Rapid and accurate interpretation of clinical exomes using Phenoxome: a computational phenotype-driven approach

Phenotype-aware prioritisation of rare Mendelian disease variants

A molecular diagnosis from the analysis of sequencing data in rare Mendelian diseases has a huge impact on the management of patients and their families. Numerous patient phenotype-aware variant prioritisation (VP) tools have been developed to help automate this process, and shorten the diagnostic odyssey, but performance statistics on real patient data are limited. Here we identify, assess, and compare the performance of all up-to-date, freely available, and programmatically accessible tools using a whole-exome, retinal disease dataset from 134 individuals with a molecular diagnosis. All tools were able to identify around two-thirds of the genetic diagnoses as the top-ranked candidate, with LIRICAL performing best overall. Finally, we discuss the challenges to overcome most cases remaining undiagnosed after current, state-of-the-art practices.

Phenotype-driven approaches to enhance variant prioritization and diagnosis of rare disease

Rare disease diagnostics and disease gene discovery have been revolutionized by whole-exome and genome sequencing but identifying the causative variant(s) from the millions in each individual remains challenging. The use of deep phenotyping of patients and reference genotype-phenotype knowledge, alongside variant data such as allele frequency, segregation, and predicted pathogenicity, has proved an effective strategy to tackle this issue. Here we review the numerous tools that have been developed to automate this approach and demonstrate the power of such an approach on several thousand diagnosed cases from the 100,000 Genomes Project. Finally, we discuss the challenges that need to be overcome if we are going to improve detection rates and help the majority of patients that still remain without a molecular diagnosis after state-of-the-art genomic interpretation.

Genomic answers for children: Dynamic analyses of >1000 pediatric rare disease genomes

PURPOSE

This study aimed to provide comprehensive diagnostic and candidate analyses in a pediatric rare disease cohort through the Genomic Answers for Kids program.

METHODS

Extensive analyses of 960 families with suspected genetic disorders included short-read exome sequencing and short-read genome sequencing (srGS); PacBio HiFi long-read genome sequencing (HiFi-GS); variant calling for single nucleotide variants (SNV), structural variant (SV), and repeat variants; and machine-learning variant prioritization. Structured phenotypes, prioritized variants, and pedigrees were stored in PhenoTips database, with data sharing through controlled access the database of Genotypes and Phenotypes.

RESULTS

Diagnostic rates ranged from 11% in patients with prior negative genetic testing to 34.5% in naive patients. Incorporating SVs from genome sequencing added up to 13% of new diagnoses in previously unsolved cases. HiFi-GS yielded increased discovery rate with >4-fold more rare coding SVs compared with srGS. Variants and genes of unknown significance remain the most common finding (58% of nondiagnostic cases).

CONCLUSION

Computational prioritization is efficient for diagnostic SNVs. Thorough identification of non-SNVs remains challenging and is partly mitigated using HiFi-GS sequencing. Importantly, community research is supported by sharing real-time data to accelerate gene validation and by providing HiFi variant (SNV/SV) resources from >1000 human alleles to facilitate implementation of new sequencing platforms for rare disease diagnoses.

Molecular Diagnostic Outcomes from 700 Cases: What Can We Learn from a Retrospective Analysis of Clinical Exome Sequencing?

Clinical exome sequencing (CES) aids in the diagnosis of rare genetic disorders. Herein, we report the molecular diagnostic yield and spectrum of genetic alterations contributing to disease in 700 pediatric cases analyzed at the Children's Hospital of Philadelphia. The overall diagnostic yield was 23%, with three cases having more than one molecular diagnosis and 2.6% having secondary/additional findings. A candidate gene finding was reported in another 8.4% of cases. The clinical indications with the highest diagnostic yield were neurodevelopmental disorders (including seizures), whereas immune- and oncology-related indications were negatively associated with molecular diagnosis. The rapid expansion of knowledge regarding the genome's role in human disease necessitates reanalysis of CES samples. To capture these new discoveries, a subset of cases (n = 240) underwent reanalysis, with an increase in diagnostic yield. We describe our experience reporting CES results in a pediatric setting, including reporting of secondary findings, reporting newly discovered genetic conditions, and revisiting negative test results. Finally, we highlight the challenges associated with implementing critical updates to the CES workflow. Although these updates are necessary, they demand an investment of time and resources from the laboratory. In summary, these data demonstrate the clinical utility of exome sequencing and reanalysis, while highlighting the critical considerations for continuous improvement of a CES test in a clinical laboratory.

