AI Tools Analyze Kidney Disease at Cellular Level to Help Tailor Treatments

Single-cell RNA sequencing has provided powerful insights, but challenges such as varying cell definitions and difficulty matching human diseases to lab models have limited its clinical use. Now, new artificial intelligence (AI)-driven solutions enable cell-level analysis to match patients with optimal treatments more precisely.

Researchers at the Perelman School of Medicine (Philadelphia, PA, USA) and the Wharton School at the University of Pennsylvania (Philadelphia, PA, USA) have developed two key resources: the SISKA 1.0 Atlas, a dataset built from over one million cells across 140 human, mouse, and rat kidney samples, and CellSpectra, an AI tool capable of analyzing a single patient’s sample in the context of species, disease, and therapy. These tools integrate statistical methods that assess gene programs rather than individual genes to better detect disease-related changes.

CellSpectra was designed to overcome limitations of current approaches, offering cross-species comparisons and enabling personalized insights at the cellular level. Both resources are open-source, making them accessible to researchers, clinicians, and scientists worldwide. This breakthrough, published this week in Nature Genetics, could impact millions suffering from kidney disease.

In a separate study published in Nature Medicine, the research team created the first comprehensive catalog of kidney proteins. The findings revealed frequent mismatches between protein abundance and gene activity, showing that RNA data alone is insufficient for fully understanding kidney disease biology. Linking protein profiles to traits like blood pressure, lipid levels, and kidney function offers new avenues for targeted therapies.

These tools and datasets could transform the staging and treatment of kidney disease by enabling precise patient stratification and therapy selection. The integration of protein-level data with RNA analysis expands the potential for discovering new drug targets. Researchers believe these methods could be adapted for other complex diseases, providing a framework for precision medicine beyond nephrology.

“We are moving from guesswork to precision. Kidney diseases are not all the same, but the use of AI helped us identify and catalog 70 distinct kinds of kidney cells that appear across human and animal samples. This improves the reliability of research and can lead to potential treatments,” said Katalin Susztak, MD, PhD, professor of Nephrology, Genetics, and director of the Penn/CHOP Kidney Innovation Center.

Related Links:
Perelman School of Medicine
Wharton School