Volume 13, Issue 4 e1651
Overview

Graph neural networks for conditional de novo drug design

Carlo Abate

Carlo Abate

Fondazione Istituto Italiano di Tecnologia, Genoa, Italy

Università degli Studi di Bologna, Bologna, Italy

Contribution: Conceptualization (equal), Formal analysis (lead), Writing - original draft (lead), Writing - review & editing (supporting)

Search for more papers by this author
Sergio Decherchi

Corresponding Author

Sergio Decherchi

Fondazione Istituto Italiano di Tecnologia, Genoa, Italy

Correspondence

Sergio Decherchi, Fondazione Istituto Italiano di Tecnologia, Genoa, Italy.

Email: [email protected]

Contribution: Conceptualization (equal), Formal analysis (supporting), Supervision (lead), Writing - original draft (supporting), Writing - review & editing (lead)

Search for more papers by this author
Andrea Cavalli

Andrea Cavalli

Fondazione Istituto Italiano di Tecnologia, Genoa, Italy

Università degli Studi di Bologna, Bologna, Italy

Contribution: Conceptualization (supporting), Supervision (supporting), Writing - review & editing (supporting)

Search for more papers by this author
First published: 12 January 2023
Citations: 1
Edited by: Peter Schreiner, Editor-in-Chief

Abstract

Drug design is costly in terms of resources and time. Generative deep learning techniques are using increasing amounts of biochemical data and computing power to pave the way for a new generation of tools and methods for drug discovery and optimization. Although early methods used SMILES strings, more recent approaches use molecular graphs to naturally represent chemical entities. Graph neural networks (GNNs) are learning models that can natively process graphs. The use of GNNs in drug discovery is growing exponentially. GNNs for drug design are often coupled with conditioning techniques to steer the generation process towards desired chemical and biological properties. These conditioned graph-based generative models and frameworks hold promise for the routine application of GNNs in drug discovery.

This article is categorized under:

  • Data Science > Artificial Intelligence/Machine Learning
  • Data Science > Chemoinformatics
  • Data Science > Computer Algorithms and Programming

Graphical Abstract

The number of tools involving deep learning in the drug discovery domain is growing exponentially. Recent advances make use of graph representations of molecules and graph neural network-based deep learning frameworks coupled with conditioning techniques.

CONFLICT OF INTEREST

Sergio Decherchi and Andrea Cavalli declare a conflict of interest as they are co-founders of BiKi Technologies s.r.l., a company selling the BiKi Life Sciences software suite for computational drug discovery.

DATA AVAILABILITY STATEMENT

Data sharing is not applicable to this article as no new data were created or analyzed in this study