Encoding and selecting coarse-grain mapping operators with hierarchical graphs

Chakraborty, Maghesree; Xu, Chenliang; White, Andrew D.

doi:10.1063/1.5040114

Cited by 45 publications

(51 citation statements)

References 76 publications

(66 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, machine learning tools have facilitated the development CG force fields [9][10][11] and graph-based CG representations [12,13]. Here we propose to use machine learning to simultaneously optimize CG representations and potentials from atomistic simulations.…”

Section: Introductionmentioning

confidence: 99%

Coarse-graining auto-encoders for molecular dynamics

2019

View full text Add to dashboard Cite

Molecular dynamics simulations provide theoretical insight into the microscopic behavior of materials in condensed phase and, as a predictive tool, enable computational design of new compounds. However, because of the large temporal and spatial scales involved in thermodynamic and kinetic phenomena in materials, atomistic simulations are often computationally unfeasible. Coarsegraining methods allow simulating larger systems, by reducing the dimensionality of the simulation, and propagating longer timesteps, by averaging out fast motions. Coarse-graining involves two coupled learning problems; defining the mapping from an all-atom to a reduced representation, and the parametrization of a Hamiltonian over coarse-grained coordinates. Multiple statistical mechanics approaches have addressed the latter, but the former is generally a hand-tuned process based on chemical intuition. Here we present Autograin, an optimization framework based on autoencoders to learn both tasks simultaneously. Autograin is trained to learn the optimal mapping between all-atom and reduced representation, using the reconstruction loss to facilitate the learning of coarse-grained variables. In addition, a force-matching method is applied to variationally determine the coarse-grained potential energy function. This procedure is tested on a number of model systems including single-molecule and bulk-phase periodic simulations.

show abstract

Section: Introductionmentioning

confidence: 99%

Coarse-graining auto-encoders for molecular dynamics

2019

View full text Add to dashboard Cite

show abstract

“…Their theoretical model focuses on preserving most information content in the lower resolution model compared to the all atom model. Chakraborty et al 8 reported a hierarchical graph method where multiple mappings of a given molecule are encoded in a hierarchical graph, which can further be used to auto-select a particular mapping using algorithms like uniform-entropy attening. 9 In a recent systematic study on the effects of CG resolution on reproducing on and off target properties of a system, Khot et al 10 hypothesized that low-resolution CG models might be information limited, instead of having a representability limitation.…”

Section: Introductionmentioning

confidence: 99%

“…Mapping operators used in practice for CG simulations are usually rule-based, 3,4 but recent advances have been made in algorithmic 5,8,[12][13][14][15] and unsupervised methods. 16 Rule-based schemes have xed resolution and must be created for each molecular functional group, limiting their application to sequence-dened biomolecules or polymers.…”

Section: Introductionmentioning

confidence: 99%

Graph neural network based coarse-grained mapping prediction

Wellawatte

Chakraborty

et al. 2020

Chem. Sci.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The general practice to define a CG mapping operators has been driven by chemical intuition [63] . However, recently there have been efforts towards choosing mapping operators more systemically [64][65][66] .…”

Section: Force Matching (Learning)mentioning

confidence: 99%

A GPU-Accelerated Machine Learning Framework for Molecular Simulation: Hoomd-Blue with TensorFlow

Barrett¹,

Chakraborty²,

Gandhi³

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

We have designed and implemented software that enables integration of a scalable GPU-accelerated molecular mechanics engine, Hoomd-blue, with the machine learning (ML) TensorFlow package. TensorFlow is a GPU accelerated, scalable, graph-based tensor computation model building package that has been the implementation of many recent innovations in deep learning and other ML tasks. Tensor computation graphs allow for designation of robust, flexible, and easily replicated computational models for a variety of tasks. Our plugin leverages the generality and speed of computational tensor graphs in TensorFlow to enable four previously challenging tasks in molecular dynamics: (1) the calculation of arbitrary force-fields including neural-network-based, stochastic, and/or automatically-generated force-fields which are differentiated from potential functions; (2) the efficient computation of arbitrary collective variables; (3) the biasing of simulations via automatic differentiation of collective variables and consequently the implementation of many free energy biasing methods; (4) ML on any of the above tasks, including coarse grain force fields, on-the-fly learned biases, and collective variable calculations. The TensorFlow models are constructed in Python and can be visualized or debugged using the rich set of tools implemented in the TensorFlow package. In this article, we present examples of the four major tasks this method can accomplish, benchmark data, and describe the architecture of our implementation. This method should lead to both the design of new models in computational chemistry research and reproducible model specification without requiring recompiling or writing low-level code. <br>

show abstract

Encoding and selecting coarse-grain mapping operators with hierarchical graphs

Cited by 45 publications

References 76 publications

Coarse-graining auto-encoders for molecular dynamics

Coarse-graining auto-encoders for molecular dynamics

Graph neural network based coarse-grained mapping prediction

A GPU-Accelerated Machine Learning Framework for Molecular Simulation: Hoomd-Blue with TensorFlow

Contact Info

Product

Resources

About