Kaitlin Susztak, MD, PhD

Professor of Medicine, Perelman School of Medicine, University of Pennsylvania

From Kidney Maps to Kidney Models: Genomics, Single-Cell Atlases, and Foundation Models for Precision Nephrology

Chronic kidney disease affects more than 800 million people worldwide, yet therapeutic progress has been slow because the causal cells, genes, and mechanisms have remained poorly defined. In this talk I will describe how human genetics, single-cell and spatial multi-omic atlases of hundreds of human kidneys, and machine learning can be integrated to move from association to mechanism. I will show how these maps identify the cell types and regulatory programs through which genetic variants drive kidney disease, and how they nominate and prioritize new therapeutic targets. Finally, I will discuss our recent work building kidney-focused AI foundation models that learn across molecular, histological, and clinical data, and how such models can support molecular diagnosis, patient stratification, and drug development in nephrology.