Penn Developed AI Tools and Datasets Help Tailor Treatments for Kidney Patients

Doctors treating kidney disease have long depended on trial-and-error to find the best therapies for individual patients. Now, new artificial intelligence (AI) tools developed by researchers in the Perelman School of Medicine and the Wharton School at the University of Pennsylvania can analyze kidney disease at the cellular level to match the most effective treatments and speed up solutions. This breakthrough, published this week in Nature Genetics, could impact millions who have kidney disease.

"We are moving from guesswork to precision," said Katalin Susztak, MD, PhD, a professor of Nephrology, Genetics and director of the Penn/CHOP Kidney Innovation Center. "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," Susztak added.

The Penn team tackled a major challenge in single-cell RNA sequencing, a cutting-edge technique that examines the genetic activity of individual cells. Until now, this method has been difficult to apply to individual patients due to inconsistent cell type definition and uncertainty about which lab models (like mice or rats) best match human diseases.

The team’s solution includes SISKA 1.0 Atlas: A massive dataset built from over 1 million cells across 140 human, mouse, and rat kidney samples. In combination with a new statistical method that examines gene programs - sets of co-regulated genes representing biological pathways - rather than individual genes, it was easier to spot disease-related problems in a person’s cells. The new, open source tool, called CellSpectra, was created right at Penn.

"We built CellSpectra to do what current methods cannot: analyze one patient’s sample at a time, and interpret it in the context of species, disease, and therapy," said Nancy Zhang, PhD, the Ge Li and Ning Zhao professor of statistics and data science at the Wharton School. "Both of these tools will be free for anyone to use. Now researchers, scientists, and clinicians will all have access to these tools that allow personalized treatments with greater precision," Susztak added.

In a separate study, the Susztak team has also created the first comprehensive catalog of kidney proteins, offering a new lens on how protein abundance, not just gene expression, contributes to disease. This work, published in Nature Medicine, found that protein levels in kidney cells often don’t match gene activity (a mismatch called discordance), showing that studying genes alone isn’t enough to understand how diseases develop.

"This is a significant step forward in understanding the biology of kidney disease - not just at the RNA level, but also at the functional protein level," said Susztak. "Linking protein profiles with traits like blood pressure, lipid levels, and kidney function opens new doors for therapies that target the right molecules in the right patients."

The studies were supported by the National Institutes of Health (2R01DK076077-15, 5R01DK087635-15, 5P50DK114786-07, 5R01DK105821-08, 5R01DK132630-02, R01 DK105821, R01 DK087635, R01 DK076077, R01 DK12345 and 1R56AG081351), the National Science Foundation DMS/NIGMS (2245575), and the Translation Genetics in Renal Medicine (TIGER) grant from the University of Pennsylvania.

Klötzer KA, Abedini A, Li S, Balzer MS, Liang X, Levinsohn J, Ha E, Dumoulin B, Hogan JJ, Quinn G, Bloom RD, Schuller M, Eller K, Halmos B, Zhang NR, Susztak K.
Analysis of individual patient pathway coordination in a cross-species single-cell kidney atlas.
Nat Genet. 2025 Aug;57(8):1922-1934. doi: 10.1038/s41588-025-02285-0

Most Popular Now

AI also Assesses Dutch Mammograms Better…

AI is detecting tumors more often and earlier in the Dutch breast cancer screening program. Those tumors can then be treated at an earlier stage. This has been demonstrated by...

RSNA AI Challenge Models can Independent…

Algorithms submitted for an AI Challenge hosted by the Radiological Society of North America (RSNA) have shown excellent performance for detecting breast cancers on mammography images, increasing screening sensitivity while...

AI could Help Emergency Rooms Predict Ad…

Artificial intelligence (AI) can help emergency department (ED) teams better anticipate which patients will need hospital admission, hours earlier than is currently possible, according to a multi-hospital study by the...

Head-to-Head Against AI, Pharmacy Studen…

Students pursuing a Doctor of Pharmacy degree routinely take - and pass - rigorous exams to prove competency in several areas. Can ChatGPT accurately answer the same questions? A new...

NHS Active 10 Walking Tracker Users are …

Users of the NHS Active 10 app, designed to encourage people to become more active, immediately increased their amount of brisk and non-brisk walking upon using the app, according to...

New AI Tool Illuminates "Dark Side…

Proteins sustain life as we know it, serving many important structural and functional roles throughout the body. But these large molecules have cast a long shadow over a smaller subclass...

Brain Imaging may Identify Patients Like…

By understanding differences in how people’s brains are wired, clinicians may be able to predict who would benefit from a self-guided anxiety care app, according to a new analysis from...

The Human Touch of Doctors will Still be…

AI-based medicine will revolutionise care including for Alzheimer’s and diabetes, predicts a technology expert, but it must be accessible to all patients. Healing with Artificial Intelligence, written by technology expert Daniele...

Deep Learning-Based Model Enables Fast a…

Stroke is the second leading cause of death globally. Ischemic stroke, strongly linked to atherosclerotic plaques, requires accurate plaque and vessel wall segmentation and quantification for definitive diagnosis. However, conventional...

Stepping for Digital Rewards

Walking is well known to have significant health benefits, but few people achieve the daily recommended steps. Fortunately, mobile health (mHealth) applications have emerged as promising tools to promote physical...

AI can Find Cancer Pathologists Miss

Men assessed as healthy after a pathologist analyses their tissue sample may still have an early form of prostate cancer. Using AI, researchers at Uppsala University have been able to...

Unlocking the 10 Year Health Plan

The government's plan for the NHS is a huge document. Jane Stephenson, chief executive of SPARK TSL, argues the key to unlocking its digital ambitions is to consider what it...