Genomics and Rare Diseases

Deciphering gain-of-function from loss-of-function variants with AlphaMissense: A case study with the mechanosensitive PIEZO1 ion channel protein

This study evaluates AlphaMissense and other tools for distinguishing gain-of-function and loss-of-function variants in PIEZO1, finding they struggle with GoF predictions and highlighting structural and biophysical insights.

Beyond thresholds: a fully Bayesian framework for quantifying allele count evidence for variant pathogenicity

This paper introduces a Bayesian framework to quantify allele count evidence for variant pathogenicity using Bayes factors, showing compatibility with existing criteria while allowing tunable parameters.

An interpretable molecular framework for predicting cancer driver missense mutations

The paper develops MutaPheno, a model that predicts cancer driver missense mutations using molecular features, achieving strong accuracy and outperforming existing tools.

FusionPath: Gene fusion pathogenicity prediction using protein structural data and contextual protein embeddings

FusionPath predicts gene fusion pathogenicity using protein structural data and contextual embeddings, outperforming existing methods with AUC scores of 0.95 and 0.87.

CHANGER: A novel genetic reference from 902 unrelated Chilean exomes to enhance South American medical genetics and research.

The paper presents CHANGER, a Chilean exome reference dataset of 902 unrelated individuals, to improve genetic research and variant interpretation in South American populations.

An AI-Integrated Framework for Precision Genomics in Coronary Artery Disease Using Whole Exome and Phenotypic Data

This paper introduces SIGMA, an AI framework that improves CAD risk prediction by integrating whole exome and phenotypic data, achieving higher sensitivity, specificity, and AUC.

Gene-centered representation of coding and regulatory variation enables outcome prediction

Volaria integrates coding and regulatory variation into gene-centered representations for disease outcome prediction using whole-genome sequencing data.

Benchmarking RNA-seq Tools for Real-World Diagnostic Applications

This study benchmarks RNA-seq tools for diagnostic applications, finding they identify 28 diagnoses from 68 samples but only 9 out of 74 undiagnosed cases.

Quantitative Optimization of Sensitivity and Specificity in Targeted and Whole-Exome Sequencing Using Reference-Standard DNA Mixtures

The paper optimizes sensitivity and specificity in targeted and whole-exome sequencing using reference-standard DNA mixtures, identifying optimal pipelines and VAF thresholds to reduce false positives and improve detection.

ProteoCast: a web server to predict, validate, and interpret missense variant effects

ProteoCast predicts and interprets missense variant effects using evolutionary and structural analysis, achieving high accuracy in pathogenicity prediction and identifying functional motifs in disordered regions.

ButterflyVI: enabling high-throughput variant interpretation and biomarker discovery with functional genomics

ButterflyVI integrates statistical learning and experimental data to annotate cancer variants and discover biomarkers, finding 60% of annotated variants were previously unknown.

How do clinician and parent reported data differ? An analysis of similarity and difference in the datasets from a cross-syndrome genetics cohort study(GenROC)

This paper compares parent-reported and clinician-reported phenotypic data in a cross-syndrome genetics cohort, finding differences in content and detail but similar quantities.

Distinguishing benign from pathogenic duplications involving GPR101 and VGLL1-adjacent enhancers in the clinical setting with the bioinformatic tool POSTRE

The paper evaluates the bioinformatic tool POSTRE to distinguish pathogenic from benign GPR101 duplications associated with X-LAG by analyzing enhancer hijacking in a tissue-specific context.

A predictive framework for stop-loss variants with C-terminal extensions

The paper develops TAILVAR, a machine-learning classifier to predict the functional impact of stop-loss variants by integrating multi-omics features, and shows it outperforms existing tools.

A Bayesian model for quantifying genomic variant evidence sufficiency in Mendelian disease

The paper introduces a Bayesian model to quantify the sufficiency of genomic variant evidence for Mendelian diseases, enabling consistent interpretation across institutions.

Automating the Clinical Interpretation of LDLR, APOB, and PCSK9 Variants: A Web-Based Platform for FH Diagnosis

A Shiny web platform pulls together gnomAD frequencies, UK Biobank LDL-C associations, AlphaMissense, LDLR deep mutational scanning and ClinVar/HGMD to draft ACMG/AMP evidence for familial hypercholesterolemia variants.

DoBSeqWF: A framework for sensitive detection of individual genetic variation in pooled sequencing data

DoBSeqWF is a pipeline for analyzing pooled sequencing data to detect rare genetic variants, improving sensitivity with machine learning filters.

Most disease gene variants show minimal population differentiation despite incomplete coverage

The paper compares methods for handling missing population data in variant interpretation and finds that most disease gene variants show minimal population differentiation despite incomplete coverage.

Identifying fundamental properties of neuromuscular disorder genes through genomic feature analysis

The paper identifies genomic features distinguishing neuromuscular disorder genes from other genes using machine learning and statistical analysis.

Variant curation of the largest compendium of FOXL2 coding and non-coding sequence and structural variants in BPES

The paper curates and classifies the largest compendium of FOXL2 sequence and structural variants in BPES, identifying key mutational patterns and improving diagnostic and genetic counseling.

PanelAppRex aggregates disease gene panels and facilitates sophisticated search

PanelAppRex aggregates and standardizes disease gene panels to enable efficient search and integration into bioinformatic workflows, recovering causal genes in all benchmarked cases.

pLM-SAV: A Δ-Embedding Approach for Predicting Pathogenic Single Amino Acid Variants

pLM-SAV predicts pathogenic SAVs using delta-embeddings from protein language models, achieving high performance on labeled data and improving ambiguous predictions over AlphaMissense.

Monte Carlo Simulation Analysis of Variant Burden Prompts Potential Oligogenic Interactions

The paper uses Monte Carlo simulations on summary-level allele counts to detect pathway-level variant burden enrichment in neuromuscular disease, identifying potential oligogenic interactions.

nAPOGEE: A machine-learning platform for clinically actionable pathogenicity assessment of all mitochondrial noncoding variants

nAPOGEE assesses pathogenicity of mitochondrial noncoding variants using machine learning, revealing consistent spatial correlation of predicted pathogenicity.

GA4GH Phenopacket-Driven Characterization of Genotype-Phenotype Correlations in Mendelian Disorders

The paper introduces GPSEA, a tool using GA4GH Phenopackets to analyze genotype-phenotype correlations in Mendelian disorders, identifying 225 significant associations across 85 cohorts.

MIRAGE: a Bayesian rare variant association analysis method incorporating functional information of variants

MIRAGE improves rare variant association analysis by incorporating variant functional information, outperforming existing methods in simulations and autism exome-sequencing data.

A fully-automated integrative workflow to streamline NGS-based analyses within Molecular Tumour Boards

This paper develops a fully-automated workflow to streamline NGS-based analyses in Molecular Tumour Boards by integrating variant interpretation and biomarker estimation.

TaoChongBao: A Large-Scale Caenorhabditis elegans Mutagenesis and Missense Variant Database Bridging Worm and Human Genomes

TaoChongBao is a database of C. elegans missense variants linked to human pathogenic sites, offering expanded mutation coverage and tools for variant interpretation and comparative genomics.

Combining MAVEs and computational predictors improves variant classification across ancestries in hereditary cancer genes

This study combines MAVEs and population-free VEPs to improve variant classification across ancestries in hereditary cancer genes, reclassifying over 90% of variants of uncertain significance.

VUSVista: Enhancing the Curation of Variants of Uncertain Significance

VUSVista is a software tool for managing Variants of Uncertain Significance, enabling automated updates and reclassification based on new evidence.