(362b) Engineering Peptides through Molecular Simulations, Machine Learning and Optimization Methods for Biological and Clean Energy Applications

Polypeptides are one of the most common and essential biomacromolecules for living systems. They have also become excellent candidates for a variety of engineering and biological applications due to their unique and tunable physical and biochemical properties. This talk demonstrates how a chemical engineer approaches and overcomes the challenges in the peptide engineering field, which include two essential aspects, fundamental mechanisms and end applications. First, we focus on combining molecular simulation and statistical mechanics to understand the thermodynamic consequence (phase transition and self-assembly) arising from complex molecular interactions between constituent building blocks spanning certain sequences. Quantitative phase diagrams are calculated for Alzheimer’s disease-related amyloid-forming peptides, and the results are validated by experimental measurements. The development of molecular models and investigations of fluid phase behavior for another common class of biomolecules such as phospholipid and small chiral molecules are also included. Second, we employ machine learning, and optimization methods to leverage our mechanistic understanding of peptide structure-activity relationship to engineer peptides for three multidisciplinary applications: (a) Design charge complementary co-assembly peptide pair (CATCH) that forms functional peptide nanofibers; (b) Design lanthanide-binding tag (LBT) peptides for green and selective separation of rare earth elements; (c) Design hybridized antimicrobial peptides (hAMP) against gram-negative and gram-positive bacteria.