SEARCH · Search PubMed

Results for “ClinVar”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 recordsLinked to original sources

PubMed2025

Imprecision medicine: Systematic gaps in reporting variants of uncertain significance (VUS) and their reclassifications.

PURPOSE: Variants of uncertain significance (VUS) are frequently encountered during clinical genetic testing. To explore the clinical burden of VUS, we developed the Brotman Baty Institute Clinical Variant Database, which is an electronic health record (EHR)-linked database of clinical germline genetic variant information from patients with rare genetic disorders seen at 2 tertiary academic medical centers. METHODS: We retrospectively reviewed EHRs and genetic testing reports from 5158 patients seen across diverse adult genetics practices at these institutions from 2015 to 2024. We also compared these EHR-based variant classifications with those in ClinVar. RESULTS: The number of reported VUS relative to pathogenic or likely pathogenic variants can vary by over 14-fold depending on the primary indication for genetic testing and 3-fold depending on self-reported race. Furthermore, at least 1.6% of variant classifications used in the EHR for clinical care are outdated based on ClinVar variant classifications, including 26 instances in which the testing lab updated ClinVar, but the reclassification was never communicated to the patient. CONCLUSION: Our findings reveal that the clinical burden of VUS in adult medical genetics is unequally distributed across patients. We also highlight a deficiency in existing systems for communicating variant reclassifications to ClinVar, patients, and providers.

Humans↗

PubMed2026

Reinforcement learning-based dynamic ensemble for missense variant effect prediction and tiered prioritization of VUS.

BACKGROUND: Accurate classification of missense variants remains a challenging task despite major advances in genomics. Numerous computational models have been developed to assist in variant classification, but often require repeated integration and benchmarking efforts. Ensemble methods have been proposed to overcome the limitations of single predictors, but mostly rely on fixed, predefined weights that constrain their ability to capture interactions among predictive signals. METHODS: We present GenixRL, a dynamic ensemble framework that reformulates model fusion as a reinforcement learning optimization problem. GenixRL uses a Q-learning agent to learn a policy that dynamically weights the probabilistic outputs of complementary predictors, including BayesDel (addAF and noAF), ClinPred, and MetaRNN. Replacing static weighting with policy learning allows GenixRL to adaptively identify optimal weightings and substantially improve classification accuracy. RESULTS: In benchmark evaluation against 25 state-of-the-art predictors, GenixRL achieved an AUROC of 0.9644 on an independent ClinVar dataset. On saturation genome editing assays for BRCA1 and BRCA2, GenixRL achieved the best performance and ranked highest on 14 of 17 clinically significant genes in a zero-shot evaluation. Applied to uncertain and conflicting ClinVar variants, GenixRL enabled tiered, evidence-based prioritization of hundreds of thousands of variants as likely pathogenic or pathogenic with high confidence, supported by orthogonal population evidence from gnomAD. CONCLUSION: GenixRL advances pathogenicity prediction for missense variants and provides an adaptive ensemble that sorts variants of uncertain significance into tiered candidates for expert curation and functional validation.

Mutation, Missense↗

PubMed2025

Tandem splice acceptor sites: Profiling their relevance to human disease.

PURPOSE: Interpretation of variation, particularly the creation or disruption of tandem splice acceptor sites (NAGNnAG variants), challenges genomic medicine practice. METHODS: We analyzed the creation and disruption of dinucleotide AG sites within ±30 bases of natural splice-acceptor sites in the GRCh37 human reference genome. These results were compared with variant data from the ClinVar and gnomAD databases, as well as with data from 779 National Institutes of Health Undiagnosed Diseases Program study participants. Using RNA sequencing, we assessed the splicing at NAGNnAG variants for 107 of the Undiagnosed Diseases Program participants and compared the empirical data with SpliceAI predictions. RESULTS: Creation or disruption of NAGNnAG sites within 30 bases of the natural splice acceptor are enriched in ClinVar compared with gnomAD; however, such variants in the 2 databases are rarely differentiated by SpliceAI scores. Empirical evaluation via RNA sequencing analysis supported novel acceptor site usage from -21 to +30; splice-altering variants did not predominate in a specific region or have SpliceAI scores invariantly, suggesting increased spliceogenicity. CONCLUSION: NAGNnAG variants within 30 bp of the natural splice acceptor have a high probability of clinical relevance and are poorly contextualized for clinical utility. Their interpretation benefits from empirical evaluation via RNA analysis.

