Physical Chemistry

Auto3D: Automatic Generation of the Low-energy 3D Structures with ANI Neural Network Potentials



Computational programs accelerate the chemical discovery processes but often need proper 3-dimensional molecular information as part of the input. Getting optimal molecular structures is challenging because it requires enumerating and optimizing a huge space of stereoisomers and conformers. We developed the python-based Auto3D package for generating the low-energy 3D structures using SMILES as the input. Auto3D is based on state-of-the-art algorithms and can automize the isomer enumeration and duplicate filtering process, 3D building process, geometry optimization and ranking process. Tested on 50 molecules with multiple unspecified stereocenters, Auto3D is guaranteed to find the stereo-configuration that yields the lowest-energy conformer. With Auto3D we provide an extension of the ANI model. The new model, dubbed ANI-2xt, is trained on a tautomer-rich dataset. ANI-2xt is benchmarked with DFT methods on geometry optimization, electronic and Gibbs free energy calculations. Compared with ANI-2x, ANI-2xt provides a 42% error reduction for tautomeric reaction energy calculations when using the gold-standard coupled-cluster calculation as the reference. ANI-2xt can accurately predict the energies with a ~10^6 factor speedup compared to DFT.


Thumbnail image of tautomer_workflow_final.pdf

Supplementary material

Thumbnail image of
SI materials
CHEMBL.smi contains 50 molecules that are used for validating the isomer enumeration step. Geometry.smi contains 2810 molecules that are used for benchmarking geometry optimization. tautomerization_E.smi contains 2824 molecules that are used to calculate tautomeric reaction energies. tautomerization_G.smi contains 162 molecules that are used to calculate tautomerization Gibbs free energies.

Supplementary weblinks

Auto3D GitHub Repository
Auto3D GitHub Repository