Differentiable simulation to develop molecular dynamics force fields for disordered proteins

Greener, Joe G.

doi:10.1039/d3sc05230c

Cited by 6 publications

(4 citation statements)

References 83 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The root mean square deviation (RMSD) is utilized to assess the stability of simulation systems with regard to simulation time [ 54 ]. The intrinsic flexibility and thermal mobility of native 1BNA may be responsible for the observed fluctuations in the RMSD plots of native 1BNA [ 55 ].…”

Section: Resultsmentioning

confidence: 99%

Elucidating the Interaction of Indole-3-Propionic Acid and Calf Thymus DNA: Multispectroscopic and Computational Modeling Approaches

Wei,

Zhang,

Pan

et al. 2024

Foods

View full text Add to dashboard Cite

Indole-3-propionic acid (IPA) is a plant growth regulator with good specificity and long action. IPA may be harmful to human health because of its accumulation in vegetables and fruits. Therefore, in this study, the properties of the interaction between calf thymus DNA (ctDNA) and IPA were systematically explored using multispectroscopic and computational modeling approaches. Analysis of fluorescence spectra showed that IPA binding to ctDNA to spontaneously form a complex was mainly driven by hydrogen bonds and hydrophobic interaction. DNA melting analysis, viscosity analysis, DNA cleavage study, and circular dichroism measurement revealed the groove binding of IPA to ctDNA and showed that the binding did not significantly change ctDNA confirmation. Furthermore, molecular docking found that IPA attached in the A-T rich minor groove region of the DNA. Molecular dynamics simulation showed that DNA and IPA formed a stable complex and IPA caused slight fluctuations for the residues at the binding site. Gel electrophoresis experiments showed that IPA did not significantly disrupt the DNA structure. These findings may provide useful information on the potential toxicological effects and environmental risk assessments of IPA residue in food at the molecular level.

show abstract

Section: Resultsmentioning

confidence: 99%

Elucidating the Interaction of Indole-3-Propionic Acid and Calf Thymus DNA: Multispectroscopic and Computational Modeling Approaches

Wei,

Zhang,

Pan

et al. 2024

Foods

View full text Add to dashboard Cite

show abstract

“…where η is the learning rate and scriptS p represents the update step function of the given optimizer. Various packages implementing differentiable molecular dynamics are available, such as JAX-MD, TorchMD, and Molly. , However, these programs lack some critical features, especially the particle mesh routines needed for the HhPF nonbonded interactions. Therefore, we have implemented our differentiable framework in ∂-HyMD based on our open-source HhPF simulator HyMD. , In ∂-HyMD we use JAX , to trace the update steps and get the gradient of the loss function as in eq 13 , while implementing the HhPF operations with fast Fourier transforms, the MD integrator, barostat and thermostat using JAX NumPy API, and taking advantage of JIT compilations whenever possible.…”

Section: Methodsmentioning

confidence: 99%

“…Various packages implementing differentiable molecular dynamics are available, such as JAX-MD, 40 TorchMD, 32 and Molly. 41 , 42 However, these programs lack some critical features, especially the particle mesh routines needed for the HhPF nonbonded interactions. Therefore, we have implemented our differentiable framework in ∂-HyMD based on our open-source HhPF simulator HyMD.…”

Section: Methodsmentioning

confidence: 99%

Learning Force Field Parameters from Differentiable Particle-Field Molecular Dynamics

Carrer,

Cezar,

Bore

et al. 2024

J. Chem. Inf. Model.

View full text Add to dashboard Cite

We develop ∂-HylleraasMD (∂-HyMD), a fully endto-end differentiable molecular dynamics software based on the Hamiltonian hybrid particle-field formalism, and use it to establish a protocol for automated optimization of force field parameters. ∂-HyMD is templated on the recently released HylleraaasMD software, while using the JAX autodiff framework as the main engine for the differentiable dynamics. ∂-HyMD exploits an embarrassingly parallel optimization algorithm by spawning independent simulations, whose trajectories are simultaneously processed by reverse mode automatic differentiation to calculate the gradient of the loss function, which is in turn used for iterative optimization of the force-field parameters. We show that parallel organization facilitates the convergence of the minimization procedure, avoiding the known memory and numerical stability issues of differentiable molecular dynamics approaches. We showcase the effectiveness of our implementation by producing a library of force field parameters for standard phospholipids, with either zwitterionic or anionic heads and with saturated or unsaturated tails. Compared to the all-atom reference, the force field obtained by ∂-HyMD yields better density profiles than the parameters derived from previously utilized gradient-free optimization procedures. Moreover, ∂-HyMD models can predict with good accuracy properties not included in the learning objective, such as lateral pressure profiles, and are transferable to other systems, including triglycerides.

show abstract

“…The Automated Topology Builder framework also takes an alternate approach to typing and parameter assignment 58 as does TAFFI, 59 and the XtalPi/Pfizer XFF force field uses a somewhat similar framework to advance an alternate force field effort. 38 Other machine learning frameworks like Espaloma, 60 , 61 Grappa, 62 MACE, 63 DMFF 64 or other differentiable frameworks 65 provide an interesting alternative and potentially promising future direction, as well.…”

Section: Introductionmentioning

confidence: 99%

The Open Force Field Initiative: Open Software and Open Science for Molecular Modeling

Wang,

Behara,

Thompson

et al. 2024

J. Phys. Chem. B

View full text Add to dashboard Cite

Force fields are a key component of physics-based molecular modeling, describing the energies and forces in a molecular system as a function of the positions of the atoms and molecules involved. Here, we provide a review and scientific status report on the work of the Open Force Field (OpenFF) Initiative, which focuses on the science, infrastructure and data required to build the next generation of biomolecular force fields. We introduce the OpenFF Initiative and the related OpenFF Consortium, describe its approach to force field development and software, and discuss accomplishments to date as well as future plans. OpenFF releases both software and data under open and permissive licensing agreements to enable rapid application, validation, extension, and modification of its force fields and software tools. We discuss lessons learned to date in this new approach to force field development. We also highlight ways that other force field researchers can get involved, as well as some recent successes of outside researchers taking advantage of OpenFF tools and data.

show abstract

Differentiable simulation to develop molecular dynamics force fields for disordered proteins

Abstract: Implicit solvent force fields are computationally efficient but can be unsuitable for running molecular dynamics on disordered proteins. Here I improve the a99SB-disp force field and the GBNeck2 implicit solvent...

Cited by 6 publications

References 83 publications

Elucidating the Interaction of Indole-3-Propionic Acid and Calf Thymus DNA: Multispectroscopic and Computational Modeling Approaches

Elucidating the Interaction of Indole-3-Propionic Acid and Calf Thymus DNA: Multispectroscopic and Computational Modeling Approaches

Learning Force Field Parameters from Differentiable Particle-Field Molecular Dynamics

The Open Force Field Initiative: Open Software and Open Science for Molecular Modeling

Contact Info

Product

Resources

About