Humans↗

PubMed2026

PubMind: literature-based genetic variant extraction and functional annotation using large language models.

Biomedical literature contains extensive functional knowledge on genetic variants, but much remains inaccessible in unstructured text. Existing resources such as ClinVar and HGMD remain limited by coverage, submission bias, update frequency, and sparse annotation. We develop PubMind, an artificial intelligence (AI) framework that uses large language models (LLMs) to triage and extract variant-function-disease associations and supporting evidence from biomedical text. PubMind captures single-nucleotide, copy-number, structural, and gene-fusion variants, and normalizes records to genomic and transcriptomic coordinates. Benchmarking shows >90% accuracy for variant recognition and 99% precision for disease extraction. Applied to >41 million PubMed abstracts and >5 million full-text articles, PubMind generates PubMind-DB, a database of ~1.3 million unique variants with contextual annotations, accessible via web interface and API. Only ~10% of PubMind variants overlap with ClinVar, and >80% of them show concordant pathogenicity labels. PubMind transforms unstructured biomedical text into structured genomic knowledge, advancing variant interpretation for precision medicine.

Large Language Models↗

PubMed2024

Rare variant analyses in 51,256 type 2 diabetes cases and 370,487 controls reveal the pathogenicity spectrum of monogenic diabetes genes.

Type 2 diabetes (T2D) genome-wide association studies (GWASs) often overlook rare variants as a result of previous imputation panels' limitations and scarce whole-genome sequencing (WGS) data. We used TOPMed imputation and WGS to conduct the largest T2D GWAS meta-analysis involving 51,256 cases of T2D and 370,487 controls, targeting variants with a minor allele frequency as low as 5 × 10-5. We identified 12 new variants, including a rare African/African American-enriched enhancer variant near the LEP gene (rs147287548), associated with fourfold increased T2D risk. We also identified a rare missense variant in HNF4A (p.Arg114Trp), associated with eightfold increased T2D risk, previously reported in maturity-onset diabetes of the young with reduced penetrance, but observed here in a T2D GWAS. We further leveraged these data to analyze 1,634 ClinVar variants in 22 genes related to monogenic diabetes, identifying two additional rare variants in HNF1A and GCK associated with fivefold and eightfold increased T2D risk, respectively, the effects of which were modified by the individual's polygenic risk score. For 21% of the variants with conflicting interpretations or uncertain significance in ClinVar, we provided support of being benign based on their lack of association with T2D. Our work provides a framework for using rare variant GWASs to identify large-effect variants and assess variant pathogenicity in monogenic diabetes genes.

Diabetes Mellitus, Type 2↗

PubMed2026

LDLR Variant Classification Through Activity-Normalized Prime Editing Screening.

BACKGROUND: Inherited variants in the LDL (low-density lipoprotein) receptor (LDLR) gene are the most common cause of familial hypercholesterolemia, significantly increasing coronary artery disease risk. Early identification of pathogenic LDLR variants enables prompt lipid-lowering therapy and cascade testing of at-risk relatives; however, most LDLR variants observed in the population have uncertain or absent clinical classifications, leaving many patients without actionable information. METHODS: We developed the first activity-normalized prime editing screening pipeline to measure the impact of 5184 LDLR coding variants on LDL-cholesterol (LDL-C) uptake. Each prime editing guide RNA is paired with a genotypic outcome reporter to correct for variable editing efficiency, overcoming a key limitation of previous pooled genome editing screens. A statistical framework further improves variant effect estimates by jointly analyzing all missense variants at each amino acid position. RESULTS: We show that prime editing of the reporter construct correlates with endogenous variant installation frequency, validating the activity normalization approach. The resulting scores capture a continuous spectrum of functional effects, robustly separate pathogenic versus benign ClinVar variants, and show concordance with LDL-C levels in UK Biobank participants. We calibrate functional evidence strengths to the ACMG/AMP variant interpretation framework, enabling integration into a clinical variant classification workflow. By combining functional, computational, population, and contextual evidence, 322 of 434 LDLR variants currently classified as variants of uncertain significance, conflicting, or absent from ClinVar appear to meet evidence thresholds for reclassification and can be prioritized for expert review, substantially expanding the pool of actionable variant classifications. The screen also reveals a cluster of gain-of-function variants in LDLR class A repeat 5, at least some of which enhance LDL-C uptake through increased apolipoprotein B interaction, with implications for therapeutic genome editing. Last, prime editing uniquely detects splice-altering coding variants missed by cDNA-based screens and pathogenicity predictors, revealing an advantage of endogenous variant installation. CONCLUSIONS: Altogether, activity-normalized prime editing provides a scalable framework for LDLR variant classification that substantially expands the proportion of variants with evidence for genetic diagnosis and reveals novel biology with therapeutic relevance.

