Beyond the exome: What's next in diagnostic testing for Mendelian conditions.

Despite advances in clinical genetic testing, including the introduction of exome sequencing (ES), more than 50% of individuals with a suspected Mendelian condition lack a precise molecular diagnosis. Clinical evaluation is increasingly undertaken by specialists outside of clinical genetics, often occurring in a tiered fashion and typically ending after ES. The current diagnostic rate reflects multiple factors, including technical limitations, incomplete understanding of variant pathogenicity, missing genotype-phenotype associations, complex gene-environment interactions, and reporting differences between clinical labs. Maintaining a clear understanding of the rapidly evolving landscape of diagnostic tests beyond ES, and their limitations, presents a challenge for non-genetics professionals. Newer tests, such as short-read genome or RNA sequencing, can be challenging to order, and emerging technologies, such as optical genome mapping and long-read DNA sequencing, are not available clinically. Furthermore, there is no clear guidance on the next best steps after inconclusive evaluation. Here, we review why a clinical genetic evaluation may be negative, discuss questions to be asked in this setting, and provide a framework for further investigation, including the advantages and disadvantages of new approaches that are nascent in the clinical sphere. We present a guide for the next best steps after inconclusive molecular testing based upon phenotype and prior evaluation, including when to consider referral to research consortia focused on elucidating the underlying cause of rare unsolved genetic disorders.

Exome and genome sequencing in a heterogeneous population of patients with rare disease: Identifying predictors of a diagnosis.

PURPOSE: Exome (ES) and genome sequencing (GS) are increasingly being utilized for individuals with rare and undiagnosed diseases; however, guidelines on their use remain limited. This study aimed to identify factors associated with diagnosis by ES and/or GS in a heterogeneous population of patients with rare and undiagnosed diseases. METHODS: In this case control study, we reviewed data from 400 diagnosed and 400 undiagnosed randomly selected participants in the Undiagnosed Diseases Network, all of whom had undergone ES and/or GS. We analyzed factors associated with receiving a diagnosis by ES and/or GS. RESULTS: Factors associated with a decreased odds of being diagnosed included adult symptom onset, singleton sequencing, and having undergone ES and/or GS before acceptance to the Undiagnosed Diseases Network (48%, 51%, and 32% lower odds, respectively). Factors that increased the odds of being diagnosed by ES and/or GS included having primarily neurological symptoms and having undergone prior chromosomal microarray testing (44% and 59% higher odds, respectively). CONCLUSION: We identified several factors that were associated with receiving a diagnosis by ES and/or GS. This will ideally inform the utilization of ES and/or GS and help manage expectations of individuals and families undergoing these tests.

Loss of function of FAM177A1, a Golgi complex localized protein, causes a novel neurodevelopmental disorder.

PURPOSE: The function of FAM177A1 and its relationship to human disease is largely unknown. Recent studies have demonstrated FAM177A1 to be a critical immune-associated gene. One previous case study has linked FAM177A1 to a neurodevelopmental disorder in four siblings. METHODS: We identified five individuals from three unrelated families with biallelic variants in FAM177A1. The physiological function of FAM177A1 was studied in a zebrafish model organism and human cell lines with loss-of-function variants similar to the affected cohort. RESULTS: These individuals share a characteristic phenotype defined by macrocephaly, global developmental delay, intellectual disability, seizures, behavioral abnormalities, hypotonia, and gait disturbance. We show that FAM177A1 localizes to the Golgi complex in mammalian and zebrafish cells. Intersection of the RNA-seq and metabolomic datasets from FAM177A1-deficient human fibroblasts and whole zebrafish larvae demonstrated dysregulation of pathways associated with apoptosis, inflammation, and negative regulation of cell proliferation. CONCLUSION: Our data sheds light on the emerging function of FAM177A1 and defines FAM177A1-related neurodevelopmental disorder as a new clinical entity.

Emery-Dreifuss muscular dystrophy Type 1 is associated with a high risk of malignant ventricular arrhythmias and end-stage heart failure.

