Ricardo Henao

This study evaluates transformer-based models for single-cell transcriptomics, finding that larger datasets don't consistently improve downstream tasks and rank-based models struggle with batch effect correction.

TAMC predicts transcription factor binding activity from ATAC-seq data using a 1D-CNN model, outperforming existing tools, especially with limited sequencing depth.

The paper connects genetic architecture of human diseases with molecular phenotypes to uncover mechanisms of COVID-19 susceptibility.

Chromatin accessibility in immune cells from COVID-19 subjects is remodeled prior to seroconversion to reflect disease severity.

The paper develops gene-expression-based prognostic models for 30-day mortality in sepsis using community-onset and hospital-acquired sepsis cohorts, showing improved prediction when combined with clinical scores.