Genomics and Rare Diseases

The paper introduces LASSIE, a model to estimate allele-specific fitness effects across human protein-coding sequences, showing its relevance to evolution and disease.

The paper estimates the risk of five disorders from paternal-age-related de novo mutations using whole-exome-sequencing data and Danish registry data.

The paper introduces a disease-specific variant pathogenicity predictor trained on clinical sequencing data from cardiomyopathy, epilepsy, and RASopathies patients.

This paper introduces GADO, a tool that uses gene expression data to predict gene-phenotype associations and improve diagnostic yield in exome sequencing by prioritizing variants in patients with unknown genetic diagnoses.

The paper develops a low-cost NGS gene panel for diagnosing intellectual disability and autism spectrum disorder, achieving a 27% diagnostic yield in 150 patients.

This study evaluates 20 variant calling pipelines using WES data and simulated datasets to identify the best combinations for detecting SNVs and InDels.

The paper introduces an AI-driven method that uses facial images to prioritize exome data, achieving 99% top-10 accuracy in identifying disease-causing genes.

The paper identifies 23 novel deleterious nsSNPs in the MEFV gene using bioinformatics tools, which may serve as diagnostic markers for Familial Mediterranean Fever.

The paper proposes a framework for classifying and reporting low-penetrance variants associated with disease risk, emphasizing data elements and criteria for clinical inclusion.

VEP-G2P filters genomic variants for diagnostic purposes, showing higher sensitivity but lower precision for de novo and inherited variants, with significant background noise in controls.

This paper introduces a weakly supervised regression model to predict pathogenicity of missense variants and infer molecular mechanisms from imprecise annotations.

This paper introduces VarCoPP, a machine-learning method that identifies pathogenic variant combinations in gene pairs, achieving high accuracy and providing interpretable explanations for predictions.

MetaDome provides a web service to analyze genetic variants by aggregating homologous human protein domains, improving interpretation of variants of unknown significance.

GenePy estimates gene-level pathogenicity from whole-exome sequencing data by combining mutational burden and biological impact, demonstrating improved sensitivity in identifying causal genes in common diseases.

This study uses a random forest algorithm to predict gene pathogenicity for Mendelian diseases, achieving 80% accuracy and identifying 10,399 protein-coding genes as potential Mendelian disease genes.

The paper uses human genetic data to analyze 3D protein structures and identifies intolerant sites that correlate with functional and drug target properties.

The paper evaluates nine FFPE DNA extraction methods for NGS, finding average WES coverage of ~20-30X and 99% genotype concordance with targeted panels.

The paper explains that population genetic measures like pLI reflect selection on heterozygotes, not dominance or haploinsufficiency.

The paper curates 164 gene-disease pairs for hearing loss, classifying their clinical validity to improve genetic testing accuracy.

Atlas-CNV identifies single-exon CNVs in targeted sequencing data with reduced false discovery rate using C-scores and ExonQC thresholds.

The paper estimates carrier rates for 416 genes across 6 ancestries and shows that screening 47 high-carrier-rate genes captures most at-risk couples.

The paper maps UniProtKB protein annotations to the human genome and demonstrates their utility in interpreting genetic variants.

S-CAP improves pathogenicity prediction for splicing-relevant variants by reducing uncertain variants by 41% with 95% sensitivity.

PathoMAN automates germline variant curation using ACMG guidelines, achieving high concordance with expert curation and proposing refinements for better ClinVar reporting.

This study computationally analyzes nsSNPs in the TAL1 gene to predict their impact on protein function and identify potential miRNA binding site alterations.

The paper develops a supervised learning method to score pathogenic non-coding variants in Mendelian diseases using purifying selection signals, achieving improved predictive performance over existing methods.

The paper evaluates 10 variant pathogenicity predictors on 63,160 benign variants from the ExAC database, finding large specificity differences and identifying the best-performing tools.

The paper identifies and characterizes TUBA1A variants in 166 individuals with tubulinopathy, highlighting common phenotypes and genotype-phenotype associations.

DeepPVP prioritizes causative variants using deep learning and phenotype similarity, outperforming existing methods in speed and accuracy.

The paper analyzes exome sequencing data to assess the contribution of splice site mutations to developmental disorders and estimates their pathogenicity and underrepresentation in public databases.