ, Roscioli T: Whole-exome  52. Sedivá M, Lassuthová P, Zámecník J, Sedlácková L, Seeman P, sequencing reanalysis at 12 months boosts diagnosis and is cost- Haberlová J: Novel variant in the KCNK9 gene in a girl with Birk effective when applied early in Mendelian disorders. Genet Med Barel syndrome. Eur J Med Genet 2020, 63:103619 2018, 20:1564e1574 53. Kim SY, Kim WJ, Kim H, Choi SA, Lee JS, Cho A, Jang SS, 66. Basel-Salmon L, Orenstein N, Markus-Bustani K, Ruhrman- Lim BC, Kim KJ, Kim J-I, Hahn SH, Chae J-H: Collagen VI-related Shahar N, Kilim Y, Magal N, Hubshman MW, Bazak L: Improved myopathy: expanding the clinical and genetic spectrum. Muscle diagnostics by exome sequencing following raw data reevaluation by Nerve 2018, 58:381e388 clinical geneticists involved in the medical care of the individuals 54. Dusl M, Moreno T, Munell F, Macaya A, Gratacòs M, Abicht A, tested. Genet Med 2019, 21:1443e1451 Strom TM, Lochmüller H, Senderek J: Congenital myasthenic syn- 67. Shashi V, Schoch K, Spillmann R, Cope H, Tan QK-G, Walley N, drome caused by novel COL13A1 mutations. J Neurol 2019, 266: Pena L, McConkie-Rosell A, Jiang Y-H, Stong N, Need AC, 1107e1112 Goldstein DB, Undiagnosed Diseases Network: A comprehensive 55. Rodríguez Cruz PM, Cossins J, Estephan EdP, Munell F, Selby K, iterative approach is highly effective in diagnosing individuals who Hirano M, Maroofin R, Mehrjardi MYV, Chow G, Carr A, Manzur A, are exome negative. Genet Med 2019, 21:161e172 The Journal of Molecular Diagnostics - jmdjournal.org 541 Bullich et al 68. Al-Nabhani M, Al-Rashdi S, Al-Murshedi F, Al-Kindi A, Al- Guan Q, Dulik MC, Conlin LK, Abou Tayoun AN, Luo M, Wu C, Thihli K, Al-Saegh A, Al-Futaisi A, Al-Mamari W, Zadjali F, Al- Cao K, Sarmady M, Bedoukian EC, Tarpinian J, Medne L, Maawali A: Reanalysis of exome sequencing data of intellectual Skraban CM, Deardorff MA, Krantz ID, Krock BL, Santani AB: disability samples: yields and benefits. Clin Genet 2018, 94:495e501 Automated clinical exome reanalysis reveals novel diagnoses. J Mol 69. Lee H, Deignan JL, Dorrani N, Strom SP, Kantarci S, Quintero- Diagn 2019, 21:38e48 Rivera F, Das K, Toy T, Harry B, Yourshaw M, Fox M, Fogel BL, 74. Jalkh N, Corbani S, Haidar Z, Hamdan N, Farah E, Abou Ghoch J, Martinez-Agosto JA, Wong DA, Chang VY, Shieh PB, Palmer CGS, Ghosn R, Salem N, Fawaz A, Djambas Khayat C, Rajab M, Dipple KM, Grody WW, Vilain E, Nelson SF: Clinical exome Mourani C, Moukarzel A, Rassi S, Gerbaka B, Mansour H, Baassiri M, sequencing for genetic identification of rare Mendelian disorders. Dagher R, Breich D, Mégarbané A, Desvignes JP, Delague V, JAMA 2014, 312:1880e1887 Mehawej C, Chouery E: The added value of WES reanalysis in the 70. Retterer K, Juusola J, Cho MT, Vitazka P, Millan F, Gibellini F, field of genetic diagnosis: lessons learned from 200 exomes in the Vertino-Bell A, Smaoui N, Neidich J, Monaghan KG, McKnight D, Lebanese population. BMC Med Genomics 2019, 12:11 Bai R, Suchy S, Friedman B, Tahiliani J, Pineda-Alvarez D, 75. Eldomery MK, Coban-Akdemir Z, Harel T, Rosenfeld JA, Richard G, Brandt T, Haverfield E, Chung WK, Bale S: Clinical Gambin T, Stray-Pedersen A, et al: Lessons learned from additional application of whole-exome sequencing across clinical indications. research analyses of unsolved clinical exome cases. Genome Med Genet Med 2016, 18:696e704 2017, 9:26 71. Nambot S, Thevenon J, Kuentz P, Duffourd Y, Tisserant E, Bruel A- 76. Alfares A, Aloraini T, Subaie LA, Alissa A, Qudsi AA, Alahmad A, L, Mosca-Boidron A-L, Masurel-Paulet A, Lehalle D, Jean- Mutairi FA, Alswaid A, Alothaim A, Eyaid W, Albalwi M, Alturki S, Marçais N, Lefebvre M, Vabres P, El Chehadeh-Djebbar S, Alfadhel M: Whole-genome sequencing offers additional but limited Philippe C, Tran Mau-Them F, St-Onge J, Jouan T, Chevarin M, clinical utility compared with reanalysis of whole-exome sequencing. Poé C, Carmignac V, Vitobello A, Callier P, Rivière J-B, Faivre L, Genet Med 2018, 20:1328e1333 Thauvin-Robinet C, Orphanomix Physicians’ Group: Clinical whole- 77. Sarmady M, Abou Tayoun A: Need for automated interactive exome sequencing for the diagnosis of rare disorders with congenital genomic interpretation and ongoing reanalysis. JAMA Pediatr 2018, anomalies and/or intellectual disability: substantial interest of pro- 172:1113e1114 spective annual reanalysis. Genet Med 2018, 20:645e654 78. Deignan JL, Chung WK, Kearney HM, Monaghan KG, Rehder CW, 72. Tan TY, Lunke S, Chong B, Phelan D, Fanjul-Fernandez M, Chao EC, ACMG Laboratory Quality Assurance Committee: Points Marum JE, Kumar VS, Stark Z, Yeung A, Brown NJ, Stutterd C, to consider in the reevaluation and reanalysis of genomic test results: Delatycki MB, Sadedin S, Martyn M, Goranitis I, Thorne N, Gaff CL, a statement of the American College of Medical Genetics and Ge- White SM: A head-to-head evaluation of the diagnostic efficacy and nomics (ACMG). Genet Med 2019, 21:1267e1270 costs of trio versus singleton exome sequencing analysis. Eur J Hum 79. Boycott KM, Hartley T, Biesecker LG, Gibbs RA, Innes AM, Riess O, Genet 2019, 27:1791e1799 Belmont J, Dunwoodie SL, Jojic N, Lassmann T, Mackay D, 73. Baker SW, Murrell JR, Nesbitt AI, Pechter KB, Balciuniene J, Temple IK, Visel A, Baynam G: A diagnosis for all rare genetic dis- Zhao X, Yu Z, Denenberg EH, DeChene ET, Wilkens AB, Bhoj EJ, eases: the horizon and the next frontiers. Cell 2019, 177:32e37 542 jmdjournal.org - The Journal of Molecular Diagnostics