Wiggle150: Benchmarking Density Functionals and Neural Network Potentials on Highly Strained Conformers

Journal of Chemical Theory and Computation, Journal Year: 2025, Volume and Issue: unknown

Published: April 10, 2025

Accurate benchmarks are key to assessing the accuracy and robustness of computational methods, yet most available benchmark sets focus on equilibrium geometries, limiting their utility for applications involving nonequilibrium structures such as ab initio molecular dynamics automated reaction-path exploration. To address this gap, we introduce Wiggle150, a comprising 150 highly strained conformations adenosine, benzylpenicillin, efavirenz. These geometries─generated via metadynamics scored using DLPNO-CCSD(T)/CBS reference energies─exhibit substantially larger deviations in bond lengths, angles, dihedrals, relative energies than other conformer benchmarks. We evaluate diverse array including density-functional theory, composite quantum chemical semiempirical models, neural network potentials, force fields, predicting challenging set. The results highlight multiple methods along speed-accuracy Pareto frontier identify AIMNet2 particularly robust among NNPs surveyed. anticipate that Wiggle150 will be used validate protocols systems guide development new density functionals potentials.

Language: Английский

DeePMD-kit v3: A Multiple-Backend Framework for Machine Learning Potentials

Journal of Chemical Theory and Computation, Journal Year: 2025, Volume and Issue: unknown

Published: May 2, 2025

In recent years, machine learning potentials (MLPs) have become indispensable tools in physics, chemistry, and materials science, driving the development of software packages for molecular dynamics (MD) simulations related applications. These packages, typically built on specific frameworks, such as TensorFlow, PyTorch, or JAX, face integration challenges when advanced applications demand communication across different frameworks. The previous TensorFlow-based implementation DeePMD-kit exemplified these limitations. this work, we introduce version 3, a significant update featuring multibackend framework that supports PaddlePaddle backends, demonstrate versatility architecture through other MLP differentiable force fields. This allows seamless back-end switching with minimal modifications, enabling users developers to integrate using innovation facilitates more complex interoperable workflows, paving way broader MLPs scientific research.

Language: Английский

Overview on Building Blocks and Applications of Efficient and Robust Extended Tight Binding

The Journal of Physical Chemistry A, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 27, 2025

The extended tight binding (xTB) family of methods opened many new possibilities in the field computational chemistry. Within just 5 years, GFN2-xTB parametrization for all elements up to Z = 86 enabled more than a thousand applications, which were previously not feasible with other electronic structure methods. xTB provide robust and efficient way apply quantum mechanics-based approaches obtaining molecular geometries, computing free energy corrections or describing noncovalent interactions found applicability targets. A crucial contribution success is availability within simulation packages frameworks, supported by open source development its program library packages. We present comprehensive summary applications capabilities different fields Moreover, we consider main software calculations, covering their current ecosystem, novel features, usage scientific community.

Language: Английский