Schatz Subproject - Convergence: RAISE: Auto-regulatory Scaffolds for Directed Evolution of Non-living Functional Materials

Directed evolution mimics the process of natural selection for evolving biomolecules toward a user-defined task. Although proteins or nucleic acids can be identified based on several rounds of selection, this approach cannot be used to evolve non-living materials for targeted applications because there is no systems-level regulatory host. This Growing Convergence RAISE proposal aims to design and realize an evolutionary host for non-living multifunctional materials based on auto-regulatory scaffolds. In such scaffolds, well-defined sensor and stimulation elements will be used for closed-loop screening of unknown materials, similar to cellular machinery in directed evolution. Realization of a non-living materials screening platform—an evolutionary host based on auto-regulatory scaffolds that incorporates synthesis, screening and feedback—that can even begin to approach the efficient, exquisite regulation in directed evolution would represent a paradigm-shift in materials discovery. The materials will be evolved directly under complex conditions needed for user-defined applications This proposed work not only intersects the NSF 10 Big Ideas but also pushes beyond the current scope. Intellectual Merit The team’s proposed materials discovery concept with negative auto-regulatory scaffolds and the development of universal principles for systems-level screening requires the integration of materials chemistry, optics, mechanics, chemical and control theory, machine learning, tissue engineering, and systems biology. Major goals of the proposal include: (1) designing and realizing auto-regulatory scaffold hosts for choices of sensing and regulation components and materials variants. Porous deformable scaffolds that contain sensing and regulatory elements will be constructed. Unknown and known materials can be soft or hard materials that undertake the user-defined sensing, signal transduction, and regulation functions. (2) Demonstrating auto-regulatory function from optical evolutionary scaffold hosts and use machine learning to screen new composite materials. Performance metrics to screen material variants until a systems-level auto-regulatory output is achieved will be determined. Metrics will be analyzed with machine-learning methods to optimize and predict performance limits. (3) Developing single component and multiscale models for auto-regulatory scaffolds. A multi-scale theory framework to describe materials evolutionary host architectures and top-down modeling of network components that lead to auto-regulation will be developed. Broader Impacts Directed evolution in non-living systems, based on auto-regulatory scaffold hosts, will result in the identification of materials and materials combinations for a diverse range of applications, from absorption-impact materials to dehumidifying coatings to stem-cell differentiation fate. The team’s educational objectives involve: (1) training graduate students and postdocs in a convergence approach to materials screening based on auto-regulatory circuits and machine learning and (2) developing outreach programs including public lectures, undergraduate courses, and published Perspectives that combine this convergence of research from different fields.