Geometric deep learning for drug discovery
Drug discovery is a time-consuming and expensive process. With the development of Artificial Intelligence (AI) techniques, molecular Geometric Deep Learning (GDL) has recently emerged for learning from molecules and accelerating drug discovery. However, there is a gap between researchers in the fiel...
Gespeichert in:
Veröffentlicht in: | Expert systems with applications 2024-04, Vol.240, p.122498, Article 122498 |
---|---|
Hauptverfasser: | , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Drug discovery is a time-consuming and expensive process. With the development of Artificial Intelligence (AI) techniques, molecular Geometric Deep Learning (GDL) has recently emerged for learning from molecules and accelerating drug discovery. However, there is a gap between researchers in the field of deep learning and biomedicine. This review provides a comprehensive overview of the recent literature on GDL for drug discovery based on molecular three-dimensional (3D) representation learning and symmetry learning, highlighting its applications of molecular property prediction, intermolecular interaction, molecular design, molecular conformation prediction, and molecular 3D pretraining. We discuss the challenges of current molecular 3D representation learning and tasks. Further, we propose future directions to promote drug discovery and deal with these challenges. The latest advances in GDL for drug discovery are summarized in a GitHub repository https://github.com/3146830058/Geometry-Deep-Learning-for-Drug-Discovery. |
---|---|
ISSN: | 0957-4174 1873-6793 |
DOI: | 10.1016/j.eswa.2023.122498 |