BACKGROUND AND AIMS: Emery-Dreifuss muscular dystrophy (EDMD) is caused by variants in EMD (EDMD1) and LMNA (EDMD2). Cardiac conduction defects and atrial arrhythmia are common to both, but LMNA variants also cause end-stage heart failure (ESHF) and malignant ventricular arrhythmia (MVA). This study aimed to better characterize the cardiac complications of EMD variants. METHODS: Consecutively referred EMD variant-carriers were retrospectively recruited from 12 international cardiomyopathy units. MVA and ESHF incidences in male and female variant-carriers were determined. Male EMD variant-carriers with a cardiac phenotype at baseline (EMDCARDIAC) were compared with consecutively recruited male LMNA variant-carriers with a cardiac phenotype at baseline (LMNACARDIAC). RESULTS: Longitudinal follow-up data were available for 38 male and 21 female EMD variant-carriers [mean (SD) ages 33.4 (13.3) and 43.3 (16.8) years, respectively]. Nine (23.7%) males developed MVA and five (13.2%) developed ESHF during a median (inter-quartile range) follow-up of 65.0 (24.3-109.5) months. No female EMD variant-carrier had MVA or ESHF, but nine (42.8%) developed a cardiac phenotype at a median (inter-quartile range) age of 58.6 (53.2-60.4) years. Incidence rates for MVA were similar for EMDCARDIAC and LMNACARDIAC (4.8 and 6.6 per 100 person-years, respectively; log-rank P = .49). Incidence rates for ESHF were 2.4 and 5.9 per 100 person-years for EMDCARDIAC and LMNACARDIAC, respectively (log-rank P = .09). CONCLUSIONS: Male EMD variant-carriers have a risk of progressive heart failure and ventricular arrhythmias similar to that of male LMNA variant-carriers. Early implantable cardioverter defibrillator implantation and heart failure drug therapy should be considered in male EMD variant-carriers with cardiac disease.

De novo EIF2AK1 and EIF2AK2 Variants Are Associated with Developmental Delay, Leukoencephalopathy, and Neurologic Decompensation.

EIF2AK1 and EIF2AK2 encode members of the eukaryotic translation initiation factor 2 alpha kinase (EIF2AK) family that inhibits protein synthesis in response to physiologic stress conditions. EIF2AK2 is also involved in innate immune response and the regulation of signal transduction, apoptosis, cell proliferation, and differentiation. Despite these findings, human disorders associated with deleterious variants in EIF2AK1 and EIF2AK2 have not been reported. Here, we describe the identification of nine unrelated individuals with heterozygous de novo missense variants in EIF2AK1 (1/9) or EIF2AK2 (8/9). Features seen in these nine individuals include white matter alterations (9/9), developmental delay (9/9), impaired language (9/9), cognitive impairment (8/9), ataxia (6/9), dysarthria in probands with verbal ability (6/9), hypotonia (7/9), hypertonia (6/9), and involuntary movements (3/9). Individuals with EIF2AK2 variants also exhibit neurological regression in the setting of febrile illness or infection. We use mammalian cell lines and proband-derived fibroblasts to further confirm the pathogenicity of variants in these genes and found reduced kinase activity. EIF2AKs phosphorylate eukaryotic translation initiation factor 2 subunit 1 (EIF2S1, also known as EIF2α), which then inhibits EIF2B activity. Deleterious variants in genes encoding EIF2B proteins cause childhood ataxia with central nervous system hypomyelination/vanishing white matter (CACH/VWM), a leukodystrophy characterized by neurologic regression in the setting of febrile illness and other stressors. Our findings indicate that EIF2AK2 missense variants cause a neurodevelopmental syndrome that may share phenotypic and pathogenic mechanisms with CACH/VWM.

A concurrent dual analysis of genomic data augments diagnoses: Experiences of 2 clinical sites in the Undiagnosed Diseases Network.

PURPOSE: Next-generation sequencing (NGS) has revolutionized the diagnostic process for rare/ultrarare conditions. However, diagnosis rates differ between analytical pipelines. In the National Institutes of Health-Undiagnosed Diseases Network (UDN) study, each individual's NGS data are concurrently analyzed by the UDN sequencing core laboratory and the clinical sites. We examined the outcomes of this practice. METHODS: A retrospective review was performed at 2 UDN clinical sites to compare the variants and diagnoses/candidate genes identified with the dual analyses of the NGS data. RESULTS: In total, 95 individuals had 100 diagnoses/candidate genes. There was 59% concordance between the UDN sequencing core laboratories and the clinical sites in identifying diagnoses/candidate genes. The core laboratory provided more diagnoses, whereas the clinical sites prioritized more research variants/candidate genes (P < .001). The clinical sites solely identified 15% of the diagnoses/candidate genes. The differences between the 2 pipelines were more often because of variant prioritization disparities than variant detection. CONCLUSION: The unique dual analysis of NGS data in the UDN synergistically enhances outcomes. The core laboratory provided a clinical analysis with more diagnoses and the clinical sites prioritized more research variants/candidate genes. Implementing such concurrent dual analyses in other genomic research studies and clinical settings can improve both variant detection and prioritization.

Arrhythmias including atrial fibrillation and congenital heart disease in Kleefstra syndrome: a possible epigenetic link.

AIMS: Kleefstra syndrome (KS), often diagnosed in early childhood, is a rare genetic disorder due to haploinsufficiency of EHMT1 and is characterized by neuromuscular and intellectual developmental abnormalities. Although congenital heart disease (CHD) is common, the prevalence of arrhythmias and CHD subtypes in KS is unknown. METHODS AND RESULTS: Inspired by a novel case series of KS patients with atrial tachyarrhythmias in the USA, we evaluate the two largest known KS registries for arrhythmias and CHD: Radboudumc (50 patients) based on health record review at Radboud University Medical Center in the Netherlands and GenIDA (163 patients) based on worldwide surveys of patient families. Three KS patients (aged 17-25 years) presented with atrial tachyarrhythmias without manifest CHD. In the international KS registries, the median [interquartile range (IQR)] age was considerably younger: GenIDA/Radboudumc at 10/13.5 (12/13) years, respectively. Both registries had a 40% prevalence of cardiovascular abnormalities, the majority being CHD, including septal defects, vascular malformations, and valvular disease. Interestingly, 4 (8%) patients in the Radboudumc registry reported arrhythmias without CHD, including one atrial fibrillation (AF), two with supraventricular tachycardias, and one with non-sustained ventricular tachycardia. The GenIDA registry reported one patient with AF and another with chronic ectopic atrial tachycardia (AT). In total, atrial tachyarrhythmias were noted in six young KS patients (6/213 or 3%) with at least four (three AF and one AT) without structural heart disease. CONCLUSION: In addition to a high prevalence of CHD, evolving data reveal early-onset atrial tachyarrhythmias in young KS patients, including AF, even in the absence of structural heart disease.

Phenotypic Expression, Natural History, and Risk Stratification of Cardiomyopathy Caused by Filamin C Truncating Variants.

BACKGROUND: Filamin C truncating variants (FLNCtv) cause a form of arrhythmogenic cardiomyopathy: the mode of presentation, natural history, and risk stratification of FLNCtv remain incompletely explored. We aimed to develop a risk profile for refractory heart failure and life-threatening arrhythmias in a multicenter cohort of FLNCtv carriers. METHODS: FLNCtv carriers were identified from 10 tertiary care centers for genetic cardiomyopathies. Clinical and outcome data were compiled. Composite outcomes were all-cause mortality/heart transplantation/left ventricle assist device (D/HT/LVAD), nonarrhythmic death/HT/LVAD, and sudden cardiac death/major ventricular arrhythmias. Previously established cohorts of 46 patients with LMNA and 60 with DSP-related arrhythmogenic cardiomyopathies were used for prognostic comparison. RESULTS: Eighty-five patients carrying FLNCtv were included (42±15 years, 53% men, 45% probands). Phenotypes were heterogeneous at presentation: 49% dilated cardiomyopathy, 25% arrhythmogenic left dominant cardiomyopathy, 3% arrhythmogenic right ventricular cardiomyopathy. Left ventricular ejection fraction was <50% in 64% of carriers and 34% had right ventricular fractional area changes (RVFAC=(right ventricular end-diastolic area - right ventricular end-systolic area)/right ventricular end-diastolic area) <35%. During follow-up (median time 61 months), 19 (22%) carriers experienced D/HT/LVAD, 13 (15%) experienced nonarrhythmic death/HT/LVAD, and 23 (27%) experienced sudden cardiac death/major ventricular arrhythmias. The sudden cardiac death/major ventricular arrhythmias incidence of FLNCtv carriers did not significantly differ from LMNA carriers and DSP carriers. In FLNCtv carriers, left ventricular ejection fraction was associated with the risk of D/HT/LVAD and nonarrhythmic death/HT/LVAD. CONCLUSIONS: Among patients referred to tertiary referral centers, FLNCtv arrhythmogenic cardiomyopathy is phenotypically heterogeneous and characterized by a high risk of life-threatening arrhythmias, which does not seem to be associated with the severity of left ventricular dysfunction.

