Artificial intelligence and laboratory automation are reshaping how researchers design new therapeutic materials. Scientists at the University of Toronto have developed an AI-powered self-driving laboratory platform that accelerates the discovery of lipid nanoparticles used for mRNA delivery.
The system, known as LUMI-lab (Large-scale Unsupervised Modeling followed by Iterative experiments), integrates artificial intelligence, robotics, and active learning to design and test candidate molecules with minimal human intervention. In a study published in Cell, the platform identified a previously unrecognized class of lipid molecules that improve the efficiency of delivering mRNA into human cells.
Self-driving laboratory systems combine machine learning models with automated experimental equipment to create closed-loop workflows. In these systems, AI algorithms predict promising molecular candidates, robotic systems synthesize and test them, and the resulting data feed back into the model to refine subsequent experiments.
Self-driving laboratory discovers new mRNA delivery lipids
LUMI-lab integrates a molecular foundation model with automated robotics to synthesize and evaluate lipid nanoparticles used in mRNA therapeutics. Lipid nanoparticles (LNPs) act as delivery vehicles that protect fragile mRNA molecules and transport them into target cells.
Over 10 active-learning cycles, the system synthesized and evaluated more than 1,700 lipid nanoparticle formulations, exploring chemical designs that had not previously been associated with mRNA delivery performance.
During these experiments, the system discovered that brominated lipid tails significantly improved transfection efficiency, a key measure of how effectively genetic material enters cells.
“Across 10 active-learning cycles, LUMI-lab synthesized and tested more than 1,700 new lipid nanoparticles, uncovering brominated-tail ionizable lipids that deliver mRNA into human lung cells more efficiently than approved benchmarks,” said Bowen Li, GSK chair in pharmaceutics and drug delivery at the University of Toronto’s Leslie Dan Faculty of Pharmacy and an affiliate scientist at Princess Margaret Cancer Centre.
Li noted that the AI system identified bromination as an important molecular design feature without prior guidance.
“The key advance of this AI-driven system is that it independently identified bromination as an important, meaningful design feature without prior hypothesis or researchers telling it to look for it first,” he said.
Expanding the design space for mRNA therapeutics
mRNA therapeutics represent one of the fastest-growing areas of pharmaceutical development, but their effectiveness depends heavily on delivery technologies. Lipid nanoparticles remain the primary platform for safely delivering mRNA into cells.
To date, only three lipid nanoparticle formulations have received FDA approval, underscoring the need for new materials to improve delivery efficiency and expand therapeutic applications.
To address data limitations in emerging areas such as mRNA delivery, researchers pretrained the LUMI model on more than 28 million molecular structures. This large-scale molecular pretraining allowed the system to recognize broader chemical patterns before refining predictions through iterative experimentation.
“When integrated into an active learning framework, the model can be continuously optimized in a closed-loop workflow, further enhancing its predictive accuracy,” Li said.
AI and robotics accelerate materials discovery
The study found that although brominated lipids represented only eight percent of the chemical library evaluated by the system, they accounted for more than half of the top-performing lipid nanoparticle candidates.
Advanced Lab Management Certificate
The Advanced Lab Management certificate is more than training—it’s a professional advantage.
Gain critical skills and IACET-approved CEUs that make a measurable difference.
Several of the newly identified lipids outperformed the lipid used in Moderna’s COVID-19 mRNA vaccine in preclinical experiments. Researchers also reported that the brominated lipids demonstrated safety profiles similar to existing clinical lipid materials.
The team now plans to expand the LUMI-lab platform to optimize multiple performance characteristics simultaneously, including safety, tolerability, and tissue targeting.
“Next, we’re expanding LUMI-lab to optimize multiple clinically relevant properties at once, not just delivery potency but also safety, tolerability, and tissue selectivity,” Li said. “By closing the loop between AI predictions and automated experiments, we aim to shorten the design cycle for new lipid materials and open up a much larger, evidence-driven chemical space for mRNA therapeutics.”
For laboratory organizations, platforms such as LUMI-lab highlight how integrating AI with automated laboratory systems may accelerate molecular discovery workflows and expand the range of materials available for emerging therapeutic technologies.
This article was created with the assistance of Generative AI and has undergone editorial review before publishing.
