scholarly journals A Deep Graph Network–Enhanced Sampling Approach to Efficiently Explore the Space of Reduced Representations of Proteins

2021 ◽  
Vol 8 ◽  
Author(s):  
Federico Errica ◽  
Marco Giulini ◽  
Davide Bacciu ◽  
Roberto Menichetti ◽  
Alessio Micheli ◽  
...  
Keyword(s):  
Md Simulations ◽  
Computer Architectures ◽  
Learning Approach ◽  
Enhanced Sampling ◽  
Algorithmic Approach ◽  
Reduced Representation ◽  
Structured Input ◽  
Reduced Representations ◽  
Speedup Factor ◽  
Automated Methods

The limits of molecular dynamics (MD) simulations of macromolecules are steadily pushed forward by the relentless development of computer architectures and algorithms. The consequent explosion in the number and extent of MD trajectories induces the need for automated methods to rationalize the raw data and make quantitative sense of them. Recently, an algorithmic approach was introduced by some of us to identify the subset of a protein’s atoms, or mapping, that enables the most informative description of the system. This method relies on the computation, for a given reduced representation, of the associated mapping entropy, that is, a measure of the information loss due to such simplification; albeit relatively straightforward, this calculation can be time-consuming. Here, we describe the implementation of a deep learning approach aimed at accelerating the calculation of the mapping entropy. We rely on Deep Graph Networks, which provide extreme flexibility in handling structured input data and whose predictions prove to be accurate and-remarkably efficient. The trained network produces a speedup factor as large as 105 with respect to the algorithmic computation of the mapping entropy, enabling the reconstruction of its landscape by means of the Wang–Landau sampling scheme. Applications of this method reach much further than this, as the proposed pipeline is easily transferable to the computation of arbitrary properties of a molecular structure.

scholarly journals Making high-dimensional molecular distribution functions tractable through Belief Propagation on Factor Graphs

2021 ◽  
Author(s):  
Zachary Smith ◽  
Pratyush Tiwary
Keyword(s):  
Belief Propagation ◽  
Conformational Dynamics ◽  
Joint Probability ◽  
Md Simulations ◽  
Distribution Functions ◽  
Factor Graph ◽  
High Dimensional ◽  
Enhanced Sampling ◽  
Small Peptides ◽  
Approximate Probability

Molecular dynamics (MD) simulations provide a wealth of high-dimensional data at all-atom and femtosecond resolution but deciphering mechanistic information from this data is an ongoing challenge in physical chemistry and biophysics. Theoretically speaking, joint probabilities of the equilibrium distribution contain all thermodynamic information, but they prove increasingly difficult to compute and interpret as the dimensionality increases. Here, inspired by tools in probabilistic graphical modeling, we develop a factor graph trained through belief propagation that helps factorize the joint probability into an approximate tractable form that can be easily visualized and used. We validate the study through the analysis of the conformational dynamics of two small peptides with 5 and 9 residues. Our validations include testing the conditional dependency predictions through an intervention scheme inspired by Judea Pearl. Secondly we directly use the belief propagation based approximate probability distribution as a high-dimensional static bias for enhanced sampling, where we achieve spontaneous back-and-forth motion between metastable states that is up to 350 times faster than unbiased MD. We believe this work opens up useful ways to thinking about and dealing with high-dimensional molecular simulations.

scholarly journals Identification of rare Lewis oligosaccharide conformers in aqueous solution using enhanced sampling molecular dynamics

2017 ◽  
Author(s):  
Irfan Alibay ◽  
Kepa K. Burusco ◽  
Neil J. Bruce ◽  
Richard A. Bryce
Keyword(s):  
Molecular Dynamics ◽  
Aqueous Solution ◽  
Md Simulations ◽  
Umbrella Sampling ◽  
Biological Action ◽  
Sialyl Lewis X ◽  
Enhanced Sampling ◽  
Sialyl Lewis ◽  
Aqueous Solvent ◽  
A Minor

