By: Richard L. Smith
A Kean University professor has helped develop a new artificial intelligence tool designed to identify whether potential medicines could cause kidney damage before they reach costly laboratory or clinical testing.
According to information released by Kean University, Assistant Professor Supratik Kar, Ph.D., co-developed KidneyTox_v1.0, or KidneyTox, with researchers from the University of Salerno in Italy.
The free, web-based platform uses AI to predict kidney toxicity in small molecules. Unlike many prediction tools, KidneyTox also provides researchers with information explaining why a particular molecule is considered potentially toxic or non-toxic.
Kidney toxicity is a significant reason promising drug candidates fail during development, often after researchers have already invested substantial time and resources.
Kar said the goal was to give scientists more than a simple toxicity prediction by identifying portions of a chemical structure that may be responsible for potential kidney damage.
Researchers could then potentially redesign a molecule before moving forward with expensive testing.
The team developed KidneyTox using data from 565 chemically diverse FDA-approved small molecules, including nearly 300 drugs associated with kidney toxicity and other compounds considered safe.
Researchers around the world have already begun using the platform to evaluate compounds before experimental testing.
The project, published in Scientific Reports, builds upon Kar’s previous work involving AI tools that predict liver toxicity.
Kar said his long-term goal is to develop a comprehensive open-source system capable of evaluating multiple types of organ toxicity, potentially helping researchers design safer medicines for patients.