BPS2026 – Developing a Martini Gold model for use in biological contexts

The success of the Martini force field is largely due to its ability to reproduce experimental phenomena, its extensive library of biological molecules, and the abundance of tools available to further develop models of small molecules. However, many inorganic materials cannot be investigated in Martini because they have no direct correspondence to a Martini bead type. One such material is gold, making it infeasible to investigate gold-protein interactions, ligand-coated gold nanoparticle aggregation, and small molecule binding to gold surfaces in Martini. To bridge this gap, we present Martini Gold: a gold model and strategy for parameterizing metals in the Martini force field. We use atomistic simulations as a foundation for the Martini Gold model. Our goal was to maintain gold cluster size, stability, and interaction energy between the coarse-grained and atomistic systems. To achieve this goal, we determined the number of gold atoms that corresponds to a Martini S-bead, matched atomistic and coarse-grained gold-gold dimerization free-energies in oil and water, and matched atomistic and coarse-grained small molecule adsorption free energies to gold surfaces. We demonstrate the model’s accuracy by comparing coarse-grained simulations of gold nanoparticle aggregation in lipid membranes with experimental data. Based on the gold parameterization scheme, we designed a workflow to parameterize new metals in Martini.