Using AI to optimize hydrogen fuel production and reduce environmental impact: Worcester Polytechnic Institute research published in Nature Chemical Engineering
Peer-Reviewed Publication
Updates every hour. Last Updated: 6-Oct-2025 17:11 ET (6-Oct-2025 21:11 GMT/UTC)
Researchers at Harvard and Northwestern have developed a machine learning method that can design intrinsically disordered proteins with custom properties, addressing nearly 30% of all human proteins that are currently out of reach of AI tools like AlphaFold. The new approach uses automatic differentiation, traditionally a deep learning tool, to optimize protein sequences for desired properties.
MIT researchers developed a framework that can help engineers design systems that involve many interconnected parts in a way that explicitly accounts for the uncertainty in each component’s performance. This could lead to electronic devices like drones and robots that are more robust and reliable in unpredictable, real-world situations.