General-purpose machine-learning interatomic potentials (MLIPs) for organic reactions need to be accurate on both the minimum energy path (MEP) for static evaluation of basic properties and the broader configurational space for simulating reaction dynamics. Existing general datasets for gas-phase organic reactions rely on quasi-static relaxation that confines configurations to the MEP vicinity, so models trained on them could fail on direct molecular-dynamics trajectories; the gap is methodological, not a question of dataset size. We introduce a spatiotemporally resolved, system-independent collective variable (CV): Cartesian RMSD within randomly partitioned local domains against an expanding list of time-averaged reference geometries. The CV drives metadynamics as the main exploration engine, supplemented by structural relaxation towards transition state (TS) to augment the coverage around TS. Within a concurrent-learning workflow, this produces OpenRxn26, a dataset of 1.8~M DFT-labeled configurations covering neutral singlet unimolecular reactions in the H/C/N/O chemical space ($N_\mathrm{heavy} \leq 30$), containing reactive atomic environments underrepresented in community datasets. Trained on OpenRxn26, a DPA3 model (denoted DPA3_rxn) achieves transferable accuracy on barrier heights and reaction energies. On off-MEP reactive trajectories, DPA3_rxn is the only model in the benchmark suite to reach 1.0 kcal/mol energy accuracy compared with the labeling method, where the domain MLIP leading on static benchmarks degrades several-fold (e.g. MACE_OMol25), showing the insufficiency of quasi-static sampling and MEP-anchored benchmarks for guaranteeing dynamics reliability of MLIPs. OpenRxn26 thus provides MD-ready reactive training data for gas-phase neutral singlet organic reactions, verifying the generality and efficiency of the sampling strategy.
A potential is presented that is universal within the class of water-free solid-state hydrogen-bond network mediated proton conductors rather than across chemistry and a database of 4.4 million first-principles configurations spanning 55 materials, offering a template for other compound classes.
Jonas Hänseroth, Rose von Stackelberg, Christian Dreßler· 0 citations
Pretrained machine-learning interatomic potentials, so-called universal or foundation models offer an appealing starting point for atomistic simulations, but their accuracy for material-specific observables often remains limited without additional reference data (fine-tuning). Here, we systematically quantify how much...
Machine learning interatomic potentials (MLIPs) are transforming atomistic simulations by accessing unprecedented length and time scales. While pretrained equivariant graph neural networks achieve robust zero-shot performance for near-equilibrium properties across broad chemical spaces, their translation to complex mat...
This new efficiency finally allows to heads-on tackle the highly dynamic evolution of working interfaces in ECS systems, where the targeted functionality like catalytic activity or ion mobility both inherently drives and results from ongoing substantial structural, compositional and morphological changes.
E. Gelžinytė, Gábor Csányi, Janine George et al.· 0 citations
MARC-TS, a two-stage framework that learns a continuous, endpoint-conditioned path, queries it at any resolution and uses local path context to refine a transition-state candidate, is introduced.
Foundation machine learning interatomic potentials (MLIPs) deliver near-ab-initio accuracy at a fraction of the computational cost, yet their promise for Metal-organic Frameworks (MOFs) remains largely unrealized as large unit cells make first-principles training data expensive to generate, fine-tuned models are scarce...
T. J. Inizan, Prathami Divakar Kamath, Alin M. Elena et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.