Automatic identification of chemical moieties

In recent years, the prediction of quantum mechanical observables with machine learning methods has become increasingly popular. Message-passing neural networks (MPNNs) solve this task by constructing atomic representations, from which the properties of interest are predicted. Here, we introduce a m...

Full description

Saved in:
Bibliographic Details
Published in:Physical chemistry chemical physics : PCCP 2023-10, Vol.25 (38), p.2637-26379
Main Authors: Lederer, Jonas, Gastegger, Michael, Schütt, Kristof T, Kampffmeyer, Michael, Müller, Klaus-Robert, Unke, Oliver T
Format: Article
Language:English
Subjects:
Chemical fingerprinting
Chemistry
Knowledge representation
Machine learning
Message passing
Neural networks
Quantum mechanics
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In recent years, the prediction of quantum mechanical observables with machine learning methods has become increasingly popular. Message-passing neural networks (MPNNs) solve this task by constructing atomic representations, from which the properties of interest are predicted. Here, we introduce a method to automatically identify chemical moieties (molecular building blocks) from such representations, enabling a variety of applications beyond property prediction, which otherwise rely on expert knowledge. The required representation can either be provided by a pretrained MPNN, or be learned from scratch using only structural information. Beyond the data-driven design of molecular fingerprints, the versatility of our approach is demonstrated by enabling the selection of representative entries in chemical databases, the automatic construction of coarse-grained force fields, as well as the identification of reaction coordinates. A versatile, transferable and differentiable method to automatically identify chemical moieties based on message passing neural network feature representations.
ISSN:1463-9076
1463-9084
1463-9084
DOI:10.1039/d3cp03845a