Coarse-Grained Dynamics Simulations

Coarse-grained molecular dynamics simulation is a simulation method between all-atom simulation and mesoscopic simulation. Compared with all-atom simulation, it can simulate large-scale dynamic behavior; compared with mesoscopic simulation, it can reflect more microscopic levels. Therefore, coarse-grained molecular dynamics simulation compensates for their respective shortcomings to a certain extent, and is a "bridge" between all-atom simulation and mesoscopic simulation.

Overall solutions

Coarse-grained simulation aims to simulate the behavior of complex systems using its coarse-grained (simplified) representation. Coarse-grained models are widely used in molecular modeling of biomolecules of various particle sizes. Various coarse-grained models have been proposed. They are usually dedicated to the computational modeling of specific molecules: proteins, nucleic acids, lipid membranes, carbohydrates or water. In these models, molecules are not represented by individual atoms, but by "pseudo atoms" that approximate groups of atoms, such as entire amino acid residues. By reducing the degrees of freedom, longer simulation times can be studied at the expense of molecular details. Coarse-grained models have found practical applications in molecular dynamics simulations.

Features

• In an all-atom simulation, the simulation of a coarse-grained (CG) system requires fewer resources and runs faster than the same system. It can achieve an order of magnitude increase in the time and length of the simulation.
• In the all-atom representation, the simulation of a coarse-grained (CG) system requires fewer resources and is faster than the simulation of the same system.

Figure 1. Illustration of replica exchange molecular dynamics (REMD) method.

Algorithm

• Residue-Based Coarse Graining
  

  In the latter case, the time step is limited to 1-2 fs, but our CG simulation can be run with a time step of 25-50 fs (because the particle mass is larger and the interaction potential is smoother). The possible speedup depends on the computer power supply at hand. For example, simulating an all-atomic system with 300,000 particles on 48 processors, we obtain dynamics of 0.1 ns in one day. The same system in the CG representation contains 30,000 particles, and the performance achieved in one day is 150 ns. This is because the number of particles per processor is small and the integration time step is large. Therefore, the speedup in this case is 1500 times.

  
  
• Shape-Based Coarse Graining
  The shape-based coarse-grained design is used to simulate the large-scale movement of macromolecular assemblies. These movements represent proteins and other biological molecules, and the fewer dotted particles, the better. Biomolecules, especially proteins, have many shapes. Generally, a single protein contains both compact domains and disordered and elongated linkers or tails. The shape-based coarse-grained method enables people to take advantage of the efficient topology preservation algorithms originally developed for neural computing. In this protocol, the versatile package VMD is used for molecular modeling and structure visualization. VMD supports computers running MacOS X, Unix, or Windows systems, and is distributed free of charge.

AI-based approaches

CD ComputaBio established four different AI-based regression models: artificial neural networks, k-nearest neighbors, Gaussian process regression and random forests to map coarse-grained models to all atomic models. The model shows better predictions than existing reverse mapping methods for selected structures, which indicates the application of our model in reverse mapping.

Why choose us?

• We can provide many different coarse-grained simulation services, including proteins, nucleic acids, water molecules, etc
• The calculation cycle is short and fast, the calculation accuracy is high, and the required funds are far less than biological or chemical experiments.
• Mature and reliable molecular simulation platform: rich project experience, participating in major projects and releasing advanced cooperation results.
• Based on high-quality data analysis, the pre-research and verification analysis content and methods are indeed mature and feasible.