The Cronin Group

Research in the Cronin Group is motivated by the fascination for complex chemical systems, and the desire to construct complex functional molecular architectures that are not based on biologically derived building blocks.


...
Machine Learning Revolutionizes Host-Guest Chemistry with High-Accuracy Molecular Design

A new study from the Cronin Group has introduced a cutting-edge machine learning model that significantly advances the design of host-guest binders. Trained on electron density data, the model has achieved over 98% accuracy in converting molecular structures into SMILES format, allowing comprehensive two-dimensional characterization. Utilizing a variational autoencoder, the model generates detailed 3D electron density and electrostatic potential representations, optimizing guest molecules via gradient descent. Successfully applied to cucurbit[n]uril and metal–organic cages, the model discovered 9 previously validated guests for CB[6] and 7 unreported guests, as well as 4 unreported guests for [Pd214]4+, paving the way for more efficient molecular discovery and design in chemistry.

The research has been published in Nature Communications, and is open access.

...
Prof. Leroy (Lee) Cronin

Prof Leroy (Lee) Cronin
Regius Chair of Chemistry
Advanced Research Centre (ARC)
Level 5, Digital Chemistry
University of Glasgow
11 Chapel Lane
Glasgow G11 6EW
Tel: +44 141 330 6650
Email: lee.cronin@glasgow.ac.uk

Latest Publications

...

502. Delocalized, asynchronous, closed-loop discovery of organic laser emitters

...

501. Investigating and Quantifying Molecular Complexity Using Assembly Theory and Spectroscopy

...

500. Electron density-based GPT for optimization and suggestion of host–guest binders

...

499. Evidence of Selection in Mineral Mediated Polymerization Reactions Executed in a Robotic Chemputer System

...

498. A programmable hybrid digital chemical information processor based on the Belousov-Zhabotinsky reaction

...

497. An integrated self-optimizing programmable chemical synthesis and reaction engine

...

496. Autonomous execution of highly reactive chemical transformations in the Schlenkputer

...

495. Universal chemical programming language for robotic synthesis repeatability

...

494. Bringing digital synthesis to Mars

...

493. An Autonomous Electrochemical Discovery Robot that Utilises Probabilistic Algorithms: Probing the Redox Behaviour of Inorganic Materials


Find us on

Copyright © 2005 - 2024 Prof Lee Cronin - The University of Glasgow
Joseph Black Building, University of Glasgow, Scotland, UK
Visitors: