Project Grant 2154447

The National Science Foundation Division of Chemistry awarded $199,252 under the Mathematical and Physical Sciences (CFDA 47.049) federal grant program to Carnegie Mellon University to support collaborative research between Professors Jian Lin and Shih-Kang Chao of the University of Missouri-Columbia and Olexandr Isayev of Carnegie Mellon University. The researchers will develop and apply a novel data-driven architecture to design new molecules with desired physical and chemical properties, combining generative modeling, reinforcement learning, and active learning algorithms. This methodology is intended to provide a general approach to the long-standing challenge of property-objective inverse molecular design. The project aims to improve understanding of molecular representations, explore novel chemical space beyond simple optimization of existing molecules, and elucidate how generative models learn chemical principles. The designed novel molecules with multiple optimized physicochemical, electronic, optical, and redox properties have the potential to transform applications in medicine, photovoltaics, catalysis, thermal storage, and organic redox flow batteries.