ROADMAP: Reflectometry-driven Optimization And Discovery of Membrane Active Peptides​

Summary

ROADMAP is a collaborative, multi-year project to map the interactions between peptides of arbitrary sequence and lipid bilayer membranes. ROADMAP will leverage a combination of NIST’s next-generation neutron instrumentation with cutting-edge AI/ML methods to unveil the structural biology of peptide/membrane interactions. The ultimate goal is to enable rapid discovery and optimization of peptides with desired membrane-targeting properties, with application to antimicrobial therapeutics, drug delivery, and novel vaccine formulations. 

Description

ROADMAP presents the opportunity to pursue collaborative, innovative research at the intersection of neutron measurement science, structural biology for public health, and AI/ML for autonomous measurements. To inquire about available positions please contact David Hoogerheide (https://www.nist.gov/people/david-p-hoogerheide) (neutron measurement science), Ella Mihailescu (https://www.ibbr.umd.edu/profiles/ella-mihailescu) (peptide biology), or Antonio Cardone (https://www.nist.gov/people/antonio-cardone) (AI/ML and autonomous measurement). 

 

ROADMAP will be a measurement platform for the rapid discovery and optimization of therapeutically relevant peptides targeting the lipid membrane. The biological expertise is provided by researchers from the NIST / University of Maryland joint Institute for Bioscience and Biotechnology Research (https://www.ibbr.umd.edu/). Neutron reflectometry (NR) will be the primary method for obtaining structural information on membrane-bound peptides. This technique can structurally resolve, for example, membrane-disrupting peptides from peripherally associated membrane-thinning peptides. However, NR has been traditionally limited to single-shot experiments, which are unsuitable for data-driven approaches. Therefore, we will develop high-throughput, autonomous experimentation neutron reflectometry capabilities that leverage the next-generation neutron instrumentation at the NIST Center for Neutron Research (https://www.nist.gov/ncnr). After completion of this milestone, an initial experimental campaign will produce larger than 1000 membrane-bound structures of therapeutically relevant antimicrobial cell-penetrating peptides. Researchers from the NIST Information Technology Laboratory will develop a self-learning AI that guides this measurement campaign. The AI will consist of an ML model that accurately predicts the interaction of such peptides with lipid bilayer model membranes under pharmaceutically relevant conditions. Furthermore, based on the prediction outcomes, an AI metrology-based component will continuously suggest target peptides for the next NR measurement that most enhance the predictive power of the ML model.