Evaluation of phenotype-driven gene prioritization methods for Mendelian diseases

It’s challenging work to identify disease-causing genes from the next-generation sequencing (NGS) data of patients with Mendelian disorders. To improve this situation, researchers have developed many phenotype-driven gene prioritization methods using a patient’s genotype and phenotype information, or phenotype information only as input to rank the candidate’s pathogenic genes. Evaluations of these ranking methods provide practitioners with convenience for choosing an appropriate tool for their workflows, but retrospective benchmarks are underpowered to provide statistically significant results in their attempt to differentiate. In this research, the performance of ten recognized causal-gene prioritization methods was benchmarked using 305 cases from the Deciphering Developmental Disorders (DDD) project and 209 in-house cases via a relatively unbiased methodology. The evaluation results show that methods using Human Phenotype Ontology (HPO) terms and Variant Call Format (VCF) files as input achieved better overall performance than those using phenotypic data alone. Besides, LIRICAL and AMELIE, two of the best methods in our benchmark experiments, complement each other in cases with the causal genes ranked highly, suggesting a possible integrative approach to further enhance the diagnostic efficiency. Our benchmarking provides valuable reference information to the computer-assisted rapid diagnosis in Mendelian diseases and sheds some light on the potential direction of future improvement on disease-causing gene prioritization methods.

Artificial intelligence (AI)-assisted exome reanalysis greatly aids in the identification of new positive cases and reduces analysis time in a clinical diagnostic laboratory

PURPOSE

Artificial intelligence (AI) and variant prioritization tools for genomic variant analysis are being rapidly developed for use in clinical diagnostic testing. However, their clinical utility and reliability are currently limited. Therefore, we performed a validation of a commercial AI tool (Moon) and a comprehensive reanalysis of previously collected clinical exome sequencing cases using an open-source variant prioritization tool (Exomiser) and the now-validated AI tool to test their feasibility in clinical diagnostics.

METHODS

A validation study of Moon was performed with 29 positive cases determined by previous manual analysis. After validation, reanalysis was performed on 80 previously manually analyzed nondiagnostic exome cases using Moon. Finally, a comparison between Moon and Exomiser was completed regarding their ability to identify previously completed positive cases and to identify new positive cases.

RESULTS

Moon correctly selected the causal variant(s) in 97% of manually analyzed positive cases and identified 7 new positive cases. Exomiser correctly identified the causal gene in 85% of positive cases and agreed with Moon by ranking the new gene in its top 10 list 43% of the time.

CONCLUSION

The use of AI in diagnostic laboratories greatly enhances exome sequencing analysis by reducing analysis time and increasing the diagnostic rate.

Powerful use of automated prioritization of candidate variants in genetic hearing loss with extreme etiologic heterogeneity

Variant prioritization of exome sequencing (ES) data for molecular diagnosis of sensorineural hearing loss (SNHL) with extreme etiologic heterogeneity poses a significant challenge. This study used an automated variant prioritization system (“EVIDENCE”) to analyze SNHL patient data and assess its diagnostic accuracy. We performed ES of 263 probands manifesting mild to moderate or higher degrees of SNHL. Candidate variants were classified according to the 2015 American College of Medical Genetics guidelines, and we compared the accuracy, call rates, and efficiency of variant prioritizations performed manually by humans or using EVIDENCE. In our in silico panel, 21 synthetic cases were successfully analyzed by EVIDENCE. In our cohort, the ES diagnostic yield for SNHL by manual analysis was 50.19% (132/263) and 50.95% (134/263) by EVIDENCE. EVIDENCE processed ES data 24-fold faster than humans, and the concordant call rate between humans and EVIDENCE was 97.72% (257/263). Additionally, EVIDENCE outperformed human accuracy, especially at discovering causative variants of rare syndromic deafness, whereas flexible interpretations that required predefined specific genotype–phenotype correlations were possible only by manual prioritization. The automated variant prioritization system remarkably facilitated the molecular diagnosis of hearing loss with high accuracy and efficiency, fostering the popularization of molecular genetic diagnosis of SNHL.

