The Transferable Water Implicit Network (TWIN) is introduced, an implicit water MLP parametrized entirely by an Equivariant Graph Neural Network and trained solely on ab initio and experimental labels, consistently outperforming previous machine-learning-based implicit solvent or coarse-grained models.
Abstract
Machine learning interatomic potentials (MLPs) have revolutionized atomistic modeling, offering the potential to replace traditional methods like Density Functional Theory (DFT). However, inference time of MLPs is orders of magnitude slower than that of classical force fields, hindering real-world applications for biomolecular systems that require timescales of microseconds and beyond. Implicit solvent MLPs can address this issue, but are faced with data challenges associated with coarse-grained modeling. Consequently, previous approaches relied on empirical force field data, thereby inherently limiting the MLP's accuracy. Here, we introduce the Transferable Water Implicit Network (TWIN), an implicit water MLP parametrized entirely by an Equivariant Graph Neural Network and trained solely on ab initio and experimental labels. We demonstrate TWIN's transferability across drug-like molecules, peptides, and proteins, achieving excellent results on ab initio and experimental crystallographic and NMR benchmarks, consistently outperforming previous machine-learning-based implicit solvent or coarse-grained models. Furthermore, TWIN closely matches DFT-based explicit solvent MLPs while providing a two-order-of-magnitude faster timestep evaluation, paving the way for efficient ab initio-level modeling of biomolecular systems in aqueous environments.
DensIP is introduced, a physics-based model of intermolecular interactions that uses machine-learned electron densities and only four universal parameters that outperforms state-of-the-art general-purpose MLFFs for long-range interactions and can be applied to molecules as large as drug ligands.
Dahvyd Wing, Mihail Bogojeski, Szabolcs Góger et al.· 0 citations
Methanol-water mixtures find use in many applications, particularly catalytic energy conversion processes. Their importance has motivated numerous computational studies, most of which employed molecular dynamics based on classical force fields. These enable simulations of large systems on long time scales but do not reliably describe reactive dynamics involving bond breaking and bond formation. In contrast, ab initio molecular dynamics (AIMD) based on density functional theory (DFT) is generally more reliable for such applications but has a high computational cost, which discourages systematic studies of alcohol-water mixtures. To remedy this, we trained a machine learning interatomic potential capable of probing the properties of aqueous methanol mixtures at the DFT level using the SCAN functional. Our results show that SCAN qualitatively reproduces multiple key experimental features arising from the amphiphilic nature of methanol, including density, diffusion coefficients, X-ray structure factors, and Kirkwood-Buff integrals. We also find that structural correlations between water molecules are somewhat overestimated, leading to a stronger preferential association than that predicted by experiments. However, increasing the temperature by 30 K mitigates this effect and also recovers the correct mobilities of both methanol and water. These results indicate that SCAN provides an accurate description of methanol-water mixtures, making it a reliable choice for investigating the reactive dynamics in such systems.
Sanghyun J. Park, A. Selloni· Journal of Physical Chemistr...· 0 citations
By identifying AIM methods that result in conformationally stable and transferable atomic properties, this study aims to improve IDP modelling and guide the development of more accurate force fields for biologically relevant systems.
While machine-learning interatomic potentials (MLIPs) have successfully learned potential energy surfaces (PES) and atomic forces, many practical applications, such as vibrational analysis and transition state search, rely heavily on the PES Hessian. Yet, standard MLIPs tend to be trained on energy and forces alone, leaving Hessian information largely unexploited. Meanwhile, existing methods that explicitly incorporate the Hessian into training objectives require architectural modifications and introduce significant computational and memory overheads due to higher-order backpropagation. To address these limitations, we propose two Hessian-derived data augmentation schemes: isotropic Gaussian displacement (\textbf{UniAug}) and normal mode-weighted displacement (\textbf{ModeAug}). Both methods utilize simple Taylor expansions, achieving effective augmentation without altering training objectives or extending the autograd graph. This allows seamless, plug-and-play integration with existing architectures and training pipelines. Comprehensive evaluations across non-equilibrium and equilibrium datasets demonstrate that our approach enhances model accuracy while providing practical, task-specific guidelines.
This work introduces implicit machine learning force fields, which replace explicit stacks of neural network layers with self-consistent fixed-point equations, and demonstrates this across three major classes of graph neural networks: invariant, equivariant Cartesian tensor, and SO(3)-equivariant spherical-tensor architectures.
J. Maeß, Leon Werner, J. Frank et al.· arXiv.org· 0 citations
This study presents a method to derive optimized CV from transition state region (TS) via an interpretable machine learning (ML) model, Elastic Net, which greatly accelerate ligand binding-unbinding transitions and achieves rapid free energy surface (FES) convergence across diverse systems.
Saikat Dhibar, B. Jana· bioRxiv· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
