We generate multiple high-throughput data sets on the genome, epigenome, and transcriptome across species and in disease-relevant tissues/cells at single-cell multi-omic resolution and use machine learning computational approaches to integrate and analyze this data. We aim to better understand the effects of genetic variation on chromatin architecture and transcriptional regulation at single-cell resolution. The major goal of the lab is to generate mechanistic knowledge about how disease susceptibility is encoded in the non-coding portion of the genome (from GWAS), with a focus on complex metabolic diseases including diabetes and related traits. We accomplish this through an interdisciplinary combination of molecular, cellular, and computational approaches.