Sevahn K. Vorperian

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.

This study characterizes transcriptomic and metabolomic profiles in human urine, identifying distinct cell type repertoires and metabolic signals compared to plasma.

Nearly 500,000 cells from 24 human tissues, many from one donor, characterize over 400 cell types and enable within-donor study of T-cell clones, B-cell mutation rates and splicing.

The paper develops methods to identify cell type contributions in cell-free RNA from blood, showing it reflects diverse cell types in health and disease.

The paper introduces scRFE, a method using random forests and recursive feature elimination to identify minimal gene sets distinguishing single-cell types in scRNA-seq data.