Citing SODA

When publishing results obtained with SODA, cite the SODA preprint, the workflow methods, and the interatomic potential or model used in your calculation. The workflows are separate scientific works with their own methods and references. The SODA reference describes the supporting execution and data-aggregation framework. Choose the references relevant to the calculation you report.

SODA preprint

Lopez, C., Lai, K. C., Rampp, M., and Scheurer, C. “A scalable framework for goal-driven task-based simulations.” ChemRxiv, preprint, version 1, 15 September 2026. doi:10.26434/chemrxiv.15008845/v1.

@article{lopez2026soda,
  author  = {Lopez, Carlos and Lai, King Chun and Rampp, Markus and Scheurer, Christoph},
  title   = {A scalable framework for goal-driven task-based simulations},
  journal = {ChemRxiv},
  year    = {2026},
  month   = sep,
  doi     = {10.26434/chemrxiv.15008845/v1},
  url     = {https://doi.org/10.26434/chemrxiv.15008845/v1},
  note    = {Preprint, version 1, posted 15 September 2026}
}

SODA software

The repository’s CITATION.cff contains the software citation metadata and the preprint as its preferred citation. Use the metadata from the release or commit used for your results, and include its version and repository URL in your software reference. For a development checkout, record the full Git commit as well.

From the source directory used to build SODA, you can record the revision with:

git describe --tags --always --dirty
git rev-parse HEAD

Keep any local modifications with your research records. A commit identifier alone does not capture changes made after that commit.

A methods statement can follow this pattern, replacing the bracketed text with the details of your calculation and adding the corresponding references:

Calculations used the [workflow] method, with [workflow software/version, where applicable]. SODA (Scalable Orchestration and Data Aggregation), version [version], commit [commit], provided parallel execution and data aggregation. Energy and force evaluations used [potential/model name and version].

Workflow references

These references describe the independent methods and frameworks that can be run through the available SODA integrations. The example READMEs identify the references relevant to particular example calculations.

APE

Lai, K. C., et al. “Automatic Process Exploration through Machine Learning Assisted Transition State Searches.” Physical Review Letters 134, 096201 (2025). doi:10.1103/PhysRevLett.134.096201.

EZGA

Lombardi, J. M., et al. Computer Physics Communications 328, 110354 (2026). doi:10.1016/j.cpc.2026.110354.

The repository reference list also records the EZGA preprint, version 1. Use the journal reference for the published work; use the preprint reference when referring specifically to that version.

NEBScape

Jung, H., et al. “From global optimization to transition state search: an automated workflow for surface reaction barriers.” npj Computational Materials 12, 300 (2026). doi:10.1038/s41524-026-02298-1.

AKMC / EON

AKMC method: Xu, L., and Henkelman, G. “Adaptive kinetic Monte Carlo for first-principles accelerated dynamics.” The Journal of Chemical Physics 129(11), 114104 (2008). doi:10.1063/1.2976010.

EON software: Chill, S. T., et al. “EON: software for long time simulations of atomic scale systems.” Modelling and Simulation in Materials Science and Engineering 22(5), 055002 (2014). doi:10.1088/0965-0393/22/5/055002.

Source provenance: SODA’s EON-derived components are based on EON, SVN revision r2545. Identify this revision alongside the method and software references when describing their origin. See Third-party notices for the incorporated components and retained license.

Potentials, models, and simulation engines

The workflow reference describes how structures or processes are explored. Also cite the method and parameterization used to calculate their energies and forces. Record the model name, version or checkpoint, and its source alongside your input files.

Used in your calculation	References to include
EAM	The EAM review and the publication for the actual EAM parameter file. For the Foiles–Baskes–Daw parameterization used in the examples, see the reference below.
GAP	The GAP method and the specific trained model. The Pd/O example model reference is listed below.
MACE	The MACE method, any additional applicable MACE references, and the publication or model card for the exact checkpoint, including the base model when fine-tuning.
Fairchem	The FAIRChem software citation, plus the method, model, and dataset references provided with the chosen checkpoint in the Fairchem project.
LAMMPS	Follow the LAMMPS citation instructions, including references for the potential styles or algorithms used.
Other calculators or models	Follow the calculator’s citation instructions and cite the specific potential, parameter set, or model used.

EAM

Daw, M. S., Foiles, S. M., and Baskes, M. I. “The embedded-atom method: a review of theory and applications.” Materials Science Reports 9(7–8), 251–310 (1993). doi:10.1016/0920-2307(93)90001-U.

GAP

Bartók, A. P., Payne, M. C., Kondor, R., and Csányi, G. “Gaussian Approximation Potentials: The Accuracy of Quantum Mechanics, without the Electrons.” Physical Review Letters 104(13), 136403 (2010). doi:10.1103/PhysRevLett.104.136403.

MACE

Batatia, I., Kovács, D. P., Simm, G. N. C., Ortner, C., and Csányi, G. “MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate Force Fields.” Advances in Neural Information Processing Systems 35, 11423–11436 (2022). doi:10.52202/068431-0830.

Fairchem

Shuaibi, M., et al. FAIRChem [software]. Zenodo. doi:10.5281/zenodo.15587497.

This follows the software citation in the project’s CITATION.cff. Record the software version you use and also cite the selected model and datasets according to its model card.

EAM example parameterization

Foiles, S. M., Baskes, M. I., and Daw, M. S. Physical Review B 33, 7983 (1986). doi:10.1103/PhysRevB.33.7983.

GAP example model

Poths, P., Lai, K. C., Cannizzaro, F., Scheurer, C., Matera, S., and Reuter, K. “ML-Accelerated Automatic Process Exploration Reveals Facile O-Induced Pd Step-Edge Restructuring on Catalytic Time Scales.” ACS Catalysis 15(1), 514–522 (2025). doi:10.1021/acscatal.4c06414.

These example references apply to those parameterizations. If you use a different potential file or retrain a model, record and cite that choice.

For copyright notices, redistribution conditions, and the retained upstream license texts, see Third-party notices.