Research

My current research is motivated by scientific questions arising in (i) understanding gene regulation from complex single-cell genomics data and (ii) translating these molecular-scale insights into clinical utility. To this end, I develop statistical methods for dimension reduction of high-dimensional single-cell data, uncovering causal structure among genomic elements, and AI/ML-powered integration of heterogeneous genomic and clinical datasets.

Single-cell 3D Genomics & Epigenomics

3D genome organization and the epigenome jointly orchestrate gene regulation across cellular contexts. I have developed statistical methods to integrate these multi-omic data.

K. Park and S. Keleş. Joint tensor modeling of single cell 3D genome and epigenetic data with Muscle, Journal of the American Statistical Association, 2024.
K. Park and S. Keleş. Sparse higher-order partial least squares for simultaneous variable selection, dimension reduction and tensor denoising, Biometrika, 2026.
C. Alberca*, K. Park*, L. Papale, A. Madrid, P. Bergmann, S. Keleş, and R. Alisch. Multi-omic data integration improves the resolution of the molecular etiology of autism in a mouse model, Molecular Psychiatry, 2026. (* denotes equal contribution)

Gene Regulation in Disease Contexts

Understanding how gene regulation is altered in disease is key to translating genomic insights into clinical applications. In collaboration with scientists, such as cancer biologists and neuroscientists, I have developed statistical methods to decipher these disease-specific mechanisms from high-dimensional single-cell data.

K. Park, Z. Sun, R. Liao, E. Bresnick, and S. Keleş. Systematic background selection with BasCoD enhances contrastive dimension reduction in single cell genomics, Nature Communications, 2026.
K. Park, T. Gao, J. Yan, and S. Keleş. MINTsC learns multi-way chromatin interactions from single cell high throughput chromatin conformation data, Nature Communications, 2026.
K. Park, H. Kang, K. Hogan, R. Alisch, and S. Keleş. Representation-powered sensitivity analysis with applications to outcome-wide observational epigenomic studies, Preprint, 2026.

Causal Inference from Genomic Perturbations to Phenotypes

Elucidating how genomic perturbations propagate through causal gene networks to phenotypes can guide precision medicine and therapeutic target identification. I have developed causal discovery/inference tools that integrate insights from perturbational and observational population-scale single-cell data.

K. Park and H. Li. Confounder-robust causal discovery and inference in Perturb-seq using proxy and instrumental variables, Journal of the American Statistical Association (accepted), 2026.
K. Park and H. Li. Causal path analysis from perturbational and population-scale single-cell data with multiscale confounding and measurement error, Preprint, 2026.
X. Hou, K. Park, and H. Li. Statistical inference for causal discovery under selection and latent variables via single-target interventions, Preprint, 2026.

Page updated

Google Sites

Report abuse