Rapid sampling of all-atom peptides using a library-based polymer-growth approach

Area of Science:

• Computational chemistry
• Molecular dynamics
• Biophysics

Background:

• Accurate molecular simulations are crucial for understanding peptide behavior.
• Existing methods for equilibrium sampling can be computationally intensive.

Purpose of the Study:

• To develop an efficient method for equilibrium sampling of peptides using atomistic force fields.
• To leverage precalculated molecular fragment libraries for faster simulations.

Main Methods:

• Adapted polymer growth strategies combined with precalculated statistical libraries of molecular fragments.
• Ensembles were generated by combining fragment configurations and reweighted to match the Boltzmann distribution.
• Applied to small peptides (4-8 residues) and larger peptides (up to 32 residues).

Main Results:

• Achieved high-quality equilibrium sampling for small peptides in under an hour.
• Generated approximate, clash-free ensembles for larger peptides in under a minute.
• Demonstrated significant speed improvements over traditional Langevin simulations.

Conclusions:

• The novel fragment-based growth method offers a computationally efficient alternative for peptide equilibrium sampling.
• This approach has potential applications in free energy calculations, protein structure prediction, and multi-resolution sampling.