Mapping the Space of Chemical Reactions Using Attention-Based Neural Networks

by   Philippe Schwaller, et al.

Organic reactions are usually assigned to classes containing reactions with similar reagents and mechanisms. Reaction classes facilitate the communication of complex concepts and efficient navigation through chemical reaction space. However, the classification process is a tedious task. It requires the identification of the corresponding reaction class template via annotation of the number of molecules in the reactions, the reaction center, and the distinction between reactants and reagents. This work shows that transformer-based models can infer reaction classes from non-annotated, simple text-based representations of chemical reactions. Our best model reaches a classification accuracy of 98.2 can be used as reaction fingerprints that capture fine-grained differences between reaction classes better than traditional reaction fingerprints. The insights into chemical reaction space enabled by our learned fingerprints are illustrated by an interactive reaction atlas providing visual clustering and similarity searching.



There are no comments yet.


page 11


A generative model for molecule generation based on chemical reaction trees

Deep generative models have been shown powerful in generating novel mole...

Chemical Transformation Motifs - Modelling Pathways as Integer Hyperflows

We present an elaborate framework for formally modelling pathways in che...

Chebyshev approximation and the global geometry of sloppy models

Sloppy models are complex nonlinear models with outcomes that are signif...

Neural Network Based in Silico Simulation of Combustion Reactions

Understanding and prediction of the chemical reactions are fundamental d...

Cayley Graphs of Semigroups Applied to Atom Tracking in Chemistry

While atom tracking with isotope-labeled compounds is an essential and s...

Anytime Prediction as a Model of Human Reaction Time

Neural networks today often recognize objects as well as people do, and ...

Evolutionary Artificial Neural Network Based on Chemical Reaction Optimization

Evolutionary algorithms (EAs) are very popular tools to design and evolv...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.