Effect of Genetic Diagnosis on Patients with Previously Undiagnosed Disease.

BACKGROUND: Many patients remain without a diagnosis despite extensive medical evaluation. The Undiagnosed Diseases Network (UDN) was established to apply a multidisciplinary model in the evaluation of the most challenging cases and to identify the biologic characteristics of newly discovered diseases. The UDN, which is funded by the National Institutes of Health, was formed in 2014 as a network of seven clinical sites, two sequencing cores, and a coordinating center. Later, a central biorepository, a metabolomics core, and a model organisms screening center were added. METHODS: We evaluated patients who were referred to the UDN over a period of 20 months. The patients were required to have an undiagnosed condition despite thorough evaluation by a health care provider. We determined the rate of diagnosis among patients who subsequently had a complete evaluation, and we observed the effect of diagnosis on medical care. RESULTS: A total of 1519 patients (53% female) were referred to the UDN, of whom 601 (40%) were accepted for evaluation. Of the accepted patients, 192 (32%) had previously undergone exome sequencing. Symptoms were neurologic in 40% of the applicants, musculoskeletal in 10%, immunologic in 7%, gastrointestinal in 7%, and rheumatologic in 6%. Of the 382 patients who had a complete evaluation, 132 received a diagnosis, yielding a rate of diagnosis of 35%. A total of 15 diagnoses (11%) were made by clinical review alone, and 98 (74%) were made by exome or genome sequencing. Of the diagnoses, 21% led to recommendations regarding changes in therapy, 37% led to changes in diagnostic testing, and 36% led to variant-specific genetic counseling. We defined 31 new syndromes. CONCLUSIONS: The UDN established a diagnosis in 132 of the 382 patients who had a complete evaluation, yielding a rate of diagnosis of 35%. (Funded by the National Institutes of Health Common Fund.).

The genetic architecture of Plakophilin 2 cardiomyopathy.

PURPOSE: The genetic architecture of Plakophilin 2 (PKP2) cardiomyopathy can inform our understanding of its variant pathogenicity and protein function. METHODS: We assess the gene-wide and regional association of truncating and missense variants in PKP2 with arrhythmogenic cardiomyopathy (ACM), and arrhythmogenic right ventricular cardiomyopathy (ARVC) specifically. A discovery data set compares genetic testing requisitions to gnomAD. Validation is performed in a rigorously phenotyped definite ARVC cohort and non-ACM individuals in the Geisinger MyCode cohort. RESULTS: The etiologic fraction (EF) of ACM-related diagnoses from truncating variants in PKP2 is significant (0.85 [0.80,0.88], p < 2 × 10-16), increases for ARVC specifically (EF = 0.96 [0.94,0.97], p < 2 × 10-16), and is highest in definite ARVC versus non-ACM individuals (EF = 1.00 [1.00,1.00], p < 2 × 10-16). Regions of missense variation enriched for ACM probands include known functional domains and the C-terminus, which was not previously known to contain a functional domain. No regional enrichment was identified for truncating variants. CONCLUSION: This multicohort evaluation of the genetic architecture of PKP2 demonstrates the specificity of PKP2 truncating variants for ARVC within the ACM disease spectrum. We identify the PKP2 C-terminus as a potential functional domain and find that truncating variants likely cause disease irrespective of transcript position.

Clinical sites of the Undiagnosed Diseases Network: unique contributions to genomic medicine and science.

PURPOSE: The NIH Undiagnosed Diseases Network (UDN) evaluates participants with disorders that have defied diagnosis, applying personalized clinical and genomic evaluations and innovative research. The clinical sites of the UDN are essential to advancing the UDN mission; this study assesses their contributions relative to standard clinical practices. METHODS: We analyzed retrospective data from four UDN clinical sites, from July 2015 to September 2019, for diagnoses, new disease gene discoveries and the underlying investigative methods. RESULTS: Of 791 evaluated individuals, 231 received 240 diagnoses and 17 new disease-gene associations were recognized. Straightforward diagnoses on UDN exome and genome sequencing occurred in 35% (84/240). We considered these tractable in standard clinical practice, although genome sequencing is not yet widely available clinically. The majority (156/240, 65%) required additional UDN-driven investigations, including 90 diagnoses that occurred after prior nondiagnostic exome sequencing and 45 diagnoses (19%) that were nongenetic. The UDN-driven investigations included complementary/supplementary phenotyping, innovative analyses of genomic variants, and collaborative science for functional assays and animal modeling. CONCLUSION: Investigations driven by the clinical sites identified diagnostic and research paradigms that surpass standard diagnostic processes. The new diagnoses, disease gene discoveries, and delineation of novel disorders represent a model for genomic medicine and science.

