CNVoyant: A Highly Performant and Explainable Multi-Classifier Machine Learning Approach for Determining the Clinical Significance of Copy Number Variants.

The precise classification of copy number variants (CNVs) presents a significant challenge in genomic medicine, primarily due to the complex nature of CNVs and their diverse impact on genetic disorders. This complexity is compounded by the limitations of existing methods in accurately distinguishing between benign, uncertain, and pathogenic CNVs. Addressing this gap, we introduce CNVoyant, a machine learning-based multi-class framework designed to enhance the clinical significance classification of CNVs. Trained on a comprehensive dataset of 52,176 ClinVar entries across pathogenic, uncertain, and benign classifications, CNVoyant incorporates a broad spectrum of genomic features, including genome position, disease-gene annotations, dosage sensitivity, and conservation scores. Models to predict the clinical significance of copy number gains and losses were trained independently. Final models were selected after testing 29 machine learning architectures and 10,000 hyperparameter combinations each for deletions and duplications via 5-fold cross-validation. We validate the performance of the CNVoyant by leveraging a comprehensive set of 21,574 CNVs from the DECIPHER database, a highly regarded resource known for its extensive catalog of chromosomal imbalances linked to clinical outcomes. Compared to alternative approaches, CNVoyant shows marked improvements in precision-recall and ROC AUC metrics for binary pathogenic classifications while going one step further, offering multi-classification of clinical significance and corresponding SHAP explainability plots. This large-scale validation demonstrates CNVoyant's superior accuracy and underscores its potential to aid genomic researchers and clinical geneticists in interpreting the clinical implications of real CNVs.

Provision and availability of genomic medicine services in Level IV neonatal intensive care units.

PURPOSE: To describe variation in genomic medicine services across level IV neonatal intensive care units (NICUs) in the United States and Canada. METHODS: We developed and distributed a novel survey to the 43 level IV NICUs belonging to the Children's Hospitals Neonatal Consortium, requesting a single response per site from a clinician with knowledge of the provision of genomic medicine services. RESULTS: Overall response rate was 74% (32/43). Although chromosomal microarray and exome or genome sequencing (ES or GS) were universally available, access was restricted for 22% (7/32) and 81% (26/32) of centers, respectively. The most common restriction on ES or GS was requiring approval by a specialist (41%, 13/32). Rapid ES/GS was available in 69% of NICUs (22/32). Availability of same-day genetics consultative services was limited (41%, 13/32 sites), and pre- and post-test counseling practices varied widely. CONCLUSION: We observed large inter-center variation in genomic medicine services across level IV NICUs: most notably, access to rapid, comprehensive genetic testing in time frames relevant to critical care decision making was limited at many level IV Children's Hospitals Neonatal Consortium NICUs despite a significant burden of genetic disease. Further efforts are needed to improve access to neonatal genomic medicine services.

A Decade's Experience in Pediatric Chromosomal Microarray Reveals Distinct Characteristics Across Ordering Specialties.

Chromosomal microarray (CMA) is a testing modality frequently used in pediatric patients; however, published data on its utilization are limited to the genetic setting. We performed a database search for all CMA testing performed from 2010 to 2020, and delineated the diagnostic yield based on patient characteristics, including sex, age, clinical specialty of providers, indication of testing, and pathogenic finding. The indications for testing were further categorized into Human Phenotype Ontology categories for analysis. This study included a cohort of 14,541 patients from 29 different medical specialties, of whom 30% were from the genetics clinic. The clinical indications for testing suggested that neonatology patients demonstrated the greatest involvement of multiorgan systems, involving the most Human Phenotype Ontology categories, compared with developmental behavioral pediatrics and neurology patients being the least. The top pathogenic findings for each specialty differed, likely due to the varying clinical features and indications for testing. Deletions involving the 22q11.21 locus were the top pathogenic findings for patients presenting to genetics, neonatology, cardiology, and surgery. Our data represent the largest pediatric cohort published to date. This study is the first to demonstrate the diagnostic utility of this assay for patients seen in the setting of different specialties, and it provides normative data of CMA results among a general pediatric population referred for testing because of variable clinical presentations.