<p>Determining the conformations accessible to carbohydrate ligands in aqueous solution is important for understanding their biological action. In this work, we evaluate the conformational free energy surfaces of Lewis oligosaccharides in explicit aqueous solvent using a multidimensional variant of the swarm-enhanced sampling molecular dynamics (msesMD) method; we compare with multi-microsecond unbiased MD simulations, umbrella sampling and accelerated MD approaches. For the sialyl Lewis A tetrasaccharide, msesMD simulations in aqueous solution predict conformer landscapes in general agreement with the other biased methods and with triplicate unbiased 10 ms trajectories; these simulations find a predominance of closed conformer and a range of low occupancy open forms. The msesMD simulations also suggest closed-to-open transitions in the tetrasaccharide are facilitated by changes in ring puckering of its GlcNAc residue away from the <sup>4</sup>C<sub>1</sub> form, in line with previous work. For sialyl Lewis X tetrasaccharide, msesMD simulations predict a minor population of an open form in solution, corresponding to a rare lectin-bound pose observed crystallographically. Overall, from comparison with biased MD calculations, we find that triplicate 10 ms unbiased MD simulations may not be enough to fully sample glycan conformations in aqueous solution. However, the computational efficiency and intuitive approach of the msesMD method suggest potential for its application in glycomics as a tool for analysis of oligosaccharide conformation.</p>

Reduced Representation by Neural Networks with Restricted Receptive Fields

1995 ◽  
Vol 7 (3) ◽  
pp. 507-517 ◽  
Author(s):  
Marco Idiart ◽  
Barry Berk ◽  
L. F. Abbott
Keyword(s):  
Neural Networks ◽  
Biological Networks ◽  
Input Data ◽  
Hebbian Learning ◽  
Receptive Fields ◽  
Principal Component ◽  
Data Sets ◽  
Reduced Representation ◽  
Learning Rules ◽  
Reduced Representations

Model neural networks can perform dimensional reductions of input data sets using correlation-based learning rules to adjust their weights. Simple Hebbian learning rules lead to an optimal reduction at the single unit level but result in highly redundant network representations. More complex rules designed to reduce or remove this redundancy can develop optimal principal component representations, but they are not very compelling from a biological perspective. Neurons in biological networks have restricted receptive fields limiting their access to the input data space. We find that, within this restricted receptive field architecture, simple correlation-based learning rules can produce surprisingly efficient reduced representations. When noise is present, the size of the receptive fields can be optimally tuned to maximize the accuracy of reconstructions of input data from a reduced representation.

Ligand binding effects on the activation of the EGFR extracellular domain

2019 ◽  
Vol 21 (15) ◽  
pp. 8141-8151 ◽  
Author(s):  
Qiang Shao ◽  
Weiliang Zhu
Keyword(s):  
Drug Discovery ◽  
Ligand Binding ◽  
Md Simulations ◽  
Extracellular Domain ◽  
Activation Mechanism ◽  
Enhanced Sampling

Enhanced sampling MD simulations were run to understand the ligand binding effects on the activation mechanism of EGFR-ECD and accordingly provide valuable information for drug discovery targeting the EGFR.

Constraint Management of Reduced Representation Variables in Decomposition-Based Design Optimization

2011 ◽  
Vol 133 (10) ◽  
Author(s):  
Michael J. Alexander ◽  
James T. Allison ◽  
Panos Y. Papalambros ◽  
David J. Gorsich
Keyword(s):  
Design Optimization ◽  
Support Vector ◽  
Management Technique ◽  
Efficient Design ◽  
Reduced Representation ◽  
Constraint Management ◽  
Target Cascading ◽  
Simple Bounds ◽  
Domain Description ◽  
Reduced Representations

In decomposition-based design optimization strategies such as analytical target cascading (ATC), it is sometimes necessary to use reduced representations of highly discretized functional data exchanged among subproblems to enable efficient design optimization. However, the variables used by such reduced representation methods are often abstract, making it difficult to constrain them directly beyond simple bounds. This problem is usually addressed by implementing a penalty value-based heuristic that indirectly constrains the reduced representation variables. Although this approach is effective, it leads to many ATC iterations, which in turn yields an ill-conditioned optimization problem and an extensive runtime. To address these issues, this paper introduces a direct constraint management technique that augments the penalty value-based heuristic with constraints generated by support vector domain description (SVDD). A comparative ATC study between the existing and proposed constraint management methods involving electric vehicle design indicates that the SVDD augmentation is the most appropriate within decomposition-based design optimization.

