×

Recommended by

Indexed by

3DReact: geometric deep learning for chemical reactions

Puck van Gerwen1,2*, Ksenia Briling1,3*, Charlotte Bunne4,2*, Vignesh Ram Somnath4,2*, Ruben Laplaza1,2*, Andreas Krause4,2*, Clemence Corminboeuf1,2,3*

1 Laboratory for Computational Molecular Design, Institute of Chemical Sciences and Engineering, École Polytechnique Fédérale de Lausanne, 1015 Lausanne, Switzerland

2 National Centre for Competence in Research − Catalysis (NCCR-Catalysis), École Polytechnique Fédérale de Lausanne, 1015 Lausanne, Switzerland

3 National Centre for Computational Design and Discovery of Novel Materials (MARVEL), École Polytechnique Fédérale de Lausanne, 1015 Lausanne, Switzerland

4 Learning & Adaptive Systems Group, Department of Computer Science, ETH Zurich, 8092 Zurich, Switzerland

* Corresponding authors emails: puck.vangerwen@epfl.ch, ksenia.briling@epfl.ch, Charlotte.Bunne@epfl.ch, vsomnath@inf.ethz.ch, ruben.laplazasolanas@epfl.ch, krausea@ethz.ch, clemence.corminboeuf@epfl.ch
DOI10.24435/materialscloud:xd-ef [version v1]

Publication date: Oct 15, 2024

How to cite this record

Puck van Gerwen, Ksenia Briling, Charlotte Bunne, Vignesh Ram Somnath, Ruben Laplaza, Andreas Krause, Clemence Corminboeuf, 3DReact: geometric deep learning for chemical reactions, Materials Cloud Archive 2024.161 (2024), https://doi.org/10.24435/materialscloud:xd-ef

Description

Geometric deep learning models, which incorporate the relevant molecular symmetries within the neural network architecture, have considerably improved the accuracy and data efficiency of predictions of molecular properties. Building on this success, we introduce 3DREACT, a geometric deep learning model to predict reaction properties from three-dimensional structures of reactants and products. We demonstrate that the invariant version of the model is sufficient for existing reaction data sets. We illustrate its competitive performance on the prediction of activation barriers on the GDB7-22-TS, Cyclo-23-TS, and Proparg-21-TS data sets in different atom-mapping regimes. We show that, compared to existing models for reaction property prediction, 3DREACT offers a flexible framework that exploits atom- mapping information, if available, as well as geometries of reactants and products (in an invariant or equivariant fashion). Accordingly, it performs systematically well across different data sets, atom-mapping regimes, as well as both interpolation and extrapolation tasks.

Materials Cloud sections using this data

No Explore or Discover sections associated with this archive record.

Files

File name Size Description
3DReact.tar.gz
MD5md5:b3f1d58d62efcedeb8aebaab749a7c36
399.3 MiB See README
README.md
MD5md5:28d4cf559ff4074788b36a7e09b17ba8
1.7 KiB README.md
README.txt
MD5md5:28d4cf559ff4074788b36a7e09b17ba8
1.7 KiB README.txt

License

Files and data are licensed under the terms of the following license: Materials Cloud non-exclusive license to distribute v1.0.
Metadata, except for email addresses, are licensed under the Creative Commons Attribution Share-Alike 4.0 International license.

External references

Journal reference (Paper that uses the data)
P. van Gerwen, K. R. Briling, C. Bunne, V. R. Somnath, R. Laplaza, A. Krause, C. Corminboeuf, J. Chem. Inf. Model. 64, 5771-5785 (2024) doi:10.1021/acs.jcim.4c00104
Software (Code and results that use the data)

Keywords

reaction barriers geometric deep learning equivariant neural networks

Version history:

2024.161 (version v1) [This version] Oct 15, 2024 DOI10.24435/materialscloud:xd-ef