Michael Thompson

This paper uses deep learning models and MPRAs to predict gene expression from sequence and identify causal regulatory motifs and grammars.

The paper experimentally quantifies amyloid nucleation for >100,000 sequences, trains an interpretable deep learning model to predict it, and reveals the model's decision-making process.

The paper introduces CONTENT, a method for multi-context genetic modeling of transcriptional regulation, which improves power and discovers more gene-phenotype associations in TWAS.

This paper improves genetic studies of MDD by integrating shallow and deep phenotypes, enhancing GWAS power and PRS accuracy while preserving specificity.

BATMAN integrates single-cell RNA-Seq datasets efficiently, outperforming existing methods by up to 80% in batch effect correction while preserving cell relationships.