scholarly journals Enhanced Sampling of Protein Conformational Transitions via Dynamically Optimized Collective Variables

2018 ◽  
Author(s):  
Z. Faidon Brotzakis ◽  
Michele Parrinello
Keyword(s):  
Degrees Of Freedom ◽  
Conformational Transition ◽  
Conformational Dynamics ◽  
Computational Cost ◽  
Md Simulations ◽  
Conformational Transitions ◽  
Enhanced Sampling ◽  
Collective Variables ◽  
Simulation Based ◽  
Combined Data

AbstractProtein conformational transitions often involve many slow degrees of freedom. Their knowledge would give distinctive advantages since it provides chemical and mechanistic insight and accelerates the convergence of enhanced sampling techniques that rely on collective variables. In this study, we implemented a recently developed variational approach to conformational dynamics metadynamics to the conformational transition of the moderate size protein, L99A T4 Lysozyme. In order to find the slow modes of the system we combined data coming from NMR experiments as well as short MD simulations. A Metadynamics simulation based on these information reveals the presence of two intermediate states, at an affordable computational cost.

scholarly journals Exploring Protocols to Build Reservoirs to Accelerate Replica Exchange MD Simulations

2020 ◽  
Author(s):  
koushik kasavajhala ◽  
kenneth lam ◽  
Carlos Simmerling
Keyword(s):  
Amino Acids ◽  
Sampling Method ◽  
Convergence Rates ◽  
Md Simulations ◽  
Replica Exchange ◽  
Conformational Sampling ◽  
Enhanced Sampling ◽  
Replica Exchange Molecular Dynamics ◽  
The Past ◽  
Biomolecular Simulations

Replica Exchange Molecular Dynamics (REMD) is a widely used enhanced sampling method for accelerating biomolecular simulations. During the past two decades, several variants of REMD have been developed to further improve the rate of conformational sampling of REMD. One such variant, Reservoir REMD (RREMD), was shown to improve the rate of conformational sampling by around 5-20x. Despite the significant increase in sampling speed, RREMD methods have not been widely used due to the difficulties in building the reservoir and also due to the code not being available on the GPUs.<br><br>In this work, we ported the AMBER RREMD code onto GPUs making it 20x faster than the CPU code. Then, we explored protocols for building Boltzmann-weighted reservoirs as well as non-Boltzmann reservoirs, and tested how each choice affects the accuracy of the resulting RREMD simulations. We show that, using the recommended protocols outlined here, RREMD simulations can accurately reproduce Boltzmann-weighted ensembles obtained by much more expensive conventional REMD simulations, with at least 15x faster convergence rates even for larger proteins (>50 amino acids) compared to conventional REMD.

scholarly journals Deep Reinforcement Learning of Transition States

2021 ◽  
Author(s):  
Jun Zhang ◽  
Yao-Kun Lei ◽  
Zhen Zhang ◽  
Xu Han ◽  
Maodong Li ◽  
...  
Keyword(s):  
Machine Learning ◽  
Molecular Dynamics ◽  
Reinforcement Learning ◽  
Transition State ◽  
Chemical Reaction ◽  
Reaction Mechanisms ◽  
Md Simulations ◽  
Learning Approach ◽  
Machine Learning Approach ◽  
Chemical Reaction Mechanisms

Combining reinforcement learning (RL) and molecular dynamics (MD) simulations, we propose a machine-learning approach, called RL‡, to automatically unravel chemical reaction mechanisms. In RL‡, locating the transition state of a...

scholarly journals Enhanced Sampling Without Borders: On Global Biasing Functions and How to Reweight Them

2021 ◽  
Author(s):  
Anna Sophia Kamenik ◽  
Stephanie Maria Linker ◽  
Sereina Riniker
Keyword(s):  
Molecular Dynamics ◽  
Time Evolution ◽  
Md Simulations ◽  
Enhanced Sampling

Molecular dynamics (MD) simulations are a powerful tool to follow the time evolution of biomolecular motions in atomistic resolution. However, the high computational demand of these simulations limits the timescales...

Sign in / Sign up

Export Citation Format

Share Document