Genomics and Rare Diseases

The paper introduces COSMIS, a framework to quantify 3D spatial constraint on amino acid sites in the human proteome, showing it outperforms existing metrics and links constraint to disease relevance.

This study evaluates genome sequencing in hospitalized newborns, finding it more effective than standard care in identifying genetic diagnoses, especially in those with specific phenotypes.

The paper analyzes whole genome sequencing data from endothelial vasculopathy patients to identify rare variants in clinically relevant genes and their association with disease severity.

The paper uses C. elegans to functionally test STXBP1 variants, classifying them as benign, pathogenic, or VUS with machine learning models.

This paper integrates monogenic, polygenic, and clinical risk factors to predict disease risk for carriers of rare variants, identifying elevated risk in 18.22% of UK Biobank participants for breast cancer, familial hypercholesterolemia, and colorectal cancer.

The paper shows that using structure-based predictors of intolerance improves the identification of pathogenic variants in whole exome sequences of patients with developmental and epileptic encephalopathies.

The paper introduces the Pleiotropic Variability Score (PVS) to quantify the phenomic associations of genomic variants using semantic reasoning and validates it with electronic health records.

This paper introduces PathoGN, a network-based method for predicting variant pathogenicity, which outperforms existing methods and identifies pathogenicity for 3,994 variants of uncertain significance.

SvAnna predicts pathogenicity of structural variants in long-read sequencing data by integrating transcript and regulatory annotations with clinical phenotype data, achieving high accuracy in prioritizing pathogenic variants.

The paper identifies genetic variants contributing to clinical heterogeneity in Lowe syndrome by analyzing genomic data from affected family members.

The paper describes the evolution of the CIViC knowledgebase to meet the needs of the cancer variant interpretation community by expanding its functionality and data types.

The paper identifies de novo variants in the 1000 Genomes Project using a GPU-based workflow, revealing cell line artifacts and cancer-related mutation profiles.

The paper quantifies disease risk associated with 37,772 clinical variants across 72,434 individuals, revealing heterogeneity in penetrance and limitations of current variant classification systems.

Adding tandem mass tag proteomics to genome and transcriptome data resolved 22% of 121 unsolved suspected mitochondrial disease cases and surfaced two new disease genes.

The paper analyzes coding variants in the FTO gene using the ExAC database to assess their potential functional impact and finds many rare variants classified as pathogenic are not detected by GWAS.

The paper uses a mini-gene splicing assay to evaluate the pathogenicity of 12 MUT/MMUT gene variants, confirming three as pathogenic and nine as non-pathogenic.

The paper identifies and validates non-synonymous SNPs in the CA27.29 gene that may affect protein stability and breast cancer susceptibility.

This study improves variant calling by integrating population data into deep learning models, enhancing precision and recall across diverse populations.

The paper identifies gene- and protein-level features distinguishing gain-of-function and loss-of-function mutations using NLP and machine learning on a new database of pathogenic mutations.

AMELIE 3 automatically reanalyzes undiagnosed Mendelian disease patients using new literature, achieving 80% diagnostic yield with 0.5-1 alerts per patient per year.

The paper evaluates a quality threshold for whole exome sequencing variants to determine when Sanger sequencing confirmation is unnecessary, finding high accuracy with specific quality score and allele fraction thresholds.

This study identifies three high-risk pathogenic nsSNPs in the TP53 gene associated with Li-Fraumeni syndrome using in silico analysis of functional and structural consequences.

This study demonstrates that whole genome sequencing can effectively detect repeat expansions in neurological disorders with high sensitivity and specificity, enabling diagnosis in previously undiagnosed patients.

This study evaluates using locus variability, measured by Shannon entropy from variant allele frequencies, to predict pathogenic SNVs in mitochondrial and splice-site regions, showing strong performance compared to existing tools.

This paper introduces targeted long-read sequencing to identify complex structural variants and disease-causing mutations missed by conventional methods, improving diagnostic accuracy in genetic disorders.

The paper introduces a new software tool, MAPS, that uses deep learning to analyze cell phenotypes from high-content microscopy, enabling rapid functional assessment of genetic variants.

The paper uses allele-specific expression analysis in a large cohort to identify high-impact, autosomal-dominant variants that are compensated by imbalanced expression.

This study identifies population-specific protein-truncating variants and introduces LoFfeR, a tool that improves the annotation of low-confidence loss-of-function alleles by filtering more variants with the same false positive rate.

The paper identifies novel essential gene candidates by analyzing human embryonic gene expression and integrating it with mouse knockout data and evolutionary conservation.