arXiv subjects

Egor Rumiantsev

Publications and source records attributed to Egor Rumiantsev.

6 recordsLinked to original sources

arXiv2026

Simultaneous Learning of Static and Dynamic Charges

Long-range interactions and electric response are essential for accurate modeling of condensed-phase systems, but capturing them efficiently remains a challenge for atomistic machine learning. Traditionally, these two phenomena can be represented by static charges, that participate in Coulomb interactions between atoms, and dynamic charges such as atomic polar tensors - aka Born effective charges - describing the response to an external electric field. We critically compare different approaches to learn both types of charges, taking bulk water and water clusters as paradigmatic examples: (1) Learning them independently; (2) Coupling static and dynamic charges based on their physical relationship with a single global coupling constant to account for dielectric screening; (3) Coupled learning with a local, environment-dependent screening factor. In the coupled case, correcting for dielectric screening is essential, yet the common assumption of homogeneous, isotropic screening breaks down in heterogeneous systems such as water clusters. A learned, environment-dependent screening restores high accuracy for the dynamical charges. However, the accuracy gain over independent dynamic predictions is negligible, while the computational cost increases compared to using separate models for static and dynamical charges. This suggests that, despite the formal connection between the two charge types, modeling them independently is the more practical choice for both condensed-phase and isolated cluster systems.

physics.chem-ph↗

arXiv2025

Learning Long-Range Representations with Equivariant Messages

Machine learning interatomic potentials trained on first-principles reference data are becoming valuable tools for computational physics, biology, and chemistry. Equivariant message-passing neural networks, including transformers, achieve state-of-the-art accuracy but rely on cutoff-based graphs, limiting their ability to capture long-range effects such as electrostatics or dispersion, as well as electron delocalization. While long-range correction schemes based on inverse power laws of interatomic distances have been proposed, they are unable to communicate higher-order geometric information and are thus limited in applicability. To address this shortcoming, we propose the use of equivariant, rather than scalar, charges for long-range interactions, and design a graph neural network architecture, LOREM, around this long-range message passing mechanism. We consider several datasets specifically designed to highlight non-local physical effects, and compare short-range message passing with different receptive fields to invariant and equivariant long-range message passing. Even though most approaches work for careful dataset-specific choices of their model hyperparameters, LOREM works consistently without such changes, with excellent benchmark performance.

physics.chem-ph↗

arXiv2024

Fast and flexible long-range models for atomistic machine learning

Most atomistic machine learning (ML) models rely on a locality ansatz, and decompose the energy into a sum of short-ranged, atom-centered contributions. This leads to clear limitations when trying to describe problems that are dominated by long-range physical effects - most notably electrostatics. Many approaches have been proposed to overcome these limitations, but efforts to make them efficient and widely available are hampered by the need to incorporate an ad hoc implementation of methods to treat long-range interactions. We develop a framework aiming to bring some of the established algorithms to evaluate non-bonded interactions - including Ewald summation, classical particle-mesh Ewald (PME), and particle-particle/particle-mesh (P3M) Ewald - into atomistic ML. We provide a reference implementation for pyTorch as well as an experimental one for JAX. Beyond Coulomb and more general long-range potentials, we introduce purified descriptors which disregard the immediate neighborhood of each atom, and are more suitable for general long-ranged ML applications. Our implementations are fast, feature-rich, and modular: They provide an accurate evaluation of physical long-range forces that can be used in the construction of (semi)empirical baseline potentials; they exploit the availability of automatic differentiation to seamlessly combine long-range models with conventional, local ML schemes; and they are sufficiently flexible to implement more complex architectures that use physical interactions as building blocks. We benchmark and demonstrate our torch-pme and jax-pme libraries to perform molecular dynamics simulations, to train ML potentials, and to evaluate long-range equivariant descriptors of atomic structures.

physics.chem-ph↗

arXiv2024

$\nabla^2$DFT: A Universal Quantum Chemistry Dataset of Drug-Like Molecules and a Benchmark for Neural Network Potentials

Methods of computational quantum chemistry provide accurate approximations of molecular properties crucial for computer-aided drug discovery and other areas of chemical science. However, high computational complexity limits the scalability of their applications. Neural network potentials (NNPs) are a promising alternative to quantum chemistry methods, but they require large and diverse datasets for training. This work presents a new dataset and benchmark called $\nabla^2$DFT that is based on the nablaDFT. It contains twice as much molecular structures, three times more conformations, new data types and tasks, and state-of-the-art models. The dataset includes energies, forces, 17 molecular properties, Hamiltonian and overlap matrices, and a wavefunction object. All calculations were performed at the DFT level ($\omega$B97X-D/def2-SVP) for each conformation. Moreover, $\nabla^2$DFT is the first dataset that contains relaxation trajectories for a substantial number of drug-like molecules. We also introduce a novel benchmark for evaluating NNPs in molecular property prediction, Hamiltonian prediction, and conformational optimization tasks. Finally, we propose an extendable framework for training NNPs and implement 10 models within it.

physics.chem-ph↗

arXiv2024

Gradual Optimization Learning for Conformational Energy Minimization

Molecular conformation optimization is crucial to computer-aided drug discovery and materials design. Traditional energy minimization techniques rely on iterative optimization methods that use molecular forces calculated by a physical simulator (oracle) as anti-gradients. However, this is a computationally expensive approach that requires many interactions with a physical simulator. One way to accelerate this procedure is to replace the physical simulator with a neural network. Despite recent progress in neural networks for molecular conformation energy prediction, such models are prone to distribution shift, leading to inaccurate energy minimization. We find that the quality of energy minimization with neural networks can be improved by providing optimization trajectories as additional training data. Still, it takes around $5 \times 10^5$ additional conformations to match the physical simulator's optimization quality. In this work, we present the Gradual Optimization Learning Framework (GOLF) for energy minimization with neural networks that significantly reduces the required additional data. The framework consists of an efficient data-collecting scheme and an external optimizer. The external optimizer utilizes gradients from the energy prediction model to generate optimization trajectories, and the data-collecting scheme selects additional training data to be processed by the physical simulator. Our results demonstrate that the neural network trained with GOLF performs on par with the oracle on a benchmark of diverse drug-like molecules using $50$x less additional data.

physics.chem-ph↗

arXiv2024

FREED++: Improving RL Agents for Fragment-Based Molecule Generation by Thorough Reproduction

A rational design of new therapeutic drugs aims to find a molecular structure with desired biological functionality, e.g., an ability to activate or suppress a specific protein via binding to it. Molecular docking is a common technique for evaluating protein-molecule interactions. Recently, Reinforcement Learning (RL) has emerged as a promising approach to generating molecules with the docking score (DS) as a reward. In this work, we reproduce, scrutinize and improve the recent RL model for molecule generation called FREED (arXiv:2110.01219). Extensive evaluation of the proposed method reveals several limitations and challenges despite the outstanding results reported for three target proteins. Our contributions include fixing numerous implementation bugs and simplifying the model while increasing its quality, significantly extending experiments, and conducting an accurate comparison with current state-of-the-art methods for protein-conditioned molecule generation. We show that the resulting fixed model is capable of producing molecules with superior docking scores compared to alternative approaches.

q-bio.BM↗