A Scalable Molecular Force Field Parameterization Method Based on Density Functional Theory and Quantum-Level Machine Learning

Galvelis, Raimondas; Doerr, Stefan; Damas, João M.; Harvey, Matt; Fabritiis, Gianni De

doi:10.1021/acs.jcim.9b00439

Cited by 75 publications

(60 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ANI-1x neural network potential [43], for example, is able to reproduce DFT-level energies ( B97X functional with 6-31G* basis set) with a 10 6 speed up. Indeed, the ANI models are so fast and reproduce quantum chemical data so well that recent approaches have integrated them into bespoke torsion refitting schemes as an alternative to costly QM torsion scans [44].…”

Section: Machine Learning (Ml) Potentials Can Reproduce Qm Energies Amentioning

confidence: 99%

Towards chemical accuracy for alchemical free energy calculations with hybrid physics-based machine learning / molecular mechanics potentials

Rufa

Macdonald

Fass

et al. 2020

Preprint

106

View full text Add to dashboard Cite

Alchemical free energy methods with molecular mechanics (MM) force fields are now widely used in the prioritization of small molecules for synthesis in structure-enabled drug discovery projects because of their ability to deliver 1–2 kcal mol−1 accuracy in well-behaved protein-ligand systems. Surpassing this accuracy limit would significantly reduce the number of compounds that must be synthesized to achieve desired potencies and selectivities in drug design campaigns. However, MM force fields pose a challenge to achieving higher accuracy due to their inability to capture the intricate atomic interactions of the physical systems they model. A major limitation is the accuracy with which ligand intramolecular energetics—especially torsions—can be modeled, as poor modeling of torsional profiles and coupling with other valence degrees of freedom can have a significant impact on binding free energies. Here, we demonstrate how a new generation of hybrid machine learning / molecular mechanics (ML/MM) potentials can deliver significant accuracy improvements in modeling protein-ligand binding affinities. Using a nonequilibrium perturbation approach, we can correct a standard, GPU-accelerated MM alchemical free energy calculation in a simple post-processing step to efficiently recover ML/MM free energies and deliver a significant accuracy improvement with small additional computational effort. To demonstrate the utility of ML/MM free energy calculations, we apply this approach to a benchmark system for predicting kinase:inhibitor binding affinities—a congeneric ligand series for non-receptor tyrosine kinase TYK2 (Tyk2)—wherein state-of-the-art MM free energy calculations (with OPLS2.1) achieve inaccuracies of 0.93±0.12 kcal mol−1 in predicting absolute binding free energies. Applying an ML/MM hybrid potential based on the ANI2x ML model and AMBER14SB/TIP3P with the OpenFF 1.0.0 (“Parsley”) small molecule force field as an MM model, we show that it is possible to significantly reduce the error in absolute binding free energies from 0.97 [95% CI: 0.68, 1.21] kcal mol−1 (MM) to 0.47 [95% CI: 0.31, 0.63] kcal mol−1 (ML/MM).

show abstract

Section: Machine Learning (Ml) Potentials Can Reproduce Qm Energies Amentioning

confidence: 99%

Towards chemical accuracy for alchemical free energy calculations with hybrid physics-based machine learning / molecular mechanics potentials

Rufa

Macdonald

Fass

et al. 2020

Preprint

106

View full text Add to dashboard Cite

show abstract

“…MD simulations were carried out using PlayMolecule (Accelera, Middlesex, UK) starting from the output models of docking experiments. The ligand was prepared by running Parametrize function based on GAFF2 force field [60]. The complex was prepared for the simulation using ProteinPrepare and SystemBuilder functions, setting pH = 7.4, AMBER force field and default experiment parameters [61].…”

Section: Molecular Dynamics Simulationsmentioning

confidence: 99%

A Drug Repurposing Approach for Antimalarials Interfering with SARS-CoV-2 Spike Protein Receptor Binding Domain (RBD) and Human Angiotensin-Converting Enzyme 2 (ACE2)

Coghi

Yang

et al. 2021

Pharmaceuticals

View full text Add to dashboard Cite

Host cell invasion by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is mediated by the interaction of the viral spike protein (S) with human angiotensin-converting enzyme 2 (ACE2) through the receptor-binding domain (RBD). In this work, computational and experimental techniques were combined to screen antimalarial compounds from different chemical classes, with the aim of identifying small molecules interfering with the RBD-ACE2 interaction and, consequently, with cell invasion. Docking studies showed that the compounds interfere with the same region of the RBD, but different interaction patterns were noted for ACE2. Virtual screening indicated pyronaridine as the most promising RBD and ACE2 ligand, and molecular dynamics simulations confirmed the stability of the predicted complex with the RBD. Bio-layer interferometry showed that artemisone and methylene blue have a strong binding affinity for RBD (KD = 0.363 and 0.226 μM). Pyronaridine also binds RBD and ACE2 in vitro (KD = 56.8 and 51.3 μM). Overall, these three compounds inhibit the binding of RBD to ACE2 in the μM range, supporting the in silico data.

show abstract

“…Recently, it has been proved that more complex atomistic structures, such as amorphous GeTe [23] and polymer materials [24], can be simulated with similar methods as well. However, the capability of ML potentials in predicting thermal conductivity of a single material in different phases has not been evaluated.…”

mentioning

confidence: 99%

A unified deep neural network potential capable of predicting thermal conductivity of silicon in different phases

Lee

Luo

2020

Materials Today Physics

View full text Add to dashboard Cite

Molecular dynamics simulations have been extensively used to predict thermal properties, but simulating different phases with similar precision using a unified force field is often difficult, due to the lack of accurate and transferrable interatomistic potential fields. As a result, this issue has become a major barrier to predicting the phase change of materials and their transport properties with atomistic-level modeling techniques. Recently, machine learning based algorithms have emerged as promising tools to develop accurate potentials for molecular dynamics simulations. In this work, we approach the problem of predicting the thermal conductivity of silicon in different phases by performing molecular dynamics simulations with a deep neural network potential. This neural network potential is trained with ab-initio data of silicon in the crystalline, liquid and amorphous phases. The accuracy of our potential is first validated through reproducing the atomistic structures during the phase transition, where other empirical potentials usually fail. The thermal conductivity of different phases is then calculated, showing a good agreement with the experimental results and ab-initio calculation results. Our work shows that a unified neural network-based potential can be a promising tool for studying phase change and thermal transport of materials with high accuracy.

show abstract

A Scalable Molecular Force Field Parameterization Method Based on Density Functional Theory and Quantum-Level Machine Learning

Cited by 75 publications

References 43 publications

Towards chemical accuracy for alchemical free energy calculations with hybrid physics-based machine learning / molecular mechanics potentials

Towards chemical accuracy for alchemical free energy calculations with hybrid physics-based machine learning / molecular mechanics potentials

A Drug Repurposing Approach for Antimalarials Interfering with SARS-CoV-2 Spike Protein Receptor Binding Domain (RBD) and Human Angiotensin-Converting Enzyme 2 (ACE2)

A unified deep neural network potential capable of predicting thermal conductivity of silicon in different phases

Contact Info

Product

Resources

About