The Automated Optimisation of a Coarse-Grained Force Field Using Free Energy Data

Abstract

Atomistic models provide a detailed representation of molecular systems, but are sometimes inadequate for simulations of large systems over long timescales. Coarse-grained models enable accelerated simulations by reducing the number of degrees of freedom, at the cost of reduced accuracy. New optimisation processes to parameterise these models could improve their quality and range of applicability. We present an automated approach for the optimisation of coarse-grained force fields, by reproducing free energy data derived from atomistic molecular simulations. To illustrate the approach, we implemented hydration free energy gradients as a new target for force field optimisation in ForceBalance and applied it successfully to optimise the un-charged side-chains and the protein backbone in the SIRAH protein coarse-grain force field. The optimised parameters closely reproduced hydration free energies of atomistic models and gave improved agreement with experiment.

Accurate Sampling of Macromolecular Conformations Using Adaptive Deep Learning and Coarse-Grained Representation

“…In this study, we resort to an intermediate approach in which we describe the system using a high-resolution coarse-grained model based on the SIRAH concept . This representation is a compromise between a fully atomistic and low-resolution coarse-grained model, which comes with its own drawbacks; low resolution coarse-grained models tend to be less accurate in modeling the interactions within the system and smoothen the energy landscape thereby reducing the energy barriers between different states which can affect the calculations of both thermodynamic and dynamic properties . Furthermore, a backmapping from coarse-grained to fully atomistic models is typically not unambiguous, and therefore, may result in inaccurate density estimations of the molecular systems in the backmapped fully atomistic representations.…”

Research on Power Grid Over-voltage Anomaly Data Management Based on the Improved Clustering Algorithm

“…For example, the low initial data accuracy causes abnormal data in subsequent processing. Or the coarsegrained data aggregation method [7] is used in the statistical process of some over-voltage data, resulting in fluctuations in the abnormal values. Therefore, it is necessary to classify and manage the grid overvoltage anomaly data according to its specific causes and characteristics.…”

“…𝑓(𝑎, 𝑏) 𝑖𝑗 = 𝑓(𝑎, 𝑏) 𝑗𝑖 −1 (7) Where 𝑎 represents the growth rate of abnormal data storage, 𝑏 represents the growth of abnormal data storage time, and 𝑖𝑗 represents the length of the time block chain. The consistency between the behavior of anomalous data management and the actual anomalous data management can be solved by the following equation.…”

Assessing SIRAH’s Capability to Simulate Intrinsically Disordered Proteins and Peptides

“…Indeed, recent modifications to the solvation energy performed by two independent groups showed significant improvements in the description of the IDP αsynuclein 28 and hydration free energies. 29 Our recently published version 2.0 implemented a series of modifications that allowed for closer water-backbone distances and higher flexibility, granting a speed-up of nearly 120 times over atomistic simulations. 30 As a result, SIRAH2.0 showed typical Root Mean Square Deviations (RMSDs) for a set of well-folded proteins of around 0.3−0.4 nm.…”

“…Therefore, conformational biases could not be ruled out. Indeed, recent modifications to the solvation energy performed by two independent groups showed significant improvements in the description of the IDP α-synuclein and hydration free energies …”