"Doctors can read about it, they can know about it, but they've never lived with it": How parents use social media throughout the diagnostic odyssey.

Parents of children with undiagnosed conditions struggle to obtain information about how to treat and support their children. It can be particularly challenging to find communities and other parents who share their experiences and can provide emotional and informational support. This study sought to characterize how parents use social media, both throughout the diagnostic odyssey and post-diagnosis, to meet their informational, social, and emotional support needs. We conducted qualitative semi-structured interviews with 14 parents from the Stanford site of the Undiagnosed Diseases Network (UDN), including five whose children had received a diagnosis through study participation. Interview recordings were analyzed using inductive, team-based coding and thematic analysis based in grounded theory using Dedoose qualitative analysis software. Through this process, we identified four key themes related to social media use. First, parents struggled to find the "right" community, often seeking out groups of similar patients based on symptoms or similar conditions. Second, though they found much valuable information through social media about caring for their child, they also struggled to interpret the relevance of the information to their own child's condition. Third, the social support and access to other patients' and families' lived experiences were described as both highly valued and emotionally challenging, particularly in the case of poor outcomes for similar families. Finally, parents expressed the need to balance concerns about their child's privacy with the value of transparency and data sharing for diagnosis. Our results suggest that the needs and experiences of undiagnosed patients and families differ from those with diagnosed diseases and highlight the need for support in best utilizing social media resources at different stages of the diagnostic odyssey.

Understanding variants of uncertain significance in the era of multigene panels: Through the eyes of the patient.

Variants of uncertain significance (VUSs) are often disclosed to patients despite ambiguous association with disease risk and lack of clinical actionability. It is important to understand how patients understand a VUS result, but few studies have assessed this. Our qualitative study explored patient recall, reaction to, and interpretation of a VUS in the context of multigene panels. We conducted 11 semi-structured phone interviews with adults who had a VUS identified on multigene panel testing in a hereditary oncology clinic, with questions focusing on the VUS result, personal and family history, and motivations for and expectations of genetic testing. Transcripts were coded iteratively, using both deductive and inductive codes. Overall, participants usually recalled that they had a VUS, despite variation in the vocabulary used. Participants responded both emotionally and intellectually to receiving information about having a VUS, which was often a result of their expectations and motivations prior to testing. Overall, participants understood the lack of clinical significance of a VUS, yet often interpreted the etiologic significance of a VUS within the context of the personal and family history. Our study provides insight into a process by which patients translate uncertain genetic testing results into a construct that fits within their current belief framework and which may be facilitated by a genetic counselor.

A toolkit for genetics providers in follow-up of patients with non-diagnostic exome sequencing.

There are approximately 7,000 rare diseases affecting 25-30 million Americans, with 80% estimated to have a genetic basis. This presents a challenge for genetics practitioners to determine appropriate testing, make accurate diagnoses, and conduct up-to-date patient management. Exome sequencing (ES) is a comprehensive diagnostic approach, but only 25%-41% of the patients receive a molecular diagnosis. The remaining three-fifths to three-quarters of patients undergoing ES remain undiagnosed. The Stanford Center for Undiagnosed Diseases (CUD), a clinical site of the Undiagnosed Diseases Network, evaluates patients with undiagnosed and rare diseases using a combination of methods including ES. Frequently these patients have non-diagnostic ES results, but strategic follow-up techniques identify diagnoses in a subset. We present techniques used at the CUD that can be adopted by genetics providers in clinical follow-up of cases where ES is non-diagnostic. Solved case examples illustrate different types of non-diagnostic results and the additional techniques that led to a diagnosis. Frequent approaches include segregation analysis, data reanalysis, genome sequencing, additional variant identification, careful phenotype-disease correlation, confirmatory testing, and case matching. We also discuss prioritization of cases for additional analyses.

Developing a genomics rotation: Practical training around variant interpretation for genetic counseling students.

