Vineel Bhat

Language models reveal evidence gaps in variants of uncertain significance

The paper uses language models to analyze ClinVar variant summaries, identifying evidence gaps in Variants of Uncertain Significance and reclassifying some as likely benign or pathogenic.

Extracting and calibrating evidence of variant pathogenicity from population biobank data

The paper uses population biobank data to identify and calibrate evidence of variant pathogenicity for 41 genes, finding that 2.6% of participants have rare variants with moderate or stronger evidence of pathogenicity.

Joint estimation and imputation of variant functional effects using high throughput assay data

The paper develops FUSE, a framework that jointly estimates and imputes variant functional effects using deep mutational scanning data, improving variant impact predictions and clinical discrimination.

Joint genotypic and phenotypic outcome modeling improves base editing variant effect quantification

This paper improves variant effect quantification in base editing screens by integrating genotypic and phenotypic data using a Bayesian network called BEAN.

Informing Variant Assessment using Structured Evidence from Prior Classifications (PS1, PM5, and PVS1 Sequence Variant Interpretation Criteria)

The paper uses prior ClinVar variant classifications to inform ACMG/AMP variant interpretation by identifying SNVs with similar functional consequences to pathogenic variants.