Greenstone Biosciences Advances AI-Driven Human Models for Drug Safety and Efficacy with NVIDIA

 Greenstone Biosciences today announced a collaboration with NVIDIA (NASDAQ: NVDA) to develop next-generation AI models that predict drug-induced toxicity using human-based cellular systems. This collaboration brings together Greenstone’s large-scale, iPSC-based experimental datasets with NVIDIA’s expertise in AI model development to move beyond traditional animal testing and accelerate safer, more precise drug discovery.

“This collaboration represents a shift toward truly human-centered preclinical testing,” said Joseph C. Wu, M.D., Ph.D., Co-Founder of Greenstone Biosciences and Director of the Stanford Cardiovascular Institute. “By integrating our iPSC-based cell systems with NVIDIA AI expertise, we can better capture how genetic differences across populations influence drug safety and response.”

This initiative integrates NVIDIA’s leading AI model development and Greenstone’s experimental dataset. The NVIDIA CodonFM model will be refined using Greenstone’s experimental datasets to better map gene-isoform expression and genotype–phenotype relationships linked to variable drug responses. These insights are also being used to explore ways to reverse disease phenotypes in patient-derived models, advancing precision and human-relevant therapeutics.

The collaboration aligns with the US Food and Drug Administration (FDA) roadmap, which outlines a strategic, stepwise shift to reduce animal testing in preclinical safety studies through the use of scientifically validated New Approach Methodologies (NAMs).

For the NVIDIA press release, click here.