George Crowley

The paper identifies core lesion cells in endometriosis that retain uterine transcriptomic identity across hormonal states and uses them to classify disease status and lesion subtype.

The paper expands Tabula Sapiens to include non-coding RNA data from 22 human tissues, revealing cell type-specific expression and subcellular localization patterns of non-coding RNAs.

The paper applies Pareto optimality to single-cell RNA sequencing data to identify cellular archetypes and their functions, finding that this principle explains transcriptomic variation across cell types.

The paper benchmarks large language models for cell-type and gene set annotation using AnnDictionary, finding high accuracy in cell-type annotation and strong functional gene set recovery.