variant effect prediction

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.

SARDINE uses EWAS summary statistics to estimate epigenetic clock-phenotype associations without individual-level data, revealing heterogeneity in biomarker associations with aging and diseases.

This paper introduces MirrorDNA, a DNA language model with a strand-equivariant tokenizer that improves prediction consistency across DNA strands.

The paper assesses the transportability of mouse gene knockout models to human complex phenotypes using GPAT framework, finding evidence for 12 phenotypes but limited for others.

The paper presents a curated dataset of 1,413 LNP cytotoxicity measurements to examine structure-activity relationships and develop computational models of lipid nanoparticle cytotoxicity.

FastaGuard is a FASTA preflight QC tool that detects structural issues in assemblies and stops pipelines before expensive analysis, with high accuracy in identifying faults.

This paper improves variant-effect prediction in sequence-to-function models using synthetic MPRA data generated from distilled model predictions.

The paper introduces eQIF, an unsupervised machine-learning method that integrates Enformer predictions to prioritize gene-regulatory variants, showing strong performance in predicting eQTLs and GWAS variants.

This paper introduces PGL, a Bayesian regression method for microbiome feature selection that uses phylogenetic trees to improve detection of exposure-associated ASVs.

DoAggregate performs rare-variant aggregation tests across genes, pathways, and GO terms, providing detailed cohort-level results and visualizations.

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

This study uses deep learning to predict cis-regulatory effects of human-specific mutations across 305 tissues, revealing their role in gene expression evolution.

The paper introduces Bvalcalc, a tool to calculate B-values, which quantify the effect of background selection on genetic diversity, and validates it using simulations and public data.

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.

TransPepMS verifies peptides not in reference proteomes by resolving ambiguities in mass-spectrometry data to distinguish genuine non-canonical peptides from false positives.

The paper introduces MEMAR, a Maximum Entropy model for predicting m6A methylation in human transcripts and influenza A viral segments, showing improved performance over existing methods.

The paper evaluates precomputed SpliceAI scores for splice-altering variant prioritization in NDD cohorts, finding inaccuracies and providing a tool to improve variant triage.

VenusGenome improves microbial genome representation learning by using a genome-aware masked autoencoder, achieving high performance in phenotype prediction and gene-level tasks.

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

This study evaluates how selective sweeps affect DFE inference using fitdadi, finding that performance declines at low recombination rates and with unfolded SFS, but remains largely unbiased with a flexible demographic model.

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

The paper evaluates sequence models in noncoding RNA assays by comparing model increments against baselines and highlights the importance of baseline content and regularization.

ONT-LibQC uses shallow nanopore sequencing to rapidly assess chromatin assay library quality before deep sequencing, showing consistent results with Illumina sequencing.

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

Sorting S-phase cells by Cyclin A2 rather than DNA content exposes a uniform genome-wide delay after CDC7 inhibition and a broad replication timing advance in RIF1 knockout.

The paper introduces DEEP-Boltz, a framework that extracts energetic potentials from protein structure predictors, outperforming existing methods in predicting mutation-induced stability changes.

This paper introduces ConcuRx, a statistical framework that improves gene prioritization across heterogeneous CRISPR drug-response datasets by modeling between-study variability, leading to better identification of MOA-associated genes.

This study uses genome-scale CRISPR base editing in yeast to assess the functional effects of 308,461 variants, identifying 18,009 with significant growth impacts.

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.