Differentiable molecular simulation can learn all the parameters in a coarse-grained force field for proteins.

Finding optimal parameters for force fields used in molecular simulation is a challenging and time-consuming task, partly due to the difficulty of tuning multiple parameters at once. Automatic differentiation presents a general solution: run a simulation, obtain gradients of a loss function with res...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Joe G Greener, David T Jones
Formato: article
Lenguaje:EN
Publicado: Public Library of Science (PLoS) 2021
Materias:
R
Q
Acceso en línea:https://doaj.org/article/60f87f3edae44e5b94a9680cd43991c8
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:60f87f3edae44e5b94a9680cd43991c8
record_format dspace
spelling oai:doaj.org-article:60f87f3edae44e5b94a9680cd43991c82021-12-02T20:08:32ZDifferentiable molecular simulation can learn all the parameters in a coarse-grained force field for proteins.1932-620310.1371/journal.pone.0256990https://doaj.org/article/60f87f3edae44e5b94a9680cd43991c82021-01-01T00:00:00Zhttps://doi.org/10.1371/journal.pone.0256990https://doaj.org/toc/1932-6203Finding optimal parameters for force fields used in molecular simulation is a challenging and time-consuming task, partly due to the difficulty of tuning multiple parameters at once. Automatic differentiation presents a general solution: run a simulation, obtain gradients of a loss function with respect to all the parameters, and use these to improve the force field. This approach takes advantage of the deep learning revolution whilst retaining the interpretability and efficiency of existing force fields. We demonstrate that this is possible by parameterising a simple coarse-grained force field for proteins, based on training simulations of up to 2,000 steps learning to keep the native structure stable. The learned potential matches chemical knowledge and PDB data, can fold and reproduce the dynamics of small proteins, and shows ability in protein design and model scoring applications. Problems in applying differentiable molecular simulation to all-atom models of proteins are discussed along with possible solutions and the variety of available loss functions. The learned potential, simulation scripts and training code are made available at https://github.com/psipred/cgdms.Joe G GreenerDavid T JonesPublic Library of Science (PLoS)articleMedicineRScienceQENPLoS ONE, Vol 16, Iss 9, p e0256990 (2021)
institution DOAJ
collection DOAJ
language EN
topic Medicine
R
Science
Q
spellingShingle Medicine
R
Science
Q
Joe G Greener
David T Jones
Differentiable molecular simulation can learn all the parameters in a coarse-grained force field for proteins.
description Finding optimal parameters for force fields used in molecular simulation is a challenging and time-consuming task, partly due to the difficulty of tuning multiple parameters at once. Automatic differentiation presents a general solution: run a simulation, obtain gradients of a loss function with respect to all the parameters, and use these to improve the force field. This approach takes advantage of the deep learning revolution whilst retaining the interpretability and efficiency of existing force fields. We demonstrate that this is possible by parameterising a simple coarse-grained force field for proteins, based on training simulations of up to 2,000 steps learning to keep the native structure stable. The learned potential matches chemical knowledge and PDB data, can fold and reproduce the dynamics of small proteins, and shows ability in protein design and model scoring applications. Problems in applying differentiable molecular simulation to all-atom models of proteins are discussed along with possible solutions and the variety of available loss functions. The learned potential, simulation scripts and training code are made available at https://github.com/psipred/cgdms.
format article
author Joe G Greener
David T Jones
author_facet Joe G Greener
David T Jones
author_sort Joe G Greener
title Differentiable molecular simulation can learn all the parameters in a coarse-grained force field for proteins.
title_short Differentiable molecular simulation can learn all the parameters in a coarse-grained force field for proteins.
title_full Differentiable molecular simulation can learn all the parameters in a coarse-grained force field for proteins.
title_fullStr Differentiable molecular simulation can learn all the parameters in a coarse-grained force field for proteins.
title_full_unstemmed Differentiable molecular simulation can learn all the parameters in a coarse-grained force field for proteins.
title_sort differentiable molecular simulation can learn all the parameters in a coarse-grained force field for proteins.
publisher Public Library of Science (PLoS)
publishDate 2021
url https://doaj.org/article/60f87f3edae44e5b94a9680cd43991c8
work_keys_str_mv AT joeggreener differentiablemolecularsimulationcanlearnalltheparametersinacoarsegrainedforcefieldforproteins
AT davidtjones differentiablemolecularsimulationcanlearnalltheparametersinacoarsegrainedforcefieldforproteins
_version_ 1718375185106599936