GWAS summary statistics

This study shows that excluding low-frequency variants and indels in fine-mapping leads to false discoveries and introduces SuSiE-OUT to address this issue.

LRCQ estimates phenome-wide heritability enrichment of genomic loci using GWAS summary statistics, identifying enriched regions and their functional annotations.

The paper benchmarks variant effect predictors using a leakage-controlled dataset and develops a gradient-boosting consensus model that outperforms existing tools.

The paper fine-maps GWAS risk variants for PDAC and identifies candidate target genes through chromatin conformation capture, finding functional interactions and eQTLs at specific loci.

The paper identifies rare SNPs associated with hypo-inflammation and adverse outcomes in trauma patients, highlighting type 17 immune responses as central to critical illness.

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

The paper introduces a pipeline to create curated, balanced variant-trait association datasets for complex disorders using GWAS and functional annotations.

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

This paper introduces S4-Oligo, an optimized polygenic score method for clinical use, which matches or exceeds existing models' accuracy while using fewer SNPs.

This study introduces PGSa,d, a polygenic scoring framework that incorporates both additive and dominance effects, showing improved accuracy for traits with significant dominance contributions.

This paper introduces GenSep, a framework to quantify genetic separability of disease subtypes, showing current PRSs fall short of theoretical limits.

This paper introduces Seal, an AI framework that models gene regulation and variant effects in the developing human brain across multiple regions, cell types, and stages, identifying regulatory mechanisms linked to neuropsychiatric diseases.

The paper imputes fluid intelligence scores in a large cohort, reducing ascertainment bias and increasing power to detect genetic associations with common and rare variants.

This study uses Mendelian randomization to assess the causal role of CACNA1C, a target of calcium channel blockers, on thoracic aortic aneurysm and dissection, finding lower risk with higher aortic expression.

The paper identifies a bias in local genetic correlation estimation due to significance filtering and proposes a correction method with a diagnostic tool.

MaveMD provides a clinical interface for MAVE data to classify variants of uncertain significance and improve genetic medicine outcomes.

The paper benchmarks variant effect predictors using MAVE data, identifies reliability ceilings, and shows that correction improves performance in some regions but not others.

The paper introduces C2BM, a deep learning model that improves SNP prioritization by integrating genomic annotations and trait relationships, showing better predictions when related trait labels are available.

This paper introduces a new unbiased linkage disequilibrium estimator called Random Probe, which outperforms existing methods in reducing bias, especially in challenging scenarios.

SPLENDID improves polygenic risk prediction across diverse populations by incorporating continuous genetic ancestry without requiring discrete ancestry labels.

The paper identifies individuals with extreme polygenic risk scores (PRS) for Alzheimer's disease (AD) and shows they have significantly elevated AD risk across multiple ancestries and APOE diplotypes.

The paper evaluates 1,752 polygenic risk score models for Alzheimer's disease and finds significant variability in results across methods and ancestries.

MissenseHMM annotates missense variants with states derived from joint modeling of pathogenicity scores, improving associations with clinical and experimental data.

This study identifies high-risk nsSNPs in TSC1 and TSC2 genes using computational tools, showing they reduce protein stability and disrupt functional sites.

This study evaluates when nonlinear machine learning methods outperform standard polygenic risk scores, finding their effectiveness depends on genetic architecture and sample size.

The paper introduces a masked-MLP neural network to improve genetic risk score calculation from noisy and missing data, showing better robustness than standard methods.

S2F-Agent improves genome interpretation by orchestrating sequence-to-function models with verifiable execution and interpretation.

This paper introduces PGSC, a polygenic score framework that incorporates locus-specific gene-context interactions, showing improved prediction across multiple traits and ancestries.

The paper compares variant effect predictors and MAVEs to understand their agreement and disagreement in assessing genetic variant impacts.

SheepEpimap combines 516 histone CUT&Tag, ATAC-seq and RNA-seq datasets from 43 sheep tissues to annotate 2.93 million cis-regulatory elements and 145,407 allele-specific variants for trait interpretation.