NewtonNet: a Newtonian message passing network for deep learning of interatomic potentials and forces

Haghighatlari, Mojtaba; Li, Jie; Guan, Xingyi; Zhang, Oufan; Das, Akshaya Kumar; Stein, Christopher J.; Heidar‐Zadeh, Farnaz; Liu, Meili; Head‐Gordon, Martin; Bertels, Luke W.; Hao, Hongxia; Leven, Itai; Head‐Gordon, Teresa

doi:10.1039/d2dd00008c

Cited by 63 publications

(49 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In cases where more or larger simulations are required such as calculating long time correlation functions for complex free energy surfaces, ML-CMD becomes even more efficient, as the initial simulations and model training only need to be performed once. Furthermore, each ML-CMD model can be shared for 36,37 Applying such a method would naturally speed up the training of ML-CMD models and should be considered for future study.…”

Section: Resultsmentioning

confidence: 99%

Centroid Molecular Dynamics Can Be Greatly Accelerated through Neural Network Learned Centroid Forces Derived from Path Integral Molecular Dynamics

Loose

Sahrmann

Voth

2022

J. Chem. Theory Comput.

View full text Add to dashboard Cite

For nearly the past 30 years, Centroid Molecular Dynamics (CMD) has proven to be a viable classical-like phase space formulation for the calculation of quantum dynamical properties.However, calculation of the centroid effective force remains a significant computational cost and limits the ability of CMD to be an efficient approach to study condensed phase quantum dynamics.In this paper we introduce a neural network-based methodology for first learning the centroid effective force from path integral molecular dynamics data, which is subsequently used as an effective force field to evolve the centroids directly with the CMD algorithm. This method, called Machine-Learned Centroid Molecular Dynamics (ML-CMD) is faster and far less costly than both standard "on the fly" CMD and ring polymer molecular dynamics (RPMD). The training aspect of ML-CMD is also straightforwardly implemented utilizing the DeepMD software kit. ML-CMD is then applied to two model systems to illustrate the approach: liquid para-hydrogen and water. The results show comparable accuracy to both CMD and RPMD in the estimation of quantum dynamical properties, including the self-diffusion constant and velocity time correlation function, but for significantly reduced overall computational cost. e.g., to better treat heterogenous systems, rare events, etc, will also provide clear benefit to the ML-CMD approach developed in this work.

show abstract

Section: Resultsmentioning

confidence: 99%

Centroid Molecular Dynamics Can Be Greatly Accelerated through Neural Network Learned Centroid Forces Derived from Path Integral Molecular Dynamics

Loose

Sahrmann

Voth

2022

J. Chem. Theory Comput.

View full text Add to dashboard Cite

show abstract

“…Recent trends have seen use of GNNs for molecular fingerprinting extensively explored for predicting molecular properties such as energy and force. Some examples of method that use GNNs for predicting energies and forces include DimeNet [ 365 ], GNNFF [ 366 ], and NewtonNet [ 367 ]. These methods have shown increased force prediction accuracy over other deep learning approaches such as PhysNet and SchNet.…”

Section: Selected Applications Of Machine Learning In Computational B...mentioning

confidence: 99%

Protein Function Analysis through Machine Learning

et al. 2022

View full text Add to dashboard Cite

Machine learning (ML) has been an important arsenal in computational biology used to elucidate protein function for decades. With the recent burgeoning of novel ML methods and applications, new ML approaches have been incorporated into many areas of computational biology dealing with protein function. We examine how ML has been integrated into a wide range of computational models to improve prediction accuracy and gain a better understanding of protein function. The applications discussed are protein structure prediction, protein engineering using sequence modifications to achieve stability and druggability characteristics, molecular docking in terms of protein–ligand binding, including allosteric effects, protein–protein interactions and protein-centric drug discovery. To quantify the mechanisms underlying protein function, a holistic approach that takes structure, flexibility, stability, and dynamics into account is required, as these aspects become inseparable through their interdependence. Another key component of protein function is conformational dynamics, which often manifest as protein kinetics. Computational methods that use ML to generate representative conformational ensembles and quantify differences in conformational ensembles important for function are included in this review. Future opportunities are highlighted for each of these topics.

show abstract

“…For energy learning, the SORF/FCHL19 39 and NequIP, with rotation orders l = 0 and l = 3. We note that FCHL19 is a comparatively simplistic atomic featurisation layer, and consequently it's expected that it does not perform as well as state-of-the-art equivariant many-body neural networks [40][41][42] such as NequiP. For a more reasonable comparison the l = 0 channel NequIP model, which contains at most 3-body terms similarly to FCHL19 has been included here.…”

Section: E Representationmentioning

confidence: 99%

GPU-Accelerated Approximate Kernel Method for Quantum Machine Learning

Browning,

Faber,

von Lilienfeld

2022

Preprint

View full text Add to dashboard Cite

Conventional kernel-based machine learning models for ab initio potential energy surfaces, while accurate and convenient in small data regimes, suffer immense computational cost as training set sizes increase. We introduce QML-Lightning, a PyTorch package containing GPU-accelerated approximate kernel models, which reduces the training time by several orders of magnitude, yielding trained models within seconds. QML-Lightning includes a cost-efficient GPU implementation of FCHL19, which together can yield energy and force predictions with competitive accuracy on a microsecond-per-atom timescale. Using modern GPU hardware, we report learning curves of energies and forces as well as timings as numerical evidence for select legacy benchmarks from atomisitic simulation including QM9, and 3BPA.

show abstract

NewtonNet: a Newtonian message passing network for deep learning of interatomic potentials and forces

Abstract: We report a new deep learning message passing network that takes inspiration from Newton's equations of motion to learn interatomic potentials and forces. With the advantage of directional information from...

Cited by 63 publications

References 33 publications

Centroid Molecular Dynamics Can Be Greatly Accelerated through Neural Network Learned Centroid Forces Derived from Path Integral Molecular Dynamics

Centroid Molecular Dynamics Can Be Greatly Accelerated through Neural Network Learned Centroid Forces Derived from Path Integral Molecular Dynamics

Protein Function Analysis through Machine Learning

GPU-Accelerated Approximate Kernel Method for Quantum Machine Learning

Contact Info

Product

Resources

About