CRISPR screening↗

PubMed2026

Distinct mutational landscapes for germline and somatic cancer variants in forty tumor suppressor genes.

Germline and somatic cancer variants in tumor suppressor genes (TSGs) share loss-of-function mechanisms, but studies of a few genes (DICER1 and CEBPA) have demonstrated differences in variant consequence and location. To systematically assess whether TSGs display distinct mutational patterns, we leveraged large public genetic databases and compared 32,941 high-quality pathogenic/likely pathogenic (P/LP) germline variants in ClinVar, with 12,907 oncogenic/likely oncogenic (O/LO) somatic tumor variants from cBioPortal across 40 TSGs. Only 3,863 (9.2%) variants were shared. Eighteen TSGs showed significantly different distributions of variant occurrences by molecular consequence, replicated with non-overlapping somatic data from the COSMIC database (chi-squared tests, false discovery rate = 5%). DICER1, TP53, and SMAD4 displayed excess somatic missense events, while nine TSGs (e.g., RB1 and APC) contained excess somatic stop-gain events throughout the coding sequence. Analysis by tumor type revealed excess stop-gain events in tissues exposed to environmental mutagens with corresponding mutation signatures. For several TSGs (WT1), germline variants predispose to tumors (Wilms' tumor) distinct from the majority source of somatic data (myeloid leukemia). Germline and somatic events are also distributed unevenly across cDNA locations, with 103 regions of preferential clustering in 39 TSGs (78 somatic and 25 germline). Twenty somatic clusters contained recurring frameshifts in homopolymer runs, many in tumors with microsatellite instability. Germline clusters contain more germline-exclusive variants, some driving non-cancer phenotypes reflecting genetic pleiotropy. Altogether, germline and somatic variants of TSGs represent unique sets with substantially different patterns shaped by selection pressures from gene-specific and somatic mutational mechanisms. Characterizing these distinctions enables more accurate clinical interpretation of TSG variants.

Humans↗

PubMed2026

Deep DNA and protein level feature integration for robust clinical variant interpretation using probabilistic gradient boosting.

A major challenge in clinical genomics is to classify genetic variations correctly, since it directly affects disease diagnosis and personal care. The existing methods tend to be based on the combination of different factors, such as protein structure, population frequencies, phenotypic annotations, and sequence conservation. Nevertheless, these methods often cannot be used to achieve the necessary interpretability, quantify uncertainty, and address rare cases. This paper presents a probabilistic gradient boosting model on variant pathogenicity prediction. The suggested framework applies biological characteristics at both level of DNA and protein levels while also scaling the level of uncertainty in clinical decision making. Our machine learning aims to solve the issues of variant interpretation by managing the features and through probability-based pathogenicity prediction. The framework formulation is aimed at generalizing over various datasets and minimizing overfitting. At the same time, it can ensure reasonable performance to facilitate clinical experiments. The model has also been tested on three standard datasets and demonstrated to be more predictive of the pathogenic effect of variants, in comparison with a variety of existing tools. The probabilistic gradient boosting model proposed had ROC AUC values of 0.9293, 0.9610, and 0.9646 on ClinVar variants, GRCh37, and GRCh38 human genome respectively. Furthermore, the dataset was ensured to include both exonic and intronic variants, and Variants of Uncertain Significance were also taken into consideration for Performance Testing. Through this it also aims to provide better clinical significance which will lead to a good interpretable tool for priority of variants for a large variety of disease conditions.

ClinVar↗

PubMed2026

A Novel BRCA1 Pathogenic Variant in Tunisian Patient With High Grade Ovarian Cancer: Favorable Therapeutic Response to Olaparib.

