Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userGaussian Mixture Based Enhanced Sampling
94
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
Many processes in chemistry and physics take place on timescales that cannot be explored using standard molecular dynamics simulations. This renders the use of enhanced sampling mandatory. Here we introduce an enhanced sampling method that is based on constructing a model probability density from which a bias potential is derived. The model relies on the fact that in a physical system most of the configurations visited can be grouped into isolated metastable islands. To each island we associate a distribution that is fitted to a Gaussian mixture. The different distributions are linearly combined together with coefficients that are computed self consistently. Remarkably, from this biased dynamics, rates of transition between different metastable states can be straightforwardly computed.
rate research
Read More
We propose a rigorous construction of a 1D path collective variable to sample structural phase transformations in condensed matter. The path collective variable is defined in a space spanned by global collective variables that serve as classifiers derived from local structural units. A reliable identification of local structural environments is achieved by employing a neural network based classification. The 1D path collective variable is subsequently used together with enhanced sampling techniques to explore the complex migration of a phase boundary during a solid-solid phase transformation in molybdenum.
We address the problem of the minus sign sampling for two electron systems using the path integral approach. We show that this problem can be reexpressed as one of computing free energy differences and sampling the tails of statistical distributions. Using Metadynamics, a realistic problem like that of two electrons confined in a quantum dot can be solved. We believe that this is a strategy that can possibly be extended to more complex systems.
The computational study of conformational transitions in RNA and proteins with atomistic molecular dynamics often requires suitable enhanced sampling techniques. We here introduce a novel method where concurrent metadynamics are integrated in a Hamiltonian replica-exchange scheme. The ladder of replicas is built with different strength of the bias potential exploiting the tunability of well-tempered metadynamics. Using this method, free-energy barriers of individual collective variables are significantly reduced compared with simple force-field scaling. The introduced methodology is flexible and allows adaptive bias potentials to be self-consistently constructed for a large number of simple collective variables, such as distances and dihedral angles. The method is tested on alanine dipeptide and applied to the difficult problem of conformational sampling in a tetranucleotide.
Forward flux sampling (FFS) provides a convenient and efficient way to simulate rare events in equilibrium or non-equilibrium systems. FFS ratchets the system from an initial state to a final state via a series of interfaces in phase space. The efficiency of FFS depends sensitively on the positions of the interfaces. We present two alternative methods for placing interfaces automatically and adaptively in their optimal locations, on-the-fly as an FFS simulation progresses, without prior knowledge or user intervention. These methods allow the FFS simulation to advance efficiently through bottlenecks in phase space by placing more interfaces where the probability of advancement is lower. The methods are demonstrated both for a single-particle test problem and for the crystallization of Yukawa particles. By removing the need for manual interface placement, our methods both facilitate the setting up of FFS simulations and improve their performance, especially for rare events which involve complex trajectories through phase space, with many bottlenecks.
Designing an appropriate set of collective variables is crucial to the success of several enhanced sampling methods. Here we focus on how to obtain such variables from information limited to the metastable states. We characterize these states by a large set of descriptors and employ neural networks to compress this information in a lower-dimensional space, using Fishers linear discriminant as an objective function to maximize the discriminative power of the network. We test this method on alanine dipeptide, using the non-linearly separable dataset composed by atomic distances. We then study an intermolecular aldol reaction characterized by a concerted mechanism. The resulting variables are able to promote sampling by drawing non-linear paths in the physical space connecting the fluctuations between metastable basins. Lastly, we interpret the behavior of the neural network by studying its relation to the physical variables. Through the identification of its most relevant features, we are able to gain chemical insight into the process.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
