AI model accurately classifies catalytic organic reaction mechanisms

Amazing stuff! This has huge potential!

"A new artificial intelligence (AI) tool can classify chemical reaction mechanisms using concentration data to make predictions that are 99.6% accurate with realistically noisy data. ...
The deep learning model ‘does not just match but surpasses what chemist experts on kinetics would be able to do with previous tools’ ...
chemistry is at a unique turning point for AI tools ...

[researchers] combined two different neural networks. First, a long short-term memory neural network tracks concentration changes over time. Second, a fully connected neural network processes what comes out of that first network.

The final model contains 576,000 trainable parameters. ...
For comparison, AlphaFold uses 21 million parameters and GPT3 uses 175 billion parameters,’ ...
Once the model has learned to recognise the characteristics of the kinetic data associated with each reaction mechanism it ‘applies those rules to new input kinetic data to classify it’ ..."

From the abstract:

"A mechanistic understanding of catalytic organic reactions is crucial for the design of new catalysts, modes of reactivity and the development of greener and more sustainable chemical processes. ... Here we show that a deep neural network model can be trained to analyse ordinary kinetic data and automatically elucidate the corresponding mechanism class, without any additional user input. The model identifies a wide variety of classes of mechanism with outstanding accuracy, including mechanisms out of steady state such as those involving catalyst activation and deactivation steps, and performs excellently even when the kinetic data contain substantial error or only a few time points. Our results demonstrate that artificial-intelligence-guided mechanism classification is a powerful new tool that can streamline and automate mechanistic elucidation. ..."

AI model accurately classifies reaction mechanisms | Research | Chemistry World

Organic reaction mechanism classification using machine learning (no public access)

The new AI model can interpret kinetic data to determine which of 20 common reaction mechanisms determine the relationship between the substrate (S), the catalyst (cat) and the product (P)