Slicing and Dicing: Optimal Coarse-Grained Representation to Preserve Molecular Kinetics

Abstract

The aim of molecular coarse-graining approaches is to recover relevant physical properties of the molecular system via a lower-resolution model that can be more efficiently simulated. Ideally, the lower resolution still accounts for the degrees of freedom necessary to recover the correct physical behavior. The selection of these degrees of freedom has often relied on the scientist's chemical and physical intuition. In this article, we make the argument that in soft matter contexts desirable coarse-grained models accurately reproduce the long-time dynamics of a system by correctly capturing the rare-event transitions. We propose a bottom-up coarse-graining scheme that correctly preserves the relevant slow degrees of freedom, and we test this idea for three systems of increasing complexity. We show that in contrast to this method existing coarse-graining schemes such as those from information theory or structure-based approaches are not able to recapitulate the slow time scales of the system.

Perspectives on Ligand/Protein Binding Kinetics Simulations: Force Fields, Machine Learning, Sampling, and User-Friendliness

“…In all cases, we recommend dutifully validating the kinetic rates obtained via CG MD calculations against experimental data to adequately estimate the acceleration factor introduced by the CG FFs. These aspects and other limitations of classic CG FFs, such as their thermodynamics-based design and configurational-based CG mapping, could be overcome by ML techniques. ,,− While we expect that these FFs, and the Martini in particular, will continue to play a leading role in investigating many phenomena, they could be integrated or replaced by ML-guided CG FFs in the future. As discussed for the atomistic FFs, the potentiality of ML methods in parametrizing, optimizing, and including additional measurables, such as kinetic data sets, could significantly improve the accuracy of FFs and, therefore, cannot be overlooked.…”

“…ML FFs hold promise, but regrettably, they are not yet fully developed for general applications. They are facing essential challenges related to instabilities, representation issues, and limitations in transferability, particularly when applied to larger systems. ,, However, their rapid development cycle is encouraging, setting them apart from traditional FFs, with numerous new ML FFs being introduced annually. ,,,,, For instance, Fu et al have provided valuable insights into the limitations of current training methods for ML FFs, which are likely to inspire further advancements in the field . The atomistic ML FFs hold great promise in achieving quantum mechanics-level accuracy in MD simulations, with the potential to introduce greater flexibility in handling configurational and conformational variations in small molecules, peptides, and proteins.…”

Transferable Implicit Solvation via Contrastive Learning of Graph Neural Networks

“…Coarse-grained (CG) force fields play a pivotal role in advancing molecular dynamics (MD) simulations of biomolecules by bridging the gap between computational efficiency and biological realism. − In these simulations, biomolecules are represented with simplified models that group several atoms into a single interaction site, reducing the computational burden without sacrificing essential structural and dynamical information. This approach is crucial for studying large and complex biomolecular systems over biologically relevant timescales, which would be infeasible with atomistic simulations.…”

Statistically Optimal Force Aggregation for Coarse-Graining Molecular Dynamics

“…Numerous aspects of the coarse-graining procedure have been studied in depth; we refer readers to recent reviews for a comprehensive overview. 8,9,22 For example, work has extensively studied the influence of the atom-to-bead mapping, 13,30,[37][38][39][40][41][42][43][44][45][46] functional form of candidate potential, 11,12,14,15,[47][48][49][50][51] and other details of the fitting routine. 19,32,33,[52][53][54][55][56] However, to our knowledge no work has directly and systematically investigated the influence of the mapping that projects fine-grained (FG) forces to the CG resolution.…”