The development of machine-learning (ML) potentials offers significant accuracy improvements compared to molecular mechanics (MM) because the inclusion quantum-mechanical effects in interactions. However, ML simulations are several times more computationally demanding than MM simulations, so there is a trade-off between speed and accuracy. One possible compromise hybrid machine learning/molecular (ML/MM) approaches with mechanical embedding that treat intramolecular interactions ligand at...

To achieve chemical accuracy in free energy calculations, it is necessary to accurately describe the system's potential surface and efficiently sample configurations from its Boltzmann distribution. While neural network potentials (NNPs) have shown significantly higher than classical molecular mechanics (MM) force fields, they a limited range of applicability are considerably slower MM potentials, often by orders magnitude. address this challenge, Rufa et al. [Rufa bioRxiv 2020,...

We recently introduced transformato, an open-source Python package for the automated setup of large-scale calculations relative solvation and binding free energy differences. Here, we extend capabilities transformato to calculation absolute After careful validation against literature results reference with PERT module CHARMM, used compute energies most molecules in FreeSolv database (621 out 642). The force field parameters were obtained program cgenff (v2.5.1), which derives missing from...

We present a comprehensive study investigating the potential gain in accuracy for calculating absolute solvation free energies (ASFE) using neural network to describe intramolecular energy of solute. calculated ASFE most compounds from FreeSolv database Open Force Field (OpenFF) and compared them earlier results obtained with CHARMM General (CGenFF). By applying nonequilibrium (NEQ) switching approach between molecular mechanics (MM) description (either OpenFF or CGenFF) net (NNP)/MM level...

Lewis pair polymerization employing N-Heterocyclic olefins (NHOs) and simple metal halides as co-catalysts has emerged a useful tool to polymerize diverse lactones. To elucidate some of the mechanistic aspects that remain unclear date better understand impact species, computational methods have been applied. Several key considered: (1) formation NHO-metal halide adducts evaluated for eight different NHOs three acids, (2) coordination four lactones MgCl₂ was studied (3) deprotonation an...

Transition state (TS) geometries of chemical reactions are key to understanding reaction mechanisms and estimating kinetic properties. Inferring these directly from 2D graphs offers chemists a powerful tool for rapid accessible analysis. Quantum methods computing TSs computationally intensive often infeasible larger molecular systems. Recently, deep learning–based diffusion models have shown promise in generating single-step reactions. However, framing TS generation as process, by design,...

The accurate prediction of reaction barrier heights is crucial for understanding chemical reactivity and guiding design. Recent advances in machine learning (ML) models, particularly graph neural networks, have shown great promise capturing complex interactions. Here, directed message-passing networks (D-MPNNs) on overlays the reactant product graphs were to provide promising accuracies property prediction. They only rely change molecular as input thus require no additional information...

The accurate prediction of reaction barrier heights is crucial for understanding chemical reactivity and guiding design. Recent advances in machine learning (ML) models, particularly graph neural networks, have shown great promise capturing complex interactions. Here, directed message-passing networks (D-MPNNs) on overlays the reactant product graphs were to provide promising accuracies property prediction. They only rely change molecular as input thus require no additional information...

We present the software package transformato for setup of large-scale relative binding free energy calculations. Transformato is written in Python as an open source project (https://github.com/wiederm/transformato); contrast to comparable tools, it not closely tied a particular molecular dynamics engine carry out underlying simulations. Instead alchemically transforming ligand L 1 directly into another 2, two ligands are mutated common core. Thus, while dummy atoms required at intermediate...

To achieve chemical accuracy in free energy calculations, it is necessary to accurately describe the system's potential surface and efficiently sample configurations from its Boltzmann distribution. While neural network potentials (NNPs) have shown significantly higher than classical molecular mechanics (MM) force fields, they limited range of applicability are slower MM potentials, often by orders magnitude. address this challenge, Rufa et al suggested a two-stage approach that uses fast...

We present a comprehensive study investigating the potential gain in accuracy for calculating absolute solvation free energies (ASFE) using neural network intramolecular energies. calculated ASFE Open Force Field (OpenFF) and compared results to previously ASFEs employing CHARMM General (CGenFF). By applying nonequilibrium (NEQ) switching approach between molecular mechanics (MM) description (either OpenFF or CGenFF) machine learning (ML)/MM level of theory (using ANI-2x as ML potential), we...

We present a comprehensive study investigating the potential gain in accuracy for calculating absolute solvation free energies (ASFE) using neural network to describe intramolecular energy of solute. calculated ASFE most compounds from FreeSolv database Open Force Field (OpenFF) and compared them earlier results obtained with CHARMM General (CGenFF). By applying nonequilibrium (NEQ) switching approach between molecular mechanics (MM) description (either OpenFF or CGenFF) net (NNP)/MM level...

To achieve chemical accuracy in free energy calculations, it is necessary to accurately describe the system's potential surface and efficiently sample configurations from its Boltzmann distribution. While neural network potentials (NNPs) have shown significantly higher than classical molecular mechanics (MM) force fields, they limited range of applicability are slower MM potentials, often by orders magnitude. address this challenge, Rufa et al suggested a two-stage approach that uses fast...

The development of machine-learning (ML) potentials offers significant accuracy improvements compared to molecular mechanics (MM) because the inclusion quantum-mechanical effects in interactions. However, ML simulations are several times more computationally demanding than MM simulations, so there is a trade-off between speed and accuracy. One possible compromise hybrid machine learning/molecular (ML/MM) approaches with mechanical embedding that treat intramolecular interactions ligand at...

To achieve chemical accuracy in free energy calculations, it is necessary to accurately describe the system's potential surface and efficiently sample configurations from its Boltzmann distribution. While neural network potentials (NNPs) have shown significantly higher than classical molecular mechanics (MM) force fields, they limited range of applicability are slower MM potentials, often by orders magnitude. address this challenge, Rufa et al suggested a two-stage approach that uses fast...