Genomics and Rare Diseases

This study analyzes genetic variation in older Lebanese adults to distinguish reference sampling biases from true allele frequency differences and genotype expectations.

The paper develops and validates disease-specific ACMG/AMP variant interpretation specifications for HBB and HBA2 in thalassemia, providing a standardized framework for globin gene variant classification.

The paper introduces Watershed-DL, a framework integrating sequence-to-omics predictions and multi-omics data to prioritize rare variants, finding it increases prioritized variants by 61.4% and identifies candidate disease alleles.

This paper identifies population-scale SNV targets for allele-specific ASO therapy design, achieving high coverage across diverse ancestry groups and supporting clinical relevance.

This study evaluates the diagnostic value of long-read genome sequencing in DEE patients, finding it particularly effective for resolving complex structural variants not detected by short-read sequencing.

The paper shows that variant effect predictors influence ClinVar classifications, creating circularity in variant classification and prediction.

The paper identifies rare genetic variants in DYM causing Dyggve-Melchior-Clausen disease through reanalysis of whole-exome sequencing data from affected individuals.

AGVarPred is a gradient-boosted classifier that improves pathogenicity prediction of germline variants by integrating AlphaGenome features, outperforming existing models across SNVs and indels.

The paper calibrates AlphaGenome Variant Impact (AVI) and SpliceAI2 scores for variant classification, showing their performance across different variant classes and comparison methods.

The paper shows inherited protein-truncating variants contribute to developmental disorders through partial gene expression loss, affecting ~20% of children with such variants.

This study uses exome sequencing to identify genetic variants in the Mexican Jewish Community, finding higher rates of carrier status and novel alleles compared to standard panels.

BUDDY is a web platform that streamlines the functional modeling of human missense variants using yeast orthologs, reducing bioinformatic preparation time and enabling non-experts to design experiments.

This study assesses the pathogenicity of VHL missense mutations using AlphaMissense, protein stability simulations, and molecular dynamics to identify critical regions and mutation effects.

Varformer prioritizes drug targets and variants by integrating variant and gene features using a transformer model, achieving high precision across diverse benchmarks.

hvantk is a framework for variant annotation and analysis that integrates multi-omic data, improving the recovery of disease genes in case-control cohorts.

The paper introduces hEVO, a genomic foundation model trained on human haplotype sequences, achieving top performance in variant interpretation and functional annotation across multiple benchmarks.

The paper introduces KASM to integrate functional annotations and estimate heritability of rare protein coding variants across 786,871 exomes, finding higher heritability than previous estimates.

Mahi-mahi accurately genotypes repeat expansions from nanopore sequencing data, matching orthogonal methods and identifying pathogenic alleles.

This paper benchmarks the AlphaGenome Variant Impact score against SpliceAI on post-release ClinVar classifications and experimental splicing assays, finding AVI performs well but not outperforms splicing predictors on AUROC.

The paper analyzes single nucleotide substitutions in the CYP21A2 gene using AlphaGenome to connect clinical annotations, population data, splicing predictions, and enzyme function.

The paper assesses the association between rare variant genetic loads and cognitive traits, intellectual disability, and psychiatric disorders in population-based cohorts.

The paper identifies structural vulnerabilities in ATP1A3 variants that lead to distinct neurological phenotypes through different mechanistic consequences.

This paper develops a calibrated functional assay to assess CACNA1A variants, showing it can accurately classify pathogenic and benign variants with high specificity and sensitivity.

IVA uses a small language model to perform local genomic variant interpretation, achieving high sensitivity but lower precision in identifying pathogenic variants.

The paper investigates missense spliceogenic variants (MSVs) using genomic and proteomic data, finding they are common and can cause disease through splicing disruption, often missed by protein-centric tools.

The paper recalculates CNV penetrance for intellectual disability, showing lower estimates due to methodological differences, and recommends phenotype-specific risk reporting.

This study benchmarks saliva-derived DNA against blood for long-read sequencing in autism, showing comparable variant detection despite nonhuman DNA contamination.

This study evaluates TruPath Genome's ability to characterize repeat expansions and variants in complex genomic regions using archived DNA samples, showing consistent results with orthogonal methods.

This paper evaluates the clinical use of long-read genome sequencing in Sweden, showing it provides molecular diagnoses and reduces diagnostic costs.