Improving conformational ensembles of folded proteins in GōMartini

The Martini coarse-grained (CG) force field enables efficient simulations of biomolecular systems but cannot reliably maintain folded protein structures. To stabilize proteins during simulation, Martini is typically combined with structure-based force fields such as elastic network models (ENMs) or Gō models. While these approaches preserve global folds and capture protein flexibility, their ability to reproduce conformational dynamics remains unclear. Here, we benchmark Martini combined with ENMs or Gō models on three folded proteins and show that both approaches struggle to sample the conformational space observed in atomistic simulations, even when uniform interaction strengths or equilibrium bond distances are adjusted. This limitation arises from the assumption of a uniform interaction network, in which all bond energies are equal. To overcome this, we present a fully automated, perturbation-based optimization approach for Gō networks, PoGō, that iteratively refines a non-uniform Gō network against a pre-computed atomistic free energy landscape in essential conformational space. Our approach converges rapidly, yielding CG ensembles in close agreement with reference atomistic simulations. As a cross-validation, the optimization also improves the root-mean-square fluctuation profile.