Data-driven synthesis of biodegradable cationic polymers using machine learning
Leverage a machine learning platform and Graph Neural Networks to accelerate the discovery of highly biodegradable cationic polymers. This solution predicts polymer properties, enabling high-throughput virtual screening and efficient lab synthesis, reducing traditional R&D timelines.
Overview
The proposed solution leverages a machine learning platform to revolutionize the discovery and synthesis of novel biodegradable cationic polymers. By using advanced Graph Neural Networks (GNNs), this approach predicts both deposition performance and biodegradability directly from polymer structures, enabling high-throughput virtual screening of extensive polymer libraries. This data-driven method is designed to minimize traditional R&D timelines by focusing lab efforts on the most promising candidates, ultimately leading to the development of innovative materials tailored to specific needs.
Technical specifications
The core of this solution is the use of Graph Neural Networks trained on both public and internal datasets to predict the properties of over 10,000 novel polymer candidates. Key features include:
- Virtual Screening: Efficiently identifies potential polymers using inherent hydrolyzable backbones like polyesters, polycarbonates, polyphosphoesters, and polyamides.
- Synthesis & Characterization: Top candidates will be synthesized using scalable methods (ROP, RAFT) and characterized with NMR, FTIR, GPC, DSC/TGA, and AFM.
- Iterative Learning Cycle: Experimental data from synthesized polymers is used to refine and improve the GNN models, ensuring iterative enhancement of model accuracy.
Technology readiness level
Currently at TRL 2, the solution is in the early stages of development. The plan involves a three-phase approach: ML design, synthesis and characterization, and validation and refinement, aiming to produce validated high-performance polymer candidates.
About University of North Dakota
The University of North Dakota is a comprehensive public research university in Grand Forks with a collaborative, industry‑oriented profile. Partners engage through an applied energy center offering bench‑ and pilot‑scale testing, a 50,000‑sq‑ft Tech Accelerator with wet‑lab suites, and a Center for UAS Research aligned with one of the nation’s largest university aerospace programs and dedicated test‑range access. Grand Forks sits inside a Northern Plains autonomous systems and energy cluster, giving companies access to airspace, infrastructure, and talent. Research is supported by competitive federal funding (e.g., NIH and NSF), and the medical school’s community‑based clinical network extends collaboration statewide. Commercialization is enabled by a dedicated tech transfer function and the Center for Innovation with incubation, SBIR/STTR support, and investor networks.
delivered to your inbox.