Henry E. Miller

This paper shows deep learning models outperform baselines when using well-calibrated metrics for genetic perturbation prediction.

BioML-bench evaluates AI agents on end-to-end biomedical ML tasks across four domains, finding agents underperform compared to humans and diverse ML strategies correlate with higher scores.

This paper identifies SB000, a single gene factor that rejuvenates cells from multiple germ layers without inducing pluripotency, rivaling the effectiveness of Yamanaka factors.

CpGPT is a foundation model for DNA methylation analysis that outperforms specialized models in aging-related tasks and can impute methylation profiles across diverse conditions.

This study investigates the effects of PLCγ2 deficiency in mouse brains using multi-omics approaches, revealing sex-specific changes in lipid and gene expression that may link to Alzheimer's disease and aging.

The paper reprocesses R-loop mapping data, develops new quality methods, and introduces RLBase, a web server for exploring and analyzing R-loop datasets.

RLSuite provides a framework for analyzing R-loop data, including preprocessing, quality control, and comparison to public datasets.

The paper reprocesses R-loop mapping datasets to assess data quality and identifies consensus R-loop regions, revealing methodological differences and biological subtypes.