Nvidia deepens drug discovery with new partnerships with Lilly, Thermo Fisher and more
AI giant Nvidia kicked off this year's JPMorgan Healthcare Conference on Monday by announcing new and expanded partnerships aimed at embedding artificial intelligence more deeply into drug discovery and drug research.
Most notably, Nvidia is partnering with Eli Lilly and Company to establish a joint innovation lab in South San Francisco. The lab's overarching goal is to accelerate drug discovery by applying advanced artificial intelligence models to long-standing biotech challenges, the partners said.
The lab will bring together Eli Lilly's biology, chemistry and medical experts with Nvidia's artificial intelligence engineers to create more powerful artificial intelligence models that can speed up the identification and validation of new drug molecules. The focus will be on building a “continuous learning system” that connects Lilly's wet labs with computational dry labs so that AI can assist around the clock and iteratively improve experiments and model development.
“Combining our vast data and scientific knowledge with Nvidia's computing power and model-building expertise can reshape drug discovery as we know it. By bringing together world-class talent in a startup environment, we are setting the stage for breakthroughs that neither company can achieve alone,” Eli Lilly CEO David Ricks said in a statement.
In addition to drug discovery, the collaboration aims to explore the use of artificial intelligence, robotics, digital twins and advanced computing in Lilly's clinical development, manufacturing and commercial operations.
The laboratory is expected to be operational early this year. Nvidia and Eli Lilly plan to spend $1 billion over the next five years on the lab's talent, infrastructure and computing, according to a press release.
BioNeMo—Nvidia’s generative artificial intelligence platform built for drug discovery—will serve as the lab’s primary software infrastructure. In addition to its biotech collaboration with Eli Lilly and Company, Nvidia said it is working with a growing number of drugmakers to integrate BioNeMo into their lab infrastructure and close the loop between computational AI workflows and real-world experiments.
The open development platform is designed to help researchers identify promising drug candidates more quickly and understand how they might behave, reducing trial and error in the lab, thereby turning large amounts of scientific data into actionable insights.
Nvidia announced that it has improved the platform with new features, including updated RNA structure models and drug synthesis utility, as well as new data processing tools.
The company also said it has begun working with Thermo Fisher Scientific to make biotech research labs more automated so scientists don't have to manually run, monitor and analyze each experiment.
By connecting Thermo Fisher's lab instruments with Nvidia's AI computing, the partners are setting out to create labs that can plan experiments, run them, check for errors, and then analyze the results with minimal human oversight. The system will be designed to process results as experiments occur and decide what to do next, rather than keeping data in silos or waiting for researchers to review it.
Nvidia's life sciences efforts aren't limited to software and analytics, though—the company is also pushing AI deeper into physical labs through automation and robotics.
The company detailed how robotics startup Multiply Labs is using its Omniverse and Isaac simulation tools to build robotic systems that can help produce complex cell therapies faster and cheaper by automating work currently done manually. Nvidia says its tools enable Multiply to create digital replicas of lab environments and virtually train robots to perform delicate tasks with high precision before touching real materials.
The partners say more such collaborations could scale up production of advanced treatments and make them more accessible.
Overall, Nvidia's announcement on Monday underscores how the company is positioning itself as a core technology provider across the pharmaceutical R&D pipeline.
Photo: Bing-Jhen Hong, Getty Images
