NSF-MUR: FET: Small: Harnessing AI for Design and Simulation of Nucleic Acid Nanodevices

This project will harness the methods of artificial intelligence (AI) to design structures and tools for bionanotechnology. Bionanostructures are small, artificial self-assembled devices that can perform tasks at nanoscale. Like a robot in a factory combining materials to build a car, these nanostructures can put together materials such as proteins, gold nanoparticles, and DNA molecules with promising applications in diagnostics, therapeutics and material science. Currently, the design of bionanostructures is a tedious iterative process, often based on costly and time-consuming trial and error approaches. This project will develop novel techniques based on generative AI algorithms to automate design and characterization of these complex nanodevices built out of DNA. It will thus allow construction of more complex devices as well as scaling up and simplifying the process, thus enabling large-scale manufacturing of new types of bionanotechnology for a variety of applications. The overarching goal of this project is to harness generative AI methods for automated design of nucleic acid nanostructures and experimentally verify them by realizing nanoscale devices. To speed up the design process, this project will introduce the following applications of AI into the bionanotechnology field: 1) Develop new methods to speed-up computational characterization of nucleic acid nanostructures based on generative deep neural network architectures trained on data from coarse-grained model of nucleic acids simulations; 2) develop reinforcement learning-based algorithms for automated design of nanostructures with feedback from simulated or experimental environment; and 3) enable human-language prompting and interactive nanostructure design using large language modeling tools trained on datasets collected through our design software. The project will create training opportunities for undergraduate and graduate students and develop AI-based tools for a biochemistry education program. It will also create easy-to-use interactive generative design tools for outreach events for the general public. This is a project jointly funded by the National Science Foundation and the Italian Ministry of Universities and Research (MUR) via the NSF-MUR Lead Agency Opportunity on Artificial Intelligence, where NSF funds the U.S. investigator and MUR funds the partner(s) in Italy. The U.S. investigator is supported by the Foundations of Emerging Technologies program and the Office of International Science and Engineering. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.