Recommended by

Indexed by

3DMolNet: a generative network for molecular structures

Vitali Nesterov1*, Mario Wieser1, Volker Roth1*

1 Department of Mathematics and Computer Science, University of Basel, Switzerland

* Corresponding authors emails: vitali.nesterov@unibas.ch, volker.roth@unibas.ch
DOI10.24435/materialscloud:g6-ft [version v1]

Publication date: Nov 28, 2021

How to cite this record

Vitali Nesterov, Mario Wieser, Volker Roth, 3DMolNet: a generative network for molecular structures, Materials Cloud Archive 2021.199 (2021), doi: 10.24435/materialscloud:g6-ft.


With the recent advances in machine learning for quantum chemistry, it is now possible to predict the chemical properties of compounds and to generate novel molecules. Existing generative models mostly use a string- or graph-based representation, but the precise three-dimensional coordinates of the atoms are usually not encoded. First attempts in this direction have been proposed, where autoregressive or GAN-based models generate atom coordinates. Those either lack a latent space in the autoregressive setting, such that a smooth exploration of the compound space is not possible, or cannot generalize to varying chemical compositions. We propose a new approach to efficiently generate molecular structures that are not restricted to a fixed size or composition. Our model is based on the variational autoencoder which learns a translation-, rotation-, and permutation-invariant low-dimensional representation of molecules. Our experiments yield a mean reconstruction error below 0.05 Angstrom, outperforming the current state-of-the-art methods by a factor of four, and which is even lower than the spatial quantization error of most chemical descriptors. The compositional and structural validity of newly generated molecules has been confirmed by quantum chemical methods in a set of experiments.

Materials Cloud sections using this data

No Explore or Discover sections associated with this archive record.


File name Size Description
169.0 KiB Novel molecules discovered with 3DMolNet


Files and data are licensed under the terms of the following license: Creative Commons Attribution 4.0 International.
Metadata, except for email addresses, are licensed under the Creative Commons Attribution Share-Alike 4.0 International license.


generative model molecules discovery three-dimensional structures MARVEL/Inc2 SNSF

Version history:

2021.199 (version v1) [This version] Nov 28, 2021 DOI10.24435/materialscloud:g6-ft