Geometric Deep Learning on Molecular Representations

TL;DR

This paper reviews the emerging field of geometric deep learning applied to molecular representations, emphasizing its applications in drug discovery, chemical synthesis, and quantum chemistry, and discusses future challenges and opportunities.

Contribution

It provides a structured overview of molecular GDL, highlighting its applications, the importance of learned features, and future prospects in molecular sciences.

Findings

GDL effectively captures molecular symmetry properties.

GDL enhances molecular feature learning beyond traditional descriptors.

Future challenges include data scarcity and model interpretability.

Abstract

Geometric deep learning (GDL), which is based on neural network architectures that incorporate and process symmetry information, has emerged as a recent paradigm in artificial intelligence. GDL bears particular promise in molecular modeling applications, in which various molecular representations with different symmetry properties and levels of abstraction exist. This review provides a structured and harmonized overview of molecular GDL, highlighting its applications in drug discovery, chemical synthesis prediction, and quantum chemistry. Emphasis is placed on the relevance of the learned molecular features and their complementarity to well-established molecular descriptors. This review provides an overview of current challenges and opportunities, and presents a forecast of the future of GDL for molecular sciences.