Personalised virtual gene panels reduce interpretation workload and maintain diagnostic rates of proband-only clinical exome sequencing for rare disorders

Purpose

The increased adoption of genomic strategies in the clinic makes it imperative for diagnostic laboratories to improve the efficiency of variant interpretation. Clinical exome sequencing (CES) is becoming a valuable diagnostic tool, capable of meeting the diagnostic demand imposed by the vast array of different rare monogenic disorders. We have assessed a clinician-led and phenotype-based approach for virtual gene panel generation for analysis of targeted CES in patients with rare disease in a single institution.

Methods

Retrospective survey of 400 consecutive cases presumed by clinicians to have rare monogenic disorders, referred on singleton basis for targeted CES. We evaluated diagnostic yield and variant workload to characterise the usefulness of a clinician-led approach for generation of virtual gene panels that can incorporate up to three different phenotype-driven gene selection methods.

Results

Abnormalities of the nervous system (54.5%), including intellectual disability, head and neck (19%), skeletal system (16%), ear (15%) and eye (15%) were the most common clinical features reported in referrals. Combined phenotype-driven strategies for virtual gene panel generation were used in 57% of cases. On average, 7.3 variants (median=5) per case were retained for clinical interpretation. The overall diagnostic rate of proband-only CES using personalised phenotype-driven virtual gene panels was 24%.

Conclusions

Our results show that personalised virtual gene panels are a cost-effective approach for variant analysis of CES, maintaining diagnostic yield and optimising the use of resources for clinical genomic sequencing in the clinic.

JARID2 haploinsufficiency is associated with a clinically distinct neurodevelopmental syndrome

PURPOSE

JARID2, located on chromosome 6p22.3, is a regulator of histone methyltransferase complexes that is expressed in human neurons. So far, 13 individuals sharing clinical features including intellectual disability (ID) were reported with de novo heterozygous deletions in 6p22-p24 encompassing the full length JARID2 gene (OMIM 601594). However, all published individuals to date have a deletion of at least one other adjoining gene, making it difficult to determine if JARID2 is the critical gene responsible for the shared features. We aim to confirm JARID2 as a human disease gene and further elucidate the associated clinical phenotype.

METHODS

Chromosome microarray analysis, exome sequencing, and an online matching platform (GeneMatcher) were used to identify individuals with single-nucleotide variants or deletions involving JARID2.

RESULTS

We report 16 individuals in 15 families with a deletion or single-nucleotide variant in JARID2. Several of these variants are likely to result in haploinsufficiency due to nonsense-mediated messenger RNA (mRNA) decay. All individuals have developmental delay and/or ID and share some overlapping clinical characteristics such as facial features with those who have larger deletions involving JARID2.

CONCLUSION

We report that JARID2 haploinsufficiency leads to a clinically distinct neurodevelopmental syndrome, thus establishing gene-disease validity for the purpose of diagnostic reporting.

Using Machine Learning to Identify True Somatic Variants from Next-Generation Sequencing

BACKGROUND

Molecular profiling has become essential for tumor risk stratification and treatment selection. However, cancer genome complexity and technical artifacts make identification of real variants a challenge. Currently, clinical laboratories rely on manual screening, which is costly, subjective, and not scalable. We present a machine learning-based method to distinguish artifacts from bona fide single-nucleotide variants (SNVs) detected by next-generation sequencing from nonformalin-fixed paraffin-embedded tumor specimens.

