Recommended by

Indexed by

Reaction-based machine learning representations for predicting the enantioselectivity of organocatalysts

Simone Gallarati1, Raimon Fabregat1, Rubén Laplaza1,2, Sinjini Bhattacharjee1,3, Matthew Wodrich1, Clemence Corminboeuf1,2,4*

1 Laboratory for Computational Molecular Design, Institute of Chemical Sciences and Engineering, Ecole Polytechnique Fédérale de Lausanne (EPFL), 1015 Lausanne, Switzerland

2 National Center for Competence in Research - Catalysis (NCCR-Catalysis), Ecole Polytechnique Fédérale de Lausanne (EPFL), 1015 Lausanne, Switzerland

3 Indian Institute of Science Education and Research, Dr. Homi Bhabha Rd, Ward No. 8, NCL Colony, Pashan, Pune, Maharashtra 4110008, India

4 National Center for Computational Design and Discovery of Novel Materials (MARVEL), Ecole Polytechnique Fédérale de Lausanne (EPFL), 1015 Lausanne, Switzerland

* Corresponding authors emails: clemence.corminboeuf@epfl.ch
DOI10.24435/materialscloud:vp-h5 [version v1]

Publication date: Mar 05, 2021

How to cite this record

Simone Gallarati, Raimon Fabregat, Rubén Laplaza, Sinjini Bhattacharjee, Matthew Wodrich, Clemence Corminboeuf, Reaction-based machine learning representations for predicting the enantioselectivity of organocatalysts, Materials Cloud Archive 2021.40 (2021), doi: 10.24435/materialscloud:vp-h5.


Hundreds of catalytic methods are developed each year to meet the demand for high-purity chiral compounds. The computational design of enantioselective organocatalysts remains a significant challenge, as catalysts are typically discovered through experimental screening. Recent advances in combining quantum chemical computations and machine learning (ML) hold great potential to propel the next leap forward in asymmetric catalysis. Within the context of quantum chemical machine learning (QML, or atomistic ML), the ML representations used to encode the structure of molecules and evaluate their similarity cannot easily capture the subtle energy differences that govern enantioselectivity. Here, we present a general strategy for improving molecular representations within an atomistic machine learning model to predict the enantiomeric excess of asymmetric propargylation organocatalysts solely from the structure of catalytic cycle intermediates. Mean absolute errors as low as 0.25 kcal mol-1 were achieved in predictions of the activation energy. This strategy opens the door for quickly and accurately predicting higher-selectivity catalysts for any reaction from available structural information.

Materials Cloud sections using this data

No Explore or Discover sections associated with this archive record.


File name Size Description
2.2 MiB See README.txt


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.

External references

Journal reference (Manuscript under consideration for publication.)
S. Gallarati, R. Fabregat, R. Laplaza, S. Bhattacharjee, M. D. Wodrich, C. Corminboeuf, Chem. Sci., under consideration.


catalysis machine learning enantioselectivity organocatalyst MARVEL

Version history:

2021.40 (version v1) [This version] Mar 05, 2021 DOI10.24435/materialscloud:vp-h5