With the wide adoption of next-generation sequencing (NGS)-based genetic tests, genetic counselors require increased familiarity with NGS technology, variant interpretation concepts, and variant assessment tools. The use of exome and genome sequencing in clinical care has expanded the reach and diversity of genetic testing. Regardless of the setting where genetic counselors are performing variant interpretation or reporting, most of them have learned these skills from colleagues, while on the job. Though traditional, lecture-based learning around these topics is important, there has been growing need for the inclusion of case-based, experiential training of genomics and variant interpretation for genetic counseling students, with the goal of creating a strong foundation in variant interpretation for new genetic counselors, regardless of what area of practice they enter. To address this need, we established a genomics and variant interpretation rotation for Stanford's genetic counseling training program. In response to changes in the genomics landscape, this has now evolved into three unique rotation experiences, each focused on variant interpretation in the context of various genomic settings, including clinical laboratory, research laboratory, and healthy genomic analysis studies. Here, we describe the goals and learning objectives that we have developed for these variant interpretation rotations, and illustrate how these concepts are applied in practice.

Yield of whole exome sequencing in undiagnosed patients facing insurance coverage barriers to genetic testing.

BACKGROUND: Despite growing evidence of diagnostic yield and clinical utility of whole exome sequencing (WES) in patients with undiagnosed diseases, there remain significant cost and reimbursement barriers limiting access to such testing. The diagnostic yield and resulting clinical actions of WES for patients who previously faced insurance coverage barriers have not yet been explored. METHODS: We performed a retrospective descriptive analysis of clinical WES outcomes for patients facing insurance coverage barriers prior to clinical WES and who subsequently enrolled in the Undiagnosed Diseases Network (UDN). Clinical WES was completed as a result of participation in the UDN. Payer type, molecular diagnostic yield, and resulting clinical actions were evaluated. RESULTS: Sixty-six patients in the UDN faced insurance coverage barriers to WES at the time of enrollment (67% public payer, 26% private payer). Forty-two of 66 (64%) received insurance denial for clinician-ordered WES, 19/66 (29%) had health insurance through a payer known not to cover WES, and 5/66 (8%) had previous payer denial of other genetic tests. Clinical WES results yielded a molecular diagnosis in 23 of 66 patients (35% [78% pediatric, 65% neurologic indication]). Molecular diagnosis resulted in clinical actions in 14 of 23 patients (61%). CONCLUSIONS: These data demonstrate that a substantial proportion of patients who encountered insurance coverage barriers to WES had a clinically actionable molecular diagnosis, supporting the notion that WES has value as a covered benefit for patients who remain undiagnosed despite objective clinical findings.

Clinical Cardiovascular Genetic Counselors Take a Leading Role in Team-based Variant Classification.

We sought to delineate the genetic test review and interpretation practices of clinical cardiovascular genetic counselors. A one-time anonymous online survey was taken by 46 clinical cardiovascular genetic counselors recruited through the National Society of Genetic Counselors Cardiovascular Special Interest Group. Nearly all (95.7%) gather additional information on variants reported on clinical genetic test reports and most (81.4%) assess the classification of such variants. Clinical cardiovascular genetic counselors typically (81.0%) classify variants in collaboration with cardiologist and/or geneticist colleagues, with the genetic counselor as the team member who is primarily responsible. Variant classification is a relatively recent (mean 3.2 years) addition to practice. Most genetic counselors learned classification skills on the job from clinical and laboratory colleagues. Recent graduates were more likely to have learned this in graduate school (p < 0.001). Genetic counselors are motivated to take responsibility for the classification of variants because of prior experiences with variant reclassification, inconsistencies between laboratories, and incomplete laboratory reports. They are also driven by a sense of professional duty and their proximity to the clinical context. This practice represents a broadening of the skill set of clinical cardiovascular genetic counselors and a unique expertise that they contribute to the interdisciplinary teams in which they work.

RExPRT: a machine learning tool to predict pathogenicity of tandem repeat loci.

Expansions of tandem repeats (TRs) cause approximately 60 monogenic diseases. We expect that the discovery of additional pathogenic repeat expansions will narrow the diagnostic gap in many diseases. A growing number of TR expansions are being identified, and interpreting them is a challenge. We present RExPRT (Repeat EXpansion Pathogenicity pRediction Tool), a machine learning tool for distinguishing pathogenic from benign TR expansions. Our results demonstrate that an ensemble approach classifies TRs with an average precision of 93% and recall of 83%. RExPRT's high precision will be valuable in large-scale discovery studies, which require prioritization of candidate loci for follow-up studies.