Search Results for author:
Found 9 papers, 6 papers with code
OpenMM 8: Molecular Dynamics Simulation with Machine Learning Potentials
Machine learning plays an important and growing role in molecular simulation.
Machine-learned molecular mechanics force field for the simulation of protein-ligand systems and beyond
The development of reliable and extensible molecular mechanics (MM) force fields -- fast, empirical models characterizing the potential energy surface of molecular systems -- is indispensable for biomolecular simulation and computer-aided drug design.
EspalomaCharge: Machine learning-enabled ultra-fast partial charge assignment
Our hybrid approach couples a graph neural network to a streamlined charge equilibration approach in order to predict molecule-specific atomic electronegativity and hardness parameters, followed by analytical determination of optimal charge-equilibrated parameters that preserves total molecular charge.
Spatial Attention Kinetic Networks with E(n)-Equivariance
Neural networks that are equivariant to rotations, translations, reflections, and permutations on n-dimensional geometric space have shown promise in physical modeling for tasks such as accurately but inexpensively modeling complex potential energy surfaces to guiding the sampling of complex dynamical systems or forecasting their time evolution.
SPICE, A Dataset of Drug-like Molecules and Peptides for Training Machine Learning Potentials
Machine learning potentials are an important tool for molecular simulation, but their development is held back by a shortage of high quality datasets to train them on.
Stochastic Aggregation in Graph Neural Networks
Graph neural networks (GNNs) manifest pathologies including over-smoothing and limited discriminating power as a result of suboptimally expressive aggregating mechanisms.
Rayleigh-Brillouin light scattering spectroscopy of air; experiment, predictive model and dimensionless scaling
Spontaneous Rayleigh-Brillouin scattering (RBS) experiments have been performed in air for pressures in the range 0. 25 - 3 bar and temperatures in the range 273 - 333 K. The functional behaviour of the RB-spectral profile as a function of experimental parameters, such as the incident wavelength, scattering angle, pressure and temperature is analyzed, as well as the dependence on thermodynamic properties of the gas, as the shear viscosity, the thermal conductivity, the internal heat capacity and the bulk viscosity.
Atmospheric and Oceanic Physics Fluid Dynamics
End-to-End Differentiable Molecular Mechanics Force Field Construction
Trained with arbitrary loss functions, it can construct entirely new force fields self-consistently applicable to both biopolymers and small molecules directly from quantum chemical calculations, with superior fidelity than traditional atom or parameter typing schemes.
Graph Nets for Partial Charge Prediction
Atomic partial charges are crucial parameters for Molecular Dynamics (MD) simulations, molecular mechanics calculations, and virtual screening, as they determine the electrostatic contributions to interaction energies.