BACKGROUND: Ovarian cancer is one of the leading causes of death from gynecological cancer worldwide. Genetic mutations in genes involved in key cellular functions such as BRCA1/2 play a central role in tumorigenesis and have major implications for targeted therapeutic strategies, especially the use of poly (ADP-ribose) polymerase (PARP) inhibitors. CASE: Herein, we described a case of a 50-year-old woman diagnosed with severe anemia secondary to heavy menometrorrhagia. Initial gynecological evaluation, including transvaginal ultrasound, was unremarkable, and endometrial biopsy was not indicated. Imaging revealed no ovarian abnormalities; however, exploratory laparotomy identified a peritoneal nodule, leading to further investigation. Targeted NGS was performed on somatic and germline DNA samples and showed a frame shift deletion of 10 bp (c.1256_1265del: p.R419Ter) in the BRCA1 gene. This variant, identified only in tumor tissues, is novel and classified as pathogenic in ClinVar and ACMG databases. Additional somatic alterations were detected in TP53 and MSH6, while germline testing revealed only a variant of uncertain significance in BARD1. After first-line chemotherapy, the patient benefited from olaparib and achieved a progression-free survival of 23 months with good tolerance and no evidence of disease recurrence. CONCLUSION: This finding highlights the importance of integrating tumor-based genomic profiling with germline testing to identify actionable mutations and guide precision oncology. The identification of a novel somatic BRCA1 mutation expands the mutational spectrum of HGSOC and underscores the need to include underrepresented populations, such as those from North Africa, in genomic studies.

Humans↗

PubMed2026

Development of RS1-specific ACMG/AMP variant classification criteria with pilot variant curation.

Gene-based therapies are being developed for retinal diseases, including RS1-related X-linked retinoschisis. Therefore it is essential to determine which variants are pathogenic and which are benign when enrolling patients. The Clinical Genome Resource (ClinGen) X-Linked Inherited Retinal Diseases (XLRD) Variant Curation Expert Panel (VCEP) brings together clinician scientists, molecular biologists, and geneticists to apply their expertise and review the clinical, genetic, population, and functional evidence for variants. American College of Medical Genetics (ACMG) guidelines have been modified for RS1 to develop a highly systematic and conservative framework for evaluating variants. The curation process involves applying 28 different codes, each with 4 strength levels (very strong, strong, moderate, supporting) across different domains of phenotype, population data, computational assessment, functional impact, and segregation. With RS1-specific rules, a total of 54 pilot variants were tested. These included 47 variants in ClinVar. Of these 21 variants were re-classified: 2 likely pathogenic variants and one likely benign were changed to variants of uncertain significance and 4 previously unclassified variants were changed to pathogenic, likely pathogenic and likely benign. Other changes resolved conflicts or multiple classifications.

Humans↗

PubMed2025

Spinocerebellar Ataxia 27 A with Episodic Ataxia: Case Series of Fibroblast Growth Factor 14 (FGF14) Microdeletions.

Spinocerebellar ataxia 27 A (SCA27A) is a form of progressive cerebellar ataxia due to pathogenic variants in the Fibroblast Growth Factor 14 (FGF14) gene. The objective of this paper is to characterise the clinical spectrum of SCA27A microdeletions (> 50 bp, <2Mbp), and report two novel cases. Literature searches of PubMed, OMIM and ClinVar were carried out. We identified SCA27A microdeletions in 32 cases across 11 families. The phenotypic presentation is: 75% (24/32) nystagmus, 46% (15/32) ataxia, 21% (7/32) episodic ataxia, 21% (7/32) tremor, 15% (5/32) dysarthria, 34% (11/32) learning disability, 28% (8/32) neuropsychiatric disease. The presentation is variable within and between families. Episodic symptoms, nystagmus, learning disability and neuropsychiatric symptoms occur at an earlier age. Patient 1 represents the first case with a 58 kb FGF14 deletion who presented with a paroxysmal movement disorder. Patient 2 carries a 545 kb deletion and developed episodic ataxia and trigeminal neuralgia, a novel feature not previously described in this cohort. We report two cases of heterozygous FGF14 microdeletions: Patient 1 (58 kb) and Patient 2 (545 kb), expanding the phenotypic spectrum of FGF14 structural variants to 32 cases across 11 families. We review potential mechanism from pre-clinical studies relating FGF14 haploinsufficiency to cerebellar, cognitive, neuropsychiatric symptoms, as well as trigeminal neuralgia. We propose the hypothesis that the episodic symptoms in SCA27A align with the molecular pathology of a channelopathy and propose management strategies based on this insight.

Humans↗

PubMed2025

A quick guide to evaluating prime editing efficiency in mammalian cells.

