How Well Do Molecular Dynamics Force Fields Model Peptides? A Systematic Benchmark Across Diverse Folding Behaviors

Area of Science:

• Computational chemistry
• Biophysics
• Drug discovery

Background:

• Linear peptides are crucial in biology and drug discovery, often mediating protein-protein interactions.
• Peptide structural plasticity presents challenges for accurate molecular simulations.
• Fixed-charge force fields are widely used but their performance for peptides is not well-established.

Purpose of the Study:

• To benchmark twelve popular and emerging fixed-charge force fields for simulating diverse linear peptides.
• To assess force field performance in capturing peptide stability, folding behavior, and structural biases.
• To provide guidance for peptide modeling and inform future force field development.

Main Methods:

• Molecular dynamics simulations of twelve peptides (structured, context-sensitive, disordered) using ten popular and two emerging fixed-charge force fields.
• Simulations initiated from both folded (200 ns) and extended (10 μs) states.
• Analysis of structural stability, folding dynamics, and force field-specific biases.

Main Results:

• Significant variations in performance were observed across the tested force fields.
• Some force fields showed strong structural biases, while others allowed for reversible conformational fluctuations.
• No single force field demonstrated optimal performance across all peptide types and simulation conditions.
• Limitations were identified in balancing intrinsic disorder and secondary structure prediction.

Conclusions:

• Current fixed-charge force fields exhibit limitations in accurately modeling the conformational landscape of linear peptides.
• The study provides a benchmark for evaluating peptide simulation methodologies.
• Results offer practical insights for selecting appropriate force fields for specific peptide modeling tasks and highlight areas for improvement in force field development.