Explainable AI in drug discovery: self-interpretable graph neural network for molecular property prediction using concept whitening

Molecular property prediction is a fundamental task in the field of drug discovery. Several works use graph neural networks to leverage molecular graph representations. Although they have been successfully applied in a variety of applications, their decision process is not transparent. In this work,...

Full description

Saved in:
Bibliographic Details
Published in:Machine learning 2024-04, Vol.113 (4), p.2013-2044
Main Authors: Proietti, Michela, Ragno, Alessio, Rosa, Biagio La, Ragno, Rino, Capobianco, Roberto
Format: Article
Language:English
Subjects:
Adaptation
Approximation
Artificial Intelligence
Classification
Computer Science
Control
Explainable artificial intelligence
Graph neural networks
Graph representations
Graphical representations
Machine Learning
Mechatronics
Molecular properties
Natural Language Processing (NLP)
Neural networks
Robotics
Simulation and Modeling
Special Issue on Discovery Science 2021
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Molecular property prediction is a fundamental task in the field of drug discovery. Several works use graph neural networks to leverage molecular graph representations. Although they have been successfully applied in a variety of applications, their decision process is not transparent. In this work, we adapt concept whitening to graph neural networks. This approach is an explainability method used to build an inherently interpretable model, which allows identifying the concepts and consequently the structural parts of the molecules that are relevant for the output predictions. We test popular models on several benchmark datasets from MoleculeNet. Starting from previous work, we identify the most significant molecular properties to be used as concepts to perform classification. We show that the addition of concept whitening layers brings an improvement in both classification performance and interpretability. Finally, we provide several structural and conceptual explanations for the predictions.
ISSN:0885-6125
1573-0565
DOI:10.1007/s10994-023-06369-y