Machine learning implicit solvation for molecular dynamics

Chen, Yaoyi; Krämer, Andreas; Charron, Nicholas; Husic, Brooke E.; Clementi, Cecilia; Noé, Frank

doi:10.1063/5.0059915

Cited by 56 publications

(73 citation statements)

References 119 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recent advances in deep learning allowed us to describe fully atomistic models using coarse-grained representations. This was exemplified by the capabilities of those models to represent explicit solvent using only a few coarse-grained beads or to account for solvent effects in a fully atomistic representation of the molecular system without explicit solvent molecules . Hierarchical approaches are also used in deep-learning models for protein structure prediction in which the backbone is initially predicted followed by prediction of the side-chain atoms …”

Section: Methodsmentioning

confidence: 99%

Accurate Sampling of Macromolecular Conformations Using Adaptive Deep Learning and Coarse-Grained Representation

Mahmoud

Masters

Lee

et al. 2022

J. Chem. Inf. Model.

View full text Add to dashboard Cite

Conformational sampling of protein structures is essential for understanding biochemical functions and for predicting thermodynamic properties such as free energies. Where previous approaches rely on sequential sampling procedures, recent developments in generative deep neural networks rendered possible the parallel, statistically independent sampling of molecular configurations. To be able to accurately generate samples of large molecular systems from a high-dimensional multimodal equilibrium distribution function, we developed a hierarchical approach based on expressive normalizing flows with rational quadratic neural splines and coarse-grained representation. Furthermore, system specific priors and adaptive and property-based controlled learning was designed to diminish the likelihood for the generation of high-energy structures during sampling. Finally, backmapping from a coarse-grained to fully atomistic representation is performed through an equivariant transformer model. We demonstrate the applicability of the method on the one-shot configurational sampling of a protein system with more than a hundred amino acids. The results show enhanced expressivity that diminish the invertibility constraints inherent in the normalizing flow framework. Moreover, the capacity of the hierarchical normalizing flow model was tested on a challenging case study of the folding/unfolding dynamics of the peptide chignolin.

show abstract

Section: Methodsmentioning

confidence: 99%

Accurate Sampling of Macromolecular Conformations Using Adaptive Deep Learning and Coarse-Grained Representation

Mahmoud

Masters

Lee

et al. 2022

J. Chem. Inf. Model.

View full text Add to dashboard Cite

show abstract

“…Using our entire reference atomistic dataset we first learn a TICA embedding into the first two non-trivial Independent Components (ICs) using all 45 pairwise α-carbon distances as features. 67,68,85,86 This learned TICA model provides us with a fixed basis set we use for constructing an FES in these two leading ICs for the atomistic and backmapped data from both our in distribution and generalization data sets.…”

Section: Thermodynamicsmentioning

confidence: 99%

“…Following the procedure in Chen et al, 85 each model was training on the delta forces; that is, each model was trained on the difference between the reference forces and the forces predicted by the prior model:…”

Section: Coarse Grain Cgschnet Force Fieldsmentioning

confidence: 99%

Temporally coherent backmapping of molecular trajectories from coarse-grained to atomistic resolution

Shmilovich¹,

Stieffenhofer²,

Charron³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Coarse-graining offers a means to extend the achievable time and length scales of molecular dynamics simulations beyond what is practically possible in the atomistic regime. Sampling molecular configurations of interest can be done efficiently using coarse-grained simulations, from which meaningful physicochemical information can be inferred if the corresponding all-atom configurations are reconstructed. However, this procedure of backmapping to reintroduce the lost atomistic detail into coarse-grain structures has proven a challenging task due to the many feasible atomistic configurations that can be associated with one coarse-grain structure. Existing backmapping methods are strictly frame-based, relying on either heuristics to replace coarse-grain particles with atomic fragments and subsequent relaxation, or parameterized models to propose atomic coordinates separately and independently for each coarse-grain structure. These approaches neglect information from previous trajectory frames that is critical to ensuring temporal coherence of the backmapped trajectory, while also offering information potentially helpful to produce higher-fidelity atomic reconstructions. In this work we present a deep learning-enabled data-driven approach for temporally coherent backmapping that explicitly incorporates information from preceding trajectory structures. Our method trains a conditional variational autoencoder to non-deterministically reconstruct atomistic detail conditioned on both the target coarse-grain configuration and the previously reconstructed atomistic configuration. We demonstrate our backmapping approach on two exemplar biomolecular systems: alanine dipeptide and the miniprotein chignolin. We show that our backmapped trajectories accurately recover the structural, thermodynamic, and kinetic properties of the atomistic trajectory data.

show abstract

“…This work uses the recently introduced CGNet 17 architecture to eventually learn and express the CG potential. This choice is rather arbitrary and the CGNet could be replaced by a different functional form, 19,[36][37][38][39][40][41][42] such as CGSchNet 19 or a just a parametric sum of analytical force field terms that are frequently employed in CG models 12,[43][44][45][46] such as Martini. 45,46 The same holds for choosing the flow model.…”

Section: Introductionmentioning

confidence: 99%

Flow-matching -- efficient coarse-graining of molecular dynamics without forces

Köhler¹,

Chen²,

Krämer³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Coarse-grained (CG) molecular simulations have become a standard tool to study molecular processes on time-and length-scales inaccessible to all-atom simulations. Learning CG force fields from all-atom data has mainly relied on force-matching and relative entropy minimization. Force-matching is straightforward to implement but requires the forces on the CG particles to be saved during all-atom simulation, and because these instantaneous forces depend on all degrees of freedom, they provide a very noisy signal that makes training the CG force field data inefficient. Relative entropy minimization does not require forces to be saved and is more data-efficient, but requires the CG model to be re-simulated during the iterative training procedure, which can make the training procedure extremely costly or lead to failure to converge. Here we present flow-matching, a new training method for CG force fields that combines the advantages of force-matching and relative entropy minimization by leveraging normalizing flows, a generative deep learning method. Flow-matching first trains a normalizing flow to represent the CG probability density by using relative entropy minimization without suffering from the re-simulation problem because flows can directly sample from the equilibrium distribution they represent. Subsequently, the forces of the flow are used to train a CG force field by matching the coarse-grained forces directly, which is a much easier problem than traditional force-matching as it does not suffer from the noise problem. Besides not requiring forces, flow-matching also outperforms classical forcematching by an order of magnitude in terms of data efficiency and produces CG models that can capture the folding and unfolding of small proteins.

show abstract

Machine learning implicit solvation for molecular dynamics

Cited by 56 publications

References 119 publications

Accurate Sampling of Macromolecular Conformations Using Adaptive Deep Learning and Coarse-Grained Representation

Accurate Sampling of Macromolecular Conformations Using Adaptive Deep Learning and Coarse-Grained Representation

Temporally coherent backmapping of molecular trajectories from coarse-grained to atomistic resolution

Flow-matching -- efficient coarse-graining of molecular dynamics without forces

Contact Info

Product

Resources

About