When Data Are Lacking: Physics-Based Inverse Design of Biopolymers Interacting with Complex, Fluid Phases

Area of Science:

• Biomolecular research
• Computational biophysics
• Protein engineering

Background:

• Traditional biomolecular research focuses on understanding mechanisms before controlling function.
• This conventional approach is time-consuming and may not yield direct functional control strategies.
• An alternative is needed to directly design functional biomolecules.

Purpose of the Study:

• To introduce and elucidate physics-based inverse design for biopolymer engineering.
• To demonstrate the application of evolutionary algorithms and coarse-grained simulations for designing peptides with specific functions.
• To explore the potential for creating novel peptide-based sensors and drugs.

Main Methods:

• Utilized evolutionary molecular dynamics (Evo-MD) simulations, combining evolutionary algorithms with the Martini coarse-grained force field.
• Directed evolution from random sequences to peptides interacting with complex fluid phases like lipid membranes.
• Employed physics-based evolution to analyze protein-lipid interactions and generate training data for predictive models.

Main Results:

• Successfully directed the evolution of peptides towards specific interactions with complex fluid phases.
• Demonstrated the ability to tailor peptides for targeting attributes like membrane curvature and lipid composition.
• Showcased the potential to extract evolutionary optimization fingerprints from native proteins.

Conclusions:

• Physics-based inverse design offers a powerful, direct route to engineer functional biopolymers.
• Evo-MD simulations provide a viable method for designing peptides for applications in sensors and therapeutics.
• This approach isolates key physicochemical principles and thermodynamic drivers for biopolymer interactions.