ConSolv: Solvent-Conditional Machine Learning Implicit Solvent Potential
- 1Multiscale Modeling of Fluid Materials, Department of Engineering Physics and Computation, TUM School of Engineering and Design, Technical University of Munich, 85748 Garching, Germany
- 2Atomistic Modeling Center (AMC), Munich Data Science Institute (MDSI), Technical University of Munich, 85748 Garching, Germany
Abstract
Implicit solvent machine learning potentials (MLPs) offer a powerful route to bridging the gap between accuracy and efficiency in molecular simulations. However, existing models have largely focused on aqueous environments, overlooking the diverse and important roles of nonaqueous solvents in areas such as organic synthesis and battery technology. Here, we present ConSolv, a solvent-conditional MLP architecture that explicitly incorporates solvent effects on solute interactions through an attention-based solvent embedding block. By combining experimental solvation free energy data with ab initio data, we train a single implicit solvent MLP that is transferable across 66 common organic solvents. ConSolv outperforms classical explicit solvent methods and selected ab initio implicit solvent approaches across multiple solvation free energy benchmarks, and demonstrates generalization to unseen solvents. Beyond solvation free energies, the model shows close agreement with experimental nuclear magnetic resonance (NMR) data for γ-fluorohydrin molecules in chloroform. ConSolv’s architecture is readily extensible to broader chemical spaces and alternative training strategies, while its attention-based design supports explainable artificial intelligence (AI) analysis that can help elucidate complex, solvent-dependent molecular interactions.
Related
- Sample-based quantum diagonalization approach for open-shell transition-metal complexes in gas and implicit-solvent
- Approximate additivity in the solvent-mediated potential of mean force for ultrasoft particle systems
- Reconstructing sign-switching dark energy histories: Scalar-field regularity, conditional potential comparison, and representative dynamics
- Physics-Constrained Conditional Generative Learning for Quantum State and Process Tomography
- Structured Hamiltonian Learning for Multiqubit Conditional Phase Gates
BibTeX
@article{zhang2026consolv,
title = {{ConSolv: Solvent-Conditional Machine Learning Implicit Solvent Potential}},
author = {Linying Zhang and Julija Zavadlav},
journal = {Journal of Chemical Theory and Computation},
year = {2026},
month = oct,
doi = {10.1021/acs.jctc.6c01253},
url = {https://doi.org/10.1021/acs.jctc.6c01253},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - ConSolv: Solvent-Conditional Machine Learning Implicit Solvent Potential
AU - Linying Zhang
AU - Julija Zavadlav
PY - 2026
DA - 2026/10/12
JO - Journal of Chemical Theory and Computation
DO - 10.1021/acs.jctc.6c01253
UR - https://doi.org/10.1021/acs.jctc.6c01253
AB - Implicit solvent machine learning potentials (MLPs) offer a powerful route to bridging the gap between accuracy and efficiency in molecular simulations. However, existing models have largely focused on aqueous environments, overlooking the diverse and important roles of nonaqueous solvents in areas such as organic synthesis and battery technology. Here, we present ConSolv, a solvent-conditional MLP architecture that explicitly incorporates solvent effects on solute interactions through an attention-based solvent embedding block. By combining experimental solvation free energy data with ab initio data, we train a single implicit solvent MLP that is transferable across 66 common organic solvents. ConSolv outperforms classical explicit solvent methods and selected ab initio implicit solvent approaches across multiple solvation free energy benchmarks, and demonstrates generalization to unseen solvents. Beyond solvation free energies, the model shows close agreement with experimental nuclear magnetic resonance (NMR) data for γ-fluorohydrin molecules in chloroform. ConSolv’s architecture is readily extensible to broader chemical spaces and alternative training strategies, while its attention-based design supports explainable artificial intelligence (AI) analysis that can help elucidate complex, solvent-dependent molecular interactions.
SN - 1549-9618
SN - 1549-9626
ER -