Research
My research spans human genetics, computational biology, and epidemiology. I work with health records, wearable sensors, and single-cell data.
Birth Order and Disease Risk Across the Phenome
Rzhetsky Lab, University of Chicago2024–2026
with Steven Kushner and Andrey Rzhetsky
Older studies of birth order typically rely on a few thousand families and test a handful of outcomes. I ran a phenome-wide scan on 10.3 million individuals from 5.1 million two-child MarketScan families, testing 569 diseases with two complementary designs: a between-family matched cohort and a within-family sibling comparison. We identified 150 Bonferroni-significant associations. The within-family design controls for stable family-level factors, while the paired designs make the remaining limits—including age-period-cohort structure and parental-age differences—explicit.
Nature HealthAccelerometry for Frailty Prediction
Rzhetsky Lab, University of Chicago2024–2026
with Yanan Long, Megan Huisingh-Scheetz, and Andrey Rzhetsky
I built a raw-device-file-to-model pipeline for seven-day free-living hip accelerometry from older adults enrolled in a longitudinal aging study and trained models to classify baseline frailty status and forecast 12-month decline. The project tested how much information a wearable stream can retain when in-person phenotyping is impractical.
medRxivCross-Disease Atlas of Human Skin Immune Cells
Bryson Lab, Ragon Institute2025
with Bryan Bryson and Robert Modlin
I harmonized 141,021 myeloid and antigen-presenting cells from 12 datasets into an 18-state reference spanning 13 skin diseases and healthy skin. The work required distinguishing reproducible cell states from dataset, tissue, and disease-composition effects rather than treating integration as a purely technical step.
Tuberculosis Granuloma Annotation
Bryson Lab, Ragon Institute2025
with Bryan Bryson and collaborators
I applied CellTypist models to tuberculosis granuloma perturbation data and carried out downstream comparative analysis.
Research background
I worked with large de-identified insurance-claims cohorts in the Rzhetsky Lab. Claims record billed care rather than true disease incidence, but their scale can support population-level questions that smaller cohorts cannot answer when the design and interpretation account for that limitation.
In the Bryson Lab at the Ragon Institute of Mass General, MIT, and Harvard, I worked on single-cell atlas harmonization, tuberculosis granuloma annotation, and de novo protein-design workflows.
Earlier work
My earlier research was in renal physiology at Case Western Reserve with Agustin Gonzalez-Vicente and Jeffrey Garvin, and in satellite-based air-pollution analysis during a Berkeley REU with Misbath Daouda. A list of the resulting publications and abstracts is on the Publications page.