Changing Your Martini Can Still Give You a Hangover

Area of Science:

• Computational biophysics
• Molecular dynamics simulations
• Force field development

Background:

• Coarse-grained (CG) models simplify complex molecular systems for large-scale simulations.
• The Martini force field is widely used for biomolecular simulations.
• Martini 3.0 was developed to improve transferability in CG models.

Purpose of the Study:

• To assess the accuracy of the Martini 3.0 force field in representing thermodynamic and structural properties.
• To compare Martini 3.0's performance against atomistic simulations and statistical mechanics principles.
• To investigate the entropy-enthalpy partitioning in Martini 3.0 potentials of mean force (PMFs).

Main Methods:

• Decomposition of potentials of mean force (PMFs) into entropic and enthalpic components for Martini 3.0 and atomistic lipid bilayers.
• Application of the reversible work theorem to lateral pair correlation functions at various temperatures.
• Analysis of the undulation spectrum of Martini 3.0 bilayers.

Main Results:

• Martini 3.0 fails to accurately partition entropy and enthalpy in PMFs compared to mapped all-atom results.
• The Martini 3.0 force field produces an incorrect undulation spectrum, particularly at intermediate length scales.
• Despite improvements from Martini 2.0, Martini 3.0 still exhibits significant inaccuracies.

Conclusions:

• The development of accurate top-down CG models like Martini requires careful consideration of entropy-enthalpy contributions.
• Temperature-dependent terms in CG force fields may be necessary but not sufficient for improved accuracy.
• Further refinements are needed for Martini 3.0 to accurately represent biophysical properties of lipid bilayers.