According to the Clinvar database, modeling the diseases associated with pathogenic mutations requires the installation of base substitutions, small insertions or deletions. Prime editor (PE) was recently developed to precisely install any base substitutions and/or small insertions/deletions (indels) in mammalian cells and animals without requiring DSBs or donor DNA templates. PE also offers greater editing and targeting flexibility compared to other precision CRISPR editing methods because the versatile editing information is encoded in the reverse-transcription template of its prime editing guide RNA. However, optimal PE system selection and experimental design can be complex, and there are various factors that can affect PE efficiency. This chapter serves as a rapid entry-level guideline for the application of PE, providing an experimental framework for using PE at a specific genomic locus. RUNX1 was selected as a representative target site to illustrate the detailed methodology for constructing PE plasmids and the process of transfecting these plasmids into 293FT cells. We further examined the efficiency of PE-mediated genome editing in mammalian cells by using next-generation sequencing.

Gene Editing↗

PubMed2025

Extracting and calibrating evidence of variant pathogenicity from population biobank data.

Genomic medicine requires a robust evidence base of variant phenotypic impacts, which remains incomplete even in extensively studied genes with monogenic disease associations. Here, we evaluated the broad potential of using population cohort data to identify evidence that can be used in variant assessment. Across 41 genes related to 18 clinically actionable monogenic phenotypes, we calculated variant-level odds ratios of disease enrichment using data from 469,803 UK Biobank participants. We found significant differences in odds ratio values between ClinVar-labeled pathogenic and benign variants in 11 phenotypes, spanning both common and rare disorders. To facilitate clinical translation, we calibrated the strength of evidence provided by variant-level odds ratios to align with American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP) interpretation guidelines (PS4 criterion) and found that odds ratios may reach "moderate," "strong," or "very strong" evidence, varying by phenotype and gene. Overall, we found that 2.6% (N = 12,350) of participants harbor a rare variant of uncertain significance (VUS) with at least moderate evidence of pathogenicity-an indication of potentially unrecognized disease risk. Finally, by incorporating computational and functional data alongside population-based odds ratios, we identified variants that met the criteria for clinical reclassification. Notably, using this approach, we identified that 12.4% of rare VUSs in LDLR seen in participants meet diagnostic criteria to be classified as likely pathogenic, demonstrating its potential to scale the reclassification of VUSs.

Humans↗

PubMed2026

CanVar-UK: A collaborative platform for germline interpretation in cancer susceptibility genes.

Germline variants in cancer susceptibility genes (CSGs) are typically inherited rather than arising de novo. Hence, wide cascade testing of families across geographies is common, meaning consistency in variant classification is particularly critical. Variant interpretation requires collation of variant-level data from diverse sources, as well as assembly of comprehensive clinical data, often necessitating sharing of information between genomic testing centers. Here, we describe CanVar-UK, a freely accessible web platform bespoke designed to support interpretation of germline CSG variants. CanVar-UK contains variant-level data for over 1.1 million single-nucleotide variants (SNVs), comprising all possible coding SNVs in 116 established CSGs. The data sources with which variants are annotated include in silico scores from 11 clinically relevant tools, population allele frequencies from gnomAD v4.1, case counts from multiple cohorts, including National Health Service (NHS) clinical laboratory testing, variant-level readouts from 47 selected functional and splicing datasets across 19 CSGs, genetic epidemiology studies, and live linkage to existing consensus classifications in the ClinVar database. The diagnostic discussion forum is only available to registered diagnostic scientist users. Through this, a variant-tagged email message can be dispatched in real time across the diagnostic forum community of >1,500 users, with all exchanges and classifications captured and stored in the platform. Already widely used by NHS diagnostic clinical scientists in the UK, CanVar-UK has a rapidly growing international diagnostic user base (>800 UK and >600 non-UK registered users). Survey of the NHS diagnostic user community illustrates the wide-ranging utility of CanVar-UK within their clinical workflows for interpretation of germline CSG variants.

Journal Article↗

PubMed2025

A dataset of estimated heterozygous individual and carrier couple frequencies for pan-ancestry carrier screening.

The data described in this publication supported the development and evaluation of pan-ancestry reproductive carrier screening panels for autosomal recessive (AR) and X-linked (XL) conditions. Raw data included combined sets of DNA variants in 1,350 AR/XL genes obtained from the ClinVar and gnomAD databases. The dataset enabled calculations of positive yield for individuals and couples across both ancestry-specific and pan-ancestry, optimised "Goldilocks"-ranked gene panels, addressing population-specific variations in the frequencies of heterozygous individuals and carrier couples. The positive yield analysis offered a performance metric for carrier screening panels, facilitating the modeling of screening performance for panels of varying sizes and composition and providing resources for optimizing panel content to ensure equity across underrepresented genetic ancestries The dataset can support ongoing research into the equitable application of carrier screening and offers significant reuse potential for refining population genetic screening practices, validating computational models, and developing frameworks to update carrier screening panels in alignment with evolving genomic data, including in underrepresented and minority populations.