METHODS

A cohort of 11278 SNVs identified through clinical sequencing of tumor specimens was collected and divided into training, validation, and test sets. Each SNV was manually inspected and labeled as either real or artifact as part of clinical laboratory workflow. A 3-class (real, artifact, and uncertain) model was developed on the training set, fine-tuned with the validation set, and then evaluated on the test set. Prediction intervals reflecting the certainty of the classifications were derived during the process to label "uncertain" variants.

RESULTS

The optimized classifier demonstrated 100% specificity and 97% sensitivity over 5587 SNVs of the test set. Overall, 1252 of 1341 true-positive variants were identified as real, 4143 of 4246 false-positive calls were deemed artifacts, whereas only 192 (3.4%) SNVs were labeled as "uncertain," with zero misclassification between the true positives and artifacts in the test set.

CONCLUSIONS

We presented a computational classifier to identify variant artifacts detected from tumor sequencing. Overall, 96.6% of the SNVs received definitive labels and thus were exempt from manual review. This framework could improve quality and efficiency of the variant review process in clinical laboratories.

Phen2Gene: rapid phenotype-driven gene prioritization for rare diseases

Human Phenotype Ontology (HPO) terms are increasingly used in diagnostic settings to aid in the characterization of patient phenotypes. The HPO annotation database is updated frequently and can provide detailed phenotype knowledge on various human diseases, and many HPO terms are now mapped to candidate causal genes with binary relationships. To further improve the genetic diagnosis of rare diseases, we incorporated these HPO annotations, gene-disease databases and gene-gene databases in a probabilistic model to build a novel HPO-driven gene prioritization tool, Phen2Gene. Phen2Gene accesses a database built upon this information called the HPO2Gene Knowledgebase (H2GKB), which provides weighted and ranked gene lists for every HPO term. Phen2Gene is then able to access the H2GKB for patient-specific lists of HPO terms or PhenoPacket descriptions supported by GA4GH (http://phenopackets.org/), calculate a prioritized gene list based on a probabilistic model and output gene-disease relationships with great accuracy. Phen2Gene outperforms existing gene prioritization tools in speed and acts as a real-time phenotype-driven gene prioritization tool to aid the clinical diagnosis of rare undiagnosed diseases. In addition to a command line tool released under the MIT license (https://github.com/WGLab/Phen2Gene), we also developed a web server and web service (https://phen2gene.wglab.org/) for running the tool via web interface or RESTful API queries. Finally, we have curated a large amount of benchmarking data for phenotype-to-gene tools involving 197 patients across 76 scientific articles and 85 patients' de-identified HPO term data from the Children's Hospital of Philadelphia.

An Improved Phenotype-Driven Tool for Rare Mendelian Variant Prioritization: Benchmarking Exomiser on Real Patient Whole-Exome Data

Next-generation sequencing has revolutionized rare disease diagnostics, but many patients remain without a molecular diagnosis, particularly because many candidate variants usually survive despite strict filtering. Exomiser was launched in 2014 as a Java tool that performs an integrative analysis of patients' sequencing data and their phenotypes encoded with Human Phenotype Ontology (HPO) terms. It prioritizes variants by leveraging information on variant frequency, predicted pathogenicity, and gene-phenotype associations derived from human diseases, model organisms, and protein-protein interactions. Early published releases of Exomiser were able to prioritize disease-causative variants as top candidates in up to 97% of simulated whole-exomes. The size of the tested real patient datasets published so far are very limited. Here, we present the latest Exomiser version 12.0.1 with many new features. We assessed the performance using a set of 134 whole-exomes from patients with a range of rare retinal diseases and known molecular diagnosis. Using default settings, Exomiser ranked the correct diagnosed variants as the top candidate in 74% of the dataset and top 5 in 94%; not using the patients' HPO profiles (i.e., variant-only analysis) decreased the performance to 3% and 27%, respectively. In conclusion, Exomiser is an effective support tool for rare Mendelian phenotype-driven variant prioritization.