Deep learning of chemical reactions

Introduction

The exploration of reactions is a central topic in chemistry. Compared to the success of machine learning for molecules, the modeling of reactions is lagging behind, especially for stereo- and regioselective reactions.

Need for New Approaches

Since current efforts toward sustainable synthesis, such as asymmetric organocatalysis or biocatalysis, rely on the accurate prediction of enantio- and regioselective reaction pathways, new modeling approaches are needed.

Project Aim

The proposed project aims to develop new, data-driven deep learning frameworks for modeling organic and enzymatic reactions, focusing on:

1. Chemo-selectivity
2. Regio-selectivity
3. Stereoselectivity arising through intermolecular interactions with the reagent, solvent, or catalyst.

Detailed Objectives

In detail, we target:

• The rule-free, stereochemistry-aware modeling and subsequent experimental validation of asymmetric organocatalysis to identify new enantioselective transformations.
• The exploration of new biocatalytic synthesis pathways, including enzymatic cascades.
• The accurate prediction of activation energies via developing new deep learning approaches.

Methodology

We will expand molecular graph-convolutional neural networks and graph transformers to reactions in a rule-free manner. Additionally, we will introduce hidden three-dimensional representations to account for stereochemistry and intermolecular interactions. This will yield a versatile, open-source toolbox for reaction deep learning.

Significance of the Approach

This approach largely surpasses current methods, which rely on:

• Two-dimensional representations
• Reaction rules
• Three-dimensional input data

It offers the opportunity to model three-dimensional aspects and atom-mapping on-the-fly, for the first time, representing a significant breakthrough in this field.

Experimental Validation

The experimental validation campaign further allows for a direct application to the identification of new asymmetric organocatalytic transformations, as well as enzymatic cascades including cofactor recycling and side-product reduction, addressing the current need for more sustainable synthesis.