Emergence of Structure and Function from Sequenceable Sequence-Defined Macrocyclic Oligourethanes

University Of Texas At Austin, Austin TX

Investigators

Abstract

With the support from the Chemical Catalysis (CAT) program and co-funding from the Chemical Synthesis (SYN) and Macromolecular, Supramolecular and Nanochemistry (MSN) programs in the Division of Chemistry, Professor Eric V. Anslyn of the University of Texas-Austin, is developing a general protocol for optimizing abiotic (non-natural) polymers for molecular recognition and catalysis. The chemical structure of natural peptides, polymers of alpha-amino acids, controls their chemical function as materials, catalytic entities, and information carriers. To develop non-natural analogues of peptides with alternative tunable reactivity, the Anslyn group will prepare macrocycles of urethanes (a common linkage) via automated methods, test their catalytic activity for the degradation of nerve agent surrogates, assess the fine structure of promising hits, and use a recursive strategy for optimization. Machine learning methods will also be integrated into the optimization protocol to link oligo-urethane fine structure information to desired reactivity and to establish fundamental connections. This carefully crafted data collection and analysis system is being used to determine how to guide the multivariable process for non-natural supramolecular catalyst synthesis to compete with and expand the activities of enzymes for the selective hydrolysis of V-agent surrogates. This program is further being used to integrate data science and automated synthesis concepts into two course-based undergraduate chemistry research programs run by Professor Anslyn, and to support research experiences for high school teachers to develop educational projects. While the field of supramolecular chemistry has had considerable success creating receptors, catalyst production has lagged considerably due to the challenges of multivariable optimization. Professor Anslyn and his research team are working toward addressing this gap in supramolecular chemistry by creating a new approach to supramolecular catalyst optimization. An integrated protocol is being developed that combines: 1) the computer-controlled synthesis of macrocyclic oligourethanes from a curated group of monomers, 2) analysis of the activity of these foldamer-type catalysts for the hydrolysis of nerve agent surrogates, 3) sequencing routines to elucidate oligourethane structure, and 4) recursive optimization using machine learning techniques. Algorithms involving partial least squares regression (PLSR) are being trained on sequence space and CD spectroscopy to predict which monomers, and their synergy, lead to improved binding and catalysis. Based upon the ML predictions, a script will be used to reprogram the synthesizer, allowing recursive cycling through synthesis, screening, and sequencing to optimize catalysis using an automated workflow. These activities will be used to train a diverse group of graduate and undergraduate researchers in the Anslyn group and further supporting course-based research experiences for undergraduates and educational research experiences for high-school teachers. 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.