Carrier screening↗

PubMed2024

Estimation of carrier frequencies of autosomal and X-linked recessive genetic conditions based on gnomAD v4.0 data in different ancestries.

PURPOSE: Monogenic rare diseases contribute significantly to infant deaths and pediatric hospitalizations and cause burden to the patients and their families. The American College of Medical Genetics and Genomics recommended in 2021 that carrier screening of autosomal recessive and X-linked conditions with a carrier frequency of ≥1/200 and a severe or moderate phenotype should be offered when planning or during pregnancy. In November 2023 gnomAD v4.0 was released. It contains in total 807,162 individuals, being nearly 5× larger than previous versions, which have been used to estimate gene carrier frequencies (GCF). METHODS: We utilized gnomAD v4.0 (GRCh38) to calculate the GCFs for available genetic ancestry groups for variants having pathogenic or likely pathogenic classification (>80% of submissions) in ClinVar. We calculated GCF separately for exomes and genomes, combined data, and at-risk couple frequencies (ACF) per genetic ancestry group. RESULTS: In total, 324 genes had a GCF ≥1/200 in at least 1 ancestry subgroup. The number of genes with GCF ≥1/200 varied greatly between subgroups. ACFs were more similar, Ashkenazi Jewish having the highest ACF of 6.11%. CONCLUSION: Improved understanding of carrier risks and updated carrier screening content would allow patients to make more informed reproductive decisions.

Humans↗

PubMed2025

Optimizing gene panels for equitable reproductive carrier screening: The Goldilocks approach.

PURPOSE: Professional organizations recommend pan-ancestry carrier screening for autosomal recessive and X-linked conditions. Advances in DNA sequencing have allowed the analysis of hundreds of genes; however, the optimal number of genes for carrier screening remains unclear. The American College of Medical Genetics and Genomics (ACMG) has proposed a tiered approach recommending screening for 113 genes. METHODS: We analyzed ClinVar and gnomAD v4.1.0, for genes associated with serious autosomal recessive and X-linked conditions and modeled screening performance across panels of varying compositions and sizes in diverse genetic ancestries. We also reevaluated the ACMG gene list using the updated gnomAD data. RESULTS: We identified potential inconsistencies in the ACMG gene lists, particularly in the carrier test performance (defined as a positive yield) for underrepresented genetic ancestry groups. Modeling of the population data for 1310 genes revealed that the screening of 152, 248, 531, and 725 genes achieved 90%, 95%, 99%, and 99.7% positive yields, respectively, in couples. Real-world data from the screening of more than 60,000 couples were used to validate the model. CONCLUSION: Our methodology optimizes the gene content of carrier screening panels for diverse ancestry groups, provides a mechanism for continually updating guidelines, ensures consistency with genomic population data, and improves equity across populations.

Humans↗

PubMed2025

Calibration of additional computational tools expands ClinGen recommendation options for variant classification with PP3/BP4 criteria.

PURPOSE: We previously developed an approach to calibrate computational tools for clinical variant classification, updating recommendations for the reliable use of variant impact predictors to provide evidence strength up to Strong. A new generation of tools using distinctive approaches has since been released, and these methods must be independently calibrated for clinical application. METHODS: Using our local posterior probability-based calibration and our established data set of ClinVar pathogenic and benign variants, we determined the strength of evidence provided by 3 new tools (AlphaMissense, ESM1b, and VARITY) and calibrated scores meeting each evidence strength. RESULTS: All 3 tools reached the Strong level of evidence for variant pathogenicity and Moderate for benignity, although sometimes for few variants. Compared with previously recommended tools, these yielded at best only modest improvements in the trade-offs between evidence strength and false-positive predictions. CONCLUSION: At calibrated thresholds, 3 new computational predictors provided evidence for variant pathogenicity at similar strength to the 4 previously recommended predictors (and comparable with functional assays for some variants). This calibration broadens the scope of computational tools for application in clinical variant classification. Their new approaches offer promise for future advancement of the field.

Humans↗