Citations & Licenses

How to cite work using Tako

Any publication using Tako should cite it as follows:

Tako.atom.li, an atomistic simulation toolkit. https://tako.atom.li (accessed yyyy-mm-dd).

Citing methods

Cite the papers for the methods that actually ran. The levelOfTheory recorded in result.json identifies the calculator; the Calculator Reference maps each level to its engine and checkpoint.

GFN2 calculator (gfn2)

Tako’s built-in tight-binding engine implements the GFN2-xTB Hamiltonian.

• C. Bannwarth, S. Ehlert, S. Grimme. GFN2-xTB—An Accurate and Broadly Parametrized Self-Consistent Tight-Binding Quantum Chemical Method with Multipole Electrostatics and Density-Dependent Dispersion Contributions. J. Chem. Theory Comput. 15, 1652–1671 (2019). doi:10.1021/acs.jctc.8b01176

When dispersion resolves to d4, additionally cite the D4 dispersion model:

• E. Caldeweyher, S. Ehlert, A. Hansen, H. Neugebauer, S. Spicher, C. Bannwarth, S. Grimme. A generally applicable atomic-charge dependent London dispersion correction. J. Chem. Phys. 150, 154122 (2019). doi:10.1063/1.5090222

For UV-Vis spectra from the sTDA operation, cite the simplified Tamm-Dancoff approach:

• S. Grimme. A simplified Tamm-Dancoff density functional approach for the electronic excitation spectra of very large molecules. J. Chem. Phys. 138, 244104 (2013). doi:10.1063/1.4811331
• S. Grimme, C. Bannwarth. Ultra-fast computation of electronic spectra for large systems by tight-binding based simplified Tamm-Dancoff approach (sTDA-xTB). J. Chem. Phys. 145, 054103 (2016). doi:10.1063/1.4959605

Machine-learned potentials

For the NequIP-family calculators (nequip-s, nequip-l, and the Nequix models), cite the NequIP architecture:

• S. Batzner, A. Musaelian, L. Sun, M. Geiger, J. P. Mailoa, M. Kornbluth, N. Molinari, T. E. Smidt, B. Kozinsky. E(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials. Nat. Commun. 13, 2453 (2022). doi:10.1038/s41467-022-29939-5

Each specific checkpoint has its own provenance, training data, and publication requirements:

Follow each upstream card or release for the checkpoint citation and any dataset citations it requires. For strict reproducibility, also record the weight SHA-256 listed in Checkpoint provenance.

Model weight licenses

Downloaded model weights are redistributed unmodified and remain under their upstream licenses; consult the checkpoint sources in the table above before redistributing weights or using them commercially.

Third-party software licenses

Tako bundles the following open-source components. Full license texts are distributed in the repository’s LICENSE file and each project’s source distribution.